UNIVERSITY OF CALIFORNIA MEDICAL CENTER LIBRARY SAN FRANCISCO DR. H. GUBNN BELL Digitized by tine Internet Arcinive in 2007 witin funding from IVIicrosoft Corporation littp://www.arcliive.org/details/bacteriologyOOIiissricli H. GLENN BELt A TEXT - BOOK OF BACTERIOLOGY .1H3 (^jjJjm'M^ A TEXT-BOOK OF B ACTE RI OLOGY A PRACTICAL TREATISE FOR STUDENTS^ ( / j AND PRACTITIONERS OF MEDICINt: , C\ BY PHILIP HANSON^SS, Jr., M.D. LATE PROFESSOR OF BACTERIOLOGY, COLLEGE OF PHYSICIANS AND SURGEONS, COLUMBIA UNIVERSITY, NEW YORK CITY AND HANS ZINSSER, M.D. PROFESSOR OF BACTERIOLOGY, COLLEGE OF PHYSICIANS AND SURGEONS, COLUMBIA UNI- VERSITY, NEW YORK city; BACTERIOLOGIST TO THE PRESBYTERIAN HOSPITAL; FORMERLY PROFESSOR OF BACTERIOLOGY AND IMMUNITY, STANFORD UNIVERSITY, CALIFORNIA; MAJOR, MEDICAL OFFICERS' RESERVE CORPS, U. S. A. WITH A SECTION ON THE PATHOGENIC PROTOZOA BY FREDERICK F. RUSSELL, M.D. MAJOR MEDICAL CORPS, U. S. A.; FORMERLY PROFESSOR OF BACTERIOLOGY AND PATHOLOGY, ARMY MEDICAL SCHOOL AND GEORGE WASHINGTON UNIVERSITY WITH ONE HUNDRED AND NINETY-EIGHT ILLUSTRATIONS IN THE TEXT, SOME OF WHICH ARE COLORED FOURTH EDITION NEW YORK AND LONDON D. APPLETON AND COMPANY 1919 154103 I V' v 4'i COPTRIGHT, 1910. 1914, 1916 AND 1918, D. APPLETON AND COMPANY Printed in New York, U. S. A. PREFACE TO THE FIRST EDITION The volume here presented is primarily a treatise on the funda- mental laws and technique of Bacteriology, as illustrated by their application to the study of pathogenic bacteria. So ubiquitous are the bacteria and so manifold their activities that Bacteriology, although one of the youngest of sciences, has already been divided into special fields— Medical, Sanitary, Agricul- tural, and Industrial — having little in common, except problems of general bacterial physiology and certain fundamental technical pro- cedures. From no other point of approach, however, is such a breadth of conception attainable, as through the study of bacteria in their rela- tion to disease processes in man and animals. ' Through such a study one must become familiar not only with the growth character- istics and products of the bacteria apart from the animal body, thus gaining a knowledge of methods and procedures common to the study of pathogenic and non-pathogenic organisms, but also with those complicated reactions taking place between the bacteria and their products on the one hand and the cells and fluids of the animal body on the other — ^reactions which often manifest themselves as symptoms and lesions of disease or by visible changes in the test tube. Through a study and comprehension of the processes underl5dng these reactions, our knowledge of cell physiology has been broadened, and facts of inestimable value have been discovered, which have thrown light upon some of the most obscure problems of infection and immunity and have led to hitherto unsuspected methods of treatment and diagnosis. Thus, through Medical Bacteriology — that highly specialized offshoot of General Biology and Pathology — have been given back to the parent sciences and to Medicine in general methods and knowledge of the widest application. It has been our endeavor, therefore, to present this phase of our subject in as broad and critical a manner as possible in the sections vi PREFACE dealing with infection and immunity and with methods of biological diagnosis and treatment of disease, so that the student and practi- tioner of medicine, by becoming familiar with underlying laws and principles, may not only be in a position to realize the meaning and scope of some of these newer discoveries and methods, but may be in better position to decide for themselves their proper application and limitations. We have not hesitated, whenever necessary for a proper under- standing of processes of bacterial nutrition or physiology, or for breadth of view in considering problems of the relation of bacteria to our food supply and environment, to make free use of illustrations from "the more special fields of agricultural and sanitary bacteriology, and si)me special methods of the bacteriology of sanitation are given in the last division of the book, dealing with the bacteria in relation to our food and environment. Iri conclusion it may be said that the scope and arrangement of subjects treated of in this book are the direct outcome of many years of experience in the instruction of students in medical and in advanced university courses iii bacteriology, and that it is our hope that this volume may not only meet the needs of such students but may prove of value to the practitioner of medicine for whom it has also been written. It is a pleasure to acknowledge the courtesy of those who furnished us with illustrations for use in the text, and our indebtedness to Dr. Gardner Hopkins and Professor Francis Carter Wood for a number of the photomicrographs taken especially for this work. P.H. H., Jr., H. Z. PREFACE TO THE SECOND EDITION Inquiry in the field of bacteriology is so active at the present day that no general text-book can maintain its usefulness long without frequent revision. In preparing the second edition of this book it has been our purpose to correct omissions and to incorporate the more im- portant researches of the last three years, rather than to alter exten- sively the plan of the text. From the wealth of material which these years have brought, we have attempted to glean those facts which have seemed to us most important and most directly bearing upon medical problems, since this book was planned, from the beginning, , to meet especially the needs of the student of infectious disease. The most extensive changes and additions have been made in the chapters on streptococci, tuberculosis, plague, leprosy, syphiHs, rabies, and pohomyelitis. Short sections on typhus fever, on the plague- hke disease of rodents, and on rat leprosy have been added, and we have inserted a tabulation of our knowledge of filtrable virus, adapted largely from the summary recently published by Wolbach. The Ander- son and McClintic method for the standardization of disinfectants, and Churchman's recent work on anihn dyes and bacterial growth, have been added. Many minor corrections and additions have been made throughout the text. In preparing these changes, valuable as- sistance has been given us by Dr. J. Gardner Hopkins, Associate in Bacteriology at Columbia University, and many helpful suggestions have been made by Drs. Dwyer and Bliss. It has been gratifying to note how' much of the work which seemed to us particularly valuable and enhghtening has emanated, during these three years, from American laboratories. We have purposely omitted making any extensive changes in the section on immunity. The function of this part of the book is to give the beginner a basis for further reading and introduce him, as simply as possible, to the difficult problems of the field. We have felt that the addition of much more detail and theory would render this section un- ^uited to the needs of a general text-book. It is a sorrowful necessity that this revision must be put fprth with- vii VUl PREFACE TO THE SECOND EDITION out the wise counsel of one of its authors. Since the first edition of this book was pubUshed Prof. PhiUp Hanson Hiss, Jr., has died. By his death we have lost a dear friend and a valued teacher, and American bacteriology has been deprived of a worker who was entering into the most brilliant period of his scientific maturity. H. Z. New York, 1914 PREFACE TO THIRD EDITION The need for a new edition of the Text Book has offered a welcome opportunity for the addition of the many new facts revealed by investi- . gation during the last two years. A thorough revision of the entire book has incidentally been made since experience with it in teaching, the questions of students, and the comments of associates have in- dicated sections, here and there, in which slight changes, elabora- tions or omissions, would add to clearness and simplicity. In this,- as in the choice of new material, the writer has again been guided chiefly by the desire to enhance the practical usefulness of the treatise for medi- cal students and workers interested primarily in infectious diseases. In the section on Biology and Technique, the changes have been relatively few. To the chapter on culture media, there have been added Kendall's modification of Endo's medium, Petroff's media for tubercle bacillus cultivation, Krumwiede's brilliant green medium, and Russell's double sugar medium. Many other minor, yet, in our opinion impor- tant, alterations have been made in this chapter, but throughout the book only such methods have been added as have been found to be actually useful in our own practice or in that of our associates. The section on Immunity has been very thoroughly revised in order to keep within the same small space a more accurately modern pres- entation of this complicated subject. It has not, of course, been possible to cover this field as thoroughly as ijb might be covered in a volume separately devoted to the subject, but the outline here given is intend- ed mainly to furnish a sound foundation for further study. In the chapter on the pathogenic microorganisms themselves, the most extensive changes have been necessary in the pneumococcus sec- tion and in the one on treponema pallidum. However, all the chapters have been revised with care and many small additions and omissions made where new facts and the revision of older opinions have made this desirable. H. Z. College of Physicians and Surgeons, Columbia University. N. Y. ix PREFACE TO THE FOURTH EDITION The most important change incorporated in the fourth edition of our Text-Book is the section on Pathogenic Protozoa written by- Major Frederick F. Russell, of the Medical Corps of the United States Army. When the first edition of this book was written the ad- dition of a section on Protozoa was seriously considered, but was finally omitted because it was the desire of both the writers at that time that nothing should go into the book that could not be based on personal knowledge, and, since neither the late Dr. Hiss nor the undersigned had worked systematically in the field of protozoology, it was thought better to limit the book to our own field of bacteriology. The book has since then gone through three editions and it has be- come more and more apparent that it has found its greatest useful- ness as a text-book for medical students and a reference book for physicians and laboratory workers. For these purposes, however, it has had the serious defect of giving no information about such microorganisms as the malaria plasmodia, the trypanosomes and other pathogenic microorganisms, which, though technically protozoa, are necessary objects of interest to all medical laboratory workers. It has appeared to us, therefore, that the practical value of the book would be greatly enhanced by the addition of a short treatise on the protozoa, if written by someone who had been in touch with this field, rather from the practical medical point of view than from that of the zoologist. Major Russell has kindly consented to supply this de- ficiency in a section which makes no claim to zoological complete- ness, but is written with the definite purpose of giving medical workers concise information concerning the important pathogenic xii PREFACE TO THE FOURTH EDITION species, with especial consideration of their common occurrence, the methods of their detection and recognition, and correlation to the diseases which they incite. In other respects the book has been thoroughly gone over. In the section on Biology and T^ichnique a few methods, which we ourselves have found it unnecessary to use, have been omitted. A few other methods have been added and a few of the most useful ones ampli- fied. Minor changes /have been made in the section on Immunity. ^ The chapter on Streptococcus has been revised and considerable addi- tions and changes have been made in the chapters on the Paratyphoid and Typhoid '9^cilli. The more recent work on the Schick Test and on the Det^mination of Virulence of the Diphtheria Bacillus has been incorporated. Minor changes have been made in the sections on Bacteria in Water and Milk. Altogether it has been attempted to bring the material up to date, especially in those features in which we think its greatest usefulness lies,|namely, as a guide for the actual performance of bacteriological work. It is a pleasure again to acknowledge the valuable assistance of Prof . J. G. Hopkins in correcting and rearranging the book. , Hans Zinsser. : College of 'Physicians and Surgeons, Columbia University, N^ew York City. CONTENTS SECTION I THE GENERAL BIOLOGY OF BACTERIA AND THE TECHNIQUE OF BACTERIOLOGICAL STUDY CHAPTER PAGE I. The Development and Scope of Bacteriology ..".... 1 IT General Morphology, Reproduction, and Chemical and Physical Properties of the Bacteria ........ 9 III. The Relation of Bacteria to Environment, and Their Classification , 25 IV. The Biological Activities of Bacteria 40 V. The Destruction of Bacteria 62 VI. Methods Used in the Microscopic Study and Staining of Bacteria 93 VII. The Preparation of Culture Media 113 VIII. Methods Used in the Cultivation of Bacteria 141 IX Methods Determining Biological Activities of Bacteria , 164 X. The Bacteriological Examination of Material from Patients ..-.•... 174 SECTION n INFECTION AND IMMUNITY CHAPTER PAGE XI, Fundamental Factors of Pathogenicity and Infection . . 181 XII. Defensive Factors of the Animal Organism 189 XIII. Toxins and Antitoxins 203 XIV. Production and Testing of Antitoxins 216 XV. Lysins, Agglutinins, Precipitins, and Other Antibodies , 224 XVI, The Technique of Serum Reactions . . . . . , . . . 250 xiii XIV Contents CHAPTER PAGE XVII. Phagocytosis 275 XVIII. Opsonins, Leucocyte Extract, and Aggressins 281 XIX. Anaphylaxis or Hypersusceptibility ........ 295 XX. Facts and Problems of Immunity in their Bearing upon THE Treatment of Infectious Diseases 305 SECTION III PATHOGENIC MICROORGANISMS CHAPTER PAGE XXI. The Staphylococci (Micrococci) 321 XXII. The Streptococci 335 XXIII. DiPLOCOCCUS PNEUMONIA 352 XXIV. MiCROCOcctrs intracellularis meningitidis (Meningococcus) . 371 XXV. DiPLOcoccus GONORRHCE^ (Gonococcus) , Micrococcus ca- tarRhalis, AND Other Gram-negative Cocci 380 XXVI. Bacilli of the Colon-Typhoid-Dysentery Group — Bacillus COLI COMMUNIS, 388 XXVII. !0AciLLi OF THE Colon-Typhoid-Dysentery Group (continued) —Bacillus of Typhoid Fever 399 XX VIII. Bacilli of the Colon-Typhoid-Dysentery Group (continued) — Bacilli Intermediate between the Typhoid and Colon X)rganisms 428 XXIX. Bacilli of the Colon-Typhoid-Dysentery Group (continued) — The Dysentery Bacilli 435 XXX. Bacillus mucosus capsulatus 447 XXXI. Bacillus tetani ' . 456 XXXII. Bacillus of Symptomatic Anthrax, Bacillus of Malignant Edema, Bacillus aerogenes capsulatus. Bacillus botu- LiNUS 465 XXXIII. The Tubercle Bacillus 479 XXXIV. The Smegma Bacillus and the Bacillus of Leprosy . . . 503 XXXV. Bacillus diphtheria. Bacillus Hoffmanni, and Bacillus xerosis 512 XXXVI. Bacillus mallei 528 XXXVII. Bacillus influenzae and Closely Related Bacteria . . . 536 i CONTENTS XV CHAPTER PAOB XXXVIII. . Bordet-Gengou Bacillus, Morax-Axenfeld Bacillus, Zur Nedden Bacillus, Ducrey Bacillus 543 XXXIX. The Bacilli of the Hemorrhagic Septicemia Group aj^d Bacillus pestis 551 XL. Bacillus anthracis and Anthrax , . 553 XLI. Bacillus pyocyaneus 577 XLII. Asiatic Cholera and the Cholera Organism 582 XLIII. Diseases Caused by Spirochetes 592 XLIV. The Higher Bacteria 618 XLV. The Yeasts 629 XLVI. Hyphomycbtes 635 SECTION IV EXANTHEMATA AND DISEASES CAUSED BY FILTRABLE VIRUS CHAPTER PAOK XLVII. Rabies 646 XLVIII. Smallpox 657 XLIX. Acute Anterior Poliomyelitis 664 L. Yellow Fever '. . 668 LI. Measles, Scarlet Fever, and Foot-and-Mouth Disease . . 675 SECTION V BACTERIA IN AIR, SOIL,' WATER, AND MILK CHAPTER PAGE LII. Bacteria in the Air and Soil 683 LIII. Bacteria in Water 689 tlV. Bacteria in Milk and MiiiK Products, Bactewa in the Inpu^TIU?^ . .. •. .. .. •. . n ••.••* ^ . "I .. . 699 xvi CONTENTS SECTION VI PATHOGENIC PROTOZOA Maj. Frederick F. Russell, Med. Corps, U. S. Army chapteb page INTRODUCTION 721* LV. Class I — Sarcodina (Rhizopoda) 723 LVI. Class II — Mastigophora (Diesing) , 738 . LVII. Class III— Sporozoa . 760 -.JiVIII, Class IV— Infusoria .792 tiiX. Technique of Blood Examinations for Protozoa . . . . . 795 \lNDEX OF AUTHORSn 799 [DEX OF SUBJECTsX 811 LIST OF ILLUSTRATIONS FIGURE PAGE 1. Types of bacterial morphology 10 2. Bacterial capsules 12 3. Arrangement of bacterial flagella 15 4. Various positions of spores in bacterial cell 17 5. Germination of spores * 17 6. Degeneration forms of Bacillus diphtheriae 19 7. Degeneration forms of Bacillus pestis » . 20 8. Hot-air sterilizer 69 9. Arnold sterilizer 70 10. Low-temperature sterilizer 71 11. Autoclave 72 12. Lentz formalin apparatus . 90 13. Breslau formaldehyde generator and section of same 91 14. Hanging drop preparation 94 15. Florence flask 114 16. Erlenmeyer flask 114 17. Petri dish 115 18. Test tubes, showing method of stoppering 116 19. Burette for titrating media 117 20. Tubing media 118 21. Media in tubes 119 22. Berkefeld filter 120 23. Berkefeld filter 121 24. Reichel filter 122 25. Kitasato filter .......' 123 26. Maassen filter, for small quantities of fluid 124 27. Platinum wires 142 28. Taking plugs from tubes before inoculation 143 29. Inoculating 144 30. Pouring inoculating medium into Petri dish 145 31. Streak plate 147 32. Deep stab cultivation of anaerobic bacteria 148 33. Deep stab cultivation of anaerobic bacteria 149 34. Cultivation of anaerobes in fluid under albolin 150 35. Wright's method of anaerobic cultivation in fluid media 151 36. Novy jar 152 37. Wright's method of anaerobic cultivation by the use of pyrogallic-acid solution 153 38. Jar for anaerobic cultivation 154 2 xvii xviii LIST OF ILLUSTRATIONS FIGURE PAGE 39. Apparatus for combining the methods of exhaustion, hydrogen replace- ment, and oxygen absorption 155 40. Simple apparatus for plate cultivation of anaerobic bacteria 156 41. Incubator 157 42. Thermo-regulator 158 43. Thermo-regulator 158 44. Moitessier gas-pressure regulator 160 45. Variations in the conformation of the borders of bacterial colonies . . 161 46. Wolffhiigel counting-plate 162 47. Types of fermentation tubes 165 48. Types of gelatin liquefaction by bacteria 169 49. Intraperitoneal inoculation of rabbit 170 50. Intravenous inoculation of rabbit 170 51. Intraperitoneal inoculation of guinea*pig 171 52. Guinea-pig cage \ 171 53. Rabbit cage J . 172 54. Blood-culture plate showing streptococcus colonies 179 55. Toxin and body cell 206 56. Toxin and antitoxin .... / 214 57. Ehrlich's conception of^cel^receistors, giving rise to lytic immune bodies 227 58. Complement, amboceptor or immune body, and antigen or immunizing substance 227 59. Microscopic agglutination reaction 229 60. Macroscopic agglutination 230 61. Ehrlich's conception of the structure of agglutinins and precipitins . . 238 82. The structure of cell-receptors and immune bodies, according to Ehrlich's conception 239 63. Neisser and Wechsberg's conception of complement deviation .... 246 64. Schematic representation of complement fixation in the Bordet-Gengou reaction 248 65. Capillary pipette for removal of exudate in doing the Pfeiffer test . .256 66. Wright's capsule for collecting blood 284 67. Pipette for opsonic work 285 68. Pipette with three substances, — corpuscles, bacteria, and serum, as first taken up 285 69. Staphylococcus pyogenes aureus 322 70. Staphylococcus colonies 323 71. Micrococcus tetragenus 334 72. Streptococcus pyogenes 336 73. Streptococcus colonies on serum agar 339 74. Streptococcus colonies from blood culture on blood-agar plate .... 345 75. Pneumococci 354 76. Pneumococci 354 77. Meningococcus, pure culture 372 78. Meningococcus in spinal fluid 373 79. Meningococcus culture (streak plate) .375 80. Gonococcus pus from urethra 38J LIST OF ILLUSTRATIONS xix MGTniB PAGE 8L Gonococcus . . . . . . ... . . ... . . . , . . . 382 82. Gonococcus colony 383 83. Bacillus coli communis 390 84. Bacillus coli communis on various media 396 85. Bacillus coli communior on various media 397 86. Bacillus typhosus 400 87. Bacillus typhosus, showing flagella 401 88. Surface colony of Bacillus typhosus on gelatin 402 89. Bacillus coU; deep colonies on Hiss plate medium 407 90. Bacillus typhosus; deep colonies in Hiss plate medium 408 9L Bacillus typhosus; colony in Hiss plate medium, highly magnified . . . 409 92. Colon and typhoid colonies in Hiss plate medium 410 93. Scheme of fermentations of the dysentery-typhoid-colon-bacilli in carbo- hydrate serum-water media 445 94. Bacillus mucosus capsulatus 448 95. Bacilli of rhinoscleroma 452 96. Bacillus tetani 457 97. Young tetanus culture in glucose agar 458 98. Older tetanus culture in glucose agar 459 99. Bacillus of symptomatic anthrax , 466 100. Bacillus of symptomatic anthrax, culture in glucose agar 467 101. Bacillus of malignant edema 469 102. Bacillus of malignant edema, culture in glucose agar 470 103. Tubercle bacilli in sputum 480 104. Culture of Bacillus tuberculosis in flask of glycerin bouillon ..... 486 105. Bacillus diphtheriae 513 106. Colonies of Bacillus diphtheriae on glycerin agar 518 107. Bacillus Hoffmanni 524 108. Colonies of Bacillus Hoffmanni on agar 625 109. Bacillus xerosis 626 110. Glanders bacillus 529 111. Glanders bacilU in tissue , 531 112. Bacillus influenzae; smear from pure culture on blood agar 537 113. Bacillus influenzae; smear from sputum 538 114. Colonies of influenza bacillus on blood agar 539 V15. Koch- Weeks bacillus 542 116. Bordet-Gengou bacillus 544 117. Morax-Axenfeld diplo-bacillus 546 118. Bacillus pestis • 655 119. Bacillus pestis, involution forms . . . . , 656 120. Bacillus anthracis; pure culture on agar 564 121. Bacillus anthracis, in kidney tissue 565 122. Bacillus anthracis, in spleen tissue 566 123. Anthrax colony on gelatin 567 124. Anthrax colony on agar 568 125. Bacillus subtilis . 576 126. Cholera spirillum .....••... 583 XX LIST OF ILLUSTRATIONS FIGURE .PAGE 127. Cholera spirillum; stab cultures in gelatin, three days old 586 128. Cholera spirillum; stab culture in gelatin, six days old 586 129. Spirochaete pallida; smear from chancre 594 130. Spirochaete pallida, in spleen of congenital syphilis 600 131. Spirochaete paUida, in hver of congenital syphilis 601 132. Spirochaete of relapsing fever 605 133. Spirochaete of relapsing fever 606 134. Spirochaete of relapsing fever 607 135. Spirochaete of Dutton, African tick fever 609 136. Smear from the throat of a case of Vincent's angina 611 137. Throat smear, Vincent's angina 612 138. Spirochaete gallinarum 616 139. Cladothrix, showing false branching 620 140. Streptothrix, showing true branching 621 141. Actinomyces granule crushed beneath a cover-glass 624 142. Actinomyces granule crushed beneath a cover-glass 625 143. Branching filaments of actinomyces / 626 144. Yeast cells . \ / 630 145. Mucor mucedo . . . . . y 636 146. Mucor mucedo V. . '^^' 637 147. Mucor mucedo .638 148. Mucor ramosus 639 149. Penicillium glaucmn 640 150. Aspergillus glaucus 641 151. Thrush 642 152. Achorion Schoenleinii 643 153. Method of drying spinal cord of rabbit for purposes of attenuation . . . 653 154. Stegomyia fasciata 671 155. Bacillus bulgaricus 717 156. Endameba histoljrtica .6 724 157. Endameba histolytica 725 158. Endameba histolytica 729 159. Endameba histolytica 730 160. Endameba histolytica o 730 161. Endameba coli . . . . = 732 162. Endameba coli .... » 733 163. Endameba coli 734 164. Endameba coli cyst ....... c ... 735 165. Endameba coli 736 166. Trichomonas intestinalis 738 167. Lamblia intestinalis. Cyst formation 739 168. Tr3rpanosoma rotatorium in blood of frog 742 169. Trypanosoma Lewsi 743 170. Most important trypanosomes parasitic in vertebrates 746 171. Dourine. Showing swelling of genitalia and plaques on the skin . . . . 749 172. Trypanosoma avium in blood of common wild birds 750 173. Trypanosoma avium in culture on blood agar 751 LIST OF ILLUSTRATIONS xxi FIGURE PAOB 174. Trypanosoma garabiense 752 175. Tsetse fly (Glossina palpalis) 753 176. Schizotrypanum cnizi in human blood 755 177. Schizotrypanum cruzi developing in tissues of guinea-pig 756 178. Leishmania Donovani 757 179. Leishmania infantum 759 180. Hsemoproteus columbse 762 181. Proteosoma precox in blood of field lark 763 182. Midgut of culex mosquito, covered with oocysts of Proteosoma praecox . 764 183. Plasmodium vivax 766 184. Plasmodium vivax 766 185. Plasmodium vivax, an atypical macrogametocyte 767 186. Plasmodium vivax 767 187. Plasmodium vivax ' 768 188. Plasmodium vivax 769 189. Plasmodium malarise 770 190. Plasmodium malarise 770 191. Plasmodium malarise 771 192. Plasmodium falciparum 771 193. Plasmodium falciparum 772 194. Plasmodium falciparum, male crescent 773 195. Comparison of culex and anopheles 779 196. Babesia bigeminum 787 197. Texas fever tick (Margaropus annulatus) 789 198. Balantidium coli in a follicle of the colon, breaking through the mucosa . 793 ' SECTION I THE GENERAL BIOLOGY OF BACTERIA AND THE TECHNIQUE OF BACTERIOLOGICAL STUDY CHAPTER I THE DEVELOPMENT AND SCOPE OF BACTERIOLOGY As we trace back to their ultimate origins the lines of development of living beings of the animal and plant kingdoms, we find them con- verging toward a common type, represented by a large group of uni- cellular organisms, so simple in structure, so unspecialized in function, that their classification in either the realm of plants or that of animals becomes little more than an academic question. However, even such microorganisms, in which the functions of nutrition, respiration, loco- motion, and reproduction are concentrated within the confines of a single cell, and in which adaptation to special conditions more readily brings about modifications leading to the production of a multitude of delicately graded transitional forms, fall into groups which, either in structure or in biological attributes show evidence of a tendency toward one or the other of the great kingdoms. Most important of these unicellular forms, for the student of medical science, are the bacteria and the protozoa. The former, by reason of their undifferentiated protoplasm, their occasional possession of cellulose membranes, their biological tendency to synthetize, as well as to break down organic compounds, and because of the transitional forms which seem to connect them directly with the lower plants, are generally placed in the plant kingdom. The latter, chiefly on the basis of metabolism, are classified with the animals. Knowledge of the existence of microorganisms as minute as the ones under discussion, was of necessity forced to await the perfection of instruments of magnification. It was not until the latter half of the seventeenth century, therefore, that the Jesuit, Kircher, in 1659, and the Dutch linen-draper, van Leeuwenhoek, in 1675, actually saw and 1 2 BIOLOGY AND TECHNIQUE described living beings too small to be seen with the naked eye. There can be no doubt that the small bodies se^n^hy these men and their many immediate successors were, at lea^st In part, bacteria.. And indeed the descriptions and illustrations of several of the earliest workers cor- respond with many oM^he forms which are well known to us at the present day. During the century following the work of these pioneers, the efforts of investigators lay chiefly in the more exact-morphological description of some of the forms of unicellular life, already known. Conspicuous among the work of this period is that of Otto Friedrich Miiller. In the generation following MuUer's work, however, a marked advance in the study of these forms was made by Ehrenberg,* who established a classification which, in some of its cardinal divisions, is retained until the present day. Meanwhile the regularity with which these "animalcula" or "in- fusion animaicula" were demonstrable in tartar from the teeth, in intes- tinal contents, in well-water, etc., had begun to arouse in the minds of the more advanced physicians of the time a suspicion as to a possible relationship of these minute forms with disease. The conception of "contagion,'' or transmission of a disease from one human being to another, was, however, even at this time, centuries old. The fact had been recognized by Aristotle, had been reiterated by medieval philos- ophers, and had led, in 1546, to the division of contagious diseases by Fracastor, into those transmitted "per contactum,'' and those con- veyed indirectly "per fomitem.'' It was for these mysterious facts of the transmissibility of disease, that clinicians of the eighteenth centurj'-, with remarkable insight, saw an explanation in the microorganisms dis- covered by Leeuwenhoek and his followers. In fact, Plenciz of Vienna, writing in 1762, not only expressed a belief in the direct etiological connection between microorganisms and some diseases, but was the first to advance the opinion that each malady had its own specific causal agent, which multiplied enormously in the diseased body. The opinions of this author, if translated into the language of our modern knowledge of the subject, came remark- ably near to the truth, not only as regards etiology and transmission, but also in their suggestion of a specific therapy for each disease. The conception of a " contagium vivum '' was thus practically es- tablished with the work of Plenciz and many others who followed in 1 " Die Infusionstierchen/' etc., Leipzig. 1838. • DEVELOPMENT AND SCOPE OF BACTERIOLOGY 3 his train, but the astonishingly shallow impression which the acute reasoning of these men left upon the medical thought of their day furnishes an excellent example of the futility of the most penetrating speculation when unsupported by experimental data. The real advancement in the scientific development of the subject was destined to be carried on along entirely different lines. In 1837, Schwann, a botanist, showed that the yeasts, found in fermenting sub- stances, were Hving beings, which bore a causal relationship to the proc- ess of fermentation. At almost the same time, similar observations were made by a French physicist, Cagniard-Latour. The opinions advanced by these men on the nature of fermentation aroused much interest and discussion, since, at that time and for a long period thereafter, fermentation was ascribed universally to proteid decomposition, a process which was entirely obscure but firmly believed to be of a purely chemical nature. Although belief in the discovery of Schwann did not completely master the field until after Pasteur had completed his classical studies upon the fermentations occurring in beer and wine, yet the conception of a " fermentum vivum " aroused much speculation, and the attention of physicians and scientists was attracted to the many analogies ex- isting between phenomena of fermentation and those of disease. The conception of such an analogy, however, was not a new thought in the philosophy of the time. Long before Schwann and Cagniard- Latour, the philosopher Robert Boyle, working in the seventeenth century, had prophesied that the mystery of infectious diseases would be solved by him who should succeed in elucidating the nature of fermenta- tion. Nevertheless, the diligent search for microorganisms in relation to various diseases which followed, led to few results, and the successes which were attained were limited to the diseases caused by some of the larger fungi, favus (1839), thrush (1839), and pityriasis versicolor (1846). During this time of ardent but often poorly controlled etiolog- ical research, it was Henle who formulated the postulates of conserva- tism, almost as rigid as the later postulates of Koch, requiring that proof of the etiological relationship of a microorganism to a disease could not be brought merely by finding it in a lesion of the disease, but that constant presence in such lesions must be proven and isolation and study of the microorganism away from the diseased body must be car- ried out. It was during this period also that one of the most fundamental 4 BIOLOGY AND/TECHNIQUEX questions, namely, that of the origin of these minlute living beings, wa^ being discussed with much passion by the scientific world. It was-held by the conservative majority that the microorganisms described by Leeuwenhoek and others after him, were produced by spontaneous generation. The doctrine of spontaneous generation, in fact, was solidly established and sanctified by tradition, and had been applied in the past not alone to microorganisms.* And it must not be forgotten that without the aid of our modern methods of study, satisfactory proof for or against such a process was not easily brought. Needham, who published in 1749, had spent much time in fortify- ing his opinions in favor of spontaneous generation by extensive ex- perimentation. He had placed putrefying material and vegetable in- fusions in sealed flasks, exposing them for a short time to heat, by immersing them in a vessel of boiling water, and had later shown them to be teeming with microorganisms. He was supported in his views by no less an authority than Buffon. The work of Needham, however, showed a number of experimental inaccuracies which were thoroughly sifted by the Abbe Spallanazani. This investigator repeated the ex- periments of Needham, employing, however, greater care fn sealing his flasks, and subjecting them to a more thorough exposure to heat. His results did not support the views of Needham, but were answered by the latter with the argument that by excessive heating he had pro- duced chemical changes in his solutions which had made spontaneous generation impossible. The experiments of Schulze, in 1836, who failed to. find living organ- isms in infusions which had been boiled, and to which air had been admitted only after passage through strongly acid solutions, and similar results obtained by Schwann, who had passed the air through highly heated tubes, were open to criticism by their opponents, who claimed that chemical alteration of the air subjected to such drastic influences, had been responsible for the absence of bacteria in the infusion. Similar experiments by Schroeder and Dusch, who had stoppered their flasks with cotton plugs, were not open to this objection, but had also failed to convince. The question was not -definitely settled until the years im- » Valleri-Radot, in his life of Pasteur, stated that Van Helmont, in the six- teenth century, had given a celebrated prescription for the creation of mice from dirty linen and a few grains of wheat or pieces of cheese. During the centu- ries following, although, of course, such remarkable and amusing beliefs no longer held sway, nevertheless the question of spontaneous generation of minute and structureless bodies, like the bacteria, still foimd learned and thoughtful partisans. DEVELOPMENT AND SCOPE OP BACTERIOLOGY 5 mediately following 1860, when Pasteur conducted a series of experi- ments which were not only important in incontrovertibly refuting the doctrine of spontaneous generation, but in establishing the principles of scientific investigation which have influenced bacteriological re- search since his time.^ Pasteur attacked the problem from two points of view. In the first place he demonstrated that when air was filtered through cotton- wool, innumerable microorganisms were deposited upon the filter. A single shred of such a contaminated filter dropped into a flask of pre- viously sterilized nutritive fluid, sufiiced to bring about a rapid and luxuriant growth of microorganisms. In the second place, he succeeded in showing that similar, sterilized " putrescible " liquids, if left in con- tact with air, would remain uncontaminated provided that the en- trance of dust particles were prohibited. This he succeeded in doing by devising flasks, the necks of which had been drawn out into fine tubes bent in the form of a U. The ends of these U-tubes, being left open, permitted the sedimentation of dust from the air as far as the lowest angle of the tube, but, in the absence of an air current, no lust was carried up the second arm into the liquid. In such flasks he showed that no contamination took place but could be immediately induced by slanting the entire apparatus until the liquid was allowed to run into the bent arm of the U-tube. Finally, by exposing a series of flasks containing sterile yeast infusion, at different atmospheric levels, in places in which the air was subject to varying degrees of dust con- tamination, he showed an inverse relationship between the purity ox the air and the contaminatioti of his flasks with microorganisms. The doctrine of spontaneous generation had thus received its final refutation, except in one particular. It was not yet clear why com- plete sterility was not always obtained by the application of definite degrees of heat. This final link in the chain of evidence was supplied, some ten years later, by Cohn, who, in 1871, was the first to observe and correctly interpret bacterial spores and to demonstrate their high powers of resistance against heat and other deleterious influences. ^ In a letter to his foremost opponent, at this period, Pasteur writes: "In experimental science, it is always a mistake not to doubt when facts do not compel affirmation." The critical spirit pervading the scientific thought of that time in France ii^ also well expressed by Oliver Wendell Holmes, who said that he had learned three things in Paris: "Not to take authority when I can have facts, not to guess when I can know, and not to think that a man must take physic because he is sick." 6 BIOLOGJV AND TECHNIQUE Meanwhile, Pasteur, parallel with his researches upon spontaneous generation, had been ^eirrying on experiments upon the subject of fermentation along t!ie lines suggested by Cagniard-Latour. As a consequence of these experiments, he not only confirmed the opinions both of this author and of Schwann concerning the fermentation of beer and wine by. yeasts, but was able to show that a number of o^her fer- mentations, such as those of lactic and butyric acid, as well as the de- composition of organic matter by putrefaction, were directly due to the action of microorganisms. It was the discovery of the living agents underlying putrefaction, especially, which exerted the most active influence upon the medical research of the day. This is illustrated by Lister's work. The suppurative processes occurring in infected wounds had long been regarded as a species of putrefaction, and Lord Lister, working directly upon the premises supplied by Pasteur, introduced into both the active and prophylactic treatment of surgical wounds, the antiseptic principles which alone have made modern surgery possible. There now followed a period in which bacteriological investigation was concentrated upon problems of etiology. Stimulated by Pasteur's successes, the long-cherished hope of finding some specific microorgan- ism as the causal agent in each infectious disease was revived. PoUender, in 1855, had reported the presence of rod-shaped bodies in the blood and spleen of animals dead of anthrax. Brauell, several years later, had made sijnilar observations and had expressed definite opinions as to the causative relationship of these rods to the disease. Convincing proof, however, had not been brought, by either of these observers. Finally, in 1863, Davaine, in' a series of brilliant investi- gations, not only confirmed the observations of the two authors men- tioned above, but succeeded in demonstrating that the disease could be transmitted by means of blood containing these rods and could never be transmitted by blood from which these rods were absent. Anthrax, thus, is the first disease in which definite proof of bacterial causation was brought. Speaking before the French Academy of Medicine at this time, Davaine suggested that the manifestations of the disease might in reality represent the results of a specific fermentation produced by the bacilli he had found. This, in a crude way, expresses the modern conception of infectious disease. Within a f^ years after this, 1868, the adherents of the parasitic theory of infectious diseases were further encouraged by the discovery, by Obermeier, of a spirillum in the blood of patients suffering from DEVELOPMENT AND SCOPE OF BACTERIOLOGY 7 relapsing fever. It is not surprising that the successes attained in these diseases, fostering hope of analogous results in all other similar condi- tions, but without the aid of adequate experimental methods, should have led to many unjustified claims and to much fantastic theorizing. Thus Hallier, at about this time, advanced a theory as to the etiology of infectious diseases, in which he attributed all such conditions to the moulds or hyphomycetes, regarding the smaller form or bacteria as developmental stages of these more complicated forms. Extravagant conjectures of this kind, however, did not maintain themselves for any length of time in the light of the critical attitude which was. already pervading bacteriological research. Progress was made during the years immediately following, chiefly in the elucidation of suppurative processes. Rindfleisch, von Reckling- hausen, and Waldeyer, almost simultaneously, described bodies which they observed in sections of tissue containing abscesses, and which they believed to be microorganisms. Notable support was given to their opinion by similar observations made upon pus by Klebs, in 1870. In view, however, of the purely morphological nature of their studies, the opinions of these observers did not entirely prevail. Satisfactory methods of cultivation and isolation had not yet been developed, and Billroth and his followers, with a conservatism entirely justified under existing conditions, while admitting the constant presence of bacteria in purulent lesions, denied their etiological significance. The contro- versy that followed was rich in suggestions which greatly facilitated the work of later investigators, but could not be definitely settled until 1880, when Koch introduced the technical methods which have made bacteriology an exact science. By the use of solid nutritive media, the isolation of bacteria and their biological study in pure culture were made possible. At about the same time the use of anilin dyes, developed by Weigert, Koch, and Ehrlich, was introduced *jito morphological study and facilitated the observation of the finer structural details which had been unnoticed while only the grosser methods employed for tissue staining had been available. With the publication of Koch's work, there began an era unusually rich in results held in leash heretofore by inadequate technical methods. The discovery of the typhoid bacillus in 1880, of the bacillus of fowl cholera and the pneumococcus in the same year, and of the tubercle bacillus in 1882, initiated a series of etiological discoveries which, ex- tending over not more than fifteen years, elucidated the causation of a majority of the infectious diseases. 8 BIOLOGY AND TECHNIQUE Coincident with the elucidation of etiological facts began the inquiry into that field which is now spoken of as the science of immunity. The phenomena which accompany the development of insusceptibility to bacterial infections in man and in animals, first studied by Pasteur, have become the subject of innumerable researches and have led to results of the utmost practical value. The problems which were encountered were firgit> studied from a purely bacteriological point of view, but their solution has shed light upon biological principles of the broadest application. Investigations into the properties of immune sera, while making bacteriology one of the most important branches of diagnostic and therapeutic medicine, have, at the same time, inseparably linked it with physiology and experimental pathology. By the revelations of etiological research, and by the study of the biological properties of pathogenic bacteria, contagion, an enemy hitherto unseen and mysterious, was unmasked, and rational campaigns of public sanitation and personal hygiene were made possible. Upon the same elucidations has depended the development of modern surgery — a science which without asepsis and antisepsis would have been doomed to remain in its medieval condition. Apart from its importance in the purely medical sciences, the study of the bacteria has shed beneficial light, moreover, upon many other fields of human activity. In their relationship to decomposition, the conditions of the soil, and to diseases of plants, the bacteria Jiave been found to occupy a position of great importance in agriculture. Knowl- edge of bacterial and yeast ferments, furthermore, has become the scien- tific basis of many industries, chiefly those concerned in the production of wine, beer, and dairy products. The scope of bacteriology is thus a wide one, and none of its various fields has, as yet, been fully explored. The future of the science is rich in allurement of interest, in promise of result, and in possible benefit to mankind. CHAPTER II GENERAL MORPHOLOGY, REPRODUCTION, AND CHEMICAL AND PHYSICAL PROPERTIES OF THE BACTERIA Bacteria are exceedingly minute unicellular organisms which maj occur perfectly free and singular, or in larger or smaller aggregations, thus forming multicellular groups or colonies, the individuals of which are, however, physiologically independent. The cells themselves have a number of basic or ground shapes which may be roughly considered in three main classes: The cocci or spheres, the bacilli or straight rods, and the spirilla or curved rod forms. The cocci are, when fully developed and free, perfectly spherical. When two or more are in apposition, they may be slightly flattened along the tangential surfaces, giving an oval appearance. The bacilli, or rod-shaped forms, consist of elongated cells whose long diameter may be from two to ten times as great as their width, with ends squarely cut off, as in the case of bacillus anthracis, or gently rounded as in the case of the typhoid bacillus. The spirilla may vary from small comma-shaped microorganisms, containing but a single curve, to longer or more sinuous forms which may roughly be compared to a corkscrew, being made up of five, six, or more curves. The turns in the typical microorganisms of this class are always in three planes and are spiral rather than simply curved. Among the known microorganisms, the bacilli by far outnumber other forms, and are probably the most common variety of bacteria in existence. Many variations from these fundamental types may occur even under normal conditions, but contrary to earlier opinions it is now positively known that cocci regularly reproduce cocci, bacilli bacilli, and spirilla spirilla, there being, as far as we know, no mutation from one form into another. The size of bacteria is subject to considerable variation. Cocci may vary from .15 /jl to 2. /x in diameter. The average size of the ordinary pus coccus varies from .8 n to 1.2 /z in diameter. Fischer has given a graphic illustration of the size of a staphylococcus by calculating that one billion micrococci could easily be contained in a drop of water hav- 9 10 BIOLOGY And technique ing a volume of one cubic millimeter. Among the bacilli the range of size is subject to even greater variations. Probably the smallest of the common bacilli is the bacillus of influenza which measures about .5 /i in length by .2 j« in thickness. The limit of the optical possibilities of the modern microscope is almost reached by some of the known micro- organisms, and it is not at all out of question that some of the diseases, for which, up to the present time, no specific microorganisms have Fig. 1. — ^Types of Bacterial Morphology. been found, may be caused by bacteria so small as to be invisible by any of our present methods. In fact, the virus causing the peripneumonia of cattle has been shown to pass through the pores of a Berkefeld filter, which are impenetrable to the smallest of the known bacteria.* MORPHOLOGY OF THE BACTERIAL CELL When unstained, most bacteria are transparent, colorless, and ap- parently homogeneous bodies with a low refractive index. The cells themselves consist of a mass of protoplasm, surrounded, in most in- stances, by a delicate cell membrane. The presence of a nucleus^ in bacterial cells, though denied by the earlier writers, has been demonstrated beyond question by Zettnow, Nakanishi,^ and others. The original opinion of Zettnow was that the entire bacterial body consisted of nuclear material intimately inter- mingled with the Cytoplasm. The opinion now held by most observers ' Nocard and Roux, Ann. Past., 12, 1898. 2 A. Fischer, Jahrbiicher f. wissen. Botanik, xxvii, • Ndkanishi, Miinch. med. Woch., vi, 1900. MORPHOLOGY, REPRODUCTION, ETC. 11 who have studied this phase of the subject favors the existence of an ectoplasmic zone which includes cell membrane and fiagella, but is definitely a part of the cytoplasm, and an entoplasm in which is con- centrated the nuclear material. Biitschli ' claims to have demonstrated within this entoplasmic substance a reticular meshwork, between the spaces of which lie granules of chromophilic or nuclear material. Confirmation of this opinion has been brought by Zettnow" and others. Nakanishi, working with a special staining method, asserts that some microorganisms show within the entoplasmic zone a well-defined, minute, round or oval nucleus, which possesses a definitely charac- teristic staining reaction.^ In the bodies of a large number of bacteria, notably in those of the diphtheria group, Ernst,* Babes,^ and others have demonstrated granular, deeply staining bodies now spoken of as metachromatic granules, or Babes-Ernst granules, or, because of their frequent position at the ends of bacilli, as polar bodies. These structures are irregular in size and number, and have a strong affinity for dyes. They are stained dis- tinctly dark in contrast to the rest of the bacterial cell with methylene blue, and may be demonstrated by the special methods of Neisser and of Roux.^ Their interpretation has been a matter of much difficulty and of varied opinion. Those who first observed them held that they were a part of the nuclear material of the cell. Others have regarded them as an early stage in spore formation, or as arthrospores.' Again, they have been interpreted as structures comparable to the centrosomes of other unicellular forms. As a matter of fact, the true nature of these bodies is by no means certain. They are present most regularly in microorganisms taken from young ^nd vigorous cultures or in those taken directly from the lesions of disease. It is unlikely that they repre- » Biitschli, " Bau der Bakterien," Leipzig, 1890. 2 Zettnow, Zeit. f . Hyg., xxiv, 1897. 3 The method of Nakanishi is carried out as follows: Thoroughly cleansed slides are covered with a saturated aqueous solution of methylene blue. This is spread over the slide in an even film and allowed to dry. After drying, the slide should be of a transparent, sky-blue color. The microorganisms to be examined are then emulsified in warm water, or are taken from the fluid media, and dropped upon a cover slip. This is placed, face downward, upon the blue ground of the slide. In this way, bacteria are stained without fixation. Nakanishi claims that by this method the entoplasm is stained blue, while the nuclear material appears of a reddish or purplish hue. '» Ernst, Zeit. f. Hyg., iv, 1888. » Babes, Zeit. f. Hyg., v, 1889. • See section on stains, p. 107. » See section on sporulation, p. 16. 12 BIOLOGY AiND TECHNIQUE sent structures in any way comparable to spores, since cultures con- taining individuals showing metachromatic granules are not more resistant to deleterious influences than are others. Their abundant presence in young vigorous cultures may indicate a relationship between them and the growth energy of the microorganisms. There is no proof, however, that these bodies affect the virulence of the bacteria. Cell Membrane and Capsule. — Actual proof of the existence of a cell membrane has been brought in the cases of some of the larger forms only,^ but the presence of such envelopes may be inferred for most bacteria by their behavior during '*A m *■ t ♦ 1 # •*' Fig. 2. — ^Bacterial Capsules, plasmolysis, where definite retrac- tion of the protoplasm from a well-defined cell outline has been repeatedly observed. The occur- rence, furthermore, of so-called "shadow forms" which appear as empty capsules, and of, occasion- ally, a well-outlined cell body, after the vegetative form has en- tirely degenerated in the course of sporulation, make the assump- tion of the presence of a cell membrane appear extremely well founded. Differing from the cell membranes of plant cells, cellulose has not, except in isolated instances, been demonstrable for bacteria, and the membrane is possibly to be regarded rather as a peripheral protoplasmic zone, which remains unstained by the usual manipula- tions. Zettnow,^ who has carefully studied the structure of some of the larger forms, takes the latter view, and regards the " ectoplasmic " zone as a part of the cell protoplasm devoid of nuclear material. Zett- now's opinion is borne out by the greatly increased size of the bacterial cells as seen by means of special stains. Many bacteria have been shown to possess a mucoid or gelatinous envelope or capsule. According to Migula,^ such an envelope is present on all bacteria, though it is in only a few species that it is sufficiently well developed and stable to be easily demonstrable and of differential * Biitschli, loc. cit. ^ Zettnow, loc. cit. 3 Migulaf "Systeme d. Bakterien," 1, p. 56. MORPHOLOGY, REPRODUCTION, ETC. 13 value. When stained, the capsule takes the ordinary anilin dyes less deeply than does the bacterial cell body, and varies greatly in thickness, ranging from a thin, just visible margin to dimensions four or five times exceeding the actual size of the bacterial body itself. This struc- ture is perfectly developed in a limited number of bacteria only in which it then becomes an important aid to identification. Most prominent among such bacteria are Diplococcus pneumoniae. Micrococcus tetra- genus, the bacilli of the Friedlander group, and B. aerogenes capsulatus. The development of the capsule seems to depend intimately upon the environment from which the baqjeria are taken. It is most easily de- monstrable in preparations of bacteria taken directly from animal tis- sues and fluids, or from media containing animal serum or milk. If cultivated for a prolonged period upon artificial media, many otherwise capsulated microorganisms no longer show this characteristic structure. Capsules may be demonstrated on bacteria taken from artificial media most successfully when albuminous substances, such as ascitic fluid or blood serum, are present in the culture media, or when the bacteria are smeared upon cover slip or slide in a drop of beef or other serum.^ Most observers believe that the capsule represents a swelling of the ectoplasmic zone of bacteria. By others it is regarded as an evidence of the formation of a mucoid intercellular substance, some of which remains adherent to the individual bacteria when removed from cultures. It is noticeable, indeed, that some of the capsulated bacteria, especially Streptococcus mucosus and B. mucosus capsulatus, develop such slimy and gelatinous colonies that, when these are touched with a platinum wire, mucoid threads and strings adhere to the loop. Exactly what the significance of the capsules is cannot yet be decided. There is, however, definite reason to believe that there is a direct relation between virulence and capsulation; capsulated bacteria are less easily taken up by phagocytes than are the non-capsulated mem- bers of the same species. Also, as Forges and others have shown, capsulated organisms are not easily amenable to the agglutinating action of immune sera. Many bacteria (plague, anthrax) which are habitu- ally uncapsulated on artificial media acquire capsules within the in- fected animal body. Also in some species (pneumococci), the loss of capsule formation as cultivated on the simpler media is accompanied by a diminution of virulence. Organs of Locomotion. — When suspended in a drop of fluid many bacteria are seen to be actively motile. It is important, however, in * Hiss, Jour. Exp. Med., vi, 1905. 14 BIOLOGY A\NrD TECHNIQUE all cases to distinguish between actual motility and the so-called Brown- ian or molecular movement which takes place whenever small particles are held in suspension in a fluid. Brownian or molecular movement is a phenomenon entirely ex- plained by the physical principles of surface tension, and has absolutely no relation to independent motility. It may be seen when particles of carmine or any other insoluble substance are suspended in water, and consists in a rapid to and fro vacillation during which there is actually no permanent change in position of the moving particle except inas- much as this is influenced by current^in the drop. The true motility of bacteria, on the other hand, is active motion due to impulses originating in the bacteria themselves, where the actual position of the bacterium in the field is permanently changed. The ability to move in this way is, so far as we know, limited almost entirely to the bacilli and spirilla, there being but few instances where members of the coccus group show active motility. In all cases, with the exception of some of the spirochetes, where motility may occasionally be due to an undulating membrane marginally placed along the body, bacterial motility is due to hair-like organs known as flagella. These flagella have rarely been seen during life, and their recognition and study has been made possible only by special staining methods, such as those devised by Loeffler, van Ermengem, Pitt, and others. In such stained preparations, the bacterial cell bodies often appear thicker than when ordinary dyes are used, and the flagella apparently are seen to arise from the thickened ectoplasmic zone. The flagella are long filaments, averaging in thickness from one-tenth to one-thirtieth that of the bacterial body, which often are delicately v/aved and undulating, and, judging from the positions in which they become fixed in preparations, move by a wavy or screw-like motion. In length they are subject to much variation, but are supposed to be generally longer in old than in young cultures. Very short flagella have been described only on nitrosomonas, one of the nitrifying bacteria discovered by Winogradsky.^ As to the finer structures of flagella, little can be made out except that they possess a higher refractive index than the cell body itself, and that they can be stained only with those dyes which bring clearly into view the supposedly true cytoplasm of the cell. Whether they penetrate this cytoplasmic membrane or whether they 1 Winogradsky, Arch, des sci. biologiques, St. Petersburg, 1892, 1, 1 and % MORPHOLOGY, REPRODUCTION, ETC. 15 are a direct continuation of this peripheral zone of the bacterial body, can not be decided. The manner in which bacteria move is naturally subject to some var- iation depending upon the number and position of the flagella possessed by them. Whether bacteria exercise or not the power of motility de- pends to a large extent upon their present or previous environment. They are usually most motile in vigorous young cultures of from twenty- four to forty-eight hours' growth in favorable media. In old cultures motility may be diminished or even inhibited by acid formation or by other deleterious products of the bacterial metabolism. At the optimum growth-temperature motility is most active, and a diminution or increase of the temperature to any considerable degree diminishes or inhibits it. Thus actively motile organisms, in the fluid drop, may be seen to diminish distinctly in activity when left for any prolonged time in a cold room, or when ^^^ 3.-Arrangement op the preparation is chilled. Any influence. Bacterial Flagella. in other words, chemical or physical, which tends to injure or depress physiologically the bacteria in any way, at the same time tends to inhibit their motility. Messea^ has proposed a classification of bacteria which is based upon the arrangement of their organs of motility, as follows: I. Gymnobacteria, possessing no flagella. II. Trichobacteria, with flagella. 1. Monotricha, having a single flagellum at one pole. 2. Lophotricha, having a tuft of flagella at one pole. 3. Amphitricha, with flagella at both poles. 4. Peritricha, with flagella completely surrounding the bac- terial body. Bacterial Spores. — ^A large number of bacteria possesses the power of developing into a sort of encysted or resting stage by a process commonly spoken of as sporulation or spore formation. The formation of spores by bacteria depends largely upon environmental conditions, and the optimum environment for spore formation differs greatly for varioua species. It is usually necessary that a temperature of over 20° C, exist in order that spores may be formed. Unfavorable factors, likcs^ acid formation, accumulation of bacterial products in old cultures, or 1 Me^sea, Cent. f. Bakt., I, Ref. ix, 1891. 16 BIOLOGY AND TECHNIQUE lack of nutrition, frequently seem to constituti^ the stimuli which lead to sporulation. In the case of some species, notably the anthrax bacillus, spores are formed only in the presence of free oxygen and are therefore never formed within the tissues of infected animals. It is claimed that some of the pathogenic anaerobes, like B. tetani and the bacillus of malignant edema, may form spores anaerobically. Nevertheless it has been observed that when an absolute exclusion of oxygen is practiced in the cultivation of these bacteria, vegetative forms only are seen in the cultures.^ The process of sporulation is by no means to be regarded as a method of multiplication, since it rarely occurs that a single bacil- lus produces more than one spore. In some species of bacteria the formation of several spores in one individual has occasionally been observed, but there can be no question about the fact that such a condition is exceptional. Varieties of spores are often recognized, the so-called arthrospores and the true spores or endospores. It is seriously in doubt whether the structures once spoken of as arthrospores should be considered as in any way comparable to true spores. They are represented by the granular and globular appearances occasionally observed in old cultures of some bacteria, notably streptococcus, cholera spirillum, diphtheria bacillus, and others. It was believed that they were due to a transformation of certain individuals of the cultures into more resistant forms. It ia probable, however, that such structures are merely to be regarded as evidences of involution or degeneration, since it has never been demon- strated that cultures containing them are more resistant either to dis- infectants or to heat, than cultures showing no evidences of such forms. The true spores or endospores are most common among bacilli, and are rarely observed among the spherical bacteria. They arise within the body of the individual bacterium as a small granule which probably represents a concentration of the protoplasmic substance. Nakanishi ^ claims that there is a definite relation between these sporogenic globules and the nuclear material of the bacterial cell. At the time at which sporulation occurs there is usually a slight and gradual thickening of the bacillary body. After the formation of this thickening, a spore mem- brane appears about the same thickened area. The completed spore is usually round or oval, has an extremely high refractive index, and a '^Zinsser, Jour. Exp. Med., viii, 1906, p. 542. » Nakanishi, Munch, med. Woch., 1900, p. 680. MORPHOLOGY, REPRODUCTION, ETC, 17 D 0 H? Fig. 4.- Various Positions of Spores IN Bacterial Cell. 0000ken of as endoenzymes- Whenever they are true secretory products, however, they can be obtained separate from the microorganisms which form them by filtration through a Berkef eld candle. From such fihrates they may, in some cases, be obtained in the dry state by precipitation with alcohoL When obtained in this way the precipitated enzyme is usually much more thermostable than when in solution, for idiile soluble enzymes in filtrates are usually destroyed by 70** C, and even less, the dried powder may occasionally withstand 140** C for as long as ten minutes.^ Apart from the general conditions of temperature and moisture, the development of these enzymes seems to depend directly upon the presence of proteids in the culture media. The number of bacterial species which produce proteoiytic enzymes is legion. Among those more com- mordy met with are staphylococci, B. subtilis, B. proteus, B. faecalis Hquefaciens, Spirillum cholerse asiaticae, B. anthracis, B. tetani, B. pyo- cyaneus, and a large number of others. The inability of any given micro- organism to Hquefy gelatin or fibrin by no means entirely excludes the formation by it of proteolytic enzymes, since these ferments may often be active for one particular class of proteid only. In order to study the qualitative and quantitative powers of any given bacterial proteolyzing enzjine or protease, it b, of course, neces- sary to study these processes in pure culture in the test tube with media of known composition. In the refuse heap, in sewage, or in rotting excreta, the process is an extremely compUcated one, for besides the bacteria which attack the proteid molecule itself, there are many other species supplementing these and each other, one species attacking the more or less complex «[id-products left by the action of the others. Exactly what the chemical reactions are which take place in these cleavages is not entirely clear. It is believed, however, that most of the cleavages are of an hydrohtic nature. In general, the action of the proteid-splitting f ermoits is ecMnpanUe to that of the pancreatic ferment trypsin, and they are most aii&a. active in an alkaline enviroimientw They differ, among themselves, chiefly in the form of proteid which they are competent to attack, and in the extent to which they are able to reduce it toward its simple radicles. A distinction is occasionally made between the terms piartfaction and decay, the former being used to rrfer to the decompositiafi taking THE BIOLOGICAL ACTIVITIES OP BACTERIA 45 place under anaerobic conditions, that is, in the absence of oxygen, a process usually resulting in incomplete cleavage of the proteid medium; the latter being used to signify decompositions under aerobic conditions and leading to a more complete splitting, the end-products often being represented by such simple compounds as carbon dioxide, water, and ammonia. In general, the products of putrefaction are largely repre- sented by the amino-acids, leucin and tyrosin, fatty acids, mercaptan, indol, and skatol. The gases generated in such decomposition are largely made up of COj, hydrogen, NH^ and HjS. The coincident presence, furthermore, of the carbohydrate-splitting bacteria and of denitrifying microorganisms renders the actual process of putrefaction a chaos of many activities in which the end-products and by-products are qualita- tively determinable onlj^ with much inexactitude, and which com- pletely defies any attempt at quantitative analysis. Ptomains. — ^There are certain products, however, resulting from the proteolytic action of bacterial enzjines upon proteids which claim more than a purely chemical interest because of their toxic action upon the animal organism, and their consequent importance as incitants of dis- ease. Pre-eminent among these are the ptomains. The word ptomain (from rrdi/xa, a dead body) is used to designate organic chemical compounds produced by the action of bacteria, which are basic in char- acter; that is, are able to combine with an acid to form a salt. They should be definitely distinguished from the so-called leucomains, a term employed to designate similar substances formed in the course of proteid metabolism within the animal body, and not bacterial in origin. Both in their basic characters and in their nitrogenous constitu- tion, the ptomains resemble the vegetable alkaloids, and for this reason are sometimes spoken of as " animal alkaloids." The ptomains must be sharply distinguished from the bacterial loxins, which are products of the bacterial growth irrespective of the medium in which they are grown, except in so far as this hinders or abets the development of the microorganisms. Thus, toxins may be developed by diphtheria organisms, for instance, in proteid-free media. As will be seen in a subsequent section, the true toxins are comparable to the enzymes themselves, rather than to their cleavage products, rep- resented in this instance by the ptomains. A great number of ptomains are chemically known. Many 5)f these possess little or no toxicity. Others, however, like putrescin (tetramethylenediamin, C4H12N3) and cadaverin (CsH.^Na) are very highly poisonous. It is to one or another of these ptomains that most 46 BIOLOGY AND TECHNIQUE cases of so-called meat poisoning (kreatoxismus) , cheese poisoning (tyrotoxismus) , or vegetable poisoning (sitotoxismus) are due. In each individual case the variety of ptomain resulting from a bac- terial decomposition varies with the individual species of microorganism taking part in the process and with the nature of the proteid upon which its development takes place. In breaking down animal excreta, the task of the bacteria is rather a simpler one than when dealing with the cadavers themselves, for here a part of the cleavage has already been carried out either by the destruc- tive processes accompanying metabolism, or by partial decomposition by bacteria begun within the digestive tract. This material outside of the body is further reduced by bacterial enzymes into still simpler sub- stances, the nitrogen usually being liberated in the form of ammonia. One example of such an ammoniacal fermentation may be found in the case of the urea fermentation by Micrococcus urese, in which the cleavage of the urea takes place by hydrolysis according to the follow- ing formula: (NH2)2 CO + 2H2 O = CO2 + 2NH3 + H2 O Similar ammoniacal fermentations are carried out, though perhaps according to less simple formulae, by a large number of microorganisms. Perhaps the most common species which possesses the power is the group represented by B. proteus vulgaris (Hauser). From what has been said it follows naturally that, so far, the decom- position of the proteid molecule from its complex structure to ammonia or simple ammonia compounds is an indispensably important function, not only for agriculture, but for the maintenance of all life processes. It is clear, on the other hand, that a further decomposition of ammonia compounds into forms too simple to be utilized by the green plants would be a decidedly harmful activity. And yet this is brought about by the so-called denitrifying bacteria which will be considered in a subsequent section. Lab Enzymes. — ^There are a number of ferments produced by bacteria which, although affecting proteids, can not properly be classified with the proteolytic enzymes. These are the so-called coagulases or lab enzymes, which have the power of producing coagulation in liquid pro- teids. Just what the chemical process underlying this coagulation is, is not known. If Hammarsten's ^ conclusions as to the hydrolytic I Hammarsten, "Textbook of Physiol. Chemistry," Translation by Mandel. THE BIOLOGICAL ACTIVITIES OF BACTERIA 47 nature of the changes produced by them are true, these enzymes are brought into close relationship to the proteolyzers, although a coagula- tion can hardly be regarded as a true katabolic process. In milk where the lab-action becomes evident by precipitation of casein, a strict dif- ferentiation must be made between this coagulation and that brought about by acids or alkalies. In the former case, casein is not only pre- cipitated and converted into paracasein, but is actually changed so that when redissolved it is no longer precipitated by lab.' Coagulating enzymes for milk proteids, blood, and other proteid solutions are produced by a large variety of bacteria. They have been observed in cultures of the cholera vibrio, B. prodigiosus, B. pyocyaneus, and several others.^ The lab enzymes are easily destroyed by temperatures of 70° C. and over, and are very susceptible to excessive acidity or alkalinity. Fat-Splitting Enzymes (Lipase). — The fat-splitting powers of bac- teria have been less studied than some of the other bacterial func- tions and are correspondingly more obscure. It is known, nevertheless, that the process is due to an enzyme and that it is probably hydrolytic in nature. The folio wing, formula represents the simplest method in which some of the molds and bacteria produce cleavage of fats into glycerin and fatty acid. C3 H5 (C^ H,„_, 02)3 + 3H2 O == C3 H3 (OH3) + 3C^ H^n O, Glycerin Fatty acid Some of the bacteria endowed with the power of producing lipase are the spirillum of cholera, B. fluorescens liquefaciens, B. prodigiosus, B. pyocyaneus. Staphylococcus pyogenes aureus, and some members of the streptothrix family. The methods of investigating this function of bacteria, originated by Ejkmann, ' consists in covering the bottom of a Petri dish with tallow and pouring over this a thin layer of agar. Upor this, the bacteria are planted. Any diffusion of lipase from the bacterial colonies becomes evident by a formation of white, opaque spots in the tallow. Carriere '* was able to demonstrate a fat-splitting ferment for the tubercle bacillus. Apart from the importance of these enzymes in nature for the destruction of fats, they are industrially important be- » Oppenheimer, " Die Fermente u. ihre Wirkung," Leipzig, 1903. « Tarini, Atti dei laborat. d. sanita, Rome, 1890. » Ejkmann, Cent. f. Bakt., I, xxix, 1901. * Carriere, Comptes rend, de la soc. de biol., 53, 1901. 5 48 BIOLOGY AND TECHNIQUE cause of their action in rendering butter, milk, tallow, and allied prod- ucts rancid, and are medically of interest for their action upon fats in the intestinal canal. Enzymes of Fermentation {The Cleavage of Carbohydrates by Bacterid). — The power to assimilate carbon dioxide from the atmosphere is possessed only by the green plants and some of the colored algae, and the sulphur or Thiobacteria. All other living beings are thus dependent for their supply of carbon upon the synthetic activities carried on by these plants to the same degree in which they are de- pendent upon similar processes for their nitrogen supply. The return of this carbon to the atmosphere is, of course, brought about to a large ex- tent by the respiratory processes of the higher animals. The carbon, which, together with nitrogen, forms a part of proteid combinations, is freed, as we have seen in a previous section, by the processes of proteid cleavage. That, however, which is inclosed in the carbohydrate mole- cule, is set free by the action of yeasts, molds, or bacteria, by an enzy- matic process similar in every respect to that described above for the process of proteid cleavage. Fermentation. — The power of carbohydrate cleavage is possessed by a large number of the yeasts and bacteria. The process, as has been indicated, is of great importance in the cycle of carbon compounds for the return of carbon to its simplest forms, and is, furthermore, as will be seen in a later section, of great utility in the industries. In each case the power to split a particular carbohydrate is a more or less specific characteristic of a given species of microorganism, and for this reason has been extensively used as a method for the biological differen- tiation of bacteria. In the course of much careful work upon this question it has been ascertained that the specific carbohydrate-splitting powers of any given species are constant and unchanged through many generations of artificial cultivation. Thus, differentiation of the Gram-negative bacteria, the members of the pneumococcus-streptococ- cus group, and the diphtheria group, can now largely be made by a study of their sugar fermentations. In most of these cases, as far as we know, the cleavage is produced by a process of hydrolysis. A convenient nomenclature which has been adopted for the designation of these ferments is that which employs the name of the converted carbohydrate adding the suffix " ase " to indicate the enzyme. There are thus ferments known as amylase, cellulase, lac- tase, etc. Amylase (Diastase or Amylolytic Ferment). — Amylases or starch- THE BIOLOGICAL ACTIVITIES OF BACTERIA 49 splitting enzymes are formed by many plants (malt) and by animal organs (pancreas, saliva, liver). Among microorganisms amylase is produced by many of the strep tothrix group, by the spirilla of. Asiatic cholera and of Finkler-Prior, by B. anthracis, and many other bacteria. A large number of the bacteria found in the soil, furthermore, have been shown to produce amylases. By cultivating bacteria upon starch- agar plates, amylase can be readily demonstrated by a clearing of the medium immediately surrounding the colonies.^ Since, of course, there are several varieties of starches, it follows that the exact chemical action of amylase differs in individual cases. The determination of the structural disintegration of starch by these fer- ments is fraught with much difficulty, owing to the polymeric constitu- tion of the starches. Primarily, however, a cleavage takes place into a disaccharid such as maltose (hexobiose) , and the non-reducing sugars and dextrin. Beyond this point, however, the further cleavages are subject to much variation and are not entirely clear. The dextrins upon further reduction yield eventually dextrose. Cellulase. — Cellulose is fermented by a limited number of bacteria, most of them anaerobes. The chemical process by which this takes place is but poorly understood.^ Gelase. — An agar-splitting ferment has been found by Gran.^ Invertase. — The enzymes which hydrolytically cause cleavage of saccharose into dextrose and levulose are numerous. The chemical process takes place according to the following formula: Ci2 H22 On + H2 0 = Cg Hi2 Og + Cg Hi2 Og Saccharose Dextrose Levulooe Invertase is produced by many of the yeasts. It is one of the most common of the enzymes produced by bacteria, and has been found in cultures of B. megatherium, B. subtilis, pneumococcus, some strepto- cocci, B. coli, and many others. Invertase is usually very susceptible to heat, being destroyed by temperatures of 70° C. and over. A slightly acid reaction of media abets the inverting action of these enzymes. Strong acids and alkalies inhibit them. Inverting enzymes may be precipitated out of solution by alcohol. Antiseptics even in weak con- centrations will inhibit their action. 1 Ejkmann, Cent. f. Bakt., xxix, 1901, and xxxv, 1904. 2 Omelianski, Lafar's '' Handb. d. techn. Mykologie," Bd. iii, Chap. 9. » Gran, Bergens Museum Aarbog, 1902, Hft. I. 50 BIOLOGY AND TECHNIQUE Lactase. — Lactose-splitting ferments are extremely common both among bacteria and among the yeasts. The process- is here again a hydrolytic cleavage resulting in the formation of the monosaccharids as dextrose and galactose. Maltase. — ^A maltose-splitting ferment has also been found in the cultures of many bacteria, leading to the formation of dextrose. Lactic Acid Fermentation. — Lactic acid (oxyproprionic acid, CgHg O3) is one of the most common substances to appear among the prod- ucts of bacterial activity, both in media containing carbohydrates and in those consisting entirely of albuminous substances. In most of these cases, the lactic acid is formed merely as a by-product accom- panying many other more complicated chemical cleavages. In some instances, however, lactic acid is produced from carbohydrates, both disaccharids and monosaccharids, as an almost pure product due to a specific bio-chemical process. The reactions taking place in this phenom- enon may be briefly expressed according to the following formulae: C,,R,,0,, + R,0= 4C3He03 Lactose Lactic acid or Cq H12 Og = 2C3 Hg O3 Dextrose Lactic acid In the same way lactic acid may be produced by bacteria from levu- lose. Examples of lactic acid formation are furnished by the streptococcus lacticus, and B. lactis aerogenes. In the case of the former, the fer- mentation may indeed proceed by the simple chemical process indi- cated in the formulae, since the action of the bacillus is entirely unac- companied by the evolution of gas. Numerous other bacteria produce large amounts of lactic acid from lactose, possibly by chemical processes less simply formulated. Among these are bacilli of the colon group, B. prodigiosus, B. proteus vulgaris, and many others. Although lactic acid is usually the chief product in the bacterial fermentation of the simpler carbohydrates, acetic, formic, and butyric acids may often be found as by-products in variable amounts.^ Oxydases {Oxydizing Enzymes). — The most common example of oxidation by means of bacterial ferments is the production of acetic acid * Buchner und Meisenheimer, Ber. d Deut. chem. Gesellsch., xxxvi, 1903. THE BIOLOGICAL ACTIVITIES OF BACTERIA 51 from weak solutions of ethyl alcohol. This process, which is the basis of vinegar production, is universally carried out by bacterial ferments. While possessed to some extent by a considerable number of microorgan- isms, acetic acid formation is a function pre-eminently of the bacterial groups described by Hansen, including ''Bacterium aceti" and "Bac- terium pasteurianum." To these two original groups a number of others have since been added. The organisms are short, plump bacilli, with a tendency to chain- formation, and occasionally showing characteristically swollen centers and many irregular involution forms. In the production of vinegar, as generally practiced by the farmer with cider or wine, these bacteria accumulate on the surface of the fluid as a pellicle or scum which is popularly known as the ''mother of vinegar." Destruction of these bacteria by disinfectants or by sterilization with heat promptly arrests the process of vinegar formation. Chemically, the conversion of the alcohol consists in a double oxidation through ethyl aldehyde into acetic acid as shown in the following formulae: 1. C2H5 (OH) + O = CH3 (COH) Alcohol Ethyl aldehyde 2. CH3 (COH) + 0 = CH3 (COOH) Acetic acid Alcoholic Fermentation {Zymase). — The formation of alcohol as an end product of fermentation is of great importance in a number of the industries, primarily in the production of wine and beer. AVhile accom- plished by a number of bacteria, thi^ form of fermentation is carried out chiefly by the yeasts. Expressed in formulae the simplest varieties of alcoholic fermenta- tion, from mono- and disaccharids, may be represented as follows: C«H,A=2C2H3 (OH) +2CO2 Dextrose Ethyl alcohol or C^H^^O,, + H2O = 4C2H3(OH) + 4CO2 Saccharose Ethyl alcohol In all cases the process may not be so simple as indicated by the equa- tions, since by-products, such as higher alcohols, glycerin, succinic and acetic acids, may often be found in small traces among the end- products of such fermentations. The conditions which favor alcoholic 52 BIOLOGY AND TECHNIQUE fermentation by the yeasts are extremely important, since, upon obser- servance of these, depends much of the uniformity of result which is so desirable in the industries mentioned above. The optimum concentra- tion of sugar for the production of the highest quantity of alcohol is at or about 25 per cent. The temperature favoring the process ranges about 30° C. Under such conditions fermentation may continue until the alcohol forms almost a 20-per-cent solution. Most of the fermenta- tions important in the wine, beer, and spirit industries, take place under anaerobic conditions, since the carbon dioxide which is formed soon shuts out any excess of air. In the industrial employment of yeasts for fermentative purposes, it is necessary to work with specific strains, and in scientifically conducted vineyards, breweries, and distilleries the study and pure cultivation of the yeasts form no unimportant part of the work. Certain races of yeasts are more uniform in their fermentative powers than others, and the by- products formed by some races differ sufficiently from those of other races to cause material differences in the resulting substances. In the wine industries, the yeasts differ much from one another according to climatic and other environmental conditions. In vineyards, natural inoculation of the grapes occurs by transportation of the yeast from the soil to the surface of the grapes by wasps, bees, or other insects, through whose alimentary canals the microorganisms pass uninjured. In the autumn the yeast is returned to the soil by falling berries and remains alive in the upper layers of the ground throughout the winter months. In actual practice this natural yeast inoculation is not de- pended upon, but pure cultures of artificially cultivated yeasts are employed for inoculation. In some of the wine-growing countries these are supplied by special government experiment stations. Denitrifying Bacteria. — Nitrogen is most readily absorbed by plants in the form of nitrates. These are furnished to the soil chiefly by the proteid decomposition induced by the proteolytic bacterial enzymes. It is self-evident, therefore, that any cleavage which reduces nitrog- enous matter beyond the stage of nitrates, to nitrites and ammonia, detracts from the value of the nitrogen as a food stuff for plants, and the eventual setting free of nitrogen in the elementary state ren- ders it entirely valueless for any but the leguminous plants. Nevertheless, this process of nitrogen waste or denitrification is constantly going on in nature. In the course of ordinary decomposition, there is a constant reduction of nitrogenous matter to nitrites and salts of ammonia, actively taken part in by a host of bacteria, as many as THE BIOLOGICAL ACTIVITIES OF BACTERIA 53 85 out of 109 investigated by Maassen ^ being found to possess this power. This, however, is not nearly so harmful a source of nitrogen waste as the process technically spoken of as true denitrification, in which nitrates are reduced, through nitric and nitrous oxides, to elementary nitrogen. This phenomenon, more widely spread among bacteria than at first believed, depends essentially upon simple oxygen extraction from the nitrates by the bacteria, and for this reason goes on most actively when the supply of atmospheric oxygen is low. The first bacteria described' as possessing this power of denitrification were the so-called B. denitri- ficans I and II, the first an obligatory anaerobe, the other a facultative aerobe. Since then numerous other bacteria, among them B. coli and B. pyocyaneus, have been shown to exhibit similar activities. It is important agriculturally, therefore, to know that many species which are able to utilize atmospheric oxygen when supplied with it, will get their oxygen by the reduction of nitrates and nitrites when free oxygen is withheld. It is thus clear that a loss of nitrogen is much more apt to proceed rapidly in manure heaps which are piled high and poorly aerated. There are other factors, however, in regard to the physi- ology of these microorganisms, which must be considered for practical purposes. In order that these bacteria may develop their denitrifying powers to the best advantage, it is necessary to supply them with some carbon compound which is easily absorbed by them. This, in decomposing material, is furnished by the products of the carbohydrate cleavage going on side by side with the proteolytic processes. It is still more or less an open question whether the facilitation of denitrification brought about in manure heaps by the presence of hay and straw is due to the carbon furnished by these materials, or whether it is due to the fact that bacilli of this group are apt to adhere to the straw which acts in that case as a means of inoculation. The actual danger of nitrogen depletion of the soil by denitrifying processes is probably much less threatening than was formerly supposed; for, in the first place, the conditions for complete denitrification are much more perfect in the experiment than they ever can be in nature, and the nitrifying processes going on side by side with denitrification make up for much of the loss sustained. Mm99en, Arb, a, d. kais. Gesundheitsamt, 1, xxviii, 1901, 54 BIOLOGY AND TECHNIQUE ANABOLIC OR SYNTHETIC ACTIVITIES OF BACTERIA Nitrogen Fixation by Bacteria. — The constant withdrawal of nitroge- nous substances from the soil by innumerable plants would soon lead to total depletion were it not for certain forces continually at work re- plenishing the supply out of the large store of free nitrogen in the atmos- phere. This important function of returning nitrogen to the soil in suitable form for consumption by the plants is performed largely by bacteria. It is well known that specimens of agricultural soil when allowed to stand for any length of time without further interference will increase in nitrogenous content, but that similar specimens, if sterilized, will show no such increase.^ The obvious conclusion to be drawn from this phenomenon is that some living factor in the unsterilized soil has aided in increasing the nitrogen supply. Light was thrown upon this problem when Winogradsky,^ in 1893, discovered a microorganism in soil which possessed the power of assimilating large quantities of nitrogen from the air. This bacterium, which he named " Clostridium Pasteurianum," is an obligatory anaerobe which in nature always occurs in symbiosis with two other facultatively anaerobic microorganisms. In sym- biosis with these, it can be cultivated under aerobic conditions and thus grows readily in the upper well-aerated la3^ers of the soil. Although, until now, no other bacteria with equally well-developed nitrogen-fixing powers have been discovered, yet it is more than likely that Clostridium Pasteurianum is not the only microorganism endowed with this function. In fact, Penicillium glaucum and Aspergillus niger, two molds, and two other bacteria described by Winogradsky, have been shown to possess this power slightly, but in an incomparably less marked degree than Clostridium Pasteurianum.^ According to the calculations of Sachse,* unsterilized soil may, under experimental conditions, gain as much as 25 milligrams of nitrogen in a season, a statement which permits the calculation of a gain of twelve kilograms of nitrogen per acre annually.^ It is very unlikely, however, that such gains actually occur in nature, where nitrogen-fixation and nitrogen-loss usually occur side by side. 1 Berthelot, Compt. rend, de la soc. de bioL, cxvi, 1893. 2 Winogradsky, Compt. rend, de la soc. de biol., cxvi, 1893, ibid., t. cxviii, 1894. 3 Tacke, Landwirtsch. Jahresber., xviii, 1889. a^ac/ise, "Agr. Chem.," 1883. « Pfeffer, Pflugers Physiologie, p. 395. THE BIOLOGICAL ACTIVITIES OF BACTERIA 55 Agriculturally of even greater importance than the free nitrogen- fixing bacteria of the soil are the bacteria found in the root tubercles of a class of plants known as "leguminosse." It has long been known that this class of plants, including clover, peas, beans, vetch, etc., not only does not withdraw nitrogen from the soil, but rather tends to enrich it. Upon this knowledge has depended the well-known method of alternation of crops employed by farmers the world over. The actual reason for the beneficial influence of the leguminosae, however, was not known until 1887, when Hellriegel and Wilfarth ^ succeeded in demonstrating that the nitrogen-accumulation was directly related to the root tubercles of the plants, and to the bacteria contained within them. These tubercles, which are extremely numerous — as many as a thousand sometimes occurring upon one and the same plant — are formed by the infection of the roots with bacteria which probably enter through the delicate root-hairs. They Vjary in size, are usually situated near the main root-stem, and, in appearance, are not unlike fungus growths. Their development is in many respects comparable to the develop- ment of inflammatory granulations in animals after infection, inas- much as the formation of the tubercle is largely due to a reactionary hyperplasia of the plant tissues themselves. They appear upon the seedlings within the first few weeks of their growth as small pink nodules, and enlarge rapidly as the plant grows. At the same time, later in the season, when the plants bear fruit, the root tubercles begin to shrink and crack. When the crops are harvested, the tubercles with the root remain, rot in the ground, and re-infect the soil. Histologically the tubercles are seen to consist of large root cells which are densely crowded with microorganisms. The microorganism itself, "Bacillus radicicola," was first observed within the tubercles by Woronin ^ in 1866. The bacilH are large, slender, and actively motile during the early development of the tubercles, but in the later stages assume a number of characteristic involution forms, commonly spoken of as "bacteroids." They become swollen, T and Y shaped, or branching and threadlike. Their isolation from the root tubercles usually presents Httle difficulty, since they grow readily upon gelatin and agar under strictly aerobic conditions. On the artificial media the bacillary form is usually well retained, involution forms appearing only upon old cultures. 1 HeUriegel und WilfaHh, Cent. f. Bakt., 1887. 2 Woronin, Bot. Zeit., xxiv, 1866. 56 BIOLOGY AND TECHNIQUE The classical experiments of Hellriegel and Wilfarth conclusively demonstrated the important relation of these tubercle-bacteria to nitro- gen assimilation by the leguminosae. These observers cultivated various members of this group of plants upon nitrogen-free soil — sand — and prevented the formation of root tubercles in some, by sterilization of the sand, while in others they encouraged tubercle formation by inoculation. An example of their results may be given as follows: ^ Lupinus luteus was cultivated upon sterilized sand. Some of the pots were inoculated with B. radicicola, others were kept sterile. Com- parative analyses were made of the plants grown in the different pots with the following striking result: Root tubercles present No root tubercles , IVllJ.^ 1COU.XU • N . added in seed, Harvested soil, and soil- Gain or dry weight N. present extract loss of N. (a) 38.919 .998 .022 + .975 (b) 33.755 .981 .023 -f.958 (c) 0.989 .016 .020 — .004 (d) 0.828 .011 .022 — .009 The great importance of this process in agriculture is demonstrated, furthermore, by a comparison made by the same observers between a legume, the pea, and one of the common nitrogen-consuming crops, oats.^ Nitrogen contents Nitrogen contents of seed and soil. of crop. Gain or loss. Oats 0.027 grams 0.007 grains —.020 Peas 0.038 " 0.459 " +.421 Exactly what the process is by which the bacteria supply nitrogen to the plant is as yet uncertain. Although the degenerating bacteroids in old nodules are bodily absorbed by the plant, this can not be con- ceived as the only method of supply, since the total nitrogen gain many times exceeds the total weight of bacteria in the nodules. It is probable that the microorganisms during life take up atmospheric nitrogen and secrete a nitrogenous substance which is absorbed by the plant cells. Although formerly the relationship between plant and bacterium was regarded as one of symbiosis and of mutual benefit, the opinions as to this subject show wide divergence. While, according to some authors, the entrance of the bacteria into the plants is regarded as a true in- fection against which the plant offers at first a determined opposition as evidenced by tissue reactions, other observers, notably A. Fischer, regard 1 Pfeffer, " Planzenphysiologie," Leipzig, 1897. 2 Hellriegel und Wilfarth, Zeit. d. Ver. f. d. Riibenzucker Industrie, 1888. Quote/ from Fischer, " Vorles. uber die Bakt.," Jena, 1903 THE BIQL.OGICAL ACTIVITIES OF BACTERIA 57 the plant as a parasite upon the bacteria, in that it derives the sole benefit from the relationship and eventually bodily consumes its host. Nitrif3ring Bacteria. — A process diametrically opposed in its chem- istry to denitrification and reduction is that which brings about an oxidation of ammonia to nitrites and nitrates. The actual increase of nitrates in soil allowed to stand for any length of time and examined from time to time has been a well-established fact for many years; but it was believed until a comparatively short time ago that this increase was due to a simple chemical oxidation of ammonia by atmospheric oxy- gen. The dependence of nitrification upon the presence of living organ- isms was finally proved by Muntz and Schlossing ^ in 1887, who demon- strated that nitrification was abruptly stopped when the soil was sterilized by heat or antiseptics. It remained, however, to isolate and identify the organisms which brought about this ammonia oxidation. This last step in our knowledge of nitrification was taken in 1890, by Winogradsky. Winogradsky ^ found that the failures experienced by others who had attempted to isolate nitrifying bacteria were due to the fact that they had used the common culture media largely made up of organic substances. By using culture media containing no organic matter Winogradsky succeeded in isolating free from the soil, bacteria which have since that time been confirmed as being the causative factors in nitrification. During his first experiments this author observed that in some of his cultures the oxidation of ammonia went only as far as the stage of nitrite formation, while in others complete oxidation to nitrates took place. Following the clews indicated by this discrepancy, he finally succeeded in demonstrating that nitrification is a double process in which two entirely different varieties of microorganisms take part, the one capable of oxidizing ammonia to nitrites, the other continuing the process and converting the nitrites to nitrates. The nitrite-forming bacteria discovered by Winogradsky, and named Nitromonas or Nitro- somonas, are easily cultivated upon aqueous solutions containing am- monia, potassium sulphate, and magnesium carbonate. According to their discoverer they develop within a week in this medium as a gelat- inous sediment. After further growth this sediment seems to break up and the bacteria appear as oval bodies, which swim actively about and develop flagella at one end. Upon the solid media in ordinary use they can not be cultivated. Special solid media suitable for their cul- » Muntz und Schlossing, Compt. rend, de I'acad. des sciences, 1887* 2 Winogradsky, Ann. Past. Inst., iv and v, 1890, 1891. 58 BIOLOGY AND TECHNIQUE tivation and composed of silicic acid and inorganic salts have been described by Winogradsky and by Omeliansky/ Other nitrite-forming bacteria have since been described by various observers, all of them more or less limited to definite localities. Some of these are similar to nitrosomonas in that they exhibit the flagellated^ actively motile stage. In others this stage is absent. The nitrite-forming bacteria, apart from their great agricultural im- portance, claim our attention because of their unique position in rela- tion to the animal and vegetable kingdoms. Extremely sensitive to the presence of organic compounds, they are able to grow and develop only upon media containing nothing but inorganic material; and this entirely without the aid of any substances comparable to the chlorophyll of the green plants. The source of energy from which this particular class of bacteria derive the power of building up organic compounds from simple substances is to some extent a mystery. The carbon which they unquestionably require for the building up of organic mate- rial may be, as Winogradsky believed, derived to a certain extent from ammonium carbonate. But it is also quite certain that they are capable of utilizing directly atmospheric CO 2. In the absence of chlorophyll or of any highly organized chemical compound, it seems likely that the energy necessary for the utilization of the carbon obtained in this simple form is derived from the oxidation of ammonia during the proc- ess of nitrification. The conversion of nitrites into nitrates is carried on by other species of bacteria also discovered by Winogradsky. These bacteria are much more generally distributed than nitrosomonas and probably include a number of varieties. The organism described by Winogradsky is an extreinely small bacillus with pointed ends. Capsules have occasionally been made out. It may be cultivated upon aqueous solutions con- taining: Sod. nitrite 1 per cent. Potass, phosphate : 05 " " Magnesium sulph 03 " " Sodium carbonate 1 " " Ferrous sulphate 04 " " The development of the organism is slow and sparse, and is directly inhibited by the presence of organic matter. It is strongly inhibited by the presence of ammonia. The Liberation of Energy by Bacteria.— Like all other living beings, 1 Omeliansky, Cent. f. Bakt., II, 5, 1899. THE BIOLOGICAL ACTIVITIES OP BACTERIA 59 bacteria in their metabolic processes liberate energy. It has been shown by several observers that slight quantities of heat are given off from actively growing cultures. The functions, furthermore, of reproduction, motility, and enzyme formation may be looked upon as forms of energy liberation. In addition to this, certain bacteria have been observed which may liberate energy in the form of light. Light Production by Bacteria. — ^The production of light by bacteria is a power possessed chiefly by certain species inhabiting salt water. Thus, much of the phosphorescence observed at sea, though more fre- quently due to Medusa and other invertebrate animals, is caused by these bacteria. Numerous species which produce this phenomenon have been isolated, too many, and too unimportant, to be individually described. All of them are aerobes and require highly complex food stuffs. They are closely allied to the putrefactive bacteria, and in the sea are usually found upon rotting animal matter.^ The production of light seems directly dependent upon the free access of oxygen, since no light appears under anaerobic conditions. Their luminous quality, moreover, is not a true phosphorescence, in that it does not depend upon previous illumination and develops as well in cultures kept in the dark as in those which have been exposed to light. ^ The Formation of Pigment by Bacteria {Chromobacteria) . — A large number of bacteria, when cultivated upon suitable media, give rise to characteristic colors which are valuable as marks of differentiation. For each species, the color is usually constant, depending, to a certain extent, upon the conditions of cultivation. In only a few of the pigmented bacteria is the pigment contained within the cell body, and in only one variety, the sulphur bacteria, does the pigment appear to hold any distinct relationship to nutrition. In most cases, the coloring matter is found to be deposited in small intercellular granules or globules. The absence of any relationship of the pigment to sunlight, as is the case with the chlorophyll of the green plants, is indicated by the fact that most of the chromobacteria thrive and produce pigment equally well in the dark as they do in the presence of light. Among the most common of the pigment bacteria met with in bacteriological work are Staphy- lococcus pyogenes aureus, Bacillus pyocyaneus. Bacillus prodigiosus, and some of the green fluorescent bacteria frequently found in feces. The chemical nature of these pigments has been investigated quite thoroughly and it has been shown that they vary in composition. 1 Pfliiger's Arch. f. Phys., xi, 1875. 2 Fisclwr, Cent. f. Bakt., iii, 1888. ^ BIOLOGY AND TECHNlQOT Some of the pigments, like that of Staphylococcus aureus, are probably non-proteid and of a fatty nature.* They are insoluble in water but soluble in alcohol, ether, and chloroform. Because of their probable composition, they have been spoken of as ''lipochromes." Other pigments, like the pyocyanin, which lends the green color to cultures of Bacillus pyocyaneus, are water soluble and are probably of proteid composition. Pyocyanin may be crystallized out of aqueous solu- tion in the form of fine needles. The crystals may be redissolved in chloroform. Aqueous solutions retain their color. Solutions in chloro- form, however, are changed gradually to yellow. The power of pigment production of various bacteria depends in each case upon cultural conditions. In most cases, this simply signifies that pigment is produced only when the microorganism, finding the most favorable environmental conditions, is enabled to develop all its func- tions to their fullest extent. Thus, a too high acidity or alkalinity of the culture medium may inhibit pigment formation. Oxygen is neces- sary for the production of color in some bacteria, since the bacteria them- selves often produce the pigment only as a leuko-body which is then oxydized into the pigment proper. A notable example of this is the pig- ment of B. pyocyaneus. In other cases, temperature plays an impor- tant role in influencing color production. Thus, Bacillus prodigiosus refuses to produce its pigment when growing in the incubator. By persistent cultivation in an unfavorable environment, colored cultures may lose their power of pigment production. Sulphur Bacteria. — Wherever the decomposition of organic matter gives rise to the formation of Hj S, in cess-pools, in ditches, at the bottom of the sea, and in stagnant ponds, there is found a curiously interesting group of microorganisms, the so-called sulphur or thiobacteria. Red, purple, and colorless, these bacteria all possess the power of utilizing sulphuretted hydrogen and by its oxidation into free sulphur obtain the energy necessary for their metabolic processes. The colorless sul- phur bacteria, the Beggiatoa and Thiothrices, usually appear as threads or chains which, in media containing sufficient H3 S, are usually well- stocked with minute globules of sulphur. If found upon decomposing organic matter, they often cover this as a grayish mold-like layer. The red sulphur bacteria, of which numerous species have been described by Winogradsky, may appear as actively motile spirilla (Thiospirillum) or as short, thick bacillary forms. » Schroeter. Gent. f. Bakt.. xviii. 1895. THE BIOLOGICAL ACTIVITIES OF BACTERIA 61 The physiology of all the sulphur bacteria, and especially of the colored varieties, is of the greatest interest in that these microorganisms are among the few members of the bacterial group which behave meta- bolically like the green plants. The higher organic substances play lit- tle or no part in the nutrition of these microorganisms. Strictly aerobic, the colorless thiobacteria are independent of sunlight, while the red and purple varieties exhibit their physiological dependence upon light by accumulating under natural conditions in well-lighted spots. Both varieties possess equally the power of oxidizing sulphuretted hydrogen as a source of energy. The sulphur is then stored as elemental sulphur within the bacterial body and when a lack of food stuffs sets in, the store of sulphur can be further oxidized into sulphurous or sulphuric anhydrides. With this sole source of energy, these bacteria are capable of flourishing aerobic ally, while an absence of HjS, even in the presence of organic food stuffs, leads to a rapid disappearance of their sulphur contents and an inability to develop. In the case of the colored thiobacteria, the red pigment appears to fulfil, to some extent, a function comparable to that of the chlorophyll of the green plants. Engelmann,* who has studied this pigment spectroscopically, has found that besides absorbing the red spectral rays there is an absorption of rays on the ultra-red end of the spectrum. The absorption of the red rays between the lines B and C of the spectrum, and of violet rays at the line F, is the same as that of the absorption spectrum of chlorophyll, and it is in the zone of these rays that the physiological effects of chlorophyll are most active. In addition to these absorp- tion bands, the bacteriopurpurin of the red sulphur bacteria shows absorption of the invisible ultra red rays of the spectrum. Engelmann, with a microspectroscope, projected a spectrum into a miscroscopic field in which green algae or, in the case under discussion, red sulphur bacteria had been placed. Other sources of light were, of course, excluded. By adding emulsions of strictly aerobic bacteria to such preparations, an accumulation of microorganisms was observed at those points in the spectrum at which most oxygen was liberated. In the case both of chlorophyll and of the red sulphur bacteria such areas of bacterial accumulation (in oxygen Hberation) occurred in the zones of the absorption bands mentioned above. Engelmann, Bot. Zeit., 1888. CHAPTER V THE DESTRUCTION OF BACTERIA GENERAL CONSIDERATIONS No branch of bacteriology has been more fruitful in practical appli- cation than that which deals with the factors which bring about the destruction of microorganisms. Upon the study of this branch has depended the growth and the development of modem surgery. The agents which affect bacteria injuriously are many, and are both physical and chemical in nature. When a procedure completely destroys bacterial life it is spoken of as sterilization or disinfection, the term disinfection being employed more especially to designate the use of chemical agents. When the procedure destroys vegetative forms only, leaving the more resistant spores un- injured, it is spoken of as "incomplete sterilization." When an agent, on the other hand, does not actually kill the microorganisms, but merely inhibits their growth and multiphcation, it is spoken of as an antiseptic. The term deodorant is indiscriminately applied to substances which mask or destroy offensive odors, and may or may not possess disinfectant or antiseptic value. Some deodorants act chemically on the noxious gases, destroying them. PHYSICAL AGENTS INJURIOUS TO BACTERIA The principal physical agents which may exert deleterious action upon bacteria are: drying, light, electricity, and heat. Drying. — Complete desiccation eventually destroys most of the path- ogenic bacteria, yet great differences in resistance to this condition are shown by various microorganisms. Ficker,^ who has made a systematic study of the influence of complete drying upon bacteria, concludes that the resistance of bacteria to desiccation is influenced by the age of the culture investigated, the rapidity with which the withdrawal of moisture 1 Ficker, Zeit. f. Hyg., xxix, 1896. 62 THE DESTRUCTION OF BACTERIA 63 is accomplished, and the temperature at which the process takes place. Microorganisms like the gonococcus and the Pfeiffer bacillus, are destroyed by drying within a few hours. The cholera vibrio dried upon a coverslip was found by Koch ^ to be killed within four hours; by Burck- holtz,^ to survive about twenty-four hours.' The spore-forms of bacteria are infinitely more resistant to this influence than are the vegetative forms, though they may be destroyed by rapid and complete drying in a desiccator. It is self-evident that many discrepancies in the experimental results of various authors may depend upon the technique of investiga- , tion, since the degree of drying attained depends intimately upon the thickness and consistence of the material investigated, and upon the methods employed for desiccation. Light. — Direct sunlight is a powerful germicide for all bacteria except a limited number of species like the thio-or sulphur bacteria, which utilize sunlight for their metabolic processes as do the green plants. Koch ^ has shown that exposure to sunlight will destroy the tubercle bacillus within two hours or less, the time depending upon the thick- ness of the exposed layers and the material surrounding the bacilli. Confirmatory researches have been published by Mignesco * and others. The powerful disinfecting influence of sunlight upon bacteria suspended in water has been shown by Buchner.^ Observations in regard to the influence of sunlight upon anthrax spores have been made by Arloing,^ and similar observations upon a number of other microorganisms have been carried out by Dieudonne, Janowski, v. Esmarch, and many others. All these observers, while differing somewhat as to the time necessary for bacterial destruction^ agree in finding definite and pow- erful bactericidal action of sunlight. Diffuse light, of course, is less active than direct sunlight. According to Buchner, typhoid bacilli are inhibited by direct sunhght in one and one-half hours, by diffuse light in five hours. A remarkable statement is made by Arloing, who claims to have found that anthrax spores are more quickly destroyed by direct sunlight than are the vegetative cells. This fact would call for further confirmation. » Koch, Arb. a. d. kais. Gesundheitsamt, iii, 1887. 2 Burkholtz, Arb. a. d. kais. Gesundheitsamt, v, 1889. 3 Koch, X Intemat. Med. Congress, Berlin, 1890. * Mignesco, Arch. f. Hyg., xxv, 1896. 6 Buchner, Cent. f. Bakt., I, xi, 1892. « Arloing, Compt. rend, de I'acad. d. scL, c, 1885, 64 BIOLOGY AND TECHNIQUE It has been shown by various authors that the influence of sunlight is not to be attributed in anyway to temperature, nor always to a direct action of the light upon the bacteria, but depends largely upon photo- chemical changes produced by the light rays in the media. Richardson ^ and Dieudonne ^ conclude that under ordinary aerobic conditions in fluid environment peroxide of hydrogen is formed under the influence of light. Novy and Freer ^ believe that the bactericidal efi^ects in fluids noticed as a result of exposure to light are too strong to be explained by the formation of small quantities of peroxide of hydrogen, and attribute this action to organic peroxides formed under the described conditions, such as the peroxides of diacetyl, benzoylacetyl, and others. These views are somewhat strengthened by the fact that exclusion of oxygen from media markedly diminishes the bactericidal power of light.^ That the photochemical changes alone, however, do not explain this action follows from the fact that dried bacteria, not surrounded by media, are subject to a similar action.^ In analyzing sunlight in regard to its bactericidal power, it has been found by various observers that the most powerful action is exerted by the ultraviolet spectral rays, whereas the yellow, red, and ultra-red rays are practically innocuous.^ It is of importance to note that sunlight has been found also to have a strong attenuating influence ^ upon some bacterial poisons, as shown by the experiments of Ferri and Celli upon tetanus toxin. Electric light exerts a distinct bactericidal action when applied in strengths of 800 to 900 candle power for seven or eight hours.^ Rontgen or a:-rays are said by Zeit,^ Blaise ^^ and Sambac, and others to be without appreciable geritiicidal power. Rieder,^^ on the other hand, has reported definite inhibition of bacterial growth after exposures of half an hour to a;-rays. 1 Richardson, Jour. Chem. Soc, i, 1893, Ref. Deut. chem. Gesells,, xxvi. 2 Dievdonne, loc. cit. 3 Novy and Freer, 3d Ann. Meeting Assn. Amer. Bacteriologists, Chicago, 1901. * Roux, Ann. Inst. Past., ix, 1887. '^ Dieudonn6, loc. cit. 8 Ward, Proc. Royal Soc, 52, 1893. .7 Fern and Celli, Cent. f. Bakt., I, xii, 1892. 8 Dieudonne, loc. cit. ^Zeit, Jour. Amer. Med. Assn., xxxvii, 1901. 1° Blaise and Sambac, Compt. rend, de la soc. de biol., 1896. " Rieder, Miinch. med. Woch., 1898. THE DESTRUCTION OF BACTERIA 65 Radium rays have a distinct inhibitory and even bactericidal power when applied at distances of a few centimeters for several hours. ^ Electricity. — If we exclude the indirect actions of heat and electro- lysis, it can hardly be said that the direct bactericidal action of electric currents has been satisfactorily demonstrated. Such action, however, has been claimed by d'Arsonville and Charrin,^ and by Spilker and Gottstein.3 Heat. — ^The most widely applicable and efficient physical agent for sterilization is heat. The dependence of bacteria for growth and vitality upon the main- tenance of a proper temperature in their environment, and the ranges of variation within which bacteria may thrive, have been discussed in a preceding section, in which a table of so-called " thermal death points " has been given. In the method of expressing these values it was seen that tv/o elements entered into the destruction of bacteria by heat, namely, that of the degree of temperature which is applied, and that of the time of application. The prolonged application of moderately high temperatures, in other words, may in certain instances, accomplish the same result as the brief use of extremely high ones. In general, the death of bacteria following prolonged exposure to temperatures but slightly exceeding the optimum is due to the inability of the anabolic processes to keep pace with the accelerated katabolic processes, gradual attenuation resulting in death. At somewhat higher temperatures death results from coagulation of the bacterial protoplasm, and at still higher degrees of heat, applied in the dry form, direct burning of the bacteria may be the cause of their destruction.' Heat may be applied in the form of dry heat or as moist heat, these methods being of great practical value, but differently applicable ac- cording to the nature of the materials to be sterilized. The two methods, moreover, show a marked difference in efficiency, temperature for tem- perature. For the recognition of this fact we are largely indebted to the early researches of Koch and Wolff hiigel," and of Koch, Gaffky, and Loeffler.^ 1 Personal observations, 2 D'Arsonville and Charrin, Compt. rend, de la soc. de biol. 3 Sjyilker and Gottstein, Cent. f. Bakt., I, 9, 1891. * Koch und Wolff hiigel, Mitt. a. d. kais. Gesundheitsamt, 1, 1882. ^ Koch, Gaffky and Loeffler, ibid. 66 BIOLOGY AND TECHNIQUE These observers were able to show that the spores of anthrax were destroyed by boihng water at 100° C. in from one to twelve minutes, whereas dry hot air was efficient only after three hours' exposure to 140° C. Extensive confirmation of these differences has been brought by many workers. An explanation of the phenomena observed is probably to be found in the changes in the coagulability of proteids brought about in them by the abstraction of water. Lewith/ working with various proteids, found that these sub- stances are coagula+ed by heat at lower temperatures when they contain abundant quantities of water, than when water has been abstracted from them On the basis of actual experiment with egg albumin he obtained the following results,^ which illustrate the point in question: Egg albumin in dilute aqueous solution, coagulated at 56° C. " . " with 25 per cent water, " " 74-80° C. u u u jg a u u u u 80-90° C. « it li o a a ti tc {( 14K0 ri Absolutely anhydrous albumin, according to Haas,^ may be heated to 170° C. without coagulation. It is thus clear that bacteria exposed to hot air may be considerably dehydrated before the temperature rises sufficiently to cause death by coagulation, complete dehydration neces- sitating their destruction possibly by actual burning. . Bacteria exposed to moist air or steam, on the other hand, may ab- sorb water and become proportionately more coagulable. The same principle, as Lewith points out, probably explains the great resistance to heat observed in the case of the highly concentrated pro- toplasm of spores. Apart from the actually greater efficiency of moist heat when com- pared with dry heat of an equal temperature, an advantage of great practical significance possessed by moist heat lies in its greater powers of penetration. An experiment carried out by Koch and his associates illustrates this point clearly. Small packages of garden soil were sur- rounded by varying thicknesses of linen with thermometers so placed that the temperature under a definite number of layers could be deter- 1 Lewith, Arch. f. exp. Path. u. Pharm., xxvi, 1890. ^Lewith, loc. cit., p. 351. 8 Haas, Prag. med. Woch., 34-36, 1876, THE DESTRUCTION OF BACTERIA 67 mined. Exposures to hot air and to steam were then made for com- parison, and the results were as tabulated:* Tempera- tures. Time of Application. Temperatures Reached within Thicknesses of Linen. Twenty Thicknesses. Forty Thicknesses. One Hundred Thicknesses. Hot air 130-140° C. 4 hours. 86° 72° Below 70° Incomplete steriliza- tion. Steam 90-105.3° 3 hours. 101° 101° 101.5° Complete steriliza- tion. This great penetrating power of steam is due presumably to its com- paratively low specific gravity which enables it to displace air from the interior of porous materials, and also to the fact that as the steam comes in contact with the objects to be disinfected a condensation takes place with the consequent liberation of he'at. When a vapor passes into the liquid state it gives out a definite amount of heat, which in the case of water vapor, at 100° C, amounts to about 537 calories. This brings about a rapid heating of the object in question. Following this process the further heating takes place by conduction, and it is, of course, well known that steam is a much better heat conductor than air.^ Moist heat may be applied as boiling water, in which, of course, the temperature varies little from 100° C, or as steam. Steam may be used as live, flowing steam, without pressure, the temperature of which is more or less constant at 100° C, or still higher efficiency may be attained by the use of steam under pressure, in which, of course, temperatures far Exceeding 100° C. may be produced, according to the amount of pressure which is used. The spores of certain bacteria of the soil which can not be killed m live steam in less than several hours may be destroyed in a few minutes, or even instantaneously, in compressed steam at temperatures ranging from 120° to 140° C." In all methods of steam sterilization, it is of great practical impor- 1 Koch, Gaffky und Loeffler, loc. cit., p. 339. 2 Gruber, Cent. f. Bakt., iii, 1888. 3 Christen, Ref. Cent. f. Bakt., V. xiii, 1893. 68 BIOLOGY AND TECHNIQUE V tance, as v. Esmarch ^ has pointed out, that the steam shall be saturated, that is, shall contain as much vaporized water as its temperature per- mits. Unsaturated, or so-called " super-heated steam " is formed when heat is applied to steam, either by passage through heated piping or over heated metal plates. In such cases the temperature of the steam is raised, but no further water-vapor being supplied, the steam exerts less pressure and contains less water in proportion to its volume than saturated steam of an equal temperature. The super-heated steam, therefore, is heated considerably over its condensation temperature and becomes literally dried. In consequence, its action is more comparable to hot air than to saturated steam, and up to a certain temperature its disinfecting power is actually less than that of live steam at 100° C. V. Esmarch, who has made a thorough study of these conditions, con- cludes that up to 125° C, the eflficiency of superheated steam is lower than that of live steam at 100° C. Above this temperature, of course, it is again active as in the case of ordinary dry heat. Practical Methods of Heat Sterilization. — Burning. — ^For ob- jects without value, actual burning in a furnace is a certain and easily applicable method of sterilization. Flaming, by passage through a Bunsen or an alcohol flame, is the method in use for the sterilization of platinum needles, coverslips, or other small objects which are used in handling bacteria in the laboratory. Hot air sterilization is carried out in the so-called "hot air chambers,'^ simple devices of varied construction. The apparatus most commonly used (Fig. 8) consists of a sheet-iron, double-walled chamber, the joints of which, instead of being soldered, are closed by rivets. The inner case of this chamber is entirely closed except for an opening in the top through which a thermometer may be introduced, while the outer has a large opening at the bottom and two smaller ones at the top. A gas- burner is adjusted under this so as to play directly upon the bottom of the inner case. A thermometer is fitted in the top in such a way that it penetrates into the inner chamber. The air in the chamber is heated directly by the flame and by the hot air, which, rising from the flame, courses upward within the jacket between the two cases and escapes at the top. To insure absolute sterilization of objects in such a chamber, the temperature should be kept between 150° and 160° C. for at least an hour. In sterilizing combustible articles in such a chamber, it should be remembered that cotton is browned at a temperature of 200° C. and 1 V. Esmarch, Zeit. f. Hyg., iv, 1888. THE DESTRUCTION OF BACTERIA 69 over. This method is used in laboratories for the sterilization of Petri dishes, flasks, test tubes, and pipettes, and for articles which may be in- jured by moisture. Both heating and subsequent cooling should be done gradually to avoid cracking of the glassware. Moist /^eaf.— Instruments, syringes, and other suitable objects may be sterilized by boiling in water. Boiling for about five minutes is amply sufficient to destroy the vegetative forms of all bacteria. For the de- struction of spores, boiling for one or two hours is usually sufficient, though the spores of certain saprophytes of the soil have been found Fi©. 8. — Hot Air Sterilizer. occasionally to withstand moist heat at a temperature of 100° C. for as long as sixteen hours. ^ The addition of 1 per cent of sodium car- bonate to boiling water hastens the destruction of spores and prevents the rusting of metal objects sterilized in this way. The addition of car- bolic acid to boiling water in from 2 to 5 per cent usually insures the destruction of anthrax spores, at least, within ten to fifteen minutes. Exposure to live steam is probably the most practical of the methods of heat steriGzation. It may be carried out by simple makeshifts of the kitchen, such as the use of potato-steamers or of wash-boilers. For 1 Christen, loc. cit. 70 BIOLOGY AND TECHNIQUE laboratory purposes, the original steaming device introduced by Koch has been almost completely displaced by devices constructed on the plan of the so-called " Arnold " sterilizer (Fig. 9) . In such an appara- tus, water is poured into the reservoir A' and flows from there into the shallow receptacle B, formed by the double bottom. The flame underneath rapidly vaporizes the thin layer of water contained in B, and the steam rises rapidly, coursing through the main chamber C. Steam which escapes through the joints of the lid of this chamber is condensed under the hood and drops back into the reservoir. Exposure to steam in Tsuch an apparatus for fifteen to thirty minutes insures the death of the vegetative forms of bacteria. In the sterilization of media by such a device, the method of fractional sterili- zation at 100° C. is employed. The prin- ciple of this method depends upon repeated exposure of the media for fif- teen minutes to one-half hour on three succeeding days. By the first exposure all vegetative forms are destroyed. The media may then be left at room tem- perature, or at incubator temperature (37.5° C.) until the following day, when any spores which may be present will have. developed into the vegetative stage. These are then killed by the second ex- posure. A repetition of this procedure on a third day insures sterility. It must always be remembered, however, that this method is applicable only in cases in which the substance to be sterilized is a favorable medium for bacterial growth in which it is likely that spores will develop into vege- tative forms. Exceptionally the method may fail even in favorable media when anaerobic spore-forming bacteria are present. Thus, it has been ob- served that anaerobic spores, failing to develop under the aerobic con- ditions prevailing during the intervals of fractional steriWzation, have developed after inoculation of the media with other bacteria, when sym- biosis had made their growth possible. Tetanus bacilli have, in this way, occurred in cultures of diphtheria bacilli employed for toxin production. Fig. 9. — Arnold Sterilizer, THE DESTRUCTION OF BACTERIA 71 In noting the time of an exposure in an Arnold sterilizer, it is important to time the process from the time when the temperature has reached 100° C. and not from the time of Hghting the flame. The principle of fractional sterilization at low temperatures is ap- plied also to the sterilization of substances which can not be sub- jected to temperatures as high as 100° C. This is especially the case in the sterilization of media containing albuminous materials, when coagulation is to be avoided, or when both coagulation of the medium and sterilization are desired. In such cases fractional sterilization may be practiced in simply con- structed sterilizers, such as a Koch inspissator or, in the case of fluids, such as blood serum, by immersion in a water-bath at a temperature Fig. 10.— Low Temperature Sterilizer (Inspissator). varying above 55° C, according to circumstances. Exposures at such low temperatures may be repeated on five or six consecutive days, usu- ally for an hour each day. The use of steam under pressure is the most powerful method of heat- disinfection which we possess. It is applicable to the sterilization of fomites, clothing, or any objects of a size suitable to be contained in the apparatus at hand, and which are not injured by moisture. In labora- tories this method is employed for the sterilization of infected appa- ratus, such as flasks, test tubes, Petri plates, etc., containing cultures. The device most commonly used in laboratories is the so-called auto- clave, of which a variety of models may be obtained, both stationary and portable. The pruiciple governing the construction of all of these 72 BIOLOGY AND TECHNIQUE is the same. The apparatus usually consists of a gun-metal cylinder supplied with a lid, which can be tightly closed by screws or nuts, and supplied with a thermometer, a safety-valve, and a steam pressure gauge. In the simpler autoclaves, water may be directly filled into the lower part of the cylinder, and the objects to be sterilized supported upon a perforated diaphragm. In this case the heat is directly applied by means of a gas flame. In the more elaborate stationary devices, steam may be let in by piping it from the regular supply used for heating purposes. Exposure to steam under fifteen pounds pressure (fifteen in addition to the usual atmospheric press- ure of fifteen pounds to the square inch) for fifteen to twenty minutes, is sufficient to kill all forms of bacterial life, including spores. In applying autoclave sterilization practically, attention must be paid to certain technical details, neglect of which would result in failure of sterilization. It is necessary always to permit all air to escape from the autoclave before closing the vent. If this is not done, a poorly conducting air-jacket may be left about the objects to be sterilized, and these may not be heated to the temperature indicated by the pressure. It is also nec- essary to allow the reduction of pressure, after sterilization, to take place slowly. Any sudden relief of pressure, such as would be produced by opening the air- vent while the pressure gauge is still above zero, will usually result in a sudden ebullition of fluid and a removal of stoppers from flasks. The temperature attained by the application of various degrees of pressure is expressed in the following table : Fig. 11. — Autoclave. Lbs. Pressure Temperature 1 102.3° 2 104.2 3 105.7 4 X07.3 Lbs. Pressure Temperature 5 108.8° 6 110.3 7 111.7 8 113 THE riiSTRUCTION OF BACTERIA 73 Lbs. Pressure Temperature 9 114.3° 10 115.6 11 116.8 12 13 14 15 16 118 119.1 120.2 121.3 122.4 Lbs. Pressure Temperature 17 123.3° 18 124.3 20 126.2 22 128.1 24 129.3 26 :.... 131.5 28 133.1 30 134.6 CHEMICAL AGENTS INJURIOUS TO BACTERIA Since the time of Koch's ^ fundamental researches upon chemical disinfectants, the known number of these substances has been enor- mously increased, and now embraces chemical agents of the most varied constitution. It is thus manifestly impossible to refer the injurious in- fluence which these substances exert upon bacteria to any uniform law of action. The efficiency of a disinfecting agent, furthermore, is not alone dependent upon the nature and concentrations of the substance itself, but depends complexly upon the nature of the solvent in which it is employed, the temperature prevailing during its application, the numbers and bio- logical characteristics of the bacteria in question, and the time of ex- posure. All these factors, therefore, must be considered in testing the efficiency of any given disinfectant. While it is true, furthermore, that all substances which in a given concentration exert bactericidal or disinfecting action upon a microorganism, will in greater dilution act •antiseptic ally or inhibitively, no definite rules of proportion exist be- tween the two values, which in each case must be determined by experi- ment. Disinfectants Used in Solution. — The actual processes which take place in the injury of bacteria by disinfectants are to a large extent unknown. In the case of strong acids, or strongly oxidizing substances, there may be destruction of the bacterial body as a whole by rapid oxidation. Other substances may act by coagulation of the bacterial protoplasm; others again by diffusion through the cell membrane are able to enter into chemical combination with the protoplasm and exert a toxic action. Again, in other cases, a' difference in tonicity between cell protoplasm and disinfectant may tend to withdrawal of water from the bacterial cell and consequent injury of the microorganism. Among the inorganic disinfectants the most important are the metallic 1 Koch, Arb. a. d. kais. Gesundheitsamt, i, 1881. 74 BIOLOGY AND TECHNIQUE salts, acids, and bases, the halogens and their derivatives, and certain oxidizing agents like peroxide of hydrogen and permanganate of potas- sium. It has been shown by Scheuerlen and Spiro,^ Kronig and Paul,^ and others, that in the case of the salts, acids, and bases, there is a distinct and demonstrable relationship between the disinfecting power of these substances and their dissociation in solution. According to the theory of electrolytic dissociation, when bodies of this class go into solution they are broken up or dissociated into an electro-positive and an electro-negative ion. Thus, metallic salts are broken up into the kation, or positive metal, and into the anion, or negative acid radicle (AgNOg = Ag, + ion and NO3, — ion). In the case of the acids, ionization takes place into the hydrogen ions and the acid radicles, while in the case of the bases the dissociation occurs into the metal, on the one hand, and the OH group on the other. The de- gree of dissociation taking place depends upon the nature of the sub- stance in solution, its concentration, and the nature of the solvent. Thus, in any such solution there appear three substances, the undis- sociated compound as such, its electro-negative ion, and its electro- positive ion, their relative concentrations depending upon an interrela- tionship calculable by definite laws. It goes without saying, therefore, that any chemical or physical reaction, taken part in by such a solution, may be participated in, not only by the dissolved undissociated residue as a whole, but by its separate ions individually as well. In the case of many disinfectants, the writers referred to above have been able to demonstrate a relationship between the degree of dissociation and the bactericidal powers. According to Kronig and Paul, double metallic salts, in which the metal is a constituent of a complex ion and in which the concentration of the dissociated metal-ions is consequently low, have very little disinfecting power. Thus potassium-silver-cyanide, which is a comparatively weak disinfectant, dissociates into the kation K and the complex anion Ag (GN) 2, this latter further dissociating to a very slight degree only. The same writers conclude that the bactericidal action of mercuric chloride and of halogen combinations with metals is directly proportionate to the degree of dissociation. This considera- tion, moreover, explains why aqueous solutions of such substances are more active than are solutions in the alcohols or in ether, since it is well 1 Scheuerlen und Sjdro, Munch, med. Woch., 44, 1897. 2 Kronig und Paul, Zeit. f. Hyg., xxv, 1897. THE DESTRUCTION OF BACTERIA 75 known that metallic salts are ionized in these substances to a much sUghter degree than they are in water. ^ On the other hand, the addition of moderate quantities of ethyl and methyl alcohol or acetones to aqueous solutions of silver nitrate or mercuric chloride, definitely increases the disinfecting action of such solutions. In the case of mercuric chloride, Kronig and Paul obtained the- most powerful effects in solutions to which alcohol had been added in a concentration of 25 per cent. For this empirical fact a satisfactory explanation has not yet been found. Kronig and Paul suggest that low percentages of alcohol may facilitate the penetration of the disinfectant through the cell membrane and thus increase its efficiency, while high percentages of alcohol have the opposite effect, by decreasing the degree of dissociation. *In this connection it has been suggested, however, that absolute and strong alcohols possibly act as desiccating agents, thus actually rendering the bacteria dry and less susceptible to dele- terious chemical influences. In the case of acids and bases the same authors have determined that the powers of disinfection of these substances are again directly proportionate to the degree of their dissociation: that is, to the concen- tration of the hydrogen or hydroxyl ions, respectively. The hydrogen ions are more powerfully active than the hydroxyl ions in equal con- centration; acids, therefore, are more efficient disinfectants than bases. A fact which appears to strengthen the opinion as to the relationship between bactericidal powers and dissociation, is that brought forward by Scheuerlen and Spiro, that the addition of NaCl to bichloride of mercury solutions reduces the disinfecting power of such solutions, in- asmuch as it diminishes the concentration of free ions. In practice, however, NaCl or NH4CI is added to bichloride of mercury solutions, since these substances aid in holding in solution mercury compounds formed in the presence of alkaUne albuminous material, blood serum, pus, etc. In regard to the halogens, Kronig and Paul have shown that the germicidal power of this class of elements is inversely proportionate to their atomic weights. Thus, chlorine with the lowest atomic weight is the strongest disinfectant of the group. Next, and almost equal to this, is 1 Water is the strongest dissociant known. Methyl alcohol has about one-half to two-thirds the dissociating power of water (Zelinsky, Zeit. f. physiol. -Chemie, xx, 1896). Ethyl alcohol allows dissociation much less than methyl alcohol; ammonia allows dissociation to about one-third to one-fourth the extent of water. See Jones, "Elements of Physical Chemistry," p. 371. Macmillan, New York, 1902. 76 BIOLOGY AND TECHNIQUE bromine. Iodine with a much heavier atomic weight than either of the former is distinctly less bactericidal. Chloride op Lime. — Of the halogen compounds used in practice, the most important is chloride of lime or bleaching powder. As to the composition of this substance, there is some difference of opinion. It was formerly believed to be a mixture of calcium hypochlorite, CaCClOg), and of calcium chloride, CaClg. The fact that the substance is not deliquescent, however, speaks against the presence of calcium chloride as such, and it is probable that it consists of a single com- pound with the formula CaOClg. The action of acids or even of atmospheric COo upon this substance results in the liberation of chlorine. For instance, CaCCl^O) + 2HC1 = CaClg + 2HC10. 2HC10 + 2HC1 = 2H2 + 2CI2. Bleaching powder is readily soluble in about twenty parts of water. According to Nissen,^ solutions of 2 in 1,000 of this substance destroy vegetative forms of bacteria in five to ten minutes. Its bactericidal action depends on the hypochlorous acid formed. After water precipi- tation an efficient dosage is 10 pounds to the million gallons. Terchloride of Iodine (ICI3) is an extremely strong disinfectant, being efficient for vegetative forms in solutions of 0.1 per cent in one minute and a 1 per cent solution destroying spores within a few minutes.^ Painting with tincture of iodine (10 per cent) is a simple and reliable method of sterilizing the skin. It is now used in many clinics in sterilizing the field of operation. Peroxide of Hydrogen is formed by the action of dilute sulphuric acid upon peroxide of barium. It readily gives up oxygen and acts upon bacteria probably by virtue of the liberation of nascent oxygen. In the presence of organic matter, such as blood, pus, etc., associated with bacteria, II2O2 is quickly reduced and weakened. It is important that the HgOg come in immediate contact with the bacteria. In prac- tice, therefore, blood and pus should be removed from wounds when applying the H2Q2 or a large excess of H2O2 should be used. Permanganate of Potassium, acting probably in the same way, is a powerful germicide. It also is readily reduced by many organic sub- stances often associated with bacteria, being rendered weaker thereby. 1 Nissen, Zeit. f . Hyg., viii, 1890. 2 y, Behring, Zeit. f. Hyg., ix, 1891. THE DESTRUCTION OF BACTERIA 77 Among organic disinfectants those of most practical importance are the alcohols, formaldehydes, iodoform, members of the phenol group and its derivatives, carbolic acid, cresol, lysol, creolin, salicylic acid, cer- tain ethereal oils, and, more recently introduced, organic silver salts such as protargol, argyrol, argonin, and others. The alcohols are but indifferent disinfectants. Koch ^ in 1881 found that anthrax spores remained alive for as long as four months when immersed in absolute and in 50 per cent ethyl alcohol. On the other hand, while absolute alcohol possesses practically no germicidal powers, possibly because of the forrnation of a protecting envelope by the coagulation of the bacterial ectoplasm, or, as suggested above, by desiccation due to the abstraction of water, dilute alcohol in a concen- tration of about 50 per cent is distinctly germicidal, destroying the vege- tative forms of bacteria in from ten to fifteen minutes or less.^ Attention has already been called to the fact that moderate ad- ditions of alcohol to aqueous solutions of mercuric chloride enhance the germicidal power of this disinfectant. Additions of ethyl and methyl alcohol to carbolic acid or formaldehyde solutions, on the other hand, progressively decrease the bactericidal activities of these substances.^ The value of boiling alcohol for the destruction of spores — especially in the sterilization of catgut — has been investigated by Saul,'* who found that boiling in absolute ethyl, methyl, or propyl alcohol is prac- tically without effect, while spores are destroyed readily in boiling dilute alcohol, the most effectual being propyl alcohol of a concentra- tion of from 10-40 per cent. Iodoform (CHIg)^ is weakly antiseptic in itself, but when introduced intD wounds where active reducing processes are taking place — often as the result of bacterial growth — iodine is liberated from it and active bactericidal action results. Carbolic acid (CgHgOH), at room temperature, consists of color- less crystals which become completely liquefied by the addition of 10 per cent of water. In contradistinction to most inorganic disinfectants, the action of carbolic acid and other members of the phenol group is 1 Koch. Arb. a. d. kais. Gesundheitsamt, i, 1881. 2 Epstein, Zeit. f. Hyg., xxiv, 1897. 3 Kronig und Paul, loc. cit. * Saul, Archiv f. klin. Chir., 56, 1898. 6 V. Behring, " Bekaempfung d. Infektions-Krankh.," Leipzig, 1894. 78 BIOLOGY AND TECHNIQUE not in any way dependent upon dissociation.^ According to Beckmann ^ and others, carbolic acid acts as a molecule and not by individual ions. The proof of this is brought out by the fact that the addition of NaCl to carbolic acid solutions, an addition which would tend to decrease the concentration of free ions, markedly increases the bactericidal powers of such solutions. On the other hand, as stated above, addi- tions of alcohol progressively diminish the efficiency of the phenols. Other members of this group of disinfectants are ortho-, meta-, and PARACRESOL (Ce5H4CH30H) , isomeric compounds differing only in the position of the OH radicle. Tricresol is a mixture of these three. The cresols are relatively more powerfully germicidal than is carbolic acid, but are less soluble in water. Lysol is a substance obtained by the solution of coal-tar cresol in neutral potassium-soap. Dissolved in water it forms an opalescent easily flowing liquid. According to Gru- ber,^ its germicidal action is slightly greater than that of carbolic acid. Creolin, another combination of the cresols with potassic soap, forms with water a turbid emulsion, v. Behring ^ expressed the relative germicidal powers of carbolic acid, cresol, and creolin for vegetative forms by the numbers 1:4 : 10, in the order named. Formaldehyde (H-COH), or methyl aldehyde, is a gas which is easily produced by the incomplete combustion of methyl alcohol. The methods of actually generating it for purposes of fumigation will be discussed in a subsequent paragraph. In aqueous solution this substance forms a colorless liquid with a characteristic acrid odor, and in this form is largely used as a preservative for animal tissues and as a germicide. It is marketed as "formalin," which is an aqueous solution containing from 35 to 40 per cent of the gas and which exerts distinctly bactericidal action on vegetative forms in further dilutions of from 1 to 10 to 1 to 20 (formaldehyde gas 1 : 400 to 1 : 800) . Anthrax spores are killed in 35 per cent formaldehyde in ten to thirty minutes.^ Unlike the phenols, the addition of salt to formaldehyde solutions does not increase its efficiency, but similar to them, additions of ethyl and methyl alcohol markedly reduce its germicidal powers. The essential oils which are most commonly used in practice — largely as intestinal antiseptics — are those of cinnamon, thyme, eucalyp- ^ Scheuerlen und Spiro, Miinch. med. Woch., 44, 1897. 2 Beckmann, Cent. f. Bakt., I, xx, 1896. 8 Gruber, Cent. f. Bakt, I., xi, 1892. *v. Behring, loc. cit., p. 111. » Kronig und Paul, loc. cit. THE DESTRUCTION OF BACTERIA 70 tus, and peppermint. Omeltschenko ^ believes that the employment of these oils in emulsions is illogical, inasmuch as their bactericidal powers depend upon their vaporization. He classifies the oils in decreasing order of their efficiencj^ as follows: Oil of cinnamon, prunol, oil of thyme, oil of peppermint, oil of camphor, and eucal3^ptol. Methods of Testing the Efficiency of Disinfectants.— The efficiency of any given disinfectant depends, as we have seen, upon a number of factors, any one of which, if variable, may lead to considerable differences in the end result. Thus, as far as the bacteria themselves are concerned, it is necessary to remember that not only do separate species differ in their resistance to disinfectants, but that different strains within the same species may show such variations as well. This fact largely ac- counts for the widely varying reports made by different investigators as to the resistance of anthrax spores, and depends possibly upon tem- porary or permanent biological differences produced in bacteria by the conditions of their previous environment. The numbers of bacteria exposed to the disinfectant, furthermore, is a factor which should be kept constant in comparative tests. The medium, moreover, in which bacteria are brought into contact with the disinfectant is a matter of great importance, inasmuch as either by entering into chemical combination with the disinfectant it m.ay detract from its concentration or by coagulation it may form a purely mechanical protection for the microorganism. Thus bacteria which may be de- stroyed in distilled water or salt-solution emulsion with comparative ease, may evince an apparently higher resistance if acted upon in the presence of blood serum, mucus, or other albuminous substances. Temperature influences bactericidal processes in that most chemical disinfectants are more actively bactericidal at higher than at lower temperatures, a fact due most likely to the favorable influence of tem- perature upon all chemical reactions.^ As far as merely inhibitory or antiseptic values are concerned, however, the temperature least favor- able for the reaction of the antiseptic is that which represents the opti- mum growth temperature for the microorganism in question and the inhibitory effects of any substance are less marked at this point than at temperatures above or below it. The important influence exerted by the solvent in which the 1 Omeltschenko, Cent. f. Bakt., I, ix, 1891. ' V. Behring, " Bekaempf. der Infektions-Krankh., Infektion u. Desinfection," Leipzig, 1894. 7 so BIOLOGY AND f ECHNIQtJE disinfectant is employed has already been discussed. For ordinary work it is customary to express absolute and comparative antiseptic and bactericidal values in terms of percentages based upon weight, and this, beyond question, is both simple and practical. For strictly scien- tific comparisons, however, as Kronig and Paul ^ have pointed out, it is by far more accurate to work with equimolecular solutions. Rideal and Walker ^ have devised a method of testing disinfectants, in which an attempt is made to establish a standard for comparisons. They choose, as the standard, carbolic acid, and establish what they call the " carboHc-acid coefficient." This coefficient they obtain in the fol- lowing way: the particular dilution of the disinfectant under investiga- tion which will kill in a given time, is divided by the strength of carbolic acid which, under the same conditions, will kill the same bacteria in the same time. We quote an example of such a test, given by Simpson and Hewlett,^ comparing formalin and carbolic acid. BACILLUS PESTIS. Dilution. Time in Minutes. Sample. 2.5 5 7.5 10 12.5 15 Formalin -] Carbolic acid -j lin 30 lin 40 1 in 100 1 in 110 growth growth growth growth growth growth growth ...... In the above table, formalin 1 in 30 killed in the same time as carbolic acid 1 in 110. Thus the carboUc-acid coefficient of formalin in this test = ^%io = .27. The Rideal- Walker method has been much used and is recommended by many workers.^ The most precise method of standardizing disinfectants is that now in use in the U. S. Public Health Service. It is a modification of the Rideal- Walker procedure devised by Anderson and McClintic.^ Stock 5 per cent solutions of the disinfectant in question and of the 1 Kronig und Paul, loc. cit 2 Rideal and Walker, Jour, of the Sanitary Ins. London, xxiv. 3 Simpson and Hewlett, Lancet, ii, 1904. * Sommerville, Brit. Med. Jour., 1904. ^Anderson and McClintic, Jour, of Inf. Dis., 1911, viil, 1- THE DESTRUCTION OF BACTERIA 81 standard (phenol) are first prepared and a series of accurate dilutions made with distilled water using graduated pipettes. (To make 1 : 70 take 4 c.c. of stock and 10 c.c. distilled water; 1:80 = 4 c.c. of stock + 12 c.c. distilled water; 1:90 = 4 c.c. stock + 14 c.c. distilled water; 1:500 = 2 c.c. of stock + 48 c.c. of distilled water. Complete dilution tables are given in their original article.) The series should include dilutions strong enough to kill B. typhosus in two and a half minutes and weak enough to fail to do so in fifteen minutes. If dilutions greater than 1- 500 are required, a second 1 per cent stock solution is prepared. They adopted the following scale for their tests : Dilutions up to 1 : 70 should vary from the next in the series by a difference of 5 (i.e., 5 parts of water). From 1:70 to 1:160 by a difference of 10 ,0 From 1 : 160 to 1 : 200 by a difference of 20 ||^' From 1:200 to 1:400 by a difference of 25 "'l' From 1 : 400 to 1 : 900 by a difference of 50 From 1 : 900 to 1 : 1800 by a difference of 100 From 1 : 1800 to 1 : 3200 by a difference of 200 and so on if higher dilutions are necessary. Short wide test tubes 1 inch by 3 inches are used in making the test. These are placed in a rack in a water bath at 20° C. Five c.c. of each dilution are measured into a series of these tubes beginning with the strongest specimen and rinsing the pipette once with each dilution before the 5 c.c. are measured out. For inoculation, a 24-hour broth culture of B. typhosus is prepared which has been transferred daily for at least 3 days. Before use it is shaken and filtered through sterile filter paper. The wide test tubes containing diluted disinfectant are inoculated with /lo c.c. of this culture with a graduated pipette. The tip of the pipette is held against the side of the tube to insure accurate measurement and the tube immediately shaken to mix the bacteria thoroughly with the disinfectant. Test inoculations are made from this mixture at proper intervals into tubes containing 10 c.c. of standard extract broth of + 1.5 acidity, using loops 4 mm. in diameter. At least four such loops should be at hand, supported on a rack or wooden block so that a fan-tail Bunsen burner may be placed under each wire in turn. Each one is sterilized after a plant is made and allowed to cool while the other three are being used in order. The test is conducted as follows: A row of ten wide tubes containing dilutions of the antiseptic is placed in the water bath at 20° C. and time allowed for them to reach the temperature of the bath. They are then inoculated in order at intervals of exactly 15 seconds. Fifteen seconds 82 BIOLOGY AND TECHNIQUE after the last tube has been inoculated a subculture is made from the first tube of the series (i.e., 23^ minutes after this first tube was inocu- lated) and from the other tubes in order at 15-second intervals. Fifteen seconds after this first series of subcultures is completed a second series of subcultures is begun which will give the result of a 5-minute exposure to the antiseptic and the subinoculations continued at 15-second intervals until all dilutions have been tested for fifteen minutes. If the strength of the antiseptic is known approximately subcultures of the lower dilu- tions for the longer periods may be omitted. It is convenient to have an assistant at hand to call time and to label the subcultures as soon as made. The^jbes may, however, be placed in order in suitable racks DETP .{m: ATION OF THE CARBOLIC-ACID COEFFICIENT •!► OF A DISINFECTANT. (Anderson and McClintic) Name "A" Temperature of Medication 20° C. Culture Used B. Typhosus 24-hr., Extract Broth, Filtered Proportion of Culture and Disinfectant 0.1 c.c. + 5 c.c. Organic Matter, None; Kind, None; Amount, None. Subculture Media Standard Extract Broth Reaction + 1.5 Quantity in Each Tube 10 c.c. Sample. Dilu- tion. Time Culture Exposed to Action of Disinfectant for Minutes Phenol Coefficient. 2y2 5 7y2 10 12^ 15 Phenol 1:80 1:90 + — — 80)375 1:100 + + + — — — 4.69 1:110 + + + + + — 110)650 5.91 Disinfectant "A".. 1:350 1:375 — — 2)10.60 1:400 + — — — 5.30 = 1:425 + + ' — — — — coefficient 1:450 + + — — — 1:500 + + -r- — — — 1:550 + + + — — — 1:600 + + + + — — 1:650 + + + + + — 1:700 + + + + + + 1:750 + + + + + + THE DESTRUCTION OF BACTERIA 83 without labelling. The subculture tubes are incubated for 48 hours at 37° C. and those in which growth is observed are recorded positive. To obtain the coefficient the weakest dilution of the unknown antiseptic which kills in 23/^ minutes is divided by the weakest dilution of phenol which kills in the same time. The same is done for the weak- est strength that kills in 15 minutes and an average is taken. The results of such a test are shown in the table on page 82. As only the 23/^-minute and 15-minute intervals are used in deter- mining this result it seems unnecessary to make plants at the intervening periods except in special cases where more detailed information is desired. The procedure may be modified by adding some organic substance such as killed bacteria to the diluted antiseptic. For many substances, e.g., bichloride of mercury, the antispetic value in presence of organic matter is much lower than in watery solution. Anderson and McClintic insist that great care in making the dilutions and rigid adherence to a uniform technique are necessary to obtain consistent results in such tests. Determination of Antiseptic Values. — The antiseptic or in- hibitive strength of a chemical substance, sometimes spoken of as the ''coefficient of inhibition," is determined by adding to definite quantities of a given culture medium, graded percent- ages of the chemical substance which is being investigated and plant- ing in these mixtures equal quantities of the bacteria in question. The medium used for the tests may be nutrient broth or melted gelatin or agar. If broth is used, growth is estimated by turbidity of -the medium and by morphological examination; if the agar or gelatin is employed, plates may be poured and actual growth observed. Thus, in the case of carbolic acid, a 5 or 10 per cent solution is prepared and added to tubes of the medium, as follows: Tube 1 contains 5% carbolic 2 c.c. + broth 8 c.c. = 1: 1,000 carbolic acid. " 2 " 5 " 1 c.c. + broth 9 c.c. = 1:200 " 3 " 5 " .5 c.c. + broth 9.5 c.c. = 1:400 " " " 4 " 5 " 7 c.c. + broth 9.8 c.c. = 1:1,000 " " 5 " 5 " .1 c.c. + broth 9.9 c.c. = 1:5,000 " " To each of these tubes a definite quantity of the bacteria is added either by means of a standard loopful of a fresh agar culture, or better by a measured volume of an even emulsion in sterile salt solution. The inoculated tubes are then incubated at a temperature corresponding to the optimum growth temperature for the microorganism in question. The tubes are examined for growth from day to day. From tubes containing higher dilutions, in which no growth is visible, transplants 84 BIOLOGY AND TECHNIQUE INHIBITION STRENGTHS OF VARIOUS ANTISEPTICS. Adapted from Flugge, Leipzig, 1902. Acids Sulphuric Hydrochloric Sulphurous Arsenous . . , Boric Alkalies Potass, hydrox . Ammon. hydrox. Calcium hydrox. Salts Copper sulphate Ferric sulphate Mercuric chlorid . Silver nitrate . . . Potass, perman Halogens and Compounds Chlorin Bromin lodin Potass, iodid Sodium chlor Anthrax Bacilli. 1 : 3,000 1 : 3.000 1:800 1:700 1:700 Organic Compounds Ethyl alcohol Acetic and oxalic acids Carbolic acid Benzoic acid Salicylic acid Formalin (4% formalde- hyde) Camphor Thymol Oil mentha pip Oil of terebinth Peroxide of hydrogen, 1 : 100,000 1 : 60,000 1 : 1,000 1 : 1,500 1 : 1,500 1 : 5,000 Other Bacteria. Choi. spir. 1 : 6,000 B. diph. 1 : 3,000 B. mallei 1 : 700 B. typh. 1 : 500 Choi. spir. 1 : 1,000 Putrefactive Bac- teria in Bouillon. B. diphth. 1 : 600 Choi. spir. 1 : 400 B. typh. 1 : 400 Choi. spir. 1 : 500 B. typh. 1 : 500 Choi. spir. 1:1,100 B. typh. 1:1,100 B. typhosus 1 : 60,000 Choi. spir. B. typhosus 1 : 50,000 1:60 1:12 1:800 1 : 1,000 1 : 1,500 1 : 1,000 1 : 10,000 1 : 3,000 1 : 8,000 B. diph. 1 : 500 B. typh. 1 : 400 Choi. spir. 1 : 600 Choi. spir. 1 : 20,000 Staphylo. 1 : 5,000 1 : 6,000 1:200 1:100 1 : 1,000 1:90 1 : 20,000 1:500 1 : 4,000 1 : 2,000 1:5,000 1:7 1:10 1:400 1 : 1,000 1 : 3,500 1 : 2,000 THE DESTRUCTION OF BACTERIA 85 BACTERICIDAL STRENGTHS OF COMMON DISINFECTANTS. Adapted from FlO^gge, Leipzig, 1902. ACID8 Sulphuric ... Hydrochloric Sulphurous . Sulphurous Boric Alkalies Potass, hydrox. . Ammon. hydrox. Calcium Salts Copper sulphate Mercuric chlor. Silver nitrate . . Potass, permang. ''' Calc. chlorid " Halogens and Com POUNDS Chlorin Trichlorid of iodin . . , Organic Compounds Ethyl alcohol Acetic and oxalic acids Carbolic acid Lysol Creolin SaHcylic acid Formalin (40% for- maldehyde) Peroxide of hydrogen . Strepto- and Staphj^lo- cocci. 5 Minutes. 1:10 1:10 1:5 1 : 10,000 to 1,000 1 : 1,000 1 :200 1 per cent. 1 :^200 70%-15 minutes 1:60 1 :300 1 : 1,000 1:10 Cone. Anthrax and Tyfjhoid Bacilli, Cholera Spirillum. 5 Minutes. 1^100 1:100 1:300 1:300 1 : 1,000 1 : 2,000 1:500 .1 per cent. 1 : 1,000 70%-lOmins Cholera 1:200 Typh. 1 : 50 1:300 1 :100 1:20 1 :200 2-24 Hours. 1 : 1,500 1 : 1,500 (Typhoid 1:700) 1:300 (Gas 10 vol. %) 1:30 1 : 10,000 1 : 4,000 1 : 2-300 1.300 1 : 3,000 1 : 1,000 1 : 500 Anthrax Spores. 1 : 50 in 10 days 1 : 50 in 10 days Cone. sol. incomplete disinfection 1 : 20 (5 days) 1:2,000 (26 hours) 1 : 20 (1 day) 1 : 20 (1 hour) 2 per cent (in 1 hr.) 1 : 1,000 (in 12 hrs.) Alcol. 50% for 4 months without killing spores. Koch.i 1 : 20 (4-45 days) (at 40° in 3 hrs.) (10% in 5 hrs.) 1 : 20 (in 6 hrs.) 1 : 100 (in 1 hr.) 3 : 100 (in 1 hr.) * Koch. Arb. a. d. kais. Gesundheitsamt, 1, 1881. 86 BIOLOGY AND TECHNIQUE are made to determine the presence of living bacteria and to distinguisli between inhibition or antisepsis and bacterial death or disinfection. The determination of the bactericidal or disinfectant value of a chemical substance upon spores may be carried out by a variety of methods. Koch,^ using anthrax spores as the indicator, dried the spores upon previously sterilized threads of silk. These were exposed to the disinfectant at a definite temperature for varying times, the disinfect- ant was then removed by washing in sterile water, and the threads planted upon gelatin or blood serum media and incubated. A serious objection to this method was pointed out by Geppert,^ who maintains that it is impossible by simple washing to remove completely the disin- fectant in which the thread has been soaked. This author suggests that, whenever possible, the disinfectant, at the end of the time of exposure, should be removed by chemical means. In the case of bichloride of mer- cury Geppert exposes emulsions of the bacteria to aqueous solutions of the disinfectant, and at the end of exposure precipitates out the bichlor- ide of mercury with ammonium sulphide. In the case of a large number of disinfectants, however, this is not possible, and, when the thread method is used, removal of the chemical agent by washing must be practised. Complete removal of the disinfectant is especially desirable, since spores previously exposed to these substances are more easily in- hibited by dilute solutions than are normal spores. The spores may be dried upon the end of a glass rod, which, after exposure, is washed in distilled water or salt solution and then immersed in sterile broth.^ A simple method is that in which graded percentages of the disin- fectant are added to the menstruum, blood, blood serum, broth, etc., in which the disinfectant is to be tested, and equal quantities of bacteria thoroughly emulsified in water or salt solution are added. Loopfuls of these mixtures are then planted from time to time in agar or gelatin plates upon which colony counts can afterward be made. In all such tests it is important to remember that the presence of organic fluids, blood serum, mucus, etc., considerably alters the efficiency of germicides, and whenever practical deductions are made, experimental imitation of the actual conditions should be attempted. Practical Disinfection. — In practical disinfection with chemical agents, the disinfectant must be chosen to a certain extent in accordance with the material to be disinfected. 1 Koch, Arb. a. d. kais. Gesundheitsamt, 1, 1881. 2 Geppert, Berl. klin. Woch., xxvi, 1889. » miL Rep. Am. Pub. Health Assn., xxiv, 1898. THE DESTRUCTION OF BACTERIA , 87 Sputum is a substance extremely difficult to disinfect because the bacteria present are surrounded by dense envelopes of mucus, through which disinfectants do not easily diffuse. For sputxmi disinfection, es- pecially tuberculous sputum, carboUc acid — 5 per cent solution — or any of the phenol derivatives in similar concentration, may be used. Bichloride of mercury is of very little use in sputum disinfection be- cause of the dense protective layers of albuminated mercury which form about the microorganisms. Sputum should always be received into cups containing the disinfectant, and contaminated handkerchiefs should be soaked in the solution. Feces from typhoid, dysentery, and cholera patients should be steril- ized by burning, if possible, or by thoroughly mixing with large quan- tities of boiling water; but if chemical disinfectants are to be used, five per cent carbolic acid or dilute formalin are convenient. Milk of lime and chloride of lime are useful, though somewhat inconvenient. Bichlo- ride of mercury is of little value in this case for the same reason that it is valueless in sputum disinfection. In all cases of feces dis- infection it is extremely important that the chemical agent should be added in large quantities and thoroughly mixed with the discharge. Ldnen, napkins, and other cloth materials which have come into con- tact with patients should be soaked for one or two hours in one per cent formaldehyde, five per cent carbolic acid, or 1 : 5,000 or 1 : 10,000 bichloride of mercury. After this, the material may be taken from the sick-room and boiled. It is extremely important that cloth material should never be removed from the sick-room in a dry state. Urine may be easily disinfected by the addition in proper con- centration of any of the disinfectants named above. The methods for sterilization of surgical instruments and the prepara- tion of the skin of the patient for operation are subject to so many local variations that it is hardly within the scope of a text-book on bacteriology to mention them. Metal instruments are usually sterilized by boiling in soda solution and may be subsequently immersed in five per cent car- bolic acid solution. Catgut may be sterilized by boiling in alcohol or by subjecting it to temperatures of 140° C. and over, for several hours in oils (albolin). The disinfection of the hands is also a matter of much variation. Two methods frequently quoted are those of Welch and of Fiirbringer. In Welch's method the hands are brushed with green soap in water as hot as it can be borne for at least five minutes. They are then rinsed and immersed for two minutes in a warm saturated solution of perman- 88 ^ BIOLOGY AND TECHNIQUE ganate of potash in which they are rubbed with a sponge or sterile cotton. They are then transferred to a saturated solution of oxalic acid, until the red color has entirely disappeared. Following this, they are rinsed in sterile water and then immersed in a 1 : 500 bichloride of mercury solution for one to two minutes. According to Fiirbringer's method, the finger nails are carefully cleaned with an orange-wood stick or nail file; the hands are then thor- oughly brushed with a nail brush in green soap and hot water for five minutes. Following this they are immersed in 60 per cent alcohol for one minute, then in 3 per cent carbolic acid solution for one minute; after which they are rinsed in sterile water and dried. Rooms, closets, and other closed spaces which are contaminated, must be disinfected largely by gaseous disinfectants. After such disinfection in the case of cellars, privies, and other places where feasible, walls and ceilings should be whitewashed. Gaseous Disinfectants for Purposes of Fumigation. — There are a large number of gaseous agents which are harmful to bacteria. Only a few, however, are of sufficient bactericidal strength to be of practical importance. Oxygen, especially in the nascent state, may exert distinct bacteri- cidal action upon some bacteria. That strictly anaerobic strains are inhibited by its presence has already been mentioned. Ozone was shown by Ransome and Fullerton ^ to exert considerable germicidal power when passed through a liquid medium in which bac- teria were suspended, but was absolutely without activity when em- ployed in the dry state. Chlorine because of its powerful germicidal action was once looked upon with favor, but has been found quite inadequate from a practical point of view because of its injurious action upon materials, and its irregularity of action. Chlorine, too, is but weakly efficient unless in the presence of moisture.^ Sulphur dioxide or sulphurous an^ydrid (SO2), formerly much used for room disinfection, is no longer regarded as uniformly efficient for general use. The gas is produced by burning ordinary roll sulphur, conveniently in a Dutch oven. To be at all effective, water should be vaporized at the same time, since the disinfectant action is dependent upon the formation of sulphurous acid. The concentration of the gas should be at least 8 per cent of the volume of air in the room. For this ^ Ransome and Fullerton, Rep. Public Health, July, 1901. * Fischer and Proskauer, Mitt. a. d. kais. Gesundheitsamt, x, 11, 1882. THE DESTRUCTION OF BACTERIA 89 purpose about three pounds of sulphur should be burned for every thousand cubic feet of space. It should be allowed to act for not less than twenty-four hours. The researches both of Wolffhiigel^ and of Koch 2 have shown that the gas is not sufficient for the destruction of spores. Park^ believes that sulphur dioxid used in quantities of four pounds of sulphur to 1,000 cubic feet is of practical value for fumigating purposes in cases of diphtheria and the exanthemata. Sulphur dioxid fumigation is more effective than formaldehyd for the destruction of insects — fleas, lice and bedbugs — a matter of importance in epidemics of typhus fever, relapsing fever, plague, etc. Of all known gaseous disinfectants by far the most reliable is form- aldehyd. In all cases where formaldehyd fumigation is intended, clothing, bed-linen, and fabrics should be spread out, cupboards and drawers freely opened. The cracks of windows and doors should be hermetically sealed with paper strips or by calking with cotton. In all cases moisture should be provided for, either in the generating appa- ratus or by a separate boiler. Direct evaporation of formaldehyd from formalin solutions has been the principle underlying most of the methods. If such evaporation is attempted from an open vessel, however, polymerization of formal- dehyd to the solid trioxymethylene occurs. To prevent this, Trillat* and others have constructed special autoclaves in which 20 per cent of calcium chloride is added to formalin which is then vaporized under pressure. The Trillat autoclave, as well as others constructed on the same principle, consists of a strong copper chamber of a capacity of about a gallon, fitted with a cover which can be tightly screwed into place. The cover is perforated by an outlet vent, a pressure gauge, and a thermometer. The whole apparatus is adjusted upon a stand and set over a kerosene lamp. Into the chamber is put about one-half to three- quarters its capacity of 40 per cent formaldehyd (commercial formalin) containing 15-20 per cent calcium chlorid. The solution used should be free from methyl alcohol, since this leads to the formation, with formaldehyd, of methylal, which is absolutely without germicidal action. For a room of about 3,000 cubic feet Trillat advises the con- ^ Wolffhiigel, Mitt. a. d. kais. Gesundheitsamt, i, 1881. 2 Koch, Mitt. a. d. kais. Gesundheitsamt, i, 1881. ' Park, "Pathogen. Bact.," N. Y., 1908. * Trillat, Compt. rend, de I'acad. des sc, 1896. 90 BIOLOGY AND TECHNIQUE tinuance of the gas flow for about an hour. • The method is not uni- formly reliable. A method which has found much favor is that in which glycerin — usually in a concentration of 10 per cent — is added to formalin. Ac- cording to Schlossmann ^ the presence of glycerin hinders polymerization. An appa- ratus in which this mixture is conveniently utilized is that of Lentz (see Fig. 12). For- malin with 10 per cent glycerin is placed in the copper tank and heated by a burner. This apparatus has been favorably endorsed by the War Department of the United States. The so-called Breslau method of generat- ing formaldehyd depends upon the evapora- tion of formaldehyd from dilute solutions, v. Brunn ^ claims that where formalin in 30 to 40 per cent strength is evaporated, water vapor is generated more rapidly than formaldehyd is liberated, and a concentration leading to poljmaerization occurs. If, however, dilution is carried out until the formaldehyd in the solution is not more than 8 per cent, the generation of water vapor and formaldehyd take place at about equal speed and no concentration occurs. Schlossmann ^ furthermore claims that polymerization in the vaporized formaldehyd does not occur if sufficient water vapor is present — a principle which may also contribute to the efficiency of the Breslau method. In prac- tice, the apparatus devised by v. Bruim (Fig. 13) consists of a strong copper kettle of about 34 cm. diameter by 7.5 cm. height. This is completely closed except for two openings in the slightly domed top, one of which is the exit vent, the other, laterally placed, is for pur- poses of filling and is closed by a screw stopper. The tank is filled with a solution of formalin of a strength of from 8 to 10 per cent (com- mercial formalin 1:4). The apparatus permits the evaporation of large quantities of fluid in a short time (3 liters in one hour). When the lamp is left in a closed room care should be taken to fill it with a quantity of alcohol proportionate to the amount of fluid to be evap- orated. This, according to v. Brunn, is about one-quarter of the Fig. 12. — Lentz Formalin Apparatus. * Schlossmann, Munch, med. Woch., 45, 1898. * V. Brunn, Zeit. f. Hyg., xxx, 1899. THE DESTRUCTION OF BACTERIA 91 ,-^irectly upon the cover-slip. The concavity on the slide, having first been rimmed with vaseline, by means of a small camel's-hair brush, the cover-slip is inverted over the slide in such a way that the drop hangs freely within the hollow space. The preparation is then ready for examination under the microscope. 1 Although the silicates of which glass is composed are extremely stable, never- theless alkaline silicates which are said to separate out on the surface, together with grease and dirt left upon the glass by handling, during blowing and cutting, neces- sitate cleansing before use. This may be accomplished by a variety of methods, A simple one suitable for general application is as follows: (1) The slides and cover- slips are thrown singly into boihng water and left there for half an hour; (2) wash in twenty-five per cent sulphuric acid; (3) rinse in distilled water; (4) wash in alcohol; (5) wipe with a clean cloth and keep dry under a bell-jar. Another method convenient for routine use is to immerse, after thorough washing in soap-suds and acid^ in ninety-five per cent alcohol and to leave in this until the time of use. 93 94 BIOI-OGY AND TECHNIQUE Another method, known as the '' hanging block method," devised by Hill,^ for the study of Uving bacteria in soUd media is carried out as fol- lows: Nutrient agar is poured into a Petri dish and allowed to solidify. Out of this layer a piece about a quarter of an inch square is cut. This is placed on a sterile slide. The upper surface of the agar block is then inoculated with bacteria by surface smearing, and the preparation covered with a sterile dish and allowed to dry for a few minutes in the incubator. A sterile cover-slip is then dropped upon the surface of the ^ Fig. 14. — Hanging Drop Preparation. block and sealed about the edges with agar. Block and cover-slip are then taken from the slide and fastened over a moist chamber with paraf- fin. The entire preparation can be placed upon the stage of a microsocpe. This method is especially designed for the study of cell-division. Living bacteria may also be studied in stained preparations by the so-called " intravital " method of Nakanishi. Thoroughly cleaned slides are covered with a saturated aqueous solution of methylene-blue. This is spread over the slide in an even film and allowed to dry. After drying the slide should appear of a trg-nsparent sky-blue color. The micro- organisms which are to be examined are then emulsified in water, or are taken from a fluid medium and placed upon a cover-slip. This is dropped, face downward, upon the blue grouni of the slide. In this way bacteria may be stained without being subjected to the often destructive proc- esses of heat or chemical fixation. According to Nakanishi, cytoplasm is stained blue, while nuclear material assumes a reddish or purplish hue. The Study of Bacteria in Fixed Preparations. — Stained preparations of bacteria are best prepared upon cover-slips, the process consisting of the following steps : (1) Spreading on cover-slip; (2) drying in air; (3) fixing; (4) staining; (5) washing in water; (6) blotting; (7) mounting. (1) Smearing. — Bacteria from a fluid medium are transferred in a small drop of the fluid, with a platinum loop, to a cover-slip and care- fully spread over the surface in a thin film. If taken from a solid medium a small drop of sterile water is first placed upon the cover-slip and the bacteria are then in very small quantity carefully emulsified in this drop with the platinum needle or loop and spread in an extremely thin film. HUl, Jour, of Med. Research, vii, 1902. MICROSCOPIC STUDY AND STAINING 95 (2) The film is allowed to dry in the air. (3) When thoroughly dried, fixation is carried out by passing the preparation, film side up, three times through a Bunsen flame, at about the rate of a pendulum swing. Fixation by heat in this manner is most convenient for routine work, but is not the most delicate method, in- asmuch as the degree of, heat applied can not be accurately controlled. The other methods which, may be employed are immersion in methyl alcohol, formalin, saturated aqueous bichloride of mercury, Zenker^s fluid, or acetic acid. If chemical fixatives are used, they must be re- moved by washing in water before the stain is applied. If a prepara- tion is made upon a slide instead of a cover-slip, passage through the flame should be repeated eight or nine times. (4) Staining. — The dyes used for the staining of bacteria are, for the greater part, basic anilin dyes, such as methylene-blue, gentian- violet, and fuchsin. These may be applied for simple staining in 5 per cent aqueous solutions made up from filtered saturated alcoholic solutions, or directly by weight. They are conveniently kept in the laboratory as saturated alcoholic solutions. The strengths of some saturated solutions are as follows: Saturated Solutions ^ (Stains Gruebler or Merck). Fuchsin (aqueous), 1.5 per cent. Fuchsin (alcohol 96 per cent), 3 per cent. Gentian-violet (aqueous), 1.5 per cent. Gentian-violet (alcohol 96 per cent), 4.8 per cent. Methylene-blue (aqueous), 6.7 per cent. Methylene-blue (alcohol 96 per cent), 7 per cent. The staining solution, in simple routine staining, is left upon the fixed bacterial film for from one-half to one and one-half minutes according to the efficiency of the stain used. Methylene-blue is the weakest of the three stains mentioned; gentian-violet the strongest. (5) The excess stain is removed by washing with water. (6) The preparation is thoroughly dried by a blotter or between layers of absorbent paper. (7) A small drop of Canada balsam is placed upon the film side of the dry cover-slip, which is then inverted upon a slide. The prepara- tion is now ready for microscopical examination. 1 After Wood, " Chemical and Microscopical Diagnosis," Appendix. N. Y., 1909. 8 96 BI0LCX5Y AND TECHNIQUE The chemical principles which underlie the staining process are still more or less in doubt. ^ Suffice it to say here that most of the dyes m common use by bacteriologists and pathologists are coal-tar derivatives belonging to the aromatic series, all of them containing at least one "benzolring" combined with what Michaelis terms a "chromophore group/' chief among which are the nitro-group (NO2), the nitroso-group (NO), and the azo-group (N = N) . Just what the actual process of stain- ing consists in, is a question about which various opinions are held, some believing that the phenomenon is purely chemical, in which a salt is formed by the combination of the dye and the protoplasm of the cells, others that there is no such salt formation, and that the process takes place by purely physical means. To support the latter view it is argued that certain substances like cellulose are stainable without possessing the property of salt formation, and that staining may often be accom- plished without there being a chemical disruption of the dye itself. Michaelis sums up his views by stating that probably both processes actually take place. A dye stuff, as a whole, may enter into and be de- posited upon a tissue or cell by a process which he speaks of as " insorp- tion." In such a case the coloring matter may be subsequently ex- tracted by any chemically indifferent solvent. On the other hand, a dye after being thus deposited upon or within a cell, may become chemically united to the protoplasm by the formation of a salt, and in such a case the color can be removed only by agents which are capable of decoiv- posing salts, such as free acids. The staining power of any solution may be intensified either by heating while staining, by prolonging the staining process, or by the addition of alkalies, acids, anilin oil, and other substances which v/ilJ be mentioned in the detailed descriptions of special staining methods. One of the most common examples of such an intensified stain is the so-called Loeffler's alkaline methylene-blue. This is made up in the following way: Saturated alcoholic solution of methylene-blue, 30 c.c. 1 : 10,000 solution potassium hydrate in water, 100 c.c. Another solution designed with a similar purpose is the Koch-Ehrlich anilin-water solution. Anilin oil, one part, is shaken up with dis- tilled water, nine parts; after thorough shaking, the mixture is filtered 1 For comprehensive reviews of the subject, the reader is referred to dissertations such as those of Mann (" Physiol. Hist. Methods and Theory," Oxford, 1902) and of Michaelis (" Einfiihrung in die Farbstoffchemie," etc., Berlin^ 1902). MICROSCOPIC STUDY AND STAINING ^7 through a moist filter paper until perfectly clear. A saturated alco- holic solution of either fuchsin or gentian-violet is added to this anilin water in proportions of about one to ten or until a slightly iridescent pellicle appears upon the surface of the solution. An extremely useful and very strong staining solution is the Ziehl carhol-fuchsin solution, rnade up as follows : ^ Fuchsin (basic) ^rVxV.^-r*^.. . .yv> .>.yV^rV??^A< 1 gm. 1. Alcohol (absolute) 10 c.c. Five per cent carbolic acid 100 c.c. y**^ ^1H To make up this staining solution, mix 90 c.c. of a five per cent aque- ous solution of carbolic acid with 10 c.c. of saturated alcoholic basic fuchsin. It may also be made up as follows: Weigh out Basic fuchsin 1 gram Carbolic acid 5 grams Dissolve in Distilled water 100 c.c. Filter and add Absolute alcohol 10 c.c. SPECIAL STAINING METHODS Spore Stains. — Abbott's Method.^ — Cover-slips are smeared and fixed by heat in the usual manner. Cover with Loeffler's alkaline methylene-blue and heat the stain until it boils, repeat the heating at intervals but do not boil continuously. Keep this up for one minute. Rinse in water. Decolorize with a mixture of alcohol eighty per cent 98 c.c. and nitric acid 2 c.c, until all blue has disappeared. Rinse in water. Dip from three to five seconds in saturated alcoholic solution of eosin 10 c.c, and water 90 c.c Rinse in water, blot, and mount in balsam. By this method the spores are stained blue, the bodies of the bacilli are stained pink. 1 Ziehl, Deut. med. Woch., 1882. ^Abbott, "Prin. of Bact.," Phila., 1905- 98 BIOLOGY AND TECHNIQUE Moeller's Method.^ — Cover-slips are prepared as usual and fixed in the flame. Wash in chloroform for two minutes. Wash in water. Cover with five per cent chromic acid one-half to two minutes. Wash in water. Invert and float cover-slip on carbol-fuchsin solu- tion in a small porcelain dish and heat gently with a flame until it steams; continue this for three to five minutes. (This step can also be done by covering the coyer-glass with carbol-fuchsin and holding over flame.) Decolorize with five per cent sulphuric acid five to ten seconds. Wash in water. Stain with aqueous methylene-blue one-half to one minute. By this method spores will be stained red, the body blue. Capsule Stains. — Welch's Method.^ — Cover-slips are prepared as usual but dried without heat. Cover with glacial acetic acid for a few seconds. Pour off acetic acid and cover with anilin water gentian-violet, renewing stain repeatedly until all acid is removed. This is done by pouring the stain on and off three or four times and then finally leaving it on for about three minutes. Wash in two per cent salt solution and examine in this solution. Hiss' Methods.^ — (1) Copper Sulphate Method. — Cover-slip prepara- tions are made by smearing the organisms in a drop of animal serum, preferably beef-blood serum. Dry in air and fix by heat. Stain for a few seconds with — Saturated alcoholic solution of fuchsin or gentian-violet 5 c.c, in distilled water 95 c.c. The cover-slip is flooded with the dye and the preparati-jn held for a second over a free flame until it steams. Wash off dye with twenty per cent aqueous copper sulphate solution. Blot (do not wash). Dry and mount. By this method permanent preparations are obtained, the capsule appearing as a faint blue halo around a dark purple cell body. Huntoon's Capsule Stain (applicable only to cultures, not to animal exudates). — This depends upon the precipitating action of lactic acid on nutrose. Requires two solutions. » Moeller, Cent. f. Bakt., I, x, 1891. 2 Welch, Johns Hopkins Hosp. Bull., 1892, iii, 81. ' Hiss, Cent. f. Bakt., xxxi, 1902; Jour. Exper. Med., vi, 1905. MICROSCOPIC STUDY AND STAINING 99 1. Diluent. — 3 per cent solution of nutrose in distilled water; place in Arnold one hour, add a small amount of carbolic as preservative, and allow to settle. 2. Stain and fixative. — 2 percent carbolic, 100 c.c; concentrated lactic acid, 0.5 c.c; 1 per cent acetic acid, 1 c.c; saturated alcoholic solution basic fuchsin, 1 c.c; carbol fuchsin, 1 c.c As to the dye employed, most anything but methylene blue or Bismarck brown may be used. Methyl violet gives the most beautiful results but is not permanent and will not photograph. I have found the above mixture the best for classroom work. Make a thin film, employing solution 1 as diluent. Dry in air. Do not fix. Cover with stain 3p seconds. Wash in water, dry, and examine. Buerger's Method.* — Cover-slip preparations are made by smear- ing in serum as in Hiss' method. As the edges of the smear begin to dry, pour over it Zenker's fluid (without acetic acid) and warm in flame for three seconds. (Zenker's fluid is composed of potassium bichromate 2.5 gm., sodium sulphate 1 gm., water 100 c.c, saturated with bichloride of mercury.) Wash in water. Flush with ninety-five per cent alcohol. Cover with tincture of iodin, U. S. P., one to three minutes. Wash with ninety-five per cent alcohol. Dr}^ in the air. Stain v/ith anilin water gentian-violet two to five seconds. Wash with two per cent salt solution. Mount and examine in salt solution. Wads worth's Method.^ — Wadsworth has devised a method of staining capsules which depends upon the fixation of smears with forma- lin. After such fixation capsules may be demonstrated both with simple stains and by Gram's method. The technique is as follows : Smear preparations, made as usual, are treated as follows: 1. Formalin, 40 per cent, two to five minutes. 2. Wash in water, five seconds. Simple Stain. Differkntial Stain (Gram's Method). 3. Ten per cent aqueous gentian-violet. 3. Anilin gentian-violet, two minutes. 4. Wash water, five seconds. 4. Iodin solution, two minutes. 5. Dry, moimt in balsam. 5. Alcohol, 95 per cent, decolorize. 6. Fuchsin, dilute aqueous solution. 7. Wash water, two seconds. 8. Dry, mount in balsam. 1 Buerger, Med. News, Dec, 1904 « Wadsworth, Jour. Inf. Dis., 1906. 100 BIOLOGY AND TECHNIQUE It is important that the formalin be fresh and the exposure to water momentary. When decolorizing in the Gram method, strong alcohol only should be used. Wadsworth also found that encapsulated pneumococci could be demonstrated in celloidin sections of pneumonic lesions hardened in strong formalin. The lungs should be distended with the formalin or the lesions cut in very thin bits, hardened, dehydrated, embedded, and cut in the usual way. The celloidin sections may be fixed on the slides by partially dissolving the celloidin in alcohol and ether and setting the celloidin quickly in water before staining. Failure to obtain pneumococci encapsulated in such sections is usually due to improper or inadequate fixation in the formalin. The differential method employed by Wadsworth for tissue staining is as follows: 1. Fix in formalin forty per cent, two to five minutes. 2. Wash in water. 3. AnUin gentian-violet, two minutes. 4. lodin solution, two minutes. 5. Alcohol, ninety-five per cent, decolorize. 6. Eosin alcohol, counterstain. 7. Clear in oil of origanum. 8. Mount in balsam. Flagella Stains. — All flagella stains, in order to be successful, neces- sitate particularly clean cover-slip preparations, best made from young agar cultures emulsified in sterile salt solution. Scrupulous care should be exercised in cleaning the glassware used. Loeffler's Method.^ — The preparation is dried in the air and fixed by heat. It is then treated with the following mordant solution: Twenty per cent aqueous tannic acid 10 parts. Ferrous sulphate aq. sol. saturated at room temperature . 5 parts. Saturated alcoholic f uchsin solution 1 part. This solution, which should be freshly filtered before using, is poured over the cover-glass and allowed to remain there for on«^-half to one minute, during which time it should be gently heated, but not allowed to boil. Wash thoroughly in water. Stain with five per cent anilin water fuchsin or anilin water gen- » Loeffler, Cent. f. Bakt., I, vi, 1889. MICROSCOPIC STUDY AND STAINING 101 tian-violet made slightly alkaline by the addition of one-tenth per cent sodium hydrate. The stain should be filtered directly upon the cover-slip. Warm gently and leave on for one to two minutes. Wash in water. Mount in balsam. Van Ermengem's Method.^ — This method requires the preparation of three solutions. (1) Twenty per cent tannic acid solution 60 c.c. Two per cent osmic acid solution 30 c.c. Glacial acetic acid 4-5 drops. The cover-slip carrying the fixed preparation is placed in this solu- tion for one hour at room temperature, or for five minutes at 100° C. (boiling) . Wash in water. Wash in absolute alcohol. Immerse the cover-slip for one to three seconds in (2) Silver nitrate, 0.25-0.5 per cent solution. Without washing, transfer to (3) Gallic acid 5 gm. Tannic acid 3 " Fused potassium acetate 10 " Distilled water 350 c.c. Immerse in this for a few minutes, moving the cover-slip about. Return to the silver nitrate solution until the preparation turns black. Wash thoroughly in water. Blot and mount. Smith's Modification of Pitfield's Method.^ — A saturated solu- tion of bichloride of mercury is boiled and is poured while still hot into a bottle in which crystals of ammonia alum have been placed in quantity more than sufficient to saturate the fluid. The bottle is then shaken and allowed to cool. Ten c.c. of this solution are added to 10 c.c. of freshly prepared ten per cent tannic acid solution. To this add 5 c.c. carbol- fuchsin sohition. Mix and filter. To stain, filter the above mordant directly upon the fixed cover-slip 1 Van Ermengem, Cent. f. Bakt., I, xv, 1894. 2 .SmtV/i, Brit. Med. Jour., I, 1901, p. 205. 102 BIOLOGY AND TECHNIQUE preparation. Heat gently for three minutes, but do not allow to boil. Wash in water and stain with the following solution: Saturated alcoholic solution gentian-violet 1 c.c. Saturated solution ammonia alum 10 c.c. Filter the stain directly upon the preparation and heat for three or four minutes. Wash in water, dry, and mount in balsam. Differential Stains. — Gram's Method.^ — By this method of staining, which is extremely important in bacterial differentiation, bacteria are divided into those which retain the initial stain and those which are subsequently decolorized and take the counterstain. The former are often spoken of as the Gram-positive, the latter as Gram-negative bacteria. Preparations are made on cover-slips or slides in the usual way. The preparation is then covered with an anilin gentian-violet solu- tion which is best made up freshly before use. The staining fluid is made up, according to Gram's original direc- tions,^ as follows: Five c.c. of anilin oil are shaken up thoroughly with 125 c.c. of dis- tilled water. This solution is then filtered through a moist filter paper. To 108 c.c. of this anilin water, add 12 c.c. of a saturated alcoholic solution of gentian- violet. The stain acts best when twelve to twenty- four hours old, but may be used at once. It lasts, if well stoppered, for three to five days. A more convenient and simple method of making up the stain is as follows: To 10 c.c. of distilled water in a test tube add anilin oil until on shaking the emulsion is opaque; roughly, one to ten. Filter this through a wet paper until the filtrate is clear. To this add saturated alcoholic solution of gentian- violet until the mixture is no longer transparent, and a metallic film on the surface indicates saturation. One part of alcohoHc saturated gentian-violet to nine parts of the anilin water will give this result. This mixture may be used immediately and lasts two to five days if kept in a stoppered bottle. Cover the preparation with this; leave on for 5 minutes. Pour off excess stain and cover with Gram's iodin solution for 2 to 3 minutes. lodin. 1 gm. Potassium iodid _ . 2 gm. Distilled water 300 c.c. 1 Gram, Fortschr. d. Med., ii, 1884. 2 Oraniy loc. cit- MICROSCOPIC STUDY AND STAINING 103 Decolorize with ninety-seven per cent alcohol until no further traces of the stain can be washed out of the preparation. This takes usually thirty seconds to two minutes, according to thinness of preparation. Wash in water. Counterstain with an aqueous contrast stain, preferably Bismarck brown.i Paltauf's Modification of Gram's Stain.^ — The staining fluid as prepared by this modification possesses the advantage of retaining its staining power for a longer period than does the anilin-water gentian- violet described in the original method. The staining fluid is prepared as follows: 3-5 c.c. anilin oil are added to 90 c.c. distilled water and 7 c.c. absolute alcohol. This mixture is thoroughly shaken and filtered through a moist filter paper imtil clear. Then add: Gruebler's gentian-violet 2 gm. The fluid should stand twenty-four hours, during which a precipi- tate forms. This is filtered before use. This gentian-violet solution retains its staining power for from 4 to 6 weeks. It is good only when a metallic luster develops on the surface. It is used in the following way: Spreads on cover-slips or slides are dried and fixed as usual. Then apply: — Anilin-water gentian-violet (as above), three minutes. Gram's iodin solution, two minutes. Absolute alcohol (with stirring), thirty seconds Counterstain, without washing in water, in aqueous fuchsin or in weak carbol-fuchsin. Sterling's Modification of Gram's Method. — 2 c.c. anilin oil + 10 c.c. 95% alcohol. Shake and add 88 c.c. distilled water. 5 grams gentian violet are ground in a mortar and the anilin solution added slowly while grinding. Filter. This solution keeps, and stains in one- half to one minute. 1 To make up Bismarck brown solution, prepare a saturated aqueous solution of the powdered dye by heating. Cool and filter. Dilute 1 to 10 with distilled water. 2 iSharnosky, Proc. N. Y. Pathol. Soc, Oct., 1909, n. s., ix, 5. 104 BIOLOGY AND TECHNIQUE Classification of the Most Important Pathogenic Bacteria According to Gram's Stain. Gram-positive. (Retain the Gentian-violet.) Micrococcus pyogenes aureus Micrococcus pyogenes albus Streptococcus pyogenes Micrococcus tetragenus Pneumococcus Bacillus subtilis Bacillus anthracis Bacillus diphtherise Bacillus tetanus Bacillus tuberculosis and other acid-fast bacilli Bacillus aerogenes capsulatus Bacillus botulinus Gram-negative. {Take Counter stain.) Meningococcus Gonococcus Micrococcus catarrhalis Bacillus coli Bacillus dysenterisB Bacillus typhosus Bacillus paratyphosus Bacillus fecalis alkaligenes Bacillus enteritidis Bacillus proteus (proteus) Bacillus mallei Bacillus pyocyaneus Bacillus influenzae Bacillus mucosus capsulatus Bacillus pestis Bacillus maligni oedematis Spirillum cholersB Bacillus Koch- Weeks Bacillus Morax-Axenfeld Stains for Acid-Fast Bacteria. — These methods of staining are chiefly useful in the demonstration of tubercle bacilli. These bacteria because of their waxy cell membranes are not easily stained by any but the most intensified dyes, but when once stained, retain the color in spite of ener- getic decolorization with acid. For this reason they are known as acid- fast bacilli. The first method devised for the staining of tubercle and allied bacilli was that of Ehrlich. Ehrlich Method.* — ^This method is now rarely used. Cover-slip preparations are prepared as usual and fixed by heat. Stain with anilin water gentian-violet, hot, three to five minutes, or twenty-four hours at room temperature. ^Ehrlich, Deut. med, Woch,. 1882. MICROSCOPIC STUDY AND STAINING 105 Decolorize with thirty-three per cent nitric acid one-half to one minute. Treat with sixty per cent alcohol, until no color can be seen to come off. Counterstain with aqueous methylene-blue. Rinse in water, dry, and mount. Ziehl-Neelson Method. 1 — Thin smears are made upon cover- slips or slides. Fix by heat. Stain in carbol-fuchsin solution as given on page 97. The slide or cover-slip may be flooded with the stain, and this gently heated with the flame until it steams, or else the cover-slip may be inverted upon the surface of the staining fluid, in a porcelain dish or watch-glass, and this heated until it steams. This is continued for three to five min- utes. Decolorize with either five per cent nitric acid, five per cent sulphuric acid, or one per cent hydrochloric acid for three to five seconds. The treatment with the acid is continued until subsequent washing with water will give only a faint pink color to the preparation. Wash with ninety per cent alcohol until no further color can be re- moved. If, after prolonged washing with alcohol, a red color still re- mains in very thick places upon the smear, while the thin areas appear entirely decolorized, this may be disregarded. Wash in water and counterstain in aqueous methylene-blue for one-half to one minute. Rinse *n water, dry, and mount. By this method the tubercle bacilli are colored red, other bacteria and cellular elements which may be present are stained blue. Gabbet's Method.^ — Gabbet has devised a rapid method in which the decolorization and counterstaining are accomplished by one solu- tion. The specimen is prepared and stained with carbol-fuchsin as in the preceding method. It is then immersed for one minute directly in the following solution: Methylene-blue 2 gms. Sulphuric acid 25 per cent (sp. gr. 1018) 100 c.c. Then rinse in water, dry, and mount. This method, while rapid and very convenient, is not so reliable as the Ziehl-Neelson method. » Ziehl, Deut. med. Woch., 1882; Keelson, Deut. med. Woch., 1883. « Gahhet, Lancet, 1887. 106 BIOLOGY AND TECHNIQUE Pappenheim^s Method/ — ^The method of Pappenheim is devised for the purpose of differentiating between the tubercle bacillus and the smegma bacillus. Confusion may occasionally arise between these two microorganisms, especially in the examination of urine where smegma bacilli are derived from the genitals, and less frequently in the examina- tion of sputum where smegma bacilli may occasionally be mixed with the secretions of the pharynx and throat. Preparations are smeared and fixed by heat in the usual way. Stain with hot carbol-fuchsin solution for two minutes. Pour off dye without washing and cover with the following mixture: Corallin (rosolic acid) 1 gm. Absolute alcohol 100 c.c. Methylene-blue added to saturation Add glycerin 2 20 c.c. This mixture is poured on and drained off slowly, the procedure being repeated four or five times, and finally the preparation is washed in water. The combination of alcohol and rosolic acid decolorizes the smegma bacilli, but leaves the tubercle bacilli stained bright red. BuNGE AND Trautenroth Method.^ — This method is designed to differentiate between the tubercle and smegma bacilli. Smear and fix by heat in the usual way. Wash with absolute alcohol to remove fat. Treat with five per cent chromic acid for fifteen minutes. Wash in several changes of water. Stain with hot carbol-fuchsin for five minutes. Decolorize with sixteen per cent sulphuric acid for three minutes. Counterstain with alcoholic methylene-blue for five minutes. Wash in water, dry, and mount. By this method the tubercle bacillus remains red, the smegma bacil- lus is decolorized. Baumgarten's Method.'* — This method is recommended by the author for differentiation between the bacillus of tuberculosis and the bacillus of leprosy and depends upon the fact that the tubercle bacillus is less easily stained than Bacillus leprae. Smears are prepared and fixed by heat in the usual way. ^ Pappenheim, Berl. klin. Woch., 1898. 2 The glycerin is added after the other constituents have been mixed. 3 Bunge und Trautenroth, Fortschr. d. Med., xiv, 1896. 5 Baumgarten, Zeit. f. wissensch. Mikrosk., 1, 1884. MICROSCOPIC STUDY AND STAINING 107 Stain in dilute alcoholic fuchsin for five minutes. Decolorize for twenty seconds in alcohol, ninety-five per cent, ten parts, nitric acid one part. Wash in water. Counterstain in methylene-blue. Wash in water, dry, and mount. The tubercle bacillus should be blue and the bacillus of leprosy red. Special Stains for Polar Bodies.— These staining methods are designed to bring into view polar bodies as found, for instance, in the bacilli of diphtheria and plague. Neisser's Method.^ — Smear and fix in the usual manner. Stain for two to five seconds in the following solution : Methylene-blue 1 gm. Absolute alcohol 20 c.c. Glacial acetic acid 50 c.c. Distilled water 1,000 c.c. Wash in water. Counterstain in two per cent aqueous Bismarck brown solution for five seconds. B}' this method polar bodies are stained blue, while the bacillary bodies are stained brown. Roux's Method.^ — Two solutions are necessary. (1) Dahlia violet 1 gm. . Alcohol 90 per cent 10 c.c. Aqua destillata ad 100 c.c. (2) Methyl-green 1 gm. Alcohol 90 per cent 10 c.c. Aqua destillata ad *. 100 c.c. Before use, one part of solution No. 1 is mixed with three parts of solution No. 2. The preparation is stained with the mixture for two minutes in the cold. Polychrome Stains. — The various polychrome stains are of value to the bacteriologist chiefly for the staining of pus and exudates where the relation of bacteria to cellular elements is to be demonstrated. They are also extremely useful in the study of fixed specimens of protozoan parasites. Th9r3 is a large number of these stains in use; a few only, 1 Neisser, Zeit. f. Hyg., xxiv, 1897. ' Roux and Yersin, Annal. de I'inst. Past., 1890. 108 BIOLOGY AND TECHNIQUE however, can be given here. In principle, all these stains depend upon a combination of eosin and methylene-blue, these elements staining not only as units, but acting together in combination. One and the same solution, therefore, contains at least three elements which color the various structures of the preparation selectively. Jenner's Method.^ — This stain, because of its simplicity, is useful for routine use. It is made up as follows: Equal parts of eosin (Gruebler, " W. G.") one and two-tenths per cent aqueous solution, and methylene- blue (medicinal, Gruebler) one per cent aqueous solution, are mixed and allowed to stand for twenty-four hours. A coarse granular precipitate is formed which appears dark, with a metallic luster on its surface. This is separated by filtration and washed with distilled water until the fil- trate appears almost clear. To make up the stain 0.5 gram of the dry precipitate is dissolved in 100 c.c. of methyl alcohol. In using the stain, preparations are not fixed, but simply dried in the air and immersed in the stain for one to two minutes. After this, wash in distilled water and examine. Wright's Modification of Leishman's Method.^ — A one per cent solution of methylene-blue (Gruebler) in five-tenths per cent solution of sodium bicarbonate in distilled water is steamed in a sterilizer at 100° C. for one hour. After this has cooled, a one-tenth per cent aqueous solution of eosin (Gruebler, W. G.) is added until a metallic scum ap- pears on the surface of the mixture. (About five parts of eosin solution to one of methylene-blue is necessary.) The precipitate which forms is collected by filtration, dried, and a saturated solution then made in methyl alcohol. This is filtered and diluted with one-quarter its bulk of methyl alcohol. To stain, cover the dried preparation with the stain for one to one and one-half minutes. Dilute by dropping upon the stain distilled water from a pipette until a metallic film appears upon the top. Leave this on for three to fifteen minutes. Wash in distilled water. Giemsa's Method.^ — The method of Giemsa is really a modification of the Romano wsky method. It is widely applicable, being of great value in the staining of the Spirochsete pallida, Vincent's spirilla, pro- tozoa, and Negri bodies. The stain has been modified several times by 1 Jenner, Lancet, i, 1889. 2 Wright, Jour. Med. Research, ii, 1902. « Oiemsa, Cent. f. Bakt., I, xxxvii, 1904. MlCHOSCOPlC STUDY AND STAINING 109 its originator, the following being the formula given by him in 1904: The substance referred to as azur II and purchasable under that name, consists of pure methylenazur chloralhydrate combined with an equal quantity of methylene-blue chloralhydrate. The substance referred to as azur Il-eosin is a combination of this substance with eosin. The staining fluid is made up as follows:^ Azur Il-eosin 3 gms. Azur II 8 gms. This mixture is thoroughly dried over sulphuric acid in a desiccator^ finely powdered, and rubbed through a fine sieve. It is then dissolved in 250 gms. of C. P. glycerin (Merck), at 60° C. To this ic added methyl alcohol (Kahlbaum) 250 c.c, previously warmed to 60° C. This mix- ture is well shaken and allowed to stand at room temperature for twenty-four hours. The mixture is now ready for use. For use 10 c.c. of distilled water are poured into a test tube and one to two drops of a one per cent potassium carbonate solution are added. Ten drops of the staining solution described above (one drop to the c.c.) are mixed with this slightly alkaline water. The preparation which is to be stained is fixed in methyl alcohol, dried, and covered with the diluted staining solution. For the staining of protozoa and ex- udates containing bacteria, ten to fifteen minutes are sufficient. For the staining of Negri bodies or Spirochete pallida, one or more hours of staining should be employed. After staining, wash in running tap water and blot. Wood's Method.^ — Wood has devised a simple staining method based on the principles of the Giemsa stain, in which azur II and eosin may be used in separate solutions. Preparations are fixed in strong methyl alcohol for five minutes and are then stained in a 0.1 per cent aqueous solution of eosin until the preparation is pink. The eosin is then poured off and the preparation is covered with a 0.25 per cent aqueous solution of azur II for one-half to two minutes. Following this, it is washed in tap water and dried by blotting. When an intense stain is desired, the solution of eosin and azur II may be flooded over the preparation together, using an excess of azur II. They are then left on from five to ten minutes. At the end of this time washing and drying as before completes the process. » It is best not to attempt to make up the undiluted staining fluid, since this is purchasable under the name of " Giemsa Losung f iir Romanowsky Farbung." 2 Wood, Med. News, 83, 1903. 110 BIOLOGY AND TECHNIQUE The Staining of Bacteria in Tissues. -^The preparation of tissue for bacterial staining is, in general, the same as that employed for purposes of cellular studies, in histology. For bacteriological studies the most useful fixative is alcohol; other fixations, such as that by formalin, Zenker's fluid, or Mueller's fluid, give less satisfaction. In other respects the details of dehydration and embedding are the same as those used in histological studies, except that it is desirable that the tissues should be handled rather more carefully than is necessary for ordinary patholog- ical work, and the changes from the weaker to the stronger alcohols should be made less abruptly.^ Embedding in paraffin is preferable to celloidin, although the latter method is not unsuccessful if carefully carried out. The chief disadvan- tages of celloidin are the retention of color by the celloidin itself and the consequent unclearness of differentiation. It is also easier to cut thin sections from paraffin blocks than from those prepared with celloidin. When staining tissue sections for bacteria, it is most convenient to carry out the process with the section attached to a slide. For cel- loidin sections this may be accomplished by means of ether vapor. For paraffin sections it is necessary to cover the slide with an extremely thin layer of a filtered mixture of equal quantities of egg albumin and glycerin, to which a small crystal of camphor or a drop or two of carbolic acid has been added. The sections are then floated upon a slide so prepared, and set away in the thermostat for four or five hours. Loepfler's Method.^ — Stain in alcoholic methylene-blue solution five to fifteen minutes, or in Loeffler's alkaline methylene-blue solution one to twenty-four hours. Wash in one to one -thousand acetic acid solution for about ten seconds. Treat with absolute alcohol by pouring the alcohol over the prepara- tion for ten to twenty seconds. Clear with xylol. Mount in balsam. When celloidin sections are stained in this way ninety-five per cent alcohol should be substituted for the absolute. A number of other staining solutions may be used in the same way, aqueous fuchsin or aqueous gentian-violet yielding good result. » For details of such work reference should be had to the standard textbooks on pathological technique, notably the very excellent one of Mallory and Wright. 2 Loejffler, Mitt. a. d. kais. Gesundheitsamt, ii. 1884. MiCROSCOiPIC STUDY AND STAINING 111 NicoUe advises the use of a ten per cent aqueous solution of tannic acid for a few seconds after washing with the acetic acid. Sections may also be stained by placing them over night into a dilute Giemsa solution (one drop to each c.c. of distilled water). When so stained the sections must not be run through the weaker alcohols but must be rapidly differentiated in absolute alcohol. Method of Staining Gram-Positive Bacteria in Tissue Sections. — Celloidin Sections. — After fixing section to the slide by pressure with a filter paper or by ether vapor, cover with anihn-water gentian-violet five minutes. Pour off excess of stain and cover with Gram's iodin solution for two minutes. Decolorize with ninety-five per cent alcohol until no more color comes out. Stain quickly with eosin-alcohol (ninety-five per cent alcohol to which enough eosin has been added to give a transparent pink color: about 1 : 15). Clear in eosin-oil of origanum (oil of origanum, 25 c.c. and eosin alcohol, as above, about 3 c.c). Blot and mount in balsam. Paraffin Sections. — Stain with anilin-water gentian-violet five to ten minutes. Wash in water. Cover with Gram's iodin solution one minute. Wash in water. Decolorize with absolute alcohol until no more color comes out. Clear in xylol. Mount in balsam. Gram-Weigert Method.^ — (For celloidin sections.) — Stain for one-half hour in the following freshly filtered solution: Carmine * 3-5 grams. Saturated aqueous solution of lithium carbonate 100 c.c. Dehydrate in ninety-five per cent alcohol. Stick section to slide with ether vapor. Stain in anilin-water gentian-violet for five to fifteen minutes (or in a saturated solution of aqueous crystal violet diluted with water one to ten, five to fifteen minutes) . Wash in physiological salt solution. Cover with Gram's iodin solution one to two minutes. Wash in water and blot. 1 Wmgert, Fortschr. d. Med., v, 1887. 112 BIOLOGY AND TECHNIQUE Decolorize with anilin oil until no more color comes off. This both decolorizes and dehydrates. Treat with xylol. Mount in balsam. Method of Staining for Tubercle Bacilli in Sections.^ — Paraffin Sections. — Stain in carbol-fuchsin solution hot for five minutes (or better cold, for twenty-four hours) . Wash in water. Decolorize and counterstain in Gabbet's methylene-blue sulphuric acid mixture for one minute. , Wash in water. Dehydrate in absolute alcohol. Clear in xylol. Mount in balsam. Celloidin Sections.'^ — Stain lightly in alum hematoxylin. » Wash in water. | Dehydrate in ninety-five per cent alcohol. I Attach the sHde by ether vapor. . ? Stain with steaming carbol-fuchsin two to five minutes. \ Wash in water. • Wash with Orth's acid alcohol (alcohol ninety per cent., 99 c.c; 1 cone. HCl, 1 c.c.) one-half to one minute. ' Wash in water several changes. ' Treat with ninety-five per cent alcohol until red color is entirely gone. ^ ^ ^ ^ I Blot and cover with xylol until clear. Mount in balsam. Method of Staining Actinomyces in Sections. ^Ma//or?/'s Method^. — 1. Stain deeply in saturated aqueous eosin ten minutes, 2. Wash in water. 3. Anilin gentian-violet two to five minutes. 4. Wash in normal saline solution. 5. Weigert's iodin solution (iodin 1, KI 2, and water 100 parts) one minute. 6. Wash in water and blot. 7. Clear in anilin oil. 8. Xylol several changes. 9. Mount in balsam. » Mallory and Wright, " Pathol. Tech.," p. 413. 2 After Mallory and. Wright. » Mallory and Wnght, " Pathol. Tech.," 1904 CHAPTER VII THE PREPARATION OF CULTURE MEDIA GENERAL TECHNIQUE The successful cultivation of bacteria upon artificial media requires the establishment of an environment which shall be suitable in regard to the presence of assimilable nutritive material, moisture, and osmotic relations. These requirements are fulfilled in the composition of the nutrient media described in another section, media which are to some extent varied according to the special -requirements of the bacteria which are to be cultivated. If cultivation, furthermore, is to have any value for scientific study of individual species, it is necessary to ob- tain these species free from other varieties of microorganisms, that is, in pure culture, and to protect such cultures continuously from con- tamination with the other innumerable species which are everywhere present. The technique which is employed for these purposes has been gradu- ally evolved from the methods originally devised by Pasteur, Koch. Cohn, and others. Bacterial cultivation is carried out in glassware of varied construc- tion, the forms most commonly employed being test tubes of various sizes, Erlenmeyer flasks, the common Florence flasks, and Petri dishes. All glassware, of course, must be thoroughly cleansed before being used. Preparation of Glassware. — The cleansing of glassware may be ac- complished by any one of a number of methods. New glassware may be immersed in a one per cent solution of hydrochloric or nitric acid in order to remove the free alkali which is occasionally present on such glass. It is then transferred to a one per cent sodium hydrate solution for a few hours, and following this is washed in hot running water. In the case of old glassware which has contained culture media, sterilization in the autoclave is first carried out, then the glassware is boiled in five per cent soda solution or in soapsuds. After this, thorough mechanical cleansing is practiced, and the glassware may be treated by^ acid and alkali followed by running water, as given above. These last 113 114 BIOLOGY AND TECHNIQUE steps, however, are not essential, thorough washing in hot water after the soapsuds or soda solution being usually sufficient to yield good results. Other workers have recommended immersion of the glassware after mechanical cleansing in five per cent to ten per cent potassium bichromate solution in twenty-five per cent sulphuric acid. This is followed by thorough washing in hot running water, and drying. Clean flasks and test tubes are then stoppered with cotton, which has been found to be a convenient and efficient seal against the bacteria of the air, catching them in the meshes of the fibers as in a filter. The technique of the stoppering or plugging of glass receptacles is important. Fig. 15. — Florence Flask. Fig. 16. — Erlenmeyer Flask. in that, when poorly plugged, sterility is not safeguarded, and the pur- pose of culture study is defeated. In almost all laboratories in this country non-absorbent cotton or " cotton batting " is used for the plug. In a few of the German labora- tories the absorbent variety is employed. The disadvantages of the latter, especially in the case of fluid media, are obvious. The plugs should fit snugly, but not so tightly that force is necessary to remove them. Care should be taken, furthermore, that no creases are left be- tween the surface of the glass and the periphery of the plug; for these, if present, may serve as channels for the entrance of bacteria. Fig. 18, accompanying, will illustrate some of the more common and un- desirable defects in poorly made plugs. The plugging itself is carried out by tearing a small piece of cotton, about 2 X 2 inches, from the roll, THE PREPARATION OP CULTURE MEDIA 115 folding over one of its corners, and, applying the smooth end of a "glass rod to the folded portion, gently pushing it into the mouth of the tube. After plugging and before media are introduced into the tubes and flasks, these should be steriHzed. This is best done in one of the " hot- air sterilizers" (see Fig. 8, p. 69), by exposing the tubes for one hour to a temperature of 150° C. If greater speed is desired exposure to 180° to 190° C. for half an hour is usually safe. If by mistake, however, the temperature is allowed to rise above 200° C, a browning of the cotton plugs occurs and the glassware is apt to be stained by the burning of the fat and other organic material derived from the cotton. Petri dishes ce Fig. 17. — Petri Dish. m after cleansing are fitted together in the manner shown in Fig. 17, and are sterilized in the hot-air chamber at 150° C. for one hour. Glassware so prepared is ready for the reception of media. Ingredients of Culture Media. — The food requirements of bacteria have been discussed in another section. From what has there been said, it is apparent that artificial culture media must, to a certain extent, be adjusted to the peculiarities of individual bacteria. In the cases of the more strictly parasitic microorganisms growth can be obtained only by the most rigid observance of special requirements. For the large majority, of pathogenic bacteria, however, routine or standard media may be employed, which, while slightly more favorable for one species than for another, are sufficiently general in their composition to per- mit the growth of all but the most fastidious varieties. The basis of many of our common media is formed by the soluble constituents of meat. These substances are best obtained by macerating 500 grams of lean beef in 1,000 c.c. of distilled water. The mixture i^ 116 BIOLOGY AND TECHNIQUE K^ V.N. allowed to infuse in the ice chest over night, and then strained through cheese-cloth. To this infusion are added the other required constituents m the manner given in the detailed instructions below. The soluble constituents of meat, however, may also be procured in a simpler way by the use of the commercial meat extracts, such as that of Liebig. These extracts are dissolved in quantities of five grams to the liter, and other constituents are added to this nutrient basis. Though simpler to make, the meat-ex- tract media are less favorable for the culti- vation of the more delicate organisms than are the media made directly from fresh meat. Nevertheless, they suffice for the cultivation of the large majority of the more saprophytic pathogenic microorganisms and hold an im- portant place in laboratory technique. The ingredients and methods used in va- rious laboratories in the preparation of such standard media should be, as much as pos- sible, uniform, in order that confusion in re- sults may be avoided ; for, as is well known, the biological characteristics of one and the same bacterial species may var}' considerably if grown on media differing in their compo- sition. A committee of the American Public Health Association,^ appointed in 1897 for the sake of standardizing the methods of preparation of media, recom- mended that the following rules should govern the choice of ingredients: 1. Distilled water should be used in all cases. 2. The meat used should be fresh, lean beef (when veal or chicken is substituted the change should be stated). 3. The pepton used should be Witte's pepton, dry, made from meat. 4. Only C. P. NaCl should be used. 5. For alkalinizing C. P. sodium hydrate should be used in normal solutions. b Fig. 18.— Test Tube (a) incorrectly stoppered; (b) correctly stoppered. ^Rep. Com. of Amer. Bact. to Com. of Amer. Pub. Health Assn. Meeting, Philadelphia, Sept., 1897. THE PREPARATION OF CULTURE MEDIA 117 6. For acidification C. P. hydrochloric acid in normal solution should be used. 7. When glycerin is used, this should be of the redistilled variety. 8. The agar-agar employed should be of the finest grade of commer- cial thread agar. 9. The gelatin should be the commercial sheet gelatin washed as free as possible of acid and impurities. 10. Chemicals and carbohydrates which are used should be as nearly chemically pure as possible. Titration of Media. — Next in importance to the actual composi- tion of media is the adjustment of their reaction. Bacteria are highly susceptible to variations in the acidity and alkalinity of media, excessive degress of either may completely inhibit development or moderate variations may lead to marked modifications of cultural characteristics. It is necessary, therefore, to adjust the reaction both for the sake of favoring growth and in order to insure uni- formity of growth characters. This is accomplished by titration which is best carried out according to the recommendations of the committee mentioned above. The color indicator employed for* the titration is a five-tenths per cent solution of phenolphthalein in fifty per cent alcohol. The chief advan- tage of this indicator over others is due to the fact that it indicates the presence of organic acid and acid compounds in its reaction. For actual titration ^ (^ normal) solutions of sodium hydrate or of hy- drochloric acid are used. Since media in the process of preparation are usually acid, the NaOH solution is the one most frequently needed. F**^ c.c. of the medium to be tested is measured accurately in a care- FiG. 19. — Burette for Titrating Media. 118 BIOLOGY AND TECHNIQUE fully washed pipette and transferred into a porcelain evaporating dish, to this are added 45 c.c. of distilled water. The mixture is thoroughly boiled for three minutes oVer a free flame. The boiling drives off CO2, giving the true neutral point, and approximates the conditions prevaiUrig during the further sterilization of the medium from which the 5 c.c. have been taken. After boiling, 1 c.c. of the phenolphthalein is added. If the medium is acid, no color is present; if alkaline, a pink or red color appears. The -^ alkali or acid solution is allowed to drop into the dish from a graduated burette. When the neutral point is approached in an acid solution, each drop of sodium hydrate added brings forth at first a deep red, which, however, upon slight stir- ring with a clean rod, completely disappears.^ The end reaction is reached when a faint but clear arid distinct pink color remains in the fluid after stirring. When titrating alkaline media^ the addition of the phenolphthalein produces a red color in the hot medium which gradtially fades upon the addition of |^ " HCl, bdcbming colorless at the end point of titration. Titration should be done quickly and in a hot solution. From the result of the titration the computation for the neutralization of the entire bulk of the rriedium can be made by a simple arithmetical process as illustrated in the following example: Let us suppose that we have used: Fig. 20. — Tubing 'Media. 2.5 c.c. of ^ NaOH to neutralize N T N then 2.5 c.c. of and 25 c.c. of NaOH will neutralize 5 c.c. of the medium, 100 c.c. " NaOH will neutralize 1,000 c.c, or one liter. See standard textbooks on volumetric analysis. THE PREPARATION OF CULTURE MEDIA 119 /:?M; 'f'o'fy" ."> . m The adjustment of the reaction of media is largely determined by the particular uses for which the media are designed. For examinations iii the practice of sanitation, such as analyses of water, ice, and milk, etc., the American Public Health Association recommends a standard reac- tion of + 1 per cent (the plus sign is used to indicate acidity, the minus alkalinity; + 1 per cent is the expression used to indicate that one per per cent of ^ sodium hydrate solution would be required to neutralize the medium or 10 c.c. to the Uter). For general work with pathogenic bacteria, the most favorable reaction for routine media is slight alka- Hnity, neutrality, or an acidity not exceeding + 1 per cent. Methods of Clearing Media. — Clearing with Eggs. — When culture media are prepared from substances containing no coagulable proteid, it is often necessary, for purposes of clearing, to add the whites of eggs, and then to heat for forty-five min- utes in the Arnold sterilizer. In the following detailed descriptions, the direction "clear with egg" has been given whenever such a step is deemed necessary. The exact tech- nique of such a procedure is as follows : In a small pot or pan, the whites of several eggs (one or two eggs to each liter of medium) are beaten up thoroughly with a little water (20 c.c). This egg white is then poured into the medium, which, if hot, as in the case of melted agar or gelatin, must first be cooled to about 50° to 55° C. The mixture is then thoroughly shaken and steamed in the Arnold sterilizer for thirty minutes. At the end of this time the flask con- taining the medium is removed from the sterilizer and thoroughly shaken so as completely to break up the coagulum which has formed. It ia then replaced and allowed to steam for another fifteen minutes. At the end of this time the medium between the coagula should be clear. It is now. ready for filtration through cotton. /■^A a be Fig. 21. — Media in Tubes: a, broth; b, agar slant; c, potato. 120 BIOLOGY AND TECHNIQUE Filtering Media through Cotton. — ^The filtration of media after clearing, either by the addition of eggs or by the coagulation of the pro- teids originally contained in it, is best done through absorbent cotton. A small spiral, improvised of copper wire, is placed as a support in the bottom of a large glass funnel. A square piece of absorbent cotton ie Fig. 22. — Berkefeld Belter. then split horizontav/ giving two squares of equal size. Ragged edges and incisures shou'd be avoided. These two layers of cotton are then placed in the funnel, one piece above the other in such a way that the direction of the fibers of the two layers is at right angles one td the other. They are then gently depressed into the filter with the closed fist. Th*^ THE PREPARATION OF CULTURE MEDIA 121 edges of the cotton are made to adhere to the sides of the funnel by allowing a thin stream of tap water to run over them, while smoothing them against the glass with the hand. The medium, when poured into such a filter, should be poured along a glass rod at first, to avoid running down the sides or bursting the filter. After filtration has begun, the filter should be kept as full as possible. The first liter or so which comes through may not be clear, but the filter gains in efficiency as the coag- ulum settles into the fibers of the cotton, and the first yield may be sent through a second time. Filtration of agar or gelatin is best done in a warm room with windows and doors closed, and the filter covered with a fid, to avoid too rapid cooling. The funnel and filter should be warmed just before use. Filtering through Paper. — Many media may be efficiently cleared by filtration through close filter paper without the aid of coagula. The Tubing of Media. — Most of the media described in the foregoing section are used in test tubes. In order to fill these tubes, the media are best poured into a large glass funnel to which a glass discharging tube has been fitted by means of a short piece of rubber tubing (see Fig. 20) . Upon this is placed a thumb cock. The plug is then re- moved from the test tube by catching it be- tween the small and ring fingers of the right hand and the glass outlet is thrust deeply into the test tube, in order to prevent the medium from touching the upper portion of the test tube where the cotton plug will be lodged. About 7 to 8 c.c. is put in each test tube. Sterilization of Media. — By Heat. — Media which contain neither sugars, gelatin, glycerin, nor animal serum may be sterilized in the auto- clave at fifteen pounds pressure for fifteen minutes to half an hour. Media which contain these or other substances subject to injury from the high temperature, must be sterilized by the fractional method, Fia. 23. — Berkeffld Filter. 122 BIOLOGY AND TECHNIQUE i.e., by twenty minutes' exposure in the live steam sterilizer (Arnold, Fig. 9, p. 70) on each of three consecutive days. During the intervals between sterilizations, they should be kept at room temperature or in the incubator, to permit the germination of spores which may be present. Media containing animal serum or other albuminous solutions which are to be sterilized without coagulation, may be sterilized in wate baths, or in hot-air chambers (Fig. 10, p. 71), at temperatures varying Fig. 24. — Reichel Filter. from 60° to 70° C, by the fractional method. In such cases five or six exposures of one hour on succeeding days should be employed. By Filtration. — It is often desirable in bacteriological work to free fluid from bacteria. This is frequently necessary for the sterilization of blood-serum or exudate fluids, or for obtaining toxins free from bac- teria. For these purposes a large variety of filters are in use. Those most commonly employed are of the Chamberland^ or Berkefeld type, which consist of hollow candles made of unglazed porcelain or dia- tomaceous earth. Both these types are made in various grades of fine- ness, upon which depend both the speed of filtration and the efficiency. They are made in various forms and models, some of which are shown 1 Pasteur and Chamberland, Compt. rend, de Tacad. des sci., 1884. THE PREPARATION OF CULTURE MEDIA 1^ in the accompanying figures. In most of the methods of filtration commonly employed the fluid which is to be filtered is sucked through the walls of the filter, either by a hand suction-pump or by some form of vacuum-pump attached to an ordinary water-tap. The hollow candle-filter may either be firmly fitted into a cylin- drical glass chimney and surrounded by the fluid which is to be filtered, or else the candle may be connected to the collecting flask with sterile rubber tubing and suspended freely in the fluid. Perfect filters of these types will hold back any of the bacteria known to us at present. Filters before use must be sterilized. The candles themselves are subjected to 150° C. in the hot-air sterilizer for one hour. The glassware and washers necessary for setting up the apparatus may be sterilized by boiling. In order that filters may be re- peatedly used with good result, it is neces- sary that they should be carefully cleaned from time to time. This is best done in the following way: Filters through which fluids from living cultures have passed are first sterilized in the Arnold steam sterilizer. Their exterior is then carefully cleaned with a fine brush. Following this a five-tenths per cent solu- tion of potassium permanganate is passed through them and this again removed by sucking through a five per cent solution of bisulphite of soda. This last is washed out by sending a considerable quantity of dis- tilled water through the filter, which is then dried and sterilized by heat. The suction necessary for filtration through these filters is usually applied by means of the ordinary suction-pump attached to a running faucet. Slanting of Media. — Solid media which are to be used in slanted form in test tubes should be inclined on a ledge (easily improvised of glass tubing) at the proper slant, after the last sterilization. Agar, the medium Fig. 25. — Kitasato Filter. 124 felOLOGY ANt) TECHNIQUE most frequently employed in this way, should be left in this position for two or three days. (See Fig. 21, b.) ACTUAL STEPS IN THE PREPARATION OF NUTRIENT MEDIA Broth. — Meat Extract Broth. — 1. To 1,000 c.c. of distilled or clear tap water add 5 gms. Liebig's meat extract, 10 gms. Witte's pepton, and 5 gms. common salt (NaCl). 2. Weigh solution with containing vessel (any suitable agate-ware vessel or glass flask will do). 3. Heat over free flame until thor- oughly dissolved, stirring constantly. 4. Weigh again and make up loss by evaporation. 5. Determine volume. 6. Titrate and adjust to required reaction, heating over free flame for five minutes. 7. Filter through paper until clear. 8. Sterihze. If medium can not be cleared by filtering through paper, clearing by white of egg msLj be resorted to and the medium filtered through cotton. Meat Infusion Broth. — 1. Infuse 500 gms.^ of lean meat, twelve to twenty-four hours, with 1,000 c.c. of distilled water in refrigerator. 2. Strain through wet cotton flan- nel or wet cheese-cloth and make up volume to 1,000 c.c. 3. Add 5 gms. common salt and 10 gms. Witte's pepton. 4. Weigh with containing vessel. 5. Warm over flame or water bath, stirring until pepton is dissolved, not allowing temperature to rise above 50° C. 6. Determine volume. Fig. 26. — ^Maassen Filter, for SMALL Quantities of Fluid. Roughly, 1 pound (1^ lb.). THE PREPARATION OF CULTURE MEDIA 125 7. Titrate and adjust reaction to neutral. 8. Heat in Arnold steriliser for thirty minutes; shake or stir well and heat again for fifteen minutes. 9. Determine weight and restore loss by evaporation. 10. Determine volume, titrate, and adjust reaction to desired point (usually one per cent acid). 11. Heat again for five minutes if adjustment of reaction has been necessary. 12. Filter through absorbent cotton, passing the filtrate through the same filter until clear. 13. Titrate and record the final reaction. Place in cotton-plugged sterile flasks or plugged sterile test tubes, and sterilize for thirty minutes in the Arnold sterilizer on three suc- cessive days, leaving at room temperature in the intervals. Sugar-Free Broth. — 1. Make 1 liter of meat infusion broth, following steps 1, 2, 3, 4, 5, 6, 7, and 8\* then filter through thin cotton filter to remove gross particles — total' clearing is not necessary. 2. Put the broth in a flask and cool. Then add 10 c.c. of a twenty- four-hour broth culture of B. coli communis. 3. Place the flask, stoppered with cotton, in the incubator at 37° C. for eighteen hours. (The bacteria will ferment and thus destroy any sugar [monosaccharid] which may be present in the broth, and thus render the broth sugar-free and acid.) 4. Heat thoroughly to kiU the bacteria. 5. Determine weight and bring to 1,015 gms. Then determine volume and titrate, and adjust to neutral. Heat thoroughly again. 6. Filter through filter paper until clear. 7. The pure sugars, dextrose, lactose, saccharose, etc., are then added to separate portions (250 c.c.) of the broth in the proportion of one per cent. 8. When the sugars are dissolved, tube the broth immediately in fermentation tubes, and sterilize by discontinuous sterilization, never heating over twenty minutes at a time, as heat tends to destroy or change the sugars. Glycerin Broth. — To ordinary, slightly acid or neutral meat in- fusion broth, add six per cent of C. P. glycerin. Sterilize by frac- tional method. * These steps refer to the regular directions for making infusion broth. Odlq liter of previously made ijof usion broth may be used instead. '™ 126 BIOLOGY AND TECHNIQUE Calcium Carbonate Broth. — This medium is designed for obtaining mass cultures of pneumococcus or strepttococcus for purposes of im- munization or agglutination. To 100 c.c. of meat infusion broth in small flasks, add one per cent of powdered calcium carbonate, and one per cent of glucose. It is a wise precaution to sterilize the dried calcium carbonate in the hot-air chamber before using. Small pieces of marble may be used as sug- gested by Bolduan. Pepton-Salt Solution (Dunham's solution): 1. Distilled water 1,000 c.c. Pepton (Witte) 10 gms. NaCl 5 " 2. Heat until ingredients are thoroughly dissolved. 3. Filter through filter paper until perfectly clear. 4. Tube twenty-five tubes, and store remainder in 250 c.c. flasks. Sterilize by discontinuous method. Nitrate Solution. — 1. Distilled water 1,000 c.c. Pepton 10 g;m.<. Potassium nitrate 0.2 " 2. Heat until ingredients are thoroughly dissolved. 3. Filter through filter paper until perfectly clear. 4. Tube twenty-five tubes, and store remainder in 250 c.c. flasks. Sterilize by discontinuous sterilization. UscMnsky^sProteid-Free Medium} — To one liter of distilled water add: Asparagin 3.4 grams. Ammonium lactate , 10 " Sodium chloride 5 . " Magnesium sulphate 0.2 " Calcium chloride . 0.1 " Potassium phosphate 1.0 " When these substances are thoroughly dissolved, add 40 c.c. of glycerin Tube and sterilize. Gelatin. — Meat-Extract Gelatin.^1. To 1,000 c.c. of distilled water add Liebig's extract 5 gms., pepton 10 gms., NaCl 5 gms., and 120 gms. of the finest French sheet gelatin.^ 1 Uschinsky, Cent. f. Bakt., 1, xiv, 1*893. 2 The acidity and consistence of the different commercial gelatins vary con- siderably and care should be taken in selecting a uniform and suitable brand, such as Hesterberg's gold label gelatin. It is advisable, when working during the summer or in hot climates, to add 130 instead of 120 grams. THE PREPARATION OF CULTURE MEDIA 127 2. Weigh with containing vessel. 3. Dissolve by warming. 4. Adjust weight, determine volume, titrate, and adjust reaction. 5. Cool to 60° C, add whites of two eggs, and stir thoroughly. 6. Heat for thirty minutes, stir thoroughly, and heat for fifteen minutes. 7. Adjust weight. 8. Filter through cotton. 9. SteriHze. Meat-Infusion Gelatin. — 1. Infuse 500 gms. lean meat twelve to twenty-four hours with 1,000 c.c. of distilled water in refrigerator, 2. Strain through wet cotton flannel or wet cheese-cloth and make up volume to 1,000 c.c. 3. Add 5 gms. common salt, 10 gms. Witters pepton, and 120 gms. of the finest French sheet gelatin. 4. Weigh with containing vessel. 5. Warm over flame or water bath, stirring till pepton and gelatin are dissolved and not allowing temperature to rise above 50° C. 6. Determine volume. 7. Titrate and adjust reaction to neutral. 8. Heat in Arnold sterilizer for thirty minutes; shake or stir well and heat again for fifteen minutes. 9. Determine weight and restore loss by evaporation. 10. Determine volume, titrate, and adjust reaction to desired point, if necessary (one per cent acid). 11. Heat five minutes over free flame, constantly stirring, if ad- justment of reaction has been necessary. 12. Filter- through absorbent cotton, passing the filtrate through the same filter until clear. 13. Titrate and record the final reaction. Place gelatin in cotton-plugged sterile 250 c.c. flasks or about 8 c.c. in plugged sterile test tubes and sterilize for thirty minutes in the Arnold sterilizer on three successive days, leaving at room temperature in the in- tervals. Never heat the gelatin for longer than is necessary to comply with directions, or it may not be solid enough for use. With some brands of gelatin it may be necessary to add thirteen per cent in order to obtain sufficient stiffness. Agar. — Meat-Extract Agar. — 1. To 1,000 c.c. of distilled water (or tap water) add 15 gms. of thread agar, 10 gms. of Witte's pepton, and 5 gms. of Liebig's meat extract, and 5 gms. of common salt. 10 128 BIOLOGY AND TECHNIQUE 2. Weigh with containing vessel. 3. Heat over free flame until agar is dissolved, thirty to forty-five minutes. (Great care should be exercised in determining that agar is completely in solution.) 4. Determine weight and make up loss by evaporation. 5. Determine volume, titrate, and adjust to desired reaction. 6. Cool to 60° C. 7. Add whites of two eggs and stir thoroughly. 8. Heat in Arnold sterilizer thirty minutes, stir, and reheat fifteen minutes. 9. Weigh and make up loss by evaporation. 10. Determine volume, titrate, and correct reaction if necessary.* 11. Heat for five minutes, if reaction is corrected. 12. Filter through cotton, tube, and sterilize. Meat-Infusion Agar? — (A) 1. Infuse 500 gms. lean meat twelve to twenty-four hours in 500 c.c. of distilled water in refrigerator. 2. Strain through wet cotton flannel or wet cheese-cloth, and make up voiiime to 500 c.c. 3. Add 10 gms. of Witte's pepton and 5 gms. of common salt. 4. Weigh solution and containing vessel. 5. Warm over free flame or water bath till pepton and salt are dis- solved, not allowing temperature to rise above 50° C. 6. Determine volume, titrate, and neutralize. (B) 7. Add 15 gms. of thread agar to 600 c.c. of distilled water and boil over free flame for thirty to forty-five minutes, watching and stirring constantly, or preferably heat in autoclave till agar is completely dis- solved. This will lose weight by evaporation ; final weight should be 515 gms. 8. Cool this to about 60° C. (C) 9. Then to the solution A of meat infusion (at 50° C.) add the solution B of agar (at 60° C). 10. Determine volume, titrate, and adjust reaction to plus one per cent acid or any desired reaction. 11. Heat for thirty minutes in Arnold sterilizer or autoclave. Shake or stir thoroughly, and heat fifteen minutes more. Adjust weight by adding water. * While titrating, care should be taken that medium does not solidify along sides of vessel. Agar may be made more quickly and successfully in autoclave. 2 Glycerin agar is made by adding 6 per cent of C. P. glycerin to meat-extract or meat-infusion agar. THE PREPARATION OF CULTURE MEDIA 129 13. Filter through absorbent cotton, passing the filtrate through the same filter until clear. 14. Titrate and record final reaction. Place agar in cotton-plugged sterile flasks or plugged sterile test tubes and sterilize for thirty minutes on three successive days. Lactose-Litmus-Agar (Wurtz). — 1. Put 1,500 c.c. distilled water in previously weighed agate-ware vessel. 2. Add 15 gms. thread agar and boil over free flame for thirty to forty-five minutes, watching and stirring constantly till the agar is completely dissolved. 3. Add 5 gms. Liebig's extract of meat, 5 gms. NaCl, 10 gms. Witte's pepton, and dissolve completely. 4. Restore loss by evaporation to 1,035 gms. 5. Determine volume, titrate, and adjust reaction to one per cent acid. 6. Place in a flask and cool to 60° C. 7. Add the whites of two eggs beaten up in 50 c.c. of water and mix thoroughly. 8. Heat for thirty minutes in Arnold sterilizer, shake thoroughly, and heat again for fifteen minutes. 9. Adjust weight. 10. Filter through absorbent cotton to clear. 11. Add two per cent pure lactose (milk sugar).* 12. Add enough pure five per cent litmus solution ^ to bring to purple color when cold. 13. Tube and sterilize. WelcNs Modification of Guamieri's Medium.^ — This medium is made on a meat-infusion basis, according to the directions given for the prep- aration of meat-infusion agar. It contains 5 grams of agar, 80 grams of gelatin, 5 grams of NaCl, and 10 grams of pepton to one liter. It should * Add lactose and litmus to 250 c.c. for 25 tubes; keep the remainder, with- out lactose, stored in small sterile flasks for further use. 2 The litmus solutions used in the preparation of media are best made up as fol- lows: Litmus in substance — Merck's purified, or Kaulbaum's — is dissolved in water to the extent of 5 per cent. The solution is made by heating in an Arnold sterilizer for about one to two hours, shaking occasionally. The solution is then filtered through paper and sterilized. It should be kept sterile, as molds will grow in it otherwise, A standard litmus solution, which is marketed for laboratory purposes, known as "Kubel and Tiemann's" solution, may be used. ' Welch, Bull. Johns Hopkins Hosp. 130 BIOLOGY AND TECHNIQUE be adjusted to a neutral reaction. It is used for stab cultures and is designed chiefly for pneumococcus cultivation and storage. Dorsett Egg Medium — This medium is chiefly useful for the cul- tivation of tubercle bacilli. 1. Carefully break eggs and drop the contents into a wide-mouthed flask. Break up the yolk with a sterile platinum wire, and shake up the flask until the whites and yolks are thoroughly mixed. 2. Add 25 c.c. of distilled water to every four eggs; strain through sterile cloth. 3. Pour 10 c.c. each into sterile test tubes and slant in an inspissator and expose to 73° C. for four to five hours on two days. 4. On the third day, raise the temperature to 76° C. 5. The sterilization may be finished by a single exposure to 100° C. in the Arnold sterilizer for fifteen minutes. Before inoculation, add two or three drops of sterile water to each tube. For a description of Petroff's medium for the isolation of tubercle bacilli, see page 484. Potato Media. — Large potatoes are selected, washed in hot water, and scrubbed with a brush. They are peeled, considerably more than the cuticle being removed The peeled potatoes are washed in running water, following which cylindrical pieces are removed with a large apple corer. The cylinders are cut into wedges. Since the reaction of the potato is normally acid, this should be cor- rected by washing the pieces in running water over night, or, better, by immersing them in a one per cent solution of sodium carbonate for half an hour. The pieces are then inserted into the large variety of test tubes known as "potato tubes." (See Fig. 21, c.) In the bottom of the tubes a small amount of water (about 1 c.c.) or a small quantity of moist absorbent cotton should be placed in order to retard drying out of the potato. The tubes are sterilized by fractional sterilization, twenty minutes to half an hour in the Arnold sterilizer on three suc- cessive days. Glycerin Potato. — In preparing glycerin potato the potato wedges are treated as above, and are then soaked in a ten to twenty-five per cent aqueous glycerin solution for one to three hours. A small quantity of a ten per cent glycerin solution should be left in the tubes. In steril- izing these tubes, thirty minutes a day in the Arnold after heating of the sterilizer will sterilize without altering the glycerin. Milk Media. — Fresh milk is procured and is heated in a flask for THE PREPARATION OF CULTURE MEDIA 131 fifteen minutes in an Arnold sterilizer. It is then set away in the ice chest for about twelve hours in order* to allow the cream to rise. Milk and cream are then separated by siphoning the milk into another flask. It is rarely necessary to adjust the reaction of milk prepared in this way, since, if acid at all, it is usually but slightly so. If, however, it should prove more than 1.5 per cent acid, it should be discarded or neutrahzed with sodium hydrate. The milk may then be poured into test tubes without further additions, or litmus solution may be added in a quantity sufficient to give a purplish blue color. The tubes are sterilized by fractional sterihzation in the Arnold sterihzer for thirty minutes on three successive days. Serum Media. — Loeffler^s Medium. — Beef blood is collected at the slaughter house in high cylindrical jars holding two quarts or more. It is desirable that attempts should be made to avoid contamination as much as is feasible by previously sterilizing the jars, keeping them covered, and exercising care in the collection of the blood. The blood is allowed to coagulate in the jars, and should not be moved from the slaughter house until coagulated. All unnecessary shaking of jars should be avoided. As soon as the coagulum is fully formed, adhesions between the clot and the sides of the jar should be carefully separated with a sterile glass rod or wire. The jars are then set away in the ice chest for 24 to 36 hours. At the end of this time clear serum will be found over the top of the clot, and between the clot and the jar. This should be pipetted off, preferably with a large pipette of 50 to 100 c.c. capacity, or siphoned off with sterile glass tubing, and transferred to sterile flasks. To three parts of the clear serum is then added one part of a one per cent glucose beef infusion or veal infusion bouillon. The mixture is filled into tubes, perferably the short test tubes commonly used for diagnostic diphtheria cultures. The tubes are then placed in a slanting position in the apparatus known as an inspissator (see p. 71). This is a double-walled copper box covered by a glass lid, cased in asbestos, and surrounded by a water jacket. It is heated below by a Bunsen flame. Together with the tubes a small open vessel containing water should be placed in the inspissator to insure sufficient moisture. The temperature of the inspissator is now raised to 70°-75° C, care being taken that the rise of temperature takes place slowly. The temperature is maintained at this point for two hours, and the process is repeated, for the same length of time, at the same temperature, on six successive days, preferably without removing the tubes from the inspissator at 132 BIOLOGY AND TECHNIQUE any time. It is also possible, though less regularly yielding good results, to sterilize in the inspissator for one day, following this on the second and third days by exposure for thirty minutes to 100° C. in the Arnold steam sterilizer. In doing this, the Arnold should be very gradually heated, at first without outer jacket, this being lowered only after thorough heating has taken place. Serum-Water Media for Fermentation Tests. — For the deter- mination of the fermentative powers of various microorganisms for purposes of differentiation, Hiss has devised the following media in which the cleavage of any given carbohydrate is indicated, not only by the production of an acid reaction, but by the coagulation of the serum proteids. Obtain clear beef serum by pipetting from clotted blood in the same way as this is obtained for the preparation of Loeffler's blood-serum medium. Add to this two or three times its bulk of distilled water, making a mixture of serum and water in proportions of one to two or three. Heat the mixture for fifteen minutes in an Arnold sterilizer at 100° C. to destroy any diastatic ferments present in the serum. Add one per cent of a five per cent aqueous litmus solution (the varia- tion in the different litmus preparations as obtained in laboratories necessitates a careful addition of an aqueous litmus solution until the proper color, a deep transparent blue, is obtained, rather than rigid adherence to any quantitative directions). Add to the various fractions of the medium thus made one per cent respectively of the sugars which are to be used for the tests. For the preparation of inulin medium, made in this way for pneu- mococcus-streptococcus differentiation, it is necessary to sterilize the inulin dissolved in the water to be added to the serum in an autoclave at high temperature (15 pounds for 15 minutes) in order to kill spores before mixing with the serum. The serum-water media are sterilized by the fractional method at 100° C, at which temperature they remain fluid. Special Media for Colon-Typhoid DifiEerentiation.^ — Hiss^ Plating Medium.^ — ^The composition of this medium is as follows: Agar 15 gms. Gelatin 15 " Liebig's meat extract 5 " Sodium chloride 5 " Dextrose 10 " Distilled water 1,000 c.c. * For details of use of these special media see also chapter on Bacillus tj^hosus. » HiaSf Jour. Exp. Med., ii, 1897; Jour. Med. Research, viii, 1902. THE PREPARATION OF CULTURE MEDIA 133 The agar is thoroughly dissolved in 1,000 c.c. of distilled water. When the agar is melted, the gelatin, meat extract, and salt are added and dis- solved by further heating. Any loss in weight is then adjusted by the addition of water. No titration or adjustment of reaction is necessary. The medium should be cleared with the whites of two eggs, and filtered through cotton. To the cleared medium is added one per cent of dex- trose, and the medium tubed, about 8 c.c. to each tube, and sterilized. Hiss' Tube Medium. — The composition is as follows: Agar 5 gms. Gelatin 80 " Liebig's meat extract 5 •" Sodium chloride 5 " Dextrose 10 " Distilled water 1,000 c.c. The method of preparation is the same as for the plating medium. The agar is thoroughly dissolved, and then the gelatin, meat extract, and salt are added and dissolved. After adjusting the loss in weight, the volume should be deter- mined, a careful titration made, and the reaction adjusted to one and five-tenths per cent acid by the addition of ^ HCl solution. The medium is cleared with eggs, filtered, and one per cent dextrose added. It is then tubed and sterilized. Hesse's Medium} — The medium devised by Hesse for typhoid-colon differentiation depends for its usefulness, as does the Hiss tube medium, upon the great motility of the typhoid bacillus. It may be used directly for the examination of feces or, as suggested by Jackson and Melia,^ after preliminary enrichment of the material to be examined by the use of the lactose-bile medium of Jackson. (See p. 138.) The Hesse medium is made up a^ follows: Agar 5 gms. (4.5 gms. absolutely dry) Pepton (Witte) 10 " Liebig's beef extract 5 " Sodium chloride 8.5 " Distilled water 1,000 c.c. Jackson and Melia found that the use of 4.5 gms. of completely dried agar give more uniform results, as the amount of moisture in commercial agar varies. The preparation of the medium is as follows: Dissolve 4.5 gms. of dry agar in 500 c.c. of distilled water over a free flame, making up for loss by evaporation. In another vessel 10 gms. of pepton, 5 gms. of beef extract, and 8.5 gms. salt are dissolved in 500 c.c. distilled water. This may be heated until dissolved and loss by evaporation made up. * Hesse, Zeit. f. Hyg., Iviii, 1908. ^ Jackson and Melia, Jour, of Inf. Dis., vi, 1909. 1^4 . BIOLOGY AND TECHNIQUE The solutions are mixed and heated thirty minutes; loss by evaporation is then made up and the solution is filtered through cotton. The reaction is ad- justed to one per cent acidity and the medium tubed — 10 c.c. to each tube. The typhoid bacillus is characteristic on the Hesse medium only when the dilution poured in the plates is so high that only a few colonies appear. The typhoid colonies are much larger than are the colon colonies and may often, be several centimeters in diameter. Conradi-Drigalshi Medium} — Original directions. (a) Three pounds of meat are infused in two liters of water for twelve hours or more. Strain, boil for one hour and add 20 gms. Witte's pepton, 20 gms. of nutrose, 10 gms. of NaCl; boil one hour and filter. Add 60 gms. of agar. Boil for three hours (or one hour in an autoclave) until agar is dissolved. Render weakly alkaline to litmus paper, filter, and boil for half an hour more. (b) Litmus solution: Two hundred and sixty c.c. of litmus solution are boiled for ten minutes. (The htmus solution used by Conradi and Drigalski is the very sensitive aqueous litmus recommended by Kubel and Tieniann, and purchasable under the name.) After boiling, 30 grams of chemically pure lac- tose are added to the litmus solution. The mixture is then boiled for fifteen minutes, and, if a sediment has formed, is carefully decanted. (c) Add the hot lactose mixture to the hot agar solution; mix well and, if necessary, again adjust to weak alkaUne reaction, litmus paper used as an in- dicator. To this mixture add 4 c.c. of a hot, sterile ten per cent solution of sodium carbonate, and 20 c.c. of a freshly made solution of crystal violet (c. p. Hochst), 0.1 gram in 100 c.c. of sterile distilled water. Surface smears are made upon large plates. These are incubated twenty-four hours. Typhoid colonies are small, blue, and transparent. Colon colonies are large, red, and opaque. Endows Medium.'^ — 1. Prepare one liter of meat infusion three per cent agar, containing 10 grams of pepton and 5 gr^ms of NaCl. 2. Neutralize and clear by filtration. 3. Add 10 c.c. of 10% sodium carbonate to render alkaline. 4. Add 10 grams of chemically pure lactose. 5. Add 5 c.c. of alcoholic fuchsin solution, filtered before using. (Endo in his original contribution does not mention the strength of this fuchsin solu- tion, which, however, should be saturated. This colors the medium red. 6. Add 25 c.c. of a 10% sodium sulphite solution. This again decolorizes the medium, the color not entirely disappearing, however, until the agar is cooled. 7. Put into test tubes, 15 c.c. each, and sterilize. 1 Conradi-Drigalski, Zeit. f. Hyg., xxxix, 1902. 2 Endo, Cent. f. Bakt., xxxv, 1904. THE PREPARATION OF CULTURE MEDIA 135 The medium should be kept in dark. Plates are poured and surface smears made. The typhoid colonies remain colorless, while those of coli become red. The preparation of Endo's medium presents difficulties due to the varying purity of sodium sulphite. Kastle and Elvove ^ recommend the use of anhydrous sodium sulphite instead of the crystallized variety. Harding and Ostenberg ^ add sodium sulphite solution to a measured amount of .5 per cent f uchsin to determine the proportions which give the greatest delicacy of reaction as tested with formaldehyde. The propor- tions so determined are then added to the hot 3 per cent agar. Although Endo described his medium as dependent upon the forma- tion of acid by the bacteria, this is not so. Acids give no coloration of the sulphite-fuchsin mixture. Indeed this mixture is used by chemists under the name of Schiff 's reagent as a test for aldehydes. Acids decol- orize the red caused by aldehydes, and this accounts for the frequent late discoloration of red colon colonies on prolonged cultivation. The medium is red when hot, and colorless when cold, because the com- pound between sulphite and f uchsin dissociates in the hot solution. Kendall's Modification of Endows Medium.^ — 1.5 per cent meat extract agar is prepared, and the reaction adjusted faintly alkaline to litmus by the addition of NaOH. This agar is stored in small flasks and it is usually convenient to keep flasks containing 100 c.c. each. Just before use, 1 per cent of lactose is added, and then decolorized fuchsin solution, as in Endows medium. Add about 1 c.c. of decolorized fuchsin solution, made up as above by mixing roughly prepared 10 per cent sodium sulphite with saturated alcoholic fuchsin. (The proportions of fuchsin and sulphite are sometimes difficult to adjust, possibly by reason of impurities in the sulphite due to formation of sulphate. The in- structions given by most workers at present are to use 10 c.c. of a 10 per cent aqueous solution of sodium sulphite, a§d to add to this 1 c.c. of a 10 per cent solution of fuchsin in 96 per cenf alcohol.) When these flasks containing the various ingredients are hot they are red or pink, but when plates are poured and allowed to harden, the medium should be either colorless or very faintly pinkish. It is best to pour a number of plates rather thickly and then allow them to dry with the covers off. Inoculations from the feces solution are then made by surface smear, with a bent glass rod. Colon colonies are pinkish and red; typhoid colonies, smaller and grayish. In concluding the description of some of the most important typhoid isola- tion media, we would like to add that a great deal seems to depend upon the ^ Kastle and Elvove, Jour. Inf. Dis., xvi, 1909. 2 Harding and Ostenberg, Jour, of Inf. Dis., xi, 1, 1909, 3 Kendall, Boston Med. & Surg. Jour. 136 BIOLOGY AND TECHNIQUE habit-acquired skill which the individual worker attains. None of these stool isolation media are ordinarily successful at once in the hands of anyone, and a certain amount of practice must be attained before one can judge of the use- fulness or uselessness of a medium. Brilliant Green Agar for Typhoid Isolation. — Krumwiede has recently devised a brilliant green agar with which he has had excellent results.* The basis is an extract agar like that used for Endo's medium: Beef Extract 0.3% Salt 0.5% Peptone 1.0% Agar 1.5% (Domestic peptones are satisfactory.) Dissolve in autoclave; clear and filter. A clear agar is essential. The final reaction of the medium is to be neutral to ^ Andrade's indicator, which in terms of phenolphthalein is 0.6-0.7% acid (normal HCl). It is more convenient to ,have the reaction set slightly alkaline to litmus at the time of preparation and to acidify each bottle as used. The agar is bottled in 100 c.c. amounts and auto- claved. When needed, the bottles are melted and the volume of each cor- rected (if necessary) to an approximate 100 c.c. Add to each bottle: One per cent Andrade's Indicator. Acid to bring to neutral point of the indicator.' One per cent Lactose.* 0.1 per cent Glucose.* Brilliant Green in 0.1 per cent aqueous solution. Two dilutions of dye are used in routine plating, corresponding to 1-500,000 and 1-330,000 in terms of soHd dye (0.2 c.c. and 0.3 c.c. of 0.1 per cent solution per 100 c.c. of agar). The sample of dye which Krumwiede has used is from Bayer, but he has also tested and found equally satisfactory samples from Griibler and Hochst. 0.1 gram of dye S accurately weighed on a foil, washed with boiling H2O into a 100 c.c. volumetric Iflask and made up to the mark when cool. The flask should be clean and neutral (by test). Fresh solutions vary in activity (see standardization tests); they keep about a month. 1 We are indebted to Dr. Krumwiede for a preliminary account of this method. ^ Andrade's Indicator: 0.5 per cent aqueous acid fuchsin 100 c.c. Normal NaOH 16 c.c. The dye is slowly (2 hours) alkalinized to the color-base; the red tint is restored by acids. 3 An agar is neutral to Andrade when, hot, the color is a deep red, but fades com- pletely on cooling. This is determined by cooling 3 or 4 c.c. of acidified hot agar in a serum tube under the tap and adjusting accordingly. * These are conveniently added from one sterile solution containing 20% lactose and 2% dextrose, 5 c.c. to 100 of agar gives the requisite concentration. THE PREPARATION OF CULTURE MEDIA 137 Each bottle is mixed and poured into six plates only (a thick layer of agar gives the most characteristic colonies). Plates are left micovered until agar has "jellied"; porous tops are used; dry plates are essential to avoid diffusion. Standardization: The agar must have proper "balance." The reaction is important; sediment reduces the activity of the dye and light colored media are better than darker ones. Different lots of agar with the same dye solution act ununiformly; a new batch or a new solution must be tested. Any variation in the composition of the media necessitates a readjustment of dye concentration; this statement cannot be over-emphasized. Brilliant green, in appropriate dilutions, not only inhibits all Gram- positive and many Gram-negative bacteria, but exhibits differential action on the colon-typhoid group. Paratyphoid and the B. lactis aerogenes are untouched, typhoid is restrained only at low dilutions, while dysentery and the other colon group are extremely susceptible. The typhoid colony on this medium is characteristic. Looking through the plate against a dark surface, in oblique light the colony has a snow- flake appearance; the edge delicately serrate. With artificial light and a hand lens, the texture is that of a coarse woolen fabHc. Acid produc- tion from the trace of glucose may tinge the colony. The colony is large. Malachite-Green Bouillon (Peabody and Pratt).' — To 100 c.c. of beef in- fusion broth add 10 c.c. of one per cent solution of malachite green, Hochst 120, made with sterile water. This is tubed. This medium is used as an enriching fluid. One drop of the suspected material (emulsified stool) is added to each tube and after incubation for eighteen to twenty-four hours inoculations may be made upon plates. Peabody and Pratt found a reaction of .5 per cent acidity to phenol- phthalein most favorable. Bile Medium^. — ^ (Recommended -for blood cultures by Buxton and Coleman.) The medium is prepared as follows: Ox-bile 900 c.c. Glycerin 100 c.c. Pepton 20 grams Put into small flasks containing quantities of about 100 c.c. each and sterilized by fractional sterilization. Jackson's Lactose-Bile Medium.^ — This medium is of great use in isolating B. typhosus and B. coli from water, and serves as a valuable enriching medium in isolating them from other sources. Jackson and ^ Peabody and Pratt, Boston Med. and Surg. Jour., clviii, 7, 1908. 2 Canradi, Deut. med. Woch., 32, 1906. « Jackson, "Biol. Studies of Pupils of W. T. Sedgwick," 1906, Univ. Chicago Press. 138 BIOLOGY AND TECHNIQUE- Melia ^ found that in this medium B. typhosus and B. coli outgrow all other microorganisms and eventually B. typhosus will even outgrow B. coli. It consists of sterilized undiluted ox-bile (or a ten per cent solution of dry, fresh ox-bile) to which is added one per cent pepton and one per cent lactose. It is filled into fermentation tubes and sterihzed by the fractional method. MacConkey's Bile-Salt Agar. — Sodium glycocholate 0.5 per cent. Pepton 2.0 " " Lactose 1.0 " " Agar 1.5" " Tap water q.s. The agar and pepton are dissolved and cleared and the lactose and sodium glycocholate added before tubing. The B. typhosus produces no change; B. coli, producing acid, causes precipitation of the bile salts. Neutral-Red Medium. — To 100 c.c. of a one or two per cent glucose agar add 1 c.c. of a saturated aqueous solution of a neutral-red. The medium is. used in tubes, stab or shake cultures. The typhoid bacillus produces no change, while members of the colon group render the medium color- less by reduction of the neutral-red and produce gas. Barsiekow's Medium.^ — To 200 c.c. of cold water, add 10 grams of nutrose and allow to soak for one-half to one hour. Pour this into 800 c.c. of boiling water, and heat for twenty minutes in an Arnold sterihzer at 100° C. Filter through cotton and to the opalescent solution of nutrose add 5 grams of NaCl, 10 grams of lactose, and sufficient aqueous litmus solution to give a pale blue color. 3 RusselVs Double Su^ar Agar.* — Russell has devised a simple medium for quick identification of typhoid bacilli. A 2% or 3% extract agar is used, about 0.8% acid to phenolphthalein 0.8%. Enough litmus solution is added to give it the ordinary deep blue. The re- action is then adjusted with sodium hydrate until neutral to litmus. Finally 1% lactose and 0.1% glucose (dissolved in a small amount of hot water) are added, the medium is carefully sterilized in the Arnold sterilizer and slanted. Inoculations are made by surface streak and stab. The typhoid bacillus grows colorless on the surface, but under the imperfect anaerobic conditions at the bottom of the stab, a bright red color is developed by acid formation. Dieudonne^s Selective Medium for cholera spirillum. See page 584. Enriching Substances Used in Media. — For the cultivation of micro- 1 Jackson and Melia, Jour. Inf. Dis., vi, 1909. 2 Barsiekow, Wien. klin. Rund., xliv, 1901. 3 Filtration may be done through paper, but takes a long time. * Russell, Jour. Med. Research, xxv, 1911, 217. THE PREPARATION OF CULTURE MEDIA 139 organisms which are sensitive to their food environment, it is often neces- sary or advisable to add to the ordinary media enriching substances, which empirical study has shown to favor the growth of the organism in question. The substances most commonly used for such enrichment are glucose, nutrose (sodium caseinate), glycerin, sodium formate, and unsohdified animal proteids. As animal and blood serum and whole blood must frequently be used in this way, an understanding of the methods employed in obtaining these substances is necessary. Method of Obtaining Blood and Blood Media. — Blood serum from beef and sheep may be collected in the manner recommended for the collection of such serum in the preparation of Loeffler's medium, pipetted into test tubes, and sterilized in the fluid state by exposure to tempera- tures ranging from 60° to 65° C, for one hour upon six consecutive days. It is not a simple matter to sterilize serum in this way and requires much time and care. The method most commonly employed, in laboratories which have access to hospitals, for obtaining clear serum depends upon the collection of exudate or transudate fluids by sterile methods directly from the pleural cavity, the abdominal cavity, or the hydrocele cavity. Sterile flasks or test tubes are prepared and the fluid is allowed to flow directly out of the cannula into these. It is necessary to avoid carbolic acid or other disinfectants in sterilizing instruments and rubber tubing used during the operation. These should be brought into the ward in the water in which they have been boiled and not in strong antiseptic solu- tions, as is frequently done. The fluid so obtained may be incubated and the contaminated tubes discarded. The serum may then be added, in proportions of one to three, to sterile broth or melted agar. Agar thus used is melted and cooled to 60° C, or below. One-third of its volume of warmed exudate fluid is added, and the plates are poured. Serum may be rendered free of bacteria by filtration through a Berkefeld or Pasteur-Chamberland filter. This is an effectual method, but requires much time and care. Whole blood may be obtained for cultural purposes by bleeding rabbits or dogs or other animals directly from a blood-vessel into tubes of melted agar. In the case of a rabbit, after the administration of an anes- thetic (ether), an incision is made directly over thie trachea, and, by careful section, the carotid artery is isolated, lying close to the side of the trachea. 140 BIOLOGY AND TECHNIQUE THE INFLUENCE OP DYE STUFFS UPON BACTERIAL GROWTH, AND AS ADDITIONS TO SELECTIVE MEDIA In describing the selective media for typhoid bacilU we have seen that malachite green and crystal violet have been found to exert a certain amount of selective action upon the tjrphoid and colon groups. The selective influence of various dyes has been recently again studied by Churchman. Churchman ^ found that the addition of gentian violet in dilutions of 1 : 100,000, to media, inhibited some bacteria, while others grew luxuriantly in its presence. Extremely interesting, both practically and theoretically, is his observation that upon such gentian violet media bacteria fall into two groups. Those which grow on gentian violet correspond in a general way to the Gram-negative bacteria; those which fail to develop on these media correspond roughly with the Gram-positive species. One strain of the enteritidis group could not be cultivated on gentian violet, and this was found to differ from the others also in its agglutination tests. Signorelli ^ claims that dahUa is useful in differentiating true cholera strains from similar spirilla. The true cholera strains grew with colored colonies, while others remain colorless, in his experiments. Krumwiede and Pratt ^ were unable recently to confirm the claims of Signorelli. However they fully confirm the findings of Churchman both as to the selective action of gentian violet and in regard to the classification of bacteria into two groups corresponding to their reaction to the Gram stain. They state that among Gram-negative bacteria a strain is occasionally found which will not grow on the gentian violet media, differing in this respect from other members of the same species. They find also that the reaction is quantitative. The streptococcus-pneumococcus group, according to their investi- gations, differs from other bacteria in being able to grow in the presence of quantities of violet which inhibit other Gram-positive species. Dys- entery bacilli show variations. Other dyes which they investigated showed no specific inhibitory properties which could be utilized for classification. 1 Churchman, Jour. Exp. Med., 16, 1912; also Churchman and Michael, ibid. 2 Sigruyrelli, Centralbl. f . Bakt., Orig. 56, 1912. ^Krumwiede and Pratt, Centralbl. f. Bakt., Orig. 68, 1913; and Proc. N. Y. Path. Soc, xiii, 1913. CHAPTER VIII METHODS USED IN THE CULTIVATION OF BACTERIA INOCULATION OF MEDIA The transference of bacteria from pathological material to media, or from medium to medium, for purposes of cultivation, is usually ac- complished by means of a platinum wire or loop. The platinum wire used should be thin and yet possess a certain amount of stiffness and be about two to three inches in length. This is fused into the end of a glass rod six to eight inches long. It is an advantage, though not necessary, to use rods of so-called "seahng-in" glass which, having the same co- efficient of expansion as platinum, does not crack during sterilization. For work with fluid media, the wire should be bent at its free end so as to form a small loop which will pick up a drop of the liquid. For the inoculation of solid media and the making of stab cultures, a straight "needle " or wire should be used. Other shapes of these wires and spat- ulae from heavy wire have been devised for various purposes and are easily improvised as occasion demands. (See Fig. 27.) When making a transfer from one test tube to another, the tubes should be held between the thumb and first and second fingers of the left hand, as shown in Fig. 28. The plugs are then removed by grasping them between the small and ring fingers and ring and middle fingers of the right hand, first loosening any possible adhesions between glass and plugs by a slight twisting motion. The platinum wire is held meanwhile by the thumb and index fingers of the right hand in the manner of a pen. The wire is heated red hot in a Bunsen flame, and is then passed into the culture tube without being allowed to touch the glass. It is held sus- pended within the tube for a few seconds to permit of cooling before touching the bacterial growth. The wire is then allowed to touch lightly the surface of the growth and a small amount is picked up. (See Fig. 29.) It is then removed from the tube without allowing it to touch the sides of the glass, and is passed into the tube which is to be inoculated. If the tube contains a slanted medium, such as agar, a light stroking motion from the bottom of the slant to its apex will deposit the bacteria 141 142 BIOLOGY AND TECHNIQUE /T^ \3 rl 13 t f^ fl ^ll upon the medium evenly along a central line. The needle may also b^ plunged downward into the substance of the nutritive material so that in the same tube both surface growth and deep growth may be observed. If a stab culture is to be made in unslanted agar or in gelatin, the needle is simply plunged straight downward as nearly as possible along the axis of the medium. If a fluid medium is being inoculated, the wire should be introduced only into the upper part of the liquid and the bac- teria gently rubbed into emulsion against the side of the glass. The needle is then removed from the tube, the stopper carefully replaced, and the platinum wire immediately sterilized in the flame. This sterilization of plati- num needles after they have been in contact with bacteria should become second nature to those working with bacteria, since an infraction against this rule may give rise to serious and widespread consequences. In burning off platinum needles it is well to re- member that a part of the glass rod, as well as the wire itself, is introduced into the tubes and may become con- taminated, and for this reason the rod itself, at least in its lower two or three inches, should be passed through the flame as well as the wire. As an extra precaution against contamina- tion, the lips of test tubes and flasks and the protruding edges of cotton plugs may be passed through the flame and singed. THE ISOLATION OF BACTERIA IN PURE CULTURE Fig. 27. — Platinum Wires. It is obvious that in many cases where bacteria are cultivated from water, milk, pathological material, or other sources, many species may be present in the same specimen. It is likewise obvious that scientific bacteriological study of any bacterium can be made only if we obtain METHODS USED IN CULTIVATION OF BACTERIA 143 this particular species entirely apart from others, in what is known as "pure culture." The earliest methods for accomplishing this were the methods of Pasteur and of Cohn who depended upon the power of one species to outgrow all others, if cultivated for a sufficient length of time in fluid media. This method, of course, was inadequate in that it was often purely a matter of chance which one of the mixture of species was finally obtained by itself. A later method, by Klebs, depends upon serial dilution, in test tubes of fluid media, by which the eventual transference of one germ only, to the last tube was attempted. Such methods, none of them of great practical value, have been entirely dis- FiG. 28. — Taking Plugs from Tubes before Inoculation. placed by those made possible by Koch's introduction of solid media which may be rendered fluid by heat. The methods now employed for the isolation of bacteria depend upon the inoculation of gelatin or agar, when in the melted state, the thorough distribution of the bacteria in these liquids by mixing, and the sub- sequent congealing of these media in thin layers. By this means the in- dividual bacteria, distributed in the medium when liquid, are held apart and separate when the medium becomes stiff. The masses of growth or "colonies" which develop from these single isolated microorganisms are discrete and are descendants of a single organism, and can be trans- ferred, by means of a process known as " colony-fishing," to fresh sterile culture media. 11 144 BIOLOGY AND TECHNIQUE Plaitng.— The first method employed by Koch for bacterial isolations was one that consisted in the use of srniple plates of glass, about 3x4 inches in size, mounted upon a leveling stand. A wooden triangle, supported upon three adjustable screw-feet, formed the base of this apparatus. Upon this was set a covered crystallizing dish which could be filled with ice water. Upon the top of this rested the sterilized plates under a bell j ar. By screwing up or down upon the supports of the triangle, leveling of the plate could be achieved and controlled by a spirit- level placed at its side. The inoculated gelatin was poured upon the Fig. 29. — Inoculating. plate and spread and rapidly cooled and hardened by the cold water contained in the crystallizing dish. The original method of Koch has been modified considerably and the method universally employed at present depends upon the use of circular covered dishes, the so-called /*ein dishes. These obviate the necessity of a leveling stand and prevent contamination of the plate when once poured. Each Petri dish plate consists of two circular glass dishes; the smaller and bottom dish has an area of 63.6 square centimeters; the larger is used as a cover for the smaller, and forms a loosely fitting lid. METHODS USED IN CULTIVATION OF BACTERIA 145 The plates when fitted together are sterilized and thus form a closed cell which, if properly handled, may remain sterile indefinitely. The technique for making a pour plate for the purpose of isolating bacteria from mixed culture is as follows: The actual "pouring" of plates is preceded by the preparation of usually three graded dilutions of the material to be examined. For this purpose three agar or gelatin tubes are melted and, in the case of the agar, are cooled to a temperature of about 42° C. in a water bath. A platinum loopf ul of the material to be examined is transferred to one of these tubes. The bacteria are then thoroughly distributed throughout the melted Fig. 30. — Pouring Inoculated Medium into Petri Plate. gelatin or agar by alternately depressing and raising the plugged end of the tube, giving it a rotary' motion at the same time. This thoroughly distributes the bacteria throughout the medium without allowing the formation of air-bubbles. Two loopfuls of this mixture are then trans- ferred to the second tube and a similar mixing process is repeated. This second tube contains the bacteria in much greater dilution than the first and the colonies which will form in the plate poured from this tube will be farther apart. A third dilution is then made by transferring five loopfuls of the mixture in the second tube to the third. This again is mixed as before. The contents of the tubes are then poured into three 146 BIOLOGY AND TECHNIQUE sterile Petri dishes. The pouring should be done with great care. The cover of the dish is raised along one margin simply far enough to permit the insertion of the end of the test tube, the plug of which has been removed and the lips passed, with a rotary movement, through the flame. The medium is poured into the dish without the lips of the tube being allowed to touch either the bottom or the cover of the dish. The cover is then replaced and the medium allowed to harden. When agar has been used, the dishes may be placed in an incubator at 37° C. It is well to place the plates upside down in the incubator. This prevents the condensation water, squeezed out of the agar dur- ing hardening, from collecting on its surface, and forming channels for the diffuse spreading of bacteria. The same end may be attained by the use of Petri plates provided with porous earthenware lids, as suggested by Hill. Simple inversion of the plates, however, usually suffices. When gelatin has been used, the plates are allowed to remain in a dark place at room temperature or in a special thermostat kept at 22°-25° C. Colonies in agar, kept at 37.5° C. , usually develop in eighteen to twenty- four hours; those in gelatin or agar at room temperature in from twenty- four to forty-eight hours, depending upon the species of bacteria which are being studied. Often in the second dilution, more frequently in the third, the colonies will be found well apart and can then be "fished." The process of " colony-fishing " is one which requires practice and should always be done with care, for upon its success depends the purity of the sub-culture obtained. Colonies should never be fished under the naked eye, no matter how far apart and discrete they may appear, since not infrequently close to the edge of or just beneath a larger colony there may be a minute colony of another species which may be too small to be visible to the naked eye, but which, nevertheless, if touched by accident will contaminate the sub-culture. For proper " fishing,'^ the Petri plate with cover"removed, should be placed upon the stage of the microscope and examined with a low power objective, such as Leitz No. 2 or Zeiss AA. The steriHzed platinum needle, held in the right hand, is then carefully directed into the line of focus of the lens, while the small finger of the hand is steadied upon the edge of the microscope stage. When the point of the needle is clearly visible through the microscope, it is gently depressed until it is seen to touch the colony and to carry away a portion of it. The needle is then withdrawn without again touching the nutrient medium or the edges of the glass or the lens, and transferred to a tube of what- ever medium is desired. In this way, individuals of one colony, de- METHODS USED IN CULTIVATION OF BACTERIA 147 scendants of a single bacterium of the original mixture, — are carried over to the fresh medium. Esmarch Roll Tubes. ^ — A simple method of obtaining separate colo- nies is that devised by von Esmarch and known as " roll-tube " cultiva- tion. Tubes of melted gelatin are inoculated with various dilutions of the bacterial mixture and, while still liquid, are laid in an almost horizon- FiG. 31. — Streak Plate. tal position upon a block of ice, which has been grooved slightly by means of a test tube filled with hot water. The test tube containing the gelatin, after being placeji in this groove, is rapidly revolved by passing the fingers of the right hand across it, while its base is carefully steadied with the left hand. If the revolving is carried out with Esmarch, Zeit. f. Hyg., i, 1886. 148 BIOLOGY AND TECHNIQUE sufficient speed, the gelatin will harden in a thin layer on the inner surface of the tube. The colonies will develop in this layer and may be " fished " by means of a platinum wire with bent point introduced into the' tube. This method is useful for certain purposes, but is too inconvenient for routine work. It is now rarely used. Separatio: i of Bacteria by Surface Streaking. — When it is necessary to isolate bacteria like the gonococcus. Bacillus influ- enzae, the pneumococcus, and others, which, because of great sensitiveness to environment and possibly a preference for free oxygen, are not readily grown in pour plates, it is often advantageous first to pour plates of suitable media, allow them to harden, and then gently smear over their surfaces dilutions of the infectious material, usually in three or four parallel streaks. (See Fig. 31.) Upon such plates, if dilutions have been prop- erly made, and this is only a question of judgment based upon an estimation of the numbers of bac- teria in the original material, discrete colonies of the microorganisms sought for may develop, and can be "fished" in the usual manner. The media most favorable for the cultivation of various microorganisms will be discussed in the sections dealing "mth the individual species. ANAEROBIC METHODS We have seen in a preceding chapter (p. 26) pjQ^ 32^ Deep ^^^^ many bacteria, the so-called anaerobes, will Stae Cultivation develop only in an environment from which free OF Anaerobic oxygen has been excluded. Bacteria. rpj^^ exclusion of oxygen for purposes of anaero- bic cultivation may be accomplished by a variety of methods, depending upon a few simple principles which have been applied, either separately or in combination, by many workers. The earliest methods depended upon the simple exclusion of air l^y mechanical devices. In other methods, the oxygen of the air is displaced by inert gases (hydrogen) , and others again depend upon the .oxygen- absorbing qualities of alkaline solutions of pyrogallol. Cultivation by the Mechanical Exclusion of Air. — Koch succeeded in METHODS USED IN CULTIVATION OF BACTERIA 149 growing anaerobic bacteria upon plates by simply dropping upon the surface of the inoculated agar or gelatin a flat piece of sterile mica. This method, however, rarely succeeds in sufficiently excluding the air. LiBORius' Method.^ — This method consists in the use of deeply filled tubes of agar or gelatin, from which all oxygen has been removed by boiling for fifteen minutes or more. It is advantageous, as has been pointed out in the section on anaerobiosis, that media used for this purpose should contain carbo- hydrates in some form, perferably glucose. After boiling, the tubes are rapidly transferred to ice water so that as little oxygen as possible may be absorbed during the hardening of the medium. The tubes are then inoculated by deep stabs. After inoculation, the medium may be covered with a thin layer of agar, gelatin, or oil (albolin), and further sealed with sealing-wax to prevent oxygen- absorption. This method may be utilized for the isolation of anaerobes (as in the original method of Libo- rius) by inoculating the medium just before it solidifies. The tubes may be gently shaken in order to distribute the bacteria throughout the medium and then rapidly cooled. In this case colonies which develop may be scattered through- out the deeper layers of the agar or gelatin, and may be "fished" after breaking the tube. Esmarch's Method.2 — Von Esmarch has applied the principles of his roll-tube to the cultivation of anaerobic bacteria. Gelatin tubes are inoculated as above and roll-tubes prepared. The tubes are then set into cold water to prevent melting of the thin gelatin layer and the interior of the tube is filled with melted gelatin. Roux's Method .2 — ^Anaerobic bacteria are culti- vated by sucking the inoculated gelatin or agar into narrow tubes, which are then closed at both ends by fusing in the flame. After growth Tias taken place the tubes are broken and the organism re- covered by "fishing." ^ Idborius, Zeit. f . Hyg., i, 1886. 2 y^^ Esmarch, loc. cit. 3 Roux, Ann. Past., i, 1887. Fig, 33. — Deep Stab Cultiva- tion OF Anaero- bic Bacteria. 150 BIOLOGY AND TECHNIQUE Fluid Media Covered with Oil. — Erlenmeyer flasks or othervessels are partially filled with glucose-bouillon over which a thin layer of al- bolin or other oil is allowed to flow. The oxygen is driven out of the liquid by vigorous boiling for fifteen minutes or more. It should be remembered whenever using this or similar methods that a layer of fluid oil does not form an impermeable seal. By covering an alkaline pyrogallol solution with oil it can easily be shown that oxygen slowly diffuses through the oil into the medi- um below. The simple exclusion of air, also, is the principle underlying the culti- vation of anaerobic bacteria in the closed arm of a Smith fermentation tube. Wright's Method.' — Wright has described a simple and excellent method for the cultivation of anaero- bic bacteria in fluid media. The ap- paratus necessary is easily improvised with the materials at hand in any laboratory. A short piece of glass tubing, constricted at both ends and fitted at each end with a small piece of soft-rubber tubing, is inserted into a test tube containing nutrient broth. The upper end of the inserted glass tubing is connected by the rubber with a pipette passed through the cotton plug in the tube. The entire apparatus, plus broth, may be steril- ized after being put together. When a cultivation is made, the fluid in the test tube is inoculated as usual. The fluid is then sucked up into the glass tubing until this is completely filled. A downflow of the fluid is then prevented by placing the finger over the pipette through which the suction has been made or by constricting a small piece of rubber tubing attached to the upper end of the pipette. The entire system of tubes is then pushed downward in such a way that both pieces of rubber tubing, attached to the ends of the httle glass 1 Wnght, J. H. Quoted from Mallory and Wright, "Path. Technique," Phila., 1904. Fig. 34. — Cultivation op Anaerobes in Fluid under Albolin. METHODS USED IN CULTIVATION OF BACTERIA 151 chamber, are kinked. The entire apparatus may then be incubated. Growth of anaerobic bacteria takes place within the air-tight chamber formed by the short glass tubing within the test tube. The fluid in the test tube, outside of this chamber, usually remains clear. When cultivation has been successful, the bacteria may be obtained either for morphological study or for further cultivation, by simply allowing the fluid to flow out of the little air-tight chamber back into the test tube. The method is simple and usually successful. Methods Based upon the Displacement of Air by Hydrogen. — The principle of air-displace- ment by hydrogen, first utilized by Hauser,* has been widely applied to the cultivation of anaerobic bacteria. In substance it consists of the conduction of a stream of hydrogen through an air-tight chamber in which plates or tubes containing inoculated media have been placed. For the production of hydrogen, the most convenient apparatus is the Kipp hydrogen generator. Hydrogen is generated from zinc and sulphuric acid and this may be passed through a series of Woulfe-bottles, containing solutions of lead acetate and of pyrogallic acid, to remove traces of sulphuretted hydro- gen and of oxygen, respectively, of Lugol's solution to absorb traces of acid vapor, and of one with a silver-nitrate solution to take up any hydrogen arsenide. For the preparation of anaerobic condi- tions where very rigid anaerobiosis is neces- sary, nitrogen may be used, which can be bought in tanks from commercial firms. For anaerobic cultivation upon solid media, the inoculated tubes or plates are placed in an apparatus such as the Novy jar. This is connected with the hydrogen apparatus and hydrogen allowed to flow through it for five or ten minutes, and the stop-cocks then closed. Fig. 35. — Wright's Method of Anaerobic Cultivation in Fluid Media. ^ Hauserj "Ueber Faulnissbakterien," 1885. 152 BIOLOGY AND TECHNIQUE In applying the hydrogen method to fluid media, flasks containing the broth are fitted with sterile, tightly fitting rubber stoppers per- forated by two holes, through which glass tubes are passed. One of these tubes, the inlet, passes below the surface of the liquid. The other one, the outlet, extends only a short distance below the stopper and is always kept above the surface of the medium. The flasks are inoculated and hydrogen is passed through the medium so that it enters the long tube, bubbles up through the fluid, and leaves by the short tube. The broth may be covered with a thin layer of liquid paraffin or albolin. The Use ot Pjnrogallic Acid Dissolved in Alkaline Solutions for Oxygen Absorption. — Buchner* has applied the principle of chemical absorption for the re^ moval of oxygen to the cultiva- tion of anaerobic bacteria. This has been made use of in a number of different ways. The method is based upon the fact that alkaline solutions of pyro- gallol possess the power of ab- sorbing large quantities of free oxygen. At first such solutions are of a light, straw-color, which becomes dark brown as oxygen is absorbed. The absorption of all the oxygen in the environ- ment may be assumed when there is no further deepening of the brown color. Buchner first utilized this principle by placing a small wire or glass holder within a large test tube, dropping dry pyrogallol (pyrogallic acid) into the bottom of this tube, then running thirty per cent sodium hydrate solution into it, and inserting within this large tube a smaller test tube containing the inoculated culture medium. The large tube was then tightly closed Fig 36.— Novy Jar. 1 Buchner, Cent. f. Bakt., I, iv, 1888. METHODS USED IN CULTIVATION OF BACTERIA 153 with a rubber stopper. In this way, the air space surrounding the smaller tube was rendered oxygen free. A simple modification of the preceding method of Buchner has been devised by Wright.^ Stab-cultures of gelatin or agar in test tubes are made in the usual way. The cotton stopper closing the tube is then thrust into the tube to such a depth that its upper end lies at least 1 cm. below the mouth of the tube. A small quantity of sodium or potassium hydrate solution in which some pyrogallic acid has been dissolved, is then allowed to flow on to the cotton of the plug and the mouth of the tube is immediately sealed by a tightly fitting rubber stopper. The cotton stopper in these cases must be made of ab- sorbent cotton; 1.5 to 2.5 c.c. of the pyro- gallicr acid solution is usually sufl^icient for test tubes of ordinary size. For cultivation of anaerobic bacteria upon agar slants, a simple technique may be appHed and easily improvised in the labora- tory as follows: The tube of slant agar is inoculated with the infectious material in the usual way. It is then, with stopper removed, inverted into a tumbler or beaker containing about a gram of dry pyrogallic acid. A small quantity of a five per cent or three per cent sodium hydrate solution is then run into the tumbler and this is covered with a thin layer of liquid paraffin or albolin before the pyrogallic acid has been completely dis- solved. In this way, completely anaerobic conditions are obtained in the tube and the growth of anaerobes takes place upon the surface of the slant. For the cultivation of anaerobes in Petri dishes, for purposes of separation, a combination of the pyrogallic acid method and the hydrogen displacement methods is often em- ployed. For this purpose the jars devised by Novy and by Bulloch are extremely convenient. 1 Wright, Jour, of the Boston Soc, of Med. Sci,, Dec, 1900, k. Fig. 37, — Wright's Meth- od OF Anaerobic Cultiva- tion BY THE Use of Pyro- gallic Acid Solution. 154 BIOLOGY AND TECHNIQUE In using the Novy jar, the inoculated plates are set upon a wire frame resting about an inch above the bottom of the jar. The cover is then tightly set in place and the air in the jar exhausted by means of a suction pump. The arrangement of the double stop-cock in the top renders it possible now, by simply turning this, to admit hydrogen from a Kipp generator into the jar. The process of alternate exhaustion and admission of hydrogen may be several times repeated. A combination of air exhaus- tion, oxygen absorption, and hy- drogen replacement may be prac- ticed in jars such as that shown in Fig. 39. Tubes or plates after inoculation are placed in this jar, on a raised wire frame. Dry py- rogallic acid is placed in the bottom of the jar and the cover tightly fitted. An opening in the side of the jar connects its interior with a bottle containing sodium or potassium hydrate so- lution. Through the stopper of this bottle pass two glass tubes, one of them of such length that it can be pushed down into the alkaline solution, or pulled up- ward above the level of the fluid. This tube connects the jar with the bottle. The other glass tube is short, passing just through the stopper and at the top made in the form of a T, one arm of the T being connected with a Kipp hydrogen generator, the other with a suction-pump. After the jar has been sealed, the glass tube connecting the jar and the bottle is raised above the level of the fluid in the bottle and, the con- nection to the hydrogen generator being closed, the air in the jar is exhausted with the suction-pump. Connection to the suction may then be closed, and the other arm of the T being open, hydrogen is allowed to flow into the jar. Alternate suction and hydrogen replacement may be Fig. 38. -Jar for Anaerobic Cul- tivation. METHODS USED IN CULTIVATION OF BACTERIA 155 carried out two or th/ee times. After the last exhaustion, the glass tube in the bottle connecting it with the jar is pushed down into the fluid, and the vacuum will draw the sodium hydrate solution into the bottom of the jar, dissolving the pyrogallol, which will then absorb any traces of free oxygen remaining in the jar. Hydrogen is again introduced and tlie jar closed. If exhaustion of oxygen has been sufficiently thorough, the pyrogallic solution in the bottom of the jar will remain light brown. A simple method for the separation of anaerobes in plates without the use of hydrogen or of specially constructed jars, may be carried out as follows ^: The apparatus used consists of two circular glass dishes, fitting one into the other as do the halves of a Petri dish, and similar to these in every respect except that they are higher, and that a slightly greater space is left bet^i^een their sides when they are placed together. The From hydrogen generator To exhaustion pump Fig. 39. — Apparatus for Combining the Methods of Exhaustion, Hydrogen, Replacement, and Oxygen Absorption. dishes should be about three-fourths to one inch in height, they need be of no particular diameter, although those of about the same size as the usual Petri dish are most convenient. An important requirement necessary for the success of the method is that the trough left between the two plates, when put together, shall not be too broad, a quarter of an inch being the. most favorable. Into the smaller of these plates the inoculated agar is poured exactly as this is done into a Petri dish in the ordinary aerobic work. Pro- longed boiling of the agar before plating is not essential. When the agar film has become sufficiently hard on the bottom of the smaller dish, the entire apparatus is inverted. The smaller dish is now lifted out of the 1 Zinsser, Jour. Exp. Med., viii, 1906. 156 BIOLOGY AND TECHNIQUE larger, and placed, still inverted, over a moist surface — a towel or the wet surface of the table — to prevent contamination. Into the bottom of the larger dish, which now stands open, there is placed a quantity (about 3 grams) of dry pyrogallic acid. Into this, over the pyrogallic acid, the smaller dish, still inverted, is then placed. A five per cent solu- tion of sodium hydrate is poured into the space left between the sides of the two dishes, in quantity sufficient to fill the receiving dish one-half full. While this is gradually dissolving the pyrogallic acid, albolin, or any other oil (and this is the only step that requires speed), is dropped from a pipette, previously filled and placed in readiness, into Fig. 40. — Simple Apparatus for Plate CuLTr7ATioN of Anaerobic Bacteria. (Zinsser.) the same space, thus completely sealing the chamber formed by the two dishes. If these steps have been performed successfully, the pyrogallic solu- tion will at this time appear of a light brown color, and the smaller plate, with its agar film, will float unsteadily above the other. Very rapidly, as the pyrogallic acid absorbs the free oxygen in the chamber, this plate is drawn down close to the other, and the acid assumes a darker hue, which remains without further deepening even after three or four days' incubation. The Use of Fresh Sterile Tissue as an Aid to Anaerobic Cultivation. — The addition of small pieces of fresh sterile tissue (rabbit or guinea- pig) to culture tubes, either solid or fluid, greatly favors the growth of anaerobic bacteria. By such a method anaerobes can be made to de- velop even when other precautions for the establishment of anaerobiosis METHODS USED IN CULTIVATION OF BACTERIA 157 are imperfectly observed. This was noticed first by Theobald Smith and by Tarozzi and has become an extremely useful reenforcement to other methods. It has been utilized most extensively by Noguchi of recent ^ , jS, Fig. 41. — Incubator. years in his technique for the cultivation of various treponemata. The simplest way to apply this method is to place a piece of freshly excised rabbit kidney, testicle, or spleen into the bottom of a high test tube (20 cm.) and then pouring over it the culture fluid. Kidney or 158 BIOLOGY AND TECHNIQUE other tissues are more suitable for this purpose than Uver tissue since the latter is not easily obtained in a sterile condition, bacteria often getting into it during life through the portal circulation. The action of the tissues depends probably upon its great reducing power. THE INCUBATION OF CULTURES After bacteria have been transferred to suitable culture media, it is necessary to expose them to a temperature favorable to their develop- ment. In the case of many saprophytes, the ordinary room temperature is suffi- ciently near the optimum to obviate the use of any special apparatus for maintain- ing a suitable temperature; in the case of most pathogenic bacteria, however, the body temperature of man, 37.5° C, is either a necessary requirement for their growth, or at any rate favors speedy and characteristic develop- ment. For the purpose of obtaining a uniform temperature of any required degree, the apparatus in general use is the so-called incubator oi: thermostat. This may be adjusted for gelatin cultivation at 20 to 22° C, or for agar, broth, or other media at 37.5° C. Incubators, while varying in detail, are all constructed upon the same prin- ciples. They consist of double-walled copper chambers, which are fitted with a set of double doors, the outer being made of asbestos- covered metal, the inner of glass. (See Fig. 41.) The space be- tween the two walls is filled with water, which, being a poor heat conductor, tends to prevent rapid changes of temperature within the chamber as the result, of changes in the external environment. Both walls are perforated above by openings to admit thermometers into the interior and one wall is perforated so that a thermo-regulator may be inserted into the water jacket. The under surface of the chamber is heated by a gas Fig. 42. Fig. 43. Fig. 42. — Thermo-regulator (Lautenschlagef.) Fig. 43.- -Thermo-regulator. (Reichert.) METHODS USED IN CULTIVATION OF BACTERIA 159 flame, the size of which is automatically regulated by the thermo- regulator. A number of thermo-regulators are on the market, all of them con- structed upon modifications of the same principle. One of the most efficient of those in common use is that shown in Fig. 42. This con- sists of a long tube of glass fitted with a metal cap through which an in- let tube (A) projects into the interior. Slightly below the middle of the tube there is a glass diaphragm separating its interior into two com- partments. In the middle of the diaphragm an aperture leads into a spiral of glass which projects into the lower compartment. The lower compartment is filled with ether and mercury. The lower end of the inlet tube (A) has a wedge-shaped slit. The gas from the supply pipe passing through the tube (A) is conducted through the slit-like opening in its lower end into the inner chamber and passes out to the burner through the elbow (B). When the temperature is raised, the ether and mercur}^ in the lower chamber expand and the n:ercury rises in the upper chamber, gradually restric ting the opening through the V-shaped slit • in the inlet tube. Thus the gas supplied to the burner is diminished,, the flame reduced, and the temperature again falls. The temperature can be arbitrarily adjusted by raising or lowering the inlet tube. A scale at the upper ,end of the inlet tube allows exact adjustment. Complete shutting off of the gas is prevented by a small circular opening placed in the inlet tube just above the slit. Another cheaper and simpler thern o-regulator is shown in Fig. 43. This consists of a long tube -open at the top and fitted about 1^ inches fro'm. the top with two hollow glass elbows. One of these elbows remains open, the other, situated on a slightly lower level, is closed by a brass screw-cap. The tube is filled with mercury to a point slightly above the level of the elbow containing the screw-cap. The height of the mercury can thus be increased or decreased by screwing in or out upon the cap. Into the upper end of the tube there is fitted another device which consists of a T-shaped system of glass tubes, one arm of the T being open and the other closed, the perperdicular leg of the T tapering to a minute opening at the bottom. The gas passes into one arm of the T down through the tapering leg and into the space immediately above the mercury. It then passes out through the open elbow of the main tube. As the mercury rises, it gradually diminishes the space between its surface and the small opening in the tapering tube above it, finally completely shutting off the gas from this source. Gas can now pass only through a minute hole perforating the vertical leg of the 12 160 BIOLOGY AND TECHNIQUE T an inch above its end. The flame decreases and the temperature again sinks. . Since gas pressure in laboratories is apt to vary, it is convenient to interpose between the gas supply and thermo-regulator some one of the various forms of gas-pressure regulators. The use of these is not ab- solutely necessary but aids considerably in the maintenance of a con- stant temperature. The one most commonly employed is the so-called Moitessier apparatus. This consists of a cylindrical metal chamber within which there is fitted an inverted metal bell. Glycerin is poured into the cylinder to the depth of about two inches. An inlet pipe con- FiG. 44. — Moitessier Gas-Pressure Regulator. ducts gas into the open space between the top of the glycerin and the bell. From the top of the bell is suspended a conical piece of metal which hangs free in the outlet pipe. As the gas pressure under the bell increases, this is raised and the opening of the outlet pipe is gradually diminished by the cone. Thus the relation between the pressure in the inlet pipe and the actual quantity of gas passing through is equalized. A cup con- nected to the top of the bell through the roof of the cylinder by a bar can be filled with birdshot and the pressure against the gas can thus be modified to conform with existing conditions. METHODS trsEr) m CUitiVAtlON 01? BACTERIA 161 Colony Study. — Cultures are usually incubated for from twelve to forty-eight hours. Considerable aid to the recognition of species ia derived from the observation of both the speed of growth andjbhe ap- pearance of the colonies. It is therefore necessary to proceed in the study of developed co onies in a systematic way. The development of colonies should be observed in all cases both upon gelatin and upon agar. In forming any judgment about colonies, the acidity or alkalinity, and the special constitution of the media should be taken into account. The colonies are carefully examined with a hand lens and with the low power (Leitz No. 2, Zeiss AA, Ocular No. 2) of the microscope. The colonies should be observed as to size, outline, transparency, texture, color, and elevation from the surface of the media. Much information, Fig. 45. — Variations in the Conformation of the Borders of Bacterial Colonies. (After Chester.) also, can be obtained by observing .whether a colony appears dry, mucoid, or glistening, like a drop of moisture. By a careful obser- vation of these points, definite differentiation, of course, can not usu- ally be made, but much corroborative evidence can be obtained which may guide us in the methods to be adopted for further identification and for a final summing up of species characteristic as a whole. The Counting of Bacteria. — It is often necessary to determine the number of bacteria per c.c. contained in water, milk, or other substances. For this purpose definite quantities of the material to be analyzed are mixed with gelatin or agar and poured into Petri plates. The exact dilutions of the suspected material must largely depend upon the number of germs which one expects to find in it. The plates, if prepared with gelatin, are allowed to develop at room temperature for twenty-four to 162 BIOLOGY AND TECHNIQUE forty-eight hours. If agar has been used, they are usually placed in the incubator at 37.5° C. At the end of this time, the colonies which have developed are enumerated. For this purpose, a Petri dish is placed upon a Wolffhugel plate. This plate consists of a disk or square of glass which is divided into small squares of one square centimeter each. Diagonal lines of these sq\iares running at right angles to each other are subdivided into nine divisions each in order to facilitate counting when the colonies are unusually abundant. The Petri dish is placed upon the plate in such a way that the center of the dish corresponds to the center of the plate. Fig. 46. — WolffhugeIj Counting Plate. The colonies in a definite number of squares are then counted. The greater the number of squares that are counted the more accurate the estimation will be. When the growth is so abundant that only a limited number of squares can be counted, these should be chosen as much as possible from different parts of the plate, and in practice one counts usually six squares in one direction and six at right angles to these, so as to preclude errors arising from unequal distribution. The final calcu- lation is then made by ascertaining the average number of colonies con- tained in each square centimeter. If standard Petri dishes have been METHODS USED IN CULTIVATION OF BACTERIA 163 used, this is multiplied by 63.6, the number of squares in the area of the dish, and then by the dilution originally used. Thus if twelve squares have been counted with a total number of one hundred and forty-four colonies — the average for each square is twelve. Twelve times 63.6 is 763.2, which represents the total number of colonies in the plate. Now if 0.1 c.c. of the original material (water or milk) has been plated, this material may be assumed to have contained 10 X 763.2, or 7,632 bacteria to each cubic centimeter. If dishes of an unusual size are employed, the square area must be ascertained by measuring the radius and multiplying its square by n (TT X R2 = area) ( 7:= 3.141592). CHAPTER IX METHODS OF DETERMINING BIOLOGICAL ACTIVITIES OF BACTERU ANIMAL EXPERIMENTATION Gas Formation. — Bacteria of many varieties produce gas from the proteid and the carbohydrate constituents of their environment. Gas formation can be observed in a very simple manner by making stab cultures in gelatin or agar containing the fermentable nutrient substances. In such cultures bubbles of gas will form along the track of the inoculation, or, in the case of such semisolid media as the tube medium of Hiss, will spread throughout the tube. In the case of some anaerobes gas formation in stab cultures will occur to such an extent that the medium v/ill split and break. It should be borne in mind in carrying out such methods that air is readily carried into the medium with the inoculating needle or loop by splitting of the medium, also that media which have been stored in the cold may absorb air. Expansion of the air in such tubes may simulate small amounts of gas formation and lead to error. It is advisable, therefore, whenever making stab inoculations with the above purpose, to heat the media and rapidly cool them before use. A more accurate method of gas determination is by the use of fer- mentation tubes, such as those devised by Smith. The gas which is formed collects in the closed arm of the fermentation tube and may be quantitatively estimated. The fermentation, with gas production, of certain substances such as carbohydrates, may be determined by adding these materials in a pure state to the media before inoculation with organisms. In the case of carbohydrates this method has proved of great differ- ential value, since the power of splitting specific carbohydrates with gas production is a species characteristic of great constancy for many forms of bacteria. Analysis of Gas Formed by Bacteria. — Carbon Dioxide. — For the estimation both qualitatively and roughly quantitatively of carbon dioxide produced by bacteria, cultures are grown in fermentation tubes containing sugar-free broth (see page 125) to which one per cent of pure dextrose, lactose, saccharose, or other sugars has been added. The tubes are incubated until the column of gas formed in the closed arm no longer X64 DETERMINING BIOLOGICAL ACTIVITIES OF BACTERU 165 increases <,cwenty-four to forty-eight hours). The level of the fluid b the closed arm is then accurately marked and the column of gas Daeasured. The bulb of the fermentation tube is then completely filled with y NaOH solution, the mouth closed with a clean rubber stopper, and the bulb inverted several times in order to mix the gas with the soda solution. The tube is then again placed in the upright position, allow- ing the gas remaining to collect in the closed arm. The gas lost may be roughly estimated as consisting of CO 2. Hydrogen. — The gas remaining, after removal of the CO 2 in the pre- ceding experiment, at least when working with carbohydrate solutions, Fig. 47. — ^Types op Fermentation Tubes. may be estimated as hydrogen. When allowed to collect near the mouth, further evidence of its being hydrogen may be gained by exploding it with a lighted match. Hydrogen Sulphide (H2S, Sulphuretted hydrogen). — In alkaline media sulphuretted hydrogen, if formed, will not collect as gas, but will form a sulphide with any alkali in the solution. For the estimation of the formation of hydrogen sulphide, bacteria are cultivated in a strong pepton solution to which 0.1 c.c. of a one per cent solution of ferric tartrate or lead acetate has been added. The addition of these substances gives rise to a yellowish precipitate in the bottom of the tubes. If, on 166 BIOLOGY AND TECHNIQUE subsequent inoculation, the bacteria produce Ho S, this precipitate will turn black. The solution recommended by Pake for this test is prepared as follows: 1. Weigh out 30 grams of pepton and emulsify in 200 c.c. of tap water at 60° 0. 2. Wash into a liter flask with 80 c.c. tap water. 3. Add sodium chloride 5 grams and sodium phosphate 3 grams. 4. Heat at 100° C. for 30 minutes, to dissolve pepton. 5. Filter through paper. 6. Fill into tubes, 10 c.c. each, and to each tube add 0.1 c.c. of a one per cent solution of ferric tartrate or lead acetate. These solutions should be neutral. 7. Sterilize.* Accurate quantitative gas analyses of bacterial cultures can be made only by the more complicated methods used in chemical labora- tories for quantitative gas analysis. The gas, in such cases, is collected in a bell jar mounted over mercury, and subjected to analysis by the usual method described in works on analytical chemistry. Acid and Alkali Formation by Bacteria. — Many bacteria produce acid or alkaline reactions in culture media, their activity in this respect depending to a large extent upon the nature of the nutrient material. Many organisms which on carbohydrate media produce acid will give rise to alkali if cultivated upon media containing only proteids. Information as to the production of acid or alkali can be obtained by the addition of one of a variety of indicators to neutral media. The indicators most often employed for this purpose are litmus or neutral red. Changes in the color of these indicators show whether acids or alkalis have been produced. Great help in differentiation is obtained by adding chemically pure carbohydrates to media to which litmus has been added and then de- termining whether or not acid is formed from the substances by the microorganisms. These tests have been of special importance in the differentiation of the typhoid-colon groups of bacilli. Quantitative estimation of the degree of acidity or alkalinity pro- duced by bacteria may be made by careful titration of definite volumes of the medium before and after bacterial growth has taken place. The variety of acid formed by bacteria depends largely upon the nature of the nutrient medium. The acids most commonly resulting from bacterial growth are lactic, acetic, oxalic, formic, and hippuric acids. Qualitative and quantitative estimation of these acids may be made by any of the methods employed by analytical chemists. * Quoted from Eyre, " Bact. Technique/' Phila., 1903. DETERMINING BIOLOGICAL ACTIVITIES OF RACTERIA 167 Indol Production by Bacteria. — Many bacteria possess the power of producing indol. Though formerly regarded as a regular accompani- ment of proteid decomposition, later researches have shown that indol production is not always coexistent with putrefaction processes and occurs only when pepton is present in the pabulum. Indol formation by bacteria is determined by the so-called nitroso- indol reaction. Organisms are grown in sugar-free pepton broth or in the pepton-salt bouillon of Dunham. (See page 126.) Media containing fermentable substaijces are not favorable for indol production since acids interfere with its formation. The cultures are usually incubated for three or four days at 37° C. At the end of this time, ten drops of con- centrated sulphuric acid are run into each tube. If a pink color appears, indol is present, and we gather the additional information that the microorganism in question has been able to form nitrites by reduction (e.g., cholera spirillum). If the pink color does not appear after the addition of the sulphuric acid alone, nitrites must be supplied. This is done by adding to the fluid about 1 c.c. of a 0.01 per cent aqueous solution of sodium nitrite. The sodium nitrite solu- tion does not keep for any length of time and should be freshly made up at short intervals. Phenol Production by Bacteria. — Phenol is often a by-product in the course of proteid cleavage by bacteria. To determine its- presence in cultures, bacteria are cultivated in flasks containing about 50-100 c.c. of nutrient broth. After three to four days' growth at 37° C, 5 c.c. of concentrated HCl are added to the culture, the flask is connected with a condenser, and about 10-20 c.c. are distilled over. To the distillate may be added 0.5 c.c. of Millon's reagent (solution of mercurous nitrate in nitric acid), when a red color will indicate phenol; or 0.5 c.c. of a ferric chloride solution, which will give a violet color if phenol is present. Reducing Powers of Bacteria. — The power of reduction, possessed by many bacteria, is shown by their ability to form nitrites from nitrates. This is easily demonstrated by growing bacteria upon nitrate broth (see page 126). Bacteria are transferred to test tubes containing this solution and allowed to grow in the incubator for four or five days. The presence of nitrites is then chemically determined.^ ^ We are indebted to Dr. J. P. Mitchell, of Stanford University, for the following technique for nitrite tests: I. SulphauiUc Acid. — Dissolve 0.5 g. in 150 c.c. of acetic acid of Sp. Gr. 1.04. 168 BIOLOGY AND TECHNIQUE In bacteriological work 4 c.c. of the culture fluid is poured into a clean test tube, and to it are gradually added 2 c.c. of the mixed test solutions. A pink color indicates the presence of nitrites, the intensity of the color being proportionate to the amount of nitrite present. The reducing powers of bacteria may also be shown by their ability to decolorize litmus, methylene-blue, and some other anilin dyes, which on abstraction of oxygen form colorless leukobases. Enzyme Action. — The action of the enzymes produced by bacteria may be demonstrated by bringing the bacteria, or their isolated fer- ments, into contact with the proper substances and observing both the physical and chemical changes produced. In obtaining enzymes free from living bacteria, it is convenient to kill the cultures by the addition either of toluol or of chloroform. Both of these substances will destroy the bacteria without injuring the enzymes. Enzymes may also be obtained separate from the bodies of the bacteria by filtration. Proteolytic Enzymes. — The most common evidences of proteolytic enzyme action observed in bacteriology are the liquefaction of gelatin, fibrin or coagulated blood-serum, and the peptonization of milk. This may be observed both by allowing the proper bacteria to grow upon these media, or by mixing sterilized cultures with small quantities of these substances.^ The products of such a reaction may be separated from the bacteria by filtration and then tested for pepton by the biuret reaction. Proteolytic ^ enzymes may also be determined by growing the bac- teria upon fluid media containing albumin solutions, blood serum, or milk serum, then precipitating the proteids by the addition of ammonium sulphate (about 30 grams to. 20 c.c. of the culture fluid) and warming between 50 to 60° C. for thirty minutes. The precipitate is then filtered off, the filtrate made strongly alkaline with NaOH, and a few drops of copper sulphate solution added. A violet color indicates the pres- ence of pepton — proving proteolysis of the original albumin. (Acetic acid of 1.04 prepared by diluting 400 c.c. of cone, of Sp. Gr. 1.75 with 70C c.c. of water.) II. A-Naphthylamin. — Dissolve 0.1 g. in 20 c.c. of water, boil, filter (if necessary), and to clear filtrate add 180 c.c. of acetic acid, Sp. Gr. 1.04. The solutions are kept separate and mixed in equal parts just before use. In carrying out the test, put 2 c.c. of each reagent in a test tube and add substance to be tested. (In ordinary water analysis use 100 c.c.) Cover tube with watch glass and set in warm water for 20 minutes. Observe presence or absence of pink color promptly. Always run a blank on the distilled water used for rinsing to avoid errors due to nitrites in the water, or in the air of the laboratory.. ^Bitter, Archiv f. Hyg., v. 1886. * Hanhin and Weshrook^ Ann. Past., vi., 1892, DETERMINING BIOLOGICAL ACTIVITIES OF BACTERIA 169 u '^^m DiASTATic Enzymes. — The presence of diastatic ferments may be determined by mixing broth cultures of the bacteria with thin starch paste. It is necessary that both the cultures and the starch paste be absolutely free from sugar. After remaining in the incubator for five or six hours, the fluid is filtered and the filtrate tested by methods used for determining the presence of sugars. Inverting Ferments. — Inverting ferments are determined by a pro- cedure similar to the above in principle. Dilute solutions of cane sugar are mixed with old cultures or culture filtrates of the respective bacteria and the mixture allowed to stand. It is then filtered, and the filtrate tested for glucose, preferably by Fehling's solution. ANIMAL EXPERIMENTATION In the study of pathogenic microorganisms, animal experi- mentation is essential in many instances. The virulence of any given organism for a definite ani- mal species and the nature of the lesions produced are character- istics often of great value in differentiation. Isolation, more- over, of many bacteria is greatly facilitated by the inoculation of susceptible animals and recovery of the pathogenic organism from the heart's blood or from the lesions produced in various organs. That investigations into the phenomena of immunity would be absolutely impossible without the use of animal inoculation is, of course, self- evident, for by this method only can the action of bacteria in relation to living tissues, cells, and body-fluids be observed . The animals most commonly employed for such observations are guinea-pigs, white mice, white rats, and rabbits. The method of inoculation may be either subcutaneous, intrapleural, intraperi- toneal, intravenous, or subdural, etc. Tt must be borne in mind always that the mode of inoculation may influence the course of an O' \^ Fig. 48. — ^Types op Gelatin Liquefac- tion BY Bacteria. 170 BIOLOGY AND TECHNIQUE infection no less than does the virulence of the microorganism or the size of the dose. Inoculations are made with some form of hypodermic needle fitted to Fig. 49. — Intraperitoneal Inoculation of Rabbit, Fig. 50. — Intravenous Inoculation of Rabbit. a syringe. The most convenient syringes are the all-glass Luer or the Debove syringes, which, however, are expensive. . Any form of steriliz- able syringe may be used. In making inoculations the hair of the DETERMINING BIOLOGICAL ACTIVITIES OF BACTERIA 171 animal should be clipped and the skin disinfected with carbolic acid or alcohol. Subcutaneous inoculations are most conveniently made in the abdom- FiG. 51. — Intraperitoneal Inoculation of Guinea-pig. Fig 52. — Guinea-pig Cage. inal wall, where the skin is thin. After chpping and sterilizing, the skin is raised between the fingers of the left hand and the needle plunged 172 BIOLOGY AND TECHNIQUE in obliquely so as to avoid penetrating the abdominal wall and entering the peritoneum. In making intraperitoneal inoculations, great care must be exercised not to puncture the gut. This can be avoided by passing the needle first through the skin in an oblique direction, then turning it into a posi- tion more vertical to the abdomen and perforating the muscles and perito- neum by a very short and carefully executed stab. Intravenous inoculations in rabbits are made into the veins running along the outer margins of the ears. The hair over the ear is clipped and the animal held for a short time head downward so that the vessels of the head may fill with blood. An assistant holds the animal firmly in Fig. 53. — Rabbit Cage. a horizontal position, the operator grasps the tip of the ears with the left hand, and carefully passes his needle into the vein in the direction as nearly as possible parallel to its course. (See Fig. 50.) Mice are usually inoculated under the skin near the base of the tail. They may be placed in a jar over which a cover of stiff wire-gauze is held. They are then grasped by the tail, by which they are drawn up between the side of the jar and the edge of the wire cover, so that the lower end of the back shall be easily accessible. The skin is then wiped with a piece of cotton dipped in carbolic solution and the needle is in- serted. Great care must be exercised to avoid passing the needle too close to the vertebral column. Mice are extremely delicate, and any injury to the' spine usually causes immediate death. DETERMINING BIOLOGICAL ACTIVITIES OF BACTERIA 173 With proper care mice or rats may be easily injected intravenously if a sufficiently fine needle is used. There are four superficially placed veins running along the tail, which stand out prominently when rubbed with cotton moistened with xylol. Into these the injections are made. When inoculating rats or guinea-pigs with bacillus pestis the KoUe vaccination method is used. The skin is merely shaved and a loopful of the culture vigorously rubbed into the shaven area. The various forms of animal holders which have been devised are rarely necessary in bacteriological work unless working unassisted, im- mobilization of the animals being easily accomplished by the hands of a skilled assistant. Autopsies upon infected animals must be carefully made. The ani- mals are tied, back down, upon pans fitted in the corners with clamps for the strings. They are then moistened either with hot water or with a weak solution of carbolic acid, so that contamination by hair may be avoided. A median cut is made, the skin is carefully dissected back, and the body cavities are opened with sterile instruments. Cultures may then be taken from exudates, blood, or organs under precautions similar to those recommended below for similar procedures at autopsy upon man. Inoculated animals should be, if possible, kept separate from healthy mimals. Rabbits and guinea-pigs are best kept in galvanized iron-wire cages, which are fitted with floor-pans that can be taken out and cleaned and sterilized. Mice ma)^ be kept in battery jars fitted with perforated metal covers. The mice should be supplied with large pieces of cotton upon batting since they are delicately susceptible to cold. CHAPTER X THE BACTERIOLOGICAL EXAMINATION OF MATERIAL FROM PATIENTS In making bacteriological examinations of material taken from living patients, or at autopsy, the validity of result is as fully dependent upon the technique by which the material is collected, as upon proper manipulation in the later stages of examination. Material taken at autopsy should be, if possible, directly transferred from the cadaver to the proper culture media. If cultures are to be taken from the liver, spleen, or other organs, the surface of the organ should first be seared with a hot scalpel and an incision made through the cap- sule of the organ in the seared area, with the same instrument. The platinum needle can then be plunged through this incision and material for cultivatipn be taken with little chance of surface contamination. When blood is to be transferred from the heart, the heart muscle may be incised with a hot knife, or else the needle of a hypodermic syringe may be plunged through the previously seared heart muscle and the blood aspirated. The same end can be accomplished by means of a pointed, freshly prepared Pasteur pipette. In taking specimens of blood at au- topsy it is safer to take them from the arm or leg, by allowing the blood to flow into a broad, deep cut miade through the sterilized skin, than from- the heart, since it has been found that post-mortem contamination of the heart's blood takes place rapidly, probably through the large veins from the lungs. Exudates from the pleural cavities, the pericardium, or the peritoneum may be taken with a sterilized syringe or pipette. Materials collected at the bedside or in the operating-room should be transferred directly to the proper media or else into sterile test tubes and so sent to the laborator}^ When the material is scanty, it may be collected upon a sterile cotton swab, which should be immediately re- placed in the sterilized containing tube and sent to the laboratory. Syringes, when used for the collection of exudates or blood, should be of some variety which is easily sterilizable by dry heat, or boiling. Most convenient of the forms in common use are the all-glass " Luer " syringe, or the cheaper ''Sub-Q" model. Instruments which can be sterilized only by chemical disinfectants should not be used. When 174 EXAMINATION OF MATERIAL FROM PATIENTS 175 fluids are collected for bacteriological examination, such as spinal fluid, paracentesis fluid, or pleural exudate, it is convenient to have them taken directly into sterilized centrifuge tubes, since it is often necessary to concentrate cellular elements by centrifugalization. By immediate col- lection in these tubes, the danger of contamination is avoided. Examination of Exudates. — Pus. — Pus should first be examined i^aorphologically by some simple stain, such as gentian-violet, and by the Gram stain. It is convenient, also, to stain a specimen by Jenner's stain, in order to show clearly the relation of bacteria to the cells. Such morphological examination not only furnishes a guide to future manipulation, but supplies a control for the results obtained by cultural methods. Specimens of the pus are then transferred to the proper media, and pour-plates made or streaks made upon the surface of previously prepared agar or serum-agar plates. A guide to the choice of media is often found in the result of the morphological examination. In most cases, it is well also to make anaerobic cultures by some simpler method. (See page 148 et seq.) ■ The colonies which develop upon the plates should be studied under the microscope, and specimens from the colonies transferred to cover-glasses and slides for morphological examination and to the various media for further growth and identification. Animal inoculation and agglutination tests must often also be resorted to. A knowledge of the source of the material may furnish considerable aid in making a bacteriological diag- nosis, though great caution in depending upon such aid is recommended. In the examination of peritoneal, pericardial, or pleural exudate it is often advantageous to use the sediment obtained by centrifugalization. A differential count of the cells present may be of aid in confirming the bacteriological findings. Morphological examination and cultural exam- ination are made as in the case of pus. Specimens should also in these cases be stained for tubercle bacilli. Whenever morphological exami- nations of such fluids are negative, no bacteria being found, and especially when among the cellular elements the lymphocytes preponderate, the search for tubercle bacilli should be continued by means of animal inocu- lation. Guinea-pigs should be inoculated intraperitoneally from speci- mens of the fluid. The animals will usually die within six to eight weeks, but can be killed and examined at the end of about six weeks if they remain alive. The chances for a positive result are considerably increased if the fluid is set away in the ice-chest until a clot has formed and the animals are inoculated with the material from the broken-up clot j3 176 BIOLOGY AND TECHNIQUE The routine examination of spinal fluid is best made upon the sedi- ment of centrifugalized specimens. The microorganisms with which we deal most frequently in this fluid are the meningococcus, the pneumococ- cus, the streptococcus, and the tubercle bacillus. If morphological ex- amination reveals bacteria resembling the first three of these in appear- ance and staining-reaction, surface smears should preferably be made upon plates of serum agar, blood agar, or upon tubes of Loeffler's co- agulated blood-serum. Failure to find organisms morphologically does not exclude their presence and careful cultivation should be done in all cases. When organisms are not found by simple morphological examina- tion and the fluid and sediment are scanty, specimens should be stained by the Ziehl-Neelson method for tubercle bacilli. In such cases it is often of advantage to set away the specimen until a thin thread-like clot of fifcrin has formed in the bottom of the tube. In smears of such a clot, tubercle bacilli are found with far greater ease than they are found in centrifugalized specimens. If these examinations are without result, inoculation of guinea-pigs should be resorted to. Examination of Urine. — Bacteriological examination of the urine is of value only when specimens have been taken with sterile catheters, and care has been exercised in the disinfection of the external genitals. Many of the numerous finds of bacillus coli in urine are unquestionably due to defective methods of collecting material. Urine should be cen- trifugalized and the sediment examined morphologically and pour- plates made and surface smears made upon the proper media. If necessary, animal inoculation may be done. In examining urine for tubercle bacilli, special care should be taken in staining methods so as to differentiate from Bacillus smegmatis. Examination of Feces. — Human feces contain an enormous num- ber of bacteria of many varieties. Klein,^ by special methods, es- timated that there were about 75,000,000 bacteria in one milligram of feces. It has been a noticeable result of all the investigations upon the feces, that although enormous numbers can be counted in morpho- logical specimens, only a disproportionately smaller number can be cultivated from the same specimen. This is explicable upon the ground that special culture media are necessary for many of the species found in intestinal contents and upon the consideration that many of the bacteria which are present in the morphological specimen are dead, show- mg that there are bactericidal processes going on in some parts of the » Klein, Ref. Cent. f. Bakt., I, xxx, 1901. EXAMINATION OF MATERIAL FROM PATIENTS 177 intestinal tract, possibly through the agency of intestinal secretions, bile, and the action of the products of metabolism of the hardier species present. By far the greater part of the intestinal flora consists of mem- bers of the colon group, bacilli of the lactis aerogenes group^ Bacillus faecalis alkaligenes, Bacillus mesentericus, and relatively smaller num- bers of streptococci, staphylococci, and Gram-positive anaerobes. Many other species, however, may be present without being necessarily con- sidered of pathological significance. Certain writers have recently laid much stress upon a preponderance of Gram-positive bacteria in speci- mens of feces, claiming that such preponderance signifies some form of intestinal disturbance. Herter ^ has recently advanced the opinion that the presence of Bacillus aerogenes capsulatus in the intestinal canal is definitely associated with pernicious anemia. The determination of these bacilli in the stools is made both by morphological examination by means of Gram stain and by isolation of the bacteria. Such isola- tion is easily done by the method of Welch and Nuttal.^ A suspension of small quantities of the feces in salt solution is made and 1 c.c. of the filtered suspension is injected into the ear vein of a rabbit. After a few minutes the rabbit is killed and placed in the incubator. After five hours of incubation, the rabbit is dissected, and if the Welch bacillus has been present in the feces, small bubbles of gas will have appeared in the liver from which the bacilli may be cultivated in anaerobic stab- cultures. Bacteriological examination of feces is most often undertaken for the isolation of Bacillus typhosus. This is accomplished with a great deal of difficulty because of the overwhelming numbers of colon bacilli which easily outgrow the typhoid germs, and because of the similarity of their colonies in most media. Many methods have been devised for this purpose, all of which depend upon the use of special media aimed at the inhibition of colon and other bacilli and the production of recog- nizable differences in the colonies of typhoid and colon bacilli. Such media are those of Eisner, Hiss, Conradi-Drigalski, Loeffler, Hesse, and others, which are described in the section upon special media. (See page 133.) The methods of using these media will be found described in the chapter on Bacillus typhosus (p. 399.) Cholera spirilla may be recognized in and isolated from the stools of patients by morphological examination, and by cultivation. (See section on Sp. cholerse.) 1 Herter, " Common Bacterial Infections of the Digestive Tract," N. Y., 1907. 2 Welch and Nuttal. See ref. p. 469. 178 BIOLOGY AND TECHNIQUE For the isolation of dysentery bacilli from feces, no satisfactory special methods have as yet been devised. Here we can depend only up- on careful plating upon agar and gelatin and extended colony "fishing," and the study of pure cultures. The complete absence of motility of these bacteria is of much aid in such identification. The determination of tubercle bacilli in stools is difficult and of questionable significance, in that they may be present in people suffer- ing from pulmonary tuberculosis as a consequence of swallowing sputum or infected food, and in that there may be other acid-fast bacilli, such as the timothy bacillus, present. Blood Cultures. — The diagnosis of septicemia can be positively made during life only by the isolation of bacteria from the blood. Such exam- inations are of much value and are usually successful if the technique is properly carried out. A large number of methods are recommended, the writers giving, however, only the one which they have found successful and simple for general use. The blood is taken by preference from the median basilic vein of the arm. If, for some reason (both forearms having been used for saline infusion), these veins are unavailable, blood may be taken from the internal saphenous vein as it turns over the internal malleolus of the ankle joint. The skin over the vein should be prepared before the specimen is taken by painting with iodine, as for a surgical operation. The syringe which is used should be of some sterilizable variety (the most convenient the Luer model), which is easily manipulated and does not draw with a jerky, irregular motion. Its capacity should be at least 10 e.c. It may be sterilized by boiling for half an hour, or preferably, when all-glass syringes are used, they may be inserted into potato-tubes and sterilized at high temperature in the hot-air chamber. Before drawing the blood, a linen bandage is wound tightly about the upper arm of the patient in order to cause the veins to stand out prominently. "When the veins are plainly in view, the needle is plunged through the skin into the vein in a direction parallel to the vessel and in the direction of the blood-stream. After perforation of the skin, while the needle is groping for the vein, gentle suction may be exerted with the piston. Great care should be exercised, however, that the piston is not allowed to slip back, and air be, by accident, forced into the vessel. In most cases no suction is necessary, the pres- sure of the blood being sufficient to push up the piston. After the blood has been drawn, it should be immediately transferred to the proper media. Epstein has recently recommended the mixture of the blood EXAMINATION OF MATERIAL FROM PATIENTS 179 with sterile two per cent ammonium oxalate solution in test tubes, by which means the clotting is prevented, and transfers can be made more leisurely to culture media. While this method is convenient in cases where blood must be taken at some distance from the laboratory, it is Fig. 54. — Blood-Culture Plate Showing Streptococcus Colonies. Note halo of hemolysis about each colony. preferable, whenever possible, to make cultures from the blood im- mediately at the bedside. The choice of culture media for blood cultures should, to a certain extent, be adapted' to each individual case. For routine work, it is best to employ glucose-meat-infusion agar and glucose-meat-infusion broth. At least six glucose-agar tubes should be melted and immersed in water at 45° C. Before the blood is mixed with the medium, the agar should be 180 BIOLOGY AND TECHNIQUE cooled to 41° in order that bacteria, if present, may not be injured by the heat. The blood is added to the tubes in varying quantities, ranging from 0.25 to 1 c.c. each, in order that different degrees of concentration may be obtained. Mixing is accomplished by the usual dipping and rotary motion, the formation of air-bubbles being thus avoided. The mouth of each test tube should be passed through the flame before pour- ing the contents into the plates. Three flasks of glucose broth, contain- ing 100 to 150 c.c. of fluid each, should be inoculated with varying quantities of blood — at least one of the flasks containing the blood in high dilution. The most stringent care in the withdrawal and replace- ment of the cotton stoppers should be exercised.' The writers have found it convenient to use, in place of one of these flasks, one containing, in addition to the glucose, 1 gm. of powdered calcium carbonate. This insures neutrality, permitting pneumococci or streptococci, which are sensitive to acid, to develop and retain their vitality. In making blood cultures from typhoid patients, Buxton and Cole- man ^ have obtained excellent results by the use of pure ox-bile con- taining ten per cent of glycerin and two per cent of peptone in flasks. The writers have had no difficulty in obtaining typhoid cultures by the use of slightly acid meat-extract broth in flasks containing 200 or more c.c. to which comparatively little blood has been transferred. Anaerobic Blood Cultures. — These cultures may be taken by mixing blood in deep tubes with glucose-ascitic agar, covering with albolene and putting into Novy jars. In estimating the results of a blood culture, the exclusion of con- tamination usually offers little difficulty. If the same microorganism appears in several of the plates and flasks, if colonies upon the plates are well distributed within the center and under the surface of the medium, and if the microorganisms themselves belong to species which commonly cause septicemia, such as streptococcus and pneumococcus, it is usually safe to assume that they have emanated from the patient 's circulation. When colonies are present in one plate or in one flask only, when they are situated only near the edges of a plate or upon the surface of the medium, and when they belong to varieties which are often found saprophytic upon skin or in air, they must be looked upon with suspicion. It is a good rule to look upon all staphylococcus albus cultures skeptically. ^ Small Florence flasks are preferable to the Erlenmeyer flasks usually employed. 2 Buxton and Coleman, Am. Jour, of Med. Sci., 1907. SECTION II INFECTION AND IMMUNITY CHAPTER XI FUNDAMENTAL FACTORS OF PATHOGENICITY AND INFECTION When microorganisms gain entrance to the animal or human body and give rise to disease, the process is spoken of as infection. Bacteria are ever present in the environment of animals and human beings and some find constant lodgment on various parts of the body. The mouth, the nasal passages, the skin, the upper respiratory tract, the conjunctivae, the ducts of the genital system, and the intestines are invariably inhabited by numerous species of bacteria, which, while sub- ject to no absolute constancy, conform to more or less definite charac- teristics of species distribution for each locality. Thus the colon organ- isms are invariably present in the normal bowel, Doderlein's bacillus in the vagina. Bacillus xerosis in many normal conjunctivae, and staph}^- lococcus, streptococcus, various spirilla, and pneumococcus in the mouth. In contact, therefore, with the bodies of animals and man, there is a large flora of microorganisms, some as constant parasites, others as transient invaders; some harmless saprophytes and others capable of becoming pathogenic. It is evident, therefore, that the production of an infection must depend upon other influences than the mere presence of the micro- organisms and their contact with the body, and that the occurrence of the reaction — for the phenomena of infection are in truth reactions be- tween the germ and the body defenses — is governed by a number of important secondary factors. In order to cause infection, it is necessary that the bacteria shall gain entrance to the body by a path adapted to their own respective cultural requirements, and shall be permitted to proliferate after gaining a foot- hold. Some of the bacteria then cause disease by rapid multiplication, progressively invading more and more extensive areas of the animal tissues, while others may remain localized at the point of invasion and 181 182 INFECTION AND IMMUNITY exert their harmful action chiefly by local growth and the elaboration of specific poisons. The inciting or inhibiting factors which permit or i3rohibit an in- fection are dependent iri part upon the nature of the invading germ and in part upon the conditions of the defensive mechanism of the subject attacked. Bacteria are roughly divided into two classes, saprophytes and parasites. The saprophytes are those bacteria which thrive best on dead organic matter and fulfill the enormously important function in nature of reducing by their physiological activities the excreta and dead bodies of more highly organized forms into those simple chemical substances which may again be utilized by the plants in their con- structive processes. The saprophytes, thus, are of extreme importance in maintaining the chemical balance between the animal and plant kingdoms. Parasites, on the other hand, find the most favorable conditions for their development upon the living bodies of higher forms. While a strict separation of the two divisions can not be made, nu- merous species forming transitions between the two, it may be said that the latter class comprises most of the so-called pathogenic or disease-producing bacteria. Strict saprophytes may cause disease, but only in cases where other factors have brought about the death of some part of the tissues, and the bacteria invade the necrotic areas and break down the proteids into poisonous chemical sub- stances such as ptomains, or through their own destruction give rise to the liberation of toxic constituents of their bodies. It is necessary, therefore, that bacteria, in order to incite disease, should belong strictly or facultatively to the class known as para- sitic. It must not be forgotten, however, that the terms are relative, and that bacteria ordinarily saprophytic may develop parasitic and pathogenic powers when the resisting forces of the invaded subject ^re reduced to a minimum by chronic constitutional disease or other causes. Organisms that are parasitic, however, are not necessarily pathoge.ic, and there are certain more or less fundamental requirements which experience has taught us must be met by an organism in order that it may be infectious (or pathogenic) for any given animal; and by infec- tiousness is meant the ability of an organism to live and multiply in the animal fluids and tissues. For instance, an organism which is shown not to grow at the body temperature of warm-blooded animals may safely be assumed not to be infectious for such animals; and experience is FACTORS OF PATHOGENICITY AND INFECTION 183 gradually teaching us that strictly aerobic organisms, those thriving only in the presence of free oxygen and not able to obtain this gas in available combination from carbohydrates, can also be safely excluded from the infectious class. We have also learned that anaerobic organ- isms, although infectious when gaining entrance to tissues not abun- dantly supplied with blood, are practically unable to multiply in the blood stream and give rise to generalized infection. The pathogenic microorganisms differ very much among themselves , in the degree of their disease-inciting power. Such power is known as virulence. Variations in virulence occur, not only among different species of pathogenic bacteria, but may occur within the same species. Pneumococci, for instance, which have been kept upon artificial media or in other unfavorable environment for some time, exhibit less viru- lence than when freshly isolated from the bodies of man or ani- mals. It is necessary, therefore, in order to produce infection,- that the particular bacterium involved shall possess sufficient virulence. Whether or not infection occurs depends also upon the number of bacteria which gain entrance to the animal tissues. A small number of bacteria, even though of proper species and of sufficient virulence, may easily be overcome by the first onslaught of the defensive forces of the body. Bacteria, therefore, must be in sufficient number to overcome local defenses and to gain a definite foothold and carry on their life processes, before they can give rise to an infection. The more virulent the germ, other conditions being equal, the smaller the number necessary for the production of disease. The introduction of a single individual of the anthrax species, it is claimed, is often sufficient to cause fatal infection; while forms less well adapted to the parasitic mode of life will gain a foothold in the animal body only after.the introduction of large numbers. The Path of Infection. — ^The portal by which bacteria gain entrance to the human body is of great importance in determining whether or not disease shall occur. Typhoid bacilli rubbed into the abraded skin may give rise to no reaction of importance, while the same microorganism, if swallowed, may cause fatal infection. Conversely, virulent strepto- cocci, when swallowed, may cause no harmful effects, while the same bacteria rubbed into the skin may give rise to a severe reaction. Animals and man are protected against invasion by bacteria in various ways. Externally the body is guarded by its coverings of skin and mucous membranes. When these are healthy and undisturbed, microorganisms are usually held at bay. While tHs is true in a gen- eral way bacteria may in occasional cases pass through uninjured 184 INFECTION AND IMMUNITY skin and mucosa. Thus the Austrian Plague Commission found that guinea-pigs could be infected when plague bacilli were rubbed into the shaven skin, and there can hardly be much doubt of the fact that tubercle baciUi may occasionally pass through the intestinal mucosa into the lymphatics without causing local lesions. Even after bacteria of a pathogenic species, in large numbers and of adequate virulence, have passed through a locally undefended area in the skin or mucosa of an animal or a human being by a path most favor- ably adapted to them, it is by no means certain that an infection will take place. The bodies of animals and of man have, as we shall see, at their disposal certain general, systemic weapons of defense, both in the blood serum and the cellular elements of blood and tissues which, if normally vigorous and active, will usually overcome a certain number of the invading bacteria. If these defenses are abnormally depressed, or the invading microorganisms are disproportionately virulent or plen- tiful, infection takes place. Bacteria, after gaining an entrance to the body, may give rise merely to local inflammation, necrosis, and abscess formation. They may, on the other hand, from the local lesion, gain entrance into the lymphatics and blood-vessels and be carried freely into the circulation, where, if they survive, the resulting condition is known as bacteriemia or septicemia. Carried by the blood to other parts of the body, they may, under favor- able circumstances, gain foothold in various organs and give rise to secondary foci of inflammation, necrosis, and abscess formation. Such a condition is known as pijemia. The disease processes arising as the result of bacterial invasion may depend wholly or in part upon the mechanical injury produced by the process of inflammation, the dis- turbance of function caused by the presence of the bacteria in the capil- laries and tissue spaces, and the absorption of the necrotic products resulting from the reaction between the body cells and the micro- organisms. To a large extent, however, infectious diseases are char- acterized by the symptoms resulting from the absorption or diffusion of the poisons produced by the bacteria themselves. Bacterial Poisons. — It was plain, even to the earliest students of this subject, that mere mechanical capillary obstruction or the absorption of the products of a local inflammation were insufl^icient to explain the profound systemic disturbances which accompany many bacterial in- fections. The very nature of bacterial 'disease, therefore, suggested the presence of poisons. It was in his investigations into the nature of these poisons that FACTORS OF PATHOGENICITY AND INFECTION 185 Brieger * was led to the discovery of th6 ptomains. These bodies, first isolated by him from decomposing beef, fish, and human cadavers, have found more extended discussion in another section. Accurately classified, they are not true bacterial poisons in the sense in which the term is now employed. Although it is true that they are produced from pro- teid material by bacterial action, they are cleavage products derived from the culture medium upon the composition of which their nature intimately depends. The bacterial poisons proper, on the other hand, are specific products of the bacteria themselves, dependent upon the nature of the medium only as it favors or retards the full development of the physiological functions of the microorganisms. The poisons, pro- duced to a greater or lesser extent by all pathogenic microorganisms, may be of several kinds. The true toxins, in the specialized meaning which the term has acquired, are soluble, truly secretory products of the bacterial cells, passing from them into the culture medium during their life. They may be obtained free from the bacteria by filtration and in a purer state from the filtrates by chemical precipitation and a vari- ety of other methods. The most important examples of such poisons are those elaborated by Bacillus diphtheriae and Bacillus tetani. If cultures of these bacteria or of others of this class are grown in fluid media for several days and the medium is then filtered through porce- lain candles, the filtrate will be found toxic often to a high degree, while the residue will be either inactive or comparatively weak. Moreover, if the residue possesses any toxicity at all, the symptoms evidencing this will be different from those produced by the filtrate. There are other microorganisms, however, notably the cholera spirillum and the typhoid bacillus, in which no such exotoxins are formed. If these bacteria are cultivated and separated from the cul- ture fluid by filtration, as above, the fluid filtrate will be toxic to only a very slight degree, whereas the residue may prove very poisonous. In these cases, we are dealing, evidently, with poisons not secreted into the medium by the bacteria, but rather attached more or less firmly to the bacterial body. Such poisons, separable from the bacteria only after death by some method of extraction, or by autolysis, were termed by Pfeiffer endotoxins. The greater number of the pathogenic bacteria seem to act chiefly by means of poisons of this class. The first to call attention to the existence of such intracellular poisons was Buchner, who formulated his conclusions from the results of experiments made ^ ^Bneger, "Die Ptomaine," Berlin, 1885 and 1886. 186 INFECTION AND IMMUNITY ^ with a number of microorganisms, notably the Friedlander bacillus and Staphylococcus pyogenes aureus, with dead cultures of which he induced the formation of sterile abscesses in animals and symptoms of toxemia. The conception of ''endotoxins," received its clearest and most definite expression in the work of Pfeiffer^ on cholera poison. Some clarity of conception, based on visual perception, may possibly be gained by comparing some of the products of pathogenic bacteria with bacterial pigments and with insoluble interstitial or intercellular substance, which may be seen accompanying bacteria in cover-glass preparations. Soluble toxic secretions are to be compared to such pig- ments as the pyocyanin of Bacillus pyocyaneus, which is so readily soluble in culture media; endotoxins proper, to pigments confined to the bacterial cell, or at least, when secreted, being insoluble in culture media, such for instance as the well-known red pigment of Bacillus pro- digiosus, which may often be seen free among the bacteria in irregular red granules like carmine powder. That bodies such as this latter might be extruded from pathogenic bacteria and not be soluble in the usual cul- ture fluids, is not improbable, and the fact that more or less insoluble interstitial substances are not infrequent among bacteria is well known. In all bacterial bodies, after removal of toxins and endotoxins, a certain proteid residue remains which, if injected into animals, may give rise to localized lesions such as abscesses or merely slight temporary inflammations. The nature of this residue has been carefully studied, especially by Buchner, who has named it bacterial protein and he believes the substance to be approximately the same in all bacteria, without specific toxic action, but with a general ability to exert a positive chemotactic effect on the white blood cells, thereby causing the forma- tion of pus. The nature of the bacterial proteins is by no means clear, and it is still in doubt whether the separation of these substances from the endotoxins can be upheld. A number of bacteria may give rise to both varieties of poisons. Thus, recently, Kraus has claimed the discovery of a soluble toxin for the cholera spirillum and Doerr for the dysentery bacillus, both of which microorganisms were regarded as being purely of the endotoxin-pro- ducing type. It is plain, moreover, that occasionally it may be very difficult to distinguish between a soluble toxin and an endotoxin. In the filtration 1 P/eiJfer, Zeit. f. Hyg., xl, 1892. FACTORS OF PATHOGENICITY AND INFECTION 187 experiment recorded above, it might well be claimed that the toxicity of the filtrate, when not very strong, may depend upon an extraction of endotoxins from the bodies of the bacteria by the medium. The final test, in such instances, lies in the power of true toxins to stimu- late in animals the production of antitoxins; for, as we shall see later, the injection of true soluble toxins into animals gives rise to antitoxins, whereas the formation of such neutralizing bodies in the serum or plasma does not, it is claimed, follow the injection of endotoxins. This distinc- tion will become clearer as we proceed in the discussion of immunity. It must not be forgotten, however, that our knowledge of bacterial poisons is by no means complete, and that sharp distinctions as those given above must be regarded to a certain extent as tentative. In resistance to chemical action and heat, the various poisons show widely divergent properties. As a general rule, most true soluble toxins are delicately thermolabile, they are destroyed by moderate heating, and deteriorate easily on standing. Their chemical nature is by no means clear, but, on precipitation of toxic solutions with mag- nesium sulphate, these poisons come down together with the globulins. The nature of the endotoxins is still less clearly understood. Most of them, while less labile than the extracellular poisons, are, nevertheless, destroyed by exposure to 70° C. On the other hand, certain specific and powerful intracellular poisons, like those of the Gartner bacillus of meat poisoning, may undergo exposure to even 100° C. and still retain their toxic properties. The nature of each individual poison will be discussed in connection with its microorganism. The Mode of Action of Bacterial Poisons. — Close study of the toxic products of various microorganisms has shown that many of the bac- terial poisons possess a more or less deKnite selective action upon special tissues and organs. Thus, certain soluble toxins of the tetanus bacillus and Bacillus botulinus attack specifically the nervous system. Again, certain poisons elaborated by the staphylococci, the tetanus bacillus, the streptococci, and other germs, the so-called "hemolysins," attack primarily the red blood corpuscles. Other poisons again act on the white blood corpuscles; in short, the characteristic affinity of specific bacterial poisons for certain organs is a widely recognized fact. In explanation of this behavior, much aid has been given by the researches of Meyer, ^ Overton,^ Ehrlich,^ and others upon the causes for i Meyer, Arch. f. exper. Pathol., 1899, 1901. 2 Overton, "Studien iib. d. Narkose," Jena, 1901. '^hrlich, "Sauerstoffs-Bediirfniss des OrganismuS;" Berlin, 1885, 188 IMMUNITY AND INFECTION the analogous selective behavior of various narcotics and alkaloids. It seems probable, from the researches of these men, that the selective action of poisons depends upon, the ability, chemical or physical or both, of the poisons to enter into combination with the specifically affected cells. From the nature of the combinations formed, it seems not unlikely that the physical factors, such as solubility in the cell plasma, may also play an important part. Observations of a more purely bacteriological nature have tended to bear out these conclusions. Wassermann and Takaki,^ for instance, have shown that tetanus toxin, which specifically attacks the nervous system, may be removed from solution by the addition of brain sub- stance. Removal of the brain tissue by centrifugation leaves the solu- tion free from toxin. In the same way it has been shown that hemo- lytic poisons can be removed from solutions by contact with red blood cells, but only when the red blood cells of susceptible species are employed. Similar observations have been made in the case of leukocidin, a bacterial poison acting upon the white blood cells specifically.^ That bacterial poisons injected into susceptible animals rapidly disappear from the circulation is a fact which bears out the view that a combination between affected tissue and toxin must take place. Donitz,^ for instance, has shown that within four to eight minutes after the injections of certain toxins, considerable quantities will have dis- appeared from the circulation. Conversely, Metchnikoff ^ has ob- served that tetanus toxin injected into insusceptible animals (lizards) may be detected in the blood stream for as long as two months after administration. 1 Wassermann und Takaki, Berl, klin. Woch., 1898. ^ Sachs, Hofmeister's Beitrage, 11, 1902. ^Donitz, Deut. med. Woch., 1897. * Metchnikoff, "L'immimit^ dans les malad. infect," CHAPTER XII DEFENSIVE FACTORS OF THE ANIMAL ORGANISM GENERAL CONSIDERATIONS We have seen that the mere entrance of a pathogenic microorganism into the human or animal body through a breach in the continuity of the mechanical defenses of skin or mucosa does not necessarily lead to the development of an infection. The opportunities for such an invasion are so numerous, and the contact of members of the animal kingdom with the germs of disease is so constant, that if this were the case, sooner or later all would succumb. It is plain, therefore, that the animal body must possess mor subtle means of defense, by virtue of which pathogenic germs are, even after their entrance into the tissues and fluids, dis- posed of, or at least prevented from proliferating and elaborating their poisons. The power which enables the body to accomplish this is spoken of as resistance. When this resistance, which in some degree is com- mon to all members of the animal kingdom, is especially marked, it is spoken of as "immunity." From this it follows naturally that the terms resistance and immunity, as well as their converse, susceptibility, are relative and not absolute terms. Degrees of resistance exist, which are determined to a certain extent by individual, racial, or species peculiarities; and persons or animals are spoken of as immune when they are unaffected by an ex- posure or an inoculation to which the normal average individual of the same species wOuld ordinarily succumb. The word does not imply, however, that these individuals could not be infected with unusually virulent or large doses, or under particularly unfavorable circumstances. Thus, birds, while immune against the ordinary dangers of tetanus bacilli, may be killed by experimental inoculations with very large doses of tetanus toxin. ^ Similarly, Pasteur rendered naturally immune hens susceptible to anthrax by cooling them to a subnormal temperature, and Canalis and Morpurgo did the same with doves by subjecting them to starvation. 1 Quoted from Abel, KoUe und Wassermann, "Handbuch," etc. 189 190 INFECTION AND IMMUNITY Absolute immunity is exceedingly rare. The entire insusceptibility of cold-blooded animals (frogs and turtles) under normal conditions to inoculation with even the largest doses of many of the bacteria patho- genic for warm-blooded animals, and the immunity of all the lower animals against leprosy, are among the few instances of absolute immu- nity known.* Apart from such exceptional cases, however, resistance, immunity, and susceptibility must be regarded as purely relative terms. The power of resisting any specific infection may be the natural heritage of a race or species, and is then spoken of as natural immunity. It may, on the other hand, be acquired either accidentally or artificially by a member of an ordinarily susceptible species, and is then called acquired immunity. Natural Immunity. — Species Immunity. — It is well known that many of the infectious diseases which commonly affect man, do not, so far as we know, occur spontaneously in animals. Thus, infection with B. typhosus, the vibrio of cholera, or the meningococcus occurs in ani- mals only after experimental inoculation. Gonorrheal and syphilitic infection, furthermore, not only does not occur spontaneously, but is produced experimentally in animals with the greatest difficulty — the consequent diseases being incomparably milder than those occurring in man. Other diseases, like leprosy, influenza, and the exanthemata,^ have never been successfully transmitted to animals. Conversely, there are diseases among animals which do not spon- taneously attack man. Thus, human beings enjoy immunity against Rinderpest, and, to a lesser degree, against chicken cholera. Among animal species themselves great differences in susceptibility and resistance toward the various infections exist. Often-quoted ex- amples of this are the remarkable resistance to anthrax of rats and dogs, and the iminunity of the common fowl against tetanus. The factors which determine these differences of susceptibility and resistance among the various species are not clearly understood. It has been suggested that diet in some instances may influence these re- lations, inasmuch as carnivorous animals are often highly resistant to glanders, anthrax, and even tuberculous infections, to which herbiv- orous animals are markedly susceptible.^ It is likely, too, that the great differences between animals of various species in their metabolism, temperature, etc., may call for special cultural adaptation on the part Lubarsch, Zeit. f. klin. Mediz., xix. With the possible exception of smallpox. 3 Hahn, in Kolle und Wassermann, vol. iv. DEFENSIVE FACTORS OF THE ANIMAL ORGANISM 191 of the bacteria. The fact that the bacillus of avian tuberculosis — whose natural host has a normal body temperature of 40° C. and above — will grow on culture media at 40 to 50° C, whereas B. tuberculosis of man can not be cultivated at a temperature above 40° C, would seem to lend some support to this view. The difference between warm- and cold-blooded animals has already been noted. The necessity for cultural adaptation, too, would seem to be borne out by the great enhancement observed in the virulence of certain microorganisms for a given species after repeated passage through individuals of this species. Racial Immunity. — ^Just as differences in susceptibility and im- munity exist among the various animal species, so the separate races or varieties within the same species may display differences in their reac- tions toward pathogenic germs. Algerian sheep, for instance, show a much higher resistance to anthrax than do our own domestic sheep, and the various races of mice differ in their susceptibility to anthrax and to glanders. Similar racial differences are common among human beings. As a general rule, it may be said that a race among whom a certain disease has been endemic for many ages is less susceptible to this disease than are other races among whom it has been more recently introduced. The appalling ravages of tuberculosis among negroes, American Indians, and Esquimaux, bear striking witness to this fact. Conversely, the compar- ative immunity of the negro from yellow fever, a disease of the greatest virulence for Caucasians, furnishes . further evidence in favor of this opinion. It must not be forgotten, however, in judging of these rela- tions, that the great differences in the customs of personal and social hygiene existing among the various races may considerably affect the transmission of disease and lead to false conclusions. In so far as the statement made above is true, however, it seems to indicate that the endemic diseases have carried in their train a certain degree of inherited immunity. In other cases *— as in the instance of the malaria-immunity of negroes — the resistance seems to be acquired rather than inherited, for, as Hirsch was first to note, death from this disease occurred frequently among the children, while adult negroes were rarely attacked. DiFFEKENCES IN INDIVIDUAL RESISTANCE. — In bacteriological ex- perimentation with smaller test animals, a direct ratio may often exist between body weight and dosage in determining the outcome of an J Hahn, in Kolle and Wassermann, loc. cit. 14 192 INFECTION AND IMMUNITY infection, provided the mode of inoculation has been the same and the virulence of the germ not excessive. It would seem, therefore, that among these animals the difference in resistance in the face of an arti- ficial infection between individuals of the same race is very slight. In higher animals, however, especially in the case of man, the ex- istence of such apparent individual differences is a well-established fact, although in judging of them we must not forget that the conditions of infection are not subject to the uniformity and control which animal experimentation permits. Of a number of persons exposed to any given infection there are always some who are entirely unaffected and there are great variations in the severity of the disease in those who are attacked. In the absence of positive evidence in support of the direct inheritance of this individual immunity, the most reasonable explanation for such differences in resistance seems to lie in attrib- uting them to individual variations in metabolism or body chem- istry. Depressions, for instance, in the acidity of the gastric secretion would predispose to certain infections of gastro-intestinal origin. Ana- tomical differences, too, may possibly influence resistance. Thus, Birch-Hirschfeld believed that certain anomalous arrangements of the bronchial tubes predisposed to tuberculosis. Instances of transient susceptibility induced by physical or mental overwork, starvation, etc., should hardly be classified under this head- ing, since the conditions in such cases correspond simply to experi- mental depression of natural species or race resistance. Acquired Immunity. — It is a matter of common experience that many of the infectious diseases occur but once in the same individual. This is notably the case with typhoid fever, yellow fever, and most of the exanthemata, and is too general an observation to require extensive illustration. A single attack of any of the diseases of this class alters in some way the resistance of the individual so that further exposure to the infective agent is usually without menace, either for a limited period after the attack, or for life. Resistance acquired in this way is often spoken of as acquired immunity. The protection conferred by certain diseases against further attack was recognized many centuries ago, and there are records which show that attempts were made in ancient China and India to inoculate healthy individuals with pus from small-pox pustules in the hope of producing by this process a. mild forn>of the disease and its consequent immunity. Pasteur, before all others, thought philosophically about the phenom- ena of acquired immunity, and, with adequate knowledge, realized the DEFENSIVE FACTORS OF THE ANIMAL ORGANISM • 193 possibility of artifically bestowing immunity without inflicting the dangers of the fully potent infective agent. The first observation which, made by him purely accidentally, inspired the hope of the achievement of such a result, occurred during his experiments with chicken cholera. The failure of animals to die after inoculation with an old culture of the bacilli of chicken cholera, fully potent but a few weeks previously, pointed to the attenuation of these bacilli by their prolonged cultivation without transplantation. With this observation as a point of departure he carried out a series of investigations with the purpose of discovering a method of so weakening or attenuating various incitants of disease that they could be introduced into susceptible individuals without en- dangering life and yet without losing their property of conferring pro- tection. The brilliant results achieved by Jenner, many years before, in protecting against smallpox by inoculating with the entirely innocu- ous products of the pustules of cowpox furnished an analogy which gave much encouraging support to this prospect. The experimental work which Pasteur carried out to solve this prob- lem not only reaped a rich harvest of facts, but gave to science the first and brilliant examples of the application of exact laboratory methods to problems of immunity. ACTIVE IMMUNITY ^ Active Artificial Immunity. — The process of conferring protection by treatment with either an attenuated form or a sublethal quantity of the infectious agent of a disease, or its products, is spoken of as " active immunization." Whatever the method employed* the immunized individuals gain their power of resistance by the unaided reactions of their own tissues. They themselves take an active physiological part in the acquisition of this new property of immunity. For this reason, Ehrlich has aptly termed these processes "active immunization." There are various methods by which this can be accomplished, all of which were, in actual application or in principle, discovered by Pasteur and his associates, and can be best reviewed by a study of their work. Active Immunization with Attenuated Cultures. — In the course of his experiments upon chicken cholera, as mentioned above, Pasteur * 1 Pasteur, Compt. rend, de I'acad. des sci., 1880, t. xa. 194 INFECTION AND IMMUNITY accidentally discovered that the virulence of the bacilli of this disease was greatly reduced by prolonged cultivation upon artificial media. This was especially noticeable in broth cultures which had been stored for long periods without transplantation. By repeated injections of such cultures into fowl, he succeeded in rendering the animals immune against subsequent inoculations with lethal doses of fully virulent strains. During the same year, 1880, in which Pasteur published his observa- tions on chicken cholera, Toussaint ^ succeeded in immunizing sheep against anthrax by inoculating them with blood from infected animals, defibrinated and heated to 55° *C. for ten minutes. Toussaint wrongly believed, however, that the blood which had been used in his immuniza- tions was free from living bacteria. In repeating this work Pasteur showed that the protection in Toussaint's cases was conferred by living bacteria, the virulence of which had been reduced by their subjection to heat. In following out the suggestions offered by these experiments, Pasteur^ discovered that he could reduce the virulence of anthrax bacilli much more reliably than by Toussaint's method, by cultivating the organisms at increased temperatures (42° to 43° C.) . By this process of attenuation he was able to produce '^ vaccines " of roughly measurable strength, with which he succeeded in immunizing sheep and cattle. A successful demonstration of his discovery was made by him at Pouilly- le-Fort, soon after, upon a large number of animals and before a commis- sion of professional men. It is a fact well known to bacteriologists that certain of the pathogenic microorganisms, when passed through several individuals of the same animal species, become gradually more virulent for this species. In his studies on the bacillus of hog cholera, Pasteur observed that when this microorganism was passed through the bodies of several rabbits it gained in virulence for rabbits, but became less potent against hogs. He suc- ceeded, subsequently, in protecting hogs against fully virulent cultures by treating them with strains which had been attenuated by their passage through rabbits. A further principle of attenuation for purposes of immunization was, at about this time, contributed by Chamberland and Roux,* who re- 1 Toussaint, Compt. rend, de Tacad. des sci., 1880, t. xci. 2 Pasteur, Ch/imberland et Roux, Compt. rend, de Tacad. des sci., 1881, t. xeii. 3 Pasteur, Compt. rend, de I'acad. des sci., 1882, t. xcv. . * Chamberland et Roux, Compt. rend, de Tacad. des sci., 1882, t. xcvi. DEFENSIVE FACTORS OF THE ANIMAL ORGANISM 195 duced the virulence of anthrax cultures by growing them in the presence of weak antiseptics (carbolic acid 1 : 600, potassium bichromate 1 : 5,000, or sulphuric acid 1 : 200) . Cultivated under such conditions the bacilli lost their ability to form spores and became entirely avirulent for sheep within ten days. A similar result was later obtained by Behring ^ when attenuating B. diphtherise cultures by the addition of terchlorid of iodin. Active Immunization wim Sublethal Doses of Fully Virulent Bacteria. — ^The use of fully virulent microorganisms in minute quantities for purposes of immunization was first suggested by Chau- veau,2 and is naturally inappHcable to extremely virulent organisms like B. anthracis. The principle, however, is perfectly valid, and has been experimentally applied by many observers, notably by Ferran ^ in the case of cholera. A similar method proved of practical value in the hands of Theobald Smith and Kilborne * in prophylaxis against the protozoan disease, Texas fever. Active Immunization with Dead Bacteria. — Suggested by Chau- veau, the method of active immunization with gradually increasing doses of dead microorganisms has been successfully employed by various ob- servers, chief among whom are Pfeiffer, Brieger, Wright, and Wasser- mann. The method is especially useful against that class of bacteria in which the cell bodies (endotoxins) have been found to be incomparably more poisonous than their extracellular products (toxins). From a practical point of view, the method is of the greatest importance in routine laboratory immunization against B. typhosus. Vibrio cholerse asiaticse, B. pestis, and a number of other bacteria. In the therapy of human disease, this method has recently come into great prominence, chiefly through the work of Wright, whose investigations will be more fully discussed in a subsequent section. Active Immunization with Bacterial Products. — Many bacteria when grown in fluid media produce extracellular, soluble poisons which remain in the medium after the microorganisms have been removed by filtration or centrifugalization. Since the diseases caused by such microorganisms are, to a large extent, due to the soluble poisons excreted by them, animals can be actively immunized against this class of bac- 1 Behring, Zeit. f. Hyg., xii, 1892. 2 Chauveau, Compt. rend, de I'acad. des sci., 1881, t. xcii. 8 Ferran, Compt. rend, de I'acad. des sci., 1895, t. ci. * Th. Smith and Kilborne, U. S. Dept. of Agri». Bureau of Ani. Indust., Wash., 1893. 196 INFECTION AND IMMUNITY teria by the inoculation of gradually increasing doses of the specific poison or toxin. This method is naturally most successful against those microorganisms which possess the power of toxin formation to a highly developed degree. Most important among these are B. diphtherias and B. tetani. The first successful application of this principle of active immunization, however, was made by Salmon and Smith^ in the case of hog cholera. PASSIVE IMMUNITY In Pasteur's basic experiments, as in those of the other scientists who followed in his footsteps, the methods of immunization were based upon the development of a high resistance in the treated subject by virtue of its own physiological activities. This process we have spoken of as? " active immunization " and it is self-evident that a method of this kind can. in the treatment of disease, be employed prophylactically onl} against possible infection, or in localized acute infections, or at thp beginning of a long period of incubation before actual symptoms have appeared, as in rabies or in chronic conditions in which the infection if not of a severe or acute nature. A new and therapeutically more hopeful direction was given to the study of immunity when, in 1890 and 1892, v. Behring and his collabora- tors discovered that the sera of animals immunized against the toxins of tetanus ^ and of diphtheria ^ bacilli would protect normal animals against the harmful action of these poisons. The animals thus pro- tected obviously had taken no active part in their own defense, but were protected from the action of the poison by the substances trans- ferred to them in the sera of the actively immunized animals. Such immunity or protection, therefore, is a purely passive phenomenon so far as the treated animal is concerned, and the process is for this reason spoken of as "passive immunization." Passive immunization of this description is practically applicable chiefly against diseases caused by bacteria which produce powerful toxins, and the sera of animals actively immunized against such toxins are called antitoxic sera. In the treatment of the two diseases men- tioned above, diphtheria and tetanus, the respective antitoxic sera have » Salmon and Smith, Rep. of Com. of Agri., Wash., 1885 and 1886. 2 V. Behring and Kitasato, Deut. med. Woch., 49, 1890. »v. Behring and Wernicke, Zeit. f. Hyg., 1892. DEFENSIVE FACTORS OF THE ANIMAL ORGANISM 197 reached broad and beneficial therapeutic application, and innumerable lives have been saved by their use. Passive immunization against microorganisms not characterized by marked toxin formation was attempted, even before Behring's dis- covery, by Richet and Hericourt,^ experimenting with cocci, and by Babes,^ in the case of rabies; and the underlying thought had been the basis of Toussaint's work upon anthrax. Microorganisms, however, which exert their harmful action rather by the contents of the bacterial cells than by secreted, soluble toxins, do not, so far as is known, pro- duce antitoxins in the sera of immunized animals. The substances which they call forth in the process are directed against the invading organisms themselves in that they possess the power of destroying or of causing dissolution of the specific germs used in their production. Such antibacterial sera are extensively used in the laboratory in the passive immunization of animals against a large number of germs, and are fairly effectual w^hen used before, at the same time with, or soon after, infection. Their therapeutic use in human disease, however, has, up to the present time, been disappointing and their prophylactic and cura- tive action has been almost invariably ineffect\ial or feeble at best, ex- cept when the antibacterial sera could be brought in direct contact with the germs, in closed cavities or localized lesions. Thus, in epidemic meningitis, such sera have proved extremely useful in the hands of Flexner, when injected directly into the spinal canal. ANTIBODIES AND THE SUBSTANCES GIVING RISE TO THEM In the foregoing sections we have seen that the process of active immunization so changes the animal body that it becomes highly resistant against an infection to which it had formerly in many in- stances been delicately susceptible. In the absence of visible anatomical or histological changes accompanying the acquisition of this new power, investigators, in order to account for it, were led to examine the physio- logical properties of the body cells and fluids of immunized subjects. While it was reasonable to suppose that all the cells and tissues were affected by, or might have taken part in, a physiological change so profoundly influencing the .individual, the blood, because of its unques- tionably close relation to inflammatory reactions, and because of the f Richet et Hericourt, Compt. rend, de Tacad. des sci., 1888. « Babes et Lepp, Ann. de Tinst. Pasteur, 1889. 198 INFECTION AND IMMUNITY ease with which it could be obtained and studied, claimed the first and closest attention. The bactericidal properties of normal blood serum noted in 1886 by Nuttall/ v. Fodor/ and Fliigge, moreover, aided in pointing to this tissue as primarily the seat of the immunizing agents. It is an interesting historical fact, that, long before this time, the EngHsh physician Hunter had noted that blood did not decompose so rapidly as other animal tissues. The study of the blood serum of immunized animals as to simple changes in chemical composition or physical properties has shed little light upon the subject. Beljaeff ^ in a recent investigation found little or no alteration from the normal in the blood sera of immunized animals as to index of refraction, specific gravity, and alkaHnity. Joachim ^ and Moll agree in stating that immune blood serum is comparatively richer in globulin than normal serum. Similar observations had been made by Hiss and Atkinson ^ and others. Important and significant as these purely chemical observations are, they have helped little in explaining the nature of the processes going on in immune sera. The first actual light was thrown upon the mysterious phenomena of immunity by the investigations of Nuttall,^ v. Fodor, Buchner, and others, who not only demonstrated the power of normal blood serum to destroy bacteria, but also showed that this property of blood serum became diminished with age and was destroyed completely by heating to 56° C. The thermolabile substance of the blood serum possessing this power was called by Buchner,^ alexin. Soon after this work, Behring, in collaboration with Kitasato ^ and Wernicke,* in 1890 and 1892, made further important advances in the elucidation of the immunizing processes by showing that the blood sera of animals actively immunized against the toxins of diphtheria and tet- anus would protect normal animals against the poisons of these diseases. He believed, at the time of discovery, that such sera contained substances which had the power of destroying the specific toxins which had been 1 Nuttall, Zeit. f. Hyg., i, 1886. 2 V. Fodor, Deut. med. Woch., 1886. 3 Beljaeff, Cent. f. Bakt., xxxiii. * Joachim, Pflugers Archiv, xciii. 6 Hiss and Atkinson, Jour. Exper, Med., v, 1900. 6 Nuttall, Zeit. f. Hyg., 1886. 7 Buchner, Cent. f. Bakt., i, 1889. 8 Behring und Kitasato, Deut. med. Woch., 1890, No. 49. • Behring und Wernicke, Zeit. f. Hyg.. 1892 DEFENSIVE FACTORS OF THE ANIMAL ORGANISM 199 used in the immunization. He called these bodies antitoxins. While Behring's first conception of actual toxin destruction soon proved to be erroneous, his discovery of the presence in immune sera of bodies specifically antagonistic to toxins was soon confirmed and extended, and stands to-day as an established fact. Ehrlich,^ soon after Behring's announcement, showed that specific antitoxins could also be produced against the poisons of some of the higher plants (antiricin, antikrotin, antirobin) , and Calmette ^ produced similar antitoxins against snake poison (antivenin) . Stimulated by these researches, other observers have, since then, added extensively to the list of poisons against which antitoxins can be produced. Kempner ^ has produced antitoxin against the poison of Bacillus botulinus, and Wassermann,* against that of Bacillus pyocyaneus. Antitoxin has been produced by Calmette ^ against the poison of the scorpion, and by Sachs ^ against that of the spider. Thus a large number of poisons of animal, plant, or bacterial origin have been found capable of causing the pro- duction of specific antibodies in the sera of animals into which they are injected. The formation of antitoxins directed against soluble poisons, how- ever, did not explain the immunity acquired by animals against bacteria like Bacillus anthracis, the cholera vibrio, and others which, unlike diph- theria and tetanus, produced little or no soluble toxin. It was evident that the antitoxic property of immune blood serum was by no means the sole expression of its protective powers. Much light was shed upon this phase of the subject by the discoveries of Pfeiffer in 1894, who worked along the lines suggested by the investigations of Nuttall and Buchner. Pfeiffer ^ showed that when cholera spirilla were injected into the peritoneal cavity of cholera-immune guinea-pigs, the microorganisms rapidly swelled up, became granular, and often underwent complete solution. The same phenomenon could be observed when the bacteria were injected into a normal animal together with a sufficient quantity of cholera-immune ^ serum. * Ehrlich, Deut. med. Woch., 1891. ^ Calmette, Compt. rend, de la soc. de biol., 1894. ^Kempner, Zeit. f. Hyg., 1897. * Wassermann, Zeit. f. Hyg., xxii. ^ Calmette, Ann. de Tinst. Pasteur, 1898. •Sachs, Hofm. Beit., 1902. » Pfeiffer, Zeit. f. Hyg., xviii, 1894. * Pfeiffer und Isaeff, ibid. 200 INFECTION AND IMMUNITY This process lie observed microscopically by abstracting, from time to time, a small quantity of the peritoneal exudate and studying it in hanging-drop preparations. The reaction was specific in that the de- structive process took place to any marked extent only in the case of the bacteria employed in the immunization. Metchnikoff ,^ Bordet, and others not only confirmed Pfeiffer's obser- vation, but were able to show that the lytic process would take place in vitro, as well as in the animal body. The existence of a specific destructive process in immune serum was thus established for the vibrio of cholera and soon extended to other microorganisms. The constitu- ents of the blood serum which gave rise to this destructive phenomenon were spoken of as hacteriolysins. Following closely upon the heels of Pfeiifer's observation came the' discovery of another specific property of immune serum by Gruber and Durham.^ These workers noticed that certain bacteria, when brought into contact with the serum of an animal immunized against them, were clumped together, deprived of motility, and firmly agglutinated. They spoke of the phenomenon as agglutination and of the substances in the serum giving rise to it as agglutinins. The list of antibodies was again enlarged by Kraus,^ who in 1897 showed that precipitates were formed when filtrates of cultures of cholera, typhoid, and plague bacilli were mixed with their specific immune sera. He called the substances which bestowed this property upon the sera precipitins. The treatment of the animal body, therefore, with bacteria or their products gives rise to a variety of reactions which result in the presence of the ". antibodies " described above. Extensive investigation has shown, however, that the power of stimulating antibody production is a phe- nomenon not limited to bacteria and their products alone. Antitoxins, we have already seen, may be produced with a variety of poisons of plant and animal origin. Lysins, agglutinins, and precipitins, likewise, may be produced by the use of a large number of different substances. Chief among these, because of the great aid they have given to the theo- retical investigation of the phenomena of immunity, are the red blood cells. Bordet ^ and, independently of him, Belfanti and Carbone ^ showed ^ Metchnikoff^ Ann. de Tinst. Pasteur, 1895. 2 Gruber und Durham, Miinch. med. Woch., 1896. ^ Kraus, R., Wien. klin. Woch., 32, 1897. * Bordet, Ann. de Tinst. Pasteur, 1898. 6 Belfanti et Carhone, Giornale della R. Acad, di Torino, July, 1^98. DEFENSIVE FACTORS OF THE ANIMAL ORGANISM 201 in 1898 that the serum of animals repeatedly injected with the defibri- nated blood of another species exhibited the specific power of dissolving the red blood corpuscles of this species. This was the first demonstration of " hemolysis " — a phenomenon which, because of the ease with which it can be observed in vitro, has much facilitated investigation. c^n Oi The knowledge that specific "endotoxins'' or cell-destroying anti- irt>-^>^ bodies could be produced by injection of red blood cells naturally sug- A'UpuJ; gested the possibility of analogous reactions for other tissue cells. It"^^. was not long, therefore, before Metchnikoff * and, independently of him, Landsteiner ^ succeeded, by repeated injections of spermatozoa, in producing a serum which would seriously injure these specialized cells. Von Dungem ^ obtained similar results with the ciliated epithelium of the trachea. Since then a host of cytotoxins have been produced with the cells of various organs and tissues. Thus, Neisser and Wechsberg * produced leucotoxin (leucocytes); Delezenne,^ neurotoxin and hepa- totoxin; Surmont,^ pancreas cytotoxin; and Bogart and Bernard,^ su- prarenal cytotoxin. One of the most interesting of the cytotoxins, moreover, is nephro- toxin — ^produced by the treatment of animals with injections of emul- sions of kidney tissue. In all cases it was supposed by those first working with these bodies, that the injection of the sera of animals previously treated with any particular tissue substance would produce specific injury upon the or- gans homologous to the ones used in immunization. It need hardly be pointed out how very important such phenomena would be in throwing light upon the degenerative pathological lesions occurring in disease. As a matter of fact, however, sera so produced have been shown to be specific for certain organs in a limited sense only. The question of specific cytotoxins has been of especial importance in the case of nephritis, where Ascoli and Figari ^ and others have suggested an autonephrotoxin as the basis of the pathology of this disease. In the hands of Pearce and others, however, the strict specificity of ^ Metchnikoff, Ann. de I'inst. Pasteur, 1898. 2 Landsteiner, Cent. f. Bakt., i, 25, 1899. ^v. Dungern, Munch, med. Woch., 1899. ^ Neisser und Wechsberg, Zeit. f. Hyg., xxxvi, 1901. ^ Delezenne, Ann. de Tinst. Past. 1900; Compt, rend, de I'acad. des sci. 1900. • Surmont, Compt. rend, de la soc. de biol., 1901. ^ Bogart et Bernard, ibid., 1891. * Ascoli and Figari, Berl. klin. Woch.^ 1902. 202 INFECTION AND IMMUNITY nephrotoxin could not be upheld and the subject is still in the ex- perimental stage. Recent experiments by Pearce ^ suggest that at least a part of the local injury to organs exerted by such "cytotoxic" sera may not be due to a specific action upon the organ cells so much as upon the hemaggiutinating action of the sera causing emboHsm and necrosis. It is a fact also that most cytotoxic sera are usually hemolytic as well. It is not easy to decide, therefore, how much of the action upon the organs is due to their true cytotoxic properties and how much is attributable to the concomitant action upon blood cells. The extrav- agant hopes at first based upon cytotoxin investigation, especially in regard to the problem of malignant tumors, have been disappointed, and much is still obscure in regard to the cytotoxins which> calls for further research. The many points of similarity existing between bacterial toxins and digestive ferments, by animal inoculation, suggested to several observ- ers the possibility of producing antibodies against the latter. As a result, a number of antiferments have been obtained, chief among which are antilab (Morgenroth ^), antipepsin (Sachs ^) , antisteapsin (Schiitze *), and antilactase (Schiitze). The stimulation of antibody formation in the sera of animals is a consequence, therefore, of the injection of a large variety of substances — some of them poisonous, some of them entirely innocuous. The sub- stances possessing this power have been conveniently named antigens or antibody-producers by German v/riters. The term antigen — though ety- mologically wrong, nevertheless is convenient and has crept into general usage. It signifies simply a substance which can stimulate the pro- duction or formation of an antibody. Such substances, so far as is known, belong to the group of proteins and are derivatives of animal or plant tissues. Being proteins, all antigens are colloids. Recently, how- ever, some crystalloidal substances have been described as possessing antigenic properties. 1 Pearce, Jour. Exper. Med., viii, 1906. 2 Morgenroth, Cent. f. Bakt., 1899. ^ Sachs, Fort. d. Med., 1902. * Schiitze, Deut. med. Woch., 1904; Zeit. f. Hyg., 1905. CHAPTER XIII TOXINS AND ANTITOXINS The Toxin-Antitoxin Reaction. — Apart from the therapeutic possi- biUties disclosed by the discovery of antitoxins, new light of inestimable value was thrown by these observations upon the biological processes involved in immunization. The most vital problem, of course, which immediately thrust itself upon all workers in this field was the question as to the nature of the reaction in which toxin was rendered innocuous by antitoxin. The simplest conception of this process would be an actual destruction of the toxin by its specific antitoxin, and it is not unnatural, therefore, that this was the view which, for a short time, found favor with some observers. Roux, and more particularly Buchner,^ however, under the sway of cellular pathology, advanced the opinion that the antitoxins in some way influenced the tissue cells, rendering them more resistant against the toxins. Antitoxin, according to this theor}^, did not act directly upon toxin, but affected it indirectly through the mediation of tissue cells. Ehrlich,^ on the other hand, conceived that the reac- tion of toxin and antitoxin was a direct union, analogous to the chem- ical neutralization of an acid by a base — an opinion in which Behring soon joined him. The conception of toxin destruction received unanswerable refuta- tion by the experiments of Calmette.^ This observer, working with snake poison, found that the poison itself (unlike most other toxins) possessed the property of resisting heat even to 100° C, while its specific anti- toxin, like other antitoxins, was delicately thermolabile. He noted, furthermore, that non-toxic mixtures of the two substances, when sub- jected to heat, regained their toxic properties. The natural inference from these observations could only be that the toxin in the original mix- ture had not been destroyed, but had been merely inactivated by the ^ Buchner, " Schutzimpfung/'etc, in Penzoldt u. Stinzing, "Handbuch d. spez. Therap. d. Infektkrank./' 1894. ^Ehrlich, Deut. med. Woch., 1891. ^Calmette, Ann. de Finst. Past., 1895. 203 204 INFECTION AND IMMUNITY presence of the antitoxin, and again set free after destruction of the antitoxin by heat. A similar observation, made soon after by Wasser- mann ^ in the case of pyocyaneus toxin and antitoxin, fully supported the results of Calmette. An ingenious proof of the direct action of antitoxin upon toxin was obtained by Martin and Cherry.^ It was found by them that veiy dense filters, the pores of which had been filled with gelatin, permitted toxin to pass through under high pressure, while the presumably larger antitoxin molecule was held back. Through such filters they forced toxin-antitoxin mixtures, under a pressure of fifty atmospheres, at vary- ing intervals after -mixing. They found that, if filtered inmiediately, all the toxin in the mixtures came through, but that, as the interval elapsing between mixing and filtration was prolonged, less and less toxin appeared in the filtrate, until, finally, two hours after mixing, no toxin whatever passed through the filter. Besides demonstrating the direct action of antitoxin upon toxin, this work of Martin and Cherry showed that the element of time entered into the toxin-antitoxin reaction, just as it enters into reactions of known chemical nature. The absolute non- participation of the living tissue cells in these reactions was demonstrated by Ehrlich himself. Kobert and Stillmarck ^ had shown that ricin pos- sessed the power of causing the red blood cells of defibrinated blood to agglutinate in solid clumps, a reaction which could easily be observed in vitro. EhrHch,'* who had obtained antiricin in 1891 by injecting rabbits with increasing doses of ricin, found that this antibody pos- sessed the power of preventing the hemagglutinating action of ricin in the test tube. By a series of quantitatively graded mixtures of ricin and. antiricin, with red blood cells as the indicator for the reaction, he succeeded in proving not only that the toxin-antitoxin neutralization was in no way dependent upon the living animal body, but that definite quantitative relations existed between the two substances entirely analogous to those which, according to the law of multiple proportions, govern reactions between different substances of known chemical nature. Similar quantitative results were subsequently obtained by Stephens and Myers ^ for cobra poison and its antitoxin, by Kossel * 1 Wassermann, Zeit. f. Hyg., xxii, 1896. ^Martin and Cherry, Proc. Royal Soc, London, Ixiii, 1898, * Kobert und Stillmarck, Arb. d. phar. Inst. Dorpat, * Ehrlich, Fort. d. Med., 1897. » Stephens and Myers, Jour, of Path, and Baet., 1898. * Kossel, Berl. klin. Woch., 1898. TOXINS AND ANTITOXINS 205 for the toxic eel blood serum, and by Ehrlich ^ for the hemolytic tetanus poison known as tetanolysin. The introduction of the test-tube experiment into the investigation of these reactions permitted of much more exact observations, and by this means, as well as by careful, quantitatively graded, animal experi- ments, the further facts were ascertained that toxin and antitoxin com- bined more speedily in concentrated than in dilute solutions, and that warmth hastened, while cold retarded, the reaction — observations^ which in every way seem to bear out Ehrlich's conception of the chemi- cal nature of the process. Ehrlich's Analysis of Diphtheria Toxin. — Shortly §tfter the discovery and therapeutic application of diphtheria antitoxin, it became apparent that no two sera, though similarly produced, could have exactly the same protective value. It was necessary, therefore, to establish some measure or standard by which the approximate strength of a given anti- toxin could be estimated. Von Behring ^ attempted to do this for both tetanus and diphtheria antitoxins by determining the quantity of immune sera which, in each case, was needed to protect a guinea-pig of known weight against a definite dose of a standard poison. He ascer- tained the quantity of standard toxin-bouillon which would suffice to kill a guinea-pig of 250 grams, and called this quantity the "toxin unit.'' This unit was later more exactlj' limited by Ehrlich, who, considering the element of time, stated it as the quantity sufficient to kill a guinea- pig of the given weight in from four to five days. Appropriating the terminology of chemical titration, v. Behring spoke of a toxin-bouillon which contained one hundred such toxin units in a cubic centimeter, as a "normal toxin solution" (" DTN^ M^^""), and designated as "normal antitoxin-" a serum capable of neutraliz- ing, cubic centimeter for cubic centimeter, the normal poison.* A cubic centimeter of such an antitoxic serum was sufficient, therefore, to neu- tralize one hundred toxin units, and was spoken of as an "antitoxin unit." In the experiments of v. Behring, toxin and antitoxin had been separately injected. Ehrlich ^ improved upon this method by mixing toxin and antitoxin before injection, thereby obviating errors arising » Ehrlich, Berl. klin. Woch., 1898. « Knorr, Fort. d. Med., 1897. » V. Behring, Deut. med. Woch., 1893. * DTN 1 M250 signifies: D, Diphtheria; TN^ Normal Toxin solution; M«», Meer- schweinchen or guinea-pig weighing 250 grams. * Ehrlich, Koesel und Wassermann, Deut. med. Woch., 1894. 206 INFECTION AND IMMUNITY from differences which may have existed in the depth of injection or rapidity of absorption. In order, however, that any such method of standardization of an- titoxin may be practically applicable, it is necessary to produce either a stable toxin or an unchangeable antitoxin. This EhrHch achieved for antitoxin by drying antitoxic serum in vacuo and preserving it in the dark, at a low temperature and in the presence of anhydrous phosphoric acid. By the use of such a stable antitoxin, various toxins may be measured and other antitoxic sera estimated against these. Given thus a constant antitoxin, the standardization of toxins would be a comparatively simple matter Avere the poison obtainable in a per- fectly pure state. Unfortunately for thp ease of measurement, how- ever, this is not the case. The problem is rendered difficult by a number of compHcating factors, many of which have been brought to light by Ehrlich ^ in his laborious researches into the quantitative relationship between the two reacting bodies. As previously stated, it had been noted by Ehrlich and others that toxin solutions would deteriorate with time; that is, a toxin-bouillon ^Toxop"hore\ Bordet, Ann. de I'inst. Pasteur, t. 14, 1900. 244 INFECTION AND IMMUNITY by mixing inactivated hemolytic serimi with its respective red blood cells, then adding the antiserum and later complement. After cen- trifugalization and separation of the corpuscles, these may be dissolved by the addition of fresh complement. This proves conclusively that there was no obstacle in the original mixture to the absorption of the immime body by the red blood cells, and that the antihemolytic prop- erties of the serum must be attributed to an anticomplement. This was the method of experimentation employed by Ehrlich and Morgen- roth.^ Antiamboceptors have been produced by the same authors as well as by Bordet ^ and Miiller,^ against hemolytic amboceptors. Complementoids. — Ehrlich and Morgenroth and Miiller have suc- ceeded in producing anticomplements by the treatment of animals with normal heated serum. They explain this by assuming that the heating has not entirely destroyed the complement in the normal serimi, but that this, analogous to toxin, possesses two groups, a haptophore and a zymophore group. Heating destroys the zymophore without affecting the haptophore group. The resulting body, which corresponds to toxoid, they call "complementoid." Further evidence for the existence of such complementoids has been claimed by Ehrlich and Sachs ^ in working with dog serum. Unheated dog seriun hemolyzes guinea-pig corpuscles. Heated to 52 degrees C. for thirty minutes, it no longer hemolyzes these corpuscles owing to complement destruction. Such heated dog serum can be reactivated by fresh guinea-pig serum (complement). If, however, the corpuscles are left in contact with the heated dog blood for two hours, reactivatio^i by the guinea-pig serum no longer occurs — that is, the addition of guinea- pig seriun no longer causes hemolysis. They conclude from this that the hemolytic amboceptor of the dog serum has been attached by its complementophile group to complementoids produced in the heating — leaving no point of attachment for the complement added later. These experiments have failed of confirmation by Gay ^ — who with Bordet de- nies the existence of complementoids. Muir, on the other hand, claims to have demonstrated the existence of complementoids by experiments too complicated to be detailed in this place. The question of complementoids must be left undecided until further work has been done. * Ehrlich und Morgenroth, loc. cit. 2 Bordet, loc. cit. 3 P. Th. Muller, Cent. f. Bakt., 1901. * Ehrlich and Sachs, "Ehrlich Collected Studies on Inamunity," trans. byBoldnau. * Gay, Cent. f. Bakt., I, xxxix, 1905. LYSINS, AGGLUTININS, PRECIPITINS, ETC. 245 Other Pacts Concerning Complement. — Muir and Browning have shown that, on the filtration of serum, amboceptor or immune body will pass through the filter, whereas alexin or complement is held back. This retention of complement by filters occurs only when new filters are used, and it is our opinion that this is probably due to absorption or complement by the finely divided substances which make up the filter and not due to retention because of the large size of the comple- ment molecule. Complement can be inactivated by shaking as well as by heat when diluted 1 : 10 and shaken for about 20 minutes in salt solution. Ac- cording to Gramenitski it is spontaneously partially reactivated on standing. Complement is dependent upon the total volume of the mixture in which it acts, i.e., upon concentration, the same actual quantity of complement acting more strongly in higher than in lower concentrations, this not being true of amboceptor or sensitizer which acts in direct proportion to its actual quantity independent of the concentration. Complement is inhibited by hypertonic salt solution and can be preserved in 15-25 per cent salt concentration for weeks in the icebox, resuming its activity when diluted to isotonicity with distilled water. Removal of salt by dialysis or other means of globulin precipitation divides the complement into two fractions, the globulin fraction and the albumin fraction, neither of which will act alone, but which to- gether possess the properties of undivided complement. The globulin fraction attaches directly to the sensitized cells and is therefore spoken of by German investigators as "mid-piece." The albumin fraction acts upon the sensitized gells only after attachment of the globulin fraction and is therefore spoken of as *' end-piece." The Fixation of Complement by Precipitates. — It has been found by Gengou ^ and confirmed by Moreschi, Gay, ^ and others, that when the serum of an animal immunized with the serum of another species or with a foreign albumin is mixed with a solution of the substance used in the immunization, the precipitate formed will remove complement from the mixture. In other words, precipitates formed by the reaction of precipitin with its antigen will fix complement. This is of great im- portance in complement-fixation tests; for because of insufficient wash- ing, the blood cells used in producing the hemolytic amboceptor, may, from the presence of serum, give rise to a precipitin as well as a hemo- 1 Gengou, Ann. Past., 1902. « Gay, Cent, f . Bakt., I, xxix, 1905. 246 INFECTION AND IMMUNITY ™ lysin. In the test done subsequently, a precipitin reaction may take place and by thus removing complement may give a false result. The absorption of complement by such precipitates takes place when the two reacting factors, the precipitin and its antigen, are in dilution — so high a visible precipitate can not be observed. This fact, together with others too complicated to be discussed in this place, have led us to the belief that the so-called precipitins are true sensitizers, exerting toward unformed proteins the same function that the so-called sensitizer or amboceptor exerts toward cellular formed antigens. (See p. 241.) Quantitative Relationship Between Amboceptor and Complement. — Morgenroth and Sachs ^ have succeeded in showing that within certain limits an inverse relationship exists between these two bodies. If for a given quantity of red blood cells a certain quantity of amboceptor and complement suffices to produce complete hemolysis, reduction of either the complement or the amboceptor necessitates an increase of the other factor. As amboceptor is increased, in other words, complement may be reduced and vice versa. This result is of great importance in arguing against the original conception of Ehrlich in supposing these substances to act together unit for unit. Deviation of the Complement (Complement-Ablenkung). — It was noticed by Neisser and Wechsberg^ that in mixing together bacteria, inactivated bactericidal immune serum (immune Hf^ T|r yf body), and complement in the test tube, a great excess of immune body hindered rather than helped bactericidal action. As the amount of immune body in the mixture was carried beyond the J^ H It experimental optimum, bactericidal action became less and less pronounced, and was finally com- pletely suspended. They explain this by assuming that free immune body, uncombined with comple- Dlj n ment, has a greater affinity for the bacterial receptor 11 11 than the immune body combined with comple- ment. The complement is consequently diverted and prevented from activating the amboceptor at- tached to the bacterial cell. Graphically, the conditions may be illustrated as follows : The above theory of Neisser and Wechsberg is here stated simply ^Morgenroth und Sachs, "Gesammel. Arb. fiir Immunitatsforschung." Berlin, Hirschwald, 1904. 2 Neisser und Wechsberg, Miinch. med. Woch., xviii, 1901, /!/ LYSINS, AGGLUTININS, PRECIPITINS, ETC. 247 because of the wide discussion it has aroused. In the light of our present knowledge concerning the relations between antigen, ambo- ceptor, and complement, their conception is obviously erroneous. Fixation of the Complement.— Bordet and Gengou^ in 1901, devised an ingenious method of experimentation by which even very small quantities of any given immune body (amboceptor) can be demon- strated in serum. The term "fixation of complement," by which their method of investigation is now generally known, explains itself, as the steps of experimentation are followed. They prepared the following mixtures : (a) (b) Bacteriolytic amboceptor Normal serum, heated {Plague immune serum, heated) + . + Plague emulsion Plague emulsion + + Complement Complement {Fresh normal serum) {Fresh normal serum) To both of these after five hours was* added Hemolytic amboceptor {Heated hemolytic serum) + Red blood cells Results: (a) showed no hemolysis. (b) showed hemolysis +. " The conclusion to be drawn from this was that in (a) the presence of immune body had led to absorption of all the complement. In (b), there being no bacteriolytic immune body to sensitize the bacteria and enable them to absorb complement, the latter substance was left free to activate the subsequently added hemolytic amboceptors. The Bordet-Gengou phenomenon has been extensively used by Wassermann and Bruck,^ Neisser and Sachs,^ and others to demonstrate the presence of immune bodies in various sera. (See p. 262.) It should be noted that this method, if valid, must presuppose the identity of the hemolytic and bactericidal complement in the activating serum. Complement fixation will be more extensively discussed in the sec- tion dealing with the Wassermann reaction. 1 Bordet et Gengou, Ann. de I'inst. Pasteur, 1901. 2 Wassermann und Bruck, Med. Klin., 1905. » Neisser und Sachs, Berl. khn. Woch., xliv, 1905, and i, 1906. 248 INFECTION AND IMMUNITY The Specificity of Hemolysins. — In the sections preceding we have seen that the blood cells of one animal, injected into an animal of an- other species, give rise to a hemolytic substance in the blood serum of the second animal, which is strictly specific for the variety of cells in- jected. Such hemolysins, when produced in one animal against blood cells of another species, are spoken of as heterolysins. In studying the nature of hemolysis,* Ehrlich and Morgenroth ^ now discovered that hemolysins could also be produced if an animal were injected with red blood cells of a member of its own species. Such hemolytic substances 1^ WK.^m;^^omp\enxe-ri\ B S^pWHHc presenf/ Togefher at 4. iTnntvine or ? °^ i 37.SO C. . aniibody n.o4- A j» i ' for one hour ' -Haetnolyitc Amt o ce pto r (s.^g — R^d bipod cell 3. ^-•-Anfi^ert If (2) present, no haemolysis. If (2) not present, haemolysis. Fig. 64. — Schematic Representation of Complement Fixation in the Bordet-Gengou Reaction. they called isolysins. In their experiments they injected goats with the washed red blood corpuscles of other goats and found that the serum of the recipient developed the power of causing hemolysis of the red blood cells of the particular goat whose blood had been used for injection. It did not, however, possess the power of producing hemolysis in the blood of all goats, nor did it produce hemolysis with the red corpuscles of its own blood. It is thus shown that the specificity of the hemolysins extends even within the limits of species, and is, to a certain extent, an individual property. The production of autolysins, that is, of substances in the blood serum which will produce hemolysis of the individual's own corpuscles, has, so far, been unsuccessful. 1 Ehrlich und Morgenroth, Berliner klin. Woch., xxi, 1900. « LYSINS, AGGLUTININS, PRECIPITINS, ETC. 249 Ehrlich and Morgenroth, in the course of these experiments, further- more succeeded in showing that the injection of isolysins into animals produced antiisolysins, and that these again were strictly specific. The almost universal failure of autolysin production has found no satisfactory explanation. It is supposed by Ehrlich and Morgenroth that the failure of autolysin production may be due to a lack of suitable receptors in the animal for its own cells. The cHnical significance of the presence of isolysins and possibly of autolysins in human beings, is too evident to require much discussion. A practical and extremely interesting result which these investigations have yielded is that of Donath and Landsteiner,^ who discovered an autolysin in the blood serum of patients suffering from paroxysmal hemoglobinuria. In these cases the sensitizing substance or ambo- ceptor appeared to be absorbed by the red blood cells only at low tem- peratures— probably in the capillaries during exposure to the cold, and hemolysis subsequently resulted in the blood stream by the action of complement. These observations have been confirmed by other writ- ers, but the phenomenon is surely not present in all cases of paroxysmal hemoglobinuria. The writers have had occasion to examine carefully several clinically typical cases with negative results. * Donath und LavdsteiTwr, Miinch. med. Woch., xxxvi, 1904. n CHAPTER XVI THE TECHNIQUE OF SERUM REACTIONS Obtaining Serum from Animals and Man. — To obtain blood serum from man, the blood may be taken from the finger or the ear, either into a sterile centrifuge tube or into a Wright capsule. When taken into a centrifuge tube, the blood is allowed to clot and the serum sep- arated by centrifugation. Larger quantities of blood may be taken with a syringe from the median basilic vein and either slanted in sterile test tubes in the ice chest or put into centrifuge tubes and centrifugalized. In bleeding small laboratory animals, a number of methods may be employed, depending upon the quantity of serum required. The animals most frequently used for laboratory purposes are rab- bits. To obtain small quantities of serum from rabbits, the animals may be bled from the marginal vein of the ear. The animal is strapped upon a tray and underneath it is placed a rubber bag filled with warm water. This is advised by Wadsworth to facilitate the flow of blood. The tray is then placed upon an easel so that the animal's head hangs downward. The skin over the ear vein is shaved and sterilized, and a Hagedorn needle plunged into the vein. The blood is caught in test tubes or centrifuge tubes. When larger quantities of blood are desired it may be taken from the carotid artery. In rabbits, the carotid may be found lying just lateral to the trachea and deeply placed, and must be carefully separated from the pneumogastric nerve by blunt dissection. The distal end of the artery is then tied off and the proximal end temporarily closed with a small clamp. The artery is then raised out of the wound on a knife or forceps handle and, with sharp-pointed scissors, a small in- cision is made into but not through the vessel. A small glass cannula is now introduced and tied into place by a thread. To this cannula a small rubber tube fitted with a pinch-cock should have been attached, the whole being sterilized. Recently we have dispensed with the can- nula, simply holding the vessel up with a pointed forceps. A larger yield of serum will be obtained if, after coagulation, the clot is sep- arated from the glass with a sterile platinum wire. 250 THE TECHNIQUE OF SERUM REACTIONS 251 In obtaining blood from larger animals, horses, sheep, etc., a cannula rmay be introduced into the jugular or internal saphenous veins. The skin is shaved and sterilized and a rubber tourniquet placed about the neck or thigh, as the case may be, in order to cause the vein to stand out. A small incision may be made through the skin over the vein, but is not necessary. A cannula, with rubber tubing attached, is then plunged into the vein and the blood caught in sterile high cylindrical jars, allowed to clot, and placed in the refrigerator. The serum is taken off after twenty-four to forty-eight hours with sterile pipettes. Agglutination Tests.— For the determination of the agglutinating power of serum it is necessary to make suitable dilutions of the serum, and to prepare an even emulsion of the microorganisms to be tested. The test may be made microscopically or macroscopically. The micro- scopic test is the one in general use in the diagnosis of typhoid fever, and is occasionally applied to some other diseases. In its application to typhoid fever it is usually spoken of as the Gruber-Widal reaction. Twelve- to eighteen-hour broth cultures of the typhoid bacillus, grown at incubator temperature, may be used. It is preferable, how- ever, to use an emulsion of a twelve to twenty-four hour old agar culture in physiological salt solution (0.85 per cent). The salt-solutron emulsion is made by adding about 10 c.c. of normal salt solution to the fresh agar slant culture, carefully detaching the culture from the surface of the agar with a flexible platimma wire, and pipetting off the emulsion thus made. With some microorganisms it is sufficient simply to allow the larger clumps to settle and to pipette off the supernatant turbid emulsion. With other microorganisms, the tendency to form clumps makes it necessary to resort to further methods of securing an even distribution of the bacteria. This may be done either by sucking the emulsion in and out through a narrow pipette held perpendicularly against the bottom of a watch glass, as in Wright's technique for the opsonic test (see section on Opsonins, p. 285), or by carefully rubbing the climips against the watch glass with a stiff platimma wire. In the case of the tubercle ba- cillus not even this suffices, but it becomes necessary to grind the moist bacillary masses in a mortar before emulsifying. With the tubercle bacil- lus, too, it is preferable to use salt solution at 1.5 per cent concentration. In preparing cultures of streptococcus and pneumococcus for ag- glutination tests, it has been found convenient by Hiss to grow for about four days in flasks of a 1% glucose, 2% pepton meat-infusion broth, to which has been added 1% of calcium carbonate (p. 126). The calcium neutralizes the inhibiting acid formed in the broth 252 INFECTION AND IMMUNITY by the microorganisms and permits the development of a mass culture. The flasks should be shaken at least once a day. The broth may be pipetted off and clumps removed by centrifugation. Without this tech- nique it is sometimes difficult to get sufficient growths of these bacteria for any quantity of emulsion unless large surfaces of agar are employed in special receptacles or by making many slant cultmes. The serum dilutions are obtained by first making a one to ten dilu- tion of serum with normal salt solution. The serum used for this pur- pose is of red blood corpuscles by centrifugation. From the 1 to 10 dilution any number of higher dilutions may be made, by mixing given parts of the 1 to 10 dilution with normal salt solution; thus one part of a 1 to 10 dilution plus an equal quantity of salt solution gives a dilu- tion of 1 to 20. One part of one to ten dilution plus two parts of normal salt solution gives one to thirty, etc. It must not be forgotten that, when equal parts of the serum and bacillary emulsion have been mixed, each one of these dilutions is doubled. In making the microscopic agglutination test, equal quantities of serum dilution and bacterial emulsion are mixed upon a cover-slip. The mixture may be made either by measuring out a drop of each sub- stance with a standard platimun loop, depositing them close together on the cover-slip, and mixing; or equal quantities may be sucked up, each to a given mark, in a capillary pipette, mixed by suction in and out, and deposited upon the cover-slip. The cover-slip is inverted over a hollow glass slide, the rim of which has been greased with vaseline. The drop is then observed through a (Leitz) No. 7 lens, ocular No. 3. Macroscopic agglutination, preferable for exact laboratory research, is made in narrow test tubes measuring about 0.5 cm. in diameter and about 5 cm. in length (Fig. 60). Equal quantities, usually 1 c.c. each, of serum dilution and emulsion are mixed. A series of tubes is prepared, in each subsequent one of which the dilution is higher. These mixtures may be placed in the in- cubator for a few hours and then kept at room temperature. After re- moval from the incubator agglutination is in some instances hastened by transference to the ice chest. When agglutination takes place in these tubes, clumps of bacteria may be seen to form, which settle to the bottom of the tube, very much like snow-flakes. The surface of the sediment is heaped up and irregular. The supernatant fluid becomes entirely clear. When the reaction does not occur the sediment is an even, granular one with a flat surface, and the emulsion remains turbid. Instead of using test tubes as described above, Wright has sug- THE TECHNIQUE OF SERUM REACTlONg 255 gested the us6 of throttle pipettes of comparatively large diameter into each of which at least three or four different dilutions can be sucked with a nipple, a small air bubble being left between the mixtures. By- sealing the distal end of these pipettes in a flame the various dilutions are kept at a distance from each other, and the pipettes may be set on end in a tumbler and observed just as are the test tubes (Fig. 68, p. 285). The special methods of carrying out agglutination tests with pneumo- cocci have been described on p. 363. Precipitin Tests. — In an earlier section on precipitins we have seen that precipitates are formed when clear filtrates of bacterial extracts or of both cultures are mixed with their specific immune sera. Such precipitin reactions are not limited to the realm of bacteria, but have a broad biological significance, in that specific precipitating sera may be produced with proteids of varied source. For carrying out a precipitin test, the following reagents are required: 1. A specific precipitating antiserimi (antibacterial or antiproteid) ; 2. A bacterial filtrate or proteid solution. Production of Precipitating Antisera.' — Antibacterial precip- itins may be produced in animals by a variety of methods. Animals, preferably rabbits, are injected with cultures of the bacteria in gradually increasing quantities. Five or six injections are given at intervals of from five to six days, the dosage and mode of administra- tion being adapted in each case to the pathogenicity of the micro- organisms in question. Myers ^ claims that specific precipitin for pepton in the culture media may be formed which may lead to error. This could not be confirmed by Norris.^ The immunized animals should be bled about 7 to 12 days after the last injection. Precipitating antisera against protein solutions are prepared by similar methods. The sera or protein solutions used should be sterile. This may be accomplished by filtration through small porcelain filters. Injections into animals may be made subcutaneously, intraperitoneally, or intravenously. The subcutaneous route has no advantages unless the substances to be used are contaminated. Nuttall advises the use of rabbits. The animals are weighed from time to time, and if considerable loss of weight ensues, the intervals should be increased. Doses from 2 to 5 c.c. should be given. In giving the later injections the danger of anaphylaxis must be remem- 1 R. Kraus, Wien. klin. Woch., 1897; Norris, Jour. Inf. Dis., 1 and 3, 1904. ^ Myers, Lancet, ii, 1900. ^ Norris, loc. cit. ^54 INFECTION AND IMMUNITY bered. A single injection of a large quantity has occasionally yielded a precipitating serum of considerable strength,^ but this method is not usually successful. Injections are made at intervals of from five to seven days. Seven to twelve days after the last injection the animals may be bled from the ear, and a preliminary test made to ascertain the precipitating value of the serum. If this is insufficient, more injections may be made. Bleeding should be done 7 to 12 days after the last injection. Such sera may be preserved in the dark and at a low tem- perature. If a preservative is added, Nuttall prefers chloroform to the phenols, because of occasional turbidity produced by these. Precipitating antisera for tests should be clear. If turbid, the sera should be filtered through small Berkefeld or porcelain candles. Preparation of Bacterial Filtrates and Protein Solutions FOR Precipitin Tests. — Bacteria may be grown in nutrient broth having an initial reaction of neutrality or five-tenths per cent acidity to phenol ophthalein. The cultures are incubated for times varying from a week to several months, and are then filtered through porcelain or Berkefeld candles until perfectly clear. Bacterial extracts may also be made by emulsifying agar cultures in salt solution, placing at 37.5° C. in the incubator for a week or longer, and filtering. More rapid ex- traction of bacteria may be accomplished by repeated, rapid freezing and thawing of salt-solution emulsions, by shaking in the shaking ma- chine or by centrifugalizing, rubbing up the sediment with dry salt, and the addition of distilled water to isotonicity. Protein solutions to be tested should be made in salt solution. When dealing with blood stains, as in doing the test for forensic purposes, the stains should be dissolved in salt solution, an approximate dilution of one in five hundred being aimed at. This solution if turbid should be filtered through a small porcelain filter. It should be clear and color- less, show a faint cloud on boiling with dilute acetic acid, and show distinct froth when shaken. When the reaction is to be done for determining the nature of meat (detection of horse-meat substitution for beef, etc.), about 20 to 40 grams of the suspected meat are macerated in a flask, and covered with 100 c.c. of salt solution. This mixture is allowed to infuse at room temperature for three to four hours, and is then placed in the refrigerator for twelve hours or more. At the end of this time 2 c.c. are shaken into a test tube. If profuse frothing ^ appears, the extract is ready for 1 Michaelis, Deut. med. Woch., 1902. 2 P. Th. Muller, "Technik. d. serodiagnos. Methoden." THE TECHNIQyE OF SERUM REACTIONS 255 use. It is then filtered clear, either through paper, or through a Buchner or Nutsche filter. Berkefeld filters may also be used. The solution is then diluted until the addition of concentrated HNO3 produces only a slight even turbidity. Before use the reaction of the meat extract should be tested, and if necessary adjusted to neutrality or slight acidity or alkalinity. In the actual test with bacterial filtrate, the procedure is as follows: In a series of narrow test tubes, the following mixtures are made: Tube 1. Antibacterial serum .5 c.c. + bacterial filtrate 1. c.c. " 2. Normal serum .5 c.c. + bacterial filtrate 1. c.c. " 3. Antibacterial serum .5 c.c. + salt solution 1. c.c. " 4. Salt solution .5 c.c. + bacterial filtrate 1. c.c. Place the tubes in the incubator at 37.5° C. Tube 1 only should .show a haziness which develops into distinct cloudiness or a flocculent precipitate within one hour. Tubes 2, 3, and 4 should remain clear. In testing an unknown protein with serum of an animal immunized with the protein sought for, the technique of the test is as follows: 1. 0.1 c.c. immune serum + 2 c.c. unknown protein solution. 2. 0.1 c.c. immune serum + 2 c.c. known protein solution of variety suspected (similarly diluted). 3. 0.1 c.c. immune serum + 2 c.c. protein solution of different nature (similarly diluted) . 4. 0.1 c.c. immune serum + 2 c.c. salt solution. 5. 2 c.c. unknown protein solution. If the test is positive a precipitate appears in tubes 1 and 2, but not in any of the others. The precipitate should appear within 15 to 20 minutes. Bactericidal and Bacteriolytic Tests. — The bactericidal and bac- teriolytic powers of serum may be tested either in the animal body or in the test tube. The in vivo test is known as Pfeiffer's phenomenon. This depends upon the fact that bacteria, when injected into the peri- toneal cavity of a guinea-pig, together with a homologous immune serum, undergo dissolution. As practiced, the test finds a double application. It may be done to determine the bacteriolytic power of a given serum against a known microorganism, or for the identification of a particular microorganism by means of its susceptibility to lysis in a known immune serum. 1. Determination of the bacteriolytic power of serum against a known microorganism in vivo: ^ ^ P. Th. Muller, "Technik d. serodiagnos. Methoden," Jena, 1909. 256 INFECTION AND IMMUNITY A number of dilutions of the serum are made with sterile neutral bouillon or salt solution, ranging from 1 in 20 to 1 in 500, or higher. It is convenient to make a first solution of 1 in 20. One c.c. of this mixed with 4 C.C. of broth will give 1 in 100. One c.c. of the 1 in 100 dilution with 1 c.c. of broth, 2 c.c. of broth and 4 c.c. of broth will give 1 in 200, 1 in 300, and 1 in 500, respectively. Into one cubic centimeter of each of these dilutions there is placed one platinum loopful of a twenty-four- hour agar culture of the microorganism against which the serum is to be tested. Into another test tube is placed 4 c.c. of broth, without serum, and with one loopful of the microorganisms. The mixtures are thoroughly emulsified in each case by rubbing the bacteria against the sides of the tube with the platinum loop. Intraperitoneal injections into guinea-pigs are then made of 1 c.c. of each of the serum-dilution-bacterial-emulsions. A control guinea-pig. (better two or three) receives Ic. c. of the broth emulsion — one-fourth as many bacteria, therefore, as the animals receiving the serum dilutions. Before making the injections, areas on the lateral abdominal walls of the guinea-pigs are shaved, and small incisions made through the 1 Fig. 65. — Capillary Pipette for Removal of Exudate in doing the Pfeiffer Test. skin, down to the muscular layers. The needle of the syringe is then introduced perpendicular to the skin until it has penetrated the peri- toneum, and then carefully slanted to avoid puncturing the gut. The animals need not be strapped down during this procedure and after- ward may be allowed to run about. After one-half hour, and again after one hour has elapsed, a drop of peritoneal exudate is removed from each guinea-pig and examined in the hanging drop for granulation and swelling of the bacteria. The method of obtaining the peritoneal exudate is as follows: Small glass tubing is drawn out into capillary pipettes, the ends of the capillaries being again drawn to fine points in a small yellow flame. A number of such pipettes should be prepared before the test is begun. The guinea- pig is then held down upon a table, either by an assistant or by the left hand of the operator, and the point of the pipette pushed through the ent in the abdominal wall into the peritoneum by a. sharp, quick thrust- THE TECHNIQUE OF SERUM REACTIONS 257 ing motion. A column of peritoneal fluid will run into the glass tubing by capillary attraction; this can then be blown out upon a cover-slip for hanging-drop examination or may be blown upon a sHde, smeared, and examined after staining. The reaction is regarded as positive if within thirty minutes to an hour the peritoneal exudates of the animals receiving immune sera contain only swollen or disintegrated microor- ganisms, while in that of the control animals only well-preserved and undegenerated bacteria are found. In dealing with typhoid bacilli and cholera spirilla, in connection with which the test is most often used, active motility in the controls is of much help. Should there be extensive degeneration of the bacteria in the exudate of the control animals the test is of no value. 2. Identification of a microorganism by observing its a susceptibility to lysis in a known immune serum in vivo: The technique for this test is practically the same as that of the preceding except that in this case we require a potent known immune serum and normal serum for control. It is necessary, furthermore, that by previous tests we should know the degree of dilution in which the immune serum will cause complete bacteriolysis of the microorganism used in its production. Thus, if we are employing a typhoid immune serum and are about to test by this method an unknown Gram-negative bacillus, we must know the titer of the serum for the typhoid bacillus itself. Mixtures are then made of dilutions of this serum and definite quantities of the microorganism to be tested. It is best, always, to employ from ten to one hundred times the amount of immune serum which suffices to produce lysis with its homologous microorganism. Thus, if the serum has been found to be active in dilutions of 1 : 1 ,000, it is employed in the test in dilutions of 1 : 1,000, 1 : 100, and 1 : 10. These dilutions are then injected into guinea-pigs in quantities of 1 c.c. together with the bacteria to be tested, and control guinea-pigs are injected with undiluted normal serum mixed with the bacteria and with salt solution and the bacteria. The exudates are then observed in the same way as in the preceding experiment. Bactericidal Reactions in the Test Tube. — Bactericidal reactions in the test tubes may be made by mixing in small sterile test tubes, definite quantities of the bacteria with inactivated s^rum and com- plement, the latter in the form of unheated normal serum. The mixtures, diluted with equal volumes of neutral broth or salt solution, are set away for a definite time three to four hours in an incubator at 258 INFECTION AND IMMUNITY 37.5° C, and equal quantities from all the tubes are then inoculated into melted agar at 40° C, and plates are poured. Control plates must be made in each case with mixtures of similar quantities of bacteria in salt solution, and similar quantities of bacteria in normal serum. By colony counting after the plates have developed, it is then possible to estimate the degree of bacterial destruction in any of the given dilutions. In actually carrying out the test, dilutions of the inactivated serum are first made, ranging from 1 : 10 to 1 : 1,000 and over. An emulsion of bacteria from a twenty-four-hour agar slant is then made in salt solu- tion, or a twenty-four-hour broth culture properly diluted may be used. Complement is obtained by taking fresh normal rabbit serum and dilut- ing it with salt solution 1 : 10 or 1 : 15. Into a series of test tubes, then, 1 c.c. of each of the serum dilutions is placed, and to each tube is added 0.5 c.c. of the diluted fresh normal rabbit serum (complement) . To these mixtures the bacteria are then added. In adding the bacterial emulsion to these tubes, the writers have found it more accurate to discard the use of the platinum loop and to measure the bacterial emulsion in a marked capillary pipette such as that used in the opsonin test. (See page 285, Fig. 68.) The controls are set up in a similar way, all of them con- taining a similar quantity of bacterial emulsion, one control containing 1.5 c.c. of salt solution, another control containing 1 c.c. of salt solution + 0.5 c.c. of the diluted complement, and the third control containing in- activated normal serum 1 c.c. +0,5 c.c. of diluted complement. Defi- nite quantities of these mixtures, taken with a standard loop, or preferably with a capiUary pipette, are plated in agar immediately after mixing. BACTERICIDAL TEST IN VITRO (To Determine the Bactericidaij Power of a Typhoid Immune Serum AGAINST Typhoid Bacilli). Plates Poured After 3 Hrs. at 37° C. .c. Immune Typh. Ser. 1 : 200 +0.5 c.c. Typh. Emulsion +0.5 c.c. Rab.Ser. 1:15 ? 0 "^ •• 1:400 +0.5 " '^ " +0.5 " " " " f Colo 1:800 +0.5 " " " +0.5 " 1:1600 +0.5 '• " '• +0.5 ' 1:3200 +0.5 " " " . +0.5 " 1:6400 +0.5 " " " +0.5 " 1:12800 + 0.5 " " " +0.5 " 1:25600 + 0.5 " " " +0.5 *• Controls Colonies. 100-1,000 Colonies More than 10,000 Colonies la ( 1.5 c.c. NaCl+0.5 Typh. Emulsion Plated immediately) More than -^1.5 " " +0.5 " " " after 3 hrs. > 10,000 IIb(l. " " +0.5 " " +0.5c.c.Rab. Ser. 1:15 " " " " ) Colonies THE TECHNIQUE OF SERUM REACTIONS 259 After Incubation for two or three hours similar quantities are again measured into tubes of melted agar with the capillary pipette. With a little practice, great accuracy in these measurements can be acquired. The inoculated agar tubes are very thoroughly mixed, and plates are poured. At the end of twenty-four hours' incubation, an enumeration of the colonies in the various plates is made and the results are compared. The in vitro bactericidal tests have been employed, practically, chiefly in the diagnosis of typhoid fever by Stern and Korte.^ While the serum of normal individuals shows practically no bactericidal power for typhoid bacilli, the sera of typhoid patients may be actively bacteri- cidal in dilutions as high as 1 : 50,000. Hemolytic Tests. — Determination of the hemolytic action of blood serum, bacterial filtrates, and of a variety of other substances, such as tissue extracts and animal and plant poisons, is frequently made in bacteriological laboratories. FamiUarity with the methods of carrying out such tests is especially essential since hemolytic tests are also em- ployed in determining other serum reactions, such as the " complement- fixation tests " discussed in another section. For these tests it is necessary to prepare washed red corpuscles of the species of animal against which the hemolysins are to be tested, and to obtain these, blood may be taken in one of the following ways: A. If small quantities of blood corpuscles are desired, the blood may be received into a sterile test tube into which a copper or other wire bent into a loop at the lower end has been introduced. This is used to prevent clotting and to remove the fibrin. Immediately after receiving the blood into this tube, the wire is twirled between the fingers so that the blood is beaten by the wire as by an egg-beater. At the end of five minutes of continuous agitation, the fibrin adhering in a mass to the wire may be lifted out. The corpuscles are then washed and centrifugalized in several changes of salt solution to remove all traces of serum, and are finally emulsified in salt solution. B. The blood may be taken into a centrifuge tube and immediately centrifugalized before clotting has taken place. The plasma is then poured off and the corpuscles are washed with salt solution, as before, to remove the serum. C. The blood may be taken directly into a solution containing five-tenths per cent sodium chlorid and one per cent sodium citrate. » Stem und Korte, Berl. klin. Woch., 1904. 260 INFECTION AND IMMUNITY The corpuscles are concentrated by centrifugalization, the citrate solu- tion is decanted, and corpuscles are washed with salt solution, as before, to remove the serum. D. When large quantities of blood are desired, either from man or from an animal, the blood may be received directly into a flask into which a dozen or more glass beads or short pieces of glass tubing have been placed. The flask is shaken for five or ten minutes, immediately after the blood has been taken and, in this way, defibrination is accom- plished. Since, for comparative tests, it is necessary to establish some stand- ard concentration of red blood cells, it is customary in these tests to employ a five per cent emulsion of corpuscles in salt solution. To obtain this, one volume of sediment of washed red blood cells is mixed with nineteen parts of 0.85 per cent salt solution.^ Such an emulsion, if kept sterile and in the refrigerator, will serve for hemolytic tests for from one to three days. An emulsion should not be used if the super- natant salt solution shows any transparent redness, as this indicates hemolysis. If the substance in which hemolysins are to be determined is serum, this should be inactivated by exposure to 56° C. in a water bath, and to each test, complement may be added in the form of fresh guinea-pig or rabbit's serum. No absolute rule for the quantity of complement to be used in these tests can be given. In each case the particular complement used should be titrated to determine the minimum quan- tity which will produce hemolysis of 1 c.c. of the sensitized cell sus- pension. In the actual test, mixtures are made of the corpuscle emulsion, the inactivated immune serum, and complement in small test tubes and the volumes of the various tubes made equal by the addition of definite quantities of salt solution. The contents of the tubes are thoroughly mixed and the tubes put in the incubator or in a water bath at 37.5° C. If complete hemolysis occurs, the fluid in the tube will assume a deep Burgundy red. If no hemolysis occurs, the fluid will remain uncolored and the corpuscles wiU settle out. Incomplete hemolysis will be evi- denced by a lighter tinge of red in the tube and by the settling out of a varying quantity of blood corpuscles. * The method here given was formerly much employed. It is now the general practice, however, to use one volume of the actual sediment to nineteen volumes of salt solution. THE TECHNIQUE OF SERUM REACTIONS 261 In all hemolytic tests the time element is important. No hemolysis should be adjudged as incomplete unless at least one hour has elapsed. Isohemolysins and Iso-agglutinins. — It is often necessary to carry out hemolytic and hemagglutinating tests on the blood corpuscles of one human being with the serum of another in order to determine the ad- visability of performing transfusion. In this case, the serum of the re- cipient is mixed with a corpuscle emulsion of the cells of the donor, and vice versa. This is conveniently done in small pipettes by the method of Ottenberg. By the determination of iso-agglutinins and isohemolysins all human beings can be divided into four groups according to Landsteiner and others. Table fob . Iso-Agglutinins — Sera I II III IV 1 2 3 4 5 6 7 8 9 10 I II - m IV ^ 5 1 0 0 0 0 0 0 0 0 0 0 2 0 0 0 0 0 0 0 0 0 0 3 0 0 0 0 0 0 0 0 0 0 4 0 0 0 0 0 0 0 0 0 0 i 5 + + + + 0 0 0 + + 0 - 6 + + + + 0 0 0 + + 0 ' 7 + + + + 0 0 0 + + 0 8 + + + + + + + 0 0 0 9 + + + 4- + + + 0 0 0 10 + + + + + + + + + 0 These groupings are permanent and inheritable, following MendeFs laws. 262 INFECTION AND IMMUNITY The Determination of Antibodies in Sera by Complement Fixation. — The principle of complement fixation, discovered by Bordet and Gengou ^ in 1901, has been utilized both in bacteriologicd investigations, and in practical diagnosis for the determination in serum of the presence of specific antibodies. The reaction depends upon the fact that when an antigen, i.e., a substance capable of stimulating the formation of anti- bodies, is mixed with its inactivated antiserum, in the presence of com- plement, the complement is fixed by the combined immune body and antigen can no longer be found free in the mixture. If such a mixture is allowed to stand at temperature for an hour or more, and to it is then added an emulsion of red blood cells together with inactivated hemolytic serum, no hemolysis will take place, since there is no free complement to complete the hemolytic system. If, on the other hand, the original mixture contains no antibody for the antigen used, the complement present is not •fixed and is available for the activation of the hemolytic serum later added. The reaction thus depends upon the fact that neither antigen ^ alone, nor amboceptor (antibody) alone, can fix complement, but that this fixation is carried out only by the combination of antigen plus ambo- ceptor. Any specific can be determined by this method, provided the homologous antigen is used; and vice versa, by the use of a known anti- body a suspected antigen may be determined. When testing immune sera for antibodies given rise to in man or animals by microorganisms which can be cultivated, either the whole bacteria or extracts of the bacteria may be used as an antigen. For the diagnosis of syphilis by this method, in the so-called " Wasser- mann reaction," the antigen employed was originally obtained by the extraction of sjrphilitic organs, in which free syphilitic antigens, i.e., uncombined products of Spirochaete pallida, were assumed to be present. It has been more recently shown, however, that the Wassermann reaction is not specific in any sense of the word, and that suitable anti- gens can be produced by the alcoholic extraction of lipoids from the normal organs of many animals and man. Bacterial extracts for complement-fixation can be made in various ways. The use of thick salt solution suspensions of the cultures them- selves is not advisable because of the anticomplementary action of such suspensions. Good bacterial antigens can be produced by cen- * Bordet and Gengou, Ann. de I'inst. Pasteur, xv, 1901. 2 Bordet and Gay, Ann. de I'inst. Pasteur, xx, 1906. THE TECHNIQUE OF SERUM REACTIONS 263 trifugalizing them from salt solution suspensions and adding to about 20 mgms., 90 mgms. of common salt, rubbing up with a glass rod for an hour, and then adding distilled water to isotonicity. This is the method of Besredka. This method has been used with success by- Miller and Zinsser in the case of tubercle bacilU for complement-fixation in tuberculosis. Wassermann and Bruck^ prepare bacterial antigen by emulsifying growths of about ten agar slant cultures in 10 c.c. of sterile, distilled water. This is shaken for twenty-four hours in a shaking apparatus. At the end of this time 0.5 per cent of carbolic acid is added and the fluid cleared by centrifugalization. The Wassermann Test for the Diagnosis of Syphilis. ^ — The sub- stances for the test are the following: I. The Antigen. — In their original experiments, Wassermann and his collaborators made use of salt-solution extracts of the organs (chiefly of the spleen) of a syphilitic fetus. The tissue was cut into small pieces and to one part by weight of this substance, four parts of normal salt solution and 0.5 per cent of carbolic acid were added. This was shaken in a shaking apparatus for twenty-four hours, and after this the coarser particles removed by centrifugalization. The reddish supernatant fluid was used as the antigen and could be preserved for a long time in dark bottles in the ice chest. Alcoholic extracts of syphilitic organs were subsequently used by a number of authors, syphilitic liver being extracted for twenty-four hours with five times the volume of absolute alcohol. This was filtered through paper and the alcohol evaporated in vacuo at a temperature not above 40° C. About 1 gram of this material was then emulsified in 100 c.c. of salt solution to which 0.5 per cent of carbolic acid has been added. It was soon found that the Wassermann antigen was a purely non- specific substance, and since this discovery was made there are few laboratories in which syphilitic organs are at all used. It appears that lipoidal extracts from almost any tissue can be employed, and that fairly useful antigens can even be obtained with solutions of commer- cial lecithin and mixtures of commercial lecithin and sodium oleate. It is apparent, therefore, that in the Wassermann reaction an even sus- * Wassermann und Bruck, Med. Klinik, 55, 1905, and Deut. med. Woch., xii, 1906. 2 Wassermann, Neisser und Bruck, Deut. med. Woch., xix, 1906; Wassermann, Neisser, Bruck und Schucht, Zeit. f. Hyg., Iv, 1906. 264 INFECTION AND IMMUNITY pension of lipoidal substances constitutes the antigen, and that the complement-fixing complex is made by these antigens in combination with some substance spoken of by Noguchi as ''lipotropic" in the syphilitic serum, which has probably no relation to true antibody. Our own work with treponema pallidum antigen would tend to con- firm this, as well as the experience of Noguchi, Craig and Nichols, Kolmer, and 6thers, who have found that a pure treponema pallidum extract gives reactions in only a few late tertiary cases, running not at all parallel to the fixations obtained with non-specific lipoidal sub- stances. Although we are, at the present writing, still in the dark as to whether the syphilitic antigen depends for its properties upon the lipoidal nature of the extracts or upon the size and dispersion of the particles present in the extracts, we can still assert that the test is reliable and, with care in execution and interpretation, of enormous value in the diagnosis of syphilis. However, it is necessary to recognize that it is surely nofto specific antigen-antibody reaction. The antigens most commonly in use today are prepared as follows: 1. Beef heart or guinea-pig heart muscle is finely chopped up and extracted in five times its volume of absolute alcohol. This mixture is kept 5 to 7 days in the incubator, being frequently shaken. It is then filtered and titrated. Human heart muscle may also be used. 2. Noguchi's Acetone Insoluble Lipoid Antigen. Fresh spleen is macerated and extracted for 5 to 7 days in the incubator in five times its volume of absolute alcohol, being frequently shaken. It is then fil- tered and evaporated to dryness with the aid of a fan. The sticky residue is taken up in a small quantity of ether and this ether solution poured into four times its volume of C. P. acetone. The floccular pre- cipitate which forms is collected and can be preserved under acetone. About 0.2 grams of this paste is dissolved in 5 c.c. of ether. This is shaken up with 100 c.c. of salt solution until the ether is evapprated. The resulting antigen is titrated. 3. Cholesterinized Antigen. According to the researches of Sachs and Rondoni, Browning and Cruikshank, and Walker and Swift, an- tigen can be made more delicate by the addition of cholesterin. Walker and Swift recommend that an alcoholic extract of human or guinea-pig heart be made up to a concentration of 0.4 per cent of cholesterin. A large number of other antigens might be mentioned, but we think that the three mentioned above represent the most important, and in principle all of those at present in common use. Before an antigen can be used for the actual test, it is necessary to THE TECHNIQUE OF SERUM REACTIONS 265 determine the quantity which will furnish a valid result. The sub- stances which are used as antigens often have the power, if used in too large quantity, of themselves binding complement. It is necessary, therefore, to determine the largest quantity of each given antigen which may be used without exerting an anti-complementary action, i.e., which will not inhibit in the presence of normal serum but which will at the same time inhibit hemolysis when syphilitic serum is used. This is done by mixing graded quantities of the antigen with a constant quantity of complement (0.1 c.c. of fresh guinea-pig serinn), in duplicate sets, adding to each tube of one set 0.2 c.c. of a normal serum, and to the other 0.2 c.c. of a known syphilitic serum. These substances are allowed to remain together for one hour and then red blood corpuscles and inac- tivated hemolytic serum are added. The quantity which has given complete inhibition with the syphilitic serum, but absolutely no inhibi- tion with normal serum, is the one to be employed in subsequent re- actions. Before actual use, it is convenient to make a dilution of antigen in salt solution in such a way that 1 c.c. shall contain the amount re- quired. Thus if 0.05 c.c. is wanted, mix 0.5 c.c. with 9.5 c.c. salt solu- tion. Then 1 c.c. of this can be added to each tube in the test. II. The Hemolytic Serum, — The hemolytic amboceptor, for the reaction, is obtained by injecting into rabbits the washed red blood corpuscles of a sheep. A 5 per cent emulsion of the corpuscles is made and of this 5 c.c, 10 c.c, 15 c.c, etc., are injected at intervals of five or six days. Three or four graded injections of this kind are usually sufficient to furnish a serum of adequate hemolytic power. The injec- tions may be made intraperitoneally or intravenously. About nine or .ten days after the last injection of corpuscles, the rabbit is bled from the carotid artery and the serum obtained by pipetting it from the clot. It is best to have a hemolytic serum of high potency in order that the quantities used for the reaction may be as small as possible. This is desirable because of the fact that the serum may contain small amounts of precipitins for sheep's serum, due to insufficient washing of the cor- puscles employed in the immunization. It is necessary to carefully titrate the hemolytic serum. For the actual reaction most observers make use of two hemolytic units. A hem- olytic unit is the quantity of inactivated immune serum which, in the presence of complement, suffices to cause complete hemolysis in 1 c.c of a 5 per cent emulsion of washed blood corpuscles. It is the custom in most laboratories today to halve all the quantities, using 0.5 c.c of the suspension instead of 1.0 c.c and other ingredients accordingly. 18 266 INFECTION AND IMMUNITY Noguchi ^ has pointed out very clearly the dangers of not delicately adjusting the quantity of amboceptor used in the reaction. He calls attention to the experiments of Morgenroth and Sachs ^ who have shown that the relationship between complement and amboceptor necessary for hemolytic reactions is one of inverse proportions. In their own words, ''in the presence of larger quantities of amboceptor, smaller quantities of complement suffice, '^ and vice versa. Noguchi, in his work, has found that, while in the presence of one unit of ambo- ceptor, 0.1 c.c. of guinea-pig's complement is required to produce hemolysis, by using four, eight, and twenty units of amboceptor, com- plete hemolysis is obtainable with one-third, one-fifth, and one-tenth of the 0.1 c.c. of complement, respectively. For this reason an excess of amboceptor might result in complete hemolysis in a test, if a small fraction of the complement were left unfixed by the syphilitic antibody. Another result of an excess of amboceptor would consist in a -partial dissociatijon of the complement from its combination with the antigen- antibody compound. As Noguchi puts it, "a quantity of syphilitic antibody just sufficient to fix 0.1 c.c. of the complement against two units of the amboceptor is no longer efficient in holding back the com- plement from partial liberation against the influence exerted by more than four units of the amboceptor." From these considerations it follows that the serum from rabbits immunized against sheep corpuscles must, in each case, be titrated in order to determine the hemolytic unit. For this purpose a number of mixtures are made in test tubes, containing each 0.1 c.c. of complement (fresh guinea-pig serum), 1 c.c. of a 5 per cent emulsion of sheep's cor- puscles, and diminishing quantities of the inactivated hemolytic serum, thus: = complete hemolysis = complete hemolysis = complete hemolysis = complete hemolysis = complete hemolysis ,0009 c.c. = partial hemolysis .0005 c.Q. = no hemolysis .0003 c.c. = no hemolysis. ^ » 1 c.c. .1 c.c. of " of 5 per cent emul- sion sheep's corpus- Inac- complement tivated fresh guinea-pig serum. - + - ' + - hemo- * lytic serum. cles. .01 c.c. .009 c.c. .005 c.c. .003 c.c. .001 c.c. 1 Noguchi, Proc. Soc. for Exper. Biol, and Med., VI, 3, 1909. 2 Morgenroth und Sachs, in Ehrlich's "Gesammelte Arbeiten," etc., Berhn, 1904. 3 In each tube the volume of the mixture should be made up to 5 c.c. with 0.85 per cent salt solution. THE TECHNIQUE OF SERUM REACTIONS 267 In the given case, 0.001 c.c. of the serum represents one unit, and 0.002 c.c, two units, is the quantity to be used for each test. III. The Complement. — The complement used in Wassermann re- action is fresh guinea-pig serum. This may be obtained in one of the following ways : A guinea-pig may be killed by an incision in the throat and the blood allowed to flow into a large Petri dish. This is set away in the ice chest until clear beads of serum have formed upon the sur- face, and these are then carefully removed with a pipette. It is more economical to puncture the heart of large guinea-pigs with a needle attached to a syringe and withdraw 5 or 6 c.c. of blood without Idlling the animal. This can be transferred to a centrifuge tube and the serum obtained by centrifugation after clotting. Serum used as complement in the Wassermann reaction must be titrated each day before reactions are done. This is, done by putting into a series of tubes 1.0 c.c. (or if half quantities are used, as with us, 0.5 c.c.) of the cell suspension sensitized with 2 units of amboceptor, and adding to these tubes varying quantities of guinea-pig serum. The guinea-pig serum is best diluted 1:10 in salt solution, and quantities ranging from 0.05 to 0.35 c.c. are added to the tubes. The unit is the amount in the tube which shows complete hemolysis at the end of an hour. The re- actions are usually complete in about 30 minutes. Two units of the com- plement are used in the ordinary test. The titration of the complement is one of the most important steps in accurate work. IV. The Sheep Corpuscles. — The sheep corpuscles for the actual re- action are obtained by receiving the blood in a small flask containing a sterile solution of a 0.5 per cent sodium citrate and 0.85 per cent sodium chloride, or into one containing glass beads or short pieces of glass tubing. In the former case, the citrate solution prevents clotting and the corpuscles may be washed free from the citrate solution and emulsified in salt solution before use in the test. In the latter case, it is necessary to shake the blood in the flask immediately after taking, and to continue the shaking motion for about ten minutes. The cor- puscles are washed free from serum by at least 3 washings in salt solu- tion. A 5 per cent suspension of the corpuscles is employed for the test, made by measuring the bulk of centrifugalized corpuscles and adding nineteen parts of sterile salt solution. V. The Serum to he Tested for Syphilitic Antibody. — The serum of the patient is best obtained in the same way that blood is obtained for blood cultures. After surgical precautions, a needle is plunged into the median basilic vein and 3 or 4 c.c. of blood are removed. Before use 268 INFECTION AND IMMUNITY ' \ for the test, the patient's serum must be inactivated by heating in a water bath to 56° C. for twenty minutes to half an hour. The Test. — The actual test for antibody in a suspected serum is carried out in the following way: In a test-tube of suitable size, 2 units of the complement, 0.2 c.c. of the inactivated suspected serum, and the antigen, in quantity determined by titration, are mixed, and the total volume brought up to 3 c.c. with normal salt solution. This mixture is thoroughly shaken, and placed for one hour in a water bath or in the incubator at 37.5° C. Recently it has been found that more delicate results are obtained when the fixation is allowed to take place in the refrigerator for three or four hours — the so-called *' ice-box method." At the end of this preliminary incubation there is added 1 c.c. of a 5 per cent emulsion of sheep's corpuscles, and two units of hemolytic amboceptor, determined by a titration of the inactivated hemolytic rabbit serum, as described above. This inixture is again placed at 37.5° C. for one to two hours. If the antibody is present in the suspected serum, no hemolysis takes place. If absent, hemolysis is complete. In our own work all tests are done in half the quantities of the original Wassermann. Hence only 0.1 c.c. of the patient's serum, and the antigen and complement as determined in titrations with 0.5 c.c. of the cells are mixed in a total volume of 1.5 c.c. At the end of the preliminary incubation, 0.5 c.c. of cells previously sensitized with 2 units of amboceptor are added. No test is of use unless suitable controls are made. The controls set up should be as follows: Control 1. For each serum tested the mixture described above, omitting antigen. Controls 2 and 3. The mixture made as in the test but with known syphilitic serum (2) with and (3) without antigen. Controls 4 and 5. The mixture made as in the test, but with normal serum (4) with and (5) without antigen. Controls 6 and 7. The hemolytic system, complement, blood cells and amboceptor, set up in order to show that the system is in working order (6) with and (7) without antigen. It is convenient to set the tubes in two rows in a rack, the front row containing antigen, the back row containing the same mixture without antigen. In a positive test, the test itself, and Control 2, alone, should show inhibited hemolysis. The other tubes should show complete solution of the hemoglobin. (See scheme, p. 259.) Modifications of the Wassermaim Test. — Since the original for- THE TECHNIQUE OF SERUM REACTIONS 269 mulation of the Wassermann reaction a great many modifications have been suggested by various workers, some of them being radical changes involving the altering of the hemolytic system; others, however, merely adding precautions here and there to increase the delicacy of the reac- tion. The literature on this subject is too voluminous to be com- pletely covered. We, indicate, therefore, some of the most important changes from the original that have been found valuable, and give in greater detail the methods as -at present in use in our own laboratory. Bauer has called attention that human serum contains a certain amount of natural- hemolysin for sheep corpuscles. In his original modification, therefore, he does not use hemolytic rabbit serum as amboceptor. His modification as a whole cannot be accepted for general use because human sera do not contain a uniform amount of hemolysin for sheep cells, and some contain none whatever. However, the presence of natural amboceptor, so-called, in human sera is taken account of by many workers, and it is important to recognize this, since naturally it adds to the amboceptor added with sensitized cells -and leads to a lack of uniformity in the dosage of amboceptor in individual tubes if included. Noguchi has worked out a test in which the difficulties presented by the pres- ence of normal sheep amboceptor are eliminated, in that he uses an antihuman hemolytic serum and human cells as the hemolytic system. It enables him also to use the cells of the patient or of any other human being, thus eliminating the neces- sity of getting fresh sheep cells. His tests are set up as follows: Tube 1. 1 drop patient's serum + complement (0.1 c.c. of 40 per cent guinea-pig serum) + antigen- Tube 2. 1 drop patient's serum 4- complement. (No antigen.) Tube 3. 1 drop known syphilitic serum + complement + antigen. Tube 4. 1 drop known syphilitic serum + complement. (No antigen.) Tube 5. 1 drop known normal serum + cojTiplement + antigen. Tube 6. 1 drop known normal serum + complement. (No antigen.) Tube 7. Complement alone (for hemolytic system control). To each tube then add 1.0 c.c. of the one per cent emulsion of human corpuscles. Shake mixtures thoroughly and incubate or place in water bath at 38-40° C. for one hour. Then add to each tube 2 units of antihuman amboceptor (serum of rabbit immunized with human cells) and replace in water bath for one hour. At the end of this time in a positive test there wiU be no hemolysis in Tubes 1 and 3 while all the other tubes will show hemolysis. The method of Noguchi is still used by a few investigators, but is not at present in common use, though we have no doubt that if sys- tematically followed it would develop as quite satisfactory. The tests are done in our own laboratory with the original sheep cell — antisheep serum hemolytic system. They are done in half quan- tities, titrations being made with 0.5 c.c. of a 5 per cent emulsion of washed sheep cells. Each day the complement (fresh guinea-pig diluted 1 : 10) is titrated with cells sensitized with 2 units of stock amboceptor. Fresh amboceptors are titrated from time to. time so that a reasonable constancy is obtained. The hemolytic system is kept as constant as possible from day to day. The unit (minimal hemolytic amount) of a new specimen of amboceptor is determined by titrating it on a num- ber of successive days with 0.5 c.c. of 5 per cent cells and 0.5 c.c. of 10 per cent guinea-pig sera, readings being made at the end of a half 270 THE TECHNIQUE OF SERUM REACTIONS hour. When complement is then subsequently titrated with 2 such amboceptor units and 0.5 c.c. of 5 per cent red cells, it is usually found that the minimal hemolytic dose of complement lies between 0.2 and 0.25 c.c, and in the actual tests twice this minimal hemolytic dose of complement is used with the 2 units of amboceptor. The daily titra- tion of complement frequently shows marked variations even though SCHEME FOR WASSERMANN TEST. ADAPTED TO ORIGINAL WASSERMANN SYSTEM AFTER SCHEME OF NOGUCHI. Test with Unknown Serum. Test with Known Positive Syphilic Serum. Test with Known Negative Normal Serum. Test without Serum to Control Efficiency of Hemolytic System. £H Serum .2 c.c. d 6 Serum ..2 c.c. Serum .2 c.c. ■ ^1 .+' CO + +. ^1 Q Complement s 0 Complement Q Complement 0 Complement g = .1 c.c. 1 .1 c.c. .1 c.c. .1 c.c. ^9 + > + + + g Salt sol. 3. c.c. 3 o Salt sol. 3. c.c. Salt sol. 3. c.c. Salt sol. 3. c.c. 2. A 4, 6. 8. Serum .2 c.c. Serum .2 c.c. Serum .2 c.c. + 6 + + o Loeh, Cent. f. Bakt., xxxii, 1902. • STAPHYLOCOCCUS PYOGENES AUREUS 325 Upon coagulated animal sera, rapid growth takes place and eventually slight liquefaction of the medium is said to occur. In nitrate solutions, reduction of the nitrates to nitrites is caused. In Dunham's broth, indol is formed. In media containing the carbohydrates — dextrose, lactose, or sac- charose— acidification takes place with the formation chiefly of lac- tic, butyric, and formic acids. There is no gas formation, however. In proteid media free from sugars, the staphylococcus produces alkali. The reducing action of staphylococcus is shown by decolorization in cultures of litmus, methylene-blue, and rosanilin.^ Pigment Formation. — Differentiation between the various members of the staphylococcus group is based largely upon the formation of pigments. These pigments, so far as we know, seem to be species characteristics. Thus, Staphylococcus pyogenes aureus is recognized primarily by its production of a yellowish-brown pigment, varying in different strains from a pale brown hue to a deep golden yellow. Pro- longed cultivation upon artificial media may lead to a diminution in the depth of color produced.^ It appears only when cultivation is carried on under freely aerobic conditions, anaerobic cultivation resulting in unpigmented colonies. The coloring matter is insoluble in water but soluble in alcohol, chloroform, ether, and benzol.^ According to Schnei- der,* the pigment belongs to the class of '4ipochromes " or fatty pig- ments, and is probably composed of carbon, oxygen, and hydrogen, without nitrogen. Treatment with concentrated sulphuric acid changes it to a green or greenish-blue.^ Resistance. — Although not spore formers, staphylococci are more resistant to heat than many other purely vegetative forms. The thermal death point given for Staphylococcus pyogenes aureus by Sternberg* lies between 56° and 58° C, the time of exposure being ten minutes. The same author states that, when in a completely dried state, the coccus is still more resistant, a temperature of from 90° to 100° C. being re- quired for its destruction. Against low temperatures, staphylococci are extremely resistant, repeated freezing often failing to sterilize cultures. » Ft. Muller, Cent. f. Bakt., xxvi, 1899. 2 Fliigge, "Die Microorg.," etc. 3 Migula, "System d. Bakt.," Jena, 1897. * Schneider, Arb. a. d. bakt. Inst., Karlsruhe, 1, vol. i, 1894. « Fischer, " Vorles. fiber die Bakt.," Jena, 1903. ^Sternberg, "Textbook," etc., N. Y., 1901, p. 375. 326 PATHOGENIC MICROORGANISMS Desiccation is usually well borne, staphylococci remaining alive for six to fourteen weeks when dried upon paper or cloth.* On slant agar, staphylococci may be safely left for three or four months without trans- plantation, and remain alive.^ The resistance of staphylococci to chemicals, a question of great surgical importance, has been made the subject of extensive researches, notably by Liibbert,^ Abbott,^ Franzott,^ and many others. According to Liibbert, inhibition of staphylococcus growth is attained by the use of boric acid 1 in 327, salicylic acid 1 in- 650, corrosive sublimate 1 in 80,000, carbolic acid 1 in 800, thymol 1 in 11,000. Staphylococci are killed by corrosive sublimate 1 in 1,000 in ten minutes, by carbolic acid 1 per cent in 35 minutes, 3 per cent in 2 minutes (Franzott). Ethyl alcohol,® even when absolute, is not very efficient as a disinfectant. Nascent iodin, as split off from iodoform in wounds, is extremely power- ful in destroying staphylococci. Pathogenicity. — Separate strains of Staphylococcus pyogenes aureus show wide variations in relative virulence. The most highly virulent are usually those recently isolated from human suppurative lesions, but no definite rule can be formulated in this respect. The virulence of a given strain, furthermore, may be occasionally enhanced by re- peated passages through the body of a susceptible animal. Prolonged cultivation upon artificial media is liable to decrease the virulence of any given strain, though this is not regularly the case. There are, more- over, unquestionably, many staphylococci constantly present in the air, dust, and water, which although morphologically and culturally not unlike the pathogenically important species, may be regarded as harmless saprophytes. The susceptibility of animals to staphylococcus infection is, likewise, subject to extreme variations, depending" both upon differ- ences between species and upon fortuitous individual differences in susceptibility among animals within the same species. Animals on the whole are less susceptible to staphylococcus than is man. Among the ordinary laboratory animals, rabbits are most sus- ceptible to this microorganism. Mice, and especially the white 1 Deslongchamps, Paris, 1897. ^Passet, Fort. d. Med., 2 and 3, 1885. » Liibbert, "Biol. Untersuch.," Wurzburg, 1886. * Abbott, Medical News, Phila., 1886. * Franzott, Zeit. f. Hyg., 1893. ^Hanel, Beit. z. klin. Chir., xxvi STAPHYLOCOCCUS PYOGENES AUREUS 327 Japanese mice, show considerable susceptibility. Guinea-pigs possess a relatively higher resistance/ Subcutaneous or intramuscular inoculation of a susceptible animal usually results in the formation of a localized abscess with much pus formation and eventual recovery. Intraperitoneal inoculation is more often fatal. Intravenous inoculation of doses of 0.5 c.c, or more, of fresh broth cultures of virulent staphylococci usually leads to pyemia with the production of secondary abscesses, located chiefly in the kid- neys and the heart and voluntary muscles, but not infrequently in other organs as well. In the kidney they occur as small foci, situated most often in the cortex, composed of a central, necrotic pus cavity, surrounded by a zone of acute inflammatory exudation. Staphylo- coccus lesions form histologically the typical "acute abscess." Not infrequently the pyemic condition is accompanied by suppurative lesions in the joints. Intravenous injections of virulent staphylococci preceded by injury to a bone is often followed by the development of osteomyelitis. Mechanical or chemical injury of the heart valves preceding intravascular staphylococcus inoculation may result in localization of the infection on or about the heart valves, leading to "malignant endocarditis." The pyemic conditions following staphy- lococcus inoculation usually lead to chronic emaciation and death after an interval dependent upon the relative virulence of the micro- organism, the amount injected, and the resistance of the infected subject. Large doseS of unusually virulent cultures cause death within twenty -four hours, or even less, without abscess formation. As above stated, the susceptibility of man to spontaneous staphy- lococcus infection is decidedly more marked than is tliat of animals. The form of infection most frequently observed is the common boil or furuncle. As Garre,^ Biidinger,^ Schimmelbusch,* and others . Jiave demonstrated by experiments upon their own bodies, energetic rubbing, of the skin with virulent staphylococcus cultures may often be followed by the development of a furuncle. Subcutaneous inoculation of the, human subject invariably gives rise to an abscess. The pathological lesions which may be produced in man by virulent staphylococqi are. naturally of great variety, depending upon the mode of inoculation, and 1 Terin, Ref. in Lubarsch und Ostertag, Ergebnisse, 1896; Lingelsheim, "AetioL d. Staph. Inf.," etc., Wien, 1900. ^Garre, Beit. z. klin. Chir., x, 1893. 3 Budinger, Lubarsch und Ostertag, Ergebnisse, etc., 1896. * Schimmelbusch, Ref. by Budinger. 328 PATHOGENIC MICROORGANISMS upon the relation between the virulence of the incitant and the resist- ance of the subject. Apart from the formation of localized abscesses, staphylococci are common as the incitants of surgical suppurations and wound infections. The large majority of acute suppurative in- flammations of bone (osteomyelitis) are caused by staphylococci. Ab- scesses of the brain, of the liver, and of the lung may be due to this microorganism. It may give rise to ascending infections of the genito- urinary tract, leading to pyelonephritis. Empyema or peritonitis may be caused by its entrance into the serous cavities from the lung or bowel. When gaining achess to the circulation from some localized focus, it gives rise to septicemia and may lead to malignant endocarditis and, by secondary locaUzation in the viscera, to general pyemia. As the incitant of septicemia it can frequently be found by blood culture during the life of the patient. Puerperal sepsis is not infrequently a staphylococcus disease. Of recent years several authors have claimed direct etiological relationship for the Staphylococcus pyogenes aureus with acute articular rheumatism.^ While not unlikely, this claim is not, at present, substantiated by sufficiently exact evidence. Apart from the local inflammatory reactions called forth by staphylococcus invasion, all such infections, if severe or prolonged, give rise to profound toxic manifestations evidenced by characteris- tically irregular temperature (the so-called "septic type"), by head- ache, nausea, and general malaise, and not infrequently by chills. Prolonged chronic infection with staphylococci may give rise to the So-called amyloid changes in liver, spleen, and kidneys. Toxic Products. — Endotoxins. — The dead bodies . of staphylococci injected into animals may occasionally give rise to abscess formation, and,^ if in sufficient quantity, may cause death. To obtain the latter result, however, large quantities are necessary, the endotoxic substances within the dead cell body of these microorganisms being probably neither very poisonous nor abundant.^ That dead cultures of Staphylococcus aureus exert a strong positive chemotaxis for leucocytes was shown beyond question by the experi- ments of Borissow.* Hemolysins. — In 1900 Kraus^ noticed the hemolytic action of ^A.H. Weis, Inaug.-Diss., Berlin, 1901. ^ Schattenfroh, Arch. f. Hyg., xxxi, 1887. 8 V. Lingelsheim, " Aetiol. u. Therap. d. Staph. Krank.," Wien, 1900. *Bori8sow Zieglers Beitr., xvi, 1894. ^Xraus, Wien. klin. Woch,, iii^ 190D. STAPHYLOCOCCUS PYOGENES AUREUS 329 staphylococci growing upon blood-agar plate cultures. Neisser and Wechsbergi then showed that this hemolytic substance, secreted by the staphylococcus, could be demonstrated in filtrates of bouillon cultures. Such hemolysins are produced by Staphylococcus aureus, and, to a lesser degree, by Staphylococcus albus. The quantity produced varies enormously with different strains and seems to be roughly proportionate to the virulence of the particular microorganism, though exceptions to this rule are not uncommon. Absolutely avirulent races do not, so far as we know, produce hemolysins. The culture medium most favor- able to the formation of these substances is, according to Neisser and Wechsberg, a moderately alkaline beef bouillon. Cultivated at 37.5° C, the bouillon contains the maximum amount of hemolytic substance be- tween the eighth and fourteenth day, and this may be separated from the bacteria by filtration through Berkefeld or Chamberland filters. The hemolytic action may be observed by the general technique for determining hemolysis (given on page 259) . It is important to wash the red blood corpuscles used for the experiments, since many animals normally possess small quantities of antihemolysin in their blood-sera (man and horse especially) .^ The red blood corpuscles of rabbits, dogs, and guinea-pigs are extremely susceptible to the action of the staphylo- hemolysin. Those of man are less easily injured by it. The hemolytic action takes place, as Todd^ and others* have shown, not only m vitro, but in the living animal as well. The staphylo-hemolysin is comparatively thermolabile. According to Neisser and Wechsberg, heating it to 56° C. for twenty minutes de- stroys it. According to some other authors, however, higher tempera- tures (60° to 80° C.) are required. Reactivation of a destroyed staphylo- hemolysin has so far been unsuccessful. The fact that antistaphylolysin is occasionally present in normal sera has been mentioned above. This antibody is most abundant in the blood of horses and of man. Arti- ficially antistaphylolysin formation is easily induced by subcutaneous inoculation of staphylolysin into rabbits. Leucocidin. — In 1894, Van de Velde ^ discovered that the pleural exudate of rabbits following the injection of virulent staphylococci showed marked evidences of leucocyte destruction. He was subse- ■ — ~ — I ■ I » Neisser und Wechsberg, Zeit. f. Hyg., xxxvi, 190L ^Neisser, Deut. med. Woch., 1900. 3 Todd, Trans. London Path. Soc, 1902. ^Kraus, Wien. klin. Woch., 1902. 6 Van de Velde, La Cellule, x, 1894 ^ 22 330 PATHOGENIC MICROORGANISMS quently able to show that the substance causing the death and partial solution of the leucocytes was a soluble toxin formed by the staphylo- coccus, not only in vivo, but in vitro as well; for cultures of Staphylo- coccus pyogenes aureus, grown in mixtures of bouillon and blood serum, contained, within forty-eight hours, marked quantities of this '^eucocidin." Other workers since Van de Velde have evolved various methods for obtaining potent leucocidin. Bail* obtained it by growing virulent staphylococcus in mixtures of one-per-cent glycerin solutions and rab- bit serum. Neisser and Wechsberg^ advise the use of a carefully titrated alkaline bouillon. To obtain the leucocidin l^e from bacteria, the cultures are passed through Chamberland or Berkefeld filters, after about eight to eleven days' growth at 37° C, at which time the con- tents in leucocidin are usually at their highest point. The action of leucocidin upon leucocytes may be observed in vivo by the pimple method of Van de Velde, of injecting virulent staphylo- cocci intrapleurally into rabbits and examining the exudate. Bail advises the production of leucocytic intrapleural exudates by the use of aleuronat and following this after twenty-four hours by an injection of leucocidin-filtrate. In vitro the phenomenon may be observed by direct examination of mixtures of leucocytes and leucocidin in the hanging drop on a warmed stage, or by the '^methylene-blue method'', of Neisser and Wechsberg. This method is based upon the fact that living leucocytes will reduce methylene-blue solutions and render them colorless, while dead leucocytes have lost this power. Leucocidin and leucocytes are allowed to remain in contact for a given time and to them is then added an extremely dilute solution of methylene-blue. If the leucocytes have been actively attacked by leucocidin, no reduction takes place. This method is particularly adapted for quantitative tests. All staphylococcus strains do not produce leucocidin to the same degree. Almost all true Staphylococcus pyogenes aureus cultures produce some of this toxin, but one strain may produce fifty- and a hundred-fold the quantity produced by another. Staphylococcus pyogenes albus gives rise to this substance but rarely, and then in small quantity. Leucocidin seems to be similar to the soluble toxins of other bacteria. It is rapidly destroyed by heat at 58° C, and deteriorates quickly in iBail, Arch. f. Hyg., xxxii, 1898. ^Neisser und Wechsberg, Zeit. f. Hyg., xxxvi, 1901. STAPHYLOCOCCUS PYOGENES AUREUS 331 Culture fluids at incubator temperatures. It is distinct from staphylo- hemolysin as shown by differences in thermostability. Soon after Van de Velde's discovery of leucocidin, Denys and Van ■ i Av.^ '■ 1 ^ .-• / ■ / V f t ^ ./ v^^;.^--. . ! / * Fig. 72. — Streptococcus pyogenes. influence in placing our knowledge of the pathogenic properties of streptococci upon a scientific basis. Morphology and Staining. — ^The individual streptococcus is a spherical microorganism measuring from 0.5 micron to 1 micron in diameter. Since the line of cleavage of cocci, when in chains, is perpendicular to the ^Fehleisen, " Aetiol. d. Erysipelas," Berlin, 1883. " Rosenbach, " Mikroorg. bei Wundinfektion," etc., Wiesbaden, 1884. 8 Passet, " Untersuch. iiber die eitrigen Phlegm.," etc., Berlin, 1885. STREPTOCOCCUS PYOGENES 337 long axis of the chain, adjacent cocci often show slight flattening of the contiguous surfaces, forming, as it were, a series of diplococci arranged end to end. As a general rule the streptococci pathogenic for man, when grown upon favorable media, have a tendency to form chains made up of at least eight or more individuals, while the more saprophy- tic,, less pathogenic varieties are apt to be united in shorter groups. Upon this basis a rough morphological distinction has been made by v. Lingelsheim,^ who first employed the terms Streptococcus "longus" and "brevis." A differentiation of this kind can hardly be re- lied upon, however, since the length of chains is to some degree de- pendent upon cultural and other environmental conditions. Species which exhibit long and tortuous chains, when grown upon suitably alkaline bouillon, or ascitic broth, may appear in short groups of three or four, or even in the diplo form, when cultivated upon solid media or unfavorable fluid media. Stained specimens often show swelling and enlargement of individual cocci, giving the chains an irregularly beaded appearance. These swollen individuals are probably to be interpreted as involution forms and are seen with especial frequency in old cultures. Streptococci do not form spores, are non-motile, and do not possess flagella. There can be no doubt that certain species of true streptococci may possess capsules, though these are not so regularly demonstrable and are more delicately dependent upon cultural conditions than are the capsules of the pneumococci.^ The capsulated streptococci will be dis- cussed more comprehensively in the section upon the differentiation of pneumococcus from streptococcus (page 367). Streptococci are easily stained by the usual anilin dyes. Stained by the method of Gram, the pyogenic streptococci are not decolorized and invariably retain the gt;ntian- violet. Certain species found in stools and described as Gram-negative, are rare and are non-pathogenic. Others of the ^'Streptococcus brevis" variety, and purely saprophytic, may stain irregularly by the Gram method. Cultivation. — ^The pyogenic streptococci are easily cultivated upon all the richer artificial media. While meat extract-pepton media may suffice for certain strains, it is usually better to employ those media which have the beef or veal infusion for a basis. For the cultivation of more delicate strains of streptococci, especially when » V. Lingelsheim, " Aetiol. u. Therap. d. Streptok. Infek." Beit. z. Exp. Ther., Hft. 1, 1899. ^ Pasquale, Zieglers Beit., xii; Bordet, Ann. de I'inst. Pasteur, 1887; Schottmilller, Munch, med. Woch., xx, 1903; Hiss, Jour. Exp. Med., vi, 1905. 338 PATHOGENIC ORGANISMS taken directly from the animal or human body, it is well to add to the media animal albumin in the form of whole blood, blood serum, or asci- tic or pleural transudates. Glucose, added in proportions of one to two per cent, likewise renders media more favorable for streptococcus culti- vation. Prolonged cultivation of all races upon artificial media renders them less fastidious as to cultural requirements. The most favorable reaction of media for streptococcus cultivation is moderate alkalinity (two-tenths to five-tenths per cent alkalinity to phenolphthalein) . Growth may be readily obtained, however, in neutral media or even in these slightly acid. The optimum temperature for growth is at or about 37.5° C. Above 43° to 45° C., development ceases. At from 15° to 20° C., growth, while not energetic, still takes place, an important point in the differentiation of these microorganisms from pneumococci. While the free access of oxygen furnishes the most suitable environment for most races of streptococci, complete anaerobiosis does not pre- vent development in favorable media. Strictly anaerobic streptococci have been cultivated from the human intestinal tract by Perrone ^ and others. In alkaline bouillon at 37.5° C., pyogenic streptococci grow rapidly, form long and tortuous chains, and have a tendency to form flakes which rapidly sink to the bottom. Diffuse clouding occurs rarely and is a characteristic rather of the shorter so-called Streptococcits brevis. When sugar has been added to the broth the rapid formation of lactic acid soon interferes with extensive development. This may be obviated, especially when mass cultures are desired, without sacrifice of the growth- increasing influence of the glucose, by adding to the sugar-broth one per cent of sterile powdered CaCOg.^ In milk. Streptococcus pyogenes grows readily with the formation of acid, followed, in most cases, by coagulation of the medium. On agar-plates at 37.5° C., growth appears within eighteen to twenty- four hours. The colonies are small, grayish, and delicately opalescent. They are round with smooth or very slightly corrugated or lace-like edges, and rise from the surface of the medium in regular arcs, like small droplets of fluid. Microscopically they appear finely granular and occasionally, under high magnification, may be seen to be composed of long intertwining loops of streptococcus chains, which form the lace-like edges. When ascitic fluid or blood serum has been added to agar, growth is more energetic and the colonies correspondingly more rapid in ^ Perrone, Ann, de I'inst. Pasteur, xix, 1905. 2 Hiss, Jour. Exp. Med., vi, 1905. STREPTOCOCCUS PYOGENES 339 appearance and luxuriant in development. In glucose-ascitic-agar, acid formation from the sugar causes coagulation of albumin with the consequent formation of flaky white precipitates throughout the medium.^ In gelatin stab-cultures growth takes place slowly, appearing after twenty-four to thirty-six hours as a very thin white line, or as discon- nected little spheres along the line of the stab. The colonies on gelatin plates are similar in form to those on agar, but are usually more opaque and more distinctly white. The gelatin is not liquefied by the pyogenic streptococci, though certain of the more saprophytic forms may occa- sionally bring about slow fluidification. On Loeffler's coagulated blood serum, growth is rapid and luxuriant, and may show a slight tendency to confluence if the medium is very moist. Good chain formation takes place on this medium. Upon potatoes, growth is said not to take place.2 On media containing red blood cells, most pathogenic streptococci cause hemolysis and decolorization (see Fig. 74, p. 345). It is useful to remember this when examining blood-culture plates, for here the yellow transparent halo of hemo- lysis and decolorization surrounding the colonies may aid in differenti- ating these bacteria from pneumococci. This is of especial importance, since many streptococci, when cultivated directly out of the human blood, do not exhibit chain formation, but appear as diplococci. In the inulin-serum media of Hiss,^ streptococci do not produce acid and coagulation. The so-called Streptococcus mucosus, a capsule- bearing, inulin-fermenting microorganism, is very probably a sub-species of the pneumococcus (see later section). Resistance. — Streptococci on the ordinary culture media, without transplantation and kept at room temperature, usually die out within ten days or two weeks. They may be kept alive for much longer periods by the use of the calcium-carbonate-glucose bouillon, if the cultures are 1 Libmun, Medical Record, Ivii, 1900. * Frosch und Kolle, in Fliigge, "Die Mikroorganismen," 1891. ' Hiss, Jour. Exp. Med., vi; 1905. Fig. 73. — Streptococcus Col- onies, ON Serum Agar. 340 PATHOGENIC ORGANISMS thoroughly shaken and the powdered marble thoroughly mixed with the bouillon from time to time.^ Preservation at low temperatures (1° to 2° C), in the ice chest, considerably prolongs the Hfe of cul- tures. Virulence is preserved longest by frequent transplantation upon albuminous media. In sputum or animal excreta, streptococci may remain alive for several weeks. Streptococci are. killed by exposure to a temperature of 54° C. for ten minutes.2 Low temperatures, and even freezing, do not destroy some races. The action of various chemical disinfectants has been thoroughly investigated by v. Lingelsheim,^ who reports among others the following results : Carbolic acid 1 : 200 kills streptococci in fifteen minutes. In the same time, bichloride of mercury is efficient in a dilution of 1 : 1,500, lysol in a dilution of 1 : 200, peroxide of hydrogen 1 : 35, sulphuric acid 1 : 150, and hydrochloric acid 1 : 150. Inhibition is exerted by car- bolic acid 1 : 550, and by bichloride of mercury 1 : 65,000. Exposure to direct sunlight kills streptococci in a few hours. Virulence and Pathogenicity. — Different races of pyogenic strepto- cocci show considerable variations in virulence, and there are few organ- isms, pathogenic both for animals and man, which show such peculiari- ties in virulence. The character or severity of the lesion in man gives little evidence as to the virulence of the organism for animals. Such differences are, to a certain extent, dependent upon inherent individual characteristics, but are rather more likely to be the consequences of pre- vious environment or habitat. Prolonged cultivation upon artificial media usually results in the reduction of the virulence of a streptococcus, while an originally low or reduced virulence may often be much en- hanced by repeated passage of the streptococci through animals. It is noteworthy, however, that while the passage of a streptococcus through rabbits will usually enhance its virulence for susceptible animals in general, repeated passages through mice may increase the virulence for these animals only, even occasionally depressing the virulence for rabbits.^ Among the donaestic animals, those most susceptible to experimental streptococcus infection are white mice and rabbits. Guinea-pigs and 1 Hiss, loc. cit. ^Sternberg, "Textbook of Bact.," 2d ed., 1901; Harlmann, Arch. f. Hyg., vii. 3y. Lingelsheim, "Aetiol. u. Therap. d. Streptoc. Inf.," etc., Beit. z. Exper. Therap., Hft. 1, 1899. ^ Knorr, Zeit. f . Hyg., xiii. • ' STREPTOCOCCUS PYOGENES 341 rats are less easily infected, and the larger domestic animals, cattle, horses, goats, cats, and dogs, are extremely refractory. Almost complete immunity toward streptococcus infections prevails among birds. The nature of the lesions following animal inoculation depends upon the manner of inoculation, the size of the dose given, and most of all upon the grade of virulence of the inoculated germ. Subcutaneous inoculations, according to the virulence of the inoculated material, may result in a simple localized abscess, differing from a staphylococcus abscess only in the more serous nature of the exudate and the frequent occurrence of edema, or in a severe general septicemia with a hardly noticeable local lesion.- Subcutaneous inoculation of mice results almost invariably in general sepsis followed by death within thirty-six to forty- eight hours, or less, and the presence of streptococci in the heart's blood and the viscera. Intrapleural or intraperitoneal inoculation of suscep- tible animals with virulent streptococci leads usually to a peculiarly hemorrhagic form of exudate, due both to the diapedesis caused by the violent inflammatory process, and to the hemolysis of the red cells by the streptococcic hemolysins. Inoculation of rabbits at the base of the ear with virulent streptococci may result in the formation of a lesion indistinguishable histologically from erysipelas in man. ^ Marbaix ^ has shown that such erysipeloid lesions could be produced in rabbits by streptococci from various and indifferent sources, provided that the virulence of each strain could be sufficiently enhanced. This marked variability of the resulting lesion as determined by the degree of virulence of the incitant, whatever its original source, forms a strong argument in favor of the opinion that all the pyogenic streptococci are members of a single species. Intravenous inoculation of rabbits with virulent cultures usually results in a rapidly fatal septicemia. An animal which has died of a streptococcus infection usually shows serosanguineous edema about the point of inoculation, multiple hemorrhagic spots upon the serous mem- branes, and congestion of the viscera. The microorganisms can almost invariably be found in the heart's blood, in the spleen, and in the exudate about the inoculated area. Microscopically, when the process has lasted sufficiently long, parenchymatous degeneration of all the organs may be observed. In the more chronic infections articular and periarticular lesions may occur.^ 1 Fehldsen, loc. cit. ; Frankel, Cent, f . Bakt., vi, 2 Marbaix, La Cellule, 1892. ^Schiitz, Zeit. f. Hyg., iii; Hiss, Jour. Med. Res., xix, 1908. 342 PATHOGENIC MICROORGANISMS Spontaneous streptococcus disease seems to occur among some of the larger domestic animals. Thus, a contagious form of inflammation of the respiratory passages of horses has been attributed to streptococcus infection.^ Among cattle these microorganisms have been found to produce purulent inflammation of the udder and occasionally post- partum uterine inflammation in cows. Among the smaller labora- tory animals, occasional streptococcus infections may be observed in rabbits. Recently an epidemic disease among white mice due to strep- tococcus was studied by Kutscher.^ As a rule, however, streptococcus disease is by far more rare among animals than it is among human beings. In man, a large variety of pathological processes may be caused by streptococci and here again the nature of the infection, whether definitely localized or generally distributed, depends upon the relationship existing between the virulence of the incitant and the resistance of the subject. The first cultivation of streptococcus from human lesions was made by Fehleisen,^ who obtained them from cases of erysipelas. It was long believed that the so-called Streptococcus erysipelatis was a similar but essentially different species from the common Streptococcus pyogenes. The production of erysipelas in animals with streptococci from other sources, however, has shown definitely that the two groups can not be separated.^ Superficial cutaneous infections are frequently caused by streptococci and these in the milder cases may be similar to the localized abscesses caused by staphylococci. In severe cases, however, infection is followed by rapidly spreading edema, lymph- angitis, and severe systemic manifestations with the development of a grave cellulitis, often threatening life and requiring energetic surgical interference. Invasion of the respiratory organs by streptococci is not rare, and may lead to bronchitis, pneumonia, or empyema. They are frequently present also as secondary invaders in pulmonary tubercu- losis.^ Streptococcus infections of the lungs and pleura not infrequently lead to pericardial involvement. Suppurations of bone may be caused by streptococci, and constitute a severe form of osteomyelitis. Such lesions when occurring in the mastoid bone are not infrequently secondary to streptococcus otitis and may lead to a form of meningitis which is in most cases fatal. In the ^ Van de Velde, Monatsheft Bakt., Thierheilk., ii. * Kvischer, Cent, f . Bakt., xlvi. ' Fehldsen, loc. cit. ^ Marhaix, La Cellule, 1892; Petruschky^ Zeit. f. Hyg., xxiii. 6 Corned, ''Die Tuberkulose," Wien, 1899. STREPTOCOCCUS PYOGENES S43 mouth and throat streptococci may give rise to pharyngitis and are one of the most frequent causes of a form of tonsillitis often clinically indis- tinguishable from diphtheria. The throat inflammation accompanying scarlatina is, almost without exception, referable to streptococcus infec- tion.^ The occasional presence of the streptococcus in the blood of scarlatina patients, moreover, has led some authors to suggest a pos- sible etiological connection between this microorganism and the disease. ^ This, however, is at present merely conjectural. In diphtheric inflammations of the throat, a secondary streptococcus infection is a frequent and serious complication. As incitants of disease of the intestines, streptococci have been found in appendicular abscesses ^ and have been described as the cause of some forms of infantile diarrhea."* From any of the local processes streptococci may pass into the circulation, causing sepsis. The septicemia occurring during the puerperium is most often caused by this microorganism. Secondary foci in the viscera may be established, leading to pyemia,^ or, if these localizations occur upon the heart valves, septic endocarditis may ensue. All such forms of general streptococcus infection, whether running acute or chronic courses, present a high rate of mortality. The diagnosis in these cases is usually easy if blood cultures are taken upon suitable media. Streptococcus throat infections have recently appeared in fulmi- nating epidemics. Several small epidemics were described in England, and three extensive outbreaks have occurred in this country; one in Boston of 1,400 cases; a second in Baltimore of about 1,000 cases, and a third in Chicago of about 10,000 cases. These outbreaks were studied by Winslow, Stokes, Davis,® and by Rosenow.^ In each case the major- ity of infections were traced to a single milk supply, though secondary cases doubtless occurred by contact. Severe complications such as suppurative adenitis, otitis, erysipelas, peritonitis, and septicemia were not uncommon. A similar organism — a capsulated, hemolytic streptococcus — was found in each epidemic. 1 Baginsky, Deut. med. Zeit., 1900. 2 Baginsky und Sommerfeld, Berl. klin. Woch., xxvii, 1900. ' Kelly, "Pathogenesis of Appendicitis." *Lanz and Tavel, Rev. de Chir., 1904; Perrone, Ann. de Tinst. Pasteur, 1905; Escherich, Jahrb. f, Kinderheilkunde, 1899. ^ Lihman, Cent. f. Bakt., xxii. 8 Cited from Capps, Jour. A. M. A., 1912, p. 1848. ' Bosenow, Jour, of Inf. Dis., 1912. 344 PATHOGENIC MICROORGANISMS Toxic Products. — In spite of extensive researches by many inves- tigators upon the nature of the poisons produced by streptococci, our understanding of these substances is still very incomplete. The grave systemic symptoms so often accompanying comparatively slight streptococcus lesions argue strongly for the production by these microorganisms of a powerful diffusible poison. Toxic filtrates of streptococcus cultures have indeed been obtained by Roger/ Marmier,^ Baginsky and Sommerfeld,^ Marmorek,^ and many others; but these have in no case been comparable in potency to the soluble toxins of diphtheria or of tetanus. When injected into young guinea-pigs in sufficient quantity, these filtrates produce rapid collapse and death. The inability to produce strong toxins is generally attributed to the difficulty of obtaining very abundant growth of these bacteria upon fluid media, development being self-hmiting, either because of the ex- haustion of specific nutritive material (Marmorek ^) , or, more probably, because of the inhibitory effects of the products of growth, chiefly acid formation. This last factor can be partially overcome by the use of the glucose-calcium-carbonate broth mentioned above, in which acid neutral- ization is constantly taking place. For toxin production, Baginsky and Sommerfeld ^ advise a strongly alkaline reaction of the media; Mar- morek ^ has used human blood-serum-bouillon with success. The toxins so produced are relatively thermostable. According to v. Lingelsheim, heating to 60° or 70° C. destroys them in part only. The endotoxins contained within the cell-bodies of streptococci them- selves have been found to possess but slight toxic qualities. Apart from these substances, some streptococci produce a hemolysin which has the power of bringing about destruction of red blood cor- puscles. The observation of this phenomenon for streptococci was first made by Marmorek ^ in 1895. According to this author, there is a direct relationship between virulence and hemolytic power* Other investigators, however, notably Schottmiiller,^ believe the hemolytic power to be a constant characteristic of certain strains unchangeable by 1 Roger, Rev. de med., 1892. 2 Marmier, Ann. de I'inst. Pasteur, ix, 1895, p. 533. ' Baginsky und Sommerfeld, Berl. klin. Woch., 1900. * Marmorek, Berl. klin. Woch., 1902. ^ Marmorek, Berl. klin. Woch., xiv, 1902. ® Loc. cit. ^ Marmorek, Ann. de I'inst. Pasteur, 1895. * Marmorek, Ann. de I'inst. Pasteur, 1895. 9 Schottmuller, Miinch. med. Woch., 1903. STREPTOCOCCUS PYOGENES 345 experimental enhancement or reduction of the virulence. Streptococcus hemolysins may be conveniently observed by cultivation of the organ- isms upon blood-agar plates. They may be produced in alkaline pepton- broth and obtained separate from the bacteria by filtration — a procedure, however, in which the quantities obtained are never large. Besredka ^ and Schlesinger ^ believe, for this reason, that the hemolytic sub- stances are closely attached to the bacterial bodies. The last-named author, furthermore, has deter- mined that, in contradistinction to the other toxic substances, strepto- coccus hemolysins are extremely labile, disappearing from culture fluids after standing for from five to seven days at ordinary room temperature. Immunization. — For reasons not wholly understood at present, re- covery from streptococcus infection does not to any marked degree produce immunity against these bacteria. Active immunity may, however, be produced in rabbits, goats, horses, and other domestic animals by treatment with gradually increasing doses of streptococcus cultures.^ In carrying out such immunizations it is necessary to use for the first injection attenuated or dead bacteria. Attenuation may be accom- plished by moderate heating or by the addition of chemicals (terchloride of iodin). Neufeld "^ advises, for the first injection in immunizing rabbits, the use of ascitic-broth cultures killed by heating to 70° C. This is followed, after ten days, by a second injection of a small quantity of fully virulent cocci. Following this, injections are made at intervals of ten days with constantly increasing doses. Modifications of, these general principles are employed in most laboratories. The sera of animals so treated contain no demonstrable antitoxic or Fig. 74. — Streptococcus Colonies FROM Blood Culture on Blood- Agar Plate. Showing areas of hemol- ysis about colonies. ^Besredka, Ann. de Tinst. Pasteur, xv, 1901, p. 880. 2 Schlesinger, Zeit. f. Hyg., xxiv, 1903. 3 Koch und Petruschky, Zeit. f . Hyg., xxiii, 1896. * Neufeld, Zeit. f. Hyg., xliv, 1903. 23 S46 PATHOGENIC MICROORGANISMS antihemolytic substances.^ They exert, however, demonstrable bacteri- cidal power both in vivo and in vitro and distinctly enhance phagocytosis when brought into contact with leucocytes and streptococci. This "opsonic" power has been noticed both intraperitoneally ^ and in vitro. ^ The protective value of streptococcus immune sera for infected animals is considerable, reaching often a potency hardly explicable by the demonstrable bactericidal or opsonic power, and thereby suggesting some other active factor not understood as yet.* Aronson ^ has produced immune sera by the treatment of horses with a streptococcus derived from a ease of scarlatina, 0.0004 c.c. of which sufficed to protect mice from ten times the fatal dose of a streptococcus culture. These high protective values, however, are obtained only when the serum injections are given simultaneously with the bacteria. Given four or six hours after infection, much higher dosage must be employed and protective results are much less regular in occurrence.^ Other antistreptococcic sera have been produced by Denys, Menger, Tavel, and others, all show- ing more or less marked potency in protecting animals.^ Since these sera, while in a general way potent against all streptococci, have been found protective chiefly against the specific microorganism em- ployed for their production. Van de Velde,^ Denys, Aronson, and others have advised the immunization of the animal with a large variety of streptococcus races, derived from many different human sources. The resulting "polyvalent" serum is more apt to exert equally high protective powers against all streptococcus infections. The therapeutic value of such sera in the treatment of human infections is still suh jiidice. Un- deniably favorable reports are published each year in increasing number, but are by no means regular or comparable to the results obtained in diphtheria with diphtheria antitoxin. Nevertheless, in mild cases or in those in which the lesions have been distinctly localized, the sera seemed to be sufficiently useful to justify their use and necessitate their stand- ardization. 1 Ldngelsheim, Zeit. f. Hyg., x, 1891. ^ Bordet, Ann. de I'lnst. Pasteur, 1897. 2 Denys et Leclef, Cellule, t. ix. ^ Benys et Marchand, "Mecanisme de Timmunit^," etc., Brussels, 1896. ^ Aronson, Berl. klin. Woch., xxxii, 1896; ibid., xlii and xliii, 1902; ibid., viii and ix, 1905. ^ Denys, "Le S6rum antistreptoc," Louvain, 1896; Van de Velde, Ann. de I'lnst. Pasteur, 1896. 7 Denys et Marchand, Bull, de I'acad. roy. de m^d. de Belgique, 1898; Menger, Berl. klin. Woch., 1902; Tavel, Corr.-Bl. f. Schw. Aerzte. 8 Van de Velde, Arch, de m€d. exp^r., 1897. STREPTOCOCCUS PYOGENES 347 Standardization is accomplished by the methods first devised by Marx ^ for the standardization of swine-plague serum, and depends upon the ability of the serum to protect animals against a measured dose of virulent streptococci. Aronson^ designates as a "normal serum" one of which 0.01 c.c. will protect a mouse against ten to one hundred times the fatal dose of virulent streptococci. One c.c. of this serum equals one serum unit. Comparisons by animal experiment with this standard serum approximately determine the value of other sera. Leucocyte extracts ^ have been employed in various forms of strep- tococcus infections of man, with success in many cases. Favorable re- sults have been obtained with these extracts in cases of erysipelas. The agglutinins found in streptococcus immune sera are usually most active toward the race of bacteria employed in the immunization. Other streptococci are also agglutinated, but in relatively higher concentra- tion. While a specific group reaction is useful in differentiating strep- tococci from other species, agglutination can not be relied upon to differentiate individual streptococci from one another (Hiss). It has been found that a serum produced with a streptococcus from one source contained a higher agglutinating value for some other streptococcus than for the one employed in its production. Agglutinins may be pro- duced by treating animals with dead as well as with the living strep- tococci. While the technique of streptococcus agglutination is not diffi- cult when we are dealing with strains which grow with even clouding in fluid media, the frequent spontaneous clumping in broth cultures necessitates the use of a special technique. The most simple of these methods is the one in which calcium-carbonate-glucose broth is used for cultivation.^ Growing in this medium and thoroughly shaken once a day, the streptococci are found evenly divided in the supernatant fluid after the settling out of the calcium-carbonate powder. Precipitins have been found by Aronson ^ in streptococcus immune horse serum. Classification. — Differences in minor cultural . characteristics and in virulence of streptococci obtained from various sources have given rise to discussion as to the identity of all races of streptococci. The earliest observers were forced to abandon their separation of the streptococci of ^ Marx, Deutsche thierarzt. Woch., vi, 1901. 2 Aronson, Berl. klin. Woch.. xliii, 1902; Otio, Arb. a. d. konigl. Inst., etc., Frank- furt a. M., Heft 2, 190G. 3 Hiss, Jour. Med. Res., xix, 1908. * Hiss, Jour. Exp. Med., vii, 1905. ^ Aronson, Deut. med. Woch., 25, 1903. 348 PATHOGENIC MICROORGANISMS erysipelas from other streptococci because of the work of Marbaix * and others, who produced erysipelas in rabbits with streptococci from non- erysipelatous lesions, after enhancement of their virulence. V. Lingel- sheim 2 proposed a purely morphological differentiation of ^'longus" and "brevis"; the former class including the streptococci usually found in pyogenic lesions with tendency to form chains of six or more links, the latter designating the short-chained varieties, including the less virulent streptococci. This classification, however, is not tenable be- cause of the dependence of chain formation upon reaction, consistency, and nutritive qualities of the media employed for cultivation, and upon the influence of animal fluids if the microorganisms are taken direct from lesions. Schottmiiller,^ in 1903, proposed a classification based both upon morphology and the appearance of cultures upon human blood agar. By this method he divided streptococci into two main groups as follows: I. Streptococcus longus seu erysipelatos, consist- ing of the most virulent varieties, with tendency to form long chains, and producing hemolysis upon blood media. II. Streptococcus mitior seu viridans, including less virulent strains, with usually shorter chain- formation, and producing green, non-hemolyzing colonies upon blood media. A third group. Streptococcus mucosus, will'receive special con- sideration in a separate section, and is probably more closely related to the pneumococci. The "viridans" type is of great importance med- ically since it is so commonly found in subacute septic endocar- ditis and has recently been associated by Rosenow and others with rheumatism. Attempts to separate the streptococci into subdivisions by their powers to ferment various carbohydrates have been made by Hiss, Gordon, and others. These attempts have, so far, been without practical result. Hiss ^ indicated a tentative division of streptococci into those which fermented monosaccharids alone, those which were also able to ferment disaccharids, and those in which the fermentative powers were extended to the polys^ccharids, starch, dextrin, and glycogen. Gordon ^ found ten different fermentation reactions among twenty pyogenic streptococci examined, and forty-eight different fermentation reactions among two hundred streptococci isolated from saliva. Other * Marbaix, loc. cit. ^v. Lingelsheim, "Aetiol. u. Therap. d. Streptokok. Krankh.," etc., Berlin, 1899. ^ Schottmuller, Munch, med. Woch., 1903. *Hiss, Cent. f. Bakt., xxxi, 1902; Jour. Exp. Med., vi, 1905. 6 Gordon, Annual Report, Local Govern. Board, 33, London, 1903. STREPTOCOCCUS PYOGENES 349 work by Andrewes and Horder and by Buerger ^ confirms the irregu- larity of the fermentation reactions within this group. Andrewes and Horder suggest the following classification: (1) Streptococcus pyogenes. Grows in long chains and ferments lactose, saccharose, and salicin; does not coagulate milk. Streptococci which cause suppurative lesions or severe systemic infections belong to this group. (2) Streptococcus mitis. A saprophytic type found frequently in the mouth which shows the same cultural characteristics as the streptococcus pyogenes, but grows in short chains. (3) Streptococcus anginosus. Found frequently in throats of scarlet-fever patients which differs from the pyogenes only in coagulating milk. (4) Streptococcus salivarius. A short-chain type which ferments lactose, saccharose, and raffinose, and coagulates milk. Streptococci of this type are found frequently in the mouth, but are rarely pathogenic. (5) Streptococcus fecalis. A short-chain type which ferments lactose, sac- charose, and mannite. This type is foimd normally in the intestine, and is occasionally pathogenic. (6) Streptococcus equinus. A short-chain type which does not ferment lac- tose. Found in horse dung and never pathogenic. Quantitative determinations of the amount of acid formed in vari- ous sugars by different races have also been made by Winslow and Palmer ^ and others, but have led to no satisfactory classification. Studies by Hopkins and Lang seem to show that the streptococci found in most human infections may be differentiated from the ordinary saprophytic types by the fact that they ferment lactose and salicin, but fail to ferment raffinose, inulin, or mannite. According to their re- sults, the usual saprophytic types found in the mouth either fail to fer- ment saUcin or ferment raffinose or inulin, whereas the usual fecal types ferment mannite. They also found in infection mannite fermenters which were apparently of fecal origin. Streptococci which gave the same fermentative reaction as the mouth saprophytes were, however, frequently found in malignant endocarditis. Probably the most reliable method of determining the interrelation- ships existing between, bacteria, not only within this group, but in all bacterial classes, is that depending upon their reactions to immune sera. The work of Aronson,^ Marmorek,^ and others has shown that streptococcus immune sera produced with any one race of pyogenic 1 Andrewes and Harder, Lancet, 1906; Buerger, Jour. Exp. Med., ix, 1907. 2 Jour, of Inf. Dis., No. viii, 1910, 1. 3 Ar(ms(m, Berl. klin. Woch., 1902; ibid., 1903. * Marmarek, Berl. klin. Woch., 1902. 350 PATHOGENIC MICROORGANISMS streptococci exerted considerable, though variable, protective action against many other strains of streptococci. The same authors, as well as many others, working with the agglutination reaction, have shown that the agglutinins produced with one streptococcus strain were active against many other streptococci. While most active usually against the particular microorganism with which they were produced, this was by no means the rule, a serum produced with a streptococcus from a case of sepsis, in one case, agglutinating a streptococcus from a case of scar- latina more highly than its own microorganism. As with other "group agglutinations,'^ the more highly immune the serum is, the more gen- eral is the agglutinating power over the whole group. Thus, while agglutination is practically useless in separating streptococci from one another, it is highly useful in differentiating these organisms from allied groups, such as the pneumococci. The immune reactions, therefore, seem to indicate a very close relationship between streptocococci as a class. Pneumococcus (streptococcus) mucosus. — First definitely de- scribed by Howard and Perkins ^ in 1901, and subsequently carefully studied by Schottmuller,^ who isolated it from cases of parametritis, peritonitis, meningitis, and phlebitis. It has since been described by many as the incitant of lobar pneumonia and of a variety of other le- sions and as an apparently harmless inhabitant of the normaL mouth. Morphologically, though showing a marked tendency to form chains, on solid media it often appears in the diplococcus form. It is enclosed in an extensive capsule, which appears with much regularity and persist- ence. Though very similar in appearance, therefore, to pneumococci, these bacteria do not appear in the typical lancet shape. Upon solid media they show a tendency to grow in transparent moist masses. The regularity with which this microorganism ferments inulin medium, make it probable that it is more accurate to place it with the group of pneumococci than with that of streptococci.^ Most of the organisms of this group show the common characteris- tics of the pneumococci and are soluble in bile. Occasional strains, such as one studied by Dochez and Gillespie, neither ferment inulin nor are bile soluble. Rarely, too, does it cause hemolysis. From the various studies carried out upon this group it must be concluded that while perfectly distinct in its formation of a heavy mucoid colony and capsulation, this group is more closely related to the pneumococci than to ^ Howard and Perkins, Jour. Med. Res., 1901, N. S., i. 2 Schottmuller, Munch, med. Woch., xxi, 1903. 3 Hiss, Jour. Exp. Med., 1905; Buerger, Cent. f. Bakt., I, xli, 1906. STREPTOCOCCUS MUCOSUS 351 the true streptococci. As would be expected from its capsulation, its virulence is very powerful and serum reactions are not easily carried out. A further discussion, of the immune serum reactions with this organism is included in the chapter on pneumococci, page 363. STREPTOCOCCI AND RHEUMATISM In 1910 Poynton and Paine ^ described a diplococcus which they obtained from eight cases of acute rheumatic fever and with which they were able to produce lesions in rabbits which they considered typical of rheumatism. The organism was recovered from the blood, the pericardial fluid, or the tonsil of their patients. They described a minute Gram-negative diplococcus growing best in acid media and under anaerobic conditions, but capable of growth on the surface of ordi- nary media. Many investigators have attempted to confirm their work, but with negative results for the most part, though some have found streptococci and diplococci from rheumatic lesions. Recently Rosenow ^ has reported the isolation of a streptococcus from the joints of seven cases of articular rheumatism. He was also able to produce non- suppurative arthritis, endocarditis, and pericarditis in rabbits with these cultures. He describes them as intermediate in character between the streptococcus viridans and streptococcus hemolyticus. More recently Rosenow^ has reported the production of gastric ulcers in rabbits and dogs with streptococci of a certain grade of viru- lence. He has also obtained streptococci from human peptic ulcers which showed a remarkable ''affinity" for the gastric mucous membranes of experimental animals. Rosenow seems inclined to think on his later work that individual strains of streptococcus may acquire predilections for definite tissues, consequently causing rheumatic lesions, muscular lesions, lesions in the gastric mucosa, etc. He bases this opinion both on experimental work and upon observations in patients. Judgment upon this point of view must for the present be held in abeyance, yet it seems to us more likely that specific localization of streptococci, especially in connection with rheumatism, may find an explanation in the hypersusceptibility of in- dividual tissues to the organism, such as that indicated in the work of Faber who sensitized joints with extracts of streptococci, subse- quently producing lesions in these joints by intravenous injections of the microorganisms themselves. 1 Poynton and Paine, Lancet, 1900, ii, 861, 932. 2 Rosenow, Jour. A. M. A., 1913, Ix, 1223. 3 Rosenow, Jour. A. M. A., 1913, Ixi, 1947, 2007. CHAPTER XXIII DIPLOCOCCUS PNEUMONIiE (Pneumococcus, Diplococcus lanceolatus) The opinion that lobar pneumonia is an infectious disease was held by many far-sighted clinicians long before the actual bacteriological facts had been ascertained. This idea, so well founded upon the nature of the clinical course of the disease, with its violent onset and equally rapid defervescence, led many of the earlier bacteriologists to make it the subject of their investigations — a subject made doubly difficult by the abundant bacterial flora found normally in the upper respiratory pas- sages, and by the fact, which is now recognized, that lobar and other pneumonias are by no means always caused by one and the same micro- organisms. Cocci of various descriptions and cultural characteristics were isolated from pneumonia cases by Klebs,^ Koch,^ Giinther,^ Talamon,^ and many others, which, however, owing to the insufficient differential methods at the command of these investigators, can not positively be identified with the microorganism now known to us as Diplococcus pneumoniae or the pneumococcus. Although thus unsuccessful as to their initial object, these early investigations were by no means futile, in that they gave valuable information regarding the manifold bacterial factors involved in acute pulmonary disease and incidentally led to the dis- covery by Friedlander ^ of B. mucosus capsulatus. Communications upon lance-shaped cocci found in saliva, and capable of producing septicemia in rabbits, were published almost simul- taneously by Sternberg ^ and by Pasteur ^ in 1880. These workers ^Klehs, Arch. f. exp. Path., 1873. ^Koch, Mitt. a. d. kais. Gesundheitsamt, Bd. 1. ^Gunther, Deut. med. Woch., 1882. •» Talamon, Progr. mM., 1883. ^Friedlander, Virchow's Arch., Ixxxvii. « Sternberg, Nat. Board of Health Bull., 1881. » Pasteur, Bull, de Tacad. de med., 1881. 352 DIPLOCOCCUS PNEUMONIA 353 beyond reasonable doubt were dealing with the true pneumococcus, but did not in any way associate the microorganisms they described with lobar pneumonia. The solution of this problem was reserved for the labors of A. Frankel ^ and Weichselbaum ^ who published their results, independently of each other, in 1886, demonstrating beyond question that the pneumococcus is the etiological factor in a large majority of cases of lobar pneumonia. Morphology and Staining. — The morphology of the pneumococcus is, in general, one of the most valuable guides to its identity. When typical, the pneumococcus is a rather large, lancet-shaped coc- cus, occurring in pairs, and surrounded by a definite and often wide capsule, which usually includes the two approximated cocci without a definite indentation opposite their lines of division. The pneumococci may, however, occur singly or in short chains, and even fairly long chains are not infrequently met with under artificial cultural conditions. This may be chiefly due to the cultural conditions or may be a promi- nent characteristic of certain strains. Apparently the capsules of or- ganisms making up the chains are continuous; wavy indentations are usually present, however, in the capsule of chains, and at times distinct divisions are observed. The chief variations from the typical morphology consist either in the assumption of a more distinctly spherical coccus type, or in an elongation approximating the bacillary form. Under certain conditions of artificial cultivation a distinct flattening of the organisms, particularly of those making up chains, may be seen, and even the impression of a longitudinal line of division, characteristic of many streptococcus cultures, is not infrequently gained. The capsules under certain conditions, especially in artificial media, may be absent or not demonstrable, and in certain strains capsules ap- parently may not be present under any conditions. Practically any of the described variations may dominate one and the same culture under different or even apparently the same conditions of cultivation, and all grades may occur in capsule development, from its typical formation through all variations, to its total and apparently permanent absence. The presence or absence of capsules depends, to a large extent, upon the previous environment of the pneumococci under observation. The most favorable conditions for the development or preservation of the pneumococcus capsule are found in the body fluids of man and animals 1 A. Frankel, Zeit. f. klin. Med., x, 1886. 2 Weichselbaum, Med. Jahrbticher, Wien, 1886. 354 PATHOGENIC MICROORGANISMS suffering from pneumococcus infection. For instance, capsules may be demonstrated with ease by the usual capsule-staining methods in the blood, serum, and inflammatory exudate of the infected rabbit and white mouse. Capsules may be equally well marked in the fresh sputum of pneumonia patients, especially in the early stages of the disease and in the exudate accompanying such pneumococcus infections as menin- gitis, otitis media, and empyema. In sputum and the* exudates of various localized infections, the organisms are, however, frequently degenerated or under chemical conditions unfavorable for capsule staining, and satisfactory results are not then easily obtained. The Fig. 75. — Pneumococci, Grown on Fig. 7G. — Pneumococct, from Rab- Loeffler's Serum. (Capsule stain bit's Heart Blood. (Capsule stain by by gentian-violet-potassium-carbonate copper-sulphate method.) method.) same is often true of the scrapings from lungs of patients dead of pneumonia, even in the stage of red hepatization. In artificial cultivation, if the nutrient medium is not milk or does not contain serum, capsules can not usually be demonstrated by the ordinary methods of preparing and staining. Capsules may, however, with much regularity be demonstrated on pneumococci, in agar, broth, or on almost all, if not all, artificial media, irrespective of the length of time the organ- isms have been under artificial cultivation if beef or rabbit serum is used as the diluent, when they are spread on the cover-glass for staining.^ The pneumococcus is non-motile and possesses no flagella. Spores are not formed. Swollen and irregular involution forms are common in cultures more than a day old. » Hiss, Cent. f. Bakt., xxxi, 1902; Jour. Exp. Med., vi, 1905. DIPLOCOCCUS PNEUMONIiE 355 The pneumococcus is stained readily with all the usual aqueous anilin dyes. Stained by the method of Gram, it is not decolorized. Special methods of staining have been devised for demonstra- tion of the capsule. The ones most generally used are the glacial acetic-acid method of Welch ^ and the copper-sulphate method of Hiss.^ More recently Buerger^ has devised a more complicated method for staining capsules, for which he claims differential value. (For methods see section on Technique, p. 98.) For simple staining of pneumococci in tissue sections, the Gram- Weigert technique is excellent. For demonstration of the capsules in tissue sections, Wadsworth * has described a simple method. Cultivation and Isolation. — The pneumococcus being more strictly parasitic than many other bacteria, presents greater difficulties in its cultivation. On meat-extract media growth does not take place with regularity. On those media, however, which have beef or veal infusion for their basis, growth can be obtained with considerable regularity, although such growth may be sparse and delicate. Growth takes place most regularly at a temperature of 37.5° C. Development does not usually occur below 25° nor above 41° C.^ At ordinary room temperature, 18-22° C, the temperature used for gelatin cultivation, growth either does not take place at all or is exceedingly slow and unenergetic. Aerobic and anaerobic conditions both permit the growth of pneumococcus, there being very little difference in speed or extent of growth along the course of deep stab cultures in favorable media. The most favorable reaction of media for the cultivation of this micro- organism is neutrality or very slight alkahnity. Slight acidity, how- ever, if not exceeding eight-tenths per cent, does not materially hamper development. The growth of pneumococci on all media may be considerably enhanced by the addition to these media of animal or human serum or whole blood. Additional substances which, among others, unquestion- ably have a favorable influence upon pneumococcus growth, are glucose, nutrose, and glycerin. The addition of the latter substances to the media, however, probably because of acid formation, hastens the death 1 Welch, Johns Hopk. Hosp. Bull, xiii, 1892. « Hiss, Cent. f. Bakt., xxxi, 1902; Jour. Exp. Med., vi, 1905. ^Buerger, Medical News, Ixxxviii, 1904. * Wadsworth, " Studies by the Pupils of W. T. Sedgwick," Chicago, 1896, •A. Frankel, Dent. med. Woch., xiii, 1886. 356 PATHOGENIC MICROORGANISMS of pneumococcus cultures. An increase of the amount of pepton used for the preparation of media is desirable for the cultivation of this microorganism; two to four per cent of pepton may be found advantageous. In suitably alkaline, nutrient broth, growth is rapid, and within twenty^four hours leads to slight clouding of the fluid. This clouding, as a rule, eventually disappears as the microorganisms, sinking to the bottom of the tube or disintegrating, leave the fluid more or less clear. In broth, pneumococci have a tendency to form short chains. When glucose has been added to the broth, growth is more rapid and profuse, but considerable acid formation causes the cultures to die out rapidly. It is possible, however, to employ glucose as a growth-enhancing element in broth cultures without interfering with the viability of the cultures by adding small quantities (one per cent) of sterile, powdered calcium carbonate. This method of cultivation in broth is especially adapted to the production of mass cultures for purposes of immunization or agglutination.^ The addition of ascitic fluid or blood serum to broth, in the proportion of one to three, makes an extremely favorable medium in which growth is rapid and profuse. Upon agar plates, pneumococcus growth is not unlike that of strepto- coccus. The colonies are small, round, and slightly , more transparent than those of the streptococci. They appear more moist than strepto- coccus colonies and often are more flat. Microscopically examined, the colonies are finely granular, with dark centers and slightly corrugated lighter-colored peripheral areas. Under high magnification no such in- tertwining convolutions can be seen as those noticed under similar magnification in streptococcus cultures. The addition of animal albu- min to agar results in the more rapid development, larger size, and deeper opacity of the colonies. Agar stab cultures show growth within twenty-four to thirty-six hours, which takes place with equal thickness along the entire course of the stab. There is nothing distinctive in these cultures to differentiate them from similar streptococcus cultures. In gelatin plate snidstab cultures at 22° C, growth, as a rule, does not take place. This, however, is not true of all races of pneumococci. Occasionally strains are met with which will grow fairly abundantly in gelatin at a temperature of 22° C. When the gelatin is rendered suffi- ciently firm to bear 25° to 26° C. without melting, growth appears Hiss, Jour. Exp. Med., vii, 1905. DIPLOCOCCUS pneumonia; 357 slowly and sparsely as minute, grayish-white, transparent colonies. The gelatin is not liquefied by the organisms. Growth upon milk is profuse, resulting in the production of acid and coagulation of the medium. Races are encountered in which this is suppressed and coagulation in milk is absent or long delayed. Upon potato, thin, gray, moist growth occurs, hardly visible and in- distinguishable from an increased moisture on the surface of the medium. Upon Loeffler's coagulated blood serum, the pneumococcus develops into moist, watery, discrete colonies which tend to disappear by a drying out of the colonies' after some days, differing in this from streptococcus colonies, which, though also discrete, are usually more opaque and whiter in appearance than those of the pneumococcus and remain un- changed for a longer time. This medium, as will be seen, is useful in differentiating pnexmiococci from the so-called Streptococcus mucosus. Upon mixtures of whole rabbit's blood and agar, the pneumococcus grows well, and forms, after four or five days, thick black surface colo- nies, not unlike sun blisters on red paint. These colonies are easily distinguished from those of streptococci, and are of considerable differ- ential value.^ Pneumococcus colonies on blood plates may cause a slight halo of hemolysis and methemaglobin formation after 48 hours or longer in the incubator. Guarnieri ^ has recommended a medium with a pepton-beef-infusion basis rendered semisolid by mixtures of agar and gelatin. A modifica- tion of this medium has been described by Welch ^ and has been much employed. Cultivation within eggs and upon egg media ^ has been used. Wadsworth ^ has recommended a medium composed of ascitic fluid to which agar has been added — sufficient to give a soft jelly-like consist- ency. He observed prolonged viability and the preservation of the virulence on this medium. For the purpose of differentiating pneumococci from streptococci, Hiss ^ devised a medium of beef serum one part, and distilled water two parts, to which is added one per cent of inulin (c. p.), and enough litmus to render the medium a clear, transparent blue. By fermentation of the inulin the pneumococcus acidifies this mixture, causing coagulation of the serum. Streptococci do not ferment inulin. 1 Hiss, loc. cit. 3 Welch, Johns Hopk. Hosp. Bull., iii, 1S92. 2 Gvjamieri, Att. dell' Acad, di Roma, 1883. ■* Sdavo, Riv. d'Igiene, 1894. B Wadsworth, Proc. N. Y. Path. Soc, 1903. • Hissj Jour. Exp. Med., vi, 1905. 358 PATHOGENIC MICROORGANISMS For the isolation of pneumococci from mixed cultures or from mate- rial containing other species, such as sputum, surface smears of the material are made upon plates of neutral glucose-agar, glucose-serum- agar, or blood agar. According to the number of bacteria present in the infected material, it may be smeared directly upon the plate, or diluted with sterile broth before planting. After incubation for twenty- four hours, the pneumococcus colonies are easily differentiated from all but those of streptococcus. With practice, however, they may be dis- tinguished from these also, by their smoother edges and greater trans- parency and flatness. For the isolation of pneumococci from sputum, the sputum may be washed by gently rinsing in successive watch glasses containing salt solution, then rubbed up in a glass tube or mortar with a little broth and injected into a white mouse, either into the base of the tail or into the peritoneum, taking great care not to puncture the liver. If virulent pneumococci are present, death will occur within 24 hours or there- abouts. Pneumococci wilL be found in pure culture in the heart's blood, and in large amounts in the peritoneal exudate. Resistance. — On artificial media, the viability of the pneumococcus is not great. Cultures upon agar or bouillon should be transplanted every third or fourth day, if the cultures are kept within an incubator. In all media in which rapid acid formation takes place, such as glu- cose media, the death of cultures may occur more rapidly. In media containing albumin and of a proper reaction, preservation for one or even two weeks is possible. The longer the particular race has been kept upon artificial media, the more profuse is its .growth, and the greater its viability, both qualities going hand in hand with diminish- ing parasitism. The length of life may be much increased by preserva- tion at low temperature, in the dark, and by the exclusion of air. In caJcium carbonate broth and kept in the ice-chest, cultures may often remain alive for months. Neufeld has succeeded in keeping pneumococci alive and virulent, by taking out the spleens of mice dead of pneumococcus infection and preserving them in a Petri dish in a dessicator, in the dark and cold. In this way, the organisms can be cultivated from the spleen, and will be found virulent for longer periods than in culture media. In sputum the viability of pneumococci seems to exceed that ob- served in culture. The studies of Guarnieri,^ Bordoni-Uffreduzzi,^ and 1 Guamieri, Att. della R. Acad. Med. di Roma, iv, 1888. ^ Bordm^'TIffrediLZzif Arch. p. 1. so. med., xv» 1891. DIPLOCOCCUS PNEUMONIiE 359 others have shown that pneumococci slowly dried in sputum may re- main alive and virulent for 1 to 4 months, when protected from light; and as long as nineteen days when exposed to diffused light at room temperature. Experiments by Ottolenghi ^ have confirmed these re- sults; the virulence seems, in Ottolenghi's experiments, to have become considerably attenuated before death of the cocci. Recent studies by Wood,2 whose attention was focused chiefly upon pneumococcus viability in finely divided sputum — in a condition in which inhalation transmission would be possible — have shown that pneumococci survive for only about one and one-half hours, under ordinary conditions of light and tempera- ture. Exposed to strong sunlight pneumococci die off within an hour. Low temperatures slightly above zero are conducive to the pro- longation of life and the preservation of virulence. The resistance of the pneumococcus to heat is low, 52° C. destroying it in ten minutes.^ To germicidal agents, carbolic acid, bichlorid of mer- cury, permanganate of potassium, etc., the pneumococcus is sensitive, being destroyed by weak solutions after short exposures. The disinfection of sputum, difficult because of the protective coat- ing of the secretions about the bacteria, has been recently studied by Wadsworth."* The conclusions reached by this writer indicate that pneumococci in exudates are most rapidly destroyed by twenty per cent alcohol, other and stronger disinfectants being less efficient, probably because of slighter powers of diffusion. Virulence and Pathogenicity. — The virulence of pneumococci is sub- ject to much variation, depending upon the length of time during which it has been cultivated. It has been mentioned above that under condi- tions such as those prevailing in dried sputum or blood ^ the virulence of pneumococci may be preserved for several weeks. Ordinarily, the virulence diminishes as the cocci adapt themselves to life upon artificial media. Upon media containing animal albumin, such as ascitic fluid or blood agar, this attenuation is less rapid than upon the simple meat- infusion preparations. In the blood of rabbits dead of a pneumococcus infection, taken directly into sterilized tubes, sealed and kept in the dark, Foa ^ has been able to preserve the virulence of pneumococci for as long as forty-five days. Preservation in the spleen of animals dead of pneumococcus infection, as practiced by Neufeld, has been mentioned above. Whether 1 Ottolenghi, Cent. f. Bakt., xxv, 1889. ^ Wood, Jour. Exp. Med., vii, 1905. 3 Sternberg, Cent. f. Bakt., xii, 1891. •* Wadsworth, Jour. Inf. Diseases, iii, 1906. 6 Quarnieri, loc. cit, ^ Foa, Zeit. f . Hyg., iv, 1888. 360 PATHOGENIC MICROORGANISMS or not the virulence of pneumococci is attenuated by sojourn within the human body during disease is uncertain. The attenuation of virulent pneumococci on artificial media may be hastened, according to Frankel/ by cultivation of the organism at or above a temperature of 41° C. The virulence of attenuated cultures may be rapidly enhanced by passage of the organisms through the bodies of susceptible animals. The virulence of strains may be so enhanced that one one-millionth of a c.c. will kill a mouse. Among the domestic animals white mice and rabbits are most sus- ceptible. Guinea-pigs, dogs, rats, and cats are much more resistant. Birds are practically immune. The results of pneumococcus inoculation into susceptible animals vary according to the size of the dose, the virulence of the introduced bacteria, the mode of administration, and the susceptibility of the subject of the inoculation. Subcutaneous inoculation of virulent pneumococci into mice and rabbits usually results in an edematous exudation at the point of inoculation, which leads to septicemia and death within twenty-four to seventy-two or more hours. When the dose has been extremely small or the culture unusually attenuated, a localized abscess may be the only result. Intravenous inoculation is usually more rapidly fatal in these animals than the subcutaneous method. Intraperitoneal inoculation in rabbits results in the formation of a rapidly spreading peritonitis in which the exudates are apt to be accompanied by a deposit of fibrin, and to lack the transparent red color so often caused by the hemolyzing streptococci. With very virulent strains, these differences are less marked. In almost all of these infec- tions death is preceded by septicemia and the microorganisms can be recovered from the heart's blood of the victims. The production in animals of lesions comparable to the lobar pneu- monia of human subjects has been the aim of many investigators. Wadsworth, ^ recognizing that such lesions probably depended upon the partial immunity which enabled the infected subjects to localize the pneumococcus processes in the lungs after infection by way of the respiratory passages, succeeded in producing typical lobar pneumonia in rabbits by partially immunizing these animals and inoculating them intratracheally with pneimiococci of varying virulence. Lamar and Meltzer ^ produced lobar pneumonia in dogs in 1912 by injecting cul- 1 Frdnkel, Deut. med. Woch., 13, 1886. 2 Wadsworth, Amer. Jour. Med. Sci., May, 1904. • 3 Lamar and Meltzer^ Jour. Exp. Med., xv, 1912. DIPLOCOCCUS PNEUMONIA 361 tures in the bronchi and blowing them into the finer bronchioles with air. Similar experiments have been made by Wintemitz and Hirsch- felder.^ In man, the most frequent lesion produced by the pneumococcus is acute lobar pneumonia. About ninety per cent of all cases of this disease are caused by the pneumococcus,^ the remainder being due to streptococci, influenza bacilli, Friedlander's bacilli, and exceptionally to other microorganisms. Lobular pneumonia is caused by the pneu- mococcus with almost equal regularity. Infection in pneumonia prob- ably occurs through the lung, and in such experiences as those of the French and Americans at Panama and those of the English in the South African diamond mines, it would appear that pneumonia might be a contagious disease. According to Cole, ^ when pneumonia is secondary to septicemia it is usually of the lobular type. Experiments of Meltzer seem to indicate that infection is facilitated by closure of the small bronchioles, and cold or chilling may possibly stimulate the mucous glands so as to plug these. In the discussions given below concerning the types of pneumococci, it will appear that the type of pneumococcus found in the lung not often coincides with that found in the mouths of normal individuals. The types I, II and III, according to Dochez and Avery,^ are found only in association with the disease, but the fourth group is the one which is found in sputum of many normal individuals. A person in contact with a pneumococcus patient may possibly carry the more virulent groups in his mouth, thus constituting a pneimiococcus carrier. During the course of these diseases the cocci are found in large numbers witljin the pulmonary alveoli, and in the capillaries and lymph vessels of the lung. Whether or not the pneumococci enter the blood stream in all these cases is a question not yet definitely settled. Frankel ^ states it as his belief that in most, if not in all, cases, the diplococci at some time during the diseases could be found in the circulating blood. Prochaska in a study of ten unselected cases obtained positive blood cultures in every one of them. A review of the literature upon the question indi- cates positive blood-culture findings in certainly over twenty-five per cent of the cases. * Wintemitz and Hirschfelder, Jour. Exp. Med., xvii, 1913. 2 better, Compt. rend, de la soc. de biol., 1890. . 3 Cole, Harvey Lecture, New York, Dec. 13, 1913. * Dochez and Avery, Jour. Exper. Med., xxi, 1915. ^Frankel, "v. Leyden Festschr.," 1902. 24 362 PATHOGENIC MICROORGANISMS In complications of pneumonia, pneumococci are found usually in the pleura where they may cause a simple dry pleurisy or even empy- ema. Less frequently they may cause pericarditis and endocarditis. Meningitis may be either secondary to pneumonia or independent. Such cases are grave, almost invariably ending in death. Other lesions which may be caused by pneumococci, either as post-pneumonic proc- esses or without previous pneumonia, are otitis media, osteomyelitis, and arthritis. Cases of pneumococcus peritonitis occur sometimes secondary to appendicular inflammations, occasionally without traceable portal of entry. Severe catarrhal conjunctivitis may be caused by these diplococci, usually during the course of a pneumonia. Ulcerative endocarditis with pneumococcus septicemia, apparently independent of a pulmonary lesion, is not infrequent. Toxic Products of the Pneumococcus. — Our knowledge of pneumococ- cus poisons is still very imperfect. Attempts to obtain soluble toxins by the filtration of cultures have been practically unsuccessful. G. and F. Klemperer,^ Mennes,^ Pane,^ Foa and Carbone,* and others failed to obtain pneumococcus filtrates of any degree of toxicity, though working- with highly virulent strains. The feeble toxin so obtained produced ii antitoxin. The general failure to procure strong soluble poisons from cultures, gives weight to the a-ssumption that the most potent toxic products of pneumococci are in the nature of endotoxins and closely bound to the cell-bodies themselves. This assumption is borne out by the more recent experiments of Macfadyen.^ This author obtained acutely poisonous substances from pneumococci by trituration of the organisms after freezing, and extracting them with a one 1 : 1,000 caustic potash solution. With the filtrates of these extracts he was able to cause rapid death in rabbits and guinea-pigs by the use of doses not exceeding 0 . 5 to 1 c.c. He found, furthermore, a striking parallelism between the degree of toxicity and the virulence of the extracted culture. Cole,^ too, in recent studies, inclines to the belief that the poisons of the pneu- mococcus are in the nature of endotoxins»and has produced toxic sub- stances by salt solution and bile extraction of the organisms. Immunization. — Recovery from a spontaneous pneumococcus in- ^ G. and F. Klemperer, Berl. klin. Woch., xxxiv and xxxv, 1891. 2 Mennes, Zeit. f . Hyg., xxv, 1897. » p^ne, Rif. med., xxi, 1898. * Foa und Carbone, Cent. f. Bakt., x, 1899. * Macfadyen, Brit. Med. Jour., ii, 1906. 6 Cole, Harvey Lecture, N. Y., Dec, 1913, DIPLOCOCCUS PNEUMONIA 363 f ection confers immunity for only a short period. Two and three attacks of lobar pneumonia in the same individual are not unusual, and it is uncertain whether even a temporary immunity is acquired in such infections. Active immunization of laboratory animals may be carried out by various methods. The method usually followed is to begin by injecting attenuated ^ or dead bacteria or bacterial extracts. Subse- quent injections are then made with gradually increasing doses of living, virulent microorganisms. Great care in increasing the dosage should be exercised since the loss of an animal after two or three weeks' treat- ment by a carelessly high dose of pneumococci is not unusual. Wads- worth centrifugalizes freshly grown pneumococcus cultures and to the pneumococcic sediment adds a definite quantity of concentrated salt solution. At the end of 12 hours, the pneumococci are dead and con- siderable destruction of the cell-bodies has taken place. Dilution with water until the solution equals 0.85 per cent NaCl now prepares the emulsion for inoculation. The sera of animals immunized with pneu- mococci contain active bactericidal substances. Specific agglutinins in pneumococcus immune sera were first thoroughly studied by Neufeld ^ and since then have been made the subject of extensive studies by Wadsworth,^ Hiss,^ and many others. For the sake of obtaining plentiful growth for agglutination purposes, Hiss has recommended cultivation in 1% glucose broth with the addi- tion of small amounts of sterile calcium carbonate to absorb acid formed from the glucose. Pneumococci do not regularly agglutinate in diluted immune sera and agglutinations are best studied in suspensions of more concentrated immune serum. Agglutination begins at the end of about 15 minutes, and can be studied both by formation of clumps and by the sediment. An entirely new turn has been given to studies of the pneumococcus group by the observation by Neufeld and Haendel that as regards reac- tions to immune serum several varieties of pneumococci exist. Studies based upon this observation of Neufeld ^ and his associates have more recently been made by Dochez and Gillespie.^ These workers have studied pneumococci isolated from many different human sources, both by means of protection in mice and by agglutination. They have accord- ingly divided pneumococci into four groups, in which they include, for 1 Radziewsky, Zeit. f. Hyg., xxxvii, 1901; Neufeld, Zeit. f. Hyg., xi, 1902. 2 Neufeld, loc. cit. ^ Wadsworth, loc. cit. * Hiss, Jour. Exp. Med., vii, 1905. ^ Neufeld and Haendel, Arb. a. d. Kais. Gesund., 1910, xxxiv. 293. 8 Dochez and Gillesvie, Jour. A. M. A., 1913, Ixj. 727 364 PATHOGENIC MICROORGANISMS reasons mentioned in another place, the streptococcus (seu pnenmococ- cus) mucosus. Both the methods of protection and agglutination per- mitted such a grouping. Groups I and II are made up of organisms that are respectively identical, since the members of each group react, in protection and the agglutination experiments, with sera produced by individuals of the same group. Group III is distinctive and repre- sents the pneumococcus mucosus, but immune reactions with members of this group are very difficult. Group IV is a heterogeneous collection of organisms which have no distinctive characteristics, and each individ- ual member seems serologically isolated. Thus, the only groups in which exact serological experiments can be made are those now classi- fied, as I and 11.^ In 223 cases of lobar pneumonia Dochez finds the various types dis- tributed as follows: Types of Number Per eumococci of Cases Cent. I 78 34.97 II 75 33.63 III 22 9.86 IV 48 21.52 Total number of cases 223 Groups I, II and III seem to be most frequently associated with disr ease, whereas members of Group IV are often found in the mouths of normal people. Groups I, II and III, moreover, are found in the sputum of normal individuals rarely except in association with the dis- ease, but often in individuals who have been closely associated with cases of pneumonia and in convalescents. Dochez and Avery ^ accord- ingly suggest the possibility of pneumonia carriers and also come to the conclusion that pneumonia is rarely an autogenous infection of the lungs with the organism harbored for a long time in the mouth, but is as a rule an infection with a virulent strain of pneumococcus from without, carriers furnishing an important element in the transmission from pa- tients to the healthy. Such contagiousness is also suggested by the apparent epidemics at Panama and in Africa, mentioned in a preceding paragraph. Precipitins have been demonstrated in pneumococcus immune serum 1 Recently Miss Olmstead, in our laboratory, has obtained results which promise the eventual subdivision of Group IV into definite divisions. ^Dochez and Avery, Jour, of Exper. Med., 1915^ xxi, 114. DIPLOCOCCUS PNEUMONIA 365 by Neufeld, ^ Wadsworth,^ Hiss,^ and others, the organism for such tests being brought into solution either with bile or with concentrated salt solution. Such sera also contain powerful opsonic, substances, or, as Neufeld and Rimpau ^ prefer to call them, '' bacteriotropins. " It seems most likely that such phagocytosis-aiding substances are most power- fully concerned in protection and cure. Clough ^ has reported an increase of opsonins at the time of crisis, and Dochez ^ has shown that protective substances may appear in the serum at or soon after the time of crisis. The outcome of a case according to Cole depends very largely on the virulence of the organism and on the ability of the body first to limit the local infection and to prevent the invasion of the blood with the organisms. In this process, of course, the protective and opsonic bacteriotropic substances would play a most important part. The history of attempts to produce sera for passive immunization in man is extensive. Washburn,^ Mennes,^ Pane^ and many others in the past have succeeded in protecting animals with such sera, but with irregular results. The rational beginning based on the recognition of different pneumococcus types was made by Neufeld and Haendel in Germany, and carried to a considerable degree of success by Cole and his associates at the Rockefeller Hospital in New York. By the immun- ization of horses with the various types of pneumococci mentioned above, considerable success has attended the use of sera produced with Type I, and less success but great promise, that of sera produced with Type II. The injection of considerable quantities of the homolog- ous sera intravenously at least aids in sterilizing the blood stream, and upon this the eventual outcome of many cases may depend. Since these methods which at present seem to promise the establish- ment of a definite specific serum therapy in this serious disease depend upon a rapid identification of the type with which the patient is infected, it becomes a part of the work of the bacteriologist to control the methods which have been developed for such determinations. The methods in use at the present writing depend mainly upon the agglutination of the strains in homologous sera. 1 Neufeld, Zeit. f. Hyg., 1902, xi.. « Wadsw(yrth, loc. cit. ^ Hiss, Jour. Exp. Med,, vii, 1905. * Neufeld and Rimpau, Deut. med. Woch., 1904. 6 Clough, Johns Hopkins Hosp. Bull., Oct., 1913. « Dochez, Jour. Exp. Med., 1913. ' Washburn, Brit. Med. Jour., 1897. « Mennes, Zeit. f. Hyg., 1897. 9 Pane, Rif . med., 1897. 366 PATHOGENIC MICROORGANISMS The patient is required to wash out his mouth with some antiseptic, and then expectorate sputum obtained by coughing, into a Petri dish. The sputum is washed by Ufting it through a series of watch glasses containing sterile salt solution and is then ground in a mortar with a little salt solution, and the even emulsion so obtained is injected intra- peritoneally into a white mouse. A fine needle is carefully introduced upward and inward from just above Poupart's ligament. Puncture of the liver may be avoided in this way. The pneumococci develop in the peritoneum, and after 4 to 8 hours the mouse is killed and the peri- toneal cavity carefully washed with salt solution by means of a capillary pipette. The washings are placed in a centrifuge tube at low speed for several minutes to bring down leucocytes. The supernatant fluid is then centrifugalized at high speed and the pneumococci so obtained are resuspended in salt solution. This suspension is then added to equal parts of the sera of Types I and II in dilutions of 1 : 2 . 5 and 1:10, with, of course, proper controls and a bile test to determine bile solubility. Agglutinations can also be made later with broth cultures made from the mouse's heart's blood or peritoneal exudate. Protection experiments for similar determinations can be carried out by the method shown in the following protocol taken from the work of Dochez and Avery: 67 (Group I) A69 (Group II) Dose of Culture 1 Con- Serum Serum Con- Serum Serum trols I II trols I II 0.1 c.c. Dead Sur- Dead Dead Dead 17 hrs. vived 41 hrs. 18 hrs. 18 hrs. 0.01 17 hrs. cc 25 hrs. 18 hrs. Sur- vived 0.001 41 hrs. tc 41 hrs. 18 hrs. it 0.0001 41 hrs. tc 41 hrs. Dead 18 hrs. It 0.00001 96 hrs. ct 41 hrs. 18 hrs. u 0.000001 48 hrs. l( 72 hrs. 18 hrs. tt (The quantities in c.c. representing dose of culture refer to young broth cultures.) DIPLOCOCCUS PNEUMONIA 367 No headway has been made up to the present time with similar treatment against Types III and IV. Occasional favorable results have been obtained by Hiss in the treat- ment of pneumococcus infection in animals, and by Hiss and Zinsser* in the treatment of pneumonia in man with aqueous leucocyte extracts. Although this work has not been extensively followed and has been dis- appointing in late years, there is no doubt in the mind of the writer that individual cases have been favorably influenced. The writer be- lieves (differing from an opinion expressed in earlier papers) that the influence of such leucocyte extracts (as, it might be incidentally stated, he believes the influence of intravenous injections of bacteria in typhoid and other conditions) consists in its chemotactic influence, i.e., its ten- dency to increase circulating leucocytes. The contrast between the good results in rabbits and the less striking results in man may easily be due to the fact that in rabbits intravenous injections were practiced. Differentiation of Pneumococcus from Streptococcus. — Pneumococci and streptococci which do not differ in morphology from their classic types can usually be differentiated from each other and identified by their morphological characters without difficulty; but 'it is equally true that certain cultures of these organisms, either at the time of their isolation or after cultivation on artificial media, approach the type of the other so closely that it may be impossible to identify them by their mor- phology alone. When such morphological variations occur there are no constant cultural or pathogenic characters as yet demonstrated which distinguish between these organisms. This lack of distinct cultural differences between pneumococci and streptococci has not infrequently led to confusion, and that uncertainty should exist and mistakes be made in identification is not surprising when one considers the characters usually depended upon to distinguish pneumococci from streptococci. Chief among these, as has just been implied, are the morphological features which are, in the case of pneu- mococci, a slightly lancet or elongated form rather than the more typical coccus form characteristic of the streptococci, and an arrangement of such cocci in pairs rather than in chains; added to these features is the possession of a more or less well-defined capsule. All of these char- acters are subject to variation or may be absent. Compared with the morphological, the cultural characters are variable and of minor im- portance. The pneumococcus colonies on coagulated blood serum and * Hiss and Zinsser, Jour. Med. Res., xix, 1908. 368 PATHOGENIC MICROORGANISMS on agar are moister and flatter, and the freshly isolated pneumococcus is usually unable to develop readily or at all on gelatin at below 22° C. The distinctness of the capsule of the pneumococcus in the body fluids of man and animals and on blood serum, milk, or serum agar, has been depended upon as the chief distinguishing and diagnostic character. Nevertheless, instances have been reported of distinct cap- sule formation by organisms which had either been previously iden- tified as Streptococcus pyogenes, or at the time of their isolatoin could not be definitely identified as belonging to this group or to the pneu- mococci, but were considered intermediate in character.^ 1 Brief Description of Organisms Reported as Capsulated Streptococci. — Bordet {Bordet, Ann. de I'inst. Pasteur, 1897, xi, p. 177), working with an organism previously- identified as Streptococcus pyogenes, described such capsule formation occurring in the peritoneal exudate of infected rabbits. Schuetz' (Schuetz, Cent. f. Bakt., Ref. 1, 1887, p. 393) Diplokokkus der Brust- seuche der Pferde, Poels and Nolen's {Poels und Nolen, Fort. d. Med., iv, 1886, p. 217) streptococcus of contagious pneumonia of cattle, and especially the organism de- scribed by Bonome (Bonome, Ziegler's Beit., viii, 1890, p. 377) as Streptococcus der meningitis cerebrospinalis epidemica, may all be looked upon as organisms differentiated on insecure grounds from either pneumococcus or streptococcus. The first two of these organisms, however, are said to be decolorized by Gram's method, and as suggested by Frosch and KoUe {Frosch und Kolle, Fliigge's " Mikro-organis.," ii, 1896, p. 161), in the case of Schuetz' organism may belong to a group inter- mediate between Fraenkel's diplococcus and the chicken-cholera group. Tavel and Krumbein (Tavel und Krumhein, Cent. f. Bakt., xviii, 1895, p. 547) describe a streptococcus with a capsule, which was isolated from a small abscess on the finger of a child. Capsules were also present in the artificial cultures, and al- though ordinarily remaining uncolored, could be stained by Loeffler's flagella stain. This organism was said to be differentiated from Fraenkel's diplococcus and also in general from streptococcus (pyogenes) by a rapid and rich growth on gelatin, agar, and potato. A pellicle was formed on broth. The organisms forming this pellicle had capsules, but those in the deeper portions of the broth generally lacked it. In 1897, Binaghi (Binaghi, Cent. f. Bakt., xxii, 1897, p. 273) described a cap- sulated streptococcus isolated from a guinea-pig dead of a spontaneous peribron- chitis and multiple pulmonary abscesses. In the pus were found some diplococci and short chains (four to six) surrounded by a capsule, shown by staining with carbol fuchsin. This organism he proposes to call Streptococcus capsulatus. Le Roy des Barres and Weinberg in 1899 (Le Roy des Barres et Weinberg, Arch, d. m6d. exper., xi, 1899, p. 399) published an account of a streptococcus with a capsule. This was isolated from a man who had apparently been infected from a horse which had died of an acute intestinal disorder. The patient neglected the infection and died. Diplococci and short chains furnished with a capsule were found in the subcutaneous tissue at the area of infection. The blood, hver, and spleen also contained these organisms. The capsule in all the preparations remained uncolored, but the authors say that its existence was not to be doubted. Ascitic DIPLOCOCCUS PNEUMONIiE 369 There are occasions, then, both within the animal body and in arti- ficial cultivations, when it is practically impossible to distinguish defi- nitely between some races of pneumococci and races of streptococci. This difficulty is especially heightened when the pneumococcus has become non-virulent, and at the same time no very typical morphology or capsule formation is to be determined and a tendency to chain-forma- tion is marked. Cultures of pneumococci in such condition can not readily be distinguished morphologically from streptococcus cultures. Under these circumstances recourse must be had to a careful bio- logical study of the organism in question. The following are the criteria mainly relied upon at present for the differentiation of these two groups. Pneumococci ferment inulin, if cultivated in inulin-serum-water me- dium. Acid formation from the inulin results within two days or more in coagulation of the serum and reddening of the litmus. Streptococci be- cause of their inability to attack the inulin leave the mediimi unchanged.' broth inoculated from the peritoneal exudate of a rabbit dying from the infection gave streptococci in extremely long chains and surrounded by capsules. These were not so distinct as in the case of the organisms in the original smear preparations. All fluid media (bouillon, milk, and ascitic broth) were said to be strongly acid after twenty-four hours. These authors report that Achard and Marmorek have assured them that they have seen capsulated streptococci, and that Marmorek showed them some preparations in which one of his streptococci presented the same characters as that isolated by them. Although Le Roy des Barres and Weinberg have used the term encapsulated, they beheve that it would perhaps be more prudent to call their organism strepto- coque aureole, since they were not able to define this capsule by staining it. Howard and Perkins {Howard and Perkins, Jour. Med. Res., 1901, iv, p. 163) have lately described an organism, probably of the foregoing type, which was present in a tubo-ovarian abscess and in the peritoneal exudate, the blood, and some of the organs of a woman dying in the Lakeside Hospital, Cleveland, Ohio. The organisms were biscuit-shaped cocci in pairs, usually arranged in chains of four, six, eight, or twenty elements, and surrounded by a wide and sharply staining capsule. In the artificial cultures special capsule stains, it was noted, failed to stain any definite area, but numerous small deeply stained jsranules were to be seen within the halo, espe- cially near its outer border. Howard and Perkins propose for the group composed of the streptococci of Bonome, Binaghi, and their own organism, the name Strepto- coccus mucosus. Streptococci isolated from cases of epidemic sore-throat have also shown capsules (p. 343). Reference to the original descriptions of these various capsulated streptococci will show that, with the exception of a rather poorly staining capsule, the majority of these organisms are separated from the typical Streptococcus pyogenes or from the pneumococcus by exceedingly slight and unstable morphological and cultural char- acters. This is true of the difference in their pathogenic action in animals. 1 Hiss, Cent. f. Bakt., xxxi, 1902; Jour. Exp. Med., vi, 1905. 370 PATHOGENIC MICROORGANISMS Cultivated on whole-blood-agar, streptococci usually cause hemo- lysis, pneumococci usually do not.^ In contradistinction to Streptococ- cus viridans which does not hemolyze, pneumococci have a tendency on these media to form the black, dry, paint-blister colonies.^ Neufeld,^ in 1900, noticed that normal rabbits' bile added in quan- tities of 0.1 c.c. to each one or two cubic centimeters, of a pneumococcus broth culture caused lysis of the bacteria, rendering the culture fluid transparent and clear. This does not occur with streptococci, and has been used to differentiate the two species. According to Libman and Rosenthal,^ great reliance may be placed upon this method. The most convenient reagent for use in the Neufeld bile test is a 10 per cent solution of sodium taurocholate in physiological salt solution. This should be sterilized or kept on ice. One-tenth volume of such a solution produces prompt lysis in a broth culture of pneumococci. Decisive differential importance may be attached to the agglutina- tions of these microorganisms in immune sera (see p. 364). The permanency of the various types in the pnemnococcus-strepto- coccus group is still open to question. E. C. Rosenow ^ has recently re- ported that he has transmuted typical pneumococci into typical hemo- lytic streptococci by methods which he has not as yet fully described, but among which were animal passage, growth in symbiosis with bacillus subtilis, and growth in an atmosphere of oxygen. The pneumococci when first altered took on the characteristics of the streptococcus viridans, later of the so-called streptococcus rheumaticus, and finally of streptococcus hemolyticus. Together with cultural characteristics the pathogenicity of these various strains for rabbits changed. The pneumococcus produced acute sepsis, the streptococcus viridans caused endocarditis, the streptococcus 'rheumaticus periarticular or serous arthritis, and hemolyticus suppurative arthritis. In intermediate stage the organisms quite regularly caused myositis. Although he was able to transmute these types one into the other in both directions, Rose- now believes that the cultural characteristics of each type correspond to a fairly definite type of pathogenicity in animals and man. This work has not as yet appeared in detail and has not been confirmed. * Schottmiiller, Miinch. med. Woch. 2 Hiss, Jour. Exp. Med., vii, 1905. * Neufeld, Zeit. f. Hyg., 1901. * Libman and Rosenthal, Proc. N. Y. Path. Soc, March, 1908. * Rosenow, J. A. M. A., 1913, Ixi, 2007, CHAPTER XXIV MICROCOCCUS INTRACELLULARIS MENINGITIDIS (MENINGOCOCCUS) Infectious processes in the meninges may be caused by many dif- ferent microorganisms. Meningitis may be primary or secondary. Secondary meningitis may often occur during the course of pneumonia, when pneumococci, carried to the meninges by the blood stream, give rise to a usually fatal form of the disease. More rarely a similar process may occur as a secondary manifestation of typhoid fever or influenza. Meningitis may also result secondarily by direct extension from suppurative lesions about the skull, such as those occurring in diseases of the middle ear or frontal sinuses or after compound fractures. In such cases the invading or- ganisms are usually staphylococci, streptococci, or pneumococci. Isolated cases of meningeal infection with B. coli, B. paratyphosus. Bacillus pestis, and Bacillus mallei have been reported. A frequent, more chronic form of the disease is caused by Bacillus tuberculosis. Primary acute meningeal infection, however, is due chiefly to two microorganisms, Micrococcus intracellularis meningitidis, and the pneu- mococcus. A tabulation of the comparative frequency with which the various microorganisms are found in the meninges has been attempted by Marschal.^ This author estimates that about 69.2 per cent of all acute cases are due to the meningococcus, 20.8 per cent to Diplococcus pneumoniae, and the remaining 10 per cent to the other bacteria mentioned. The cases caused by the pneumococcus and the other less frequent incitants usually occur sporadically. When the disease occurs in epi- demic form, it is almost always due to the meningococcus. Diplococcus intracellularis meningitidis was first seen in menin- geal exudates by Marchiafava and Celli ^ in 1884. These authors not only described accurately the morphological characteristics now recog- 1 Marschal, Diss. Strassburg, 1901, Quoted from Weichselbaum, in Kolle und Wassermann, " Handbuch." 2 Marchiafava and Celli, Gaz. degli ospedali, 8, 1884. 871 372 PATHOGENIC MICROORGANISMS nized, but also called attention to the intracellular position of the micro- organism and to its gonococcus-like appearance. They failed, however, to cultivate it. Observations confirmatory of the Italian authors were, soon after, made by Leichtenstern.^ Cultivation and positive identification as a separate species was not accomplished, however, until Weichselbaum,^ in 1887, reported his observations upon six cases of epidemic cerebro- ' y^ ^ \, ". .V ^^ • # ^^ e # # • ' '^' ^ . •» »•••'• • • % % " It * ♦* ♦ % .V. •. W- ' .li ■ Fig. 77. — Meningococcus, Pure Culture. (Very highly magnified.) spinal meningitis in which he not only found the cocci morphologically, but was able to study their biological characteristics in pure culture. The researches of Weichselbaum were soon confirmed and extended by elaborate studies ^ which left no doubt as to the specific relationship between the microorganism cultivated by him and the clinical condition. 1 Leichtenstern, Deut. med. Woch,, 1885. 2 Weichselbaum, Fort. d. Med., 1887. ^Councilman, Mallory, and TFngrA^, Special Rep. Mass. Board of Health, 1898; AUbrecht und Ghon, Wien. klin. Woch., 1901. MICROCOCCUS INTRACELLULARIS MENINGITIDIS 373 Morphology and Staining. — Stained in the spinal fluid from an in- fected patient, the meningococcus bears a striking similarity to the gon- ococcus. The microorganisms appear intra- and extracellularly, usually in diplococcus groups, sometimes as tetrads, or even in larger agglomer- ations. The individual diplo-forms are flattened on the sides facing each other, presenting somewhat the biscuit-form of the gonococcus. The variation in size of the cocci in the same smear is a noticeable feature Fig. 78, — Meningococcus in Spinal Fluid. - and of some diagnostic importance. This dissimilarity in size is notice- able also in cultures, which, especially when older than twenty-four hours, contain forms double or even triple the size of the average coccus. These may possibly be involution forms. The meningococcus is non-motile and non-spore forming. It stains easily with all the usual aqueous anilin dyes. Its behavior toward Gram's stain was long a subject of controversy, owing to the error of Jaeger,^ who claimed to have found it Gram-positive. There ^ Jaeger, Zeit. f. Hyg., xix, 1895. 374 PATHOGENIC MICROORGANISMS is no question now, however, that the cocci decolorize by Gram's method when this is carefully carried out. In spinal fluid very beautiful preparations may be obtained by staining in Jenner's blood stain. Councilman, Mallory, and Wright ^ were the first to notice that, when stained with Loeffler's methylene-blue, meningococcus stains irregularly, showing metachro- matic granules in the center of the cell bodies. These granules can be demonstrated more clearly with the Neisser stain employed for similar demonstration in the case of B. diphtherise (see p. 107) and have some value in differentiating meningococcus from gonococcus. Cultivation. — Micrococcus intracellularis meningitidis grows readily upon all the meat-infusion culture media. It may even be culti- vated upon meat-extract media, but growth upon these is not profuse. Upon agar, colonies appear within eighteen to twenty-four hours as grayish, glistening spots with smooth edges and raised granular pentere. These show a tendency to enlargement and eventual confluence. Growth is more luxuriant and rapid upon media to which animal proteid in the form of blood serum or ascitic fluid has been added. Co- agulated serum is not liquefied. For cultivation of the meningococcus directly from the human body it is wise to use the richer serum or blood media, ability to grow easily upon simple agar being occasionally acquired only after previous cultivation upon richer media. Agar to which whole rabbit's blood has been added forms an excellent medium, both for cul- tivation and for keeping the organism alive. Loefjiefs blood serum is also very favorable. It is advisable, too, when cultivating directly from spinal fluid, to plant rather large quantities (1 to 2 c.c), since many of the cocci in the exudate will fail to develop colonies, possibly because of their prolonged exposure either to the body fluids or to their own products in a closed space. Upon broth, growth is slow and takes place chiefly upon the surface, the sediment consisting mainly of dead bacteria. Glucose added to agar or to broth renders the medium more favorable for rapid growth, but, owing to acid formation, tends to cause a more rapid death of the culture. In flasks of broth containing glucose one per cent, and CaCOg one per- cent, however, cultures have been kept alive for as long as fourteen months (Hiss). On milk, growth takes place without coagulation of the casein. Potatoes are not a favorable medium, though growth occasionally takes place. ^ Councilman, Mallory, and Wright, Rep. Mass. State Bd. of Health, 1898. MICROCOCCUS INTRACELLULARIS MENINGITIDIS 375 While slight alkahnity or acidity does not inhibit, the most favor- able reaction of media is neutrality. Oxygen is necessary for development. Complete anaerobiosis, while not absolutely inhibitory, is extremely unfavorable, unless proper carbohydrates be present. While growth may take place at temperatures ranging from 25° 1 /,'■. ^i^.^ 'I' 1 ^ 1 Fig. 79. — ^Meningococcus Culture. Streak culture from spinal fluid on serum agar-plate. to 42° C, the optimum is 37.5° C. Apart from the remarkable viability displayed upon calcium-carbonate broth, the average length of time during which the meningococcus will remain alive without transplanta- tion is rather short. Recently isolated cultures grown on agar or serum- agar may die within two or three days. Accustomed to artificial cul- tivation throu-gh a number of generations, however, the cultures become 376 PATHOGENIC MICROORGANISMS more hardy and transplantation may safely be delayed for a week or even longer. Albrecht and Ghon ^ have kept a culture alive on agar for one hundred and eighty-five days. It is a strange fact that after prolonged artificial cultivation some strains of meningococcus may gradually lose their growth energy and finally be lost because of their refusal to develop in fresh transplants. Storage is best carried out at incubator temperatures. At room temperatures or in the ice chest, the diplococcus dies rapidly .^ Resistance. — ^The meningococcus is killed by exposure to sunlight or to drying within twenty-four hours. ^ It is extremely sensitive to heat and cold and by the common disinfectants is killed in. high dilutions and by short exposures. At 0° C. it usually dies within two or three days. Pathogenicity. — As stated above, the form of meningitis caused by the diplococcus of Weichselbaum occurs usually in epidemics, though isolated sporadic cases are seen from time to time in all crowded com- munities. Epidemics have been numerous and widespread, and their records far antedate the discovery of their causative agent. As a rule, these epidemics have occurred during the winter and spring months, and have attacked chiefly that part of the population which is forced by poverty to live in crowded unhygienic surroundings. The manner in which the microorganism enters the human body is still a subject for investigation. Weichselbaum,* Ghon and Pfeiffer,^ and, more recently, Goodwin and v. Sholly ® of the New York Department of Health, have succeeded in demonstrating culturally the presence of the meningococ- cus in the nasal cavities, not only of patients suffering from the disease, but occasionally in those of healthy subjects as well. Similar findings have been reported by many others; but in many cases morphological examination only was made, which, owing to the danger of confusion with Micrococcus catarrhalis, a frequent inhabitant of the nose, renders such reports valueless. The careful work of the writers mentioned, how- ever, has given ground for the theory that meningeal infection, which is 1 Albi^echt und Ghon, Wien. klin. Woch., 1901. 2 A very thorough biological study of meningococcus and related organisms has recently been made by Elser and Huntoon (Jour. Med. Res., N. S. vol. xv, 1999), which may be consulted for a more detailed description of cultural characteristics. 3 Councilman, Mallory, and Wright, Boston, 1898; Albrecht and Ghon, loc. cit. * Weichselbaum, Fort. d. Med., 1887. ^Ghon und Pfeiffer, Zeit. f. klin. Med., xliv, 1901. « Goodwin und v. Sholly, Jour. Inf. Dis., Suppl. 2, Feb., 1906. MICROCOCCUS INTRACELLULARIS MENINGITIDIS 377 often preceded by nasal catarrh, may take place along the paths of the lymphatics, passing out of the nose and its accessory cavities toward the base of the skull. These facts, together with the low resistance shown by the meningococcus against drying, and the general failure so far to demonstrate it in air, dust, or fomites, would seem to indicate that trans- mission usually occurs directly from one human being to another. The disease produced in man consists anatomically in a suppurative lesion of the meninges, involving the base and cortex of the brain and the surface of the spinal cord. The nature of the exudate may vary from a slightly turbid serous fluid to that of a thick fibrinous exudate. In chronic cases encephalitis and dilatation of the ventricles may take place. Apart from their presence in the meninges and in the naso- pharynx, meningococci have not been satisfactorily demonstrated in any of the complicating lesions of the disease. Reports of their presence in the conjunctivae, in the bronchial secretions from broncho- or lobar pneumonia, and in otitis media, have usually been based upon insuf- ficient bacteriological evidence. The occurrence of this microorganism in the circulating blood of men- ingitis cases has been definitely proved by Elser,^ who found it in ten cases. Animals are not very susceptible to infection with Diplococcus meningitidis. Subcutaneous inoculation is rarely followed by more than a local reaction unless large quantities are used. White mice are rather more susceptible than other species. Intraperitoneal and intra- venous inoculation of sufficient quantities usually results in the death of mice, rabbits, guinea-pigs, and dogs. Occasional strains have been found to possess a not inconsiderable degree of toxicity for rabbits, grave symptoms or even death following intravenous injection of but moderate quantities without any traceable development of the micro- organisms in the organs of the animals. Similar observations have been made by Albrecht and Ghon,^ who succeeded in killing white mice with dead cultures. It would seem therefore that the effect of this coccus upon animals depends chiefly upon the poisonous substances contained in the bacterial bodies (endo- toxins) . Lepierre ^ has obtained the meningococcus toxin by alcohol precipitation of broth cultures. Weichselbaum himself succeeded in producing meningeal suppura- 1 Elser, Jour. Med. Res., xiv, 1906. 2 Albrecht und Ghon., loc. cit. * Lepierre, Jour, de phys. et de path. g6n., v, No. 3. 25 378 PATHOGENIC MICROORGANISMS tion and, in one case, brain abscess, by subdural inoculation of dogs. Councilman, Mallory, and Wright ^ produced a disease in many re- spects similar to the human disease by intraspinous inoculation of a goat. Recently, Flexner^ has succeeded in producing in monkeys a condition entirely analogous to that occurring in human beings. Agglutination. — Immunization of animals by repeated inoculations of meningococcus ^ results in the formation in the blood serum of agglutinins. KoUe and Wassermann * obtained from horses a serum which had an agglutinating value of I : 3,000 for the homologous strain, and of as much as 1 : 500 for other true meningococcus strains. Similar experiments by Dunham ^ and others have proved the unquestionable value of agglutination for species identification of this group. Great differences m^y, however, exist between individual races in their aggiutinability in the same immune serum. Kutscher has recently called attention to the fact that strains which can not be agglutinated in specific sera at 37° C. will often yield positive results when subjected to 55° C, a fact of some practical im- portance if confirmed. Elser and Huntoon ^ have shown that in the serum of infected human subjects agglutination of some strains takes place in dilutions as high as 1 : 400. Serum Therapy of Meningitis. — During recent years, attempts have been made to treat epidemic meningitis by injections, subcutaneous and intraspinous, of meningococcus-immune serum. Wassermann,^ in 1907, reported results of such treatment in one hundred and two patients, with a recovery of 32 . 7 per cent. The serum, manufactured by Was- sermann and his associates, was obtained from horses immunized with cultures of meningococcus and with toxic meningococcus extracts. More recently Flexner and Jobling ^ have used a similar serum in the United States with apparently excellent results. The serum, in Flexner's cases, as in the technique of Jochmann, is injected intraspinously after a quantity of spinal fluid had been withdrawn. The cases treated by Flexner and Jobling' s method have now reached 1 Councilman, Mallory, and Wright, loc. cit. 2 Flexner, Jour. Exp. Med., 1906. 3 Albrecht and Ghon, Wien. klin. Woch., 1901. ■ Kolle und Wassermann, Deut. med. Woch., 15, 1906. * Dunham, Jour. Inf. Dis., 11, 1907. « Elser and Huntoon, loc. cit. » Wassermann, Deut. med. Woch., 39, 1907. . « Flexner and Joblinq, Jour. Exper. Med., x, 1908. MICROCOCCUS INTRACELLULARIS MENINGITIDIS 379 large numbers, both in this and foreign countries and the value of the serum as a therapeutic agent seems firmly established. Hiss and Zinsser ^ have treated a number of meningitis patients with subcutaneous injections of leucocyte extracts and believe that they have favorably influenced the course of the disease. Pseudomeningococcus. — Elser and Huntoon ^ have described a diplo- coccus very similar to the meningococcus which they differentiated from it only by serum reactions. This diplococcus could be identified only by agglutinin absorption tests. They named it pseudomeningo- coccus. Parameningococcus. — Dopter^ described a Gram-negative diplococ- cus which was identical with meningococcus in its cultural and fer- mentative characteristics. It failed to give agglutination and pre- cipitation reactions in anti-meningitis serum, though it did give a complement-fixation reaction. These organisms, at first isolated from normal throats, have since been found in the spinal fluid of cases of meningitis. In a recent study of strains from meningitis Amoss and Wollstein found that besides the two distinct types — normal menin- gococci and parameningococci — there were a number of intermediate varieties. Olmstead and Dubois* and Neal and Schweitzer^ also found marked antigenic differences among the strains of meningococci which they studied. The use of representatives of each of the types of meningococci and parameningococci has been found essential for the production of a potent anti-meningitis serum. 1 Hiss and Zinsser, Jour. Med. Res., Nov., 1908. 2 Elser and Huntoon, Jour. Med. Res., xxi, 1909. 3 Dopter, "Conte Rendu de la Society de Biologie," 1909, Ixvii, p. 74, * Olmstead and Dubois, Jour. Exper. Med., xxiii, p. 403. 5 Neal and Schweitzer, Jour, of Immunol., 1916, i, p. 307* CHAPTER XXV DIPLOCOCCUS GONORRHCEiE (GONOCOCCUS), MICROCOCCUS CATARRHALIS, AND OTHER GRAM-NEGATIVE COCCI DIPLOCOCCUS GONORRHCEiE 4 Neisser/ in 1879, described diplococci which he had found regulai^^ in the purulent secretions of acute cases of urethritis and vaginitis arfd in the acute conjunctivitis of the new-born. His researches were purely morphological, as were the numerous confirmatory investigations which rapidly followed his announcement. Cultivation of this diplococcus, now usually spoken of as gonococcus, was not definitely successful until 1885, when Bumm ^ obtained growth upon tubes of coagulated human blood serum. Bumm was not only able to keep the organisms alive by transplantation in pure culture, but produced the disease by inoculation of his cultures upon the healthy urethra. Morphology and Staining. — The gonococcus is usually seen in the diplococcus form, the pairs being characteristically flattened along the surfaces facing each other. This gives the cocci a peculiar coffee-bean or biscuit shape. The size of the diploforms is about 1.6 micra in the long diameter, about 0.8 micron in width. Stained directly in gonorrheal pus from acute cases, the microorganisms are found both intra- and extracellularly, a large number of them crowded characteristically within the leucocytes. They are never found within the nucleus. The phagocytosis. which produces this picture has been shown by Scholtz ^ and others to take place in the free secretions, not in the depth of the tissues. The intracellular position, which is of considerable diagnostic importance, is lost to a great extent in secretions from chronic cases. In smears made from pure cultures the arrangement in groups of two may often be less marked than in pus, clusters of eight or more being common. 1 Neisser, Cent. f. d. med. Wiss., 1879. 2 Bumm, " Beitr. z. Kenntniss des Gonococcus," Wiesbaden, 1885. 3 Scholtz, Arch, f, Dermat,, 1899. 330 DIPLOCOCCUS GONORRHCEiE^ 381 The gonococcus is non-motile and does nof form spores. It is easily stained with aqueous anilin dyes. Methylene-blue alone, or eosin followed by methylene-blue, give good results. An excellent picture is obtained with the Pappenheim-Saathof stain consisting of Methyl green 0 . 15 Pyronin 0.5 96% alcohol 5.0 • Glycerin 20.0 2% carbolic acid water ad .• 100 . 0 Fix stain 1-2 min. Gram's method of staining, however, is the only one of differential value, gonococcus being Gram negative. The Gram stain applied to pus from the male urethra, while not absolutely reliable, is, for practical purposes, sufficiently so to make a diagnosis. In exudates from the vagina or from the eye the morphological pic- ture is not so reliable, owing to the frequent presence in these regions of other Gram-negative cocci. The great scarcity of gonococci in very chronic discharges necessitates thorough cultural investigation; negative morphological examina- tion in such cases can not be regarded as con- clusive.^ Cultivation. — The gonococcus is delicate and difficult to cultivate. Bumm ^ obtained his first growths upon human blood serum which had been heated to partial coagulation. The medium most commonly used at the present day was introduced by Wertheim,^ and consists of a mixture of two or three parts of meat infusion-agar with one part of uncoagulated human ascitic fluid, hydrocele fluid, or blood serum. The agar is melted and cooled to 45° is added. The mixture may then be slanted or poured into a Petri plate. Fig. 80. — Gonorrheal Pus FROM Urethra, Showing the Cocci WITHIN A Leucocyte. before the serum in the test tube One per cent of glucose may be added. Cultures in fluid media may be obtained by similar additions of serum to meat-infusion-pepton-broth. Whole rabbit's blood added to agar, ^ Heinian, Medical Record, 1896. " Bumm, Deut. med. Woch., 1885. 3 Wertheim, Arch, f . Gynakol., 1892. \.' •• % 9 % 4 • -'Z , • • -•* • • • '• .••.■. •* .%. •# .. ^■■.; /.• M • j • • •• • • . y % «• "T. ••:.. i 382 PATHOGENIC MICROORGANISMS or the swine-serum-nutrose medium of Wassermarin ^ may occasionally be used with success. Plates may be made by smearing for enrichment a drop of blood from the finger over the surface of agar in the manner of Pfeiffer's method for influenza-bacillus cultivation. Inoculations from gonorrheal material are best made by surface smearing upon plates, since the gono- coccus grows best in the presence of free oxygen. Growth becomes more luxuriant after prolonged cultivation upon artificial media. The most favorable reaction of media is neu- trality or slight acidity. When the gonococcus has been successfully cultivated from pus upon media without serum additions, the success has probably been due to the Fig. 81. — Gonococcus. Smear , , . , • -i f ,, substances carried over in the pus. from pure culture. i . . . The ease of cultivation differs con- siderably with different strains of gonococci. Some grow very heavily after first isolation, but the majority show a very delicate growth even on rich ascitic glucose agar. After several generations of growth on artificial media, however, the organism develops with increasing ease and on simpler media. It may eventually be cultivated on plain agar, especially when this is made of veal infusion. Recently a medium upon which gonococci after first cultivation can be grown with ease has been recommended by Edward B. Vedder.^ The medium consists of a 1 . 5 to 1 . 75 per cent agar made with beef infusion neutral to phenol- phthalein, and after clearing, 1 per cent of corn starch added. The corn starch is best added after grinding with a little agar to avoid clumps, this then being poured into the bulk of the agar and thoroughly mixed. The medium should be sterilized at not over 15 lbs. pressure to avoid changes in the starch. Recently we have isolated several strains of gonococci which grew very heavily on simple media without ascitic fluid in the second culture generation. ^ Wassermann, Berl. klin. Woch., 1897. (Fifteen c.c. swine-serum, 35 c.c. of water, 3 c.c. glycerin, with two per cent nutrose. The nutrose is dissolved by boiling and the solution sterilized. This is then added to agar, in equal parts, and used in plates.) 2 Vedder, Jour. Infec. Dis., May 15, 1915, xvi, 385. DIPLOCOCCUS GONORRHCE^ 383 The gonococcus will develop sparsely under anaerobic conditions, but has marked preference for aerobiosis. The optimum temperature is 37.5° C. Growth ceases above 38.5° and below 30°. Upon suitable media colonies appear as extremely delicate, grayish, opalescent spots, at the end of twenty-four hours. The separate colo- nies do not tend to confluence and have slightly undulated margins. Touched with a platinum loop their consist- ency is found to be slimy. In fluid media, growth takes place chiefly at the surface. Until recent years the gonococci were regarded as a compact group, within which individual members, isolated from different cases, were identical in all respects. How- ever, work done by Torrey,* and by Teague and Torrey,^ has shown that immunologic- ally gonococci can be divided into a number of different groups. The situation here is closely analogous to that which has developed in the case of the pneumo- coccus group. By cross agglutinations and agglutinin absorption expe- riments, Torrey has been able to show that gonococci fall into about 10 groups, A, B, C, G, K, L, N, O, Q, S, which are serologically separable one from the other. The practical importance of this is that in all diagnostic serum reactions, such as the complement-fixation test, it is necessary to make a polyvalent antigen in which these various groups shall be represented. Otherwise, of course, many cases would react negatively if infected with a particular group which is not included in the antigen. Resistance. — Recent cultures of gonococcus, if not transplanted, usually die out within five or six days at incubator temperature. At room temperature they die more rapidly. The resistance of the gonococcus to light and heat is slight. A tem- perature of 41° to 42° kills it after a brief exposure. Complete drying 1 Torrey, Jour, of Med. Research, 1907, xvi, 329. 2 Teague and Torrey, Jour, of Med. Research, Dec, 1907, xvii, 223. Fig. 82. — Gonococcus Colony. Low power of magnification. (After Mallory and Wright.) 384 PATHOGENIC MICROORGANISMS destroys it in a short time. Incompletely dried, however, and pro- tected from light (gonorrheal pus) it may live, on sheets and clothing, for as long as eighteen to twenty-four hours.^ It is easily killed by most disinfectant solutions ^ in high dilution and seems to be almost specifically sensitive to the \arious silver salts, a fact of therapeutic importance. Pathogenicity. — Gonorrheal infection occurs spontaneously only in man. True gonorrheal urethritis has never been experimentally pro- duced in animals. In human beings, apart from the infection in the male and female genital tracts, and in the conjunctivae, the gonococcus may produce cystitis, proctitis, and stomatitis. It may enter the cir- culation, giving rise to septicemia, ^ to endocarditis and arthritis. Iso- lated cases of gonorrheal periostitis and osteomyelitis have been re- ported.^ The acute infections of the genito-urinary passages are often fol- lowed by prolonged chronic infection, which, though quiescent, may for many years be a source of social danger. In children, especially females, the infection is not rare, and may assume epidemic characters, traveling from bed to bed in institutions. Such hospital epidemics can be stopped only by the most rigid isolation. It is advisable, in view of this danger, ^to examine all female children applying for admission to a hospital, by 'vaginal smear and, if possible, to keep them in a receiving ward for twenty-four hours in order that the examination may be repeated before admission to the general wards. In the best-equipped institutions, fur- thermore, separate thermometers, bed linen, and diapers are set aside for each child in order to preclude any possibility of accidental transmis- sion from cases which may have escaped detection by smear examination. While inoculation of animals has never resulted in active prolifera- tion of the gonococcus upon the new host, local necrosis, suppuration, and temporary systemic reactions have been produced by subcutaneous and intraperitoneal inoculation. A toxin has been isolated by Niko- laysen ^ by extraction from the bacterial bodies with distilled water or sodium hydrate solutions. It was found to be resistant to a tempera- ture of 120° and to remain potent after complete drying. The same author found that the isolated toxin and dead cultures were fully as toxic for animals as living cultures, 0.01 gram killing a white mouse. ^ Heiman, Medical Record, 1896. 2 Schaeffer und Steinschneider, Kong. Deut. Dermat. Gesells., Breslau, 1894. ' Review of cases of Gon. Septicemia, Faure-Beaulieu, Thesis, Paris, 1906. * ^Jllmann, Wien, med. Presse, 1900. ^ Nikolaysen, Cent. f. Bakt., 1897. 1I4ICR0C0CCUS CATARRHALIS 385 Specific injury to the nervous system by injections of gonococcus toxin has been reported by Moltschanoff.^ The secretion of a true soluble toxin by the gonococcus, asserted by Christmas,^ is denied by Wassermann,^ Nikolaysen,^ and others. Christ- mas,^ and, more recently, Torrey,^ hare reported successful active im- munization of animals by repeated injections of whole bacteria. Torrey and others apparently have successfully treated human cases by injec- tions of the serum of immunized animals. Antibodies to Gonococcus. — Patients infected with gonococci seem to produce antibodies against the organisms. Although in the or- dinary gonorrheal urethritis, or vaginitis, it is relatively simple to make the diagnosis by finding gonococci in the discharges, diagnosis may be difficult in cases of gonorrheal rheumatism, or endocarditis, when iso- lation of the bacteria fails or when the connection between the local venereal disease and the general condition is obscure. Various sero- logical diagnostic methods have been attempted, and of recent years the complement-fixation test has been found to be very useful. The method has been especially developed by Archibald McNeil, at the New York Department of Health. It consists in making a polyvalent antigen, using the 10 Torrey strains which are kept in stock transplants on glucose ascitic agar. It has been found that the best medium for antigen production is an agar made of "bob veal." For the production of antigen, stock cultures are transplanted on "bob veal" agar, without salt, glucose or ascitic fluid, the reaction carefully adjusted to an acidity of 0.1 per cent to 0.2 per cent. Twenty-four hour growths on this me- dium are scraped off and emulsified in neutral sterile distilled water. The emulsion is autolyzed one hour in a water bath at 56° and heated one hour at 80° C. It is then filtered through a sterile Berkfeld filter. The filtrate is aseptically bottled and sterilized 3 days at 56°, half an hour each day. It is then made isotonic and is ready for titration. Vaccine therapy in systemic gonorrheal infection has been tried and is reasonably successful. The vaccine, if possible, should be made with the organism isolated from the patient, for reasons described above. Passive immunization with the serum of gonococcus-immune animals has also been attempted, but records on it at present are not sufficiently complete to permit definite judgment. ';:.' 1 Moltschanoff, Munch, med. Woch., 1899, ^Christmas, Ann. de Tlnst. Pasteur, 1897. 3 Wassermann, Zeit, f, Hyg., xxvii, 1897. * Nikolaysen, Fort. d. Med., xxi, 1897. ^ Christmas, loc, cit. * Torrey, Jour. Amer. Med. Assn., xlvi, 1906. 386 PATHOGENIC MICROORGANISMS MICROCOCCUS CATARRHALIS Micrococcus catarrhalis is a diplococcus described first by R. Pfeiffer*, who found it in the sputum of patients suffering from catarrhal in- flammations of the upper respiratory tract. It was subsequently care- fully studied by Ghon and H. Pfeiffer.^ According to these authors the pathogenic significance of the micrococcus is slight, though occasion- ally it may be regarded as the causative factor in catarrhal inflammations. Its chief claim to attention, however, lies in its similarity to the meningo- coccus and the gonococcus, from neither of which it can be morphologi- cally distinguished. It is decolorized by Gram's stain, appears often in the diplococcus form, and has a tendency, in exudates, to be located intracellularly. Not unlike the two microorganisms mentioned, too, it shows but slight pathogenicity for animals. Differentiation from gonococcus i§ extremely simple in that Micro- coccus catarrhalis grows easily on simple culture media and shows none of the fastidious cultural requirements of the gonococcus. From meningococcus the differentiation is less simple and, because of the presence of both microorganisms in the nose, is of great importance. Distinction between the two is made entirely upon cultural charac- teristics and agglutination reactions. Culturally, Micrococcus catar- rhalis grows more heavily than meningococcus upon the ordinary culture media. The colonies of Micrococcus catarrhalis are coarsely granular and distinctly white in contradistinction to the finely granu- lar, grayish meningococcus colonies.^ Micrococcus catarrhalis will develop at temperatures below 20° C, while meningococcus will not grow at temperatures below 25° C* Dunham,^ who has recently made a comparative study of meningo- coccus and other Gram-negative diplococci from the nose and throat, states that while some of the supposed Micrococcus catarrhalis cul- tures are easily distinguished from meningococcus simply by the char- acteristics of their growths upon two-per-cent glucose agar, others offer great difficulties to differentiation. He recommends as a differential medium a mixture of sheep serum and bouillon containing 1% of glucose. Upon this medium all true meningococci produce acid, but no coagulation, » Flugge, "Die Mikroorg.," 3d ed., 1896. « Ghxm und H. Pfeiffer, Zeit. f. klin. Med., 1902. » Ghon und Pfeiffer, loc. cit. * Weichselbaum, in Kolle und Wassermann, Bd. iii, p. 269. ^ Dunham, Jour. Inf. Dis.. 190'^ GRAM-NEGATIVE COCCI 387 with 24 hours. Cultures from the nose and throat, however, produce acid and coagulation, or else produce an alkaline reaction. OTHER GRAM-NEGATIVE COCCI Micrococcus pharyngis siccus. — First described by von Lingelsheim^ in 1906. It is described by Elser and Huntoon as readily differentiable from meningococcus and other Gram-negative cocci by the firm adher- ence and dryness of its colonies. It is similar to Micrococcus catarrhalis from which it may, however, be differentiated by fermentation tests. Diplococcus mucosus. — This organism was also described by von Lingelsheim together with the preceding one. Its colony formation is similar to that of meningococcus, but slightly more sticky and mucoid. Stained by the capsule methods, it is seen to possess a distinct capsule. Chromogenic Gram-negative Cocci. — These microorganisms all pro- duce a greenish-yellow pigment on the ordinary culture media. When pigment is absent, as is frequently the case when grown upon sugar-free media, these microorganisms can be distinguished from meningococcus only by sugar fermentation and senrni reactions. An exhaustive study of Gram-negative micrococci has recently been made by Elser and Huntoon.^ These authors, in studying the differ- ential value of sugar fermentation in the diagnosis of these bacteria, have constructed the following table: Strains Tested Meningococcus Pseudomeningococcus Gonococcus Micrococcus catarrhalis Micrococcus pharyngis siccus Chromogenic group I Chromogenic group II Chromogenic group III Jaeger meningococcus, Krai. . Diplococcus crassus, Krai Strains Dex- Malt- Levu- Saccha- Lac- trose ose lose rose tose 200 + + 0 0 0 6 + + 0 0 0 15 + 0 0 0 0 64 0 0 0 0 0 2 + + + 0 28 + + + 0 11 + + + 0. 0 9 + + 0 0 0 1 + + + + + 1 + + + + + Gal- actose 1 V. Ldngelsheim, Klin. Jahrb., 15, 1906, * EUer and Huntoon, loc. cit. CHAPTER XXVI BACILLI OF THE COLON-TYPHOID-DYSENTERY GROUP The bacilli belonging to this group of microorganisms/while present- ing great differences in their pathogenic characteristics, possess many points of morphological and biological similarity which have made their differentiation extremely difficult. Among pathogenic bacilli, they are probably the ones most commonly encountered and because of the fact that some of them are specifically pathogenic, while others are essen- tially saprophytic and are pathogenic only under exceptional conditions, the necessity of accurate differentiation is a daily occurrence in bacteri- ological laboratories. It has been through the study of this group par- ticularly that many of the modern differential methods of bacteriology have been developed. The group includes the colon bacillus and its allies, the typhoid bacillus, the paratyphoid organisms, the several varieties of dysentery bacillus and numerous closely related species, and Bacillus fecalis alka- ligenes. Closely related to the group though not properly within it, are Bacillus lactis aerogenes, B. acidi lactici bacilli of the Friedlander or mucosus capsulatus group, and a number of less important subdivisions of this last group. All bacilli of the group possess morphological characteristics which^ although exhibiting slight differences, are insufficient to permit accurate morphological diagnosis. They are none of them spore-bearing. Stained by Gram's method they are decolorized. Cultivated upon artiftcial media, they grow readily both at room and at incubator temperatures. None of them liquefies gelatin. Though showing, often, distinct differences in the speed and luxuriance of growth upon ordinary media, these differences are, nevertheless, too slight to become the basis of differentiation. In order to distinguish between the individual members of this group, therefore, we are forced to a careful biological and cultural study. This is carried out by the observation of the cultural character- istics upon special media and by the study of serum reactions in speci- fic immune sera. Our mainstays in the accurate differentiation of these 388 BACILLUS COLl COMMUNIS 389 bacilli are the observation of their fermentative action upon carbohy- drate media, and their agglutinating reactions in immune sera. These 'points will be alluded to in the description of the individual microor- ganisms, and will be again summarized in the differential tables given at the end of the chapters dealing with this group. BACILLUS GOLI COMMUNIS AND MEMBERS OF THE COLON BACILLUS GROUP Under the name of " colon bacilli " are grouped a number of bac- terial varieties differing from one another somewhat in minor character- istics, but corresponding in certain cardinal points which stamp them as close relatives and amply warrant their consideration under one heading. While usually Hving as harmless parasites upon the animal and human body, and capable of leading a purely saprophytic existence, they may, nevertheless, under certain circumstances become pathogenic and thus, both culturally and in their pathological significance, form a link between pure saprophytes like Bacillus lactis aerogenes, on the one hand, and the more strictly pathogenic Gram-negative bacilli of the paratyphoid, typhoid, and dysentery groups, on the other. As a type of the group we may consider in detail its most prominent and thoroughly studied member, Bacillus coli communis. . BACILLUS COLI COMMUNIS This microorganism was seen and described by Buchner ^ in 1885. It was thoroughly studied in the years immediately following, especially by Escherich,^ in connection with the intestinal contents of infants. Morphology. — Bacillus coli communis is a short, plump rod about 1-3 micra long, and varying in thickness from one-third to one-fifth of its length. Under varying conditions of cultivation, it may appear to be more slender than this or shorter and even coccoid in form. In stained preparations, it usually appears singly, but occasionally may be seen in short chains. It stains readily with the usual anilin dyes and decolorizes by Gram's method. Spores are not formed. It is motile, and flagella staining reveals eight or more flagella peripherally arranged. Its motility is subject to wide variations. Young cultures, in the first gen- 1 Buchner, Arch. f. Hyg., 3, 1885. ^Escherich, "Die Darmbakt. des Sauglings," Stuttgart, 1880; Cent. f. Bakt., 1, 1887. 390 PATHOGENIC MICROORGANISMS eration, after Isolation from the body, may be extremely motile, while old laboratory strains may show almost no motility. Independent of these modifying conditions, however, separate races may show individual characteristics as to motility, varying in range between a motility hardly distinguishable from Brownian movement and one which is so active as to be but little less than that of the typhoid bacillus. Ordi- narily, the colon bacillus possesses a motility intermediate between these two extremes. Cultivation. — The bacillus is an aerobe capable of anaerobic growth under suitable cultural conditions. It grows well on the simplest media i^iG. 83. — Bacillus coli communis. at temperatures ranging from 20° to 40° C, but finds its optimum growth at about 37.5° C. Upon broth it grows rapidly, giving rise to general clouding; later to a pgllicle and a light, slightly slimy sediment. Withm moderate ranges, it is not delicately susceptible to reaction, growing equally well on media slightly acid and on those of a moderate alkalinity. Upon agar, it forms grayish colonies which become visible within twelve to eighteen hours, gradually becoming more and more opaque as they grow older. The deep colonies are dense, evenly granular, oval. BACILLU& COLI COMMUNIS 391 or round. Surface colonies often show a characteristic grape-leaf structure, or may be round and flat, and show a definitely raised, glisten- ing surface. Upon agar slants, growth occurs in a uniform layer. On gelatin the colon bacillus grows rapidly, causing no liquefaction. Surface colonies are apt to show the typical grape-leaf formation. Deep colonies are round, oblong, and glistening. In gelatin stabs growth takes place along the entire line of inoculation, spreading in a thin layer over the surface of the medium. On potato, growth is abundant and easily visible within eighteen to twenty-four hours, as a grayish-white, glistening layer which later turns to a yellowish-brown, and in old cultures often to a dirty green- ish-brown color. In pepton solution indol is formed. In milk there is acidity and co- agulation. In lactose-litmus-agar acid is formed, the medium becom- ing red, and gas-bubbles appear along the line of the stab inoculation. In carbohydrate broth, gas is formed in dextrose, lactose, and mannit, but not in saccharose. Levulose, galactose, and maltose are also fer- mented with the formation of acid and gas. Cultures of the colon bacillus are characterized by a peculiar fetid odor which is not unlike that of diluted feces. The acids formed by the colon bacillus from sugars are chiefly lactic, acetic, and formic acids. The gas it produces consists chiefly of CO 2 and hydrogen. The bacillus grows well on media containing urine and on those containing bile. Upon the latter fact some methods for the isolation of the colon bacillus from water and feces have been based. Isolation of the colon bacillus from mixed cultures is most easily accomplished by plating upon lactose-litmus-agar, the Conradi-Drigal- ski medium, or the Endo medium after preliminary enrichment of the. specimen to be tested in bile or malachite-green broth. (In the case of feces such enrichment is superfluous.) Distribution. — The colon bacillus is a constant inhabitant of the intestinal canal of human beings and animals. It is also found occasion- ally in soil, in air, in water, and in milk and is practically ubiquitous in all neighborhoods which are thickly inhabited. When found in nature its presence is generally taken to be an indication of contamination from human or animal sources. Thus, when found in water or milk, much hygienic importance is attached to it. Recently, Papasotiriu ^ and, independently of him, Prescott,^ have reported finding bacilli apparently » Papasotiriu, Arch, f, Hyg., xU. ^ Presoott, Cent. f. B^kt , Ref., xxxiii, 1903, 302 PATHOGENIC MICROORGANISMS identical with Bacillus coli upon rye, barley, and other grains. They believCj upon the basis of this discovery, that Bacillus coli is widely distributed in nature and that its presence, unless it appears in large numbers, does not necessarily indicate recent fecal contamination. These reports, however, have not found confirmation by the work of others, and can not, therefore, be as yet accepted. In man. Bacillus coli appears in the intestine normally soon after birth, at about the time of taking the first nourishment.^ From this time on, throughout life, the bacillus is a constant intestinal inhabitant ap- parently without dependence upon the diet. Its distribution within the intestine, according to Gushing and Livingood,^ is not uniform, it being found in the greatest numbers at or about the ileocecal valve, diminish- ing from this point upward to the duodenum and downward as far as the rectum. Adami ^ and others claim that, under normal conditions, the bacillus may invade the portal circulation, possibly by the inter- mediation of leucoeytic emigration during digestion. After death, at autopsy, Bacillus coli is often found in the tissues and the blood with- out there being visible lesions of the intestinal mucous membrane.^ It is probable, also, that it may enter and Hve in the circulation a few hours before death during the agonal stages. Extensive investigations have been carried out to determine wheth- er or not the constant presence of this microorganism in the intestinal tract is an indication of its possessing a definite physiological function of advantage to its host. It has been argued that it may aid in the fermen- tation of carbohydrates. The question has been approached experiment- ally by a number of investigators. Nuttall and Thierfelder ® delivered guinea-pigs from the mother by Cesarean section and succeeded in keeping them without infection of the intestinal canal for thirteen days. Although no microorganisms of any kind were found in the feces of these animals, no harm seemed to accrue to them, and some of them even gained in weight. Schottelius,^ on the other hand, obtained con- tradictory results when working with chicks. Allowing eggs to hatch in an especially constructed glass compartment, he succeeded in keeping the 1 Schild, Zeit. f. Hyg., xix, 1895; Lembke, Arch. f. Hyg., xxvi, 1896. 2 Gushing and Livingood, " Contributions to Med. Sci. by Pupils of Wm. Welch," Johns Hopk. Press, 1900. 3 Adami, Jour, of Amer. Med. Assn., Dec, 1899. * Birch-Hirschfeld, Ziegler's Beitr., 24, 1898. » NiUtall und Thierfelder, Zeit. f. Physiol. Chemie, xxi and xxii • Schot^lius, Arch. f. Hyg., xxxiv^ 1889. BACILLUS COLI COMMUNIS 393 chicks and their entire environment sterile for seventeen days. During this time they lost weight, did not thrive, and some of them were mori- bund at the end of the second week, in marked contrast to the healthy, well-noumished controls, fed in the same way, but under ordinary en- vironmental conditions. Although insufficient work has been done upon this important question, and no definite statement can be made, it is more than likely that the function of the Bacillus coli in the intestine is not inconsiderable if only because of its possible antagonism to cer- tain putrefactive bacteria, a fact which has been demonstrated in inter- esting studies by Bienstock ^ and others.^ Pathogenicity. — The pathogenicity of the colon bacillus for animals is slight and varies greatly with different strains. Intraperitoneal in- jections of 1 c.c. or more of a broth culture will often cause death in guinea-pigs. Large doses intravenously administered to rabbits may frequently cause a rapid sinking of the temperature and death with symptoms of violent intoxication within twenty-four to forty-eight hours. Subcutaneous inoculation of moderate doses usually results in nothing more than a localized abscess from which the animals recover. It is likely that, even in fatal cases, death results chiefly from the action of poisons liberated from the disintegrating bacteria and not from the multiplication of the bacilli themselves, for often no living organism can be found unless large doses are given. In man, a large variety of lesions produced by Bacillus coli have been described. It is a surprising fact that disease should be caused at all, in man, by a microorganism which is so constantly present in large numbers in the intestine and against which, therefore, it is to be expected that a certain amount of immunity should be developed. A number of explanations for this state of affairs have been advanced, none of them entirely satisfactory. It is probable that none of the poi- sonous products of the colon bacillus is absorbed unchanged by the healthy unbroken mucosa and that, therefore, the microorganisms are, strictly speaking, at all times, outside of the body proper. Under these circumstances, no process of immunization would be anticipated. It is also possible that, whenever an infection with Bacillus coli does occur, the infecting organism is one which has been recently acquired from another host, having no specific adaptation to the infected body. Viru- lence may possibly be enhanced by inflammatory processes caused by other organisms. Considering the subject from another point of view, » Bienstock, Arch. f. Hyg., xxxix, 190L ^ a Tissier and Martelly, Ann. de I'inst. Pasteur, 1902. 394 PATHOGENIC MICROORGANISMS colon-bacillus infection may possibly take place simply because of unu- sual temporary reduction of the resistance of the host. Whether or not altered cultural conditions in the intestine may lead to marked enhance- ment in the virulence of the colon bacilli can not at present be decided. The opinion has been frequently advanced, however, without adequate experimental support. Septicemia, due to the colon bacillus, has been described by a large number of observers. It is doubtful, however, whether many of these cases represent an actual primary invasion of the circulation by the bacilli, or whether their entrance was not simply a secondary phenomenon occurring during the agonal stages of another condition. A few unques- tionable cases, however, have been reported, and there can be no doubt about the occurrence of the condition, although it is probably less frequent than formerly supposed. The writers have observed it on two occasions in cases during the lethal stages of severe systemic disease due to other causes. An extremely interesting group of such cases are those occurring in new-born infants, in which generalized colon-bacillus infection may lead to a fatal condition known as WinckeFs disease or hemorrhagic septicemia.^ Prominent among disease processes attributed to these microorganisms are various diar- rheal conditions, such as cholera nostras and cholera infantum. The relation of these maladies to the . colon bacillus has been studied es- pecially by Escherich,^ but satisfactory evidence that these bacilli may specifically cause such conditions has not been brought. While it is not unlikely that under conditions of an excessive carbohydrate diet, colon bacilli may aggravate morbid processes by a voluminous formation of gas, they do not, of themselves, take part in actual putrefactive proc- esses. It is likely, therefore, that in most of the intestinal diseases formerly attributed purely to bacilli of the colon group, these micro- organisms actually play but a secondary part.^ It is equally difficult to decide whether or not these bacilli may be regarded as the primary cause of peritonitis following perforation of the gut. Although regularly found in such conditions, they are hardly ever found in pure culture, being accompanied usually by staphylococci, streptococci, or other microorganisms, whose relationship to disease is far more definitely established. Isolated cases have been reported, however, one of them by Welch, in which Bacillus coli was present in 1 Kamen, Ziegler's Beitr., 14, 1896. 2 Escherich, loc. cit. 8 Herter, " Bact. Infec of Digest. Tract," N. Y., 1907. BACILLUS COLI COMMUNIS 395 the peritoneum in pure culture without there having been any intestinal perforation.* Granting that the bacillus is able to proliferate within the peritoneum, there is no reason for doubting its ability of giving rise to a mild suppurative process. Infiammator}^ conditions in the liver and gall-bladder have fre- quently been attributed to the colon bacillus. It has been isolated from liver abscesses, from the bile, and from the center of gall-stones. Welch has reported a case of acute hemorrhagic pancreatitis in which the bacillus was isolated from the gall-bladder and from the pancreas. In the bladder. Bacillus coli frequently gives rise to cystitis and oc- casionally to ascending pyonephrosis. No other microorganism, in fact, is found so frequently in the urine as this one. It may be present, often, in individuals in whom all morbid processes are absent. The condition is frequently observed during the convalescence from typhoid fever. It may disappear spontaneously, or cystitis, usually of a mild, chronic variety, may supervene. Localized suppurations due to this bacillus may take place in all parts of the body. They are most frequently localized about the anus and the genitals. They are usually mild and easily amenable to the simplest surgical treatment. Poisonous Products of the Colon Bacillus. — The colon bacillus belongs essentially to that group of bacteria whose toxic action is supposed to be due to the poisonous substances contained within the bacillary body. Culture filtrates of the colon bacillus show very little toxicity when in- jected into animals; whereas the injection of dead bacilli produces symptoms almost equal in severity to those induced by injection of the live microorganisms. Corroborative of the assumption of this endotoxic nature of the colon-bacillus poison is the fact that, so far, no antitoxic bodies have been demonstrated in serum as resulting from immuniza- tion. Immunization with the Colon Bacillus. — The injection into animals of gradually increasing doses of living or dead colon bacilli gives rise to specific bacteriolytic, agglutinating, and precipitating substances. The bacteriolytic substances may be easily demonstrated by the technique of the Pfeiffer reaction. In vitro bacteriolysis is less marked than in the case of some other microorganisms such as the cholera spiril- lum or the typhoid bacillus. Owing probably to the habitual presence of colon bacilli in the intestinal tracts of animals and man, considerable 1 Welch, Med. News, 59, 1891. 396 PATHOGENIC MICROORGANISMS bacteriolysis may occasionally be demonstrated in the serum of normal individuals. Agglutinins for the colon bacillus have often been produced in the sera of immunized animals in concentration sufficient to be active in dilu- tions of 1 : 5,000 and over. The agglutinins are produced equally well by the injection of live cultures and of those killed by heat, if the tem- perature used for sterilization does not exceed 100° C. It is * a notice- 12 3 Fig. 84. — Bacillus coli communis. Grown in: 1. Dextrose, 2. Lactose, 3. Saccharose broth. The bacillus forms acid and gas from dextrose and lactose, not from saccharose. Note the absence of growth in the closed arm of the sac- charose tube, in which no acid or gas is formed. able fact that the injection of any specific race of colon bacilli produces, in the immunized animal, high agglutination values only for the individual culture used for immunization, while other strains of colon bacilli, although agglutinated by the serum in higher dilution than are paratyphoid or typhoid bacilli, require much higher concen- tration than does the original strain. The subject has been extensively studied by a number of observers and illustrates the extreme individual » Wolff, Cent. f. Bakt., xxv, 1899. BACILLUS COLI COMMUNIS 397 specificity of the agglutination reaction. Thus a serum which will agglutinate its homologous strains in dilutions of one 1 : 1,000 will often fail to agglutinate other races of Bacillus coli in dilutions of 1:500 or 1 : 600. The normal serum of adult animals and man will often agglutinate this bacillus in dilutions as high as 1 : 10 or 1 : 20 — a phenomenon pos- sibly referable to its habitual presence within the body. Corrobo- rating this assumption is the observation of Kraus and Low/ that the serum of new-born animals possesses no such agglutinating powers. The fact that agglutinins for the colon bacillus are increased 1 2 Fig. 85. — Bacillus coli communior. Grown in: Saccharose broth. 1. Dextrose, 2. Lactose, 3, in the serum of patients convalescing from typhoid fever or dysentery is probably to be explained, partly by the increase of the group agglu- tinins produced by the specific infecting ageiit, and partly by the in- vasion of colon bacilli, or the absorption of its products induced by the diseased state of the intestinal mucous membrane. Varieties of the Colon Bacillus. — During the earlier days of bacterio- logical investigations, a large number of distinct varieties of colon bacilli Were described, many of which may now be dismissed as based simply 1 Kraus und Low, Wien. klin. Wocb., X899, 398 PATHOGENIC MICROORGANISMS upon a temporary depression of one or another cultural characteristic of Bacillus coli communis, while others can be definitely included within other closely related, but distinct groups. That secondary features, such as dimensions, motility, and luxuri- ance of growth upon various media, may be markedly altered by arti- ficial cultivation is a common observation. It has not, however, been satisfactorily shown that cardinal characteristics, such as the forma- tion of indol from pepton, or the power to produce gas from dextrose and lactose, can be permanently suppressed without actual injury or inhibition of the normal vitality of the microorganism. Such alter- ation is, in fact, contrary to experience, which demonstrates that whenever such changes do occur, they are purely temporary and a few generations of cultivation under favorable environmental conditions will regularly restore the organism to its normal activity. Bacillus coli communior. — Distinct and constant varieties of Bacil- lus coli, however, do occur. The most common of these is one whicb Dunham has named Bacillus coli communior, because of the -fact that he believes it to be more abundant 4n the human and animal intestine than is coli communis itself. This bacillus possesses all the cardinal characteristics of the colon group. It is a Gram-negative bacillus, moderately motile, non-sporulating, and morphologically indistinguish- able from the communis variety. It does not liquefy gelatin, it produces indol from pepton, coagulates and acidifies milk, and grows characteristically upon. agar and potato. It differs from Bacillus coli communis in that it produces acid and gas from saccharose as well as from dextrose and lactose, whereas the former does not form acid or gas from saccharose. CHAPTER XXVII BACILLI OF THE COLON-TYPHOID-DYSENTERY GROUP (Continued) THE BACILLUS OF TYPHOID FEVER {BacUlus typhosus, Bacillus typhi abdominalis) Typhoid fever, because of its wide distribution and almost con- stant presence in most communities, has from the eariiest days been the subject of much etiological inquiry. A definite conception as to its infectiousness and transmission from case to case was formed as early as 1856 by Budd.^ But it was not until 1880 that Eberth ^ discovered in the spleen and mesenteric glands of typhoid- fever patients who had come to autopsy, a bacillus which we now know to be the cause of the disease. Final proof of such an etiological connection was then brought by Gaffky,^ who not only saw the bacteria referred to by Eberth, but succeeded in obtaining them in pure culture and studying their growth characteristics. Morphology and Staining. — The typhoid bacillus is a short rod from 1-3.5 /i in length with a varying width of from .5 to .8 At. In appear- ance it has nothing absolutely distinctive which could serve to differen- tiate it from other bacilli of the typhoid-colon group, except that it has a general tendency to greater slenderness. Its ends are rounded without ever being club-shaped. Contrary to the descriptions of the earliest observers, typhoid bacilli do not form spores. They are actively motile and have twelve or more flagella peripherally arranged. The bacilli stain readily with the usual anilin dyes. Stained by Gram's method, they are decolorized. Cultivation. — Bacillus typhosus is easily cultivated upon the usual laboratory media. It is not delicately susceptible to reaction, but will grow well upon media moderately alkaline or acid. It is an aerobic and facultative anaerobic organism, when the proper nutriment is present. Upon agar plates growth appears within eighteen to twenty-four hours » Budd, " Intestinal Fever/' Lancet, 1856. _ 2 Eberth, Virch, Archiv, 81, 1880, and 83, 1881. » Gaffky, Mitt. a. d. kais, Gesimdheitsamt, 2, 1884. 399 . 400 PATHOGENIC MICROORGANISMS as small grayish colonies at first transparent, later opaque. Upon agar slants growth takes place in a uniform layer. There is nothing charac- teristic about this growth to aid in differentiation. In broth, the typhoid bacillus grows rapidly, giving rise to an even clouding, rarely to a pellicle. Upon gelatin, the typhoid bacillus grows readily and does not lique- fy the medium. In stabs, growth takes place along the entire extent of the stab and over the surface of the gelatin in a thin layer. In geW tin plates the growth may show some distinction from that of other mem- bers of this group, and this medium was formerly much used for isolation Fig. 86. — Bacillus typhosus, from twenty-four-hour culture on agar. of the bacillus from mixed cultures. Growth appears within twenty- four hours as small, transparent, oval, round, or occasionally leaf-shaped colonies which are smaller, more delicate, and more transparent than contemporary colonies of the colon bacillus. They do not, however, show any reliable differential features from bacilli of the dysentery group. As the colonies grow older they grow heavier, more opaque, and lose much of their early differential value. ^: \ v On potato the growth of typhoid bacilli is distinctive, and this medium was recommended by Gaffky ^ in his early researches for purposes of ^Gaffky,\oc. cit. BACILLUS OF TYPHOID FEVER 401 identification. On it typhoid bacilli, after twenty-four to forty-eight hours, produce a hardly visible growth, evident to the naked eye only by a slight moist glistening, an appearance which is in marked contrast to the grayish-yellow or even brown and abundant growth of the colon bacilli. If the potato medium is rendered neutral or alkaline, this distinction disappears, the typhoid bacillus growing more abundantly. In milk, typhoid bacilli do not produce coagulation. In litmus-milk, during the first twenty-four hours, the color is changed to a reddish or violet tinge by the formation of acid from the small quantities of mono- FiG. 87. — Bacillus typhosus, showing flagella. (After Frankel and Pfeiffer.) saccharid present. Later the color becomes deep blue from the forma- tion of alkali. In Dunham's pepton solution no indol is produced. According to Peckham, however, continuous cultivation in rich pepton medi^ may lead to eventual indol formation by typhoid bacilli. This fact appears to have no bearing on the value of the indol test, as indol is never formed under the usual cultural conditions. In dextrose, mannite, lactose, and saccharose broth, the typhoid bacil- lus produces no gas. A comparative summary of the action of other bacilli of this group in these sugar media will be given in the final dif- ferential table on page 443. 402 PATHOGENIC MICROORGANISMS f^ Tested for its power to form acid from sugars commonly used in differential tests, typhoid bacilli form acid, but no gas, on the mono- saccharides, on mannit, maltose and dextrin, but not on lactose and saccharose. (See Table, p. 44.) In the Hiss tube medium (see section on Media, page 133), the typhoid bacillus within eighteen to twenty-four hours produces an even clouding by virtue of its motility, but does not form gas. In contradis- tinction to this, dysentery bacilli grow only along the line of inocula- FiG. 88. — Surface Colony of Bacillus typhosus on Gelatin. (After Heim.) tion, while bacilli of the colon group move in irregular sky-rocket-like figures away from the stab, at the same time breaking up the medium by the formation of gas-bubbles. Some actively motile colon bacilli cloud the medium, but the ruptures caused by the gas are always evident. The differentiation of the typhoid bacillus in pure culture from similar microorganisms by means of its growth upon media has been the sub- ject of many investigations. It is neither practicable nor desirable to enumerate all the various media which have been devised and reported. BACILLUS OF TYPHOID FEVER 403 The aim has been chiefly the differentiation of typhoid bacilli from the bacilli of the colon group, and most of the media have been devised with this end in view. (See section on Media.) Rothberger ^ has devised a mixture of glucose agar to which is added one per cent of a saturated aqueous solution of neutral-red. Shake-cul- .tures or stab-cultures are made in tubes of this medium. The typhoid bacillus causes no changes in it, while members of the colon group, by reduction of the neutral-red, decolorize the medium and produce gas by fermentation of the sugar. Utilizing the fact that bile-salts are precipitated in the presence of acids, Macconkey devised a medium composed of sodium glycocholate, pepton, lactose, and agar (the composition of this medium is given on page 138), in which Bacillus typhosus grows without causing nmch change, but distinct clouding results from the growth of the colon bacillus which, producing acid from the lactose, causes precipitation of the bile- salts. On Wurtz's lactose-litmus-agar (see page 129) the typhoid bacillus produces no acid, but eventually deepens the purple color to blue; the colon bacillus produces acid and in stab-cultures gas bubbles and the color changes to red. In Barsiekow's (see page 139) lactose-nutrose-litmus mixture the typhoid bacillus causes no change, while the eolon bacillus produces coagulation and an acid reaction. Cultural Differences Within the Typhoid Group. — ^Recent work by H. Weiss in this laboratory has shown that not all typhoid bacilli are culturally alike, there being two distinct groups, one which ferments xylose and the other which does not. This work is being elaborated. Since there are also antisrenic differences it may be necessary in the future to speak rather of a typhoid group than of the typhoid bacillus. Biological Considerations. — The typhoid bacillus is an aerobic and facultatively anaerobic organism growing well both in the presence and in the absence of oxygen when certain sugars are present, showing a slight preference, however, for well aerated conditions. It grows most luxu- riantly at temperatures about 37.5° C, but continues to grow within a range of temperature lying between 15° and 41° C. Its thermal death point, according to Sternberg, is 56° C. in ten minutes. It remains alive in artificial cultures for several months or even years if moisture is sup- plied. In carefully sealed agar tubes Hiss has found the organisms ^Rothberger, Cent. f. Bakt., xxiv, 18Q8. 404 PATHOGENIC MICROORGANISMS alive after thirteen years. In natural waters it may remain alive as long as thirty-six days, according to Klein/ In ice, according to Prud- den,^ it may remain alive for three months or over. Against the ordi- nary disinfectants, the typhoid bacillus is comparatively more resistant than some other vegetative forms. It is killed, however, by 1 : 500 bichlorid or five-per-cent carbolic acid within five minutes. Pathogenicity. — In animals, some early investigators to the contrary, typhoidal infection does not occur spontaneously and artificial inocula- tion with the typhoid bacillus does not produce a disease analogous to typhoid fever in the human being. Frankel ^ was able to produce intes- tinal lesions in guinea-pigs by injection of the bacilli into the duodenum, and recovered the bacteria from the spleens of the animals after death, but the disease produced was in no other respect analogous to typhoid fever in the human being. It is probable that typhoid bacilli injected into animals do not multiply extensively and that most of the symp- toms produced are due to the endotoxins liberated from the dead bac- teria. In corroboration of this view is the observation that inoculation with dead cultures is followed by essentially the same train of symp- toms as inoculation with live cultures."* The injection of large doses into rabbits or guinea-pigs intravenously or intraperitoneally is usually followed by a rapid drop in temperature, often by respiratory em- barrassment and diarrhea. Occasionally blood may be present in the stools. According to the size of the dose or the weight of the ani- mal, death may ensue within a few hours, or, with progressive emacia- tion, after a number of days, or the animal may gradually recover. Welch and Blachstein ^ have shown that typhoid baciUi injected into the ear vein of a rabbit appear in the bile and may persist in the gall- bladder for weeks. Typhoid bacilli isolated from different sources may show considerable variations in virulence and toxicity. Doerr,^ Koch,^ Morgan,^ and more recently Johnston^ have all confirmed this, the last named showing that the typhoid bacillus could 1 Klein, Med. Officers' Report, Local Govern. Bd., London, 1894. 2 Prudden, Med. Rec, 1887. 3 Frankel, Cent, f . klin. Med., 10, 1886. * Petruschky, Zeit. f. Hyg., xii, 1892. ^ Welch and Blachstein, Bull. Johns Hop. Hosp., ii, 1891. ' 8 Doerr, Centralbl. f . Bakt., 1905. ' Koch, Zeitschr. f . Hyg., 1909. ^Morgan, Jour, of Hyg., 1911. » Johnston, Jour, of Med. Res., xxvii, 1912. ' BACILLUS OF TYPHOID FEVER 405 not only remain latent for a long time in the gall-bladder of rabbits, but would appear in the blood stream with considerable regularity after the seventh or ninth day, and persist for as long as 125 days. Gay and Claypole^ have been able to produce the carrier state in rabbits with great regularity by growing the typhoid cultures used for inoculation upon agar containing 10 per cent defibrinated rabbit's blood. Such cultures are not as readily agglutinated by immune serum as are those grown on plain agar, and it may well be that they have acquired a certain degree of resistance to the serum antibodies which renders them more competent to survive in the body of the rabbit. Gay has used rabbits inoculated with such cultures for the determination of the efficacy of his sensitized vaccines. In man the large majority of typhoid infections take the form of the disease clinically known as typhoid fever. For a description of the clinical course and pathological lesions of the disease, the reader is referred to the standard text-books of medicine and pathology. During the course of the disease, and during convalescence, the bacilli may be cultivated from the circulating blood, the rose spots, the feces, the urine, and in exceptional cases from the sputum. At autopsy the bacilli may be obtained from these sources as well as from the lesions in the intestine, the spleen, and often from the liver, kidneys, and from the gall-bladder. Though formerly regarded as primarily an intestinal disease, recent investigations have brought convincing proof that the disease is in its inception actually a bacteriemia. It is not unlikely that the intestinal lesions are largely the result of toxic products which are excreted through the intestinal wall. Typhoid Bacilli in the Blood during the Disease. — The investigations of many workers have shown that typhoid bacilli are present in the circulating blood of practically all patients during the early weeks of the disease. Series of cases have been studied by Castellani,^ Schottmtil- ler,^ and many others. More recently Coleman and Buxton^ have reported their researches upon 123 cases, and have at the same time analyzed all cases previously reported. Their analysis of blood cultures taken at different stages in the disease is as follows: ^Gay and Claypole, Arch, of Inf. Med., Dec, 1913. 2 CasteUani, Riforma medica, 1900. ^ Schottmueller, Deut. med. Woch., xxxii, 1900, and Zeit. f. Hyg., xxxvi, 1901. * Coleman and BuxtoUf Am. Jour, of Med. Sci., 133, 1907. 406 PATHOGENIC MICROORGANISMS Of 224 cases during first week, 89 per cent were positive. Of 484 cases during second week, 73 per cent were positive. Of 268 cases during third week, 60 per cent were positive. Of 103 cases during fourth week, 38 per cent were positive. Of 58 cases after fourth week, 26 per cent were positive. The technique recommended by Coleman and Buxton for obtaining blood cultures is that recommended by Conradi/ slightly modified. The blood is taken into flasks each containing about 20 c.c. of the following mixture: Ox-bile 900 c.c. Glycerin 100 c.c. Pepton 20 grams. About 3 c.c. of blood are put into each flask. The ox-bile, besides preventing coagulation, may possibly neutralize the bactericidal sub- stances present in the drawn blood. The flasks are incubated for eigh- teen to twenty-four hours, at the end of which time streaks are made upon plates of lactose-litmus-agar and the organisms identified by agglutination or by cultural tests. European workers have generally preferred to make high dilution of the blood in flasks of bouillon, small quantities of blood, 1 to 2 c.c, being mixed with 100 to 150 c.c. of nutrient broth. Epstein 2 has reported excellent results from mixing the blood in considerable concentration with two-per-cent glucose agar and pouring plates. The writers in hospital work have had equally good results with the bile medium and with broth in flasks, rather less uniform but still satis- factory results with the plating method. In general it may be said that any one of these methods carried out with reasonable accuracy may be satisfactorily employed. Typhoid Bacilli in the Stools. — The examination of the stools for typhoid bacillus is performed for diagnostic purposes chiefly in obscure cases. It may, furthermore, furnish information of great hygienic importance. Thus Drigalski ^ and Conradi have succeeded in isolat- ing typhoid bacilli from the stools of ambulant cases so mild that they were not clinically suspected. It la oy means of such examina- tions that the so-called typhoid-carriers are detected, cases which, 1 Conradi, Deut. med. Woch., xxxii, 1906. ^Epstein, Proc. N. Y. Path. Soc, N. S., vi, 1906. * Drigalski and Conradi, Zeit. f . Hyg., xxxix, 1902. BACILLUS OF TYPHOID FEVER 407 though perfectly well themselves, may be a means of spreading the disease. Such cases have been known to harbor the bacilli for periods as long as several years. The examination itself is fraught with difficulties, owing to the pre- ponderating numbers of colon bacilli found in all feces and the difficulty of isolating the typhoid bacilli from such mixtures. Reviewing the data collected by a number of investigators, it seems probable that the bacilli do not appear in the stools, at least in numbers sufficient for recognition, much before the middle of the second week, or, in other words, as pointed out by Hiss, about the time that the intestinal lesions are well ad- vanced and ulceration is occurring. Thus Wiltschour ^ could not determine their pres- ence before the tenth day; Redtenbacher,^ in reviewing the statistics, states that in a majority of cases the bacilH first appear toward the end of the second week, and Horton-Smith ^ could not find the bacilli be- fore the eleventh day. Hiss,^ in an investi- gation of the same subject, obtained the following results: First to tenth day, inclusive, twenty- fiq. 89.— Bacillus coli. eight cases examined; typhoid bacilli isolated Deep colonies on Hiss plate from three; percentage of positive cases medium. 10.7 per cent. Eleventh to twentieth day, inclusive, forty-four cases examined; ty- phoid bacilli from twenty-two; percentage of positive cases 50 per cent. Twenty-first day to convalescence, sixteen cases examined; typhoid- bacilli isolated from thirteen; percentage of positive cases 81.2 per cent. The difficulties encountered in such examinations have led to the development of a large number of methods. The first method which yielded successful results was that of Eisner,^ who employed a potato- extract gelatin containing one per cent of potassium iodid, a medium which prevented the growth of many intestinal bacteria, allowing only 1 Wiltschour, Cent. f. Bakt., 1890. — 2 Redtenbacher, Zeit. f . klin. Med., xix, 1891. * Horton-Smith, Lancet, May, 1899. * Hiss, Med. News, May, 1901. 6 Eisner, Zeit. f . Hyg., xxi, 1895. 0 1 o Of ^ ■%«' ■ 0 408 PATHOGENIC MICROORGANISMS colon, typhoid, and a few others to develop. This medium is at present rarely used. Hiss ^ has employed with success an agar-gelatin mixture containing one per cent of glucose, the preparation of which has been described in detail in the section on media. The actual technique of the test is as follows: One to two loopfuls of feces are transferred to a tube of broth, making the broth fairly cloudy. From this emulsion five or six plates are made by transferring in series one to five loopfuls of the emulsion to tubes containing the melted plate medium, and then pouring the con- tents of these tubes into Petri dishes. These dishes, after the medium Fig. 90. — ^Bacillus typhosus. Deep colonies in Hiss plate medium. has hardened, are placed in an incubator at 37° C, and allowed to re- main for eighteen to twenty-four hours, when they are ready for examina- tion. If typhoid bacilli are present they will be found as small, usually glistening colonies with a fringe of threads growing out like flagella from their peripheries (see Figs. 90 and 91). These colonies are smaller and quite distinct from those of colon bacilU, which are heavier and darker and do not display the fringing threads. Suspicious colonies may be fished and transferred to the Hiss tube medium (see page 133) or iden- tified by other reliable methods. A method which has been found useful, especially in Europe, is that in which smears of diluted feces are made upon large plates of the Conradi-Drigalski medium. (For preparation see page 135.) The principles underlying the use of this medium are the formation of acid from the lactose by the colon bacilli and the inhibition of cocci and many other bacteria by the crystal-violet. In practice, an emul- 1 Hiss, Jour, of Exp, Med., ii, 1897; Med. News, May, 1901; and Jour. Med. Res., N. S., iii, 1902. BACILLUS OF TYPHOID FEVER 409 sion is made of a loopful of feces in a tube of broth. Into this is dipped a bent glass smearing rod, the excess of fluid is allowed to drip off, and smears are made upon plates of the medium, several plates being smeared without redipping the rod. Colonies of the colon bacillus on these plates will appear opaque, comparatively large, and will produce an acid reaction with consequent reddening of the medium. Typhoid colonies will be smaller, transparent, and without acid formation. These colonies are fished and the microorganisms may Fig. 91. — Bacillus typhosus. Colony in Hiss plate medium, highly magnified. be identified by agglutination or by stab cultures in the Hiss tube medium. The malachite-green media of Loeffler and others have found less general use than was originally expected, because of the difficulty in obtaining uniform preparations of malachite-green. Peabody and Pratt ^ have applied the principle of colon-bacillus in- hibition by malachite-green, by adding this dye to broth in the manner described in the section on media (page 137), planting the feces directly into this broth, and, after incubation for several hours, making smears from these tubes upon plates of the Conradi-Drigalski medium. I As a routine method for isolation of the bacilli from stools we our- 1 Peabody and Pratt, Boston Med. and Surg. Jour., 1908. 27 410 PATHOGENIC MICROORGANISMS selves use largely the Endo fuchsin-agar or Kendall's modified Endo medium. (See p. 135.) Emulsions of feces are made in tubes of ordinary broth or salt solution and smears of this emulsion are made upon several large plates of the fuchsin-agar by means of a glass smearing rod. The colonies of Bacillus coli, after eighteen or more hours of incubation, will be found to have brought back a deep red color to the medium, whereas the typhoid colonies are small, more transparent, and have left the medium uncolored. An excellent medium recently devised is that of Krumwiede. (See p. 136.) In all cases where plates are prepared from broth emulsions of feces, it is desirable t9 allow the emulsion to stand at incubator temperature Fig. 92. — Colon and Typhoid Colonies in Hiss Plate Medium. from stool. Note the small thread-forming typhoid colonies.) (Planted for an hour. Subsequent removal of fluid from the upper layers of the medium is likely to bring away a comparatively larger number of the organisms. The methods given above do not exhaust the records of work done upon this problem. It is not satisfactory to compare any two methods as to practical value, since all of them require famiharity with organisms and media. In fact, it may be said that all of the methods given are satisfactory if consistently employed by a worker who has become thoroughly accustomed to the peculiarities of the typhoid colonies upon the medium with which he is working. In all of these methods when suspicious colonies are found they are identified morphologically and transplanted to such media as the Russell double-sugar agar, or the Hiss tube medium. For rapid diagnosis BACILLUS OF TYPHOID FEVER 411 agglutination may be done in a strong immune serum, either directly from the colony by the hang-drop method, or better macroscopically from the growth in the transplant. Typhoid Bacilli in the Urine. — Careful investigation has revealed typhoid bacilli in the urine in about twenty-five per cent of all patients. Neimiann ^ discovered the bacilli in eleven out of forty-six and Kar- linsky ^ in twenty-one out of forty-four cases. Investigations by Pe- truschy,^ Richardson,^ Horton-Smith,^ Hiss,^ and others have confirmed these results. In general the bacilli have not been found before the fifteenth day of the disease, and examination of the urine, therefore, can be of little early diagnostic value. A series of seventy-five cases examined by Hiss before the fourteenth day of the disease did not once reveal typhoid bacilli in the urine. On the other hand, they have been found to be present for weeks, months, and, in isolated cases, for years after convalescence, the examination thus having much hygienic im- portance. They are probably present in about twelve per cent of cases during the early days of convalescence. In most of these, albumin is present in the urine in considerable quantities. The bacilli usually appear and disappear with the albuminuria. \ An obstinate cystitis caused by typhoid bacilli may follow in the path of typhoid fever. Such cases have been reported by Blumer,^ Richardson,^ and others. Suppurative processes in the kidneys are less frequent. It is noteworthy, also, that in the course of, and fol- lowing, typhoid fever there often occurs voiding of Bacillus coli with the urine. This may obstinately persist for considerable periods after convalescence. The reasons for this are not entirely clear. Typhoid Carriers and Typhoid Bacilli in the Gall-Bladder. — Typhoid bacilli have been frequently observed in the gall-bladder at autopsy. They have also been found present in this organ, at operations for cholecystitis, months and years after the occurrence of typhoid fever. Miller * has reported a case in which typhoid bacilli were present in the gall-bladder seven years after the disease; v. Dungern^^ has cultivated them from an inflamed gall-bladder fifteen years after the disease. Zinsser has had occasion^^ to observe a case in which an operation for 1 Neumann, Berl. klin. Woch., xxvii, 1890. « Hiss, Med. News, May, 1901. 2 Karlinsky, Prag. med. Woch., xv, 1890. "^ Blunter, Johns Hopk. Hosp. Rep., 5, 1895. ' Petruschy, Cent. f. Hyg., xxiii, 1898. ^ Richardson, loc. cit. ^ Richardson, Jour. Exp. Med., 3, 1898. ^ Miller, Johns Hopk. Hosp. Bull., 1898. ^ Horton-Smith, Lancet, May, 1899. " v. Dungern, Miinch. med. Woch., 1897. » Zimser, Proc. N. Y. Pathol. Soc, 1908. 412 PATHOGENIC MICROORGANISMS gall-stone seventeen years after the occurrence of typhoid fever revealed the presence of the bacilli in the gall-bladder. In such cases typhoid bacilli may be constantly discharged from the intestine with the feces and prove a menace to the health of the community. An extremely in- teresting example of such a typhoid carrier has been carefully studied and reported by Park.^ Typhoid carriers are much more common than formerly supposed. Lentz 2 believes that about 4 per cent of all cases become chronic car- riers and Goldberger, averaging the cases of other workers, calculates that of 1,782 cases of typhoid, 53 or about 3 per cent became chronic carriers. This problem therefore is of the utmost sanitary import- ance. Even when detected the cure of such carriers is very difficult, and perhaps, in some cases impossible without cholecystectomy. Vac- cine treatment has not been encouraging and no other form of uni- formly successful treatment has so far been devised. Typhoid Bacilli in the Rose Spots. — Neufeld ^ obtained positive re- sults in thirteen out of fourteen cases. According to his researches and those of Frankel,^ the bacilli are localized not in the blood, which is taken when the rose spots are incised, but are crowded in large num- bers within the lymph spaces. Typhoid Bacilli in the Sputum. — In rare cases typhoid bacilli have been found in the sputum of cases complicated by bronchitis, broncho- pneumonia, and pleurisy. Such cases have been reported by Chante- messe and Widal,^ Frankel,® and a number of others. Empyema, when it occurs in connection with such cases, is usually accompanied by a mixed infection. From a hygienic point of view the spread of typhoid fever by means of the sputum is probably of rare occurrence. Suppurative Lesions Due to Typhoid Bacillus. — In the course of typhoid convalescence or during the latter weeks of the disease, sup- purative lesions may occur in various parts of the body. The most frequent localization of these is in the periosteimi, especially of the long bones, and in the joints. A large number of such lesions have been described by Welch, Richardson,^ and others. They usually take the form of periosteal abscesses, often located upon the tibia, occurring ^Park, "Pathogenic Bacteria," N. Y., 1908. 2 Lentz, Hyg. Rundscha,u, vol. 16, 1906. ' Neufeld, Zeit. f. Hyg., xxx, 1899. * Frankel, Zeit. f. Hyg., xxxiv, 1900. ^ Chantemesse and Widal, Arch, de physiol. norm, et path., 1887. * Frankel, Deut. med. Woch., xv and xvi, 1899. ' Richardson, Jour. Boston Soc. Med. Sci., 5, 1900. BACILLUS OF TYPHOID FEVER 413 either late in the disease or months after convalescence, and are char- acterized by very severe pain. Osteomyelitis may also occur, but is comparatively rare. Subcutaneous abscesses and deep abscesses in the muscles, due to this bacillus, have been described by Pratt. ^ Synovitis may also occur. Meningitis, due to the typhoid bacillus, occurs not infrequently, usually during convalescence from typhoid fever. A case of primary typhoid meningitis has been reported by Farnet.^ Peritoneal abscesses, due to the typhoid bacillus, have been re- ported. The writer ^ has reported a case in which typhoid bacilli were found free in the peritoneal cavity during typhoid fever without per- foration of the gut. Isolated instances of typhoid bacilli in abscesses of the thyroid and parotid glands and in brain abscesses have been observed. Typhoid Fever without Intestinal Lesions. — A considerable number of cases have been reported in which typhoid bacilli have been isolated from the organs after death or from the secretions during life of pa- tients in whom the characteristic lesions of typhoid fever have been backing. Most of these cases must be regarded as true typhoid septi- cemias. In some cases the bacilli were isolated from the spleen, liver^ or kidneys; in others, from the urine or the gall-bladder. In a case observed by Zinsser the bacilli were isolated from an infarct of the kidney removed by operation. In this case the clinical course of the disease had pointed only toward the existence of an indefinite fever ac- companied by symptoms referable to the kidneys. The Widal test, however, was positive. An excellent summary of such cases, together with several personally observed, has been given by Flexner."^ Hygienic Considerations. — Although typhoid fever is frequently spoken of as an epidemic disease, it is, more truly, endemic in character in almost all parts of the world, but subject to occasional epidemic ex-, acerbations. i In the larger communities of the temperate zones these epidemics take place chiefly in the autumn and are circumscribed by the distribution of a water or milk supply. Since the disease never occurs except by transmission, directly or in- directly from a previous case, it is amenable more than most other mal- adies to sanitary regulation, and it may be said that extensive preva- lence of typhoid in a large community is a consequence of defect in the 1 Pratt, Jour. Boston Soc. Med. Sci., 3, 1899. 2 Famet, Bull, de la soc. med. des hop. de P., 3, 1891. » Zinsser, Proc. N. Y. Path. Soc, 1907. * Flexner, Johns Hopk. Rep., 5, 1896. 414 PATHOGENIC MICROORGANISMS system of sanitation. The disease is acquired by ingestion of the spe- ^ cifie bacteria. Infection by other channels has not been demonstrated. Prophylactic measures in typhoid fever, therefore, should begin with the isolation of the patient and the disinfection of excreta, dis- charges, linen, and all utensils which have been in contact with the patient. The bacilli leave the body in the feces and the urine and the dangers of contaminatiori, by these substances, of all objects in imme- diate contact with the patient are considerable. Excreta should there- fore be either mixed with boiling water or chemically disinfected, pref- erably by means of thoroughly mixing with carbolic acid, lysol, or a solu- tion of freshly slaked lime, and, if possible, destroyed by burning. Linen, tableware, and eating utensils should be soaked in similar solu- tions and boiled. The observance of such measures, furthermore, should not be discontinued until bacteriological examination has demon- strated the absence of the bacilli from feces and urine. Disregard of this last precaution may well be one of the main causes of the endemic persistence of the disease in large cities — especially considered in the light of our recent knowledge of 'Hyphoid carriers." Typhoid fever, in the large majority of cases, is transmitted by the agency of water. In an analysis of six hundred and fifty tjrphoid epidemics Schiider ^ found four hundred and sixty-two reported, upon reasonable evidence, as originating from water. The technical difficul- ties attending the isolation of typhoid bacilli from contaminated water have prevented actual bacteriological proof in most epidemics; never- theless, indirect evidence of pollution of the suspected water-supply, correspondence of the distribution of this supply with that of the dis- ease, and reduction of typhoid morbidity upon the substitution of an uncontaminated supply are sufficiently convincing to remove reasonable doubt. Added to this is our knowledge, from the experiments of Jor- dan, Russell, and Zeit ^ and others, that typhoid bacilli may remain alive in natural waters for as long as five days. Prudden has demon- strated that the bacilli may survive freezing as long as three months. Next to water, the most important source of typhoid fever is con- taminated milk. In the summary by Schiider,^ one hundred and ten of the four hundred and sixty epidemics were attributable to milk. No visible modifications in milk occur, which makes this source es- pecially insidious. Contamination of milk has been traceable to water used in washing cans or to dairy attendants. * Schiider, Zeit. f. Hyg., xxxviii, 1901. 2 Jordan, Russell, and Zeit, Jour, of Inf. Dis., 1, 1904. ^ Schiider, loc. cit. BACILLUS OF TYPHOID FEVER 415 Excluding water and milk, all remaining causes of typhoid dissemi- nation constitute about twelve per cent and are found chiefly in the use of vegetables contaminated from infected soil, and other food prod- ucts. Recently Conn ^ suggested that oysters grown in waters close to sewage discharges may be the means of typhoid transmission. An epidemic at Wesleyan University was traced to this cause. Experi- ments by Foote ^ demonstrated that typhoid bacilli may be found alive within oysters for three weeks after they disappeared from the surrounding water. The importance of this mode of infection is un- certain. Rosenau, Lumsden, and Castle ^ found it to be negligible in the District of Columbia. Indirect contamination of food and water by the intermediation of flies and other insects has been emphasized by Veeder * and is unques- tionably of great importance. Typhoid Carriers. — One of the important factors in the spread of typhoid fever is the existence of a considerable number of ''car- riers ' ' in every community. These are individuals who harbor typhoid bacilli in their intestinal canals, probably with a nidus in the gall bladder. They may intermittently or continually discharge typhoid bacilli with the feces and therefore be a constant menace to others. There are now a considerable number of well-known cases where series of small epidemics have been started by carriers, and in the mobiliza- tion of armies the incorporation of carriers with troops may be a seri- ous menace. Carriers may be discovered by isolation of the bacilli from the feces as described in another place. Poisons of the Tsrphoid Bacillus. — The investigation of the toxic products of the typhoid bacillus has occupied the attention of a large number of workers. The first to do experimental work upon the sub- ject was Brieger ^ soon after the discovery and cultivation of the micro- organism. That toxic substances can be obtained from typhoid cultures is beyond question. There is, however, a definite difference of opinion as to whether these poisons are so-called endotoxins only, or whether they are in part composed of soluble toxins comparable to those of diphtheria and tetanus, follovnng the injection of which antitoxic sub- stances may be formed. The evidence so far seems to bear out the original contention of 1 Conn, Med. Record, Dec, 1894. 2 poote, Med. News, 1895. ' Rosenau, Lumsden, and Castle, Bull. 52, Hyg. Lab. U. S. Pub. Health Service, 1908. * Veeder, Med. Record, 45, 1898. ^ Brieger, Deut. med. Woch., xxvii, 1902. 416 PATHOGENIC MICROORGANISMS Pfeiffer/ who first advanced the opinion that the poisonous substances are products of the bacterial body set free by destruction of the bacteria by the lytic substances of the invaded animal or human being. These poisons, when injected into animals for purposes of immunization, in Pfeiffer's experiments, did not incite the production of neutralizing or antitoxic bodies, but of bactericidal and lytic substances. That these endotoxins constitute by far the greater part of the toxic products of the typhoid bacillus can be easily demonstrated in the laboratory, by the simple experiment of filtering a young typhoid culture (eight or nine days old) and injecting into separate animals the residue of bacilli and the clear filtrate respectively. In such an experiment there will be little question as to the overwhelmingly greater toxicity of the bacillary bodies as compared with that of the culture filtrate. On the other hand, if such cultures, especially in alkaline media, are allowed to stand for several months and the bacilh thus thoroughly extracted by the broth, the toxicity of the filtrate is found to be greatly increased. Nevertheless, more recent experiments by Besredka,^ Macfadyen,^ Kraus and Stenitzer,* and others have tended to show that, together with such endotoxic substances, typhoid bacilli may produce a true toxin which is not only obtainable by proper methods from compara- tively young typhoid cultures, but which fulfils the necessary require- ment of this class of poisons by producing in treated animals a true antitoxic neutralizing body. The typhoid endotoxins may be obtained by a variety of methods. Hahn^ has obtained what he calls " typhoplasmin '^ by subjecting them to a pressure of about four hundred atmospheres in a Buchner press. The cell juices so obtained are cleared by filtration. Macfadyen has obtained typhoid endotoxins by triturating the bacilli after freezing them with liquid air and extracting in 1 : 1,000 potassium hydrate. Besredka obtained toxic substances by emulsifying agar cultures of bacilli in salt solution, sterilizing them by heating to 60° C. for about one hour, and drying in vacuo. The dried bacillary mass was then ground in a mortar and washed in sterile salt solution which was again heated to 60° C. for two hours. The remnants of the bacterial 1 Pfeiffer, Deut. med.Woch., xlviii, 1894; Pfeiffer und Kolle, Zeit. f. Hyg., xxi, 1896. 2 Besredka, Ann. de Tinst. Pasteur, 1895, 1896. ^Macfadyen and Rowland, Cent. f. Bakt., I, xxx, 1901; Macfadyen, Cent, f. Bakt., I, 1906. * Kraus und Stenitzer, Quoted from "Handb.d. Tech., "etc., 1, Fischer, Jena, 1907. « Hahn, Miinch. med. Woch., xxiii, 1906. C:. BACILLUS OF TYPHOID FEVER ^17 bodies settle out and the slightly turbid supernatant fluid contains the toxic substances. Vaughan ^ has obtained poisons from typhoid bacilli by extracting at 78° C. with a two-per-cent solution of sodium hydrate in absolute alcohol. In this way he claims to separate by hydrolysis a poisonous and a non-poisonous fraction. He claims, moreover, that this poison- ous fraction is similar to the poisons obtained in the same way from Bacillus coli and the tubercle bacillus, and other proteid substances, beheving that the specific nature of such proteids depends upon the non-toxic fraction. A simple method of obtaining toxins from typhoid bacilli is carried out by cultivating the microorganisms in meat-infusion broth, rendered alkaline with sodium hydrate to the extent of about one per cent. The cultures are allowed to grow for two or three weeks and then steril- ized by heating to 60° C. for one hour, and allowed to stand for three or four weeks at room temperature. At the end of this time the cul- tures may be filtered through a Berkefeld or Pasteur-Chamberland filter and will be found to contain strong toxic substances. The accounts concerning the thermostability of the various toxins obtained are considerably at variance. In general, corresponding with other endotoxins, observers agree in considering them moderatiely re- sistant to heat, rarely being destroyed at temperatures below 70° C. Intravenous inoculation of rabbits with typhoid endotoxins, if in sufficient quantity, produces, usually within a few hours, a very marked drop in temperature, diarrhea, respiratory embarrassment, and death. If given in smaller doses or by other methods of inoculation — ■ subcutaneous or intraperitoneal — rabbits are rendered extremely ill, with a primary drop in temperature, but may live for a week or ten dayS, and die with marked progressive emaciation, or may survive. Guinea- pigs and mice are susceptible to the endotoxins, though somewhat less so than rabbits. • Immunity in Typhoid Fever. — As a rule, one attack of typhoid fever protects against subsequent ones. Although exceptions to this rule may occur, they are so rare that the history of a previous attack of this disease practically excludes its consideration in the diagnosis of any obscure condition. Animals may be actively immunized by the injection of typhoid bacilli in gradually increasing doses. In actual practice, this is best 1 Vaughan, Am. Jour, of Med. Sci., 136, No. 3, 1908. 418 PATHOGENIC MICROORGANISMS accomplished by beginning with an injection of about 1 c.C. of broth culture heated for ten minutes at 60° in order to kill the bacilli. After five or six days, a second injection of a larger dose of dead bacilli is administered; at similar intervals, gradually increasing doses of dead bacilli are given and finally considerable quantities of a living and full)'' virulent culture may be injected without serious consequences to the animal. While this method is convenient and usually successful, it is also possible to obtain satisfactory immunization by beginning with very small doses of living microorganisms, according to the early method of Chantemesse and Widal,^ and others. Such active immunization, successfully carried out upon rabbits and guinea-pigs, within a short time after the discovery of the typhoid bacil- lus, was believed to depend upon the development of antitoxic sub- stances in immunized animals. This point of view, however, was not long tenable, and was definitely disproven by the investigations of Pfeif- fer and KoUe ^ in 1896. These investigators, as well as a large number of others working subsequently, have shown satisfactorily that there are present in the blood serum of typhoid-immune animals and human beings, bacteriolytic, bactericidal, and agglutinating substances, and to a lesser extent, precipitating and opsonic bodies. Bactericidal and Bacieriolytic Substances. — The bacteriolytic sub- stances in typhoid-immune serum may be demonstrated either by the intraperitoneal tachnique of Pfeiffer or in vitro. In the former experi- ment a small quantity of a fresh culture of typhoid bacilli is mixed with the diluted immune serum and the emulsion injected into the peritoneal cavity of a guinea-pig. Removal of peritoneal exudate with a capillary pipette and examination in the hanging drop will reveal, within a short time, a swelling and granulation of the bacteria — the so-called Pfeiffer phenomenon. The test in vitro, as recommended by Stern and Korte,^ may be carried out by adding definite quantities of a fresh agar culture of typhoid bacilli to progres^vely increasing dilu- tions of inactivated immune serum together with definite quantities of complement in the form of fresh normal rabbit or guinea-pig serum. At the end of several hours' incubation at 37.5° C. definite quantities of the fluid from the various tubes are inoculated into melted agar and plates are poured to determine the bactericidal action. Careful colony counting in these plates and comparison with proper controls ' Chantemesse and Widal, Ann, de llnst. Pasteur, 1892. 2 Pfeiffer und Kolle, Zeit. f. Hyg., xxi, 1896. 3 Stern und Korte, Berl. klin. Woch., x., 1904. BACILLUS OF TYPHOID FEVER 419 will not only definitely demonstrate the presence of bactericidal sub- stances in the immune serum, but will furnish a reasonably accurate quantitative estimation. (For these tests see p. 255.) Although normal human serum contains in small quantity substances bactericidal to typhoid bacilH, moderate dilution, 1 : 10 or 1 : 20, of such sei*um will usually suffice to eliminate any appreciable bactericidal action. The bactericidal powers of immune serum, on the other hand, are often active, according to Stem and Korte, in dilutions of over 1 : 4,000 and in one case even of 1 : 4,000,000. The specificity of such reactions gives them a considerable degree of practical value, both in the biological identification of a suspected typhoid bacillus in known serum and in the diagnosis of typhoid fever in the human patient by the action of the patient's serum on known typhoid bacilli. In the publication of Stem and Korte, quoted above, it was found that typhoid patients during the second week often" possess a bactericidal power exceeding 1 : 1,000, whereas the blood of normal human beings was rarely active in dilu- tions exceeding 1 : 50 or 1 : 100. While scientifically accurate, the prac- tical application of bactericidal determinations for diagnosis presents considerable technical difficulties, and gives way to the no less accurate method of agglutination. Agglutinins. — Agglutinins are formed in animals and man inoculated with typhoid bacilli, and in the course of typhoid fever. It was, in fact, while studying the typhoid bacillus that the agglutinins were first dis- covered by Gruber and Durham. In animals, by careful immunization, specific typhoid agglutinins may easily be produced in sufficient quantity to be active in dilution of 1 : 10,000, and occasionally even 1 : 50,000 or over. In the blood of typhoid patients, the agglutinins may often be found in dila- tions of 1 : 100 and over. It is interesting to note that irrespec- tive of the agglutinin contents of any given serum, there may occasionally be noted differences in the agglutinability of various typhoid cultures, a point which is practically important in the choice of a typhoid culture for routine diagnosis work. Weeny ^ has called attention to the fact that bacilli which do not readily agglutinate when directly cultivated from the body, may often be rendered more sensitive to this reaction by several generations of cultivation upon artificial media. Walker has noted ^ a loss of agglutinability if the baciUi » Weeny, Brit. Med. Jour., 1889. « Walker, Jour, of Path, and Bact., 1892; Totsuka, Zeit. f. Hyg., xlv, 1903. 420 PATHOGENIC MICROORGANISMS are cultivated in immune serum. A similar alteration in the agglutin- abilit}^ of typhoid bacilli was noted by Eisenberg and Volk * when they subjected the microorganism to moderate heat or to weak acids such as f HCl. The practical application of agglutination to bacteriological work is found, as in the case of the bactericidal substances, in the identification of suspected typhoid bacilli, and in the diagnosis of typhoid fever. When it is desired to determine by means of agglutination whether or not a given bacillus is a typhoid bacillus, mixtures may be made of young broth cultures, gr preferably of emulsions of young agar cul- tures in salt solution, with dilutions of immune serum. The tests are made microscopically in the hanging-drop preparation or, preferably, macroscopically in small test tubes. In all cases it is desirable first to determine the agglutinating power of the serum when tested against a known typhoid culture. (For detailed technique, see chapter on Technique of Serum Reactions, p. 250.) In scientific investigations, specific agglutinations in high dilutions of immune serum constitute very strong proof of the species of the micro- organism and may often furnish much information as to the biological relationships between similar species. It is found in immunizing ani- mals with any given strain of typhoid bacilli, that there are formed the "chief" or ''major" agglutinins which are specific and active against the species used in immunization, and the "group " or " minor " agglutinins, active also against closely related microorganisms. The following extract from a table will serve to illustrate this point in the case of typhoid and allied bacilli. • Highly Immune Typhoid Serum. 1:100 1:250 1 :500 1 : 1,000 1 : 2,500 B. tvph + + + + + + + + B . paratyph. (Schottmiiller) B. enteritidis , . B. coli communis. . . . The sera of most adult normal animals and human beings usually contain a small amount of agglutinin for these bacilli. Immuniza- tion with the typhoid bacillus, while increasing chiefly the agglutinin ^Eisenberg und Volk, Zeit. f. Hyg., xlv, 1903. BACILLUS OF TYPHOID FEVER ^1 for this bacillus itself, also to a slighter extent increases the group ag- glutinins for other closely allied species. That these group agglu- tinins are separate substances and not merely a weaker manifestation of the action of the typhoid agglutinin itself upon these other micro- organisms, may be demonstrated by the experiments of agglutiniiL absorption. (See section on Agglutinins, page 234.) In the clinical diagnosis of typhoid fever, the phenomenon of agglutination was first utilized by Widal.^ This observer called at- tention to the fact that during the last part of the first or the earlier days of the second week of typhoid fever, as well as later in the dis- ease and in convalescence, the blood serum of patients would cause agglutination of typhoid bacilli in dilutions of 1 : 10, or over, whereas the serum of normal individuals usually exerted no such influence. Upon this basis he recommended, for the diagnosis of the disease, the employment of a microscopic agglutiv"iation test carried out by the usual hanging-drop technique. The reaction of Widal is, at present, widely depended upon for diagnostic purposes and although not uni- versally successful, owing to irregularities in agglutinin formation in some patients and because of differences in agglutinability of the cultures employed, it is nevertheless of much value. The fact that the recent work of Hooker and of Weiss has shown that typhoid bacilli differ in antigenic properties, and may on the basis of agglutinatioii and agglutinine absorption be divided into a number of groups, will necessitate the use of several cultures for Widal reactions. The orig- inal conclusions as to the dilutions of the serum which must be em- ployed, have, however, necessarily been modified. Owing to the; fact that Gruber,^ Stern,^ Frank^l,^ and a number of others have found that occasionally normal serum will give rise to agglutination of typhmd bacilli in dilutions exceeding 1 : 10, it has been found necessary, when- ever making' a diagnostic test, to make several dilutions, the ones most commonly employed being 1:20, 1:40, 1:60, and 1:80. The wide application of the method has given rise to the development of a^ number of technical procedures, all of them devised with a view toward simplification. In ordinary hospital work, it is most con- venient to keep on hand upon slant agar, a stock typhoid culture, the agglutinability of which is well known. From this stock culture, fresh ^ Widal, Bull, de la soc. med. des hopit., vi, 1896; Widal et Sicard, Ann. de I'inst. Pasteur, xi, 1897. 2 Gruber, Verhand. Congr. f. inn. Med,, Wiesbaden, 1896. *Stem, Gent. f. inn. Med., xlix, 1896. |, » Frankel, Deut. med. Woch., ii, 1897o 422 PATHOGENIC MIRCOORGANISMS inoculations upon neutral bouillon should be made each day, so that a young broth culture may always be on hand to furnish actively motile, evenly distributed bacteria. These bouillon cultures may be grown for from six to eight hours at incubator temperature or for from twelve to eighteen hours at room temperature. The temperatures at which the broth cultures are kept must depend, to a certain extent, upon the peculiarities of the typhoid bacillus employed, since some strains are rather more actively motile and furnish a more suitable emulsion if kept at a temperature lower than 37.5° C. A false clumping in the broth cultures due to a too high acidity of the bouillon or a too prolonged incubation, must be carefully guarded against. It is also possible to use for this test an emulsion of typhoid bacilli prepared by rubbing up a small quantity of a young agar culture in salt solution. Uniformity in the preparation of broth cultures or of emulsions should be observed, since the quantitative relationship between typhoid bacilli and agglu- tinins will markedly affect the completeness or incompleteness of the reaction. In high dilutions an excess of typhoid bacilli may bring about complete absorption of all the agglutinins present, without agglutinat- ing all the microorganisms. The blood of the patient to be used for a Widal test may be obtained in a number of ways. The most convenient method is to bleed the pa- tient from the ear or finger into a small glass capsule, in the form of that used in obtaining blood for the opsonin test, or into a small centrifuge tube. About 0.5 to 1 c.c. is amply sufficient. These capsules or tubes, after clotting of the blood, may be placed in the centrifuge which in a few revolutions will separate clear serum from clot. . The dilutions of the serum are then made. It is best to use sterile physiological salt solu- tion as a diluent, but neutral broth may be used. The dilutions may be made either by means of an ordinary blood-counting pipette or by means of a capillary pipette upon which a mark with a grease pencil, made about an inch from the tip, furnishes a unit of measure, and upon which suction is made by means of a rubber nipple. It is convenient to have at hand a small porcelain palette such as that used by painters, in which the various cup-like impressions may be utilized to contain the various dilutions. Dilutions of the serum are made, ranging from 1 : 10 to 1 : 50. A drop of each of these dilutions is mixed with a drop of the typhoid culture or emulsion upon the center of a cover-slip and the cover- slip inverted over a hollow slide. A control with normal serum and the same culture should always be made and also one with the culture alone to exclude the possibility of spoHtaneous clumping. Mixture BACILLUS OF TYPHOID FEVER 423 with the typhoid culture, of course, each time doubles the dilutions so that, for instance, a drop of serum dilution 1 : 10, plus a drop of the typhoid culture, gives the final dilution of 1 : 20. The prepara- tions may be examined with a high power dry lens or an oil im- mersion lens. In a positive reaction, the baciUi, which at first swim about actively, singly or in short chains, soon begin to gather in small groups and lose much of their activity. Within one-half to one hour, they will be gathered in dense clumps between which the fluid is clear and free from bacteria, and only upon the edges of the agglutinated masses may slight motility be observed. The degree of dilution and the time of exposure at which such a reaction may be regarded as of specific diagnostic value, have been largely a matter of empirical de- termination. It is generally accepted at present that complete agglu- tination within one hour in dilutions from 1 : 40 to 1 : 60 is definite prcof of the existence of typhoid infection. Exceptions, however, to this rule may occur. Agglutinations of typhoid bacilli in dilutions of 1 : 40, and over, have occasionally been observed in cases of jaundice and of tuberculosis, and these conditions must occasionally be consid- ered, though their importance was formerly exaggerated. The method of making the Widal test from a drop of whole blood, dried upon a slide, is not to be recommended, as accuracy in dilution by this method is practically impossible. As stated above, the agglutinin reaction rarely appears in typhoid fever before the beginning of the second week. It may continue during convalescence for as long as six to eight weeks and occasionally, in cases where there is a chronic infection of the gall-bladder, a Widal reac- tion may be present for years after an attack. For very exact work, even in clinical cases, the microscopic agglu- tination method may be replaced by macroscopic agglutination, ac- cording to the technique described in another section (page 229). In order to avoid both the necessity of keeping alive typhoid cultures for routine agglutination tests and also to preclude the danger of in- fection by the use of living culture, Ficker ^ has recommended the use of typhoid bacilli killed by formalin. This method has no advan- tages for practical purposes and in scientific bacteriological work it is, of course, not to be considered in comparison with the other exact methods. Precipitins, — The investigations of Kraus ^ in 1897, by which the 1 Ficker, Bed. klin. Woch., xlviii, 1903. « Kraus, Wien. klin. Woch., xxxii, 1897. 424 PATHOGENIC MICROORGANISMS precipitins were discovered, revealed specific precipitating substances, among others, also in typhoid immune sera. Since Kraus' original in- vestigation, these substances have been studied by Norris ^ and others.^ Opsonins. — A number of observers have shown that opsonins specific for the typhoid bacillus are formed in animals immunized with these organisms. Opsonins are formed also in patients suffering from typhoid fever, but exact opsonic estimations in all these cases are extremely difficult because of the rapid lysis which these bacteria may undergo both in the serum, and intracellularly after ingestion by the leucocytes. Klein ^ has attempted in part to overcome this difficulty by working with dilutions of serum and at the same time using comparatively thick bacterial emulsions and exposures to the phagocytic action not exceed- ing ten minutes. Chantemesse * has claimed that the opsonic index of typhoid patients was increased after treatment with a serum obtained by him from immunized horses, and Harrison ^ has reported similar results in patients treated by a modification of Wright's method of active immunization. Klein claims to have demonstrated that in typhoid-immune rabbits, after five injections, the opsonic contents of the blood were increased to an equal extent with the bactericidal sub- stances. He concludes from this interesting observation that it may well be that the opsonins are quite as important in typhoid immunity as are the latter substances. For diagnostic purposes in typhoid fever the estimation of the opsonic index, so far, has not been proven to be of great value. Specific Therapy in Typhoid Fever. — The failure to produce a soluble toxin from typhoid cultures has naturally so far precluded the possibility of an antitoxic therapy, such as that which has been successful in diph- theria. In the light of our present knowledge of the poisonous products of the typhoid bacillus it seems but natural that attempts by earlier investigators to apply the principles of Behring's work to typhoid fever were doomed to fail. Attempts to employ specific bactericidal and bac- teriolytic sera for therapeutic purposes in this disease have also been without favorable result. Active Prophylactic Immunization. — We have seen that work by Pfeiffer and KoUe and later by many others has shown that it is com- 1 Nfftris, Jour, of Inf. Dis., I, 3, 1904. 3 Barker and Cole, 22d Ann. Session, Assn. of Amer. Phys., Wash., 1897* 3 Klein, Bull. Johns Hopkins Hosp., 1907. * Chantemesse, 14th Internatl. Cong, for Hyg., Berlin, 1907.. ^Karrispn, Jour. Royal Army Med. Corps, 8, 1QQ7.. BACILLUS OF TYPHOID FEVER 425 paratively easy to immunize animals actively against typhoid infection by the systematic injection of graded doses, at first of dead bacilli, later of fully virulent live cultures. Attempts to. apply these principles pro- phylactically have been made recently on a large scale by Wright and his associates upon English soldiers in South Africa, and by German observers in German East Africa. The first recorded experiment of this sort which was done upon human beings was that of Pfeiffer and KoUe,* who in 1896 treated two in- dividuals with subcutaneous injections of an agar culture of typhoid bacilli which had been sterilized at 56° C. The first injection was made with two milligrams of this culture. Three or four hours after the in- jection the patient suffered from a chill, his temperature gradually rose to 105° F., and there was great prostration and headache, but within twenty-four hours the temperature had returned to normal. This experiment showed that such injections could be practiced upon human beings without great danger. Simultaneously with the work of Pfeiffer and Kolle, Wright ^ con- ducted similar experiments on officers and privates in the English army. The actual number of persons treated directly or indirectly under Wright's ^ supervision in an investigation covering a period of over four years comprised almost one hundred thousand cases. The methods employed by Wright have been modified several times in minor details; the principles, however, have remained consistently the same. In the first experiments Wright employed an agar culture three weeks old, grown at 37° C, then sterilized at a temperature below 60° C, and pro- tected from contamination by the addition of five-tenths per cent of carboUc acid*. Later, Wright^ employed bacilli grown in a neutral one-per-cent pepton bouillon in shallow layers or flasks. Great importance is attached both to the virulence of the typhoid strain, which may to a moderate extent be standardized by passage through guinea-pigs, and to care in using bw temperatures for final sterilization. The temperature recommended, by Harrison,^ is 52° €).,, after which the cultures are aarboUzed.. 1 Pfeiffer. nnd KoUCf Deut. med. Wocb., xxii, 1896; xxiv, 1898. 2 Wright, Lancet, Sept., 1896. 3 Wright and Semple, Brit. Med. Jour., 1897; Wright and Leishman, Brit. Med. Jour;, Jan., 1900. * Wright, Brit. Med. Jou^r,, 1901; Lancet, Sept., 1902; Brit. Medr Jpuj*., Oct.,, 1903. 5 Uarrison, Jour. Iio3^^1 , Arpay , Medical Corps^ 1907, , 426 PATHOGENIC MICROORGANISMS It is nevertheless extremely difficult to tabulate satisfactory statis- tics from a mass of experiments observed by a large number of indi- viduals. On the whole, however, it seems fair to state that advan- tageous results followed the active immunization practiced by Wright. Wright's own estimation, in a careful attempt to present the subject fairly, gives a reduction of the morbidity from typhoid fever in the British army of fifty per cent, and a reduction of the mortality of those who became infected in spite of inoculations of fifty per cent also. It is not at all impossible that a number of different strains will have to be used eventually for the ideal vaccine, inasmuch as the antigenic differences which have been recently discovered and alluded to would make it seem that no si»gle strain can be expected to produce anti- bodies which would protect against all other strains. It is not impos- sible that some individual strain may combine the antigenic properties of the entire group. This, however, has still to be worked out. The method of Peiffer and Kolle consists in the injection of salt- solution emulsions of fresh agar cultures sterilized at 60° C. The results reported were in general favorable. Recent extensive tests in the United States Army, observed by Rus- sell,^ seem to have removed any doubt which may have existed as to the efficacy of prophylactic typhoid vaccination. Russell's statistics show a steady decline of typhoid in the U. S. Army beginning with the introduction of compulsory vaccination in 1910. In 1913 there was but one case among over 80,000 men. The method at present employed is as follows: The ''Rawlings" strain of typhoid, obtained from Wright, is used. " Eightoen-hour agar cultures in Kolle flasks are washed off with sterile saline to an approxi- mate concentration of one billion to the c.c. The suspension is killed at 53° C. for one hour and 0.25 per cent tricresol is added. Aerobic and anaerobic culture controls are made and a rabbit and mouse inocu- lated to insure sterility. For immunization 3 to 4 doses are given ranging in quantity from 500 million to one billion at 7 to 10 day intervals. The protection probably lasts about 2 years, though this is not certain. Another point of importance in this connection has recently been raised by Metchnikoff and Besredka.^ They vaccinated chimpanzees with typhoid bacilli and found that when emulsions of the clear bac- 1 Russell, Am. Jour, of Med. Sc, cxlvi, 1913.. 2 Metchnikoff and Besredka, Am. de Tinst. Past., 1911. BACILLUS 0^ TYPHOID FEVE^ 427 teria were used, protection was only slight. Better results were ob- tained— that is, apparently complete protection within 8 to 10 days — when living sensitized bacteria were injected. (Bacteria which had been exposed to the action of inactivated immune serum.) Brough- ton ^ has applied this method to human beings. Gay ^ has also pre- pared a sensitized dead typhoid vaccine which he has already used in a considerable number of cases. It will take some time, however, before a statistical estimation of the superiority of this method over the older vaccination with dead bacteria will be possible. Attempts have recently been made to treat active typhoid fever by intravenous injections of sensitized vaccines. The cases are as yet too few to permit final judgment. BACILLUS FECALIS ALKALIGENES In 1896 Petruschky ^ described a bacillus which is a not infrequent inhabitant of the human intestine, being found chiefly in the lower part of the small intestine and the large intestine. This organism, which he called Bacillus fecalis alkaligenes, is of little pathogenic im- portance, although Neufeld states that he has seen a case of severe gastroenteritis in which the watery defecations contained this bacillus in almost pure culture. As a rule, however, this organism cannot be regarded as pathogenic, and is important chiefly because of the ease with which it may be mistaken for Bacillus typhosus. Bacillus fecalis alkaligenes is an actively motile, Gram-negative bacillus, ' possessing, like the typhoid bacillus, numerous peritrichal flagella. On the ordinary culture media it grows like the typhoid bacillus. It does not coagulate milk. It produces no indol, and on sugar media in fermentation tubes produces no acid or gas. On potato, its growth, while somewhat heavier than that of the typhoid bacillus, is not sufficiently so to permit easy differentiation. It differs from Bacillus typhosus in that it produces no acid on any of the sugar media, and is therefore easily differentiated by cultivation upon Hiss serum-water media or on pepton waters containing sugars. On the Hiss semi-solid tube-medium Bacillus fecalis alkaligenes, while cloud- ing the medium throughout, grows most heavily on the surface, where, eventually, it forms a pellicle. 1 Broughton, C. R. de I'Acad. des Sc, cliv, 1911. 2 Gay, Arch, of Int. Med., 1914. » ?eiruschky. Cent. f. Bakt., I, xix, 1896- CHAPTER XXVIII "BACILLI OF THE COLON-TYPHOID-DYSENTERY GROUP , ifiordinued) BACILLI INTERMEDIATE BETWEEN THE TYPHOID AND COLON ORGANISMS {Bacilli of Meat Poisoning and Paratyphoid Fever) There is an extensive group of Gram-negative bacilli which be- cause of their morphology, cultural behavior, and pathogenic properties, are classified as intermediate between the colon and the typhoid types. The microorganisms belonging to this group have been described, most of them, within the last fifteen years, but few of them have been fully identified with one another. They have been variously designated as thp "hog-cholera group," "the enteritidis group," the "paracolon group" or "paratyphoid group," because of the pathological conditions with which the chief members under investigation have been found associated. Attempts to systematize the group by the comparative study of a large number of its members have been made, notably by Buxton^ and by Durham,^ and the work of these writers, based on cultural and agglutinative studies, has added materially to our knowledge of these organisms. The microorganisms of this group are morphologically indistinguish- able from the colon and typhoid bacilli. They are Gram-negative and possess flagella. Their motility is variable, but usually approaches that of the typhoid bacilli in activity. They correspond, furthermore, to the two other groups in their cultural characteristics upon broth, agar, and gelatin. On 'potato, they vary, some of them approaching in deli- cacy the typhoid growth upon this medium, others more closely approximating the heavy brownish growth of B. coH. Indol is rarely formed by them, though this has not been absolutely constant in all descriptions. As a group, they are easily distinguished from Bacillus 1 Buxton, Jour. Med, Res., N. S., iii, 1900. » Durham, Jour. Exper. Med., v, 1901. 428 BACILLI BETWEEN TYPHOID AND COLON ORGANISMS 429 typhosus on the one hand, and from Bacillus coli on the other, by the following simple reactions tabulated by Buxton.^ B. coli. Intermediates. B. typhosus. Coagulation of milk + + + + + _ Production of indol Fermentation of lactose with gas Fermentation of dextrose with gas Agglutination in typhoid-immune serum. + Characteristics of the three groups as»shown by fermentation tests follow: Gas upon Dextrose. Gas upon Lactose. Gas upon Saccharose. Gas on Dulcit. B typhosus . . + + + + + + + + • + + + Intermediates B. coli communis + B. coli communior B. acidi lactici B. lactis aerogenes ^ Pathogenically, the bacilli of this "intermediate group*' have attracted attention chiefly in connection with meat poisoning, and with protracted fevers indistinguishable from mild typhoidal infections. In 1888, Gartner ^ described a bacillus which he isolated from the meat of a cow, the ingestion of which had produced the symptoms of acute gastrointestinal catarrh in 57 people. One of these died of the disease and the bacilli could be demonstrated in the spleen and blood of the patient. This bacillus, called Bacillus enteritidis by Gartner, was actively motile, formed no indol, but produced gas in dextrose media. Acute gastrointestinal symptoms could be induced by feeding the organisms to mice, guinea-pigs, rabbits, and sheep, and the bacilli could be recov- ered from the infected animals. An interesting observation, which has since become important in characterizing the group of these bacilli concerned in meat poisoning, was the fact that the bacterial bodies themselves were foimd by Gartner to be extremely toxic, containing a poison which, in contradistinction to the endotoxins of many other microorganisms, was extremely resistant to heat. Sterilized cultures showed the same pathogenic effects as the living bacilli. Epidemics 1 Buxton, loc. cit. 2 Jackson. * Gartner f Corresp. Bl. d. Aerzt. Vereins, Turingen, 1888. ' 430 PATHOGENIC MICROORGANISMS of meat poisoning similar to the one described by Gartner, in which similar bacteria were isolated, were those described by Van Ermen- gem,^ occurring at Morseele in 1891, the one described by Holst,^ the Rotterdam epidemic described by Poels and Dhont,^ the one described by Basenau, and many others. Bacillus Morseele of Van Ermengem, Bacillus hovis morhificans of Basenau,* and bacilli isolated in similar epidemics by others, are, ex- cept for slight differences in minor characteristics, almost identical with Gartner's microorganism. In 1893, Theobald Smith and Moore ^ noted a great similarity be- tween the so-called hog-cholera bacillus, the bacilli of the Gartner group, and Bacillus typhi murium isolated by Loeffler. These ob- servers first used the term *' hog-cholera" group for the organisms under discussion. In 1899 Reed and Carroll® noted that Bacillus icteroides, asso- ciated by Sanarelli with yellow fever, was culturally similar to the bacillus of hog cholera. Meanwhile, other observers had been isolating bacilli, similar to those spoken of above, from cases of protracted fevers in human beings, often closely simulating typhoid infections. The first cases of this kind on record were those of Achard and Bensaude.'^ In 1897, Widal and Nobecourt ^ described a bacillus which they had isolated from an esophageal abscess following typhoid fever, which closely resembled Bacillus psittacosis of Nocard,® and which, following a nomenclature previously suggested by Gilbert,-^^ they designated the paracolon bacillus. This microorganism, isolated from a parrot by Nocard, showed a close resemblance to bacilli of the Gartner group. There are a large number of apparently nonpathogenic organisms sometimes referred to as paratyphoid C, but better perhaps as ''hetero- genous types,'' which are culturally identical with the paratyphoid * Van Ermengem, Bull. Acad. d. m^d. de Belgique, 1892; "Trav. de lab. de Puniv. de Gand," 1892. 2 Hoist, Ref . Cent. f. Bakt., xvii, 1895. ' Poels und Dhont, Holland Zeit. f. Tierheilkunde, xxiii, 1894. * Basenau, Arch, f . Hyg., xx, 1894. " Th. Smith and Moore, U. S. Bureau of Animal Industry Bull., vi, 1894. « Reed and Carroll, Medical News, Ixxiv, 1899. ' Achard and Bensaude, Bull, de la soc. d. h6pitaux de Paris, Nov., 1906. 8 Widal et Nobecourt, Semaine m€d., Aug., 1897. ^ Nocard, Ref. Baumgarten's Jahresb., 1896. " Gilbert, Semame m6d., 1895. BACILLI BETWEEN TYPHOID AND COLON ORGANISMS 431 but do not agglutinate in either paratyphoid A or B sera. An anti- serum produced with these types usually reacts with the homologous strain only. Such organisms have been studied by Krumwiede and many others, including ourselves. Strains recently isolated at this laboratory came from cases of nephritis, German measles, jaundice, and the stools of healthy soldiers, done as a matter of routine. In 1898, Gwyn ^ reported a case at the Johns Hopkins Hospital, which presented all the symptoms of typhoid fever, but lacked serum agglutinating power for Bacillus typhosus. From the blood of the pa- tient, Gwyn isolated an organism, with cultural characteristics similar to those of the Gartner bacillus, which he called a ' ' paracolon bacillus. ' ' This bacillus was agglutinated specifically by the serum of the patient. Cushing,^ in 1900, isolated a similar microorganism from a costo- chondral abscess, appearing during convalescence from typhoid fever. In the same year, Schottmiiller ^ reported five cases from which similar bacilli were isolated. Careful cultural studies of the microor- ganisms here obtained showed that they could be divided into two simi- lar, yet distinctly different types, one of them, the ''Miiller*' organism, approaching closely to the typhoid type, especially in its growth upon potato; the other, the '*Seeman'' type, .corresponding more closely to the Gartner bacilli. Similar cases were reported by Kurth,* Buxton and Coleman,^ Libman,^ and others. The two types of organisms, paratyphoid A and B, described by Schottmiiller and studied by many other observers, can be culturally differentiated, though not without difficulty. Type A is more delicate in its growth on various media than B, growing with almost invisible growth on potato, and differing from typhoid in its gas formation on dextrose broth. Milk is not coagu- lated, but remains turbid, not being finally cleared by solution of the casein as in similar cultures of type B. Lactose whey is acidified and remains acid. This organism has been isolated from the normal intes- tines of animals by Morgan.'^ Kutscher for this reason suggests that essentially and except in rare instances this organism is a non-patho- * Gwyn, Johns Hopkins Hosp. Bull., 1898. 2 Gushing, Johns Hopkins Hosp. Bull., 1900. ^ Scholtmuller, Deut. med. Woch., 1900; Zeit. f. Hyg., xxvi. * Kurth, Deut. med. Woch., 1901. 6 Buxton and Goleman, Proc. N. Y. Pathol. Soc, Feb., 1902. « lAbman, Jour. Med. Res., N. S., iii, 1902. ' Morgan, cited from Kutscher, KolU und WassermanUf Handbuch. Erganzungs, I. 432 PATHOGENIC MICROORGANISMS genie saprophyte. Recently Krumwiede, Pratt, and Kohm have shown that the members of this group and some of the so-called paratyphoid C group may be distinguished from all the other paratyphoids by inability to ferment xylose, and Weiss has shown paratyphoid A may be subdivided on the basis of antigenic properties. Type B grows more heavily on all media than A, especially on potato (though this is irregular). Milk is slightly acidified at first, but eventually is rendered alkaline and cleared, possibly by casein solution. Subdivision of this group was made by Weiss and others on the basis of inosite fermentation. Eventual differentiation must be made by agglutination. The diseases caused by these bacteria may be divided as follows : I. Those which fall into the category of meat poisoning with sudden onset of gastroenteric symptoms following ingestion of meat; and II. Those in which the disease simulates a mild typhoid fever, dif- fering from ttis only by the absence of the specific agglutination. The differential diagnosis between the second type of case and true typhoid fever may be extremely difficult. In contradistinction to true typhoid the temperature reaction of this case may set in more abruptly and remain more irregular throughout the disease. Gastric symptoms, vomiting, and nausea are often more prominent than in typhoid fever and enlargement of the spleen is less regularly present. Owing to the low mortality of paratyphoid fever (in 120 cases observed by Lentz ^ less than 4 per cent, and in many other smaller epidemics no deaths have occurred) , we have remained relatively ignorant concerning the pathologic anatomy of the dis- ease. Longcope ^ observed a case, fatal after two weeks of illness, in which there was no enlargement of Peyer's patches and no sign of even beginning ulceration. Most other observers have also found less involvement of the lymphatics of the bowel than is found in typhoid fever. During the disease the bacteria can often be cultivated from the blood, and the serum of the patient may agglutinate specifically paratyphoid strains. In this way the diagnosis can often be made. Libmann ^ has isolated the organism from the fluid aspirated from the gall bladder in a case operated on for cholecystitis. Most of these microorganisms possess pathogenicity for mice, guinea- ^ Lentz, Klin. Jahrb., xiv, 1914. 2 Longcope, Amer. Jour, of Med. Sciences, cxxiv, 1902. * Libmann, Jour, of Med. Res., viii, 1902. BACILLI BETWEEN TYPHOID AND COLON ORGANISMS 433 pigs, and rabbits, which exceeds that of the colon or typhoid bacilli. A number of the bacilli of this group, furthermore, especially those most closely similar to the original B. enteritidis of Gartner, contain an endo- toxin which shows a high resistance to heat, which may explain the fact that illness has occasionally followed the ingestion of infected meat even after preparation by cooking. Bacteriological correlation of these bacilli has been attempted, as stated above, by Durham and by Buxton, and more recently by Kutscher and Meinicke.^ The subject is a difficult one and for ultimate clearness will require much further work. Harding and Ostenberg ^ have examined a series of organisms of the intermediate group on various sugars, and find that by the use of xylose and arabinose three definite groups can be established. I. Those making aldehyd (red) on fuch'oin-sulphite agar with both arabinose and xylose — both Schottmiiller types A and B and strains of Bacillus enteritidis. II. Red on arabinose and not on xylose — typhi murium, para- typhoid Gwyn, paratyphoid Loomis, and three others. III. Red on xylose and not on arabinose — B. hog cholera. This work was carefully carried out and may possibly point toward an ultimate classification. However, the strains employed were too few to permit definite conclusions at present. Durham,^ on the basis of cultural and agglutinative studies, has formulated a classification of the Gram-negative bacilli of the typhoid- colon and aUied groups, which, though hardly final, aids considerably in throwing light upon the interrelationships of the various species. Durham's divisions are as follows: Division I. Typhoid-like Morphology (motile). A. No sugars fermented. Type B. fecalis alkaligenes. B. Acid in dextrose, but no gas. Type B. typhosus. Agglutination in typhoid serum. C. Acid in dextrose, but gas only when other constituents are favor- able. No acid or gas from lactose or saccharose. No agglutination in typhoid serum. Includes Bacillus "Gwyn" and Bacillus "0" of Gushing. D. Acid and gas from dextrose. No acid or gas from lactose or. ^ Kutscher und Meinicke, Zeit. f. Hyg., lii, 1906. 2 Harding and Ostenberg, Jour, of Inf. Dis., ii, 1912. ' Durham^ loc. cit. 434 PATHOGENIC MICROORGANISMS saccharose. Grows more rapidly than typhoid. No agglutination in colon-immune serum. Slight reaction with some typhoid sera. Includes Gartner's B. enteritidis, B. Morseele, Gunther's meat-poisoning bacillus, hog cholera bacillus, B. psittacosis, B. morbificans bo vis, Durham's Bacillus "A," B. typhi murium. Division II. Colon-like Morphology (motile). E. Acid and gas from dextrose, none from lactose or saccharose. Rate of growth and colony appearance more Uke colon than typhoid. F. Acid and gas from dextrose, and no gas from lactose. Types isolated by Durham. G. Acid and gas from dextrose; acid, no gas, from lactose. Differ from F in serum reactions. H. B. coli communis. Acid and gas from dextrose and lactose; none from saccharose. /. B. coli communior. Acid and gas from dextrose, lactose, and saccharose. Division III. Non-motile. Polysaccharide splitters (starch). Type B. lactis aerogenes. Includes baciUi of mucosus capsulatus group, and Friedlander's bacillus. CHAPTER XXIX BACILLI OF THE COLON-TYPHOID-DYSENTERY GROUP {Continued) THE DYSENTERY BACILLI Although acute dysentery has been an extremely prevalent disease, occurring almost annually in epidemic form in some of the Eastern coun- tries and appearing sporadically all over the world, its etiology was obscure until 1898 when Shiga ' described a bacillus which he isolated from the stools of patients suffering from this disease in Japan, and es- tablished with scientific accuracy its etiological significance. Since the discovery of Shiga's bacillus a number of other bacilli have been de- scribed by various workers, all of which, while showing slight biological differences from Shiga's microorganism, are sufficiently similar to it culturally and pathogenically to warrant their being classified together with it in a definite group under the heading of the " dysentery bacilli." The manner in which Shiga made his discovery furnishes an in- structive example of the successful application of modern bacteriological methods to etiological investigation. Many workers preceding Shiga had attempted to throw light upon this subject by isolations of bacilli from dysenteric stools, and by extensive animal inoculation. Shiga, following a suggestion made by Kitasato, approached the problem by searching for a microorganism in the stools of dysentery patients which would specifically agglutinate with the serum of these patients. His labors were crowned with success in that he found, in thirty-six cases, one and the same microorganism which showed uniform serum agglu- tinations. Further, he found that this bacillus was not present in the dejections of patients suffering from other diseases nor in those of normal men, and that when tested against the blood serum of such people it was not agglutinated. Morphology. — Shiga's bacillus is a short rod, rounded at the ends, » iS/iigia, Cent. f. Bakt., xxiii, 1898; ibid., xxiv, 1898; Deut. med. Woch., xliii, xliv, and xlv, 1901. 435 436 PATHOGENIC MICROORGANISMS morphologically very similar to the typhoid bacillus, and, like it, inclined to involution forms. The organism generally occurs singly, more seldom in pairs. It is decolorized by Gram's method of staining. With the»ordinary anilin dyes it stains easily, showing a tendency to stain with slightly greater intensity at the ends. The organism is an aerobe and facultative anaerobe. Although described at first by Shiga as being motile, its motility has not been satisfactorily proven,- and most observers agree in denying the presence of flagella and affirming the complete absence of motility. Cultural Chaxacteristics. — On agar the colonies are not characteristic, resembling those of the typhoid bacillus. On gelatin, the colonies appear very much like typhoid colonies and the- gelatin is not liquefied. On 'potato, the growth, like that of typhoid, is at first not visible, but after about a week turns reddish brown. In broth, there is clouding, with moderate deposits after some days. No pellicle is formed. Milk is not coagulated. Litmus milk shows a slight primary acidity, later again becoming alkaline and taking on a progressively deeper blue color. Indol is not formed in pepton water by all varieties. No gas is formed in media containing dextrose, lactose, saccharose, or other carbohydrate. While not delicately susceptible to reaction^ the bacillus prefers slightly alkaline media. . Shiga differentiated his organism from the typhoid bacillus chiefly by supposed differences in colony characters and by the agglutination reaction. Following the work of Shiga, a large number of investigators turned Cheir attention to the subject of dysentery, with the result that many new forms were discovered and at first a considerable amount of con- fusion prevailed. Flexner * in 1899 investigated dysentery in the Philippines, and isolated a bacillus which, he considered, corresponded to Shiga's organism. Strong and Musgrave/' in 1900 described a bacillus isolated from ^Flexner, Phila. Med. Jour., vi, 1900, and Bull. Johns Hopkins Hosp., xi, 1900. • Strong and Musgrave, Report Surg. Gen. of Army, Washington, 1900. THE DYSENTERY BACILLI 437 dysentery cases in the Philippines which was essentially like that of Flexner. Nearly simultaneously with the papers of Flexner and of Strong and Musgrave, Kruse ^ published investigations of an epidemic of dysentery occurring in Germany. His observations were of the greatest importance and largely formed the starting point of the further advances which have been made in the etiology ^f dysentery. Kruse's organism was described as forming colonies on gelatin and agar, practically like those of Bacillus typhosus. Like this bacillus, no gas was formed from grape sugar, and the growth in milk and on potato, and even in Piorkowski's urine gelatin^ resembled that of Bacillus typhosus. According to Kruse, this organism was absolutely with- out motility. In 1901 Kruse ^ contributed a second paper. In this, besides con- firming his previous observations, he described another class of organ- ism coming from cases which he designated as "pseudo-dysentery of insane asylums." In the case of one patient, and at two autopsies, he isolated organisms which he could not distinguish morphologically or cul- turally from the true dysentery bacillus, but which showed differences in their serum reaction. By careful study of the. behavior of these bacilli in the serum of patients and in immune serum from animals, he not only showed that they were different from his original cultures from cases of epidemic dysentery which, no matter what their source, were found to be alike, but that they showed differences among themselves and apparently fell into two or more varieties. One of these organisms culturally and by its serum reactions showed itself practically identical with one of the cultures he had received from Flexner. Spronck ^ in 1901 described an organism isolated in Utrecht from dysentery cases, which showed great similarity to the Shiga-Kruse organism; but, when tested in the serum of a horse immunized against true dysentery bacillus, showed practically no agglutination. He placed this organism in the group designated by Kruse as the '^ pseudo-dysentery bacilli." His communication is of importance, since it is the first re- ported instance in which any investigator had recognized and associated the so-called pseudo-dysentery bacilli with dysentery approaching the acute epidemic form in type. Following this work a number of investigators, including Vedder ^ Kruse, Deut. med. Woch., xxvi, 1900. ' Kruse, Deut. med. Woch., xxvii, 1901. •/Sp'onc/c, Ref, Baumgarten's Jahresber., 1901. 438 PATHOGENIC MICROORGANISMS and Duval/ Flexner, and Shiga ^ himself, published communications ir» which they claimed identity for the various forms previously described. In 1902 Park ^ and Dunham described an organism which they found in a small outbreak of dysentery occurring in Maine. This organism differed from most of those previously described in that it was found to produce indol in pepton solutions. In the same year Martini * and Lentz published an article in which they attempted to differentiate various dysentery bacilli by means of agglutination. This research is of importance in that it supported the work of Kruse and of Spronck, indicating a difference between the ag- glutinative character of the Kruse organism and the so-called " pseudo- dysentery" type, in which Flexner's organisms were included. It is of further interest, since it indicated a marked difference between Flexner's Philippine cultures and the Philippine culture of Strong, the Strong organism refusing to agglutinate not only in ^' Shiga'' immune serum, but also in " Flexner " immune serum. Simultaneously with this article Lentz ^ published the results of com- parative cultural researches with dysentery and "pseudo-dysentery" bacilli, in which he made the important observation that the original Shiga-Kruse bacilli did not affect mannit, while the " pseudo-dysentery " bacilli, including Flexner's and Strong's Philippine cultures, fermented mannit, giving rise to a distinct acid reaction in the medium. The Flexner organisms and others of the "pseudo-dysentery" bacilli, how- ever, fermented maltose, while the Shiga-Kruse type, as well as Strong's bacillus, left it unchanged at the end of forty-eight hours. In January, 1903, Hiss and Russell ® described a bacillus (" Y") from a case of fatal diarrhea in a child, which by ordinary cultural test and absence of motility was found to resemble the Shiga-Kruse and Flexner bacilli. Immediately upon its isolation, it was found, however, to differ from the Kruse culture by its ability to ferment mannit. This observa- tion was made independently of Lentz's work, which, at that time, had not become known in America. In the comparative study of Hiss and Russell on the fermentative abiHties of various dysentery cultures, the serum water media (described on page 132). were used. By the use of ' Vedder and Duval, Jour, Exp. Med., vi, 1902. ^ Shiga, Zeit. f. Hyg., 41, 1902. « Park and Dunham, N. Y. Univ. Bull, of Med. Sci., 1902. * Martini und Lentz, Zeit. f. Hyg., xli, 1902. » Lentz, Zeit. f. Hyg., xli, 1902. •Hiss and Russell Med. News, Feb., 1903. THE DYSENTERY BACILLI 439 these media, it was found that the Kruse culture, a culture of Flexner's bacillus from the Philippines, and Duval's " New Haven '* culture fer- mented dextrose with the production of a solid acid coagulum, but did not affect mannit, maltose, saccharose, or dextrin. The culture of Hiss and Russell, on the other hand, fermented not only dextrose but also mannit with the production of acid and coagulation of the medium. Maltose, saccharose, and dextrin were not fermented. The " Y " bacillus, furthermore, was shown to differ entirely from the cultures of Shiga, Kruse, and "New Haven" in the serum of immunized animals. This serum had for bacillus " Y " a titer of 1 : 500 while the three other above- named organisms did not agglutinate in it at any dilution. In normal beef serum, the Hiss-Russell organism was found to agglutinate as highly at 1 : 320, while the other three cultures gave no reaction in dilutions of over 1 : 10 or 20. Park and Carey/ in March, 1903, described an epidemic of dysen- tery occurring in the town of Tuckahoe, near New York City, and isolated an organism which resembled the Shiga-Kruse bacilli in not fermenting mannit, but produced indol in pepton solution after five days. It corresponded in agglutination with the cultures " New Haven '' and "Shiga" when tested in the serum of a goat immunized against the mannit-fermenting culture "Baltimore," i.e., did not react at 1:50, whereas Flexner's " Manila " and " Baltimore " cultures. Park and Dun- ham's "Seal Harbor" culture, and some New York cultures, all fer- menting mannit, agglutinated up to two thousand dilution in the " Bal- timore" serum. The preceding review of a part of the literature, by which our knowl- edge of the dysentery bacilli was developed, demonstrates sufficiently that we have to deal in this group with a number of different micro- organisms. This, as we have seen, was a fact first recognized by Kruse when he spoke of his true dysentery and his pseudo-dysentery strains. In spite of much confusion at first, the careful study of fermentation phenomena, of specific, agglutinations, and, more recently, by Ohno ^ and others, of the bacteriolytic phenomena in immune sera, has made it possible to distinguish sharply between a number of groups. Basing the grouping of these microorganisms upon a careful study of fermentations, Hiss^ has divided them as follows: » Park and Carey, Jour, Med' Res., ix, 1903. » Ohno, Philippine Jour, of Sci., 1, ix., 1906. *Hiss, Jour. Med. Res., N. S., viii, 1904. 440 PATHOGENIC MICROORGANISMS "Kruse," > Ferment dextrose. Group I. "New Haven" "Y" (Hiss and Russell type) "SealHarbbr" " Diamond " "Ferra" "Strong" (type) " Harris " (type) "Gray" "Baltimore" "Wollstein" Ferment dextrose and mannit. Group II. Ferments dextrose, mannit, saccharose Group III. Ferment dextrose, mannit, maltose, saccha- rose, dextrin. Fermentation of saccharose (as a rule) only after 6 days. Group IV. It was noticed, it should be mentioned, however, that in the case of the a Y " "Diamond," and "Ferra" there was usually delayed acid fermen- tation of maltose, never any of dextrin. In studying the agglutinative characters of these groups, furthermore, it was found that fermentation tests and agglutinations went hand in hand. The following table will illustrate this point:/ ' Serum of Rabbit immunized against GRobp I. (Shiga's culture). Bacilli of Group I. : " Shiga " (homologous) 20,000 "Kruse" .............;..!'.. 20,000 "New Haven" •. 20,000 Bacilli of Group II.: "Y" 200 "Ferra" ..-.. 200 "Seal Harbor" 200 Bacilli of Group. IV.: "Baltimore" 800 "Harris" 800 "Gray" ...,....,......,.................,..> 800 "Wollstein" ...... ^ ......! . ... .......... .^ . . . . ....,.., 800 Serum of Rabbit immunized against Group II. ("Y" culture, Hiss and Russell). Bacilli of Group I.: ^ "Shiga" less than 100 "Kruse" 100 "New Haven" .. . . . .' 100 1 Hiss, Jour, of Med. Research, 13, N. S.", viii, 190i^. THE DYSENTERY BACILLI 441 Bacilli of Group II. : " Y " (homologous) 6,400 "Ferra" 6,400 "Seal Harbor 6,400 Bacilli of Group IV.: "Baltimore " 1,600 "Gray" 1,600 "Harris" 1,600 "WoUstein" 1,600 Serum of Rabbit immunized against Group IV. ("Baltimore" culture) . Bacilli of Group I.: "Shiga" less than 100 "Kruse" 100 "New Haven " 100 Bacilli of Group II. : " Y" 400 "Ferra" 400 "Seal Harbor" 400 BaciUi of Group IV.: . "Baltimore" (homologous) ,. . 3,200 "Harris" 3,200 "Gray" 3,200 "WoUstein" 3,200 In common, all these groups possess an identical morphology, the Gram-negative staining characteristics, the lack of motility with close adherence to the line of inoculation in the Hiss tube medium, the in- ability to liquefy gelatin, the inability to form acid from lactose, and the inability to produce gas from any carbohydrate media. Biological Considerations. — The dysentery bacilli in neutral broth or upon agar slants may remain alive without transplantation for periods of several months. They are aerobes and facultative anaerobes when proper sugars are present, preferring, however, the aerobic environ- ment. They are easily destroyed by heat, an exposure to 60° C. killing them usually in a short time (ten minutes). Against cold they show considerable resistance, surviving freezing for a period of several weeks. They show little resistance to the usual strengths of the common chem- ical disinfectants. Pathogenicity. — There is practically no doubt at the present time as to the etiological connection between the bacilli of this group and the dis- eases clinically classified as acute dysentery. A more chronic form of 29 442 PATHOGENIC MICROORGANISMS dysentery due to a protozoan, the Amoeba coli, though presenting much clinical resemblance to the bacillary dysenteries is, nevertheless, an entirely distinct disease. Infection takes place, probably, entirely by ingestion of the bacteria with infected water or food contaminated from the feces of dysentery patients. A small epidemic occurring in a hospital in New York City and caused by the bacillus ^'Y" of Hiss and Russell was indirectly traced to milk by Zinsser.^ Endemic in a large part of the world, especially in the warmer climates, the disease most frequently occurs in epidemics of more or less definite localization, usually under conditions which accompany the massing of a large number of human beings in one place, such as those which occur in the crowded quarters of unsanitary towns, in insti- tutions such as insane asylums, or in military camps. The mortality of such epidemics may be very large. According to Shiga,^ the disease in Japan frequently shows a mortality of over twenty per cent. The disease in human beings usually begins as an acute gastro- enteritis which is accompanied by abdominal pain and diarrhea. As it becomes more severe, the colicky pains and diarrhea increase, the stools lose their fecal character, becoming small in quantity and filled with mucus and flakes of blood. There is often severe tenesmus at this stage, and the bacilli are present in large numbers in the dejecta. Owing to the absorption of toxic products, symptoms referable to the nervous system, such as muscular twitching, may supervene, and if the disease is at all prolonged, there are marked inanition and prostration. At autopsy in early stages there may be found only a severe catar- rhal inflammation of the mucous membrane of the large intestine. In the later stages there are extensive ulcerations, and the bacteria are histologically found lodged within the depths of the mucosa and sub- mucosa. Occasionally they may penetrate to the mesenteric glands, but as far as we know there is no penetration into the general circulation. Poisonous Products of the Dysentery Bacilli. — The separate types of dysentery bacilli vary exceedingly in their powers to pro luce toxic substances. Of all the various types which have been described, the strongest poisons have been produced with baciUi of the Shiga-Kruse variety, less regularly active ones with bacilli of the Flexner and of the "Y" type. In fact, investigations carried out with the Shiga bacillus have tended to show that the disease itself is probably a true toxemia, ^Zinsser, Proc. N. Y. Path. Soc, 1907 2 Shiga. Cent. f. Bakt., xxiii. 1898. THE DYSENTERY BACILLI 443 its symptoms being referable almost entirely to the absorption of the poisonous products of the bacillus from the intestine. The earliest investigations, carried on chiefly upon rabbits, which are more suscjptible to this poison than any other animals, showed that even small doses of cultures of this bacillus administered intravenously or subcutaneously would produce death within a very short time. Conradi,^ Vaillard^ and Dopter, and others, finding that toxic symptoms were almost as pronounced when dead cultures were given as when the living bacilli were administered, came to the conclusion that the poisons of this bacillus were chiefly of the endotoxin type. More recently Todd,^ Kraus,'' and Rosenthal^ have claimed independently that they were able to demonstrate strong soluble toxins, similar in every way to diph- theria toxin. Kraus and Doerr,^ moreover, claim to have further cor- roborated this by producing specific antitoxins with these substances. It is easy to obtain poisonous substances from dysentery cultures in considerable strength, both by extracting the bacilli themselves and by filtration of properly prepared cultures. It is therefore not unlikely that both types of poison are produced by the bacilli. Neisser and Shiga ^ obtained toxins by emulsifying agar cultures in sterile salt solution, killing the bacilli at 60° C, and allowing them to extract at 37.5° C. for three days or more. The filtrates from such emul- sions were extremely toxic. The simplest method of obtaining poisons from these bacilli is to cultivate them for a week or longer upon moder- ately alkahne meat-infusion broth. At the end of this time, the micro- organisms themselves may be killed by heating to 60° and the cultures filtered. According to Doerr,^ the toxins may be obtained in the dry state by precipitation with ammonium sulphate and re-solution of the precipitate in water. The action of th.e dysentery toxin upon animals is extremely characteristic and throws much light upon the disease in man. The injection of a large dose intravenously into rabbits causes a rapid fall in temperature, marked respiratory embarrassment, and a violent 1 Conradi, Deut. med. Woch., 1903, 2 Vaillard et Dopter, Ann, de Tinst, Pasteur, 1903. 3 Todd, Brit. Med, Jour,, Dec, 1903, and Jour, of Hyg,, 4, 1904. ^ Kraus, Monatschr, f. Gesundheit, Suppl. 11, 1904. »/2osen^/iaZ, Deut, med. Woch., 1904. * Kraus und Doerr, Wien. klin. Woch,, xlii, 1905. 7 Neisser and Shiga, Deut. med, Woch., 1903. 8 Doerr, " Das Dysenterietoxin," Jena, 1907. 444 PATHOGENIC MICROORGANISMS diarrhea. This is at first watery, later contains large amounts of blood. If the animals live a sufficient length of time, paralysis may occur, the animal may fall to one side or may drag its posterior extremities. It is a remarkable fact that intravenous inoculation gives rise to intestinal inflammation of a severe nature, unquestionably due to the excretion of the poison by the intestinal mucosa and limited, usually, to the ce- cum and colon, rarely attacking the small intestine. Flexner,^ who has experimented extensively upon this question, believes it probable that most of the pathological lesions occurring in the intestinal canal of dysen- tery patients are referable to this excretion of dysentery toxin, rather than to the direct local action of the bacilli. Toxins from the Shiga-Kruse type are the most potent and those which cause paralysis. Immunization with Dysentery Bacilli. — The immunization of small animals, such as rabbits and guinea-pigs, against dysentery bacilli, especially those of the Shiga type, is attended with much difficulty, owing to the great toxicity of the cultures. Nevertheless, successful results may be accomplished by the administration of extremely small doses of living or dead bacilli, increased very gradually and at sufficient intervals. Horses may be more easily immunized. The serum of such actively immunized animals contains agglutinins in considerable con- centration and of a specificity sufficiently illustrated in the preceding section dealing with the identification of the various species. For diagnostic purposes in human beings, the agglutination reaction, accord- ing to the technique of the Widal reaction for typhoid fever, has been utilized by Kruse ^ and others. According to most observers, normal \uman serum never agglutinates dysentery bacilli in dilutions greater than one in twenty, while the serum of dysentery patients will often be active in dilutions as high as one in fifty. Bactericidal substances have been demonstrated in the serum of im- munized animals as well as in the serum of diseased human beings. These have been determined, in vitro, by Shiga,^ and by the intraperito- neal technique of Pfeiffer by Kruse."* Bacteriolysis may take place in high dilutions of the serum, and has recently been used for the differen- tiation of the types of the dysentery bacilli by Ohno.^ True antitoxins in immune sera have been recently described by Kraus and Doerr,* » Flexner, Jour. Exp. Med., 8, 1906. 2 Kruse, Deut. med. Woch., 1901. 3 Shiga, Zeit. f. Hyg., xli. * Kruse, Deut. med. Woch., 1903, « Ohno, Philippine Jour, of Sci., vol. i, 1906. « Kraus und Doerr, loc. cit. THE DYSENTERY BACILLI 445 Q K ^ is H O < m 1 V pi h 3 E 1 •5 f 1 1 1 1 t 1 1 1 J) ■Co 1^ 3^ ■ ■ a o c X ■ ■ ■ D 1 ■ ■ ■ a 1 ■ ■ D a . J ■ D D D J E M D D 1 ■ D D a J E 0 D D 1 1 ■ ■ D a n n V c 1 a M i 0 a. 3> V <« 2 « 0 « « •0 I i n 3 E N 0 & V •> I 3* b tt s « c V w u ! E 1 5 •1 0 a ! «) o d a 1 Ol. M « i a. y t a. Q w 3 .2 i E 8 1 c I 0 0 s S 0) a 3 II Is is .2 C • •502 >. 22 lis 446 PATHOGENIC MICROORGANISMS Passive immunization of animals and human beings with the serum of highly immunized horses has been variously attempted by Shiga/ Kraus/ Gay/ and others. All these observers have reported distinct benefit to the patients and a reduction of the mortality by the use of such sera. Striking and rapid reductions of temperature and rapid con- valescence, after a sjngie injection, have occasionally been observed. The earlier workers were inclined to attribute the beneficial results of these sera entirely to their bactericidal value. Todd has recently demonstrated that the mixture of such an immune serum with solutions of toxin and exposure of the mixture at 37.5° C. for a half hour would produce almost complete neutralization of the poison, thus demonstrating that at least a large part of the beneficial action of the immune sera was due to a true antitoxic process. Be- cause of the different varieties of dysentery bacilli, polyvalent serum has been recommended. Prophylactic vaccination of human begins with dead dysentery cultures has, so far, led to no practical result. Shiga, Deut. med. Woch., 1901, ^ Kraus, loc. cit. 8 Gay, Penn. Med. Bull., 1902. CHAPTER XXX BACILLUS MUCOSUS CAPSULATUS, BACILLUS LACTIS AEROGENESj BACILLUS PROTEUS BACILLUS MUCOSUS CAPSULATUS (Bacterium pneumonice, Friedldnder' s bacillus, Pneumohacillus) In 1882, Friedlander ^ announced the discovery of a microorganism which he believed to be the incitant of lobar pneumonia and which, in his original communications, he described as a " micrococcus." A superficial morphological resemblance between Friedlander's microorganism and Diplococcus lanceolatus, now recognized as the most frqquent cause of lobar pneumonia, led, at first, to much confusion, and it was not until several years later, owing to the careful researches of Frankel ^ and of Weichselbaum,^ that the " micrococcus " of Friedlander was recognized as a short, encapsulated bacillus which occurred in lobar pneumonia exceptionally only. Similar bacilli were subsequently found by other observers, bacilli which, mainly upon morphological grounds, are classified together as the ^'Friedlander group,'' or the ''group of Bacillus mUcosus capsulatus." Morphology and Staining. — The Friedlander bacillus is a short, plump bacillus with rounded ends, subject to great individual variations as to size. Its average measurements are from 0.5 to 1.5 micra in width and 0.6 to 5 micra in length. Forms approaching both extremes may be met with in one and the same culture. The short, thick forms, frequently found in animal and human lesions, are almost coccoid and account for Friedlander's error in first describing the bacillus as a tnicrococcus. The bacilli may be single, in diplo-form, or in short chains. They are non-motile and possess no flagella. Spores are not formed. The bacillus is characteristically surrounded by a well-developed capsule which is most perfectly demonstrated in preparations taken directly from some animal fluid, such as the secretion or exudate from infected areas. It is also seen, however, in smears made from agar ^Friedlander, Virchow's Arch., Ixxxvii, 1882; Fort. d. Med., i, 1883; ibid., ii. 1884. 2 Frankel, Zeit. f. klin. Med., x, 1886. • Weichselbaum, Med. Jahrb., Wien, 1886. 447 448 ^ PATHOGENIC MICROORGANISMS" or gelatin cultures. The capsule is usually large, twice or three times the size of the bacillus itself. When seen in chains or in groups, several bacilli may appear to be inclosed in one capsule. Prolonged cultivation on agar or gelatin may result in disappearance of the capsule. The bacil- lus is easily stained with the ordinary dyes, but is decolorized when stained by the Gram-method. Capsules may often be seen when the more intense anilin dyes are employed. They are brought out with much regularity by any of the usual capsule stains. Cultivation. — B. mucosus capsulatus is easily cultivated. It grows • ■ m. t^ ' ?*■ ■ » \ • * * * • • • * ■ %■ %» * - ' • ■«»■■■ <»'' ,. ■ ■ ♦ .♦ %. . " 1 • «9«- ^ * .» • '«>'. - ' ■'"if , ■% • * • ,, .'-V ■• .: • _fe. *»: . \" • * • -* •• . # ■1. ■ * ■ . % . .V ■» Fig. 94. — Bacillus mlcosus capsulatus. readily on all the usual culture media, both on those haying a meat- infusion basis and on those made with meat extract. Growth takes place at room temperature (18° to 20°) and more rapidly at 37.5° C. A temperature of 60° C. and over kills the bacilli in a short time. The ther- mal death-point according to Sternberg is 56° C. Growth ceases below 10° to 12° C. Kept at room temperature and protected from drying, the bacillus may remain alive, in cultures, for several months. The bacillus is not very fastidious as to reaction of media, growing BACILLUS MUCOSUS CAPSULATUS 449 equally well on moderately alkaline or acid media. It is aerobic and facultatively anaerobic; growth under anaerobic conditions, however, is not luxuriant. On agar, growth appears in the form of grayish-white mucus-hke colonies, having a characteristically slimy and semi-fluid appearance. Colonies have a tendency to confluence, so that on plates, after three or four days, a large part of the surface appears as if covered with a film of glistening, sticky exudate, which, if fished, comes off in a tenacious, stringy manner. It is often possible to make a tentative diagnosis of the bacillus from the appearance of this growth. In broth, there is rapid and abundant growth, with the formation of a pellicle, general clouding, and later the development of a profuse, stringy sediment. Stab cultures in gelatin show, at first, a white, thin line of growth along the course of the puncture. .Soon, however, rapid growth at the top results in the formation of a grayish mucoid droplet on the surface, which, enlarging, gives the growth a nail-like appearance. This nail-shape was originally described by Friedlander and regarded as diag- nostic for the bacillus. The gelatin is not fluidified. As the culture grows older the entire surface of the gelatin tube may be covered with growth, flowing out from the edges of the nail-head. The gelatin acquires a darker color and there may be a few gas bubbles below the surface. Micro- scopically, colonies on gelatin plates have a smooth outline and a finely granular or even homogeneous consistency. On blood serum, a confluent mucus-like growth appears. On 'potato, abundant growth appears, slightly more brownish in color than that on other media. In pepton solutions, there is no indol formation. In milk, there is abundant growth and marked capsule develop- ment. Coagulation occurs irregularly. In considering the general cultural characteristics of the Fried- lander bacillus, it must hot be forgotten that we are dealing with a rather heterogeneous group, the individuals of which are subject to many minor variations. Capsule development, lack of motility, in- ability to fluidify gelatin, failure to form indol, and absence of spores, are characteristics common to all. In size, general appearance, gas forma- tion, and pathogenicity, individual strains may vary much, one from the other. Strong ^ has studied various races as to gas formation and ^ Strong, Cent. f. Bakt., xxv, 1899. 450 PATHOGENIC MICROORGANISMS concludes that most strains form gas from dextrose and levulose, but that lactose is fermented by some only. About two-thirds of the gas formed is hydrogen, the rest CO2. Acid formation, according to Strong, is also subject to much variation among different races. Similar studies by Perkins ^ show that most of the ordinary cultural characteristics of bacilli of this group are extremely variable and can not serve as a basis for differentiation. Reactions on sugars, however, are more constant. Perkins suggests the following tentative division classes on this basis: I. All carbohydrates fermented with the formation of gas. II. All carbohydrates, except lactose, fermented with the formation of gas. III. All carbohydrates, except saccharose, fermented with the formation of gas. Type I. corresponds to B. aerogenes (Migula), Type II. to B. Friedlander or Bacterium pneumoniae (Migula), and Type III. to Bacillus lactis aerogenes. Differentiation by means of sermn reactions has not proved satis- factory. 2 Pathogenicity. — When Friedlander first described this microorganism, he assumed it to be the incitant of lobar pneumonia. Subsequent re- searches by Weichselbaum ^ and others have shown it to be etiologically associated with pneumonia in about seven or eight per cent of all cases. The percentage in this country is probably lower. Such cases can often be diagnosed by the presence of the bacilli in the sputum, which is pecul- iarly sticky and stringy. Cases of Friedlander pneimionia are extremely severe and usually fatal. The bacillus has been found in cases of ulcer- ative stomatitis and nasal catarrh; in two cases of severe tonsillitis in children; in the pus from suppurations in the antrum of Highmore and the nasal sinuses (Frankel and others), and in cases of fetid cory- za (ozena), of which disease it is supposed by Abel^ and others to be the specific cause. Whether the ozena bacillus represents a separate 1 Perkins, Jour, of Infect. Dis., I, No. 2, 1904. 2 J, G. Fitzgerald, who has recently made a careful study of the mucosus cap- sulatus group has concluded that present methods do not permit a subdivision of these organisms into separate species. He offers the following ' ' tentative suggestion ' ' : It is conceivable that mutations based on the necessity of maintaining a parasitic existence have caused Gram-negative badlli found normally in the body elsewhere than in the intestinal tract to develop capsules for protection and a new group has arisen which we designate B. mucosus capsulatus; and the varieties B. aerogenes and B. acidi lactici connect the group with the non-encapsulated colon group." 3 Weichselbaum, loc. cit, * Abel, Zeit. f . Hyg., xxi. BACILLUS OF RHINOSCLEROMA 451 species or not, can not at present be decided. The bacillus of Fried- lander has been found in empyema fluid, in pericardial exudate (after pneumonia), and in spinal fluid. ^ Isolated cases of Friedlander bacillus septicemia have been described. ^ Being occasionally a saprophytic inhabitant of the normal intestine, it has been believed to be etiologic- ally associated with some forms of diarrheal enteritis. B. mucosus capsulatus is pathogenic for mice and guinea-pigs, less so for rabbits. Inoculation of susceptible animals is followed by local in- flammation and death by septicemia. If inoculation is intraperitoneal, there is formed a characteristically mucoid, stringy exudate. The question of immunization against bacilU of the Friedlander group is still in the stage of experimentation. Immunization with care- fully graded doses of dead bacilh has been successful in isolated cases. Specific agglutinins in immune serum have been found by Clairmont,^ but irregularly and potent only against the particular strain used for the immunization. OTHER BACILLI OF THE FRIEDLANDER GROUP Bacillus of Rhinoscleroma. — This bacillus, discovered by v. Frisch * in 1882, is a plump, short rod, with rounded ends, morphologically almost identical with Friedlander's bacillus; it is non-motile and pos- sesses a distinct capsule. Although at first described as Gram-positive, it has been shown to be decolorized with this method of staining. Cul- turally it is almost identical with B. mucosus capsulatus. It forms slimy colonies, has a nail-like appearance in gelatin stab cultures, and in pepton solutions produces no indol. It differs from B. mucosus cap- sulatus (Wilde ^) in forming no gas in dextrose bouillon, in producing no acid in lactose bouillon, and in never coagulating milk. Pathogenicity. — The bacillus of rhinoscleroma is but moderately pathogenic for animals delicately susceptible to the bacillus of Fried- lander. Rhinoscleroma, the disease produced by this bacillus in man, consists of a slowly growing granulomatous inflammation, located usu- ally at the external nares or upon the mucosa of the nose, mouth, pharynx, or larynx. It is composed of a number of chronic, hard, nodular swellings, which, on histological examination, show granulation tissue and productive inflammation. In the meshes of the abundant 1 Jdger, Zeit. f. Hyg., xix. ^ Howard, Johns Hopkins Hosp. Bull., 1899. ' Clairmont, Zeit. f. Hyg., xxxix. ^ v. Frisch, Wien. med. Woch., 1882. 6 Wilde, Cent. f. Bakt., xx, 1896. 452 PATHOGENIC MICROORGANISMS connective tissue lie many large swollen cells, the so-called "Mikulicz cells." 1 The rhinoscleroma baciUi He within these cells and in the intercellular spaces. They can be demonstrated in histological sections and can be cultivated from the lesions, usually in pure culture. Rhino- scleroma is rare in America. It is most prevalent in Southeastern Europe. The disease is slowly progressive and comparatively intract- able to surgical treatment, but hardly ever affects the general health unless by mechanical obstruction of the air passages. B. Ozaense. — The work of Abel 2 and others has shown that ozena, or fetid nasal catarrh, is almost al- ways associated with a bacillus morphologically and culturally almost iden- tical with B. mucosus capsulatus. The bacillus can not be definitely sepa- rated from the latter. Ac- cording to Wilde ^ it forms no gas in dextrose bouillon and is less pathogenic for mice than B. Friedlander. Whether it is a separate species, or merely an atypical form changed by environment, can not be stated at present. Perez Bacillus of Ozsena. — Perez ^ in 1899 described another micro- organism which he connects etiologically with ozaena. The Perez bacillus is Gram-negative, pleomorphic, non-motile and non-capsulated. It grows easily on ordinary media, does not liquefy gelatin, and makes indol. Its cultures have a characteristic fetid odor. Intravenously injected into rabbits it seems to produce a localized lesion in the nasal cavity on the turbinated bones. Hofer ^ has also isolated it, but recent work leaves its importance as the causative agent in doubt. 1 Mikulicz, Arch, f . Chir., xx, 1876. 2 ^5^;^ Zgit. f. Hyg., xxi. ^ Wilde, loc. cit. * Perez, Animal de I'Inst. Past. 1899. •^ Hofer, Wien. klin. Woch., vol. 26, pp. 1011 and 1628. Fig. 95. — Bacillus of Rhinoscleroma. Sec- tion of tissue showing the microorganisms within MikuUcz cells. (After Frankel and Pfeiffer.) BACILLUS LACTIS AEROGENES 453 BACILLUS LACTIS AEROGENES Bacillus lactis aerogenes is the type of a group which is closely similar to the colon group and often distinguished from it with difficulty. It was first described by Escherich ^ in 1885 who isolated it from the feces of infants. Since then it has been learned that this bacillus is almost constantly present in milk, and, together with one or two other micro- organisms, is the chief cause of the ordinary souring of milk. Apart from its occurrence in milk, moreover, the bacillus is widely distributed in nature, being found in feces, in water, and in sewage. It is distinguish- able from the colon bacillus chiefly by the fact that it is less motile, possesses no flagella, hardly ever forms chains, and, when cultivated upon suitable media, especially milk, it possesses a distinct capsule. It differs from the colon bacillus, furthermore, in that it is capable of fermenting polysaccharids, such as starch, and does not form indol upon pep- ton media. It is distinguishable from the bacillus of Friedlander (B. mucosus capsulatus), according to Wilde,^ by its more energetic gas formation in dextrose broth, its abiUty to produce acid on lactose media, and its invariable coagulation of milk. Unlike the colon bacilli, it does not form gas on Dulcit.^ It differs from the other important non-duldt fermentation, the bacillus acidi lactici, in fermenting saccharose. Ten- tative differentiations of these bacilli may be made as follows: Dextrose. Lactose. Saccharose. Dulcit. B. coli communis + + + + + + + + + B. coli communior B. lact. aerogenes B. lact. acidi _ It grows upon the simplest media, is a facultative anaerobe, and grows most abundantly at a temperature between 25° and 30° C. Upon agar and gelatin it grows with a heavy white growth, the colonies of which have a tendency to confluence and are more mucoid in appearance than are those of Bacillus coli. In hroth, it causes a general clouding and a pellicle. The cultures have a slightly sour or cheesy odor. On potato, the growth is heavy and gas is formed. On milk, there is rapid coagulation and acid formation. It is charac- teristic of this bacillus that it is capable of producing a large amount of acid, chiefly lactic, and of being able to withstand these large amounts of acid without being injured by them. 1 Escherich, Fort. d. Med., 16, 17, 1885. 2 Wilde, Cent, f . Bakt., xx, 1896. ^ Jackson. 454 PATHOGENIC MICROORGANISMS The pathogenicity of Bacillus lactis aerogenes for man is slight. Its chief claims to importance lie in its milk-coagulating properties and its almost constant presence in the human intestine. In infants, it may give rise to flatulence and it has been occasionally observed as the sole incitant of cystitis. Among such cases rare instances have been observed in which it has formed gas in the bladder (pneumaturia). When this occurs the urine is not ammoniacal but remains acid. Different strains of this bacillus vary much in their pathogenicity for animals. Wilde claims that it is more pathogenic for white mice and gionea-pigs than is the bacillus of Friedlander. He speaks of it as the most virulent member of this group. Kraus, writing in Fluegge's "Mikroorganismen/' rates its pathogenicity less high. Closely related to this bacillus, as well as to those of the Friedlander group, is an encapsulated bacillus isolated from a case of broncho- pneumonia by Mallory and Wright,^ which is strongly pathogenic for mice, guinea-pigs, and rabbits. BACILLI OP THE PROTEUS GROUP , The bacilli of this group have little pathological interest, but are im- portant because of the frequency with which they are encountered in routine bacteriological work. They may confuse the inexperienced because of a superficial similarity to bacilli of the colon-typhoid group. In form they may be short and plump or long and slender, staining easily with anilin dyes and decolorizing with Gram's method. They are ac- tively motile and possess many flagella. Individuals stain irregularly, often showing unstained areas near the center. The so-called Bacillus proteus vulgaris described by Hauser ^ m 1885 is the type of the group. Bacilli of this group are widely distributed, being found in water, soil, air, and wherever putrefaction takes place. In fact, proteus is one of the true putrefactive bacteria possessing the power to cause the cleav- age of proteids into their simplest radicles. Bacillus proteus vulgaris grows best at temperatures at or about 25° C. and develops upon the simplest media. It is a facultative anae- robe and forms no spores. In hroth, it produces rapid clouding with a pellicle and the formation of a mucoid sediment. In gelatin, the colonies are characteristically irregular, giving the name to this group. Gelatin is rapidly liquefied. Liquefaction, however, is diminished or even inhibited under anaerobic conditions. 1 Mallory and Wright, Zeit. f. Hyg., xx, 1895, . ^Hauser, "Ueber Faulniss-Bakt.," Leipzig, 1885. BACILLUS PROTEUS 455 On agar and other solid media, as well as upon gelatin before lique- faction has taken place, characteristic colonies are produced. From the central flat, grayish-white colony nucleus, numerous irregular streamers grow out over the surrounding media, giving the colony a stellate appearance. On potato, it forms a dirty, yellowish growth. In milkf there is coagulation and an acid reaction at first; later the casein is redissolved by proteolysis. Blood serum is often liquefied, but not by all races. A great many really dissimilar bacteria have been described under the name of Proteus. The type of the group is the so-called Proteus vulgaris (Hauser, 1885). Other organisms spoken of as proteus are the Proteus mirahilis, which differs in slower gelatin liquefication from vulgaris, the Proteus Zenkeri, which does not liquefy gelatin, the Proteus septicus, and the Bacillus Zopfi, a Gram-positive organ- ism. A good many of these were formerly classified as of Bacterium termo. Closely related is the slow liquefying organism known as Bacillus cloacce, common in sewage. There is no group which so urgently requires study as this, since organisms belonging here are so often found in the human body and human excreta. In urine we have encountered a non-gelatin liquefy- ing Gram-negative bacillus belonging to this group which has given us much trouble in identification. As far as we can establish any general characteristics for the group at all, we may say that they are Gram-negative, non-spore-bearing, motile bacilli, which on the surface of gelatin plates show colonies characterized by spreading streamers, most of which liquefy gelatin, a few of which, however, do not. All of them ferment dextrose and saccharose with gas, but do not attack lactose. The pathogenic powers of proteus are slight. Large doses injected into animals m^y give rise to localized abscesses. In man proteus in- fections have been described in the bladder, in most cases, however, together with some other microorganism. The Urohacillus lique- faciens septicus described by Krogius was a variety of this group. Epidemics ^ of meat poisoning have been attributed to the proteus family by some observers. Thus Wesenberg^ cultivated a proteus from putrid meat which had caused acute gastroenteritis in sixty- three individuals. Similar epidemics have been reported by Silber- schmidt, ^ Pfuhl, * and others. 1 Schnitzler, Cent, f . Bakt., viii, 1890. 2 Wesenberg, Zeit. f. Hyg., xxviii, 1898. 3 Silberschmidt, Zeit. f . Hyg., xxx, 1899. _ * Pfuhl, Zeit. f, Hyg., xxxv, 1900. CHAPTER XXXI BACILLUS TETANI Lockjaw or tetanus, though a comparatively infrequent disease, has been recognized as a distinct clinical entity for many centuries. The infectious nature of the disease, however, was not demonstrated until 1884, when Carlo ^ and Rattone succeeded in producing tetanus in rabbits by the inoculation of pus from the cutaneous lesion of a human case. Nicolaier,^ not long after, succeeded in producing tetanic symp- toms in mice and rabbits by inoculating them with soil. In connec- tion with the lesions produced at the point of inoculation, Nicolaier described a bacillus which may have been Bacillus tetani, but which he was unable to cultivate in pure culture. Kitasato,^ in 1889, definitely solved the etiological problem by obtaining from cases of tetanus pure cultures of bacilli with which he was able again to produce the disease in animals. Kitasato succeeded where others had failed because of his use of anaerobic methods and his elimination of non-spore-bearing con- taminating organisms by means of heat. His method of isolation was as follows: The material containing tetanus bacilli was smeared upon the surface of agar slants. These were permitted to develop at incubator temperature for twenty-four to forty-eight hours. At the end of this time the cultures were subjected to a temperature of 80^ C. for one hour. The purpose of this was to destroy all non-sporulating bacteria, as well as aerobic spore-bearers which had developed into the vegetative form. Agar plates were then inoculated from the slants and exposed to an atmosphere from which oxygen had been com- pletely eliminated and hydrogen substituted. On these plates colonies of tetanus bacilli developed. Morphology and Staining. — ^The bacillus of tetanus is a slender bacil- lus, 2 to 5 micra in length, and 0.3 to 0.8 in breadth. The vegetative forms which occur chiefly in young cultures are slightly motile and are » Carlo e Rattone, Giomale d. R. Acad. d. Torino, 1884. ' Nicolaier, Inaug, Diss,, Gottingen, 1885. 8 Kitasato, Deut. med. Woch., No. xxxi, 1889. 456 BACILLUS TETANI 457 seen to possess ^ numerous peritrichal flagella, when stained by special methods. After twenty-four to fgrty-eight hours of incubation, the length of time depending somewhat on the nature of the medium and the degree of anaerobiosis, the bacilli develop spores which are char- acteristically located at one end, giving the bacterium the diagnostic drumstick appearance. As the cultures grow older the spore-bearing forms completely super Fig. 96. — ^Bacillus tetani. Spore stain. sede the vegetative ones. Very old cultures contain spore-bearing bacilli and spores only. The tetanus bacillus is easily stained by the usual anilin dyes, and reacts positively to Gram's stain. Flagella staining is successful only when very young cultures are employed. Distribution. — In nature, the tetanus bacillus has been found by Nicolaier and others to occur in the superficial layers of the soil. The 1 Vottaler, Zeit. f. Hyg., xxvii. 458 PATHOGENIC MICROORGANISMS earth of cultivated and manured fields seems to harbor tnis organism with especial frequency, probably because of its presence in the dejecta of some of the domestic animals. Biological Characteristics. — The bacillus of tetanus is generally de- scribed as an obligatory anaerobe. While it is unquestionably true that growth is ordinarily obtained only in the complete absence of oxygen, various observers, notably Ferran ^ and Belfanti,^ have successfully habituated the bacillus to aerobic conditions by the gradual increase of oxygen in cultures. Habituation to aerobic condi- tions has usually been accompanied by diminution or loss of pathogenicity and toxin-formation. Anaerobic conditions may likewise be dispensed with if tetanus bacilli be grown in symbiosis with some of the aerobic bacteria. The addition to culture media of suitable carbohydrates, and of fresh sterile liver tissue, has also been found to render it less exacting as to absolute anaerobiosis.^ Anaerobically cultivated. Bacillus tetani grows readily upon meat-infusion broth, which it clouds within twenty-four to thirty-six hours. Anaerobic broth cultures may be simply made by covering the surface of the medium with a layer of albolin or any other oil, and removing the air by boiling. Upon meat-infusion gelatin at 20° to 22° C. the tetanus bacillus grows readily, ■ growth becoming visible during the second or third day. There is slow fluidification of the gelatin. On agar, at 37.5° C, growth appears within forty- eight hours. Colonies on agar plates present a rather characteristic appearance, consisting of a compact center surrounded by a loose meshwork of fine fila- ments, not unlike the medusa-head appearance of subtilis colonies. In agar stabs, fine radiating processes growing out in all directions from the central stab tend to give the culture the appearance of a fluff of cotton. Milk is a favorable culture medium and is not coagulated. On potato, growth is delicate and hardly visible. 1 Ferran, Cent. f. Bakt., xxiv, No. 1. 2 Belfanti, Arch, per le sci. med., xvi. « Th. Smith, Brown, and Walker, Jour. Med. Res,, N, S., ix, 1906. Fig. 97. — Young Tetanus Culture IN Glucose Agar. BACILLUS TETANI 459 The most favorable temperature for the growth of this bacillus is 37.5° C. Shght alkalinity or neutrality of the culture media is most ad- vantageous, though moderate acidity does not altogether inhibit growth. All the media named may be rendered more favorable still by the ad- dition of one or two per cent of glucose, maltose, or sodium formate.^ In media containing certain carbohydrates, tetanus bacilli produce acid. In gelatin and agar, moderate amounts of gas are produced, consisting chiefly of CO 2, but with the admixtures of other volatile substances which give rise to a characteristically unpleasant odor, not unlike that of putrefying organic matter. This odor is due largely to H2 S and methylmercaptan. The vegetative forms of the tetanus bacillus are not more resistant against heat or chemical agents than the vegetative forms of other microorganisms. Tetanus spores, however, will resist dry heat at 80° C. for about one hour, live steam for about five minutes; five per cent carbolic acid kills them in twelve to fifteen hours; one per cent of bichlo- rid of mercury in two or three hours. Direct sunlight diminishes their virulence and eventually destroys them.^ Protected from sunlight and other deleterious influences, tetanus spores may remain viable and virulent for many years. Henrijean * has reported her success in producing tetanus with bacilli from a splinter of wood infected eleven years before. Pathogenicity. — The comparative infrequency of tetanus infection is in marked contrast to the wide distribution of the bacilli in nature. Introduced into the animal body as spores, and free from toxin, they may often fail to incite disease, easily falling prey to phagocytosis and other protective agencies before the vegetative forms develop and toxin is formed. The protective importance of phagocyto- sis was demonstrated by Vaillard and Rouget,'* who introduced tetanus spores inclosed in paper sacs into the animal body. By the paper cap- 1 Kitasato, Zeit. f. Hyg., 1891. 2v. Eisler und Pribram, in Levaditi, "Handbuch," etc., Jena, 1907. * Henrijean, Ann. de la soc. med. chir. de Liege, 1891. * Vaillard et Rouget, Ann. de I'inst. Pasteur, 1892. 30 Fig. 98.— Older Tetanus Culture IN Glucose Agar. 460 PATHOGENIC MICROORGANISMS sules the spores were protected from the leucocytes, not from the body fluids. Nevertheless, tetanus developed in the animals. The nature of the wound and the simultaneous presence of other microorganisms seem to be important factors in determining whether or not the tetanus bacilli shall be enabled to proliferate. Deep, lacerated wounds, in which there has been considerable tissue destruction, and in which chips of glass, wood splinters, or grains of dirt have become embedded, are particularly favorable for the development of these germs. The injuries of compound fractures and of gunshot wounds are especially liable to supply these conditions, and the presence in such wounds of the common pus cocci, or of other more harmless parasites, may aid materially in furnishing an environment suitable for the growth of the tetanus bacilli. Apart from its occurrence following trauma, tetanus has been not infrequently ob- served after childbirth,^ and isolated cases have been reported in which it has followed diphtheria and ulcerative lesions of the throat.^ A definite period of incubation elapses between the time of infection with tetanus bacilli and the development of the first symptoms. In man this may last from five to seven days in acute cases, to from four to five weeks in the more chronic ones. Experimental inoculation ( f guinea-pigs is followed usually in from one to three days by rigidity of the muscles nearest the point of infection . This spastic condition rapidly extends to other parts and finally leads to death, which occurs within four or five days after infection. Autopsies upon human beings or animals dead of tetanus reveal few and insignificant lesions. The initial point of infection, if at all evident, is apt to be small and innocent in appearance. Further than a general and moderate congestion, the organs show no pathological changes. BacilH are found sparsely even at the point of infection, and have been but rarely demonstrated in the blood or viscera. Nicolaier succeeded in producing tetanus with the organs of infected animals in but eleven out of fifty-two cases. More recently, Tizzoni ^ and Creite ^ have suc- ceeded in cultivating tetanus bacilli out of the spleen and hearths blood of infected human beings. The researches of Tarozzi ^ and of Canfora ^ have shown also that spores may be transported from the site of inoculation to the liver, spleen, and other organs, and there he dormant for as long as fifty-one days. If injury of the organ is experimentally practised and dead tissue 1 Baginsky, Deut. med. Woch., 1893. ^ Creite, Cent, f . Bakt., xxxvii. 2 Foges, Wien. med. Woch., 1895. ^ Tarozzi, Cent, f, Bakt. Orig. xxxviii. ' Tizzoni, Ziegler's Beit., vii. ^ Canfora, Cent, f . Bakt.* Orig. xlv. BACILLUS TETANI 461 6r blood clot produced, the spores may develop and tetanus ensue. These experiments may explain, cases of so-called cryptogenic tetanus. Tetanus Toxin. — The pathogenicity of the tetanus bacillus depends entirely upon the soluble toxin which it produces. This toxin is produced in suitable media by all strains of virulent tetanus bacilli, individual strains showing less variation in this respect than do the separate strains of diphtheria bacilli. While partial aerobiosis does not completely elimi- nate toxin formation, anaerobic conditions are by far more favorable for its development. The jnedium most frequently employed for the production of tetanus toxin is neutral or slightly alkaline beef -infusion bouillon containing five- tenths per cent NaCl and one per cent pepton. Glucose, sodium formate, or tincture of litmus may be added, but while these substances increase the speed of growth of the bacilli they do not seem to enhance the de- gree of toxicity of the cultures. Glucose is said even to be unfavorable for strong toxin development. It is important, too, that the bouillon shall be freshly prepared.^ There does not seem to be any direct relationship between the amount of growth and the degree of toxicity of the cultures. Under anaerobic conditions in suitable bouillon and grown at 37.5° C, the maximum toxin content of the cultures is reached in from ten days to two weeks. After this time the toxin deteriorates rapidly. Tetanus toxin has been produced without resort to anaerobic methods by several observers, notably by Debrand,^ by cultivating the bacilli in bouillon in sjrmbiosis with Bacillus subtilis. By this method, Debrand claims to have produced toxin which was fully as potent as that produced by anaerobic cultivation. The tetanus toxin, in solution in the bouillon cultures, may be sepa- rated from the bacteria by filtration through Berkefeld or Chamberland filters. Since the poison in such filtrates deteriorates very rapidly, much more rapidly even than diphtheria toxin, various methods have been devised to obtain the toxin in the solid state. The most useful of these is precipitation of the poison out of solution by oversaturation with ammonium sulphate.^ Very little of the toxin is lost by this method and, thoroughly dried and stocked in vacuum tubes, together with an- hyd]X)us phosphoric acid, it may be preserved indefinitely without dete- rioration. The precipitate thus formed is easily soluble in water or 1 Vaillard et Vincent, Ann. de I'inst. Pasteur, 1891. 2 Debrand, Ann. de I'inst. Pasteur, 1890, 1902. 3 Brieger und Cohn, Zeit. f . Hyg., xv. 462 PATHOGENIC MICROORGANISMS salt solution, and therefore permits of the preparation of uniform solu- tions for purposes of standardization. Brieger and Boer ^ have also succeeded in precipitating the toxin out of broth solution with zinc chloride. Vaillard and Vincent ^ have procured it in the dry state by evaporation in vacuo. Brieger and Cohn,^ Brieger and Boer,^ and others have attempted to isolate tetanus poison, removing the proteids from the ammonium sul- phate precipitate by various chemical methods. The purest preparations obtained have been in the form of fine yellowish flakes, soluble in water, insoluble in alcohol and ether. Solutions of this substance have failed to give the usual proteid reactions. The toxin when in solution is extremely sensitive to heat. Kita- sato ^ states that exposure to 68° C. for five minutes destroys it com- pletely. Dry toxin is more resistant,^ often withstanding temperatures of 120° C. for more than fifteen minutes. Exposure to direct sunlight destroys the poison in fifteen to eighteen hours. ^ Interesting experiments as to the action of eosin upon tetanus toxin have been carried out by various observers. Flexner and Noguchi ^ found that five per cent eosin added to the toxin would destroy it within one hour. This action is ascribed to the photodynamic power of the eosin. The toxin exerts an extremely low osmotic pressure and is easily destroyed by electric currents. Tetanus toxin is one of the most pov/erful poisons known to us. Filtrates of broth cultures, in quantities of 0.000,005 c.c, will often prove fatal to mice of ten grams weight. Dry toxin obtained by ammo- nium sulphate precipitation ^ is quantitatively even stronger, values of 0.000,001 grams as a lethal dose for a mouse of the given weight not being uncommon. Brieger and Cohn^^ succeeded in producing a dry toxin capable of killing mice in doses of 0.000,000,05 gram. Different species of animals show great variation in their suscepti- 1 Brieger und Boer, Zeit. f . Hyg., xxi. 2 Vaillard et Vincent, Ann. de I'inst. Pasteur, 1891. 2 Brieger und Cohn, loc. cit. * Brieger und Boer, Zeit. f . Hyg., xxi. ^ Kitasato, Zeit. f . Hyg., x. ^ Morax et Marie, Ann. de I'inst. Pasteur, 1902. ' Fermi und Pemossi, Cent, f . Bakt., xv. 8 Flexner and Noguchi, "Studies from Rockefeller Inst.," v., 1905. ^ Brieger und Cohn., loc. cit. ^^ Brieger und Cohn., Zeit. f. Hyg., xv. BACILLUS TETANI 463 bility to tetanus toxin. Human beings and horses are probably the most susceptible species in proportion to their body weight. The common domestic fowls are extremely resistant. Calculated for grams of body weight, the horse is twelve times as susceptible as the mouse, the guinea- pig six times as susceptible as the mouse. The hen, on the other hand, is 200,000 times more resistant than the mouse. After the inoculation of an animal with tetanus toxin there is always a definite period of incubation before the toxic spasms set in. This period may be shortened by increase of the dose, but never entirely eliminated.^ When the toxin is injected subcutaneously, spasms begin first in the muscles nearest the point of inoculation. Intravenous inoculation,^ on the other hand, usually results in general tetanus of all the muscles. The feeding of toxin does not produce disease, the poison being passed through the bowel unaltered. The harmful action of tetanus toxin is generally attributed to its affinity for the central nervous system. Wassermann and Takaki ^ show that tetanus toxin was fully neutralized when mixed with brain substance. Other organs — liver and spleen, for instance — showed no such neutralizing power. The central origin of the tetanic contractions was made very evident by the work of Gumprecht,'^ who succeeded in stop- ping the spasms in a given region by division of the supplying motor nerves. The manner in which the toxin reaches the central nervous system has been extensively investigated, chiefly by Meyer and Ransom, and Marie and Morax. Meyer and Ransom ^ from a series of careful experi- ments reached the conclusion that the toxin is conducted to the nerve centers along the paths of the motor nerves. Injected into the circu- lation,® the toxin reaches simultaneously all the motor nerve endings," producing general tetanus. In this case too, therefore, the poison from the blood can not pass directly into the central nervous system, but must follow the route of nerve tracts. These observations have been of great practical value in that they pointed to the desirability of the injection of tetanus antitoxin directly into the nerves and the central nervous system in active cases. 1 Courmont et Doyen, Arch, de phys., 1893. 2 Ransom, Deut. med. Woch., 1893. ' Wassermann und Takaki, Berl. klin. Woch., 1898. * Gumprecht, Pfliiger's Arch., 1895. ^ Meyer und Ransom, Arch, f . exp. Pharm. u. Path., xlix. « Marie et Morax, Ann. de I'inst. Pasteur, 1902. 464 PATHOGENIC MICROORGANISMS Tetanolysin. — Tetanus bouillon contains, besides the "tetano- spasmin*' described above which produces the familiar symptoms of the disease, another substance discovered by Ehrlich ^ and named by him "tetanolysin." Tetanolysin has the power of causing hemolysis of the red blood corpuscles of various animals, and is an entirely separate substance from tetanospasmin. It may be removed from toxic broth by admixture of red blood cells, is more thermolabile than the tetano- spasmin, and gives rise to an antihemolysin when injected into animals. For the production, standardization, and use of tetanus antitoxin, see p. 220 et seq. Therapeutic Value of Tetanus Antitoxin. — Until recently tetanus antitoxin was chiefly useful as a prophylactic, from 1,000 to 5,000 units being given by deeply subcutaneous or intramuscular injection. Its use after the onset of symptoms had, however, been fraught with much disappointment, largely because it was not possible to influence the toxin which had already become united with the substance of the nerve tis- sues. Recently, however, more perfect methods of administration have been devised with which better results have been achieved. The most successful of these seems to be the method of Park and Nicoll.^ They carry out a spinal puncture taking off a moderate amount of spinal fluid, and then inject slowly by gravity from 3,000 to 5,000 units of tetanus antitoxin in a volume of from 3 to 10 c.c. At the same time 10,000 units are given intravenously or intramuscularly. 1 Ehrlich, Berl. klin. Woch., 1898. 2 Park and Nicoll, Jour, of the A. M. A., vol. 63, July, 1914, p. 235. CHAPTER XXXII BACILLUS OF SYMPTOMATIC ANTHRAX, BACILLUS OF MALIGNANT EDEMA, BACILLUS AEROGENES CAPSULATUS, BACILLUS BOTULINUS BACILLUS OF SYMPTOMATIC ANTHRAX {Bacillus anthracis symptomatid, Rauschbrand, Charbon symptomatique, Sarcophysematos hovis) Symptomatic anthrax is an infectious disease occurring chiefly among sheep, cattle, and goats. It is spoken of as ''quarter-evil" or "blackleg." The disease has never been observed in man. It was formerly confused with true anthrax, because of a superficial similarity between the clinical symptoms of the two maladies. Bacteriologically, the two microorganisms are in entirely different classes. Symptomatic anthrax is of wide distribution and infection is usually through the agency of the soil in which the bacillus is present, in the form of spores which may retain viability for several years. Morphology and Staining. — The bacillus of symptomatic anthrax is a bacillus with rounded ends, being about four to six micra long, and five-tenths to six-tenths micra wide. It is usually seen singly and never forms long chains. The bacillus in its vegetative form is actively motile and possesses numerous flagella placed about its periphery. In artificial media it forms spores which are oval, broader than the rod itself, and placed near, though never actually at, the end of the bacillary body. This gives the bacillus a racket-shaped appearance. It is readily stained with the usual anilin dyes, but is easily decolor- ized by Gram's method of stainii^g. However, von Hibler claims that when verv carefullv stained the bacillus can be shown to be Gram- positive — at least when taken from the animal body.^ Cultivation. — The bacillus is a strict anaerobe. It was obtained in pure culture first by Kitasato.^ Under anaerobic conditions it is easily ^ von Hibler, Kolle, Wassermann, etc., p. 792, vol. iv. 2 Kitasato, Woch. f . Hyg., 1889. 465 466 PATHOGENIC MICROORGANISMS cultivated upon the usual laboratory media, all of which are more favorable after the addition of glucose, glycerin, or nutrose. In all media there is active gas formation, which, owing to an admixture of butyric acid, is of a foul, sour odor. The bacillus is not very delicate in its requirements of a special reaction of media, growing equally well on those slightly acid or slightly alkaline. On gelatin plates, at 20° C, colonies appear in about twenty-four hours, usually round or oval, with a compact center about which fine radiating filaments form an opaque halo. The gelatin is fluidified. V \ Pig. 99. — Bacillus of Symptomatic Anthrax. (After Zettnow.) Surface colonies upon agar plates are circular and made up of a slightly granular compact center, from which a thinner peripheral zone emanates, containing microscopically a tangle of fine threads. In agar stabs, at 37.5° C, growth appears within eighteen hours, rapidly spreading from the line of stab as a diffuse, fine cloud. Gas formation, especially near the bottom of the tube, rapidly leads to the formation of bubbles and later to extensive splitting of the medium. In gelatin stab cultures growth is similar to that in agar stabs, though less rapid. Pathogenicity.— Symptomatic anthrax bacilli are pathogenic for cattle, sheep, and goats. By far the largest number of cases, possibly the only spontaneous ones, appear among cattle. Guinea-p'igs are very susceptible to experimental inoculation. Horses are very little suscep- BACILLUS OF SYMPTOMATIC ANTHRAX 467 tible. Dogs, cats, rabbits, and birds are immune. Man also appears to be absolutely immune. Spontaneous infection occurs by the en- trance of infected soil into abrasions or wounds, usually of the lower extremities. Infection depends to some extent upon the relative de- gree of virulence of the bacillus — a variable factor in this species. Twelve to twenty-four hours after inoculation there appears at the point of entrance a soft, puffy swelling, which on palpation is found to emit an emphysematous crack- ling. The emphysema spreads rapidly, often reaching the abdomen and chest within a day. The course of the disease is extremely acute, the fever high, the general prostration extreme. Death may result within three or four days after inoculation. At autopsy the swollen area is found to be infiltrated with a thick exudate, blood-tinged and foamy. Subcutaneous tissue and muscles are edematous and crackle with gas. The internal organs show parenchymatous degeneration and hemorrhagic areas. The bacilli, immediately after death, are found but sparsely distributed in the blood and internal organs, but are demonstrable in enormous numbers in the edema surrounding the central focus. If carcasses are allowed to lie unburied for some time, the bacilli will attain a general distribution, and the entire body will be found bloated with* gas, the organs filled with bubbles. Practically identical conditions are found after experimental inocula- tion. Toxins. — According to the investigations of Le- clainche and Vallee,^ the bacillus of symptomatic anthrax produces a soluble toxin. It is not formed to any extent in ordinary broth, but is formed in considerable quantities in broth containing blood or albuminous ani- mal fluids. The best medium for obtaining toxin, according to the same authors, is the bouillon of Martin,^ made up of equal parts of veal infusion and a Fig. 100.— Ba- cillus OF Symp- tomatic Anthkax. Culture in glucose agar. 1 Leclainche et Vallee, Ann. de I'inst. Pasteur, 1900. 2 Martin, Ann. de I'inst. Pasteur, 1898. 468 PATHOGENIC MICROORGANISMS pepton solution obtained from the macerated tissues of the stomachs of pigs. The toxin contained in filtrates of such cultures is quite resistant to heat, but rapidly deteriorates if free access of air is allowed. Immunity. — Active immunization against the bacillus of symptom- atic anthrax was first accomplished by Arloing ^ and his collaborators by the subcutaneous inoculation of cattle with tissue-extracts of in- fected animals. The work of these authors resulted in a practical method of immunization which is carried out as follows: Two vaccines are prepared. Vaccine I consists of the juice of in- fected meat, dried and heated to 100° C. for six hours. Vaccine II is a similar meat-juice heated to 90° C, for the same length of time. By the heating, the spores contained in the vaccines are attenuated to relatively different degrees. Vaccine I in quantities of 0.01 to 0.02 c.c. is emulsified in sterile salt solutions and injected near the end of the tail of the animal to be protected. A similar quantity of Vaccine II is injected in the same way fourteen days later. This method has been retained in principle, but largely modified in detail by various workers. Kitt ^ introduced the use of the dried and powdered whole meat instead of the meat juice, and made only one vaccine, heated to 94° C, for six hours. This method has been largely used in this country.^ Passive immunization with the serum * of actively immunized sheep and goats has been used in combination with the methods of active immunization. BACILLUS OF MALIGNANT EDEMA (Bacillus oedernatis maligni, Vibrion septique) In 1877, Pasteur ^ described a bacillus which he had found in guinea- pigs and rabbits experimentally inoculated with putrefying animal tissues. This bacillus, which he named "Vibrion septique," he suc- ceeded in cultivating only under anaerobic conditions and in an impure state, and described as its pathognomonic characteristics the formation of an extensive edema in and about the point of inoculation. ^ Arloing, Cornevin, et Thomas, " Le Charbon Sympt.," etc., Paris, 1887. Kei. from Grassberger und Schattenfroh, Kraus und Levaditi, "Handbuch," etc., vol i, pt. 2. 2 Kitt, Ref . from Grassberger und Schattenfroh, loa cit. » Ti<»nnrt of Bureau of Animal Ind., Wash., 1902. * Arloing, Leclainche, et Vallee, loc. cit. 6 Pasteur, Bull, de I'acad. de mM., 1877, p. 793. BACILLUS OF MALIGNANT EDEMA 469 Koch/ who studied this infection in connection with his work upon anthrax in 1881, called attention to the fact that the bacillus described by Pasteur did not produce a true septicemia, and suggested the term "Bacillus of malignant edema," which is now in general use. Gaffky ^ found that, apart from its presence in putrid material, the bacillus occurred in the upper layers of garden soil and in dust. It has since been found to be widely distributed in nature and in the intestines of animals and of man. Its wide diistribution is unques- tionably due to the great resistance of its spores. Morphology and Staining. — The bacillus of malignant edema is a ^^^J Fic3. 101. — Bacillus of Malignant Edema. (After Frankel and Pfeiffer.) long slender rod, not unlike the anthrax bacillus, but decidedly more slender. Its average measurements are 1 micron in thickness and 3 to 8 micra in length. It usually occurs as single rods, but frequently appears in long threads showing irregular subdivisions. Often no sub- divisions can be seen and the threads appear as long, homogeneous filaments. These threads are less frequently seen in preparations from solid media than in those from bouillon or edema fluid. 'The bacilli are motile and possess numerous laterally placed flagella. Their motil- ity is never very marked and is often entirely absent. The bacillus » Koch, Mitt. a. d. kais. Gesundheitsamt, i, 1881, p. 52 et seq. 2 Gaffky, Mitt. a. d. kais. Gesundheitsamt, 1881. 470 pathogejnic microorganisms produces spores at temperatures above 20° C, which are oval, irregularly- placed either in the center or slightly nearer one or the other end, and cause a bulging of the bacillary body. It is readily stained by any of the usual anilin dyes. Stained by Gram's method it is decolorized. Cultivation. — Bacillus oedematis maligni is strictly anaerobic. Under anaerobic conditions it develops readily upon any of the usual artificial media. The bacillus is not very sensitive to the reaction of media and grows more luxuriantly in all media to which glucose has been added. In all media it forms, by the cleavage of proteids, putridly offensive gases. In gelatin at room temperature, colonies develop in about three days as small grayish spherical growths, which microscopically show an arrangement in radial filaments. The gelatin is fluidified. In gelatin stab cultures growth begins as a white column extending to within a centimeter of the top of the medium. Soon irregularly radiating processes de^^elop laterally and gas bubbles appear, breaking up the medium. Stab cultures in agar show growth within twenty- four to thirty-six hours at 37.5° C, appearing at first as a white line, but soon showing a cloud-like lateral extension along the entire line of the stab. If sugar is present bubbles appear throughout the medium. In broth there is general clouding and a granular sediment; no pellicle is formed. Milk is slowly coag- ulated. On blood serum growth is very luxuriant. On potato, a medium used in the earliest studies of the bacillus by Gaffky, the bacillus grows readily. Isolation may be accomplished by any of the ordinary anaerobic . plating methods. The bacillus can usually be obtained for subsequent isolation by injection of a susceptible animal with soil, especially that of gardens or manured fields. Pathogenicity. — The bacillus is pathogenic for mice, guinea-pigs, rabbits, horses, dogs, sheep, pigs, some birds, and man. Cattle were formerly regarded as immune, an opinion which has since been found to Vie erroneous. Fig. 102.— Ba- cillus OF Ma- lignant Edema. Culture in glu- cose agar. BACILLUS AEROGENES CAPSULATUS 471 Subcutaneous inoculation of pure culture into a susceptible subject produces, within twenty-four to thirty-six hours, an acute edematous inflammation about the point of inoculation. The edema extends throughout the subcuticular and deeper layers, and consists of thin, slightly bloody fluid. Neighboring lymph nodes become swollen and hemorrhagic. In the mixed infections of accidental inoculation, but more rapidly in experimental inoculations with pure cultures, gas is formed and consequent subcutaneous emphysema. Together with this there are symptoms of general toxemia. In the smaller test animals this disease is usually fatal. At autopsy the bacilli are found in the edema fluid about the local lesion. At autopsies done soon after death, the organisms are not found in the blood or internal organs. Later they may be generally distributed throughout the body. In mice only may the bacilli enter the blood stream before death. The internal organs of animals dead of this infection usually show parenchymatous degen- eration and occasionally hemorrhages. Malignant edema is not a frequent disease. It has been occasionally observed in horses, in cattle, and in sheep. In man the infection usually appears after traumatism or secondarily after compound fractures or upon the site of suppurating wounds. Isolated cases have been de- scribed as arising after hypodermic injections. One case has been reported as arising in the uterus after instrumental abortion. Immunity. — Recovery from an infection with the bacillus of malig- nant edema produces immunity against subsequent infections.^ The bacillus in fluid media produces small amounts of a soluble toxin which in bacteria-free filtrates is capable of killing guinea-pigs. Relatively large quantities of filtrate must be employed. Roux and Chambcrland,^ the first to work upon these toxins, were able, by means of them, to immunize guinea-pigs. Similar immunity could be produced by treat- ment with the toxic, filtered sera of animals dead of the disease.^ BACILLUS AEROGENES CAPSULATUS The bacillus finds wide distribution in nature, being found in soil, dust, and brackish water, and in the normal intestinal tracts of human beings and mammals.^ ^ Arlaing et Chxiuveau, Ann. de med, vet., 1884. 2 Rouxet Chamberland, Ann. de I'inst. Pasteur, 1887. ^Sanfelice, Zeit. f. Hyg., xiv. * Although B. aerogenes capsulatus and B. phlegmonis emphysematosa are sep- arately described in many books, notably Migula's ''System d. Bakteriologie," the microorganisms have been shown beyond question to be identical and are acknowl- edged to be so by Frankel himself. 472 PATHOGENIC MICROORGANISMS There has been endless confusion in the classification of the so- called aerogenes capsulatus organisms, a group which has been known in different countries by a large variety of names. Its first description is probably that by Welch and Nuttall ^ in 1892, who gave it the name B. aerogenes capsulatus, by which name, or briefly Welch bacillus, it is most commonly designated by American bacteriologists. Similar or identical organisms are tfie B. phlegmonis emphysematosoe of Fraenkel, the B. enteriditis sporogenes of Klein, the B. perf ring ens of Veillon and Zuber, and the Granulobacillus saccharohutyricus liquefaciens immohilis of Schattenfroh and Grassberger, and a number of others less prominent in the literature. The task of classifying these various organisms is one of great technical difficulty, and, as in so many other groups of bacteria orig- inally described as a single species, we are now learning that instead of complete identification of individual isolations, one with another, we can merely draw a circle about a certain group, recognizing close relationship morphologically, culturally, and in relation to infections of man and animals. After a very careful study of many different strains, Simmonds comes to the conclusion that the Welch bacillus, so-called, is roughly identical with those mentioned, including also the bacilli once associated by Achalme with rheumatic infections. The term B. Welchii, he concludes, does not represent a fixed species but a closely related group of bacteria not yet fully classified even by his own extensive studies. It is often hard to determine in the case of organisms of such varied activities exactly what can be regarded as a cardinal characteristic important for classifying purposes. The organisms treated of in this group are all large Gram-positive, non-motile bacilli, with rounded ends, rarely occurring in chains, and anaerobic. Spore formation is in- constant and occurs only in an alkaline medium. When fermentable sugars, and consequently free acid, are present, no spore formation takes place. Milk is fermented with the formation of butyric acid. The intravenous injection of animals, especially rabbits, usually pro- duces death with an enormous swelling of the body by the formation of gas, which burns with a pale blue flame. Capsules may be seen when the smears are made directly from animal tissues, but are almost universally missed in organisms taken from culture. As to minor cultural characteristics, so many variations are ob- served that it is hardly worth while to summarize them. * See Simmonds, Monographs of the Rockefeller Institute, No. 5, Sept. 27, 1915. BACILLUS AEROGENES CAPSULATUS 473 Pathogenicity. — ^It is extremely interesting that, though a path- ogenic anaerobe, no one has so far been able to show satisfactorily the production of a true exotoxin, though such a claim has been made sev- eral times. Poisonous products that have been obtained from cultures have not conformed with the characteristics of true exotoxins as we recognize them now. Moreover, the reaction of toxin-containing fil- trates, as Simmonds points out, has been almost always acid, and Mc- Campbell and others have suggested that the toxicity of such cultures is due to the presence of butyric acid. It is interesting in this connec- tion also that Herter, who gave a great deal of attention to the presence of these organisms in the bowel and attributed to them etiological rela- tionship with many intestinal disorders, believed that diarrhea following intestinal putrefaction was due to a very large extent to the irritation produced by ammonium butyrate, the Welch bacilli being- responsible for the appearance of the butyric acid. ^ '^ The search for endotoxins, too, seems to have been unsuccessful, and hemolysins have been irregularly present in cultures studied by various investigators. Great differences in virulence have been observed in the study of various strains. A good many of the strains seem to have great patho- genicity for the ordinary laboratory animals, especially for guinea-pigs and rabbits. Yet certain non-sporulating forms have shown little or no virulence. However, variation in virulence seems to fluctuate con- siderably according to the culture medium and the symbiotic conditions under which the organisms have been cultivated. In man, fortunately, there seems to be a relatively powerful resistance against Welch bacillus infections. In fact, many investigators have believed that the Welch bacillus is primarily a saprophyte and requires the presence of dead tissue or physiological injury before it can penetrate and thrive in the human body. It has been observed in gas cysts of the brain, in "gangrenous" pneumonia, in appendix abscesses, in puerperal sepsis after abortion, in general septicemia, and in many other conditions. However, in many of these cases it may be quite reasonable to question the primary character of the "gas bacillus" infection, and in others again, the Welch bacillus may be regarded as an ante-mortem invasion, a supposition strengthened by the study of terminal infections made by Flexner, in which such a possibility is plainly demonstrated. Thus, while the Welch bacillus may unquestionably invade the human body and do much injury, and even be regarded in many cases as probably the direct cause of death, its invasion in many of these cases must be 474 PATHOGENIC MICROORGANISMS regarded as not a primary invasion, but as one associated with and made possible by either traumatic injury or the co-operation of other invasive germs such as streptococci, staphylococci and others. There may, of course, in addition to these be rkre instances in which the Welch bacillus, owing to excessive virulence of the strain or relatively low resistance of the infected individual, becomes the primary invader. It is easy to understand under these conditions why severe trauma associated with the carrying of soil would form ideal conditions for the production of Welch bacillus infections. The most interesting chapter in the pathogenicity of the Welch bacillus is its relation to intestinal disease. Simmonds confirms by his own work the work of many others that the Welch bacillus may be regarded as a normal inhabitant of the intestines of man. In 19 babies under one year of age he found Welch bacilli in the stools of 8. He points out that it has been found even in nursing infants by a number of investigators. ■ Herter ^ has associated the presence of the Welch bacillus in the bowel with pernicious anaemia. However, no such relationship can at present be assumed on the basis of available work. Isolation. — The bacillus may, of course, be isolated by anaerobic plating methods. It is best isolated, however, from mixed cultures by animal inoculation. If, for instance, it is desired to obtain it from a mixed culture or from feces, a suspension of about 1 c.c. of the sus- pected material is made in 5 c.c. of sterile. salt solution. This is thor- oughly emulsified and filtered through a sterile paper. One to two c.c. of this suspension is then injected into the ear vein of a rabbit. After four or five minutes the rabbit is killed. It is then placed in the incu- bator for five to eight hours. At the end of this time, the animal is usually found tensely distended with gas. At autopsy, gas bubbles will be found distributed throughout th-e organs, most characteristically in the liver, where isolated bubbles are found covering the surface. From these bubbles cultures or smears may be taken for identification. Identification is easily made from its morphology, its capsule, lack of motility, and gas formation. The interest of bacteriologists in the Welch bacillus has lately been attracted particularly by the frequent occurrence of the co-called gas bacillus infections in soldiers in the trenches. Apparently, these infections are due to the fact that dirty clothing and dirt from the skin 1 Herter, " Bacterial Infection of the Intestinal Tract," New York, Macmillan, 1907. BACILLUS BOTULINUS 475 are carried into the wounds with the missiles, and the degree of trauma furnishes the conditions under which the bacillus can grow. These infections are characterized by a very marked destruction of muscle fibers, histological examination showing a complete disinte- gration of the cells. It is probable that a great deal of acid, especially butyric acid, is formed from the glycogen of the muscle by the gas bacillus, and that this contributes to the intoxication of the patient. Stewart and West have studied such infections in guinea-pigs in our laboratory and have noticed great frequency of gastric ulcer formation in guinea-pigs succumbing from the bacillus. Since they could produce similar ulcers by intravenous injections of acetic acid, they concluded that much of the systemic injury caused by the bacillus may be associated with acid production. Weinberg and Sacquep^e have recently claimed a true toxin in cul- tures isolated from wounded soldiers and have produced an intoxication against this poison. No such true toxin has been found in the cultures studied in our laboratory and this, together with the experiences of many other workers, inclines us to believe that the organism studied by Weinberg and Sacquepee is distinct from the bacteria ordinarily classified with the Welch bacillus. BACILLUS BOTULINUS Meat poisoning was formerly regarded as universally dependent upon putrefactive changes taking place in infected meat, resulting in the production of ptomains or other harmful products of bacterial putre- faction. It was not until 1888 that certain of these cases were definitely recognized as true bacterial infections, in which the preformed poison probably aided only in establishing the infection. Gartner, in that year, discovered the Bacillus enteritidis, a microorganism belonging to the group of the paracolon bacilli, and demonstrated its presence both in the infecting meat and in the intestinal tracts of patients. The char- acteristics of this type of meat poisoning have been discussed more particularly in the section describing the bacillus of Gartner and its allied forms. There is another type of meat poisoning, however, which is not only much more severe (ending fatally in almost 25 per cent of the cases), but is characterized by a clinical picture more significant of a profound systemic toxemia than of a mere gastroenteric irritation. The etio- logical factor underlying this type of infection was first demonstrated 31 476 PATHOGENIC MICROORGANISMS by Van Ermengem/ in 1896, and named Bacillus botulinus. Van Ermengem isolated the bacillus from a ham, the ingestion of which had caused disease in a large number of persons. Of the thirty-four individuals who had eaten of it, all were attacked, about ten of them very severely. Van Ermengem found the bacilli in large numbers lying between the muscle fibers in the ham, and was able to cultivate the same microorganism from the stomach and spleen of one of those who died of the infection. The results of Van Ermengem have been confirmed by Romer,^ and others. Morphology and Staining. — Bacillus botuHnus is a large, straight rod with rounded ends, 4 to 6 micra in length by 0.9 to 1.2 micra in thickness. The bacilli are either single or grouped in very short chains. Involu- tion forms are numerous on artificial media. The bacillus is slightly motile and possesses from four to eight undulated flagella, peripherally arranged. Spores are formed in suitable media, most regularly in glucose-gelatin of a distinctly alkaline titer. The spores are oval and usually situated near the end of the bacillus, rarely in its center. Spores are formed most abundantly when cultivation is carried out at 20° to 25° C, and are usually absent when higher temperatures are em- ployed. The bacillus is easily stained by the usual aqueous aniline dyes, and retains the anilin-gentian-violet when stained by Gram. It is necessary, however, in carrying out Gram's stain to decolorize carefully with alco- hol since overdecolorization is easily accomplished, leaving the result doubtful. Cultivation. — The bacillus is a strict anaerobe. In anaerobic en- vironment it is easily cultivated on the usual meat-infusion media. It grows most readily at temperatures about 25° C, less luxuriantly at temperatures of 35° C. and over. The bacillus is delicately susceptible to the reaction of media, growing only in those which are neutral or moderately alkaline. In deep stab cultures in one per cent glucose agar, growth is at first noticed as a thin, white column, not reaching to the surface of the medium. Soon the medium is cracked and split by the abundant formation of gas. On agar plates, the colonies are yellowish, opalescent, and round, and show a finely fringed periphery. On gelatin, at 20° to 25° C., growth is rapid and abundant, and 1 Van Ermengem, Cent, f . Bakt., xix, 1896; Zeit. f. Hyg., xxvi, 1897. 2 Romer, Cent. f. Bakt., xxvii, 1900. BACILLUS BOTULINUS 477 differs little from that oh agar, except that, besides the formation of gas, there is energetic fluidification of the medium. On glucose-gelatin plates, Van Ermengem describes the colonies as round, yellowish, transparent, and composed of coarse granules which, along the periphery in the zone of fluidification, show constant motion. The appearance of the surface colonies on glucose-gelatin plates is regarded by the discov- erer as diagnostically characteristic. In glucose hroth there is general clouding and large quantities of gas are formed. At 35° C. and over, the gas formation ceases after four or five days, the broth becoming clear with a yellowish-white flocculent sediment. At lower temperatures this does not occur. Milk is not coagulated and disaccharids and polysaccharids are not fermented The gas formed in cultures consists chiefly of hydrogen and methane. All cultures have a sour odor, like butyric acid, but this is not so offensive as that of some of the other anaerobic organisms. The bacillus lives longest in gelatin cultures, but even upon these, transplantations sHould be done every four to six weeks, since the spores of this bacillus show less viability and resistance than do those of most spore-formers. Pathogenicity. — Botulism, or allantiasis, as noticed in human beings, is, as far as we know, always due to ingestion of infected meat, usually of ham, canned meats, or sausages (botulus = sausage). Symptoms appear only after a definite period of incubation which varies from twenty-four to forty-eight hours. The first definite symptoms are chilliness, trembling, and giddiness. These manifestations are soon followed by headache, occasionally by vomiting. In contradistinction' to the meat poisonings caused by other microorganisms, those, due to Bacillus botulinus may show few or no symptoms directly referable to the intestinal tract. The chief diagnostic characteristics of the disease are a group of symptoms referable to toxic interference with the cranial nerves. Loss of accommodation, dilated pupils, ptosis, aphonia, and dysphagia may occur. Fever is usually absent. Consciousness is rarely lost. The characteristic symptoms may be produced in various animals by injection of living cultures or culture filtrates, i.e., toxins. The most susceptible animals are guinea-pigs. These may be killed by the injec- tion of minute quantities of bouillon cultures or of toxin. Preceding death, which occurs within twenty-four to thirty-six hours, there m.ay be general motor paralysis, dyspnea, and hypersecretion of mucus from nose and mouth. Guinea-pigs may be infected per os as well as by 478 *" PATHOGENIC MICROORGANISMS hypodermic injections. Cats, mice, and monkeys are highly susceptible; rabbits are less so, but still favorable subjects for experimental studies. Birds, especially pigeons, are highly resistant, but react typically to large doses. Autopsies upon man or animals dead of botulism show general hyperemia of the organs with much parenchymatous degenera- tion and many minute hemorrhages. The bacilli have been found in the spleen after death,i but Van Ermengem does not believe that they are generally distributed during the course of the disease. It is believed by most of those who have studied this disease that poisoning in the human subject is due to the toxins preformed and ingested within the infected meat by this bacillus. Experiments have shown that little or no poison is produced by the bacilli after gaining entrance to the himian or animal body. The Toxin of B. botulinus — Bacillus botulinus produces disease chiefly by means of a strong soluble toxin secreted by it, and absorbed by the infected subject. This toxin is active in animals and presumably in man, not only when injected subcutaneously, but also when intro- duced through the gastrointestinal canal. The poison has been par- ticularly studied by Brieger and his collaborators. It is obtained in filtrates of alkaline bouillon cultures. It has been precipitated out of the filtrate by Brieger and Kempner ^ by means of a three per cent zinc chlorid solution (2 volumes of 3 per cent ZnCl2). The toxin thus ob- tained was sufficiently powerful to kill a 250-gram guinea-pig in fifty hours. Specific action of the toxin for the nerve-cells of the spinal ganglia has been shown by Marinesco.^ A specific antitoxin has been produced by Kempner and Pollack.^ 1 Stricht, Quoted from Van Ermengem, in KoUe und Wassermann. 2 Brieger und Kempner, Deut. med. Woch., xxxiii, 1897. 3 Marinesco, Compt. rend, de I'acad. des sci., Nov., 1896. * Kempner und Pollack, Deut. med. Woch., xxxii, 1897. CHAPTER XXXIII THE TUBERCLE BACILLUS In view of the clinical manifestations of tuberculosis, it is not sur- prising that the infectious nature of the disease has been suspected for many centuries. Even Fracastor had remarkably modern ideas concern- ing its transmission. Inoculation by means of tuberculous material was first successfully accomplished by Klencke, in 1843, and, more elabo- rately, by Villemin,^ in 1865. It was not until 1882, however, that Koch 2 succeeded in isolating and cultivating the tubercle bacillus. Baumgarten ^ had previously seen the bacillus in tissue sections, but his researches were limited to purely moi-phological observations. Koch, in addition to demonstrating the bacillus in tuberculous tissues from various sources, produced characteristic lesions in guinea-pigs and other animals by infecting them with pure cultures, and established beyond doubt the etiological relationship of the bacillus to the disease. Morphology. — ^Tubercle bacilli appear as slender rods, 2 to 4 micra in length, 0.2 to 0.5 micra in width. Their ends are usually rounded. The rods may be straight or slightly curved; their diameters may be uniform throughout; more often, however, they appear beaded and irregularly stained. The beaded appearance is due to different causes. Unstained spaces may occur along the body of the bacillus, especially in old cultures. These are generally regarded as vacuoles. The bodies of the bacilli, on the other hand, may bulge slightly here and there, often in three or four places, showing oval or rounded knobs which stain with great depth and are very resistant to decolorization. These thickenings were formerly regarded as spores, but in view of the fact that the bacilli are not more resistant against heat and disinfectants than other vegeta- tive forms, this interpretation is probably incorrect. The bacilli are said to possess a cell membrane which confers upon them their resistance against drying and entrance of stains. This membrane gives a cellulose reaction and is believed to contain most of the waxy substances which can be extracted from the cultures. 1 Villemin, Gaz. hebdom., 1865. 2 Koch, Berl. klin. Woch., 1882; Mitt. a. d. kais. Gesundheitsamt, 1884. ^Baumgarten, Virchow's Arch., Ixxxii. 479 480 PATHOGENIC MICROORGANISMS Various observers, notably Nocard and Roux/ Mafucci,^ and Klein,^ have demonstrated branched forms of the tubercle bacillus. These ob- servations, variously extended and confirmed, make it probable that Bacillus tuberculosis is not a member of the family of schizomycetes, 0 "'' "^ #. % ■ 4 ■ . ' \ .> ■ -> ' \ i| ■■«'■ 1"* y f ' lii H « • ^Ik ltf»\' ^..' Fig. 103. — ^Tubercle Bacilli in Sputum. but belongs rather to the higher bacteria, closely related to the actino- myces. Staining. — Tubercle bacilli do not stain easily with the ordinary anilin dyes; to these they are made permeable only by long exposure or by heating of the staining solution. Once stained, however, the dye is tenaciously retained in spite of treatment with alcohol and strong 1 Nocard et Roux, Ann. de I'inst. Pasteur, 1887. ^ Mafucci, Zeit. f. Hyg., ii. 8 KUin, Cent, f . Bakt., 1890. THE TUBERCLE BACILLUS 481 acids. For this reason, this bacillus, together with some other bacteria to be mentioned later, is spoken of as '' acid-fast." The acid-fast nature of the bacillus seems to depend upon the fatty substances contained in it,^ and has furnished the basis for differential staining methods. All the staining methods devised for the recognition of the tubercle bacillus thus depend upon the use of an intensely penetrating staining solution, followed by vigorous decolorization which deprives all but the acid-fast group of their color. Counterstains of any of the weaker dyes may then be used to stain the decolorized elements. One of the first of the staining solutions to be of practical use was the anilin-water-gentian- violet solution of Ehrlich ^ (11 c.c. saturated alcoholic gentian- violet to 89 c.c. 5 per cent anilin water). This dye, although of sufficient penetrating power, has the disadvantage of deteriorating rapidly and has in practice been almost entirely displaced by ZiehFs ^ carbol-fuchsin solution. (Fuchsin 1 gm. in 10 c.c. alcohol absolute, added to 90 c.c. 5 per cent carbolic.) This staining solution is the one now in general use and is employed as follows: Thin smears, on slides or cover-slips, are covered with the dye and gently heated. In the case of cover-glasses, these may be floated, face downward, in staining dishes filled with the dye. The dye is allowed to act for about three miimtes, steaming but not allowed to boil. At the end of this time the preparation is washed either with 5 per cent nitric acid, 5 to 20 per cent sulphuric acid, or 1 per cent hydro- chloric acid, until most of the red color has disappeared (a few seconds), and the preparation appears pale pink. This results in decolorization of all microorganisms with the exception of members of the acid-fast group. Thorough washing in 80 to 95 per cent alcohol is now employed to complete the decolorization. The preparation is then rinsed in water and counterstained with 1 per cent aqueous methylene-blue. Tubercle-bacillus staining has been further simplified by Gabbett,'' who combines decolorization and counterstaining. In this method preparations are stained with Ziehl's carbol-fuchsin as in the preceding; they are then rinsed in water and covered with a solution containing methylene-blue 1 gram, concentrated sulphuric acid 25 grams, and distilled water 100 c.c. This is allowed to act for from two to four 1 Bienstock, Fort. d. Med., 1886; Weyl, Deut. med. Woch., 1891. 2 Ehrlich, Deut. med. Woch., 1882; Weigert, Deut. med. Woch., 1885. 3 Ziehl, Deut. med. Woch., 1883; Neelsen, " Lehrb. d. allg. Path.," 1894. » GabbetL Lancet, 1887. 482 PATHOGENIC MICROORGANISMS minutes, at the end of which time all elements in the preparation except the acid-fast bacilli will be decolorized and counterstained. Tubercle bacilli in very young culture are often not acid-fast and it is not always possible to demonstrate acid-fast bacilli in pus from cold abscesses in sputum, in serous exudates, and in granulomatous lesions of the lymph nodes which can be shown by animal inoculation to be tuberculous. Much ^ demonstrated in such material Gram-positive granules which lay singly in short chains or in irregular clumps, and which he believed to be non-acid-fast tubercle bacilli. He found similar granules in cultures of tubercle bacilli which showed on further incuba- tion numerous acid-fast bacillary forms. His work has been repeatedly confirmed, and there seems little doubt but that these granules are really tubercle bacilli. Their demonstration is not, however, of great diag- nostic value, as other bacilli form granules of the same appearance. Small rods and splinters are also found which stain by Gram's method, but not by carbol-fuchsin.^ To find "Much's granules," smears or sections are steamed in a solution of methyl violet B.N. (10 c.c. of saturated alcohoHc solution of the dye in 100 c.c. of distilled water containing 2 per cent phenol). They are then treated with Gram's iodine solution 1-5 minutes; 5 per cent nitric acid 1 minute; 3 per cent hydrochloric acid 10 seconds; ab- solute alcohol and acetone equal parts, until decolorized. The granules may be stained by other modifications of Gram's method. Weiss ^ has devised a combination stain. One part of Much's methyl violet is mixed with three parts of Ziehl's carbol-fuchsin and filtered; slides are stained for 24 to 48 hours in the mixture. They are then decolorized as in Much's method and counterstained with Bismarck brown or safranin 1 per cent. Both acid-fast and Gram-positive forms are stained by this method and in the red may be seen blue-black granules. While the acid-fast group of bacteria is composed of a number of organisms to be mentioned later, a few only of these offer difficulties of differentiation from the tubercle bacillus. Those to be considered practically are the bacillus of leprosy and that of smegma. The latter bacillus, because of its distribution, is not infrequently found to ccgi- taminate feces, urine, or even sputum, and it is important to apply to suspected specimens one or the other of the stains devised for the differentiation of the smegma bacillus from Bacillus tuberculosis. The 1 Much, Berl. klin. Woch., 1908, xlv, 700. 2 Liebermeister, Deutsche med. Woch., 1909, xxxv, 1324. 3 Weiss, Berl. klin. Woch., 1909, xlvi, 1797. THE TUBERCLE BACILLUS 483 one most frequently employed is that of Pappenheim. ^ The preparations are stained in hot carbol-fuchsin as before; the carbol-fuchsin is then poured off without washing and the preparation immersed in a solution made by saturating a 1 per cent alcoholic solution of rosolic acid with methylene-blue and adding 20 per cent of glycerin. In such prepara- tions tubercle bacilli remain red, smegma bacilli appear blue. Stained by Gram, tubercle bacilli retain the gentian-violet. When tubercle bacilli are very sparsely present in sputum and other material it may be impossible to find them by direct examination, and often the only method of finding them will be animal inoculation. However, a number of methods have been devised by which the bacilli may be concentrated in such a way that they may be found even when a few only are present. One of these is to add peroxide of hydrogen to the sputum. By this the mucus is dissolved out and the soUd particles settle or may be centrifugalized. A method very commonly employed to-day is that which depends on the use of ''antiformin." This is a preparation used extensively for the cleansing of vats in breweries. It is described by Rosenau ^ as consisting of equal parts of liquor sodse chlorinatse and a 15 per cent solution of caustic soda. The formula for liquor sodse chlorinatee he gives as: Sodium carbonate 600 Chlorinated lime 400 Distilled water 4,000 If sputum is poured into a 10 to 15 per cent solution of antiformin and allowed to stand for several hours, most of the other elements of the sputum, cells, and bacteria, will dissolve out, and acid-fast baciUi be left in the residue. Strangely enough they are not killed by this- process and if sufficiently washed may be cultivated or can produce lesions in guinea-pigs. Isolation and Cultivation. — Tubercle bacilli are not easily cultivated. Their slowness of growth precludes isolation by plating. The first isola- tions by Koch ^ were made upon coagulated blood serum from tuber- culous tissue. Isolation from tuberculous material may be aided by inoculation into guinea-pigs. These animals will withstand the acute infection .-K^ i>h^ Pappenheim, Berl. klin. Woch., 1898. }•, -^ Rosenau, "Preventive Medicine and Hygiene," D. Appleton, N. Y., 1913; Uhlenhuth, Berl. klin. Woch., No. 29, 1908. , 3 Kochf loc. cit. 484 PATHOGENIC MICROORGANISMS produced by the contaminating organisms and succumb later (four to six weeks) to tuberculosis. The bacilli may then be obtained by culti- vations from lymph nodes or other foci which contain only tubercle bacilli. When isolation from sputum is attempted, whether directly or by means of animal inoculation, the sputum may be rendered com- paratively free from contaminating bacteria by washing. The sputum is rinsed in running water to free it from pharyngeal mucus. It is then washed in eight or ten changes of sterile water. The material selected is taken from the center of the washed mass, if possible from the flakes of caseous material visible in such sputum. For the isolation of tubercle bacilli from sputum and other materials in which contaminating bacteria of other species are present, Petroff ^ has devised an excellent method which has been tried, out and used with success in our laboratory by Dr. H. R. Miller. The principles on which Petroff's method rests are, first of all, the bactericidal power of 3% sodium hydroxid on non- acid-fast bacteria, and the selective action of dyes like gentian violet on bac- terial growth, as first practically utilized by Churchman (See page 140). The medium used by Petroff is made as follows: I. Meat Juice. 500 grams of beef or veal are infused in 500 c.c. of a 15% solution of glycerin in water, in a cool place. After 24 hours the meat is squeezed in a sterile press and the infusion collected in a sterile beaker. II. Eggs. The shells of the eggs are sterilized by 10 minute immersion in 70% alcohol. They are broken into a sterile beaker, well mixed and filtered through "sterile gauze. One part of meat juice is added to two parts of egg by volume. III. Gentian Violet. 1% alcoholic solution of gentian violet is added to make a final proportion of 1 : 10,000. The three ingredients are well mixed. The medium is tubed and inspissated as usual. Petroff recommends for sputum the following technique: Equal parts of sputum and 3% sodium hydroxid are shaken and incubated at 38° C. for 15 to 30 minutes, the time depending on the consistency of the sputum. The mixture is neutralized to litmus with hydrochloric acid and centrifugalized. The sedi- ment is inoculated into the medium described above. Pure cultures are ob- tained in a large proportion of cases. Petroff's method has been applied by him to feces, in which the problem is made more difficult by the presence of many spore-formers which resist sodium hydroxid. Feces is collected and diluted with three volumes of water, and then filtered through several thicknesses of gauze. The filtrate is satu- rated with sodium chlorid and left for half an hour. The floating film of bac- 1 Petroff, Johns Hopkins Hosp. Bull., vol. xxvi, No. 294, August, 1915, p. 276. THE TUBERCLE BACILLUS 485 teria is collected in a wide-mouthed bottle and an equal volume of normal sodium hydroxid is added. This is shaken and left in the incubator for three hours, shaking every half hour. It. is then neutralized to litmus with normal hydrochloric, centrifugalized, and the sediment planted. On Mood serum at 37.5° C, colonies become visible at the end of eight to fourteen days. They appear as small, dry, scaly spots with corrugated surfaces. After three or four weeks, these join, covering the surface as a dry, whitish, wrinkled membrane. Coagulated dog serum is regarded by Theobald Smith ^ as a favorable media for the growth of tubercle bacilli. Slants of agar, to which whole rahbWs blood has been added in quan- tities of from 1 to 2 c.c. to each tube, make an excellent medium. Cultivation methods, were simplified by the discovery by Roux and Nocard that glycerin facilitates cultivation. Upon glycerin-agar (gly- cerin 3 to 6 per cent), at 37.5° C, colonies become visible at the end of from ten days to two weeks. Glycerin houillon (made of beef or veal with pepton 1%, glycerin six per cent, slightly alkaline) is a favorable medium. It should be filled, in shallow layers, into wide-mouthed flasks, since oxygen is essential. Transplants to this medium should be made by carefully floating flakes of the culture upon the surface. In this mediimi the bacilli will spread out upon the surface, at first as a thin, opaque, floating membrane. This rapidly thickens into a white, wrinkled, or granular layer, spreading over the entire surface of the fluid in from four to six weeks. Later, portions of the membrane sink. In old cultures, the membrane becomes yellowish. These cultures emit a peculiar aromatic odor. Glycerin potato forms a favorable culture medium for the bacillus. Hesse ^ has devised a medium containing a proprietary preparation known as ^'Nahrstoff Heyden," upon which tubercle bacilli are said to proliferate more rapidly than other bacteria. His method has yielded excellent results. It is prepared as follows: "Nahrstoflf Heyden"' 10 grams Sodium chlorid 5 " Glycerin. 30 " Agar 10 " Normal sodium solution 5 c.c. Aq. dest 1,000 " 1 Th. Smith, Jour. Exp. Med., iii, 1898. 2 Hesse, Zeit. f. Hyg., xxxi. ^ "Nahrstoff Heyden" is prepared in Germanyf It is a white powder similar to nutrose. 486 PATHOGENIC MICROORGANISMS Biological Considerations. — The tubercle bacillus is dependent upon the access of oxygen. Its optimum temperature is 37.5° C. Tem- peratures below 30° and above 42° C. inhibit its growth. In fluid media, the bacilli are killed by 60° in fifteen to twenty minutes, by 80° in five minutes, by 90° in one to two min- ' utes. They will withstand dry heat at 100° C. for one hour. They are resistant to cold. The comparatively high powers of resistance of the bacil- lus are attributed to the protective qualities of the waxy cell membrane.^ The life of cultures, kept in favor- able environment, is from two to eight months, varying to some extent with the nature of the culture medium. The viability of the bacilli in sputum is of great hygienic importance. In most sputum they may remain alive and virulent for as long as six weeks, in dried sputum for more than two months.^ Five per cent carbolic acid kills the bacilli in a few minutes.^ Used for sputum disinfection, where the bacilli are protected, complete disin- fection requires five to six hours. Bi- chloride of mercury is not very efficient for sputum because of the formation of albuminate of mercury. For room disinfection, formaldehyde gas is efficient. Direct sunlight kills in a few hours. Pathogenicity. — The tubercle bacillus gives rise in man and suscep- tible animals to specific inflammation which is so characteristic that a diagnosis of tuberculosis may be made by histological examination, even without the finding of tubercle bacilli. The foci known as tubercles have been studied by Baumgarten ^ and many others ^ Th. Smith, Jour. Exper. Med., 1899; Grancher et Ledoux-Lebard, Arch, de med. exper;, 1892; Galtier, Compt. rend, de I'acad. des sci., 1887. 2 Schell und Fischer, Mitt. a. d. kais. Gesundheitsamt, 1884. 3 De Toma, Ann. di med., 1886. . * Baumgarten, Berl. klin. Woch., 1901. Fig. 104. — Culture of Bacillus Tuberculosis in Flask of Glyc- erin Bouillon. THE TUBERCLE BACILLUS 48? and descriptions may be found in any text-book of pathological anatomy. In man, tuberculosis is the most common of diseases. Naegeli's sta- tistics, based on a large series of autopsies, show not. only the frequency of the disease, but its relation to age. Before one year of age, he finds it very rare. From the first to the fifth year it is rare, but usually fatal. From the fifth to the fourteenth year, one-third of his cases showed tuberculosis; from the fourteenth to the eighteenth year, one-half of the cases. Between eighteen and thirty, almost all the cases examined showed some trace of tuberculous infection. Three-quarters of these were active, one-quarter healed. Two-fifths of all deaths occurring at these ages were due to tuberculosis. After the age of thirty, active lesions gradually diminished, healed lesions increased. In 1900 Pry or ^ stated that the average yearly mortality from tuberculosis in New York amounted to 6,000, and that in Manhattan alone there were constantly 20,000 tuberculous persons. Cornet ^ esti- mates that in 1894 the deaths in Germany from all other infectious diseases amounted to 116,705, those from tuberculosis alone to 123,904. Similar statistics might be chosen from the health reports of any large city. While the disease is less common in rural districts than in large towns, the difference is not so striking as is generally supposed. In man, pulmonary infection is the commonest type. Besides this, tuberculous processes may be found in the skin, the bones, the joints, the organs of special sense, and the abdominal viscera and peritoneum. No part of the human body is exempt from the danger of infection. Infection in man may take place by inhalation, through the skin or the digestive apparatus. V. Behring ^ has expressed the belief that a large percentage of all cases of tuberculosis originate in childhood from infection through the intestinal tract. He determined that tubercle bacilli may penetrate the intestinal mucosa without causing lesions. Behring' s contention has caused a great deal of discussion, and the question he has raised is intimately bound up with the problem of the virulence of bovine tubercle bacilli for human beings, as he as- sumes that the infection is due to the use of infected milk. The problem is plainly of the greatest importance, and for this reason has been diligently investigated during the last few years. The only reliable method of approaching it has been to isolate the tubercle 1 Pry(yr, Med. News, Ixxvii, 1900. « C(ymet, "Die Tuberculose," Wien, 1899, p. 1. » V, Behring, Deut. med. Woch., 39, 1903. 488 PATHOGENIC MICROORGANISMS bacilli from diseased human beings and determine for each case whether the guilty organism belonged to the human or bovine type. These types can be differentiated definitely by cultural characteristics and pathogenicity, and it is not likely that the type changes during the sojourn in the human body. Granted this permanence of type, it is naturally of much value in revealing the source of an infection, to de" Combined Tabulation, Cases Reported and Own Series of Cases (From Park and Krumwiede, loc. cit.) Diagnosis. Adults 16 Years and Over, Human Bovine Children 5 to 16 Years. Human Bovine Children Under 5 Years. Human Bovine Pulmonary tuberculosis Tuberculous adenitis, axillary or inguinal Tuberculous adenitis, cervical Abdominal tuberculosis Generalized tuberculosis alimentary origin . . . Generalized tuberculosis Generalized tuberculosis, including meninges, alimentary origin Generalized tuberculosis, including meninges . Tuberculous meningitis. Tuberculosis of bones and joints Genito-urinary tuberculosis Tuberculosis of skin Miscellaneous Cases: Tuberculosis of tonsils Tuberculosis of mouth and cervical nodes Tuberculous sinus or abscesses Sepsis, latent bacilli Totals 568 2 22 15 6 28 18 11 1 2 677 20 7 3 1 11 4 33 7 2 4 1 7 2 26 1 1 99 33 12 2 15 6 13 28 3 45 14 21 1 161 20 13 10 5 8 1 2 59 Mixed or double infections, 4 termine whether or not a human being is harboring a bacillus of the human type or one of the bovine type. One of the most valuable contributions made to this problem during the last three years is that of Park and Krumwiede.^ The above tabulation is taken from their paper and represents a summary of their own cases and those reported by others. * Park and Krumwiede, Jour, of Med. Res., Oct., 1910. THE TUBERCLE BACILLUS 489 From this table it is evident that out of a total of 1,042 cases, 101 only were bovine in origin and over 50 per cent of these occurred in children under five years of age. Fifty-one out of the 59 cases occurring in the 161 infants were directly or indirectly traced to alimentary in- fection. It seems reasonably accurate, therefore, to state the case as follows: Human adults are relatively insusceptible to bovine infection. Such infection can take place, but is unusual. Below 16 years of age the human race is relatively more susceptible and up to this age the danger of milk infection is unquestionably great, this source accounting for about one-third of the cases. Below 5 years the danger is greatest. This table alone should form sufficient evidence to silence absolutely any doubts as to the dangers of milk infection and remove any objections to the most rigid sanitary control of milk supplies. On the other hand, it also shows that Behring's original claims were far too sweeping and can not be upheld. Rosenberger ^ has recently reported finding tubercle bacilli in the circulating blood of all cases of human tuberculosis which he examined. This announcement aroused much interest and has led to many investi- gations by other workers. Rosenberger's results were obtained by mor- phological examination of smears of citrated blood taken from the patients, dried upon slides and laked with distilled water. Many other observers have failed to confirm Rosenberger's results. Anderson ^ ex- amined 47 cases in which tubercle bacilli were found in the sputum and one case of joint tuberculosis. In none of these 48 cases was he able to obtain tubercle bacilli, either by morphological examination nor by guinea-pig inoculation. Brem ^ subsequently found that laboratory distilled water may frequently contain acid-fast saprophytes — a fact which may account in many cases for errors when morphological examination alone is relied upon and blood examined by the technique of Rosenberger. This, too, is suggested by the finding of acid-fast bacilli in the blood of perfectly healthy individuals. Therefore, although the bacilli may be present in the blood in a certain number of cases it does not seem likely that they are so distributed in anything like the high percentages found by Rosenberger.'* Bacillus tuberculosis (typus humanus) is pathogenic for guinea- 1 Rosenberger, Am. Jour, of Med. Sc, cxxxvii, 1909. 2 Anderson, U. S. P. H. Service, Hygienic Lab., Bull. 57, 1909. 3 Brem, Jour. A. M. A., liii, 1909. ^Suzuki and Takaki, CentralbL f. Bakt.^ Ixij 1911. 490 PATHOGENIC MICROORGANISMS pigs, less markedly for rabbits, and still less so for dogs. It is slightly pathogenic for cattle, a question spoken of more extensively below. Chemical Analysis of Tubercle Bacilli.^ — Diligent efforts by many investigators to isolate the specific toxins which lend tubercle bacilli their pathogenic properties have led to careful chemical analysis of the organisms. About 85.9 per cent of the bacillus consists of water; 20 to 26 per cent of the residue can be extracted with ether and alcohol. This material consists of fatty acids and waxy substances (fatty acids in combination with the higher alcohols). The residue after alcohol- ether extraction is composed chiefly of proteids. These can be extracted with dilute alkaline solutions, and consist chiefly of nucleo-albumins. A nuclein present in this fraction shows extremely high toxicity and has,2 therefore, been suspected of being the pathogenic principle of the bacillus. After these extractions the remainder contains ''cellulose," supposed to represent the framework of the cell membrane, and an ash rich in calcium and magnesium. Toxins of the Tubercle Bacillus. — The Tuberculins. — Filtrates of bouillon cultures of Bacillus tuberculosis ^ will occasionally produce slight emaciation when injected into guinea-pigs, and when administered to tuberculous subjects in sufficient quantity will give rise to marked increase of temperature. It is likely, therefore, that the tubercle bacillus actually secretes a soluble toxin.^ The chief toxic principles, however, of Bacillus tuberculosis are probably endotoxins or bacterial proteins, bound during cell life to the body of the bacillus. Dead bacilli will produce sterile abscesses when injected into animals. Prudden and Hodenpyl,^ Straus and Gamaleia,® and others,^ moreover, have shown that the injection of dead and care- fully washed cultures of this bacillus will produce lesions histologi- cally similar to those occurring after infection with the living germs, and will often lead to marasmus and other systemic symptoms of poisoning. The hope of actively immunizing with substances obtained from ^ Hammer schlag, Cent. f. kKn. Med., 1891; Weyl, Deut. med. Woch., 1891; De Schweinitz and Dorset, Jour. Amer. Chem. Soc, 1895; Hammerschlag, loc. cit. 2 Behnng, Berl. klin. Woch., 1899. ^Straus and Gamdleia, Arch. med. exp., 1891. * Denys, "Le Bouillon Filtre," Louvain, 1905. ^ Prudden and Hodenpyl, N. Y. Med. Jour., June, 1S91; Prudden, ibid., Deo. 5.. ^ Straus and Gamaleia, loc. cit. 7 Majucd, Cent, f . allg. Path., 1890. THE TUBERCLE BACILLUS 491 dead bacilli led Koch to employ various methods of extraction of cultures for the manufacture of tuberculin. ''Old Tuberculin*''^ (Koch) ("T.A.K.")-— The first tuberculin made by Koch is produced in the following manner: Tubercle baciUi are grown in slightly alkaline 5 per cent glycerin-pepton bouillon for six to eight weeks. At the end of this time, growth ceases and the corrugated peUicle of tubercle bacilli, which during growth has floated on the surface, begins, here and there, to sink to the bottom. The entire culture is then heated on a water-bath at about 80° C, until reduced to one-tenth of its original volume. It is then filtered either through sterile filter paper or through porcelain filters. The resulting filtrate is a rich brown, syrupy fluid, containing the elements of the original cul- ture medium and a 50 per cent glycerin extract of the tubercle bacilli. While the glycerin is of sufficient concentration to preserve it indef- initely, 0.5 per cent phenol may be added as an additional precaution. Dilutions of this fluid are used for diagnostic. and therapeutic purposes. "New Tuberculin'' ^ (Koch) (TA, TO, TR).— Koch believed that the immunity resulting from treatment with the old tuberculin was purely an antitoxic immunity, devoid of all antibacterial action. The use of whole dead tubercle bacilli for immunization purposes, however, was impracticable; because, injected subcutaneously, they were not absorbed, and introduced intravenously they were deposited in the lungs and gave rise to lesions. Koch was led, therefore, to resort to more energetic extraction of the bacilli in the hope of procuring a substance which could be easily absorbed and would at the same time give rise, when injected, to antibodies more definitely bactericidal. By extract- ing tubercle bacilli with decinormal NaOH, for three days, filtering through paper and neutralizing, he obtained his TA (alkahne tubercu-. lin). This preparation seemed to fulfil some of the hopes of its dis- coverer, but had the disadvantage of often producing abscesses at the points of injection. Koch then resorted to mechanical trituration of the bacilli. The method he subsequently followed for tuberculin pro- duction is now extensively used, and is carried out as follows: ^ Virulent cultures of tubercle bacilli are dried in vacuo and thoroughly ground in a mortar. Grinding is continued until stained preparations reveal no intact bacilli. (This is done by machinery in all large manu- factories.) One gram of the dry mass is shaken up in 100 c.c. of sterile distilled water. This mixture is then centrifugalized at high speed. 1 Koch, Cent. f. Bakt., 1890; Deut. med. Woch., 1891. 2 Koch, Deut. med. Woch., 14, 1897. ^ Ruppel, Lancet, March 28, 1908. 32 492 PATHOGENIC MICROORGANISMS i The supernatant fluid, known as TO (Tuberculin-Oberschicht), contains the water-soluble constituents of the bacillus, gives no precipitate on the addition of 50 per cent glycerin, and has the 'same physiological action as the old tuberculin. The residue TR (TubercuUn-Ruckstand), after pouring off TO, is again dried and ground up, and again shaken in water and centrifugalized. This process is repeated several times, and eventually, after three or four repetitions, all the TR goes into emulsion. The total volume of water used for these TR extractions should not exceed 100 c.c. All of the TR emulsions are then mixed to- gether. This gives TR a precipitate with 50 per cent of glycerin, and is supposed by Koch to contain substances important in producing an antibacterial immunity. For purposes of standardization, the amount of solid substance in 5 c.c. of the TR is determined by evaporation in vacuo and drying. To the rest are added a little glycerin and formalde- hyde and enough water to allow each cubic centimeter of the solution to contain 0.002 grams of solid material. Thus the culture and the medium remaining the same, fairly accurate standardization is possible. ''New TubercuUn-Bacillary Emulsion.'' ^ — In 1901, Koch combined *'T0" and "TR" by putting forth a preparation referred to as "Bazillenemulsion." This consists of an emulsion of pulverized bacilli 1 : 100 in distilled water. After several days of sedimentation to re- move the coarser particles, the supernatant fluid is poured off and fifty per cent volume of glycerin is added to it for purposes of pi'eservation. This preparation contains 5 milligrams of solid substance in each cubic centimeter. Bouillon Filtre (Denys).^ — This preparation consists of the filtrate (through Chamberland filters) of 5 per cent glycerin-pepton-bouillon cultures of Bacillus tuberculosis. Phenol 0.25 per cent is added to insure sterility. The filtered bouillon corresponds to the unconcentrated old tuberculin of Koch, but, not having been heated, is supposed by Denys to contain important soluble and possibly thermolabile secretory products of the bacillus. Tuberculoplasmin (Buchner and Hahn).^ — Buchner and Hahn, by crushing tubercle bacilli by subjecting them to a pressure of 400 atmospheres, obtained a cell-juice in the form of an amber fluid, to which they attributed qualities closely analogous to those of TR. 1 Koch, Deut. med. Woch,, 1901. 2 Denys, "Le Bouillon Filtre," Louvain, 1905. ' Buchner und Hahn, Munch, med. Woch., 1897; Hahn, ibid THE TUBERCLE BACILLUS 493 Other tuberculins are those of Beraneck/ highly recommended clinically by Sahli,^ that of Klebs,^ and the tuberculin produced from bovine tubercle bacilli by Spengler.'^ Diagnostic Use of Tuberculin. — Subcutaneous Use. — The preparation usually employed for diagnostic purposes is Koch's "Old Tuberculin'* (Alttuberculin). This preparation is administered by hypodermic injec- tion of small quantities obtained by means of dilutions. The dilutions are best made with a 0.5 per cent aqueous carboUc acid solution. In practice a 1 per cent solution is made by pipetting 0.1 c.c. of tuberculin into 9.9 c.c. of the 0.5 per cent carbolic solution. A cubic centimeter of this then contains 0.01 c.c. of tuberculin. One c.c. of this solution added to 9 c.c. of 0.5 per cent carboUc acid gives a solution in whieh each cubic centimeter contains 0.001 c.c, or 1 milligram of tuberculin. The initial dosage in adults in Koch's ^ early work, and as used by Beck ^ on a large number of patients, was 1 milligram. This, according to present opinions, is too high, and most clinicians to-day prefer 0.1 to 0.2 of a milligram. If after three or four days no reaction has occurred, a second dose of 1 milligram is given. The reaction itself is recognized chiefly by the changes in tem- perature. In a positive reaction the patient's temperature will begin to increase within six to eight hours after injection, rising sharply within a few hours to 0.5 or 1.5° higher than the temperature before in- jection. It then sinks more gradually than it rose, the reaction usually being complete within thirty to thirty-six hours. "With the temperature there may be nausea, a chill, rapid pulse; and general malaise. Locally visible tuberculous processes, such as lupus, lymph nodes, etc., may become more tender or swollen, and if the tuberculosis is pulmonarjr, there may be coughing and increased expectoration. The temperatures of persons subjected to the test should be taken regularly for three or four days before tuberculin is used. Ophthalmo-TuhercuUn Reaction. — "Wolff-Eisner^ and, soon after him, Calmette,^ proposed a method of using tuberculin for diagnostic * Beraneck, Compt. rend, de I'acad. des scL, 1903. 2 Sahli, Corrbl. d. Schw. Aerzte, 1906. 3 Kiehs, Cent. f. Bakt., 1896; Deut. med. Woch., 1907. * Spengler, Deut. med. Woch., xxxi, 1904; xxxi and xxxiv, 1905. 6 Koch, Deut. med. Woch., 1890. « Beck, Deut. med. Woch., 1899. • 7 Wolff-Eisner, Berl. med. Gesell., May 15, 1907. * Calmette, Acad, des sci., June 17, 1907. t 494 PATHOGENIC MICROORGANISMS purposes by instillation into the conjunctival sac. In tuberculous patients this process is followed by a sharp conjunctival congestion lasting from one to several days. The preparation used for this purpose is produced in the following way: **01d Tuberculin" is treated with double the quantity of 95 per cent alcohol, and the precipitate allowed to settle and the alcohol then filtered off through paper. The sediment is washed with 70 per cent alcohol until the filtrate runs clear, then pressed between layers of filter paper to remove excess of moisture, scraped into a dish, dried in vacuo over H2SO4, and broken up in a mortar under a hood. Solutions of the powder are made in sterile normal salt solution, 1 per cent by weight, boiled and filtered. The solutions are used in strengths of 0.5 to 1 per cent, a drop of which is instilled into the con- junctival sac.^ Cutaneous Tuberculin Reaction. — ^Von Pirquet ^ has suggested the cutaneous use of tuberculin for diagnostic purposes. A 25 per cent solution of *'01d Tuberculin" was first used. At present the undi- luted substance is employed. After sterilization of the patient's forearm, two drops of this solu- tion are placed upon the skin about 6 cm. apart. Within each of these drops scarification is done, and the skin between them sacrificed as a control. Within twenty-four to forty-eight hours, in tuberculous patients, erythema, small papules, and herpetiform vesicles will ap- pear. According to recent investigations, about 70 per cent of adults show a positive reaction. This reverses its diagnostic value for adults. Moro ^ has modified this by simply making a 50 per cent ointment of tuberculin in lanolin and rubbing it into the skin without scari- fication. Complement Fixation in Tuberculosis.^ — The problem of comple- ment fixation in tuberculosis for diagnostic purposes has been very actively investigated of recent years. The most promising results have been reported with an antigen made by Besredka of a filtrate of an egg-meat-broth, upon which culture the tubercle bacilli had been grown for several weeks ; a similar filtrate of cultures on a watery extract of 1 Method in use at Saranac and kindly communicated by Dr. Baldwin. 2 V. Pirquet, Berl. klin. Woch., xx, 1907; Med. Klinik, xl, 1907. » Moro, Munch, med. Woch., 1906, p. 216. * A review of complement fixation tests in tuberculosis will be found in an article by H. R. Miller, Jour. I ab. & Clin. Med., 1916, i, 816. THE TUBERCLE BACILLUS 495 potato with glycerin, used by Petroff ; and an antigen made by Miller and Zinsser by triturating dead tubercle bacilli with, dry crystals of NaCl and adding distilled water to isotonieity. Craig, Bronfenbren- ner and the above-named writers have reported good results with these various antigens, and, although too early to say which will prove most useful, it is clear that complement fixation methods can aid in the diagnosis of active tuberculosis. We can, of course, judge con- cisely only of the metliod used in our laboratory, where Miller has followed carefully a considerable number of cases on which the method has been used. It would appear at present that between 80 and 90 per cent of the fixation results correspond accurately with clinical findings. The Tuberculin Test as Applied to Cattle. — In cattle, the symptoms of tuberculosis are not easily detected by methods of physical diag- nosis until the disease has reached an advanced stage. In conse- quence, cows may be elements of danger without appearing in any way diseased to those who handle them. In consequence, routine examination of herds by the tuberculin test has become one of the necessary measures in 'public sanitation. According to Mohler,^ an accurate diagnosis may be established in at least 97 per cent of the cases. It is natural that a good deal of objection to the test is encoun- tered on the part of dairy farmers and cattle raisers, and recently it has been publicly claimed that the cattle are injured by the test. There is, however, no scientific basis for this belief, if the test is carried out care- fully and intelligently. As a matter of fact, the systematic use of the test would eventually be distinctly advantageous to the owners of the cattle themselves, since it has been shown that cows, even in the early stages of the disease, may expel tubercle bacilli, either during respira-' tion or in the feces, and thus become a menace to healthy members of the herd. The tuberculin test on cattle should be made as follows: (The directions given below are taken directly from the circular sent out from the Bureau of Animal Industr>^ at Washington.) 1. Begin to take the rectal temperature at 6 a.m., and take it every two hours thereafter until midnight. 2. Make the injection at midnight. 3. Begin to take the temperature next morning at 6 a.m., and con- tinue as on preceding day. To those who have large herds to examine, or are unable to give the 1 MoAZer,. Pub. H. and.Mar. Hosp. Serv. Bull. 41,. 1908.. 496 PATHOGENIC MICROORGANISMS time required by the above directions, the following shortened course is recommended : 1. Begin to take the temperature at 8 a.m., and continue every 2 hours until 10 p.m. (omitting at 8 p.m., if more convenient); or take the temperature at 8 a.m., 12 m., and 10 P.M. 2. Make the injection at 10 p.m. 3. Take the temperature next morning at 6 or 8 a.m., and every 2 hours thereafter until 6 or 8 p.m. Each adult animal should receive 2 c.c. of the tuberculin as it is sent from the laboratory. (The tuberculin sent out from the central labora- tory at Washington is already diluted; 2 c.c. represents 0.25 c.c. of the concentrated ''Old Tuberculin'' of Koch.) Yearlings and two-year-olds, according to size, should receive from 1 to 1.5 cubic centimeters. Bulls and very large animals may receive three cubic centimeters. The injec- tion should be made beneath the skin of the neck or shoulders behind the scapula, after washing the area with a weak carbolic acid solution. There is usually no marked local swelling at the seat of the injection. There are now and then uneasiness, trembling, and the more fre- quent passage of softened dung. There may also be sHght acceleration of the pulse and of the breathing. The febrile reaction in tuberculous cattle following the subcutaneous injection of tuberculin begins from six to ten hours after the injection, reaches the maximum nine to fifteen hours after the injection, and returns to normal eighteen to twenty-six hours after the injection. A rise of two or more degrees Fahrenheit above the maximum tem- perature observed on the previous day should be regarded as an indica- tion of tuberculosis. For any rise less than this a repetition of the injection after four or six weeks is highly desirable. It is hardly necessary to suggest that for the convenience of the one making the test the animals should not be turned out, but fed and watered in the stable. It is desirable to make note of the time of feed- ing and watering and of any temperature fall after watering. The tuberculin should not be used later than six weeks after the date on the bottle, nor if there is a decided clouding of the solution. Therapeutic Uses of Tuberculin. — Tuberculin was first used therapeu- tically, shortly after its discovery, by Koch.^ Hailed with the most optimistic enthusiasm, its possibilities were overestimated and hope- less cases were treated unskilfully, with unsuitable dosage. The conse- quence was that harm was done, the method was attacked by Virchow 1 Koch, Deut. med. Woch., iii, imh 'tttti ITTBERCLE BACILLUS 49^ and others and the new therapy fell into almost complete neglect. At present, the use of tuberculin has again been revived, but with greater caution and with a thorough understanding of its limitations. The tendency has been toward smaller dosage and the limitation of the agent to early cases. No two institutions use tuberculin in exactly the same manner, and it is, therefore, impossible to do more than outline the general scheme of treatment. It must never be forgotten, however, that all forms of tuberculin treatment consist in an "active immuniza- tion" in which, for the time being, the toxemia of the patient is increased rather than neutralized. It is obvious, therefore, that only such cases are at all suitable for treatment in which the process is not a very acute one. The general principle of modern tuberculin therapy seems to lie in choosing doses so small that no marked general reaction shall follow. The preparations most frequently employed are Koch's "Alttuber- cuHn," his "TR," his "Neu Tuberkulin-Bazillen Emulsion," and the Bouillon filtre of Denys. Initial doses of Alttuberculin range from 0.1 to 0.01 of a milligram. In case of successful avoidance of a reaction, the injection may be repeated, gradually increasing, about twice a week. The occurrence of a reaction should be the signal for a longer interval and a slower advance in the size of the dose. The initial dose of "TR" is, as advised by Koch,i about 0.002 mgm. This usually causes no reaction. The dose is doubled, at reason- able intervals, up to 1 mgm. After this, further increase is care- fully gauged by the clinical indications. The maximum dose is about 20 mgm. "Neu Tuberkulin-Bazillen Emulsion," ^ is begun with a dose of 0.001 mgm. Gradual increase as with the other preparations is then prac- ticed. The maximum dose is about 10 mgm. Bouillon filtre has been used chiefly by Denys ^ and with apparently excellent results. Denys is very Emphatic in advising the absolute avoidance of any reaction. He begins with a millionth or even the tenth of a millionth of a cubic centimeter of the bouillon and in- creases with extreme caution. His dilutions are made with glycerin broth. Passive Immunization in Tuberculosis. — ^Numerous attempts have been made to immunize tuberculous subjects with the sera of actively 1 Koch, Deut. med. Woch., xiv, 1897. ^Bandelier und Roepke, ''Lehrb. d. spezifisch. Tub. Ther.," Wurzburg, 1908; Koch, Deut. med. Woch., 1901. ^ Denys, "Le Bouillon filtre," Louvain, 190.5. 498 PATHOGENIC MICROORGANISMS immune animals. The most widely used method of producing such serum is that of Maragliano. Maragliano^s Serum} — Maragliano believes that a toxalbumin is present in tubercle-bacillus cultures which is destroyed by the heating employed in the usual tuberculin production. He procures this sub- stance by filtration of unheated cultures and precipitation with alcohol (tossina praecipitata) . He furthermore makes an aqueous extract of the bacillary bodies. With these two substances he immunizes horses. He draws blood from these after four to six months of treatment. The serum is extensively used in Italy. Its value is, at present, very doubtful. Marmorek^s Serum.^ — Marmorek claims that the poisons produced by Bacillus tuberculosis depend largely upon the medium on which it is grown. He advanced the view in 1903 that the substances obtained in tuberculin were not the true toxins of the tubercle bacillus, that there was a marked difference between these and the poisons elaborated by a younger (primitive) phase of the bacillus as it occurs only within the animal body or on media composed of animal tissue. He consequently grows his cultures on a medium composed of a leucotoxic serum (pro- duced by inoculating calves with guinea-pig leucocytes) and liver tissue. Such cultures, he claims, contain no tuberculin. To the sera produced by immunization with these cultures he attributes high curative powers. Bacilli Closely Related to the Tubercle Bacillus. — The Bacillus of Bovine Tuberculosis. — Tuberculosis of cattle (Perlsucht) was studied by Koch ^ in connection with his early work on human tuberculosis. Koch did not fail to recognize differences between, the reactions to in- fection in the bovine type of the disease and that of man. He attrib- uted these, however, to the nature of the infected subject rather than to any differences in the infecting agents. This point of view met with little authoritative contradictibn, until Theobald Smith, "^ in 1898, made a systematic comparative study of bacilli isolated from man and from cattle and pointed out differences between the two types. The opinion of Smith was fully accepted by Koch ^ in 1901. Since that time, the question, because of its great importance to prophylaxis, has been the subject of many investigations, most of them ^Maragliano, Berl. klin, Woch., 1899; Soc. de biol, 1897. 2 Marmorek, Berl. klin. Woch., 1903, p. 1108; Med. Klinik, 1906. ^ Koch, Arb. a. d. kais. Gesundheitsamt, 11, 1882, * Th. Smith, Jour. Exp. Med., Ill, 1898. 6 i^oc/i, Deut. med. Woch., 1901. '' THE TUBERCLE BACILLUS 499 confirming Smith's original work. Morphologically, Smith ^ found that the bovine bacilli were usually shorter than those of the human type and grew less luxuriantly than these upon artificial media. He determined, furthermore, that, grown upon slightly acid glycerin bouillon, the bovine bacillus gradually reduces the acidity of the culture medium until the reaction reaches neutrality or even shght alkalinity. Fluctuations, after this, do not exceed 0.1 or 0.2 per cent on either side of neutrality. In the case of the human bacillus, on the other hand, there is but slight reduction of the acidity during the first weeks of growth; after this acidity increases and, though subject to fluctuations, never reaches neutrality. This behavior is probably due to action exerted upon the glycerin, since on ordinary bouillon no such differences between the two varieties can be noticed. These observations of Smith were confirmed by Ravenel,^ Vagedes,^ and others. The cultural differences between the two types have been studied with especial care by Wolbach and Ernst,^ and Kossel, Weber, and Heuss.^ All of these observers bear out Smith's contention that luxuriance and speed of growth are much more marked in the human than in the bovine variety. Marked differences, furthermore, have been shown to exist in the pathogenic qualities of these bacilli toward various animal species. Guinea-pigs inoculated with the bovine type ® die more quickly and show more extensive lesions than those infected with himian bacilli. The difference in the pathogenicity of the two organisms for rabbits is sufficiently striking to be of diagnostic value. The bovine bacilli usually kill a rabbit within two to five weeks; the himaan bacilli produce a mild and slow disease, lasting often /for six months, and occasionally fail to kill the rabbits at all. The practical importance of distinguishing between the two types, of course, attaches to the question as to whether the bovine and the human disease are mutually intercommunicable. Extensive attempts to infect cattle with bacilli of the human type have been made,^ for the most part with very httle or no success. Infections of human beings with 1 Th. Smith, Jour. Exp. Med., 1905. 2 Ravenel, Lancet, 1901; Univ. Penn. Med. Bull., 1902. 3 Vagedes, Zeit. f. Hyg., 1898. ^Wolbach and Ernst, "Studies from the Rockefeller Inst.," 11, 1904. ^ Kossel, Weber, und Hems, Arb. a. d. kais. Gesundheitsamt, 1904 and 1905. ^ Smith, loc. cit., and Medical News, 1902. 7 Beck, "Festsch. R. Koch," 1902; SmUh, loc. cit. 500 PATHOGENIC MICROORGANISMS bovine bacilli, however, have been reported and proved beyond reason- able doubt, by Smith,i Ravenel,^ Kossel, Weber, and Heuss,^ Park and Krumwiede,^ and others. Most of these infections have been in children. It is likely, therefore, that while cattle are to a considerable degree im- mune against the bacillus of the human type, human beings do not enjoy the same safeguard in respect to the bovine bacillus. During adult life, the danger of such infection, however, is far less than it is during infancy and early youth. This question has been discussed on p. 487. The Bacillus of Avian Tuberculosis. — A disease resembling in many features the tuberculosis of man is not uncommon among chickens, pigeons, and some other bird species. Koch was the first to discover in the lesions of diseased fowl bacilli much resembling Bacillus tuberculosis. It was soon shown, however, by the studies of Nocard and Roux,^ Mafucci,^ and others, that the'bacillus of the avian disease represented a definitely differentiable species. Morphologically, and in staining characteristics, the bacillus is almost identical with that of the human disease. In culture, however, growth is more rapid and takes place at a temperature of 41° to 45° C.^ (the normal temperature of birds), while the human type is unable to thrive at a temperature above 40°. , Guinea-pigs, very susceptible to human tuberculosis, are very refractory to infection with the avian type; while, on the other hand, rabbits which are resistant to the human type, succumb rapidly to in- fection with avian tuberculosis.^ Prolonged cultivation and passage through the mammahan body is said to cause these bacilli to approach more or less closely to the mammalian type. Conversely, Nocard ^ succeeded in rendering mammalian tubercle bacilli pathogenic for fowl by keeping them in the peritoneal cavities of hens in celloidin sacs for six months. Recently Koch and Rabinovitsch ^^ have isolated from the spleen of ^ Smith, Trans. Assn. Amer. Phys., 1903. 2 Ravenel, Univ. Penn. Med. Bull., 1902. ^ Kossel, Weber, und Heitss, loc. cit. * Park and Krumwiede, Jour. Med. Res., 1910. ^ Nocard et Roux, Ann. de I'inst. Pasteur, 1887. ^ Mafucci, Zeit. f . Hyg., xi. ' Mafucci, loc. cit. 8 Straus et Gamaleia, Arch, de m^d. exp^r., 1891; Courmont et Dor, Arch, de med. exp., 1891. 8 Nocard, Ann. de I'inst. Pasteur, 1898. ^^Koch und Rabinovitsch, Virch. Arch., Beiheft to Bd. 190, 1907. THE TUBERCLE BACILLUS 501 a young man dead of tuberculosis, a microorganism which, culturally, morphologically, and in its pathogenic action upon birds, seemed to belong to the avian type. Lowenstein^ describes a similar organism cultivated from a human case which seems to be a transitional type. Observations of this order are, however, too few at the present time to be used as the basis of a definite opinion as to the relationship between the two varieties. Tuberculosis in Cold-blooded Animals. — The bacillus isolated by Dubarre and Terre ^ resembles Bacillus tuberculosis in morphology and in a certain degree of acid-fastness. It grows at low temperatures, 15° to 30° C. It is non-pathogenic for animals, but kills frogs within a month. Except for the acid-fastness it has little in common with Bacillus tuberculosis. Similar acid-fast bacilli have been isolated from other cold-blooded animals (carp, frogs, turtles, snakes) by many observers. There have been many attempts to show a close relationship between the tubercle bacilli of cold-blooded and those of warm-blooded animals. Moeller, Hansemann, Friedmann, Weber, Ktister, and others have given this subject particular attention and it has gained especial interest because of the recent notorious claims of Friedmann that he has suc- ceeded in obtaining, from turtles, a strain of acid-fast bacilli which can be successfully used in actively immunizing human beings. In 1903 Friedmann ^ described two cases of spontaneous infection of a salt-water turtle (Chelone corticata) with acid-fast bacilli, presenting lesions in the lungs which simulated pulmonary tuberculosis in the human being (cavity formation and miliary nodules) . The organisms cultivated from these lesions presented much similarity to those of the human type and, according to Friedmann,^ unlike other acid-fast bacilli of cold-blooded* animals, could be grown at 37.5° C. As a possible human origin for the turtle infections Friedmann mentions that the attendant who fed these turtles suffered from a double pulmonary tuberculosis. Upon inoculation into guinea-pigs localized lesions only were pro- duced, and dogs, rats, and birds were inmiune. The implication of Friedmann's work is that his culture represents a human strain attenu- ^ Lowenstein, quoted from Koch and Rabinovitsch, loc. cit. 2 Dubarre et Terre, Compt. rend, de la soc. de biol., 1897. ^Fnedmann, D. Med. Woch., No. 2, Jan., 1903, 25. 4 FHedmann, D. Med. Woch., No. 26, 464, 1903, and Centralbl. f . Bakt., I, xxxiv, 1903, also Zeitschr. f. Tuberkulose, iv, Heft 5, 1903. 502 PATHOGENIC MICROORGANISMS ated for man by passage through the turtle, although, as far as we are aware, no definite statement as to this has been made. Summarizing the work of many investigators (Weber, Taute, Ktister, Allegri, Eertarelli, and others) Kiister ^ makes a statement which is, in essence, as follows: In the carp, in snakes, turtles, and frogs spontaneous tuberculosis may occur. The organisms which cause these diseases are specific for cold-blooded animals, similar in many respects to the tubercle bacillus of warm-blooded animals, but in the latter do not produce progressive disease. Human, bovine, and avian tubercle bacilli inoculated into cold-blooded animals can produce lesions which histologically simulate tuberculosis. These micro- organisms can remain a year in cold-blooded animals without losing their pathogenicity for guinea-pigs. Mutation of the tubercle bacillus of warm-blooded animals into cold-blooded ones has not been proven. For these reasons it is quite impossible to exclude, in the apparently positive work of Friedmann and others, the isolation of a true " cold- blooded" type organism, rather than a mutation form originally of the warm-blooded type. What Friedmann's present claims in this respect are for his culture has not been stated as far as we know. The possibility of a positive immunizing value of organisms isolated from cold-blooded animals in human beings, though remote, is not out of ques- tion. The problem is so serious and important, and the experience of many workers is, so far, so inconclusive that the time has not come for commercial exploitation and the cruel arousing of false hopes. The subject, however, deserves carefully controlled further investigations. Bacillus of Timothy. — Moeller isolated from timothy-grass and from the dust in haylofts acid-fast bacilli, like Bacillus tuberculosis. Thfey grow rapidly on agar, soon showing a deep red or dark yellow color. ' Bacillus butyricus {Butter Bacillus). — Slightly acid-fast bacilli 'i'e- sembling Bacillus tuberculosis have been isolated from milk and butter by Petri,2 Rabinovitsch,^ Korn,^ and others. These bacilli are easily differentiated from Bacillus tuberculosis cul- turally. They are slightly pathogenic for guinea-pigs, but riot for man. Bacillus smegmatis and the bacillus of leprosy will be discussed in separate sections. The differentiation of these organisms by staining reactions has been discussed in the section on staining methods. ^ Kolle und Wassermann's Handbuch, 2d edition, v, 767. 2 Petri, Arb. a. d. kais. Gesundheitsamt, 1897. 3 Rabinovitsch, Zeit. f . Hyg., 1897. * Korn, Cent, f . Bakt., 1899. CHAPTER XXXIV THE SMEGMA BACILLUS AND THE BACILLUS OF LEPROSY BACILLUS SMEGMATIS In 1884, Lustgarten ^ announced that he had succeeded in demon- strating, in a number of syphilitic lesions, a characteristic bacillus, which he declared to be the etiological factor in the disease. The great importance of the subject of Lustgarten's communication caused nu- merous investigators to take up the study of the microorganisms found upQn the genitals of normal and diseased individuals. As a result of these researches the presence of the Lustgarten bacilh upon the genitals of many syphilitics was confirmed; but at the same time bacilli, which in all essential particulars were identical with them, were found in the secretions about the genital organs and anus of many normal persons. The first to throw doubt upon the etiological significance of Lustgarten's bacillus, and to describe in detail the microorganism now recognized as Bacillus smegmatis, were Alvarez and Tavel.^ Similar studies were made soon afterward by Klemperer,^ Bitter,^ and others. The smegma bacilli are now known to occur as harmless sapro- phytes in the preputial secretions of the male, about the external genital organs of the female, and within the folds of thighs and buttocks. They are usually found, in these situations, in clumps upon the .mucous mem- brane, and occasionally in the superficial layers of the epithelium, intra- and extra-cellularly. Morphology. — The smegma bacilli are very similar to tubercle bacilli, but show greater variations in size and appearance than do the latter. In length the individuals may vary from two to seven micra. They are usually straight or slightly curved, but according to Alvarez and Tavel may show great polymorphism, including short comma-like forms, and occasional S-shaped spiral forms. 1 Lustgarten, Wien. med. Woch., 47, 1884. 2 Alvarez et Tavel, Arch. d. physiol. norm, et path., Oct., 1885. 3 Klemperer, Deut. med. Woch., xi, 1885. * Bitter J Virchow's Arch., ciii. 503 504 PATHOGENIC MICROORGANISMS They are not easily stained, and though less resistant in this respect than the tubercle bacillus, they yet belong distinctly to the group of acid-fast bacilli. Once stained by the stronger dyes, such as carbol- fuchsin or anilin-water-gentian-violet, they are tenacious of the dye, though less so than tubercle bacilli. The identification of the smegma bacillus by staining methods has become of great practical importance since Fraenkel,^ Miiller,^ and others have demonstrated the occasional presence of acid-fast bacilli, probably of the smegma group, in sputum, and in secretions from the tonsillar crypts and throat. The methods of differentiation which have been found most practical are those which depend upon differences in the retention of stain shown by these bacilli. While it may be stated as a general rule that the smegma bacilli are more easily decolor- ized than tubercle bacilli, it is nevertheless important that a con- trol, as suggested by Wood, be made with known tubercle bacilli whenever a slide of suspected smegma bacilli is examined. For the actual differentiation an excellent method is that of Pappenheim, described in detail in the section on Staining, page 106. This method depends upon the fact that prolonged treatment with alcohol and rosolic acid decolorizes the smegma bacilli but not the tubercle bacilli. Coles ^ has stated that smegma bacilli will resist Pappenheim's decolorizing agent for four hours at the most, while tubercle bacilli will retain the stain, in spite of such treatment, for as long as twenty- four hours. Although minor differences between the smegma bacillus and that of Lustgarten have been upheld by Doutrelepont * and others, never- theless, the etiological significance of Lustgarten's bacillus in syphilis has been finally discredited, and, if not identical with the smegma bacillus, it at least belongs to the same group. The smegma bacilli have no pathogenic significance. They are found upon human beings as harmless saprophytes, and all attempts to infect animals have so far been unsuccessful. They are cultivated with great difficulty, first cultivations from man being successful only upon the richer media containing human serum or hydrocele fluid. After prolonged cultivation upon artificial media they may be kept alive upon glucose agar or ascitic agar. Their growth is slow; 1 Fraenkel, Berl. klin. Woch., 1898. ^Mailer, Deut. med. Woch., 1898. » Coles, Jour, of State Med., 1904. » Doutrelepont, quoted from Klemperer, loc. cit. BACILLUS LEPR^ AND LEPROSY 505 and the colonies, appearing within five or six days after inoculation, are yellowish white, corrugated, and not unlike tubercle-bacillus colonies. BACILLUS LEPRAE AND LEPROSY The bacillus of leprosy was first seen and correctly interpreted as the etiological factor in the disease in 1879, by G. Armauer Hansen/ a Norwegian observer. Hansen found the bacilli in the tissues of the nodular lesions of patients, lying in small clumps, intra- and extra- cellularly, as well as in the serum oozing from the tissue during its removal. Hansen's observation was the fruit of over six years of careful study and as to his priority in making this great discovery, there can be no doubt. Almost simultaneously with his publication, however, Neisser ^ published similar results, obtained by him during a brief stay at Bergen, during the preceding summer. The bacilli described by these workers are now recognized as being unquestionably the cause of the various forms of the disease known as leprosy. Morphology and Staining. — The leprosy bacillus is a small rod measuring about 5 to 7 [x in length and has a close morphological re- semblance to Bacillus tuberculosis, except in that it is less apt to display the beaded appearance and is slightly less slender than the latter. It is non-motile, possesses no flagella, and forms no spores. Like tubercle bacilli, furthermore, the leprosy bacilli belong to the class of so-called acid-fast bacteria, being stained with much difficulty; but when once stained they are tenacious of the color, offering con- siderable resistance to the decolorizing action of acids. It is necessary for differential diagnosis, however, to note that both the difficulty of staining and the resistance to decolorization are less marked in the case of this microorganism than in the case of Bacillus tuberculosis. It was this peculiar behavior to stains that caused the delay of several years in Hansen's publications, since he failed in obtaining good morphological specimens until the work of Koch upon bacterial staining had supplied him with proper methods. The bacillus is stained most easily with anilin-water-gentian-violet or with carbol-fuchsin solution. Stained by Gram's method, it is not decolorized and appears a deep blue. Differ- ential staining by the Ziehl-Neelsen method shows the bacillus stained red unless decolorization by means of the acid and alcohol are prolonged 1 Hansen, Virch. Arch., 79, 1879. 2 Neisser, Breslauer arztl. Zeitschr., 20, 1879, 506 PATHOGENIC MICROORGANISMS for an unusual time. A differentiatian from tubercle bacilli by virtue of greater ease of decolorization is of value only in the hands of those having much experience with these bacilli, and follows no regular laws of acid-strengths or time of application which can be generally applied by the inexperienced. In tissues, the bacilli are easily stained by the methods used for staining tubercle bacilli. The sections are* left in the Ziehl carbol-fuchsin solution either from two to twelve hours at incu- bator temperature or for twenty-four hours at room temperature. Subsequent treatment is that employed in the case of tuberculous tissue sections (see p. 112). Cultivation. — Cultivation of the leprosy bacillus has not met with success. Hansen and others who have approached the problem with a thorough knowledge of the microorganism, combined with a com- petent bacteriological training, have failed in all their attempts. Numerous positive results reported by observers have always lacked adequate confirmation. Recently, Rost,^ of the British Army Medical Corps, has claimed success in cultivation of leprosy bacilli upon salt-free bouillon, his point of departure being the previous observation that salt-free media favored the growth of tubercle bacilli. His results have not been confirmed. In 1909 Clegg^ succeeded in growing an acid-fast bacillus from leprous tissue, obtaining his results by inoculating leprous material upon agar plates upon which ameba coli had been grown in symbiosis with other bacteria. On such plates the acid-fast bacilli multiplied, and, subsequently, pure cultures were obtained by heating the cultures to 60° C, which destroyed the ameba coli and other bacteria. These results were confirmed by other, workers and, soon after that, Duval ^ not only succeeded in repeating Clegg's experiments, but obtained cul- tures of an acid-fast bacillus directly from leprous lesions without the aid of ameba. He first observed that the leprosy organism would multi- ply around a transplanted piece of leprous tissue upon ordinary blood agar tubes upon which influenza bacilli and meningococci were grown. He concluded that such growth depended upon chemical changes in the media and believed the formation of amino-acids essential for the initial growth. The method he subsequently described depended upon supplying these substances either by adding tryptophan to nutrient agar or by pouring egg albumen and human blood serum in Petri dishes, 1 Rost, Brit. Med. Jour., 1, 1905. 2 Clegg, Philippine Jour, of Sc, iv, 1909. 3 Duval, Jour. Exp. Med., xii, 1910, and ibid., 15, 1912. BACILLUS LEPR^ AND LEPROSY 507 inspissating, at 70° C, for three hours and, after inoculating with leprous tissue, adding a 1 per cent solution of trypsin. Indirectly the same result was obtained by employing culture media containing albu- minous substances and inoculating with bacteria capable of producing amino-acids from the medium. After leprosy bacilli had been grown on this medium for several generations, they could easily be cultivated on agar slants without special additions or preliminary treatment. In spite of extensive work upon this very important problem opinions are still divided as to the specific nature of the organisms cul- tivated by Clegg and by Duval. Animal experiments with these cultures have remained inconclusive. The cultures after prolonged preservation upon artificial media grow heavily, often lose their acid-fast charac- teristics, develop into streptothrix-hke or diphtheroid forms and become markedly chromogenic, all these characteristics suggesting saprophytism. In a recent communication, Duval and Wellman ^ state their opinion as follows : From 29 cases of leprosy, 22 successive cultivations of acid- fast bacilli were made; in 14 of them a chromogenic organism, similar to that of Clegg, was found. This grows either as a non-acid-f ast strep- tothrix in subsequent cultivations or as non-acid-fast diphtheroid forms. From eight cases an organism distinctly different from the former t^s cultivated which grows only on specific media and by serological tests seems to give reaction which differentiates it from Clegg's organism. Du- val believes that there is no reason to assume specific etiological relation- ship for the first organism mentioned. In the case of the second, he admits that not sufficient proof has been brought, but states his belief that its etiological significance is probable. Pathogenicity. — Innumerable attempts to transmit leprosy to ani- mals by inoculation have been unsuccessful. Nicolle,^ however, has recently claimed 'successful experiments upon monkeys (macacus) in whom inoculation with tissue from infected human beings was followed, in sixty-two days, by the development of a small nodule at the site of inoculation, in which, upon excision, leprosy bacilli were found. In most cases, however, inoculation has given rise merely to a transient inflammatory reaction. Among human beings, leprosy has been a widely spread disease since the beginning of history, and much evidence is found in ancient lit- erature which testifies to a wide distribution of the disease long before the Christian era and throughout the Middle Ages. At the present day, 1 Duval and Wellman, Jour, of Inf. Dis., xi, 1912. 2 Nicolle, Sem. medicale, 10, 1905. 508 PATHOGENIC MICROORGANISMS leprosy is most common in the eastern countries, especially in India and China. In Europe the disease is found in Norway, in Russia, and in Iceland. In other European countries, while the disease occurs, it is not at all common. In the United States, there are, according to Osier, three important centers of leprosy situated in Louisiana, in California, and among the Norwegian settlers in Minnesota. The disease is also present in several provinces of Canada. In all countries in which segregation of lepers is rigidly practiced, the disease is diminishing. In Norway, according to Hansen, proper sanitary measures have reduced the number of lepers from 2,870 in 1856, to 577 in 1900. Clinically, the disease appears in two chief varieties, tubercular leprosy and the so-called anesthetic leprosy. In the former variety, hard nodular swellings appear, usually in the face, but often on other parts of the body as well. These lead to frightful disfigurement and are accompanied by a falling-out of hair and a loss of sensation in the affected areas. In the anesthetic form, there is usually at first pain in definite areas of the extremities and the trunk, which is soon followed by the formation of flat or slightly raised pigmented areas, within which there is absolute anesthesia with, later, atrophy and often secondary necrosis in the atrophied parts. The disease is usually chronic in its course. The bacilli are found in large numbers in the cutaneous lesions. In the knobs of the nodular variety, they lie in clumps between the con- nective-tissue cells and within the large spheroidal cells which make up the nodules. They are found, also, in advanced cases, in the liver and in the spleen, lying within the cells, and, to a slighter extent, in the intercellular spaces. They have also been found within the kidneys, the endothehum of the blood-vessels, and in the testicles.^ In the blood, the bacilli have frequently been demonstrated, especially during the febrile attacks which occur during the disease. Westphal and Uhlen- hut ^ have found the bacilli within the central nervous system, and these observers, as well as others, have found them Ijdng within the substance of the peripheral nerves, thus explaining the anesthesia. A fact of enormous importance to the question of transmission is the observation made by various observers, more especially by Sticker, that the bacilli are found with great regu- larity in considerable numbers in the nasal secretions of persons suffering from the disease. Sticker is inclined to regard the nose 1 Sticker, Miinch. med. Woch., 39, 1897. 2 Westphal uud UhUnhut, Klin. Jahrb., 190X. BACILLUS LEPR.E AND LEPROSY 509 as the primary path of infection. Whether or not this be true can not, at present, be decided. As a source of infection, however, the nasal mucus and, secondarily, the saliva, are certainly the vehicles by which large numbers of the bacilli leave the infected patient, and, therefore, tend to spread the disease. The contagiousness of leprosy is far less than is that of most other bacterial diseases. Physicians and others who come into direct contact with large numbers of leprous patients, observing at the same time the ordinary precautions of cleanliness, rarely contract the disease. On the other hand, intimate contact with lepers without such precautions is the only possible means of transmission. The demonstration of leprosy bacilli in dust, soil, etc., must always be looked upon with sus- picion, since, apart from actual human inoculation, there is no method of positively differentiating the baciUi from similar acid-fast organisms. Instances of transmission by contact are on record, not the least famous of which is the case of Father Damien, who contracted the disease while taking care of the lepers upon the island of Molokai. Hansen states that in his knowledge no case of leprosy can be found in which careful examination of the past history will not reveal direct contact with a previous case. Direct inoculation of the human being with material from a leprous patient has been successfully carried out by Arning,^ upon a Hawaiian criminal. In this case a piece of a leprous nodule was planted into the subcutaneous tissue of the left arm. One month after the inoculation, pain appeared in the arm and shoulder, and four and a half months later a typical leprosy nodule was formed. Four years after the inoculation, the patient was a typical leper. Although our inabihty to cultivate the leprosy bacillus, and the lack of success attending animal inoculation, have made it impossible to study more closely the toxic action of this microorganism, there is, neverthe- less, some evidence which points toward the production of a poisonous substance of some kind by the bacillus. Rost,^ who claims to have cultivated the bacillus, manufactured from his cultures, by the technique for the production of ^'Old Tuberculin," a substance which he (Jailed "leprolin," and which he employed therapeutically in the same manner in which tuberculin is employed in tuberculosis. As stated before, the results of Rost still lack confirmation. Of far greater importance, both in demonstrating the probability of the existence of a definite toxin as well as in indicating the close relationship between the leprosy bacillus 1 Aming, Vers. d. Naturfor. u. Aerzte, 1886. ^ Rost, loc. cit. 510 PATHOGENIC MICROORGANISMS and the Bacillus tuberculosis, are the investigations upon the action of tuberculin upon leprous patients. When tuberculin is adminis- tered to lepers, a febrile reaction occurs usually twenty-four or more hours after the administration. The fever differs from that produced by the use of the same substance in tuberculous patients in that it is of late occurrence and lasts considerably longer. At the same time, there may be marked redness and tenderness of the nodules. In isolated cases, Babes ^ has noticed alarmingly high and prolonged fever together with systemic symptoms such as nausea, headache, and even uncon- sciousness, following the injection of tuberculin. The same writer claims to have extracted from the organs of lepers, which contained enormous numbers of bacilli, substances which showed an action similar to that of the tuberculin. RAT LEPROSY Stefansky ^ first observed this disease among rats in Odessa, and since then it has been observed in Berlin (Rabinovitsch ^), in London (Dean^), in New South Wales (Tidswell ^), and in San Francisco (Wherry ^ and McCoy ^) . The disease occurs spontaneously among house rats and is characterized by subcutaneous induration, swelling of lymph nodes, with, later, falling out of the hair, emaciation, and some- times ulceration. Its course is protracted and rats may live with it for six months or a year. When a rat suffering from this disease is dis- sected there is usually found, under the skin of the abdomen or flank, a thickened area which has the appearance of adipose tissue except that it is more nodular and gray and less shiny than fat. It is so like fat, however, that it is often possible to overlook it as evidence of disease by one unfamiliar with the condition. In this area acid-fast bacilli looking like the Bacillus leprae are found in large numbers. These bacilli are also found in the lymph nodes and sometimes in small nodules which appear in the liver and lung. 1 Babes, in KoUe und Wassermann, "Handbuch," etc., Erst. Erganz. Bd., 1907. ^Stefansky, Centralbl. f. Bakt., xxxiii, 481. ' Rabinovitsch, Centralbl. f. Bakt., xxxiii, 577. ^ Dean, Centralbl. f. Bakt., xxxiv, 222; Jour. Hyg., xcix. ^ Tidswell, cited by Brinkerhoff in "The Rat and Its Relation to Public Health," Treas. Dept., Wash., 1910. 6 Wherry, J. A. M. A., June 6, 1908, p. 1903; Jour. Inf. Dis., dvii, Rep. U. S. P. H., and M. H. S., xxiii, 1841. "" McCoy, Rep. U. S. P. H. and M. H. S., xxiii, 981; Abstr. in J. A. M. A., Aug. 22, 1908, 690. RAT LEPROSY 511 The disease can be transmitted experimentally from rat to rat and probably is transmitted naturally from rat to rat by the agency of fleas (Wherry, McCoy). Although clinically not exactly like human leprosy the condition is sufficiently like it to arouse much hygienic interest. The distribution of the disease in various parts of the world does not correspond with the distribution of leprosy. A peculiar feature of its distribution is the fact that in San Francisco, as the writer was told by McCoy, almost all the rats that suffered from this disease came from the district in which the retail meat business is located, known as "Butchertown." The organisms were made to multiply in vitro by Zinsser and Cary in plasma preparations of growing rat spleen. Chapin has succeeded in cultivating them by a method analogous to the trypsin- egg albmnen method employed by Duval. In the experiments of Zinsser and Cary it was found that although the organisms may retain their acid-fast characteristics for many weeks within leucocytes they degen- erate rapidly within the spleen cells, a fact which seems to have some bearing on the mechanism of resistance possessed by the body against acid-fast organisms. 34 CHAPTER XXXV BACILLUS DIPHTHERIiE, BACILLUS HOFFMANNI, AND BACILLUS XEROSIS BACILLUS DIPHTHERIiE Since 1821, when Bretonneau of Tours published his observa- tions, diphtheria has been an accurately recognized clinical entity. Our knowledge of the disease in the sense of modern bacteriology, however, begins with the first description of Bacillus diphtheriae by Klebs in 1883. Klebs ^ had observed in the pseudomembranes from diphtheritic throats, bacilli which in the light of more recent knowledge we can hardly fail to recognize as the true diphtheria organism. His v/ork, however, was purely morphological and, therefore, inconclusive. One year after this announcement, Loeffler ^ isolated and cultivated an organism which corresponded in its morphological characters to the one described by Klebs. He obtained it from thirteen clinically unques- tioned cases of diphtheria, and, by inoculating it upon the injured mucous surfaces of animals, succeeded in producing lesions which resembled closely the false membranes of the human disease. His failure to find the bacillus in all the cases he examined, his finding it, in one instahce, in a normal throat, and his inability to explain to his own satisfaction some of the systemic manifestations of the infection which we now know to be due to the toxin, caused him to frame his conclusions in a tone of the utmost conservatism. The second and third publications of Loeffler,^ however, and the inquiry into the nature of the toxins produced by the bacillus, published in 1888 by Roux and Yersin/ eliminated all remaining doubt as to the etiological relationship existing between this organism and the disease. Innumerable observations, both clinical and bacteriological, by other workers, have, since that time, confirmed the early investigations, 1 Klebs, Verb. d. 2. Kongr. f. inn. Medizin, Wiesbaden, 1883, 2 Loeffler, Mittheil. a. d. kais. Gesundheitsamt, 1884. ^Loeffler, Cent. f. Bakt., 1887 and 1890. * Roux and Yersin, Ann. de I'inst. Pasteur, 1888 and 1889. 512 BACILLUS DIPIITHERIiE 513 and it is to-day a scientific necessity to find the bacillus of Klebs and Loeffler in the lesion before a diagnosis of '' diphtheria '^ can properly be made. Morphology and Staining. — While Bacillus diphtherise presents certain characteristic appearances which facilitate its recognition, it is, at the same time, subject to a nimiber of morphological variations with i«i i^ «? A"^?^ ?irit>i JJ*-*: *> %>'l\yC ! ^- It V « ^ /^i Fig. 105. — Bacillus diphtheria. all of which it is important to be familiar. These variations are, to a limited extent, dependent upon the age of the culture and upon the constitution of the medium on which it has been grown. These factors, however, do not control the appearance of the organism with any degree of regularity, and any or all of its various forms may occur in one and the same culture. It is likely that these different appear- ances represent stages in the growth and degeneration of the indi- vidual bacilli, but there does not seem to be any just reason for believing that, as several observers have stated, there is definite correla- tion between its microscopic form and its biological characteristics, such as virulence, toxicity, etc. 514 PATHOGENIC MICROORGANISMS The bacilli are slender, straight, or slightly curved rods. In length they vary from 1.2 micra to 6.4 micra, in breadth from 0.3 to 1.1. As seen most frequently when taken from the throat they are about 4 to 5 micra in length. They are rarely of uniform thickness throughout their length, showing club-shaped thickening at one or both ends. Occasionally they may be thickest at the center and taper toward the extremities. When thickened at one end only, a slender wedge-shape results. Such forms are usually straight, of smaller size than their neighbors, and are more often stained with great uniformity. These are spoken of by Beck ^ as the "ground type," and assumed, for in- sufficient reasons, to be the young individuals. Branched forms have been described by some investigators. They are rare and probably to be regarded as abnormal or involution forms due to un- favorable environment. The organisms stain with the aqueous anilin dyes. A characteristic irregularity of staining which is of great aid in diagnosis is best obtained with Loeffler's " alkaline methylene-blue." (For preparation see section on Staining, p. 96.) Stained with this solution for five to ten minutes many of the bacilli appear traversed by unstained transverse bands which give them a striped or beaded appearance. The longer indi- viduals often have a strong resemblance to short chains of strepto- cocci. Others may appear unevenly granular. In cultures which are about eighteen hours old, ma^y of the bacilli may show deeply stained oval bodies situated most frequently at the ends. These are the so-called " polar '^ or "Babes-Ernst" bodies.^ Special stains have been devised for the demonstration of these appearances. One of these was originated by Neisser,^ who claims for it differential value in distinguishing these organisms from pseudodiphtheria and xerosis baciUi. His method requires two solutions: 1. Methylene blue (Griibler) 1 gram. Alcohol, 96 per cent 20 c.c. Glacial acetic acid 50 " Water 950 " 2. Bismarck brown 2 grams. Water 1,000 c.c. 1 Beck, in KoUe imd Wassermann, ii, p. 773. a Babes, Zeit. f. Hyg., Bd. v, 1889. . » Neisser, Zeit. f. Hyg., xxiv, 1897. BACILLUS DIPHTHERIA 515 The cover-slip preparation, after having been fixed, is stained with so- lution No. 1 for one to three seconds. It is then washed in water and immersed for from three to five seconds in solution No. 2. With this stain the bodies of the bacilli appear brown, the polar granules blue. Another method which has been extensively used is that of Roux. The solutions required for this are: 1 . Dahlia violet 1 gram. Alcohol, 90 per cent 10 c.c. Aq. dest ad 100 " 2. Methyl green ' 1 gram. Alcohol, 90 per cent 10 c.c. Aq. dest ad 100 " The two solutions are mixed, one part of 1 being added to three parts of 2. Preparations are stained in this mixture for two minutes. The polar bodies appear a dark violet. Other methods for the staining of polar bodies have been recommended. The significance of the polar bodies is not well understood. Their discoverer, Ernst, regarded them as bodies analogous to the spores of other organisms. The ease with which they are stained, however, and the low temperatures to which the bacteria succumb make this appear very unlikely. A more probable interpretation seems to be that of Escherich ^ who regards them as chromatic granules. Stained by Gram's method, the diphtheria bacilli retain the gentian- violet. In stained smears from the throat or from cultures a characteristic grouping of the bacilli has been observed. They lie usually in small clusters, four or five together, parallel to each other, or at sharp angles. Two organisms may often be seen attached to each other by their cor- responding ends while their bodies diverge to form a ''V" or "Y"' shape. Biological Characteristics. — The diphtheria bacillus is a non-motile, non-flagellated, non-spore-forming aerobe. Its preference for oxygen is marked, but it will grow in anaerobic environment in the presence of suitable carbohydrates. It does not liquefy gelatin. The bacillus grows at temperatures varying between 19° C. and 42° C, the most favorable temperature for its development being 37.5° C. Temperatures above 37.5°, while not entirely stopping its growth, impede the development of its toxin. Resistance. — The thermal death point of this organism is 58° C. for • ten minutes, according to Welch and Abbott. Boiling kills it in about ^Escherich, "Aetologie, etc., d. Diphth.," Wien, 1894. "" 516 PATHOGENIC MICROORGANISMS one minute. Low temperatures, and even freezing, are well borne. Desiccation and exposure to light are not so fatal to this organism as to most of the other pathogenic bacteria. Sternberg ^ has found it alive in dried bits of the pseudomembrane after fourteen weeks. It is easily killed by chemical disinfectants in the strengths customarily employed. H2O2 seems especially efficacious in killing the organisms rapidly. Cultivation. — The diphtheria bacillus grows readily on most of the richer laboratory media. It will grow upon media made of meat ex- tract, but develops more luxuriantly on all those which have a meat infusion as their basis. While it will grow upon both acid and alkaline media, it is sensitive to the extremes of both, the most favorable reaction for its development being probably about 0.5 per cent alkalinity ex- pressed in terms of f NaOH. Animal proteids added to the media, in the form of blood serum, ascitic fluid, or even whole blood, increase greatly the rapidity and richness of its growth. Horse serum is sup- posed by some to be especially favorable. ^ Loeffler's Medium. — ^The most widely used medium for the cultiva- tion of this bacillus is the one devised by Loeffler. This consists of: Beef blood serum 3 parts One per cent glucose meat-infusion bouillon 1 part The mixture is coagulated at 70° C. in slanted tubes and sterilized at low temperatures by the fractional method. Upon this medium the diphtheria bacillus in twelve to twenty-four hours develops minute, grayish-white, glistening colonies. These enlarge rapidly, soon out- stripping the usually accompanying streptococci. The medium seems to possess almost selective powers for the bacillus and, for this reason, it is especially valuable for diagnostic purposes. Meat-Infusion Agar. — Upon slightly alkaline meat-infusion agar the bacillus develops readily, though less so than on Loeffler's serum. Or- ganisms which have been on artificial media for one or more genera- tions may grow with speed and luxuriance upon this medium. When planted directly from the human or animal body upon agar, however, growth may occasionally be slow and extremely delicate. Colonies on agar appear within twenty-four to thirty-six hours as small, rather trans- lucent, grayish specks. The appearance of these colonies is quite char- acteristic and easily recognized by the practiced observer. Surface colo- 1 Sternberg, "Manual Bac," p. 455. 2 Michel, Cent. f. Bakt., 1897. BACILLUS DIPHTHERIiE 517 nies are irregularly round or oval, showing a dark, heaped-up, nucleus- like center, fringed about by a loose, coarsely granular disk. The edges have a peculiarly irregular, torn appearance which distinguishes them readily from the sharply defined streptococcus colonies. For these reasons agar is the medium most commonly used for purposes of iso- lation. The addition of dextrose 1 per cent, nutrose 2 per cent, or glycerin 6 per cent, renders agar more favorable for rapid growth, but unfits it for the preservation of cultures, the organism dying out more rapidly, probably because of acid formation. Meat-Infusion Broth. — Upon beef or veal broth the diphtheria bacil- lus grows rapidly, almost invariably forming a pellicle upon the surface, — another expression of its desire for oxygen. The broth remains clear. Broth tubes with such growth, therefore, have a characteristic appear- ance. Meat-infusion gelatin is a favorable medium for the Klebs-Loeffler bacillus, but growth takes place slowly because of the low temperature at which this medium must be kept. Gelatin is not fluidified. Milk is an excellent medium, and for this reason may even occa- sionally be a vehicle of transmission. There is no coagulation of the milk. Upon potato, B. diphtheriae will grow only after neutralization of the acid. It is, at best, however, a poor nutrient medium. Upon the various pepton solutions the bacillus of diphtheria produces no indol. Many special media have been recommended for the cultivation of this organism. The most important of these are the modification of Loeffler's serum devised by Beck,^ the horse-blood-fibrin cake used by Escherich, and Wassermann's ascitic-fluid-nutrose-agar, called by him "Nasgar." None of these has sufficient advantages over the simpler media, however, to make its substitution desirable. Isolation. — Cultures are taken from throats upon Loeffler's blood serum. These are permitted to grow at 37.5° C. for from eighteen to twenty-four hours. At the end of this time about 5 c.c. of bouillon are poured into the tubes and the growth is gently emulsified in the broth with a platinum loop. Two or three loopfuls of this emulsion are then streaked over the surface of glucose agar, serum agar, or nutrose agar. After twenty-four hours' incubation these plates show characteristic . 1 M, Beck, KoUe imd Wasserjnann; Brit. Med. Jour. 518 PATHOGENIC MICROORGANISMS colonies which can be easily fished and again transferred to Loeffler tubes or any other suitable medium. Diagnosis. — Cultures from suspected throats are taken on Loeffler 's blood serum medium and incubated at 37.5° C. for 12 to 18 hours. At the end of this time morphological examination by staining with Loef- fler's alkaline methylene blue and by some polar body stain like that of Neisser is carried out. Occa- sionally direct smears from the throat may show the bacilli, but it is rarely possible to make a satis- factory diagnosis in this way. Williams has pointed out that in throat cultures in which the diph- theria bacilli are few in number it is of advantage to inoculate a tube of ascitic broth with the mixed culture. The diphtheria ba- cilli will appear in eighteen to twenty-four hours as a pellicle on the surface. A portion of this pellicle may then be plated on ascitic agar and isolated in pure culture from the colonies. Pathogenicity. — Bacillus diphtherias causes a more or less specific local reaction in mucous membranes, which results in the formation of the so-called * * pseudo-membranes. ' ' When these are characteristically present, infection with this bacillus should always be suspected. The consequent disease depends, in part, upon the mechanical disturbance caused by these false membranes and, in part, upon the systemic poi- soning with the toxin which the bacilli produce. Although the diph- theria bacillus has been found after death in the spleen and liver, we have no data which would justify the assumption that a true diph- theria-septicemia may occur during life. It is probable that in those cases which Baginsky ^ has called the septicemic form of diphtheria. Bacillus diphtherias has merely opened a path by which accompanying streptococci have gained access to the lymphatics and the blood stream. The most frequent sites of diphtheritic inflammation are the mucous membranes of the throat, larynx, and nose. They have also been Fig. 106. — Colonies of Bacillus DIPHTHERIA ON GlYCERIN AgAR. ^Baginsky^ "Lehrbuch d, Kinderkrankheiten/ BACILLUS DIPHTHERIA 619 found in the car, upon the mucous membrane of the stomach and the valva, and upon the conjunctiva and the skin. According to Loeffler, Strelitz/ and others, the bacillus may, by extension from the larynx, give rise to a true diphtheritic broncho-pneumonia. For the usual laboratory animals the diphtheria bacillus is very pathogenic. Dogs, cats, fowl, rabbits, and guinea-pigs are susceptible. Rats and mice are resistant. False membranes, analogous to those found in human beings, have been produced in many animals, but only when inoculation had been preceded by mechanical injury of the mucosa. Small quantities (0.5 to 1 c.c.) of a virulent broth culture, given subcutaneously to a guinea-pig, may produce the gravest symp- toms and within six to eight hours the animal may show signs of great discomfort. Death occurs usually within thirty-six to seventy-two hours. Upon autopsy the point of inoculation is soggy with serosan- guineous exudate; neighboring lymph-nodes are edematous. Lungs, liver, spleen, and kidneys are congested. There may be pleuritic and peritoneal exudates. Pathognomonic is a severe congestion of both suprarenal bodies. The gastric ulceration recently described by Rose- nau and Anderson ^ may occur, but are by no means regularly found (two out of fifty in our series ^). Determination of Virulence. — When diphtheria or diphtheria-like bacilli are isolated from the throats of patients not showing typical clinical diphtheria, or from healthy individuals suspected of being carriers, it is important to determine whether these organisms are viru- lent. The usual criterion is their virulence for guinea-pigs. Two c.c. of a forty-eight-hour broth or ascitic broth culture are injected sub- cutaneously into a normal guinea-pig. This dose will kill the pig in three to five days if the culture is virulent. A control injection should always be made into another pig of the same weight, which has previously received an injection of antitoxin (at least 250 units). Recently Neisser has suggested that the intracutaneous injection of the suspected bacilli may be used for the determination of virulence. This has the advantage of economy, as several tests can be carried out on the same pig. The method as applied by Zingher and Soletsky * has been to use the following modification of Neisser 's method: Two guinea-pigs of about 250 gr. are used for the test. The abdominal 1 Strelitz, Arch. f. Kinderheilk., 1891. 2 Rosenau and Anderson, Jour. Inf. Dis., iv, 1907. ' Zinsser, Journ. Med. Res., xvii, 1907. * Zingher and Soletsky, Jour. Inf. Dis., 1916, xvii, 54. 620 PATHOGENIC MICROORGANISMS wall is prepared by shaving or plucking out the hair. A twenty-four- hour pure culture on Loeffler's medium is emulsified in 20 c.c. of normal salt solution and 0.15 c.c. of this suspension is injected intra- cutaneously at a corresponding site into each of the two guinea-pigs. One of these animals is given at the same time an intracardial injec- tion of about 250 units of antitoxin, or is prepared by an intraperi- toneal injection of antitoxin twenty-four hours before the tests are made. Six cultures may be tested in this way on two animals. Viru- lent strains produce a definitely circumscribed local infiltrated lesion, which shows superficial necrosis in two to three days. In the control pig the skin remains normal. This method, in the hands of Zingher, gives results parallel to those obtained with the subcutaneous tests. Diphtheria Toxin.^ — Animals and man infected with B. diphtheriae show evidences of severe systemic disturbances and even organic de- generations, while the microorganism itself can be found in the local lesion only. This fact led even the earliest observers to suspect that, in part at least, the harmful results of such an infection were attrib- utable to a soluble and diffusible poison elaborated by the bacillus. The actual existence of such a poison or toxin was definitely proved by Roux and Yersin ^ in 1889. They demonstrated that broth cultures in which B. diphtherias had been grown for varying periods would remain toxic for guinea-pigs after the organisms themselves had been removed from the culture fluid by filtration through a Chamberland filter. Methods of Production op Diphtheria Toxin. — While toxin can be produced with almost all of the virulent diphtheria bacilli, there is great variation in the speed and degree of production, dependent upon the strain of organisms employed and upon the ingredients and reac- tion of the medium upon which they are grown. Most laboratories possess one or several strains of bacilli which are empirically known to be especially potent in this respect. One of the most extensively used, not only in this country but in Europe as well, is the strain known as ''Culture Americana," or '' Park- Williams Bacillus No. 8,'* an organism isolated by Dr. Anna Williams of the New York Depart- ment of Health in 1894. Throughout more than ten years of cultiva- tion this bacillus has retained its great power of toxin production. Because of the severity of cases of diphtheria in which the diph- theria bacilli were associated with streptococci, many observers were led to believe that the presence of streptococci tended to increase the 1 Loeffler, Cent, f . Bakt., 1887. 2 ^ux and Yersin, loc. cit. BACILLUS DIPHTHERLE 521 toxin-producing power of B. diphtheriae. Experiments by Hilbert, ^ Theobald Smith,^ and others seem to have given support to this view. The medium most frequently employed for the production of toxin is a beef-infusion broth. There are minor differences of opinion as to the most favorable constitution of this medium for the production of toxin. All agree, however, in recognizing the importance of peptone, without which, according to Madsen,^ no satisfactory toxin has yet been produced. This is added in proportions of from one to two per cent. The presence of sugars in the medium is not desirable in that it leads to acid production; L. Martin* removes the sugars from the meat by fermentation with yeast. Smith ^ accomplishes the same purpose with B. coli. According to Park and Williams,^ however, this is super- fluous, the quantity of sugar present in ordinary butcher's meat not being sufficient to exert unfavorable influence. Experience has shown that a primary alkaline reaction offers the most favorable conditions for toxin production. In all cultures of B. diphtherise in non-sugar free broth, there is, at first, a production of acid and, while this continues, there is, as Spronk ^ has shown, little or no evidence of toxin elaboration. Park and Williams,^ in an inquiry into the question of reaction, came to the conclusion that the best results are obtained with a broth to which, after neutralization to litmus, Y NaOH is added in an amount of 7 c.c. to the liter. In such a medium the largest yield of toxin is obtained after about five to eight days' growth at a temperature of 37.5° C. Free access of oxygen to the culture medium during the growth of the organisms has been found to be of great importance. Roux ob- tained this by passing a stream of oxygen through the bouillon. The supply is quite sufficient for practical purposes, however, if the medium is distributed in thin layers in large-necked Erlenmeyer flasks. Chemical Nature and Physical Properties op Diphtheria Toxin. — The chemical composition of diphtheria toxin is not known. Brieger and Frankel,^ by repeated precipitation with alcohol, suc- 1 Hilbert, Zeit. f. Hyg., xxix, 1898. ^ Smith, Medical Rec, May, 1896. ' Madsen, Kraus und Levaditi, "Handbuch d. Technic," etc., 1907. 4 L. Martin, Ann. de Tinst. Pasteur, 1897. 6 Th. Smith, Jour. Exp. Med., iv, 1899. « Park and Williams, Jour. Exp. Med., 1897. ''Spronk, Ann. de I'inst. Pasteur, 1895. 8 Park and Williams, Jour. Exp. Med., 1897. ' Brieger und Frankel, Berl. klin. Woch., xi-xii, 1889. 522 PATHOGENIC MICROORGANISMS ceeded in extracting from toxic bouillon a white, water-soluble powder which possessed most of the poisonous properties of the broth itself. This, in solution, gave many of the useful proteid reactions, but dif- fered from proteids in failing to coagulate when boiled and in not giving precipitates when treated with magnesium sulphate, sodium sulphate, or nitric acid. It was believed by them to be closely related to the albumoses, bodies representing intermediate phases in the pep- tonization of albumins. Similar results have been obtained by Was- sermann and Proskauer,^ Brieger and Boer,^ and others. Uschinsky,^ on the other hand, has disputed the proteid nature of toxins in gen- eral, having produced diphtheria toxin by growing the organism upon a medium entirely free from albuminous bodies. Uschinsky believes that the protein reactions observed by others may be due to ingredi- ents of the precipitates other than the toxin. It is not impossible, however, that the organisms may have produced proteid substances by synthesis from the simpler substances in Uschinsky 's medium. The production of toxin from such a medium, therefore, is not a conclusive argument against the proteid nature of toxins. Accurate chemical iso- lation and analysis of diphtheria toxin have not yet been accomplished. Diphtheria toxin is destroyed,* when in the fluid form, by tempera- tures of 58° to 60° C. In the dry state, it resists a temperature of 70° C. and over, without change. Light and free access of air produce rapid deterioration. Sealed, protected from light, and kept at almost freezing point, the toxin remains stable for long periods. Electrical currents passed through toxic broth have little or no effect upon it. Transmission. — Diphtheria is transmitted from one individual to another directly or indirectly by contact or droplet infection — as in coughing, etc. It has been found that individuals may retain virulent diphtheria bacilli in nose and throat for long periods after recovery from the disease. These are the so-called "diphtheria carriers.*' The problem of diphtheria carriers has become one of considerable importance and has been given special prominence of recent years by the studies of Von Scholly, Moss, and Nuttall and Graham Smith. Anderson, Goldberger and Hachtel ^ studied 4,0^9 healthy people in ^ Wassermann und Proskauer, Deut. med. Woch., 1891, p. 585. 2 Brieger und Boer, Deut. med. Woch., 1896, p. 783. 3 Uschinsky, Cent. f. Bakt., xxi, 1897. * Roux et Yersin, loc. cit. « Goldberger, Williams and Hachtel, Bull. No. 101, of the Hygienic Laboratories, of the U. S. Public Health Service. BACILLUS DIPHTHERLE 523 the city of Detroit, and found that 0.928% harhored bacilli identical morphologically with the Klebs-Loeffler bacillus. This figure is rather lower than those of some other investigators, but would indicate, as the writers stated, that there were from 5,000 to 6,000 diphtheria car- riers in the city of Detroit. Of 19 cultures isolated from 19 of the carriers, only 2 were virulent, which would indicate that only 0.097% of the people examined carried organisms capable of producing disease. An interesting further point is that the bacillus Hoffmanni was present in at least 41.9% of over 2,000 individuals examined, and that 47 cultures, morphologically identified as Bacillus Hofi:manni, were avirulent. This would confirm the impression gained, we believe, by most experienced laboratory workers that a true Hoffmanni can be distinguished with considerable certainty from a Klebs-Loeffler bacillus by morphological examination alone, and that its significance is probably that of a frequently present saprophyte of the throat and pharynx. The studies of Goldberger, Williams and Hachtel also indicate that in examining for diphtheria carriers it is best not to confine oneself either to the nose or throat, but that cultures should be taken from both places in every ease. Bacteria Similar to Bacillus Diphtherise. — Bacii^lus Hoffmanni {Pseudodiphtheria hacillus). — Hoffmann- Wellenhoff,^ in 1888, and, at almost the same time, Loeffler,^ described bacilli which they had culti- vated from the throats of normal persons and in several instances from those of diphtheritic persons, which were in many respects similar to true B. diphtheria, but differed from this chiefly in being non-patho- genic for guinea-pigs. These organisms were at first regarded by some observers as merely attenuated diphtheria bacilli. More recent inves- tigations, however, prove them to be unquestionably a separate species, easily differentiable by proper methods. They differ from B. diph- therise in so many important features, moreover, that the term ** pseu- dodiphtheria bacillus*' is hardly an appropriate one for them. Morphology. — Bacillus Hoffmanni is shorter and thicker than Ba- cillus diphtheria. It is usually straight and slightly clubbed at one end, rarely at both. Stained with Loeffler's blue it occasionally shows unstained transverse bands ; unlike B. diphtheriae, however, these bands hardly ever exceed one or twofn number at most. In many cultures the single transverse band gives the bacillus a diplococcoid appearance. 1 Hoffmann-Wellenhoff, Wien. med. Woch., iii, 1888. 2 Loeffler, Cent, f . Bakt., ii, 1887. 624 PATHOGENIC MICROORGANISMS :?' S%J'- '^ I staining. — Stained by Neisser's or Roux's method, no polar bodies can be demonstrated. The bacillus forms no spores, is non-motile, and possesses no flagella. Cultivation. — On the usual culture media B. Hoffmann! grows more luxuriantly than B. diphtheriae, developing even in first isolations from the human body upon the simple meat-extract media. On agar plates its colonies are larger, less transparent, and whiter than are those of true diphtheria bacilli. In fluid media there is even clouding and less ten- dency to the formation of a pellicle than with B. diphtheriae. A positive means of distinction between the two is given by the inability of B. Hoffmanni to form acid upon various sugar media. The differentiation on a basis of acid formation was first attempted by Cob- bett ^ and has been recently worked out systematically by Knapp,^ and con- firmed by various observers.^ The re- /p. ^ « ft.*^ / - r-4 ^^^^^ ^^ ^^i^ work, carried out with the t' " V^^ ' ~^^ V^ H serum-water media of Hiss, to which '■J. ^J-.^'^ k^^ ^2 various sugars were added, show that s%^ '*"■' , ,, ^jC^ 1 B. Hoffmanni forms acid upon none of ' ■■' -^>> . ^^ — J the sugars used, while B. diphtheriae Fig. 107. — Bacillus Hoffmanni. acidifies and coagulates media contain- ing monosaccharids and several of the more complex sugars, as given in the diagram in the section following, dealing with B. xerosis. Differentiation can finally be made on the basis of animal patho- genicity, B. Hoffmanni being entirely innocuous to the ordinary labora- tory animals. B. Hoffmanni forms no toxins, and animals immunized with it do not possess increased resistance to B. diphtherias. Bacillus xerosis. — In 1884, Kutschert and Neisser* described a bacillus, isolated from the eyes of patients suffering from a form of chronic conjunctivitis known as xerosis. This bacillus, which, morpho- logically, is almost identical with B. dif)htheriae, they believed to be the etiological factor of the disease. The frequency with which it has been 1 Cohbett, Cent. f. Bakt., 1898. ^ Knapp, Jour. Med. Res., vii, 1904. ' Graham Smith, Jour, of Hyg., vi, 1906; Zinsser, Jour. Med. Res., xvii, 1907. * Kutschert und Neisser, Deut. med. Woch., xxiv, 1884. BACILLUS DIPHTHERLE 525 isolated from normal eyes, precludes this etiological relationship, and it may safely be regarded as a harmless parasite which may indeed be more abundant in the slightly inflamed than in the normal conjunctiva. Morphology. — B* xerosis resembles B. diphtherias closely. It is occasionally shorter than this, but on the whole no absolute morpho- logical differentiation between the two is possible. It forms no spores and is non-motile. Polar bodies may occasionally be seen. Cultivation. — On Loeffler's Mood serum, on agar, glycerin agar, and in hroth, its growth is very similar to that of B. diph- therias, but more delicate through- out. It cannot easily be cultivated upon the simple meat-extract media, nor will it grow on gelatin at room temperature. Its colonies on glycerin or glucose agar are microscopically identical with those of B. diphtherise. Differentiation. — It differs from B. diphtherias distinctly in its acidifying action on sugar media. These relations were first worked out by Knapp for various sugars and the alcohol mannit, and have been extensively confirmed by others. See table on page 526. A reference to the table shows that differentiation may be made by the use of two sugars — saccharose and dextrin. B. diphtherias forms acid from dextrin, not from saccharose ; B. xerosis from saccha- rose, not from dextrin ; B. Hoffmanni does not form acid from either. B. xerosis is non-pathogenic to animals and forms no toxin. The Diphtheroid Bacilli. — In addition to the bacteria mentioned above, there is a large group of microorganisms spoken of as the diph- theroid bacilli, largely because of their morphological resemblance to the diphtheria bacillus. For this group, Lehman and Neumann have suggested the term corynehacterium. The characteristics of this group are a morphological similarity to the diphtheria bacillus, that they are Gram-positive, non-motile, often show metachromatic granules and have no spores. It is not, at the present writing, possible to formulate Fig. 108. — Colonies op Bacillus Hoffmanni on Agar. 526 PATHOGENIC MICROORGANISMS a classification of these organisms. They are apparently very numer- ous and have been isolated from a great many different sources, both Hiss serum-water media plus 1% B. Diphtheriae B. Xerosis B. Hoffmanni Dextrose + + + + Saccharose _ Dextrin _ ^ « gH '-^ ^ tor temperature. Recent studies by Anna Williams at the New York Department of Health seem to indicate that the Koch-Weeks bacillus may be merely a variety of the true influenza bacillus. Bacillus of Pleuro-Pneumonia of Rabbits. — This is a small Fig. 115. — Koch- Weeks Bacillus. Gram-negative bacillus, described by Beck, not unlike that of influenza. These microorganisms are slightly larger than the Pfeiffer bacilli and grow upon ordinary media even without animal sera or hemoglobin. Bacillus murisepticus and Bacillus RHUsioPATHiiE. — While mor- phologically similar to the microorganisms of this group, these bacilli are culturally easily separated because of their luxuriant growth on simple media. (The last two microorganisms are more closely related to the groups of the bacilli of hemorrhagic septicemia. See page 561.) 1 Koch, Arb. a. d. kais. Gesundheitsamt, iii; Cent. f. Bakt., 1, 1887. 2 Weeks, N. Y. Eye and Ear Infirmary Rep., 1895; Arch. f. Augenheilk., 1887. 3 Kamen, Cent. f. Bakt., xxv, 1899; Weichselbaum and Miiller, Arch. f. Ophthalm., xlvii, 1899; Knapp, Studies from Dept. of Path., Coll. of P. and S., 1903. CHAPTER XXXVIII BORDET-GENGOU BACILLUS, MORAX-AXENFELD BACILLUS, ZUR NEDDEN'S BACILLUS, DUCREY BACILLUS BORDET-GENGOU BACILLUS {" Microbe de la Coqueluche,'' Pertussis hacUlus, Bacilliis of whooping- cough.) In 1900 Bordet and Gengou ^ observed in the sputum of a child suffering from pertussis a small ovoid bacillus which, though similar to the influenza bacillus, showed a number of morphological characteristics which led them to regard it as a distinct species. As they were at first unable to cultivate this organism, their discovery remained ques- tionable until 1906, when cultivation succeeded and the biology of the microorganism was more fully elucidated. Morphology. — The morphology of this organism is described by them as follows: The organism in the sputum, early in the disease, is scattered in enormous numbers indiscriminately among the pus cells, and at times within the cells. It is extremely small and ovoid, and frequently is so short that it resembles a micrococcus. Often its poles stain more deeply than the center. In general, the form of the or- ganisms is constant, though occasionally slightly larger individuals are encountered. -They are usually grouped separately, though occa- sionally in pairs, end to end. Compared with the influenza bacillus in morphology, the bacillus of pertussis is more regularly ovoid and somewhat larger. It has, furthermore, less tendency to pleomorphism and involution. Staining. — The Bordet-Gengou bacillus may be stained with alkaline methylene-blue, dilute carbol-fuchsin, or aqueous fuchsin solutions. Bordet and Gengou recommended as a staining-solution carbolated toluidin-blue made up as follows: Toluidin-blue 5 gms. Alcohol 100 c.c. Water 500 c.c. 1 Bordet et Gengou, Ann. de I'inst. Pasteur, 1906. 36 543 544 PATHOGENIC MICROORGANISMS Allow to dissolve and add 500 c.c. of 5 per cent carbolic acid in water. Let this stand one or two days and filter. Stained by the method of Gram, the bacillus of Bordet and Gengou is decolorized. Cultivation. — Early attempts at cultivation made by the discov- erers upon ©rdinary ascitic agar or blood agar were unsuccessful. They finally obtained successful cultures from sputum by the use of the following medium: One hundred grams of shced potato are put into 200 c.c. of 4 per cent glycerin in water. This is steamed in an autoclave and a glycerin extract of potato obtained. To 50 c.c. of this extract 150 c.c. of 6-per- cent salt solution and 5 grams of agar are added. The mixture is melted in the autoclave and the fluid filled into test tubes, 2 to 3 c.c. each, and sterilized. To each tube, after sterilization, is added an equal volume of sterile de- fibrinated rabbit blood or preferably human blood, the substances are mixed, and the tubes slanted. On such a medium, inoculated with sputum, taken preferably during the par- oxysms of the first day of the disease, colonies appear, which are barely visible after twenty-four hours,^ plainly visible Fig. 116.-Bordet-Gengou ^^^^^ forty-eight hours. They are small. Bacillus. grayish, and rather thick. After the first generation the organisms grow with mark- edly greater luxuriance and speed. On the potato-blood medium, after several generations of artificial cultivation, they form a grayish glistening layer which, after a few days, becomes heavy and thick, almost resembling the growth of typhoid bacilli. In these later generations, also, they develop readily upon plain blood agar or ascitic agar and in ascitic broth or broth to which blood has been added. In the fluid media they form a viscid sediment, but no pellicle. Culturally, the 'bacillus varies from B. influenzae in growing less readily on hemoglobin media than the latter, on first cultivation from the sputum. Later it grows much more heavily on such media and shows less dependence upon the presence of hemoglobin than does B. influenzae. It also grows rather more slowly than the influenza bacillus. It is 1 Wollstein, Jour. Exp. Med., xi, 1909. w ",^"' - • " 1? . '*'* '* ^^.' ^^■,-'- U: .c i?' •' «.»** • ' '^•■ ' " «^ --^•x 1 '- ' *•• « BORDET-GENGOU BACILLUS 545 strictly aerobic and' in fluid cultures is best grown in wide flasks with shallow layers of the medium. The Bordet-Gengou bacillus grows moderately at temperatures about 37.5° C, but does not cease to grow at temperatures as low as 5° to 10° C. On blood agar and in ascitic broth it may remain alive for as long as two months (Wollstein). Pathogenicity. — As regards the pathogenicity and etiological spe- cificity of this organism for whooping-cough, no positive statement can as yet be made. The fact that it has been found in many cases in almost pure cultures during the early paroxysms, renders it likely that the organism is the specific cause of the disease. However, in early cases true influenza bacilli have been often found, and these latter seem to remain in the sputum of such patients for a longer period and in larger numbers than the bacillus of Bordet and Gengou. Endotoxins have been obtained from the cultures of the bacilli by Bordet and Gengou by the method of Besredka.^ The growth from slant cultures is washed up in a little salt solution, dried in vacuo, and ground in a mortar with a small, measured quantity of salt. Finally, enough distilled water is added to bring the salt into a solution of 0.75 per cent and the mixture is centrifugalized and decanted. One to two c.c. of such an extract will usually kill a rabbit within twenty-four hours after intravenous inocula- tion. Subcutaneous inoculation produces non-suppurating necrosis and ulceration without marked constitutional symptoms. Inoculation of monkeys with the bacilli themselves by the respira- tory path has failed to produce the disease. Specific agglutinins may be obtained in immunized animals which prove absolutely the distinctness of this organism from Bacillus in- fluenzae.^ In the serum of afflicted children the agglutination is too irregular to be of value. Specific complement fixation with the serum of patients is reported by Bordet and Gengou, but failed in the hands of Wollstein. MORAX-AXENFELD BACILLUS In 1896 Morax^ described a diplo-bacillus, which he associated etiologically with a type of chronic conjunctivitis to which he applied the name " conjonctivite suhaigueJ^ Soon after this, a similar micro- organism was found in cases corresponding to those of Morax by Axen- 1 Bordet, Bull, de la Soc. Roy. de Brux., 1907. ^ Wollstein, loc. cit. ^ Morax, Ann. de I'inst. Pasteur, 1896. 546 PATHOGENIC MICROORGANISMS feld.i The condition which these microorganisms characteristically produce is a catarrhal conjunctivitis which usually attacks both eyes. The inflammation is especially noticeable in the angles of the eye, most severe at or about the caruncle. There is rarely much swelling of the conjunctiva and hardly ever ulceration. The condition runs a subacute or chronic course. Its diagnosis is easily made by smear preparations of the pus which is formed with especial abundance during the night. Morphology. — In smear preparations from the pus, the microorgan- isms appear as short, thick bacilli, usually in the form of two placed end to end, but not infrequently singly or in short chains. Their ends Fig. 117. — Morax-Axenfeld Diplo-Bacillxjs. are distinctly rounded, their centers slightly bulging, giving the bacillus an ovoid form. They are usually about two micra in length. They are easily stained by the usual anilin dyes, and, stained by the method of Gram, are completely decolorized. Cultivation. — The Morax-Axenfeld bacillus can be cultivated only upon alkaline media containing blood or blood serum. It grows poorly, or not at all, at room temperature. Upon Loeffler^s blood serum, colonies appear after twenty-four to thirty-six hours as small indentations which indicate a liquefaction of the medium. Axenfeld states that eventually the entire medium may 1 Axenfeldy Cent, f . Bakt., xxi, 1897. ZUR NEDDEN'S BACILLUS 547 become liquefied. Upon serum agar delicate grayish drop-like colonies are formed which are not unlike those of the gonococcus. In ascitic bouillon general clouding occurs within twenty-four hours. Pathogenicity. — Attempts to produce lesions in the lower animals with this bacillus have been universally unsuccessful. Subacute con- junctivitis, however, has been produced in human beings by inocula- tions with pure cultures. ZUR NEDDEN'S BACILLUS In ulcerative conditions of the cornea, Zur Nedden has frequently found a bacillus to which he attributes etiological importance. The bacillus which he has described is small, usually less than one micron in length, often slightly curved, and generally found singly. It may be found in the diplo form but does not form chains. It is stained by the usual dyes, often staining poorly at the ends. Stained by Gram's method it is decolorized. The bacillus is non-motile. Cultivation. — It is easily cultivated upon the ordinary laboratory media. Upon agar it forms, within twenty-four hours, transparent, slightly fluorescent colonies which are round, raised, rather coarsely granular, and show a tendency to confluence. Gelatin is not liquefied. Milk is coagulated. Upon potato, there is a thick yellowish growth. Upon dextrose media, there is acid formation, but no gas. The bacillus forms no indol in pepton solutions. Pathogenicity. — Corneal ulcers have been produced by inoculation of guinea-pigs. BACILLUS OP DUCREY The soft chancre, or chancroid, is an acute inflammatory, destructive lesion which occurs usually upon the genitals or the skin surrounding the genitals. The infection is conveyed from one individual to an- other by direct contact. It may, however, under conditions of surgical manipulation, be transmitted indirectly by means of dressings, towels, or instruments. The lesion begins usually as a small pustule which rapidly ruptures, leaving an irregular ulcer with undermined edges and a necrotic floor which spreads rapidly. It differs cHnically from the true or syphilitic chancre in the lack of induration and in its violent inflammatory 548 PATHOGENIC MICROORGANISMS nature. Usually it leads to lymphatic swellings in the groin which, later, give rise to abscesses, commonly spoken of as "buboes." In the discharges from such lesions, Ducrey,^ in 1889, was able to demonstrate minute bacilli to which he attributed an etiological rela- tionship to the disease, both because of the regularity of their presence in the lesions and the successful transference of the disease by means of pus containing the microorganisms. Morphology and Staining. — The Ducrey bacillus is an extremely small bacillus, measuring from one to two micra in length and about half a micron in thickness. It has a tendency to appear in short chains and in parallel rows, but many of the microorganisms may be seen irregularly grouped. It is not motile, possesses no flagella, and does not form spores. Stained by the ordinary anilin dyes, it has a tendency to take the color irregularly and to appear more deeply stained at the poles. By the Gram method, it is decolorized. In tissue sections, it may be demon- striited by Loeffler's methylene-blue method, and in such preparations has been found within the granulation tissues forming the floor of the ulcers. In pus, the bacilli are often found within leucocytes. Cultivation and Isolation. — Early attempts at cultivation of this bacillus were universally unsuccessful in spite of painstaking experi- ments with me^ia prepared of human skin and blood serum. In 1900, Besangon, Griffon, and Le Sourd ^ finally succeeded in obtaining growths upon a medium containing agar to which human blood had been added. They were equally successful when dog's or rabbit's blood was substi- tuted for that of man. Since the work by these- authors, the cultiva- tion by similar methods has been carried out by a number of investiga- tors. Coagulated blood, which has been kept for several days in sterile tubes, has been found to constitute a favorable medium. Freshly clotted blood can not be employed, probably because of the bacteri- cidal action of the serum. Serum-agar has occasionally been used with success, but does not give results as satisfactory as those obtained by the use of the whole blood. The best method of obtaining pure cultures upon such media con- sists in puncturing an unruptured bubo with a sterile hypodermic needle and transferring the pus in considerable quantity directly to the agar. If possible, the inoculation of the media should be made immediately 1 Ducrey, Monatschr. f. prakt. Dermat., 9^ 1889. 2 Besangon, Griffon, et Le Sourd, Presse m^d., 1900. BACILLUS OF DUCREY 549 before the pus has had a chance to cool off or to be exposed to light. When buboes are not available, the primary lesion may be thoroughly cleansed with sterile water or salt solution, and material scraped from the bottom of the ulcer or from beneath its overhanging edges with a stiff platinum loop. This material is then smeared over the surface of a number of blood-agar plates. Upon such plates, isolated colonies appear, usually after forty-eight hours. They are small, transparent, and gray, and have a rather firm, finely granular consistency. The colonies rarely grow larger than pin- head size, and have no tendency to coalesce. At room temperature, the cultures die out rapidly. Kept in the incubator, however, they may remain alive and virulent for a week or more. On the simpler media, glucose-agar, broth, or gelatin, cultivation is never successful. On moist blood-agar and in the condensation water of such tubes, the bacilli have a tendency to grow out in long chains. Upon media which are very dry, they appear singly or in short chains. Pathogenicity. — Besangon, Griffon, and Le Sourd, and others, have succeeded in producing lesions in man by inoculation with pure cultures. Inoculation of the lower animals has, so far, been entirely without result. MICROCOCCUS MELITENSIS (MALTA FEVER) (Bacillus melitensis) Malta fever is a disease occurring along the Mediterranean coast and its islands. It has been recently found to " occur also in South America, South Africa, China, and in the West Indies. The disease is not very unlike typhoid fever, though more irregular and with a lower mortality. It is accompanied by joint pains, sweating, constipa- tion, and occasionally orchitis. The spleen is almost always enlarged. Recent investigations into the manner in which this disease is con- veyed have revealed that it is primarily an infection of goats. A large percentage of the goats on Malta were shown to be infected and passed the organism with the milk. Forty per cent of the goats gave positive agglutination tests and the organisms have been found in the milk in about 10 per cent of the animals. The most susceptible animals seem to be goats, but horses and cows are also susceptible. In guinea-pigs and rabbits the disease can be ex- perimentally produced, but usually takes a protracted course. Monkeys 550 PATHOGENIC MICROORGANISMS are susceptible, and the disease produced in these animals is in many- features identical with that of man. Transmission seems to take place chiefly by the ingestion of infected milk. Direct cutaneous infection or through mucous membranes may also occur. In human beings, suffering from the disease, the organisms may be isolated from the blood stream during the entire course of the disease and as early as the second day. The disease is rarely fatal, death occurring in less than 2 per cent of the cases (Eyre, loc. cit.)} The microorganism causing the disease was isolated in 1887 by Bruce,^ a British army surgeon. Morphology. — Micrococcus melitensis is a minute bacterium ap- pearing coccoid in smears from agar cultures, in broth cultures assum- ing the form of a short, slightly wedge-shaped bacillus resembling B. influenzae. Babes ^ regards it as unquestionably a bacillus. Eyre de- scribes it as a minute coccus, and beUeves the bacillus-hke individuals to represent involution forms. It appears in irregularly parallel groups, and occasionally forms short chains. It is easily stained with the ordinary dyes, and is decolorized by Gram's method. Cultivation. — Micrococcus melitensis can usually be cultivated from the spleens of those who have succumbed to the disease and from the blood stream in active cases. It grows on nutrient agar at 37.5° C, forming small, pearly white colonies at the end of two or three days. It grows easily on all of the ordinary laboratory media. Both in patients and in injected animals, infection with this bacte- rium produces specific agglutinins which are of great practical aid in diagnosis.^ 1 British Commission Report cited from Eyre in Kolle und Wassermanrit Handbuch, etc., Erganzungsband, Heft 2. 2 Bruce, Practitioner, 1887. ' Babes, Kolle und Wassermann, iii, p. 443. * Wright and Lamb, Jour. Path, and Bact., v, 1899. CHAPTER XXXIX THE BACILLI OF THE HEMORRHAGIC SEPTICEMIA GROUP AND BACILLUS PESTIS. In many of the lower animals there occur violently acute bacterial infections characterized by general septicemia, usually with petechial hemorrhages throughout the organs and serous membranes and severe intestinal inflammations. These diseases, spoken of as the "hemor- rhagic septicemias,'' are caused by a group of closely allied bacilli, first classified together by Hueppe ^ in 1886. Some confusion has existed as to the forms which should be considered within Hueppe's group of "hemorrhagic septicemia," a number of bacteriologists including in this class bacilli such as Loefl^er's Bacillus typhi murium, and Salmon and Smith's hog-cholera bacillus, microorganisms which, because of their motility and cultural characteristics, belong more properly to the "Gartner," " enteritidis," or "paratyphoid" group, intermediate be- tween colon and typhoid. The organisms properly belonging to this group are short bacilli, more plump than are those of the colon type, and showing a marked ten- dency to stain more deeply at the poles than at the center. They are non-motile, possess no flagella, and do not form spores. They grow readily upon simple media, but show a very marked preference for oxygen, growing but slightly below the surface of media. By some observers they are characterized as "obligatory aerobes," but this is undoubtedly a mistake. \Miile showing considerable variations in form and differences in minor cultural characteristics, the species characteristics of polar stain- ing, decolorization by Gram, immobility, lack of gelatin liquefaction, and great pathogenicity for animals, stamp alike all members of the group. Its chief recognized representatives are the bacillus of chicken cholera, the bacillus of swine-plague (Deutsche Schweineseuche) , » Hueppe, Berl. klin. Woch., 1886. 551 552 PATHOGENIC MICROORGANISMS and the Bacillus pleurosepticus which causes an acute disease in cattle and often in wild game. Because of certain cultural and pathogenic characteristics, it seems best to consider the bacillus of bubonic plague with this group. BACILLUS OF CHICKEN CHOLERA (Bacilltcs avisepticus) The bacillus of chicken cholera was first carefully studied by Pas- teur ^ in 1880. It is a short, non-motile bacillus, measuring from 0.5 to 1 micron in length. Stained with the ordinary anilin dyes, it displays marked polar staining qualities, which often give it the appearance of being a diplococcus. It is decolorized by Gram's method. It does not form spores, but may occasionally form vacuolated degeneration forms, not unlike those described for Bacillus pestis. The bacillus is easily cultivated from the blood and organs of infected animals, it grows well upon the simplest media at temperatures vary- ing from 25° to 40° C. In broth , it produces uniform clouding with later a formation of a pellicle. Upon agar it forms, within twenty-four to forty-eight hours, minute colonies, white or yellowish in color, which are at first transparent, later opaque. Upon gelatin^ it grows without liquefaction. Upon milk, the growth is slow and does not produce co- agulation. According to Kruse,^ indol is formed from pepton bouillon. Acid, but no gas, is formed in sugar broth. Among barnyard fowl, this disease is widely prevalent, attacking chickens, ducks, geese, and a large variety of smaller birds. The infection is extremely acute, ending fatally within a few days. It is accompanied by diarrhea, often with bloody stools, great exhaustion, and, toward the end, a drowsiness bordering on coma. Autopsy upon the animals re- veals hemorrhagic inflammation of the intestinal mucosa, enlargement of the liver and spleen, and often bronchopneumonia. The specific bacilli may be found in the blood, in the organs, in exu- dates, if these are present, and in large numbers in the dejecta. Infection takes place probably through the food and water contaminated by the discharges of diseased birds.^ Subcutaneous inoculation or feeding of such animals with pure cultures, even in minute doses, gives rise to a quickly developing septicemia which is uniformly fatal. The bacillus is extremely patho- » Pasteur, Comptes rend, de I'acad. des sci., 1880. siiTn^e, in Fliigge's "Die Mikroorganismen." 3 Salmon, Rep. of the Com. of Agriculture, 1880, 1881, and 1882. BACILLI OF HEMORRHAGIC SEPTICEMIA GROUP 553 genie for rabbits, less so for hogs, sheep, and horses, if infection is prac- ticed by subcutaneous inoculation. Infection by ingestion does not seem to cause disease in these animals. Historically, the bacillus of chicken cholera is extremely interesting, since it was with this microorganism that Pasteur ^ carried out some of his fundamental researches upon immunity, and succeeded in immu- nizing chickens with attenuated cultures. The first attenuation ex- periment made by Pasteur consisted in allowing the bacilli to remain in a broth culture for a prolonged period without transplantation. With minute doses of- such a culture (vaccin I) he inoculated chickens, fol- lowing this, after ten days, with a small dose of a fully virulent culture. Although enormously important in principle, the practical results from this method, as applied to chicken cholera, have not been satisfactory. It was with this bacillus, furthermore, that Pasteur was first able to demonstrate the existence of a free toxin which could be separated from the bacteria by filtration. BACILLUS OP SWINE PLAGUE {Bacillus suisepticuSj Schweineseuche) This microorganism is almost identical in form and cultural charac- teristics with the bacillus of chicken cholera. It is non-motile, forms no spores, is Gram-negative, and does not liquefy gelatin. The bacillus causes an epidemic disease among hogs, which is characterized almost regularly by a bronchopneumonia followed by general septicemia. There is often a sero-sanguineous pleural exudate, a swelling of bronchial lymph glands and of liver and spleen. The gastrointestinal tract is rarely affected. The bacilli at autopsy may be found in the lungs, in the exudates, in the liver and spleen, and in the blood. The disease is rarely acute, but, in young pigs, almost uniformly fatal. It is probable that spontaneous infection usually occurs by inhala- tion. Experimental inoculation is successful in pigs, both when given subcutaneously and when administered by the inhalation method. Mice, guinea-pigs, and rabbits are also susceptible, dying within three or four days after subcutaneous inoculation of small doses. Active and passive immunization of animals against Bacillus suisep- ticus has been attempted by various observers. Active immunization, if carried out with care, may be successfully done in the laboratory. 1 Pasteur, loc. cit. 554 PATHOGENIC MICROORGANISMS Passive immunization of animals with the serum of actively immunized horses has been practiced by Kitt and Mayr/ Schreiber,^ and Wasser- mann and Ostertag. The last-named observers, working with a poly- valent serum produced with a number of different strains of the bacillus, have obtained results of considerable practical value. The researches of Kitt and Mayr have revealed a fact pointing to the interrelationship of the bacilli of the "hemorrhagic septicemia" group. They were able to show that the serum of horses immunized with chicken cholera bacilli was able to protect, somewhat, against Bacillus suisepticus. Infection with the bacillus of swine plague, in hogs, is often ac- companied by an infection with the hog-cholera bacilhis (Schweinepest). The latter, as we have seen, is a microorganism belonging to the enteri- tidis group, intermediate between Bacillus coli and Bacillus typhosus, and differing from suisepticus in being actively motile, possessing flagella, not showing the polar staining, having a more slender morphology, and producing gas upon dextrose broth. A confusion between the two bacilli frequently occurs because of their nomenclature. Bacteriologic- ally and pathogenically, they are quite distinct. Bacillus suisepticus produces an acute septicemia, accompanied by bronchopneumonia and usually not affecting the gastro-intestinal canal. The bacillus of hog cholera produces an infection localized in the intestinal canal. BACILLUS PESTIS (Bacillus of Bubonic Plague) The history of epidemic diseases has no more terrifying chapter than that of plague.^ Sweeping, time and again, over large areas of the civilized world, its scope and mortality were often so great that all forms of human activity were temporarily paralyzed. In the reign of Justinian almost fifty per cent of the entire population of the Roman Empire perished from the disease. The "Black Death" which swept over Europe during the fourteenth century killed about twenty-five million people. Smaller epidemics, appearing in numerous parts of the world during the sixteenth, seventeenth, and eighteenth centuries, hare claimed innumerable victims. In 1893, plague appeared in Hong Kong. During the epidemic which followed, Bacillus pestis, now recognized as the etiological factor of the disease, was discovered by » Kitt und Mayr, Monatsh. f. prakt. Thierheilk., 8, 1897. ^Schreiber, Berl. tierarztl. Woch., 10, 1899. « Hirsch. " Handb. d. histor.-geogr. Path.," 1881. BACILLUS PESTLS 555 Kitasato * and by Yersin,^ independently of each other. By both ob- servers the bacillus could invariably be found in the pus from the buboes of afflicted persons. It could be demonstrated in enormous numbers in the cadavers of victims. The constancy of the occurrence of the bacillus in patients, shown in the innumerable researches of many bacteriologists, would alone be sufficient evidence of its etiological relationship to the disease. This evidence is strengthened, moreover, by accidental infections which occurred in Vienna in 1898, with labora- tory cultures. Morphology and Staining.— Bacillus pestis is a short, thick bacillus with well-rounded ends. Its length is barely two or two and a half times Fig. 118. — Bacillus pestis. (After Mallory and Wright.) its breadth (1.5 to 1.75 micra by 0.5 to 0.7 micron). The bacilli appear singly, in pairs, or, more rarely, iii short chains of three or more. They show distinct polar staining. In size and shape these bacilli are sub- ject to a greater degree of variation than are most other microorganisms. In old lesions or in old cultures the bacilli show involution forms which may appear either as swollen coccoid forms or as longer, club-shaped, diphtheroid bacilli. Degenerating individuals appear often as swollen, oval vacuoles. All these involution forms, by their very irregularity, a?e of diagnostic importance. They appear more numerous in artificial fiultures than in human lesions. According to Albrecht and Ghon,^ the plague bacillus may, by » Kitasato, Lancet, 1894. 2 Yersin, Ann. de I'inst. Pasteur, 1894. » Albrecht und Ghon, Wien, 1898. 556 PATHOGENIC MICROORGANISMS special methods, be shown to possess a gelatinous capsule. It does not possess flagella and does not form spores. The plague bacillus is easily stained with all the usual anilin dyes. Diluted aqueous fuchsin and methylene-blue are most frequently employed. With these stains the characteristically deeper staining of the polar portions of the bacillus is usually easy to demonstrate. Special polar stains have been devised by various observers. Most of these depend upon avoidance of the usual heat fixation of the prepara- tions, which, in some way, seems to interfere with good polar staining. Fixation of the dried smears with absolute alcohol is. therefore, prefer- able. The bacillus is decolorized by Gram's method. Fig. 119. — Bacillus pestis. Involution Forms. (After Zettnow.) Isolation and Cultivation. — The bacillus is easily isolated in pure culture from the specific lesions of plague patients, during life or at autopsy. It grows readily and luxuriantly upon the meat-infusion media. The optimum temperature for its cultivation is about 30° C. Below 20° C. and above 38° C, growth is sparse and delayed, though it is not entirely inhibited until exposed to temperatures below 12° C, or above 40° C. The most favorable reaction of culture media is neu- trality or moderate alkalinity, though slight acidity does not prevent development. On agar, growth appears within twenty-four hours as minute colonies with a compact small center surrounded by a broad, irregularly indented, granular margin. BACILLUS PESTIS 557 On gelatin, similar colonies appear after two or three days at 20° to 22° C. The gelatin is not liquefied. In bouillon, the plague bacilli grow slowly. They usually sink to the bottom or adhere to the walls of the tube as a granular deposit and may occasionally form a deUcate pelhcle. Chain-formation is not un- common. In broth cultures, moreover, a peculiar stalactite-like growth is often seen, when the culture fluid is covered with a layer of oil. Delicate threads of growth hang down from the surface of the medium into its depths like stalactites. Characteristic involution forms are brought out best when the bacilli are grown upon agar containing 3 per cent NaCl. Milk is not coagulated. In litmus-milk there is slight acid forma- tion. On potato and on blood serum the growth shows nothing char- acteristic or of differential value. On pepton media no indol is formed. Biological Considerations. — Bacillus pestis is aerobic. Absence of free oxygen is said to prevent its growth, at least under certain condi- tions of artificial cultivation. It is non-motile. Outside of the animal body the bacilli may retain viability for months and even years if preserved in the dark and in a moist environment. In cadavers they may live for weeks and months if protected from dryness. In pus or sputum from patients they may live eight to fourteen days. These facts are of great hygienic importance. Complete drying in the air kills the bacilli within two or three days.^ Thoroughly dried by artificial means they die within four or five hours. Dry heat at 100° C. kills the bacillus in one hour.^ Live steam or boil- ing water is effectual in a few minutes. The bacilli possess great resist- ance against cold, surviving a temperature of 0° C. for as many as forty days. Direct simlight destroys them within four or five hours. The common disinfectants are effectual in the following strengths: carboUc acid, one per cent kills them in two hours, five per cent in ten minutes; bichloride of mercury 1 : 1,000 is effectual in ten minutes. In a recent communication to the New York Pathological Society, Dr. Wilson reported that plague cultures which he had kept sealed for as long as ten years in the ice chest were found living and virulent at the end of this time. This ability to go into a quasi latent stage under suitable conditions is of the greatest importance in connection with the problem of prevention. 1 Kitasato, Lancet, 1894. 2 ^5^^^ Cgnt, f^ Bakt., xxi, 1897. 558 PATHOGENIC MICROORGANISMS Pathogenicity. — In man, plague is acquired^ by entrance of the bacil- lus either through the skin or by the respiratory tract. The period of incubation is about three to seven days. Two distinct clinical types of the disease occur, depending upon the mode of infection. When cutaneous infection has occurred the disease is first locaUzed in the lymph nodes nearest the point of inoculation. If the respiratory tract has been the portal of entrance the disease primarily takes the form of a pneumonia. Infection may take place through the most minute lesions, hardly visible to the naked eye. Even the unbroken skin may admit the microorganisms if these are rubbed in with sufficient energy. From the primary lymphatic swellings, the bacilli enter the blood and may pro- duce secondary foci. The pneumonic form of plague usually begins with symptoms not unlike a typical pneumonia and is usually fatal within four or five or even fewer days. This form of the disease is especially menacing as a means of dissemination, because of the enormous numbers of plague bacilU in the sputum. One of the chief characteristics of the general systemic plague infec- tion is the very marked cardiac depression. The bacteriological diagnosis during life may be made by finding the bacilli in the sputum or in aspiration fluid from a bubo. The micro- organisms are identified morphologically, culturally, by animal experi- ment, and by agglutination reaction. Blood cultures from plague ph- tients often yield positive results, especially when the blood is well diluted in neutral broth to prevent any inhibiting action of the anti- bodies in the serum. At autopsy, in man, the bacilli are found in the primary lesions, in the blood, and in the spleen, the liver, and the lymphatics. There may be hemorrhages into the serous cavities. When pneumonia exists, it usually takes the form of a bronchopneiunonia with extensive swelling of the bronchial lymph nodes. In cases in which the disease is prolonged, there are often tubercle- like foci in the spleen, liver, and lungs. Histologically these foci show central necrosis surrounded by the usual inflammatory cell reactions. In more chronic cases connective-tissue encapsulation may appear. Bacillus pestis is extremely pathogenic for rats, mice, guinea-pigs, rabbits, and monkeys. The most susceptible of these animals are rats 1 Gottschlich, Zeit. f . Hyg., xxxv, 1900. BACILLUS PESTIS 559 and guinea-pigs, in whom mere rubbing of plague bacilli into the un- broken skin will often produce the disease. This method of experimen- tal infection of guinea-pigs is of great service in isolating the plague bacillus from material contaminated with other microorganisms. For the same purpose, infection of rats subcutaneously at the root of the tail may be employed. Such inoculation in rats is invariably fatal. The studies of McCoy ^ upon guinea-pigs and white rats show that individual plague cultures may vary considerably in virulence. The size of the dose, always excepting enormous quantities such as a whole agar culture, seems to make little difference in the speed with which the animals die. There may be considerable variation in the suscep- tibility of individual animals. Prolonged cultivation on artificial media may gradually reduce the virulence of plague bacilli, though, as stated above, this has not been the experience of all observers. In rats, spontaneous infection with plague is common and plays an important role in the spread of the disease. Rats become infected from the cadavers of plague victims or by gnawing the dead bodies of other rats dead of the disease. The pneumonic type of the disease is common in these animals and has been produced in them by inhalation experi- ments. During every well-observed plague epidemic, marked mortality among the domestic rats has been noticed. Since the examination of rats for plague is an important phase of the study of epidemics, it may be well to review the typical lesions in these animals as described by an experienced American student of plague, George W. McCoy.^ McCoy, agreeing with the Indian Plague Com- mission, states that the naked eye is superior to the microscopical ex- amination. There is engorgement of the subcutaneous vessels and a pink coloration of the muscles. The bubo when present is sufficient for diagnosis. Marked injection surrounds it and sometimes there is hemorrhagic infiltration. The gland itself is firm but usually caseous or occasionally hemorrhagic. In the liver there is apparent fatty change, but this is due to necrosis. Pin-point spots give it a stippled appear- ance as though it had been dusted with pepper. Pleural effusion is an important sign. The spleen is large, friable, and often presents pin- point granules on the surface. One or two per cent of rats may present no gross lesions. Cultures should of course be made. The method of examination consists in immersing the rat in any convenient antiseptic 37 1 McCoy, Jour, of Inf. Dis., vi, 1909. 2 Gewge W. McCoy, PubUc Health Reports, July^ 1912, 560 PATHOGENIC MICROORGANISMS to kill fleas and other ectoparasites. The rats are nailed by their feet to a shingle and the skin is reflected from the whole front of the body and neck so as to expose the cervical, axillary, and inguinal regions. The thoracic and abdominal cavities are then opened and examined. Wherry,^ McCoy ,2 and others have found that the California ground squirrel was infected with plague, during the recent occurrence of plague on the Pacific coast, and several cases of plague in man were traced to this source. In studying these and other American ro- dents McCoy found that ground squirrels as a species were highly susceptible, never showing natural immunity. Field mice were but moderately susceptible. Gophers were highly resistant. McCoy has also described a case of spontaneous infection in a brush rat (Neo- toma fuscipes). Rock squirrels were found by McCoy to be readily infected. Wu Lien Teh (G. L. Tuck) ^ has recently found that the Manchurian tarbagan or marmot (Arctomys bobac), an animal trapped for its fur, occasionally suffers from plague. The disease is never extensive and the animal of much less importance in spreading the disease than is the rat. The two principal species of rats to be considered in this connection are Epimys norvegicus and Epimys rattus. The spread from rat to rat, according to the Second Indian Commission, is entirely by means of infected fleas. The ordinary spread of the disease to man, according to this same commission, comes from Epimys rattus, which lives in close relation- ship with man and is conveyed to man largely by the rat flea, Xenop- sylla cheopis. This flea leaves the dead rat in about three days, and is capable of living for three or four weeks on man's blood. The plague bacilli need about three days' incubation in the body of the flea. Summarizing the knowledge at present available about the spread of the plague, it seems likely that, excepting in the case of pneumonic plague, the ordinary method is by means of rat fleas. It is a curious fact observed by various bacteriologists that plague bacilli isolated from pneumonic cases are particularly apt to cause pneumonic lesions, having, as it were, acquired a selective pathogenicity for the lung. A most valuable contribution to our knowledge of pneu- 1 Wherry, Jour. Inf. Dis., v, 1908. 2 McCoy, Jour. Inf. Dis., vi, 1909; vii, 1910. 5 Wu Lien Teh, JouT. of Hyg., xiii, 1913. 6ACILLUS PESTIS 561 monic plague has recently been made by Strong, Teague, and Barber ' in their report of the American Red Cross Expedition to Manchuria dm-ing the plague epidemic of 1910-11. Their investigations were made with remarkable courage and skill under difficult conditions. The chief points of interest in their reports may be sunmiarized as follows: Expired air of plague patients rarely contains the bacilli; these are thrown out in coughing or hawking. Transmission is, in this form, direct from patient to patient and not indirect through animals. The first localization (Strong, Teague, and Crowell) is in the bronchi from which extension takes place. Septicemia soon follows the pneumonic process. Spreading occurs most likely in the wet and cold of winter, since the bacteria are rapidly destroyed by drying. Toxin Formation. — The systemic symptoms of plague are largely due to the absorption of poisonous products of the bacteria. Albrecht and Ghon,2 Wernicke,^ and others were unable to obtain any toxic action with broth-culture filtrates and concluded that the poisons of B. pestis were chiefly endotoxins, firmly attached to the bacterial body. Kossel and Overbeck,^ however, on the basis of a careful investigation, came to the conclusion that, in addition to the endotoxin, there is formed in older broth cultures a definite and important true, soluble toxin. • Immunization. — A single attack of plague usually protects human beings from reinfection. A second attack in the same individual is extremely rare. Immunization in animals produces specific agglutinat- ing and bacteriolytic substances which are of great importance in the bacteriological diagnosis of the bacillus. The agglutinating action of the serum of patients is clinically important in the diagnosis of the disease, even in dilutions of one in ten, since undiluted normal human serum has no agglutinating effect upon plague bacilli. Active immunization of animals ^ is accomplished by inoculation of the whole dead bacteria. Haffkine has attempted active immunization in human beings by subcutaneous treatment with sterilized broth cul- tures of B. pestis. Gaffky ^ and his collaborators recommend, for similar purposes, forty-eight-hour agar cultures of a bacillus of standard viru- lence, emulsified in bouillon and sterilized at 65° C. 1 Strong, Teague, and Barber, Philippine Jour, of Sc, Sect. B, vii, 1912, No. 3. * Albrecht und Ghon, loc. cit. 3 Wernicke, Cent. f. Bakt., Ref., xxiv, 1898. ; * Kossel und Overbeck, Arb. a. d. Gesundh., xviii, 1901. ^ Yerdn, Calmette, et Roux, Ann. de I'inst. Pasteur, 1895. ^ Gaffky, PfeiffeTy Stickery und Dieudonne, Aih. a. d. kais. Gesundheitsamt, xvi, 1899. 562 PATHOGENIC MICROORGANISMS The curative plague serum prepared by Yersin and others by the immunization of horses with plague cultures has been extensively used in practice and though often disappointing, a definitely beneficial in- fluence on the milder cases has been noted. The sera are standardized by their protective power as measured in white rats. THE PLAGUE-LIKE DISEASE OF RODENTS (McCOY) ^ Bacterium Tularense (McCoy and Chapin)^ McCoy has described a disease occurring in Californian ground squirrels (Citellus beechyi) which presents lesions very similar to those of plague in these animals. In fact the disease was noticed in the course of the systematic examination of rodents by McCoy at the Federal Laboratory in San Francisco. Although McCoy was able to transmit the disease to guinea-pigs, mice, rabbits, monkeys, and gophers, and plague-like lesions could be produced in most of the animals, he was at first entirely unable to cultivate any organism from these lesions. In 1912 McCoy and Chapin finally succeeded in growing the specific bacterium on an egg medium made entirely of the yolk. Mor- phologically it is a very small rod, 0.3 to 0.7 micron in length and often capsulated. The rods stain poorly with methylene blue, better with carbol fuchsin or gentian violet. They are found in large numbers in the spleens of animals dead of the disease. 1 McCoy, U. S. Public Health Bull. 43, 1911. 2 McCoy and Chapin, Jour, of Inf. Dis., x, 1912. CHAPTER XL BACILLUS ANTHRACIS AND ANTHRAX {Milzbrand, Charhon) . Anthrax is primarily a disease of the herbivora, attacking especially cattle and sheep. Infection not infrequently occurs in horses, hogs, and goats. In other domestic animals it is exceptional. Man is susceptible to the disease and contracts it either directly from the living animals or from the hides, wool, or other parts of the cadaver used in the industries. The history of the disease dates back to the most ancient periods and anthrax has, at all times, been a severe scourge upon cattle- and sheep- raising communities. Of all infections attacking the domestic animals no other has claimed so many victims as anthrax. In Russia, where the disease is most common, 72,000 horses are .said to have succumbed in one year (1864) .^ In Austro-Hungary, Germany, France, and the Eastern countries, each year thousands of animals and numerous human beings perish of anthrax. In England and America the disease is relatively infrequent. No quarter of the globe, however, is entirely free from it. Especial historical interest attaches to the anthrax bacillus in that it was the first microorganism proved, definitely to bear a specific etio-* logical relationship to an infectious disease. The discovery of the an- thrax bacillus, therefore, laid, as it were, the cornerstone of modem bacteriology. The bacillus was first observed in the blood of infected animals by PoUender in 1849, and, independently, by Brauell in 1857. Davaine,^ however, in I'SBS, was the first one to produce experimental infection in animals with blood containing the bacilli and to suggest a direct etiological relationship between the two. Final and absolute proof of the justice of Davaine's contentions, however, was not brought until the further development of bacteriological technique, by Koch,^ had made it possible for this last observer to isolate the bacillus upon 1 Quoted from Sobernheim, Kplle und Wassermann., vol. ii. 2 Davaine, Comptes rend, de I'acad. des sci., Ivii, 1863. » Koch, Cohn's '' Beitr. z. Biol. d. Pflanzen," ii, 1876. 563 564 PATHOGENIC MICROORGANISMS artificial media and to reproduce the disease experimentally by inocu- lation with pure cultures. Morphology and Staining. — The anthrax bacillus i^ a straight rod, 5 to 10 micra in length, 1 to 3 micra in width. It is non-motile. In preparations made from the blood of an infected animal, the bacilli are usually single or in pairs. Grown on artificial media they form tangles of long threads. Their ends are cut off squarely, in sharp con- FiG. 120. — Bacillus anthracis. From pure culture on agar. trast to the rounded ends of many other bacilli. The comers are often sharp and the ends of bacilli in contact in a chain often touch each other only at these points, leaving in consequence an oval chink between the ends of the organisms. The appearance of a chain of anthrax bacilli, therefore, has been not inaptly compared to a rod of bamboo. On artificial media the anthrax bacillus forms spores. Oxygen is necessary for the formation of these spores and they are consequently not found BACILLUS ANTHRACIS AND ANTHRAX 565 in the blood of infected subjects. The spores are located in the middle of the bacilli and are distinctly oval. They are difficult to stain, but may be. demonstrated by any of the usual spore-staining procedures, such as Moller's or Novy's methods. The bacilH themselves are easily stained by the usual anilin dyes, and gentian-violet or fuchsm in aque- ous solution may be conveniently employed. They are not decolorized by Gram's method. In preparations from animal tissues or blood, stained by special pro- FiG. 121. — Bacillus anthkacis. In section of kidney of animal dead of anthrax. cedures, the anthrax bacillus may occasionally be seen to . possess a cap- sule. The capsule is never seen in preparations from the ordinary artificial media. Some observers have demonstrated them in cultures grown in fluid blood serum. In chains of anthrax bacilli, the capsule when present seems to envelop the entire chain and not the individual bacteria separately. Isolation. — Isolation of the anthrax bacillus from infected material 566 PATHOGENIC MICROORGANISMS is comparatively simple, both because of the ease of its cultivation and because of the sharply characteristic features of its morphological and cultural appearance. Cultivation. — The anthrax bacillus is an aerobic, facultatively anaero- bic bacillus. While it may develop slowly and sparsely under anaerobic conditions, free oxygen is required to permit its luxuriant and charac- teristic growth. The optimum temperature for its cultivation ranges about 37.5° C. It is not, however, delicately susceptible to moderate variations of tem- Fig. 122. — Bacillus anthracis. In smear of spleen of animal dead of anthrax. perature and growth does not cease until temperatures as low as 12° C. or as high as 45° C. are reached. By continuous cultivation at some of the temperatures near either the higher or the lower of these limits, the bacillus may become well adapted to the new environment and attain luxuriant growth.^ The anthrax bacillus may be cultivated on all of the usual artificial media, growing upon the meat-extract as well as upon the meat-infusion media. 1 Dieudonne, Arb. a. d. kais. Gesundheitsamt, 1894. BACILLUS ANTHRACIS AND ANTHRAX 567 It may be cultivated also upon hay infusion, various other vegetable media, sugar solutions, and urine. While moderate acidity of the medium does not prevent the growth of this bacillus, the most favorable reaction for media is neutrality or slight alkalinity. On gelatin plates, colonies develop within twenty-four to forty-eight hours as opaque, white disks, pin-head in size, irregularly round and rather flat. As the colonies increase in size their outlines become less regular and under the microscope they are seen to be made up of a hair-like tangle of threads spreading in thin wavy layers from a more compact central knot. The microscopic appearance of these colonies has been aptly described as resembling a Medusa head. Fragments of a rv^>^'5^5 Fig. 123. — ^Anthrax Colony on Gelatin. (After Giinther.) colony examined on a slide with a higher power show the individual threads to be made up of parallel chains of bacilli. After a day or two of further growth, the gelatin about the colonies becomes fluidified. In gelatin stab cultures, growth appears at first as a thin white line along the course of the puncture. From this,- growth proceeds in thin spicules or filaments diverging from the stab, more abundantly near the top than near the bottom of the stab, owing to more active growth in well oxygenated environment. The resulting picture is that of a small inverted "Christmas tree.'' Fluidification begins at the top, at first a shallow depression filled with an opaque mixture of bacilli and fluid. 568 PATHOGENIC MICROORGANISMS Later the bacilli sink to the bottom of the flat depression, leaving a clea? supernatant fluid of peptonized gelatin. In hroth, growth takes place rapidly, but does not lead to an even, general clouding. There is usually an initial pellicle formation at the top where the oxygen supply is greatest. Simultaneously with this a slimy mass appears at the bottom of the tube, owing to the sinking of Fig. 124. — Anthrax Colony on Agar. bacilli to the bottom. Apart from isolated flakes and threads the inter- vening broth is clear. Shaken up, the tube shows a tough, stringy mass, not unlike a small cotton fluff, and general clouding is produced only by vigorous mixing. Upon a^ar plates, growth at 37.5° C. is vigorous and colonies appear BACILLUS ANTHRACIS AND ANTHRAX 569 within twelve to twenty-four hours. They are irregular in outline, slightly wrinkled, and show under the microscope the characteristic tangled-thread appearance seen on gelatin, except that they are more compact than upon the former medium. The colonies are slightly glisten- ing and tough in consistency. On agar slants, the colonies usually become confluent, the entire surface soon being covered by a grayish, tough pellicle which, if fished, has a tendency to come away in thin strips or strands. On potato, growth is rapid, white, and rather dry. Sporulation upon potato is rapid and marked, and the medium is favorable for the study of this phase of development. Milk is slowly acidified and slowly coagulated. This action is chiefly upon the casein ; very few, if any, changes being produced either in the sugars or in the fats of the milk. The acids formed are, according to Iwanow,^ chiefly formic, acetic, and caproic acids. Biological Considerations. — The anthrax bacillus is aerobic and facul- tatively anaerobic. It is non-motile and possesses no flagella. In the animal body it occasionally forms capsules. In artificial cultures in the presence of oxygen, it sooner or later invariably forms spores. The spores appear after the culture has reached its maximum of develop- ment. Sporulation never occurs in the animal body, probably because of the absence of sufficient free oxygen. Spores are formed most exten- ^vely ^ at temperatures ranging from 20° C. to 30° C. Spore formation ceases below 18° C. and above 42° C. For different strains these figures may vary slightly, as has been shown from the results of variouS observers. Spores appear most rapidly and regularly upon agar and potato media. The spore — one in each bacillus — appears as a small, highly refractile spot in the center of the individual bacterium. As this enlarges, the body of the bacillus around it gradually undergoes granular degenera- tion and loses its staining capacity.^ If anthrax bacilli are cultivated for prolonged periods upon media containing hydrochloric or rosolic acid or weak solutions of carbolic acid,* cultures may be obtained which do not sporulate and which seem permanently to have lost this power, without losing their virulence to the same degree. Similar results may be obtained by continuous cul- 1 Iwanow, Ann. de I'inst. Pasteur, 1892. 2 Koch, loc. cit. 3 Behring, Zeit. f. Hyg., vi and vii, 1889; Deut. med. Woch., 1889. * Chamberland et Roux, Comptes rend, de I'acad. des sci., xcvi, 1882. 570 PATHOGENIC MICROORGANISMS tivation at temperatures above 42° C. By this procedure, however, virulence, too, is considerably diminished. Resistance. — Because of its property of spore formation, the anthrax bacillus is extremely resistant toward Chemical and physical environ- ment. The vegetative forms themselves are not more resistant than most other npn-sporulating bacteria, being destroyed by a temperature of 54° C. in ten minutes. Anthrax spores may be kept in a dry state for many years without losing their viability.^ While different strains of .anthrax spores show some variation in their powers of resistance, all races show an extremely high resistance to heat. Dry heat at 140° C. kills them only after three hours.^ Live steam at 100° kills them in five to ten minutes. Boiling in water destroys them in about ten min- utes. Low temperatures have but little effect upon them. Ravenel ^ found that, frozen by liquid air, they were still viable after three hours. The variability shown by different strains of spores in their resistance to heat is even more marked in their behavior toward chemicals.^ Some strains will retain their viability after exposure to five-per-cent carbolic acid for forty days,^ while others are destroyed by the same solution in two days. Corrosive sublimate, 1 : 2,000, kills most strains of anthrax in forty minutes. Direct sunlight destroys anthrax spores within six to twelve hours.^ Pathogenicity. — ^The anthrax bacillus is pathogenic for cattle, sheep, guinea-pigs, rabbits, rats, and mice. The degrees of susceptibil- ity of these animals differ greatly, variations in this respect existing even among different members of the same species. Thus, the long-haired Algerian sheep show a high resistance, while the European variety are highly susceptible; and, similarly, the gray rat is much more resistant than the white rat. Dogs, hogs, cats, birds, and the cold-blooded ani- mals are relatively insusceptible. For man the bacillus is definitely pathogenic, though less, so than for some of the animals mentioned above. While separate races of anthrax bacilli may vary much in their de- gree of virulence, a single individual strain remains fairly constant in this respect if preserved, dried upon threads or kept in sealed tubes, in 1 Surmont et Arnould, Ann. de Tinst. Pasteur, 1894. 2 Koch und Wolff hiigel, Mitt. a. d. kais. Gesundheitsamt, 1881. » Ravenel, Medical News, vii, 1899. * Frankel, Zeit. f. Hyg., vi, 1889. ' Koch, loc. cit. • Momont, Ann. de Finst. Pasteur, 1892. BACILLUS ANTHRACIS AND ANTHRAX 571 a cold, dark place. Virulence may be reduced ^ by various attenuating laboratory procedures which are of importance in that they have made possible prophylactic immunization. Heating the bacilli to 55° C. for ten minutes considerably reduces their virulence. Similar results are obtained by prolonged cultivation at temperatures of 42° to 43° C, or by the addition of weak disinfectants to the culture fluids.^ Once reduced, the new grade of virulence remains fairly constant. Increase of virulence may be artificially produced by passage through animals. Experimental infections in susceptible animals are most easily accom- plished by subcutaneous inoculations. The inoculation is followed, at first, by no morbid symptoms, and some animals may appear perfectly well and comfortable until within a few hours or even moments before death, when they suddenly become visibly very ill, rapidly go into collapse, and die. The length of the disease depends to some extent, of course, upon the resistance of the infected subject, being in guinea- pigs and mice from twenty-four to forty-eight hours. The quantity of infectious material introduced, on the other hand, has little bearing upon the final outcome, a few bacilli, or even a single bacillus, often sufficing to bring about a fatal infection. Although the bacilli are not demonstrable in the blood until just before death, they nevertheless invade the blood and lynfph streams immediately after inoculation, and are conveyed by these to all the organs. This has been demonstrated clearly by experiments where inoculations into the tail or ear were im- mediately followed by amputation of the inoculated parts without pre- vention of the fatal general infection. The bacilli are probably not at first able to multiply in the blood. At the place of inoculation and probably in the organs they proliferate, until the resistance of the in-, fected subject is entirely overcome. At this stage of the disease, no longer held at bay by any antagonistic qualities of the blood, they enter the circulation and multiply within it. Autopsy upon such animals reveals an edematous hemorrhagic infiltration at the point of inocu- lation. The spleen is enlarged and congested. The kidneys are con- gested, and there may be hemorrhagic spots upon the serous mem- branes. The bacilli are found in large numbers in the blood and in the capillaries of all the organs. The mode of action of Bacillus anthracis is as yet an unsettled point. It is probable that death is brought about to a large extent by purely ' Toussaint, Comptes rend, de Tacad. des sci., xci, 1880; Pasteur, Chamberland et Roux, Comptes rend, de I'acad. des sci., xcii, 1881. 2 Chamberland et Roux, ibid., xevi, 188?, 572 PATHOGENIC MICROORGANISMS mechanical means, such as capillary obstruction. Neither a true secretory toxin nor an endotoxin has been demonstrated for the anthrax bacillus. The decidedly toxemic clinical picture of the disease, however, in some animals and in man, precludes our definitely concluding that such poisons do not exist. It is a matter of fact, however, that neither culture filtrates nor dead bacilli have any noticeable toxic effect upon test animals, and exert no appreciable immunizing action. Spontaneous infection of animals takes place largely by way of the alimentary canal, the bacilli being taken in with the food. The bacteria are swallowed as spores, and therefore resist the acid gastric juice. In the intestines they develop into the vegetative forms, increase, and gradually invade the system. The large majority of cattle infections are of this type. Direct subcutaneous infection may also occur sponta- neously when small punctures and abrasions about the mouth are made by the sharp spicules of the hay, straw, or other varieties of fodder. When infection upon a visible part occurs, there is formed a diffuse, tense local swelling, not unlike a large carbuncle. The center of this may be marked by a black, necrotic slough, or may contain a pustular de- pression. Infection by inhalation is probably rare among animals. Trans- mission among animals is usually by the agency of the excreta or un- bumed carcasses of infected animals. The bacilli escaping from the body are deposited upon the earth together with animal and vegetable matter, which forms a suitable medium for sporulation. The spores may then remain in the immediate vicinity, or may be scattered by rain and wind dver considerable areas. The danger from buried car- casses, at first suspected by Pasteur, is probably very slight, owing to the fact that the bacilli can not sporulate in the anaerobic environment to which the burying-process subjects them. The disease, in infected cattle and sheep, is usually acute, killing within one or two days. The mortality is extremely high, fluctuating about eighty per cent. In man the disease is usually acquired by cutaneous inoculation. It may also occur by inhalation and through the alimentary trabt. Cutaneous inoculation occurs usually through small abrasions or scratches upon the skin in men who habitually handle live-stock, and in butchers, or tanners of hides. Infection occurs most frequently upon the hands and forearms. The primary lesion, often spoken of as " malig- nant pustule," appears within twelve to twenty-four hours after inocula- tion, and resembles, at first, an ordinary small furuncle. Soon, however, its center will show a vesicle filled with sero-sanguineous, later sero BACILLUS ANTHRACIS AND ANTHRAX 573 purulent fluid. This may change into a black central necrosis sur- rounded by an angry red edematous areola. Occasionally local gangrene and general systemic infection may lead to death within five or six days. More frequently, however, especially if prompt excision is practiced, the patient recovers. The early diagnosis of the condition is best made bacteriologically by finding the bacilli in the local discharge. The pulmonarj^ infection, known as "wool-sorter's disease," occurs in persons who handle raw wool, hides, or horse hair, by the inhala- tion or by the swallowing of spores. The disease is fortunately rare in this country. The spores, once inhaled, develop into the vegetative forms ^ and . these travel along the lymphatics into the lungs and pleura. The disease manifests itself as a violent, irregular pneumonia, which, in the majority of cases, leads to death. The bacilli in these cases can often be found in the sputum before death. Infection through the alimentary canal may occasionally, though rarely, occur in man, the source of infection being usually ingestion of the uncooked meat of infected animals. This form of infection is rare, because in many cases the bacilli have not sporulated in the animal and the ingested vegetative forms are injured or destroyed by the acid gastric juice. When viable spores enter the gut, however, infection may take place, the initial lesion being localized usually in the small intes- tine. The clinical picture that follows is one of violent enteritis with bloody stools and great prostration. Death is the rule. The diagnosis is made by the discovery of the bacilli in the feces. General hygienic prophylaxis against anthrax consists chiefly in the destruction of infected animals, in the burying of cadavers, and in the disinfection of stables, etc. The practical impossibility of destroying, the anthrax spores in infected pastures, etc., makes it necessary to re- sort to prophylactic immunization of cattle and sheep. Immunity against Anthrax. — Minute quantities of virulent anthrax cultures usually suffice to produce death in susceptible animals. Dead cultures are inefficient in calling forth any immunity in treated subjects. It is necessary, therefore, for the production of active immunity to resort to attenuated cultures. The safest way to accomplish such at- tenuation is the one originated by Pasteur,^ consisting in prolonged cultivation of the bacillus at 42° to 43° C. in broth. Non-spore-forming races are thus evolved. The longer the bacilli are grown at the above temperature the greater I Epjringer, Wieri. med. Woch., 1888. 2 Pasteur, loc. dt. 574 PATHOGENIC MICROORGANISMS is the reduction in their virulence. Koch, Gaffky, and Loeffler/ utilizing the variations in susceptibilities of different species of animals, devised a method by means of which the relative attenuation of a given culture may be estimated and standardized. Rabbits are less susceptible than guinea- pigs, and virulent anthrax cultures, grown for two or three days under the stated conditions, lose their power to kill rabbits, but are less virulent for guinea-pigs. After ten to twenty days of further cultivation at 42° C. the virulence for the guinea-pig disappears, but the culture is potent against the still more susceptible mouse. Even the virulence for mice may be entirely eliminated by further cultivation at this temperature. The method of active immunization first practiced by Pasteur, and still used extensively, is carried out as follows: Two ianthrax cultures of varjnng degrees of attenuation are used as vaccins. The premier vaccin is a culture which has lost its virulence for guinea-pigs and rabbits, and is potent only against mice. The deuxieme vaccin is a cul- ture which is still definitely virulent for mice and guinea-pigs, but not potent for rabbits. Forty-eight-hour broth cultures of these strains, grown at 37.5° C, form the vaccin actually employed. Vaccin I is subcutaneously injected into cattle in doses of 0.25 c.c, sheep receiving about half this quantity. After twelve days have elapsed similar quan- tities of Vaccin II are injected. Pasteur's method has given excellent results and confei-s an im- munity which lasts about a year. Chauveau ^ has modified Pasteur's method by growing the bacilli in bouillon at 38° to 39° C, at a pressure of eight atmospheres. Cul- tures are then made of races attenuated in this way, upon chicken bouillon and allow^ed to develop for thirty days. Single injections of 0.1 c.c. each of such cultures are said to protect cattle. Active immunization of small laboratory animals is very difficult, but can be accomplished by careful treatment with extremely attenu- ated cultures. Passive immunization by means of the serum of actively immune animals was first successfully accomplished by Sclavo.^ The subject of passive immunization has been especially investigated and practically applied by Sobernheim.^ The serum used is produced by actively immunizing sheep. It is necessary to carry immunization to an 1 Koch, Gaffky, und Loeffler, Mitt. a. d. kais. Gesundheitsamt, 1884. 2 Chauveau, Comptes rend, de I'acado des sci., 1884. 3 Sclavo, Cent. f. Bakt., xviii, 1895. « Sobemheim, Zeit. f. Hyg., xxv, 1897; xxxi, 1899. BACILLUS ANTHRACIS AND ANTHRAX 575 extremely high degree in order to obtain any appreciable protective power in the serum. This is accomplished by preliminary treatment with Pasteur's or other attenuated vaccines, followed by gradually increasing doses of fully virulent cultures. Treatment continued at intervals of two weeks, for two or three months, usually produces an effective serum. Horses and cattle may also be used for the process, but they are believed by Sobernheim to give less active sera than sheep. Bleeding is done about three weeks after the last injection. The sera are stable and easily preserved. Injections of 20 to 25 c.c. of such a serum have been found to protect animals effectually from anthrax and to confer an immunity lasting often as long as two months. Animals already infected are said to be saved by treatment with 25 to 100 c.c. of the serum. Neither specific bactericidal nor bacteriolytic properties have, so far, been demonstrated in these immune sera. In fact, these properties are distinctly more pronounced against Bacillus anthracis in the normal sera of rats and dogs. Agglutinins have not been satisfactorily demon- strated in sera, partly because of the great technical difficulties en- countered in the active chain-formation of the bacillus in fluid media. An increase of opsonic power of such serum over normal serum has not been satisfactorily demonstrated. Bacteria Closely Resembling Bacillus anthracis. — In most laboratory collections there are strains of true anthrax bacilli so attenuated that they are practically non-pathogenic. These do not differ from the virulent strains in any morphological or cultural characteristics. Besides such strains there are numerous non-virulent bacteria culturally not identical with Bacillus anthracis, but resembling it very closely. B. ANTHRACOiDEs {Hueppe and Wood ^) . — A Gram-positive bacillus, morphologically different from B. anthracis in that the ends are more rounded. Culturally, somewhat more rapid in growth and more rapid in gelatin fluidification. Non-pathogenic. Otherwise indistinguishable from B. anthracis. B. RADicosus (Wurzel Bacillus) . — Cultivated from water — city water supplies. Morphologically somewhat larger than Bacillus anthracis, and the individual bacilli more irregular in size. Very rapid fluidification pf gelatin and growth most active at room temperature. Non-pathogenic. B. suBTiLis {Hay Bacillus). — Although not very closely related tc the anthrax group, this bacillus is somewhat similar and conveniently » Hueppe und Wood, Berl. klin. Woch., xvi, 1889. 576 PATHOGENIC MICROORGANISMS described in this connection. It is of importance to workers with patho- genic bacteria, because of the frequency with which it is found as a saprophyte or secondary invader in chronic suppurative lesions. Morphology. — Straight rod, 2 to 8 micra long, 0,7 micron wide. Spores formed usually slightly nearer one pole than the other. Grows in long chains and only in such chains are spores found. It does not decolor- ize by Gram's method. Is actively motile in young cultures in which Fig. 125. — Bacillus Subtilis. (Hay Bacillus.) the bacilli are single or in pairs. In older cultures chains are formed and the bacilli become motionless. Gelatin is liquefied. On gelatin and agar the bacilli grow as a dry corrugated pellicle. Microscopically, the colonies are made up of interlacing threads, being irregularly round with fringed edges. There is a tendency to confluence. The bacillus is found in brackish water, infusions of vegetable matter, etc., and is practically non-pathogenic, occurring only occasionally as a saprophyte in old sinuses and infected wounds. CHAPTER XLI BACILLUS PYOCYANEUS It is a matter of common surgical experience that many suppurating wounds, especially sinuses of long standing, discharge pus which is of a bright green color. The fact that this peculiar type of purulent inflam- mation is due to a specific chromogenic microorganism was first demon- strated by Gessard ^ in 1882. The bacillus which was described by Ges- sard has since become the subject of much careful research and has been shown to hold a not unimportant place among pathogenic bacteria.^ Morphology and Staining. — Bacillus pyocyanieus is a short rod, usu- ally straight, occasionally slightly curved, measuring, according to Fliigge, about 1 to 2 micra in length by about 0.3 of a micron in thickness. The bacilli are thus small and slender, but are subject to considerable variation from the measurements given, even in one and the same cul- ture. While ordinarily single, the bacilli may be arranged end to end in short chains of two and three. Longer chains may exceptionally be foTmed upon media which are especially unfavorable for its growth, such as very acid media or those containing antiseptics. Spores are not found. The bacilli are actively motile and possess each a single flagellum placed at one end. Bacillus pyocyaneus is stained easily with all the usual dyes, but is decolorized by Gram's method. Irregular staining of the bacillary body is common, but is always an indication of degeneration, and not a normal characteristic, as, for instance, in the diphtheria group. Cultivation. — The pyocyaneus bacillus is aerobic and facultatively anaerobic. It can be adapted to absolutely anaerobic environments, but does not produce its characteristic pigment without the free access of oxygen. The bacillus grows readily upon the usual laboratory media and is not very sensitive to reaction, growing equally well upon moder- ately alkahne or acid media. Development takes place at temperatures as low as 18° to 20° C, more rapidly and luxuriantly at 37.5° C. » Gessard, These de Paris, 1882. ^Charrin, " La maladie pyocyanique, " Paris, 1889. 677 578 PATHOGENIC MICROORGANISMS On agar slants, growth is abundant and confluent, the surface of the agar being covered by a moist, grayish or yellowish, glistening, even layer. The pigment which begins to become visible after about eighteen hours soon penetrates the agar itself and becomes diffused throughout it, giving the medium a bright green fluorescent appearance, which grows darker as the age of the culture increases. In gelatin stabs, growth takes place much more rapidly upon the surface than in the depths. A rapid liquefaction of the gelatin takes place, causing a saucer-shaped depression. As this deepens, pigment begins to form in the upper layers, often visible as a greenish pellicle. In gelatin plates, the colonies have a characteristic appearance. They are round and are composed of a central dense zone, and a peripheral, loosely granular zone, which extends outward into the peripheral fluidi- fied area in a fringe of fine filaments. When first appearing, they are grayish yellow, later assuming the characteristic greenish hue. In broth, growth is rapid and chiefly at the surface, forming a thick pellicle. Below this, there is moderate clouding. The pigment is formed chiefly at the top. In old cultures there is a heavy flocculent precipitate. In fluid media containing albuminous material, strong alkalinity is produced. On potato, growth develops readily and a deep brownish pigment ap- pears, which is not unlike that produced by B. mallei upon the same medium. Milk is coagulated by precipitation of casein and assumes a yellowish- green hue. In older cultures the casein may again be digested and liquefied The pigment of Bacillus pyocyaneus has been the subject of much investigation. It was shown by Charrin ^ and others that this pigment had no relation to the pathogenic properties of the bacillus. It is found in cultures as a colorless leukobase which assumes a green color on the addition of oxygen. Conversely, the typical green "pyocyanin," as the pigment is caUed, may be decolorized by reducing substances. This explains the fact that it is not found in cultures sealed from the air. Pyo-. cyanin may be extracted from cultures with chloroform and crystallized out of such solution in the form of blue stellate crystals. These, on chemical analysis, have been found to belong to the group of aromatic compounds, with a formula, according to Ledderhose,^ of Ci4Hi4N20. Besides pyocyanin. Bacillus pyocyaneus produces another pig- » Charrin, loc. cit. « Ledderhose, quoted from Boland, Ceiio. f. Bakt., xjcv, 1889. BACILLUS PYCXIYANEUS 579 ment which is fluorescent and insoluble in chloroform, but soluble in water. ^ This pigment is common to other fluorescent bacteria, and not peculiar to Bacillus pyocyaneus. The reddish-brown color seen in old cultures^ and supposed by some writers to be a third pigment, is probably a derivative from pyocyanin by chemical change. Chloroform extraction of pyocyanin from cultures may serve oc- casionally to distinguish the pyocyaneus bacilli from other similar fluorescent bacteria. Ernst has claimed that there are two types of B. pyocyaneus, an a-type which produces only the fluorescent, water- soluble pigment, and a /5-type which produces both this and pyocyanin.^ Pathogenicity. — Bacillus pyocyaneus is one of the less virulent patho- genic bacteria. It is widely distributed in nature and may be found frequently as a harmless parasite upon the skin or in the upper respira- tory tracts of animals and men. It has, however, occasionally been found in connection with suppurative lesions of various parts of the body, often as a mere secondary invader in the wake of another incitant, or even as the primary cause of the inflammation. In most cases where true pyocyaneus infection has taken place, the subject is usually one whose general condition and resistance are abnormally low."* Thus pyocyaneus may be the cause of chronic otitis media in ill-nour- ished children. It has been cultivated out of the stools of children suf- fering from diarrhea, and has been found at autopsy generally distributed throughout the organs of children dead of gastro-enteritis.^ It has been cultivated from the spleen at autopsy from a case of general sepsis following mastoid operation. The bacillus has been found, further- more, during life in pericardial exudate and in pus from liver abscesses.® Brill and Libman,^ as well as Finkelstein,^ have cultivated B. pyocyaneus from the blood of patients suffering from general sepsis. Wassermann ^ showed the bacillus to have been the etiological factor in an epidemic of umbilical infections in new-born children. Similar exam- ples of B. pyocyaneus infection in human beings might be enumerated in large numbers, and there is no good reason to doubt that, under given • Boland, loc. cit. ^Gessard, Ann. de Tinst. Pasteur, 1890, 1891, and 1892. « Ernst, Zeit. f. Hyg., ii, 1887. » Rohner, Cent. f. Bakt., xi, 1892. • Neumann, Jahrb. f. Kinderheilk,, 1890. • Kraunhals, Zeit. f. Chir., xxxvii, 1893. » Brill and Lihman, Amer. Jour. Med. Sci., 1899. • Finkelstein, Cent. f. Bakt., 1899. f • Wassermann, Virchow's Arch., clxv, 1901. 580 PATHOGENIC MICROORGANISMS conditions, fatal infections may occur. Such cases, however, are still to be regarded as depending more upon the low resistance of the individual attacked than upon the great pathogenicity of B. pyocyaneus. Many domestic animals are susceptible to experimental pyocyaneus infection, chief among these being rabbits, goats, mice, and guinea- pigs. Guinea-pigs are killed by this bacillus with especial ease. Intra- peritoneal inoculation with a loopful of a culture of average virulence usually leads to the death of a young guinea-pig within three or four days. Toxins and Immunization. — Emmerich and Low have shown that filtrates of old broth cultures of B. pyocyaneus contain a ferment-Hke substance which possesses the power to destroy some other bacteria, apparently by lysis. They have called this substance " pyocyanase " and claim that, with it, they have succeeded in protecting animals from anthrax infection. During recent years pyocyanase has been employed locally for the removal of diphtheria bacilli from the throats of convales- cent cases. Broth-culture filtrates evaporated to one-tenth their volume in vacuo are used for this purpose. Pyocyanase is exceedingly thermostable, resisting boiling for several hours, and is probably not identical with any of the other toxins or peptonizing ferments produced by B. pyocyaneus. The toxins proper of B. pyocyaneus have been the subject of much investigation, chiefly by Wassermann.^ Wassermann found that filtrates of old cultures were far more poisonous for guinea-pigs than extracts made of dead bacteria. He concludes from this and other observations that B. pyocyaneus produces both an endotoxin and a soluble secreted toxin. The toxin is comparatively thermostable, resisting 100° C. for a short time. Animals actively immunized with living cultures of B. pyo- cyaneus give rise in their blood serum to bacteriolytic antibodies only. Immunized with filtrates from old cultures, on the other hand, their serum will contain both bacteriolytic and antitoxic substances. The true toxin of B. pyocyaneus never approaches in strength that of diph- theria or of tetanus. Active immunization of animals must be done carefully if it is desired to produce an immune serum, since repeated injections cause great emaciation and general loss of strength. Specific agglutinins have been found in immune sera by Wassermann^ and others. Eisenberg ^ claims that such agglutinins are active also against some of the fluorescent intestinal bacteria. 1 Wassermann, Zeit. f. Hyg., xxii, 1896. ^ Wassermann, Zeit. f. Hyg., 1902. ^ Eisenberg, Cent. f. Bakt., 1903. BACILLUS PYOCYANEUS 581 Bulloch and Hunter * have recently been able to show that old broth cultures of B. pyocyaneus contain a substance capable of hemolyzing the red blood corpuscles of dogs, rabbits, and sheep. This " pyocyanolysin " seems intimately attached to the bacterial body. Prolonged heating of cultures does not destroy it. Heating of hemolytic filtrates, however, destroys it in a short time. The filtration of young cultures yields very little pyocyanolysin in the filtrate. In old cultures, however, a considerable amount passes into the filtrate. Whether or not the hemolytic power is due to a specific bacterial product or is dependent upon changes in the culture fluid, such as alkalinization, etc., can not yet be regarded as certain. Gheorghiewski ^ claims to have found a leucocyte-destroying ferment in pyocyaneus cultures. > Bulloch imd Hunter, Cent. f. Bakt., xxviii, 1900. * Gheorghiewski, Ann. de Tinst. Pasteur, xiii, 1899. CHAPTER XLII J^IATIC CHOLERA AND THE CHOLERA ORGANISM (Spirillum cholerce asiaticce, Comrna Bacillus) The organism of Asiatic cholera was unknown until 1883. In this year, Koch/ at the head of a commission established by the German government to study the disease in Egypt and India, discovered the " comma bacillus " in the defecations of patients, and satisfactorily de- termined its etiological significance. Koch's investigations were carried out on a large number of cases and many investigations have since then corroborated his results. The numerous morphologically similar spirilla which were later found in normal individuals and in connection with other conditions, have been shown by accurate bacteriological methods to be closely related, but not identical. Apart from the" evidence of the constant association of the cholera vibrio with the disease, the etiological relationship has been clearly demonstrated by several accurately recorded accidental infections oc- curring in bacteriological workers, and by the famous experiment of Pettenkofer and Emmerich, who purposely drank water containing cholera bacilli. Both observers became seriously ill with typical clini- cal symptoms of cholera, and one of them narrowly escaped death. Morphology and Staining. — The vibrio or spirillum of cholera is a small curved rod, varying from one to two micra in length. The degree of curvature may vary from the slightly bent, comma-like form to a' more or less distinct spiral with one or twt) turns. The spirals do not lie in the same plane, being arranged in corkscrew fashion in three dimensions. The spirillum is actively motile and owes its motility to a single polar flagellum, best demonstrated by Van Ermengem's flagella stain. Spores are not found. In young cultures the comma shapes predominate, in older growths the longer forms are more nu- merous. Strains which have been cultivated artificially for prolonged » Koch, Deut. med. Woch., 1883 and 1884. 582 ASIATIC CHOLERA AND THE CHOLERA ORGANISM 583 periods without passage through the animal body have a tendency to lose the curve, assuming a more bacillus-Hke appearance. The spirilla are stained with all the usual aqueous anilin dyes. They are decolor- ized by Gramas method. In histological section they are less easily stained, but may be demonstrated by staining with alkaline methylene blue. Cultivation. — The cholera spirillum grows easily upon all the usual culture media, thriving upon meat-extract as well as upon meat-infusion FiG. 126.— Cholera Spirillum. (After Frankel and Pfeiffer.) media. Moderate alkalinity of the media is preferable, though slight acidity does not prevent growth. In gelatin plates growth appears at room temperature within, twenty- four hours as small, strongly refracting yellowish-gray, pin-head colonies. As growth increases the gelatin is fluidified. Under magnification these colonies appear coarsely granular with margins irregular because of the liquefaction. Liquefaction, too, causes a rapid development in such colonies of separate concentric zones of varying refractive power. Old strains, artificially cultivated for long periods, lose much of their liquefying power. In gelatin stab cultures fluidification begins at the surface, rapidly giving^ rise to the familiar funnel-shaped excavation. Upon agar plates, within eighteen to twenty-four hours, grayish, opalescent colonies appear, which are easUy differentiated by their 584 PATHOGENIC MICROORGANISMS transparency from the other bacteria apt to appear in feces. Agar plates, therefore, are important in the isolation of these organisms. Coagulated blood serum is fluidified by the cholera vibrio. On potato, growth is profuse and appears as a brownish coarse layer. In milk, growth is rapid and without coagulation. In broth, general clouding and the formation of a pellicle result. The rapidity and luxuri- ance of growth of the cholera spirillum upon alkaline pepton solutions render such solutions peculiarly useful as enriching media in isolating this microorganism from the stools of patients. In pepton solution, too, the cholera spirillum gives rise to abundant indol, demonstrated in the so-called "cholera-red" reaction. This reaction has a distinct diagnostic value, but is by no means specific.^ In the case of the cholera vibrio the mere addition of strong sulphuric acid suffices to bring out the color reaction. This is due to the fact that, unlike some other indol- producing bacteria, the cholera organism is able to reduce the nitrates present in the medium to nitrites, thus itself furnishing the nitrite necessary for the color reaction. The medium which is most suitable for this test is that proposed by Dunham,^ consisting of a solution of 1 per cent of pure pepton and .5 per cent NaCl in water. Dieudonn^^ has recommended a selective medium upon which cholera spirilla will grow well, but upon which the colon bacillus will grow either very sparsely or not at all. Its preparation is very simple. To 70 parts of ordinary 3 per cent agar, neutralized to litmus, there are added 30 parts of a sterile mixture of defibrinated beef blood and jiormal sodium hydrate. The latter is sterilized by steam before being added to the agar. This pure alkali agar is poured out in plates and allowed to dry several days at 37° or 5 minutes at 60°. The material to be examined is smeared upon the surface of these plates with a glass rod. The principle of this medium is that cholera will grow in the presence of an amount of alkali which inhibits other fecal bacteria. The medium has been studied by Krimiwiede, Pratt, and Grund,* who have recom- mended a modification. They find the following combination sat- isfactory and an improvement upon Dieudonne's medium because transparent and more easily prepared. They prepare the following mixtures : * See indol reaction, p. 167. ^ Dunham, Zeit. f. Hyg., ii, 188;?. ^ Dieudonni, A., Cent. Bakt.j 1., orig., 1909. * Krumwiede, Pratt, and Grund, Jour, of Inf. Dis., x, 1912, ASIATIC CHOLERA AND THE CHOLERA ORGANISM 585 Egg-White Medium. A. White of egg and water a.a. Sodium carbonate cryst. 12 per cent. Mix in equal parts, steam in Arnold sterilizer for 20 minutes. B. Meat pepton 3 per cent agar, neutral to litmus. 30 parts of A are added to 70 parts of B. Another modification recommended by them is as follows: Whole-Egg Medium. A. Whole egg and water a.a. Sodium carbonate 12 to 13.5 per cent. Mix in equal parts, steam for 20 minutes. B. Meat free agar, viz., pepton, salt, and 3 per cent agar. 30 parts of A are mixed with 70 parts of B while the agai* is boiling hot as above. The medium is poured on the plates in a thick layer and allowed to stand open for 20 to 30 minutes and then the inoculation is carried out by surface streaking. Isolation. — Isolation of the cholera vibrio from the feces, while easy in many cases, is occasionally attended with some difficulty owing to the large number of other bacteria present. The most satisfactory method of procedure is to inoculate a set of gelatin plates, another of agar plates, and a number of Dunham's pepton- broth tubes, with small quantities of the suspicious material. When the spirilla are numerous they can frequently be fished directly from sus- picious colonies in the plates and isolated for further identification. When less numerous, they can usually be found in relatively increased numbers after eight or ten hours at 37.5° C, in the topmost layers of the Dunham broth, which is an almost selectively favorable medium for these organisms. They collect at the surface where free oxygen is readily obtained. From the pepton broth, plate dilutions can then be prepared and colonies fished.^ Once isolated, the spirilla are identified by their morphology, by the appearance of their colonies, by their manner of growth upon gelatin stabs, by the cholera-red reaction, and, finally, by agglutinative and bacteriolytic tests in immune sera. Owing to the existence of other spirilla morphologically and cultu- rally similar, the serimi reactions are the only absolutely positive dif- ferential criteria. 1 Abel und Claussen, Cent, f . Bakt., xvii, 1895. 586 PATHOGENIC MICROORGANISMS For isolation of the bacteria from water, it is, of course, necessary to use comparatively large quantities. Fliigge ^ and Bitter advise the distribution of about a Uter of water in ten or twelve Erlenmeyer flasks. To each of these they add 10 c.c. of sterile pepton-salt solution (pepton ten per cent, NaCl five per cent). After eighteen hours at 37.5° C. the surface growths in these flasks are examined both microscopically and culturally as before. Biological Considerations. — The cholera spirillum is aerobic and facultatively anaerobic. It does not form spores. The optimum tem- perature for its growth is about 37.5° C. It grows easily, however, at a temperature of 22° C. and does not cease to grow at temperatures as high as 40°. Frozen in ice, these bacteria may live for about three or four days. Roil ins; destroys them immediately. A temperature of h I Fig. 127. Fig. 128. Fig. 127.^ — Cholera Spirillum. Stab Culture in Gelatin, three days old. Fig. 128. — Cholera Spirillum. Stab Culture in Gelatin, six days old. (After Frankel and Pfeiffer.) 60° C. kills them in an hour. In impure water, in moist linen, and in food stuffs, they may live for many days. Associated with sapro- phytes in feces and other putrefying material, and wherever active acid formation is taking place, they are destroyed within several days. Complete drying kills them in a short time. The common disin- fectants destroy them in weak solutions and after short exposures (carbolic acid, five-tenths per cent in one-half hour; bichlorid of mercury, 1 : 100,000 in ten minutes; mineral acids, 1 : 5,000 or 10,000 in a few minutes) .2 Pathogenicity. — Cholera is essentially a disease of man. Endemic in India and other Eastern countries, it has from time to time epidemically invaded large territories of Europe and Asia, not infrequently assuming -^ _ 1 /?'%^e, Zeit. f. Hyg., xiv, 1893. 2 Forster, Hyg. Rundschau, 1893. ASIATIC CHOLERA AND THE CHOLERA ORGANISM 587 pandemic proportions and sweeping over almost the entire earth.^ Five separate cholera epidemics of appalling magnitude occurred during the nineteenth century alone; several of these, spreading from India to Asia Minor, Egypt, Russia, and the countries of Central Europe, reached even to North and South America. The last great epidemic began about 1883, traveled gradually westward, and in 1892 reached Germany where it appeared with especial virulence in Hamburg, and thence, fol- lowing the highways of ocean commerce, entered America and Africa. During this epidemic in Russia alone 800,000 people fell victims to the disease. In man the disease is contracted by ingestion of cholera organisms with water, food, or any contaminated material. The disease is essen- tially an intestinal one. The bacteria, very sensitive to an acid reaction, may often, if in small numbers, be checked by the normal gastric secre- tions. Having once passed into the intestine, however, they proUferate rapidly, often completely outgrowing the normal intestinal flora. Fatal cases, at autopsy, show extreme congestion of the intestinal walls. Occasionally ecchymosis and localized necrosis of the mucosa may be present and swelling of the solitary lymph-follicles and Peyer's patches. Microscopically the cholera spirilla may be seen to have penetrated the mucosa and to lie within its deepest layers close to the submucosa. The most marked changes usually take place in the lower half of the small intestine. The intestines are filled with the characteristically fluid, slightly bloody, or "rice-water" stools, from which often pure cultures of the cholera vibrio can be grown. The microorganisms can be cultivated only from the intestines and their contents, and the parenchymatous degenera- tions taking place in other organs must be interpreted as being purely of toxic origin. In animals, cholera never appears as a spontaneous disease. Nikati and Rietsch ^ have succeeded in producing a fatal disease in guinea-pigs by opening the peritoneum and injecting cholera spirilla directly into the duodenum. Koch ^ succeeded in producing a fatal cholera-like disease in animals by introducing infected water into the stomach through a catheter after neutralization of the gastric juice with sodium carbonate. At the same time, he administered opium to prevent active peristalsis. A method of infection more closely analogous to the infec- ^Hirsch, "Handb. d. histor.-geogr. Path.," 1881. 2 Nikati und Rietsch, Deut. med. Woch., 1884; 3 Koch, Deut. med. Woch., 1885. 58§ PATHOGENIC MICROORGANISMS tion in man was followed by Metchnikoff/ who successfully produced fatal disease in young suckling rabbits by contaminating the maternal teat. Subcutaneous inoculation of moderate quantities of cholera spirilla into rabbits and guinea-pigs rarely produces more than a temporary illness. Intraperitoneal inoculation, if in proper quantities, generally leads to death. It will be remembered that when working with intra- peritoneal cholera inoculations the phenomenon of bacteriolysis was discovered by Pfeiffer.^ Different strains of cholera spirilla vary greatly in their virulence. The virulence of most of them, however, can be enhanced by repeated passages through animals. Most of our domestic animals enjoy consid- erable resistance against cholera infection, though under experimental conditions successful inoculations upon dogs, cats, and mice have been reported. Doves are entirely insusceptible.^ Hygienic Considerations.: — The cholera spirillum leaves the body of the infected subject with the defecations only. Infection takes place, so far as we know, only by way of the mouth. From these two facts it follows that the chief source of danger for a community Ues in infection of its water supply. As a matter of fact the bacteria have been fre- quently found in the wells, lakes, rivers, and harbors of afflicted terri- tories, and in several cases it has been possible to define the Hmits of an epidemic almost precisely by the distribution of the contaminated water supply. A classic example of this is that of the Hamburg epi- demic, during which Altona, a town as close to Hamburg as Brooklyn is to New York, with unrestricted interurban traffic but with separate water supply, was almost spared, while Hamburg itself was iftidergoing one of the most virulent epidemics of its history. It has been statistically noted, moreover, chiefly by Koch, that cholera in its spread not infre- quently follows the water courses. Apart from infection through the water supply, cholera may be transmitted directly or indirectly by con- tact with contaminated linen, bedclothes, etc., the organism being con- veyed to the mouth by the fingers, or by infected food. Epidemics due to this mode of infection alone, however, are apt to be more narrowly localized and more sporadic in their manifestations. It is probable that this mode of infection is of great importance in countries where the disease 1 Metchnikoff, Ann. d. I'inst. Pasteur, 1894 and 1896. 2 pfeiffer, loc. cit. ' 3 Pfeiffer und Nocht, Zeit. f . Hyg., vii, 1889. ASIATIC CHOLERA AND THE CHOLERA ORGANISM 589 is endemic, but its significance in producing epidemics is limited owing to the fortunately low resistance of the spirillum to desiccation. The sudden appearance of cholera in a place far distant from the seat of a prevalent epidemic may be explained by the occasional presence of cholera spirilla in the dejecta of convalescents as late as two or three weeks after apparent recovery from the disease and consequent release from quarantine. Cholera Toxin. — ^The absence of the cholera spirilla from the in- ternal organs of fatal cases, in spite of the severe general symptoms of the disease, points distinctly to the existence of a strong poison pro- duced in the intestine by the microorganisms and absorbed by the patient. It was in this sense, indeed, that Koch first interpreted the clinical picture of cholera. Numerous investigations into the nature of these toxins have been made, the earUer ones defective in that definite identification of the cultures used for experimentation were not carried out. Pfeiffer,Vin 1892, was able to show that filtrates of young bouillon cultures of cholera spirilla were but sHghtly toxic, whereas the dead bodies of carefully killed agar cultures were fatal to guinea-pigs even in small quantities. In consequence, he regarded the cholera poison as consisting chiefly of an endotoxin.^ The opinion as to the endotoxic nature of the cholera poison is not, however, shared by all workers. Metchnikoff, Roux, and Salimbeni,^ in 1896, succeeded in producing death in guinea-pigs by introduction into their peritoneal cavities of cholera cultures enclosed in celloidin sacs. Brau and Denier,^ and, more recently, Kraus,^ claim that they have succeeded not only in demonstrating a soluble toxin in alkaline broth cultures of cholera - spirilla, but in producing true antitoxins by immunization with such cultures. It appears, therefore, that the poisonous action of the cholera organisms may depend both upon the formation of true secretory toxins and upon endotoxins. Which of these is paramount in the produc- tion of the disease can not be at present definitely stated. In favor of the great importance of the endotoxic elements is the failure, thus far, to obtain successful therapeutic results with supposedly antitoxic sera. 1 Pfeiffer, Zeit. f. Hyg., xi, 1892. 2 Pfeiffer und Wassermann, Zeit. f. Hyg., xiv, 1893. 3 Metchnikoff, Roux, et Salimheni, Ann. d.e I'inst. Pasteur, 1896. ^ Brau et Denier, Comptes rend, de I'acad. des sci., 1906. 6 R, Kraus, Cent, f . Bakt., 1906. 590 PATHOGENIC MICROORGANISMS Cholera Immunization. — One attack of cholera confers protection against subsequent infection. Active immunization of animals may be accomplished by inoculation of dead cultures, or of small doses of living bacteria. In the serum of immunized animals specific bacterio- lytic and agglutinating substances are found. The discovery of bacte- riolytic immune bodies, in fact, was made by means of cholera spirilla. Both the bacteriolysins and the agglutinins, because of their specificity, are of great importance in making a bacteriological diagnosis of true cholera organisms. Protective inoculation of man has been variously attempted by Ferran ^ and others. Experiments on a large scale were done, more re- cently, by Haffkine,^ who succeeded in producing an apparently dis- tinct prophylactic immunization by the subcutaneous inoculation of dead cholera cultures. Similar immunization with, bacterial filtrates has been attempted by Bertarelli.^ Strong reconunends immunization with filtrates from autolyzed cultures. CHOLERA-LIKE SPIRILLA The biological group of the vibriones, to which the cholera spirillum belongs, is a large one, numbering probably over a hundred separate species. Most of these are of bacteriological importance chiefly because of the difficulties which they add to the task of differentiation, for while some of them simply bear a morphological resemblance to the true cholera vibrio, others can be distinguished only by their serum reac- tions and pathogenicity for various animals. Additional difficulty, too, is contributed by the fact that within the group of true cholera organisms occasional variations in agglutinability and bacteriolytic reactions may exist. Certain strains, too, the six El Tor cultures isolated by Gottschlich, while in every respect similar to true cholera spirilla, are considered as a separate sub-species by Kraus,^ because of their ability to produce hemolytic substances, a function lacking in other cholera strains. Spirillum Metchnikovi. — ^This spirillum was discovered by Gamaleia ^ in the feces and blood of domestic fowl, in which it had caused an in- testinal disease. Morphologically and in staining reactions it is identical 1 Ferran, Comptes rend, de I'acad. des sciences, 1885. 2 Haffkine, BuU. med., 1892. ^ Bertarelli, Deut. med. Woch., 33, 1904, ^Kraus, Kraus and Levaditi; "Handbuch," vol. i, p. 186. ^ Gamaleia, Ann, de Tinst. Pasteur, 1883. ASIATIC CHOLERA AND THE CHOLERA ORGANISM 591 with Spirillum cholerse asiaticse. It possesses a single polar flagellum, and is actively motile. Culturally it is identical with Vibrio cholerae except for slightly more luxuriant growth and more rapid fluidification of gelatin. It gives the cholera-red reaction in pepton media. It is differentiated from the cholera vibrio by its power to produce a rapidly fatal septicemia in pigeons after subcutaneous inoculation of minute quantities.^ It is much more pathogenic for guinea-pigs than the cholera vibrio. It is not subject to lysis or agglutinated by cholera immune sera. Spirillum Massaua. — This organism was isolated at Massaua by Pasquale ^ in 1891 from the feces of a clinically doubtful case of cholera. Culturally and morphologically it is much like the true cholera vibrio, but in pathogenicity is closer to Spirillum Metchnikovi, in that small quantities produce septicemia in birds. It possesses four flagella. It does not give a specific serimi reaction with cholera immune serum. Spirillum of Finkler-Prior.^ — Isolated by Finkler and Prior from the feces of a case of cholera nostras. Morphologically it is like the true cholera spirillum, though slightly larger and less uniformly curved. Culturally it is much like the cholera vibrio, but grows more rapidly and thickly upon the usual media. It does not give the cholera-red reaction, nor does it give specific serum reactions with cholera im- mune serum. • Spirillum Deneke.'* — ^A vibrio isolated by Deneke from butter. Much like that of Finkler-Prior. It does not give the cholera-red reaction. 1 Pfeiffer und Nocht, Zeit. f . Hyg., vii, 1889. 2 Pasquale, Giorn. med. de r. eserc. ed. R. Marina, Roma, 1891. ' Finkler und Pricyr, Erganz. Hefte, Cent. f. allg. ges. Phys., 1884. * Deneke, Deut. med. Woch., iii, 1885. ^ CHAPTER XLIII DISEASES CAUSED BY SPIROCHETES (TREPONEMATA) The microorganisms known as spirochsetes are slender, undulatingj corkscrew-like threads which show definite variations both structurally and culturally from the bacteria as a class. Most important among them are the spirochsete of relapsing fever, Spirochaete pallida of syphilis, the spirillum of Vincent, Spirochsete refringens. Spirillum gallinarum, a microorganism which causes disease in chickens, Spirochaete anserina, which causes a similar condition in geese, and several species which have been found as parasites, both in animal ° and in man, without having definite etiological connection with disease. The classification of these various species in one group is rather more a matter of convenience than one of scientific accuracy, since our knowl- edge of them is not far advanced, and our inability to cultivate almost all of them has not permitted their detailed biological study. Formerly many of these organisms were regarded as bacteria belonging to the gen- eral group of the spirilla. Recently Schaudinn,^ the discoverer of the syphilis spirochaete, has claimed, upon the basis of a careful morphological study, that many of these forms are actually protozoa. He based this claim upon the observation that stained preparations often showed undu- lating membranes extending along the long axis of the microorganisms and that definite nuclear structures were demonstrable. This observer also claimed that many of the spiral forms reproduce by cleavage along the longitudinal axis. Other observers have not agreed with this view, Laveran,^ Novy and Knapp,^ and others assertiifg that their own obser- vations indicate a close relationship of these microorganisms to the true bacteria. Whatever the final conclusion may be, the question is more or less an academic one, in that our ideas as to the exact line of division between the unicellular animals and the unicellular plants is not by any means founded upon a sound basis. In common with the bacteria, most * Schaudinn, Arb. a. d. kais. Gesundheitsamt, 1904. . ^Laveran, Comptes rend, de Tacad. des sci., 1902 and 1903. 3 Novy and Knapp, Jour, of Infec. Dis., 3, 1906. 592 DISEASES CAUSED BY SPIROCILETES 593 of these microorganisms have the power of .multipHcation by transverse fission. They possess flagella and, in the case of some of them at least, definite immune bodies can be demonstrated in the serum of infected subjects similar to those produced by bacteria during infection. The undulating membranes and the definite differentiation between nucleus and cytoplasm claimed for them by some observers have not been uni- formly confirmed, and their similarity to the trypanosomes has not therefore been established. On the other hand, none of these micro- organisms has so far been successfully cultivated upon artificial media, with the exception of the spirilla which occur in Vincent's angina. For some of the diseases caused by this class of parasites, moreover, trans- mission by an intermediate host, in which the spirilla undergo multipli- cation, has been definitely shown, a fact which corresponds with the conditions observed in many protozoan infections. Upon a careful re- view of these various data, it seems to be fully justified, on the basis of our present knowledge, to group these microorganisms, as KoUe and Hetsch 1 have done, in a class between bacteria and protozoa. The terms spirochsete and spirillum have been indiscriminately used. In the original classification of Migula the difference between the two groups was based upon the rigidity of the cell body in the case of the spirilla and the sinuous or flexible nature of the cell in the case of the spirochaetse. Although the term spirillum is still colloquially used for some members of this group, merely because of past usage, it would be better to speak of all the microorganisms here grouped together by the term ''spirochsetes." SYPHILIS AND SPIROCUffiTA PALLIDA {Treponema pallidum) The peculiar manifestations of syphilis, its mode of transmission, and the fact that its primary lesion was always located at the point of contact with a preceding case, have always stamped it as unques- tionably infectious in nature. Until very recently the microorgan- ism which gives rise to syphilis was unknown. Many bacteriologists had attacked the problem and many microorganisms for which defi- nite etiological importance was claimed had been described. Most of these announcements, however, aroused little more than a sensational interest and received no satisfactory confirmation. A bacillus described ^KolU und Hetsch, "Die experimentelle Bakt.," Berlin, 1906. 594 PATHOGENIC MICROORGANISMS by Lustgarten ^ in 1884 seemed, for a time, to have solved the mystery. The Lustgarten bacillus was an acid-fast organism very similar to Bacillus tuberculosis, and found by its discoverer in a large num- ber of syphilitic lesions. The observation, at first, aroused much interest and received some confirmation. Later extensive investigations, how- ever, failed to uphold the etiological relationship of this bacillus to the disease and practically identified it with the smegma bacillus, so often a saprophyte upon the mucous membranes of the normal genitals. In 1905, Schaudinn,2 a German zoologist, working in collaboration with Hoffmann, investigated a number of primary syphilitic indurations and secondarily enlarged lymph nodes, and in both lesions discovered a spirochaete similar to, but easily distinguished from, the spirochaetes Fig. 129. — ^SriROCHiETA pallida. Smear preparation from chancre stained by -the india-ink method. already known. He failed to find similar microorganisms in uninfected human beings. The microorganism described by him as "Spirochseta palUda" is an extremely delicate undulating filament measuring from four to ten micra in length, with an average of seven micra, and varying in thickness from an immeasurable delicacy to about 0.5 of a micron. It is thus distinctly smaller and more delicate than the spirochaete of relapsing fever. Ex- amined in fresh preparations it is distinctly motile, its movements con- sisting in a rotation about the long axis, gliding movements backward and forward, and, occasionally, a bending of the whole body. Its con- volutions, as counted by Schaudinn, vary from 3 to 12 and differ from those observed in many other spirochaetes by being extremely steep, or, in other words, by forming acute, rather than obtuse, angles. The ends of the microorganism are delicately tapering and come to a point. In ^ Lustgarten, Wien. med. Woch., xxxiv, 1884. 2 Schaudinn und Hoffmann, Arb. a. d. kais. Gesundheitsamt, 22, 1905. DISEASES CAUSED BY SPIROCHETES 595 his first investigations, Schaudinn was unable to discover flagella and believed that he saw a marginal undulating membrane similar to that noticed in the trypanosomes. Later observations by this observer, as well as by others, revealed a delicate flagellum at each end, but left the existence of an undulating membrane in doubt. Uncertain, in his later investigations, whether the, microorganisms described by him could scientifically be classified with the spirochsete proper, Schaudinn sug- gested the name of " Treponema pallidum.^' In the same preparations in which Spirochaeta pallida was first seen, other spirochetes were present, which were easily distinguished from the former by their coarser contours, their flatter and fewer undula- tions, their more highly refractile cell bodies, and, in stained prepara- tions, their deeper color. These microorganisms were not found regularly, and were interpreted merely as fortuitous and unimportant companions. To them Schaudinn gave the name of " Spirochseta re- fringens." The epoch-making discovery of Schaudinn and Hoffmann was soon confirmed by many observers, and the etiological relationship of Spiro- chaeta pallida to syphilis may now be regarded as an accepted fact. Although our inability to cultivate the microorganism has made it impossible to carry out Koch's postulates, nevertheless indirect evi- dence of such a convincing nature has accumulated that no reasonable doubt as to its caustive importance can be retained. The spirochaetes have been found constantly present in the primary and secondary lesions of all carefully investigated cases, and, so far, have invariably been absent in subjects not afflicted with syphilis. Schaudinn himself, not long after his original communication, was. able to report seventy cases of primary and secondary syphilis in which these microorganisms were found. Spitzer^ found them constantly present in a large number of similar cases. Sobernheim and Tomas- czewski^ found the spirochaetes in fifty cases of primary and secondary S3rphilis, but failed to find them in eight tertiary cases. Mulzer,^ who found the microorganisms invariably in twenty cases of clinical syphiHs, failed to find them in fifty-six carefully investigated non-syphilitic sub- jects. The voluminous confirmatory literature which has accumulated upon the subject can not here be reviewed. The presence of these spirochaetes in the blood at certain stages of the disease has been demon- 1 S-pitzer, Wien. klin. Woch., 1905. 2 Sobernheim und Tomasczewski, Miinch. med. Woch., 1905. ' MtUzer, Berl. klin. Woch., 1905, and Archiv f . Dermat. u. Syph., 79, 1906. 596 PATHOGENIC MICROORGANISMS strated by Bandi and Simonelli^ who found them in the blood taken from the roseola spots, and by Levaditi and Petresco ^ who found them in the fluid of blisters produced upon the skin. In tertiary lesions the spirochsetes have been found less regularly than in the primary and secondary lesions, but positive evidence of their presence has been brought by Tomasczewsjd,^ Ewing,^ and others who succeeded in demonstrating them in gummata. Noguchi and Moore ^ have recently found the Spirochseta pallida in the brain of patients dead of general paresis. In congenital syphilis, many observers, have found Spirochseta pallida in the lungs, liver, spleen, pancreas, and kidneys, and, in isolated cases, in the heart muscle. The organisms were always present in large numbers and practically in pure culture. These results more than any others seem to furnish positive proof of the etiological relationship be- tween the spirochsete and the disease. Demonstration of Treponema pallidum. — In the living state the spirochsetes have been observed in the hanging drop or under a cover- slip rimmed with vaseline. It is extremely important, in preparing such specimens from primary lesions or from lymph glands, to obtain the material from the deeper tissues, and thus as uncontaminated as possible by the secondary infecting agents present upon the surface of an ulcer, and also as free from blood as possible. An ordinary microscope and condenser may be used, provided that the light is cut down con- siderably by means of the iris diaphragm. This method is, however, difficult and uncertain. It is better to employ a special device known as a "condenser for dark-field illumination'' (Dunkel-Kammer- Beleuchtung). This apparatus is screwed into the place of the Abbe condenser. The preparation is made upon a slide and covered with a cover-slip as usual. A drop of oil is then placed upon the upper sur- face of the dark chamber and the slide laid upon it so that an even layer of oil, without air-bubbles, intervenes between the top of the dark chamber and the bottom of the slide. The preparation is then best examined with a high-power dry lens. An arc light furnishes the most favorable illimoination. In such preparations the highly refractive cell- 1 Bandi und SirrumeUi, Cent, f . Bakt., 40, 1905. * Levaditi et Petresco, Presse m6d., 1905. ' Tomasczewski, Miinch. med. Woch., 1906. * Swing, Proc. N. Y. Path. Soc, N. S., 5, 1905. \ * Noguchi and Moore, Jour. Exp. Med., xvii, 1913.^ DISEASES CAUSED BY SPIROCHiETES 597 bodies stand out against the black background, and the motility of the organisms may be observed.^ The dark-field condenser is without question the easiest method of finding the Spirochseta pallida. Its use is easily learned and the appara- tus is sufficiently cheap so that it lends itself to the use of the clinic and the office. With very little practice it is possible to detect the spiro- chsete in suspension if care is taken that not too much blood or other solid particles are mixed with the preparation. Should it be impossible to obtain the material scraped from syphilitic lesions in a sufficiently dilute condition it is best to emulsify it in a drop or two of human ascitic fluid. Examination in Smears. — The Spirochseta pallida can not be stained with the weaker anilin dyes, and even more powerful dyes, such as carbol-fuchsin and gentian-violet, give but a pale and unsatisfactory preparation. , The staining method most commonly used is the one originally recommended by Schaudinn and Hoffmann. This depends upon the use of Qiemsa's azur-eosin stain employed in various modi- fications. The most satisfactory method of applying this solution is as follows: Make smears upon slides or cover-slips, if possible from the depth of the lesions, as free as possible from blood. Fix in methyl alcohol for ten to twenty minutes and dry. Cover the preparation with a solution freshly prepared as follows: Distilled water 10 c.c. Potassium carbonate 1 : 1,000 5-10 gtt. Add to this: Giemsa's solution {Jilr Romanowski Fdrbung) 10-12 gtt. This staining fluid is left on for one to four hours, preferably in a moist chamber. Wash in running water. Blot. By this method Spirochseta pallida is stained characteristically with a violet or reddish tinge. A rapid and convenient method for staining such smears consists in the use of azur I and eosin in aqueous solutions as recommended by Wood (see section on Staining, page 109). The smears are fixed in methyl alcohol as before and are then flooded with the azur I solution. The eosin solution is then dropped on the preparation until an iridescent 1 For a critical summary of the various methods of dark-field illumination, the reader is referred to an article by Siedentopf, Zeit. f. wiss. Mikrosc, xxv, 1908. 598 PATHOGENIC MICROORGANISMS pellicle begins to form. Satisfactory preparations may be obtained by this method after ten or fifteen minutes of staining. An excellent method of staining the treponema pallidum in smear- preparations is that of Fontana.^ For this method, the following solu- tions are necessary: 1 . Acetic acid 1 c.c. Formalin 2 c.c. Distilled water 100 c.c. Leave in one minute; wash in water. 2. Phenol 86% (Uquefied crystals) 1 c.c. Tannic acid 5 grams Distilled water 83 c.c. Cover preparation with this and steam gently one-half minute; wash. 3 . Silver nitrate 0 . 25 grams Distilled water 100 c.c. Ammonia q. s. Add ammonia drop by drop until the precipitate which first appears goes into solution. Steam one-half minute; wash. Recently a rapid and extremely simple and reliable method for the demonstration of Spirochseta pallida in smears, by the use of India ink, has been described. Smears are prepared in the following way: A drop of the fluid squeezed out of the syphilitic lesion, as free as possible from blood cells, is mixed, on a slide, with a drop of India ink (best variety is '' Chin chin" Giinther-Wagner Liquid Pearl ink), and the mixture smeared with the edge of another slide as in making blood smears. When the smear dries, which takes about a minute, it^may be immediately examined with an oil-immersion lens. The organisms are seen unstained on a black back- ground. (See Fig. 129, p. 594.) Demonstration of Spirochetes in Tissues. — Ordinary histo- logical staining methods do not reveal the spirochaetes in tissue sections. It is customary, therefore, to employ some modification of Cajal's silver impregnation. The technique most commonly employed is that known as Levaditi's method,'^ which is carried out as follows: The fresh tissue is cut into small pieces which should not be thicker than 2 to 4 millimeters. Fix in 10% formalin (4% formaldehyde) for twenty-four hours. Wash in water. ^See Levaditi and Bankowski: Ann. de I'lnsti. Past., 1913, XXVII, p. 583. 2 Levaditi, Comptes rend, de la see. de bioL, 59, 1905. DISEASES CAUSED BY SPIROCHiETES 599 Dehydrate in 96% alcohol twenty-four hours. Wash in water. Place in a 3% silver-nitrate solution at incubator temperature (37.5° C.) and in the dark for 3 to 5 days. Wash in water for a short time. Place in the following solution (freshly prepared) : PjTogallic acid 2-4 grams. Formalin 5 c.c. Distilled water 100 " Leave in this for twenty-four to forty-eight hours at room temperature. Wash in water. Dehydrate in graded alcohols. Embed in paraffin and cut thin sections. The sections may be examined without further staining, or, if desired, may be weakly counterstained with Giemsa's solution or hematoxyUn. A modification of this method which has been much recommended is that of Levaditi and Manouelian} The directions given by these authors are as follows: Fix in formahn as in previous method. Dehydrate in 96% alcohol 12 to 24 hours. Wash in distilled water. Place in a 1% silver-nitrate solution to which 10% of pyridin has been added just before use. Leave in this solution for 2 to 3 hours at room temperature and from four to six hours at 50° C. approximately. Wash rapidly in 10% pyridin. Place in a solution containing 4% of pyrogallic acid to which 10% of C. P. acetone, and 15% (per volume) of pyridin have been added just before use. Leave in this solution 2 to 3 hours. Wash in water, dehydrate in graded alcohols, and embed in paraffin by the usual technique. Examined after treatment by either of these methods, the spiro- chaetes appear as black, untransparent bodies lying chiefly extracellu- larly. They are characteristically massed about the blood-vessels of the organs and only exceptionally seem to penetrate into the interior of the parenchyma cells. Attempts at cultivating Spirochseta pallida were at first unsuccessful. Recently Schereschewsky ^ has reported that he has succeeded in ob- taining multiplication of the organisms on artificial media as follows: Sterile horse serum in centrifuge tubes was coagulated at 60° C. until it assumed a jelly-like consistency. It was then placed in the incubator ^ Levaditi et Manoiielian, Comptes rend, de la see. de bid., 60, 1906. 2 Schereschewsky, Deut. med. Woch., N. S., xix and xxix, 1909. 600 « PATHOGENIC MICROORGANISMS at 37.5° C. for three days before being used. The cultures were planted by snipping off a small piece of tissue from a sj^hilitic lesion, dropping it into such a tube, and Causing it to sink to the bottom by means of centrifugalization. The tube was then tightly stoppered with a cork. In such anaerobic serum cultures Schereschewsky claims to have grown the organisms for several generations, though not in pure culture. Miihlens also obtained growth of Spirochseta pallida in horse serum agar by a method which is very similar to that of Schereschewsky. The most ex- tensive and convincing work on trepon- ema pallidum has been done recently by Noguchi. Noguchi ^ began his work in Fig. 130. — SpIROCH.ETA pal- mirk ^ inii XS' n j. r 1 1 CI .1 u lyiO and 1911. His first successful cul- LiDA. Spleen, congenital syph- ilis. (Levaditi method.) tivations were made from the syphilis- infected testicles of rabbits, and after many unsuccessful attempts, with slightly varying media and technique, he finally succeeded in the following way: He pre- pared tubes (20 cm. high and 1.5 cm. wide), containing 10 c.c. of a serum-water made of distilled water, three parts; and horse, sheep, or rabbit serum, one part. These were sterilized by the fractional method in the usual way (15 minutes each day). Into them was then placed a small piece of sterile rabbit kidney or testicle and a bit of the testicle of a syphilitic rabbit, in which many spirochaetes were present. The fluid was then covered with sterile paraffin oil and placed in an anaerobic jar. After 10 days at 33.5° C. the spirochaetes had multiplied considerably, in all but one case, together with bacteria. He obtained pure cultures from these initial cultivations after much diffi- culty, by a number of methods. At first he succeeded only by allowing the spirochaetes to grow through Berkefeld filters, which they did on the fifth day. A better method more recently adopted by him consists in preparing higl:i tubes of three parts of very slightly alkaline or neutral agar to which a piece of sterile tissue has been added. These tubes are then inoculated from the impure cultures with a long pipette. Close to the tissue and along the stab the spirochaetes and bacteria will grow and, after about ten days to two weeks, the spirochaetes will have wan- dered away from the stab and will be visible as hazy colonies. They can ^Noguchi, Jour. Exp. Med., xiv, 1911; xvii, 1913. DISEASES CAUSED BY SPIROCH.ETES 601 then be fished, after cutting the tubes, and directly transplanted to other serum-agar-tissue tubes prepared as before, and eventually will grow in pure culture. By this method Noguchi has also cultivated pure cultures from lesions in monkeys, and has produced lesions both in rab- bits and monkeys with his pure cultures. He has thus for the first time carried out Koch's postulates with syphilis and established beyond the shadow of a doubt the etiological significance of Spirochseta pallida in syphilis. The writer, with Hopkins, has successfully applied Noguchi's method and has found that, after once cultivated artificially, the treponema pallidum can be obtained in quantity best by cultivation in flasks con- FiG. 131. — SpiROCHiETA PALLIDA. Livcr, congenital syphilis. (Levaditi method.) taining heated or unheated rabbit kidney with ascitic broth and sealed with paraffin. Recently we have been using modifications of a method worked out in our laboratory by Miss Gilbert, in which slanted egg, with or without glycerin, made as for tubercle cultivation, is used in- stead of kidney tissue. This is put up in high tubes and ascitic broth and paraffin oil added. By this method, large quantities of culture, pallida are obtained within two weeks and can be concentrated in large quantities. Animal Pathogenicity. — Until very recently, all experimental inoc- ulation of animals was unsuccessful. During the year 1903 Metchnikoff and Roux ^ finally succeeded in transmitting the disease to monkeys. The monkey first used by these observers was a female chimpanzee. At the point of inoculation, the clitoris, there appeared, twenty-six days after inoculation, a typical indurated chancre, which was soon followed by swelling of the inguinal glands. Fifty-six days after the inoculation there appeared a typical secondary eruption, together with swelling of the spleen and of the lymph nodes. Similar successful experiments were 1 Metchnikoff et Roux, Ann. de Tinst. Pasteur, 1903, 1904, and 1905. 602 PATHOGENIC MICROORGANISMS made soon after this by Lassar.^ Soon after the experiments of Metch- nikoff and Roux, successful inoculations upon lower monkeys (maca- cus) were carried out by Nicolle.^ Since that time, it has been found by various observers that almost all species of monkeys are susceptible. Simple subcutaneous injection is not sufficient to produce a lesion. The technique which has given the most satisfactory results consists in the cutaneous implantation of small quantities of syphilitic tissue obtained by excision or curetting of primary and secondary lesions. A small pocket is made under the mucous membrane of the genitals or of the eyebrows and the tissue placed in this under aseptic precautions. The inoculation may be made directly from the human being, but can also be successfully carried out from monkey to monkey for many generations. Attempts at transmission from tertiary lesions have so far been unsuccessful. The spirochsetes can be demonstrated both in the primary lesions of the inoculated animal and in the secondarily enlarged glands. The successful inoculation of rabbits with syphilis has been recently performed by Bertarelli.^ He obtained ulcerative lesions by inoculation upon the cornea and into the anterior chamber of the eye and was able to prove the syphilitic nature of these lesions by finding the spirochsete within the tissue. In these animals, as well as in the lower monkeys, the disease usually remains localized. In 1907, Parodi showed that syphilitic lesions could be produced by direct inoculation into the testicles of rabbits. This method of inocula- tion has been subsequently studied by many investigators, especially by Uhlenhuth and Mulzer.^ It is the easiest method of obtaining spirochaete in any quantity from lesions in man. The spirochaete-con- taining lesions may be either excised or scraped as conditions permit and rubbed up in a mortar with sterile sand, in a few centimeters of sterile human ascitic fluid. This emulsion is then injected directly into the substance of rabbit testicles. A swelling supervenes which is often noticeable after two weeks, and is usually at its height in 5 to 7 weeks. At this time the testicle is much larger than normal, some- times evenly swollen and sometimes nodular, and of a firm elastic con- sistency. When taken out at castration it oozes a sticky fluid, both from testicle and tunica, which is rich in actively motile spirochaetes. By. continuous transinoculation from one rabbit to another such a strain can be indefinitely carried along. It can be inoculated from rabbits 1 Lassar, Berl. klin. Woch., xl, 1903. ^ Nicolle, Ann. de I'inst. Pasteur, 1903. » Bertarelli, Cent. f. Bakt., xli, 1906. ^ Uhlenhuth und Mvlzer, Arb. a. d. k. Gesimdh'tsamt., xxxiii, 1909. DISEASES CAUSED BY SPIROCHiETES 603 to monkeys and vice versa. This method as well as Noguchi's cul- tivations have opened a new era of spirochsete investigation. It is stated by some observers that intravenous inoculation of rabbits may be followed by localization in the testis and occasionally gummatous infections in other parts of the body have been induced after such in- oculation by Uhlenhuth, Mulzer, and others. Immunization in Syphilis. — It is a well-known fact observed by clinicians that during active syphilis the patient cannot be superin- fected. That this resistance develops quite rapidly was shown by Metchnikoff and Roux, who found that reinfection of a monkey was possible if attempted within two weeks of the first inoculation, but was unsuccessful if delayed beyond this period. On the basis of this knowledge, Metchnikoff,^ Finger and Land- steiner,^ and others have made attempts to devise some method of im- munization. They attempted to attenuate the syphilitic virus by re- peated passage through monkeys. These experiments were unsuccess- ful, the last-mentioned observers finding absolutely no attenuation after twelve generations of monkey inoculation. Bertarelli and others have shown that the production of a syphilitic lesion on the cornea of one eye does not protect against an inoculation done on the other. Rabbits that haye been inoculated jvith spirochsete material and that have not developed syphilitic disease can be success- fully inoculated on subsequent attempts. The offspring of female rab- bits with syphilis of the cornea are, according to Muhlens, not immune. There is no evidence so far that specific therapy or treatment with spirochaete material has had favorable influence upon the disease. Chemotherapy has had results analogous to those obtained in- man.^ Attempts at passive immunization have been entirely without success. Investigations carried on in our own laboratory in the last three years have shown definitely, we think, that immunization of animals with culture pallida produces antibodies, agglutinins, treponemacidal substances, entirely analogous to similar substances produced against bacteria. However, there is a biological change which takes place when treponema pallidum is cultivated. The antibodies produced with the culture pallida have no action whatsoever upon the virulent 1 Metchnikoff, Arch. g^n. de m6d., 1905. 2 Finger und Landsteiner, Sitzungsber. d. Wien. Akad. d. Wiss., 1905. » Von Prowazek, "Handbuch der pathogenen Protozoen," i, 1912, Leipzig, Bartscli. 604 PATHOGENIC MICROORGANISMS organisms. The latter, indeed, seem to be entirely insulated against such antibodies and do not induce antibody formation to any great extent, in either the infected animal or man. Both active and passive immunization with culture pallida and the sera produced with them have no effect. We have obtained some evidence, however, that in rab- bits a purely local resistance developes in the tissue praviously the site of a lesion. The occurnence of a Wassermann reaction was formerly supposed to indicate the existence of specific syphilitic antibodies in the serum of patients. More recent- information regarding this reaction seems to show that it depends upon the presence in the serum of syphilitic patients of substances produced indirectly because of the presence of syphilitic infection. It may be a relative increase of globulins or, as Schmidt has suggested, a change in the physical state of the globulins or other substances present in the serum. At any rate it has been found that the fixation of complement in the Wassermann reaction does not depend upon the occurrence of a specific antigen-antibody reaction. In the first place the antigens most commonly used, and successfully so, in the Wassermann reactions, are non-specific lipoidal extracts of normal organs. Again it has been demonstrated that extracts of cultures of the Spirochseta pallida as well as extractions from the testes of syphilitic rabbits do not furnish an antigen suitable for the Wassermann reaction. This has followed especially from the work of Noguchi,^ of Craig and Nichols,^ and ourselves. This forms a corollary to the other experi- ments previously mentioned and shows that, whatever the Wassermann reaction may be, it is not a specific complement fixation in the sense of Bordet and Gengou. It must be admitted, therefore, that our knowl- edge of syphilis immunity is in its infancy and that we know very little about the systemic reactions which follow infection with the Spirochseta pallida. The fact that the syphilitic virus does not pass through a filter has been demonstrated by Klingmiiller and Baermann,^ who inoculated themselves with filtrates from syphilitic material. ^Zinsser and Hopkins, Jour, of Exp. Med., xxi, 1915, p. 576; xxiii, 1916, p. 323; xxiii, 1916, p. 329; xxiii, 1916, p. 341. 2 Noguchi, Jour. Am. Med. Assoc, 1912. 3 Craig and Nichols, Jour. Exp. Med., xvi, 1912. * Klingmiiller und Baermann, Deut. med. Woch., 1904. DISEASES CAUSED BY SPIROCH.ETES 605 THE SPIROCHiETES OF RELAPSING FEVER The microorganisms causing relapsing 'fever were first observed in 1873, by Obermeier/ who demonstrated them in the blood of patients suffering from this distinct type of fever. Since his time extensive ^0 Fig. 132. — Spirochete of Relapsing Fever, and Flournoy.) (After Norris, Pappenheimer, studies by many other observers have proven beyond question the etiological connection between the disease and the organisms. Morphology and Staining. — The spirochsete of Obermeier is a delicate spiral thread measuring from 7 to 9 mi era in length (Novy), and about 1 micron in thickness. While this is its average size, it may, according to some observers, be considerably longer than this, its undulations varying from 4 to 10 or more in number. Compared with the red blood ^Obermeier, Cent. f. d. med. Wiss., 11, 1873. 606 PATHOGENIC MICROORGANISMS cells among which they are seen, the microorganisms may vary from one -half to 9 or 10 times the diameter of a corpuscle. In fresh prepara- tions of the blood, very active corkscrew-hke motility and definite lateral oscillation are observed. In stained preparations no definite cellular structure can be made out, the cell body appearing homogeneous, except in degenerated individuals, in which irregular granulation or beading has been observed. Flagella have been described by various observers. Fig. 133. — Spiroch.ete of Relapsing Fever. Citrated normal rat blood. (After Norris, Pappenheimer, and Flouriioy.) Novy and Knapp^ believe that the organisms possess only one terminal flagellum. Zettnow,^ on the other hand, claims to have demonstrated lateral flagella by special methods of staining. Norris, Pappenheimer, and Flournoy,^ in smears stained by polychrome methods, have described long, filamentous tapering ends which they interpreted as bipolar, terminal flagella, never observing more than one at each end. Spores are not found. Cultivation. — Innumerable attempts to induce these microorganisms to multiply upon artificial media have been made. Novy and Knapp succeeded in keeping the microorganisms alive and virulent in the ^ Novy and Knapp, Jour, of Infec. Dis., 3, 1906. 2 Zettnow, Deut. med. Woch., 32, 1906. 3 Norris, Pappenheimer, and Flournoy, Jour, of Inf. Dis., 3, 1906. DISEASES CAUSED BY SPIROCILETES 607 original blood for as long as forty days, and call attention to the fact that the length of time for which they may be kept aUve depends to a great extent upon the stage of fever at which the blood is removed from the patient. They do not, however, believe that extensive multiphca- tion, or, in other words, actual cultivation, had taken place in their experiments. Norris, Pappenheimer, and Flournoy, on the other hand, have obtained positive evidence of multiplication of the spirochaetes in fluid media. They obtained their cultures by inoculating a few drops of spirochaetal rat blood into 3 to 5 c.c. of citrated human or rat blood. Smears made from these tubes, after preservation for twenty-four hours at room temperature, showed the microorganisms in greater number Fig. 134. — SpiROCHiETE of Relapsing Fever. (From preparation furnished by Dr. G. N. Calkins.) than in the original infected blood. A similar multiplication could be observed in transfers made from these '^ first-generation" tubes to other tubes of citrated blood. Attempts at cultivation for a third generation, however, failed. Noguchi ^ has lately successfully cultivated the spirochaete of Ober- meier in ascitic fluid containing a piece of sterile rabbit's kidney and a few drops of citrated blood under anaerobic conditions. Four different, probably distinct varieties of spirochsete have been described in connection with relapsing fever, all of which have been cultivated by Noguchi by means of this method. The first is known as the spirochsete of Obermeier mentioned above. Probably distinct 40 ^Noguchi, Jour. Exp. Med., xvii, 1913. 608 PATHOGENIC MICROORGANISMS are the Spirochseta Duttoni, described by Dutton and Todd^ inl905, the Spirochseta Kochi, and the Spirochaeta.Novyi,^ the organism studied by Norris and Flournoy and Pappenheimer, and regarded as a different species by them. Pathogenicity. — Inoculation with blood containing these spirochsetes produces disease in monkeys, rats, and mice. Attempts to transmit the disease experimentally to dogs, rabbits, and guinea-pigs have so far been unsuccessful. The subcutaneous inoculation of monkeys is followed after from two to four days by a rise of temperature which occurs abruptly as is the case in the disease in man and which may last several days. During this time the spirochsetes can be found in the blood of the animals just as it is found in that of infected human beings. The temperature subsides after a day or more, when it again rapidly returns to normal. As a rule, the paroxysms are not repeated. Occa- sionally, however, two or three attacks may supervene before immunity is established. In rats, an incubation time of from two to five days occurs. At the end of this time the spirochsetes may be found in large numbers in the blood, and the animals show symptoms of a severe systemic infection. The attack lasts from four to five days, at the end of which time the microorganisms again disappear. Occasionally even in these animals relapses have been observed. Gross pathological changes are not found, with the exception of an enlargement of the spleen. In man the disease caused by the spirochsete of Obermeier, commonly known as relapsing fever, is common in India, Africa, and most of the warmer countries. It has, from time to time, been observed epidemically in Europe, especially in Russia, and a few epidemics have occurred in the Southern United States. The disease comes on abruptly, beginning usually with a chill accompanied by a sharp rise of temperature and gen- eralized pains. Together with the rise of temperature, which often ex- ceeds 104° F., there are great prostration and occasioifally delirium. Early in the disease the spleen becomes palpable and jaundice may appear. The spirochsetes are easily detected in the blood during the persistence of the fever, which lasts usually from three to ten days. At the end of this time the temperature usually drops as suddenly as it rose, and the general symptoms rapidly disappear. After a free interval of from one to three weeks a relapse may occur, which is usually less severe and of shorter duration than the original attack. Two, three, or 1 Dutton and Todd, Brit. Med. Jour., 1905. ^ Novy and Fraenkel, cited from Noguchi. DISEASES CAUSED BY SPIROCHiETES 609 even four attacks may occur, but the disease is not very often fatal. When patients do succumb, however, the autopsy findings are not particularly characteristic. Apart from the marked enlargement of the spleen, which histologically shows the changes indicatin"g simple hyperplasia, and a slight enlargement of the Hver, no lesions are found. The diagnosis is easily made during the febrile stage by examination of a small quantity of blood under a cover-slip or in the hanging-drop preparation. Several types of relapsing fever have been described. In Africa the disease has long been prevalent in many regions and the investigations of Ross ami Milne, ^ Koch,^ Button and Todd,^ and others have brought Fig. 135. — Spiroch^ete of Dutton, African Tick Fever. (From prepara- tion furnished by Dr. G. N. Calkins.) to light that many conditions occurring among the natives, formerly regarded as malarial, are caused by a species of spirochsete. Whether" or not the microorganisms observed m the African disease are exactly identical with the spirochaete observed by Obermeier is yet a question about which several opinions are held. Dutton and Todd believe that the same microorganism is responsible for both diseases. Koch, on the other hand, beUeves that the slightly smaller size of the African spiro- chaete and the milder course of the clinical symptoms indicate a defi- nite difference between the two. Animal experiments made with the African organism, furthermore, usually show a much more severe in- fection than do similar inoculations with the European variety. The 1 Ross and Milne, Brit. Med. Jour., 1904. 2 Koch, Deut. med. Woch., xxxi, 1905. 3 DiUton and Todd, Lancet, 1905, and Jour, of Trop. Med., 1906. 610 PATHOGENIC MICROORGANISMS spirochsete found in the African disease is usually spoken of at present as "Spirochaeta Duttoni.'' Novy and Knapp/ after extensive studies with the microorganisms from various sources, have come to the conclu- sion that,' although closely related, definite species differences exist be- tween the two types mentioned above, and that these again are definitely distinguished from similar organisms described by TurnbulP as occurring in a similar disease observed in India. The mode of transmission of this disease is not clear for all types. Dutton and Todd, however, were able to show satisfactorily that, in the case of the African disease at least, transmission occurs through the intermediation of a species of tick. The conditions under which such intermediation occurs have been carefully studied by Koch.^ The tick (Ornithodorus moubata) infects itself when sucking blood from an infected human being. The spirochsete may remain alive and demonstrable within the body of the tick for as long as three days. Koch has shown, furthernore, that they may be found also within the eggs laid by an infected female tick. He succeeded in producing experi- mental infection in monkeys by subjecting the animals to the bites of the infected insects. For the European variety of the disease no such intermediate host has as yet been demonstrated. Immunity. — It has long been a well-known fact that recovery from an attack of relapsing fever usually results in a more or less definite immunity. The blood of human beings, monkeys, and rats which have recovered from an attack of this disease show definite and specific bactericidal and agglutinating substances, and Novy and Knapp have demonstrated that the blood serum of such animals may be used to ponfer passive immunity upon others. VINCENT'S ANGINA The condition known as Vincent^s angina consists of an inflamma- tory lesion in the mouth, pharynx, or throat, situated most frequently upon the tonsils. The disease usually begins as an acute stomatitis, pharyngitis, or tonsillitis, which soon leads to the formation of a pseudo- membrane, which, at this stage, has a great deal of resemblance to that caused by the diphtheria bacillus. At later stages of the disease there may be distinct ulceration, the ulcers having a well-defined margin 1 Novy and Knapp, loc. cit. 2 Tumbull, Indian Med. Gaz., 1905. » Koch, Berl. med. Woch., 1906. DISEASES CAUSED BY SPIROCH.ETES 611 and "punched-out" appearance, so that clinically they have of ten been erroneously diagnosed as syphilis. Apart from the localized pain, the disease is usually mild, but occasionally moderate fever and systemic disturbances have been observed. Unlike diphtheria and syphilis, this peculiar form of angina usually yields, without difficulty, to local treat- ment. The nature of lesions of this peculiar kind was not clear until Plaut,^ i . ^^ • V. \ x,/ *| ' . * *l| ¥ i ^1 . Fig. 136. — Smear prom the throat of a Case of Vincent's Angina. Giemsa Stain. Vincent,^ and others reported uniform bacteriological findings in cases of this description. These observers have been able to demonstrate in smears from the lesions a spindle-shaped or fusiform bacillus, to- gether with which there is usually found a spirillum not unlike the spirillum of relapsing fever. The two microorganisms are almost 1 Plant, Deut. med. Woch., xlix, 1894. 2 Vincent, Ann. de I'inst. Pasteur, 1896, and Bull, et m6m. de la see. med. des h6p. de P., 1898. 612* PATHOGENIC MICROORGANISMS always found together in this form of disease and were regarded by the first observers as representing two distinct forms dwelHng in sym- biosis. More recently Tunnicliff/ on the basis of experimental work, has claimed identity for the two forms, believing that they represent different developmental stages of the same organism. The fusiform hacilli described by Vincent, Plant, Babes, and others, are from 3 to 10 micra in length, and have a thickness at the center varying from 0.5 to 0.8 micron. From the center they taper gradually ^ ./ ) r ^ \ Fig. 137. — ^Throat Smear. Vincent's Angina. Fusiform bacilli and spirilla. toward the ends, ending in blunt or sharp points. The length of these bacilli may vary greatly within one and the same smear preparation. They are usually straight, sometimes slightly curved. They do not stain very easily with the weaker anilin dyes, but are readily stained by Loeffler's methylene-blue, carbol-fuchsin, or better, by Giemsa's stain. Stained by Gram, they are usually decolorized, though in this respect the writers have found them to vary. Stained preparations show a charac- teristic inequality in the intensity of the stain, the bacilli being more 1 Tunnidiff, Jour, of Infec. Dis., 3, 1906. . DISEASES CAUSED BY SPIROCH.ETES 613 deeply stained near the end, and showing a banded or striped alternation of stained and unstained areas in the central body. Their staining quaUties in this respect are not unlike those of the diphtheria bacillus, and according to Babes ^ the dark areas are to be interpreted as meta- chromatic granules. The bacilli are not motile. The spirilla found in Vincent's angina are usually somewhat longer than the fusiform bacilli, and are made up of a variable number of un- dulations, shallow and irregular in their curvatures, unlike the more regularly steep waves of Spirochseta pallida. They are stained with even more difficulty than are the bacilli and usually appear less distinct in the preparations. The stain, however, is taken without irregu- larity, showing none of the apparent metachromatism observed in the bacilli. By the earlier observers cultivation of these microorganisms was attempted without success. Recently, however, it has been shown that cultivation could be carried out under anaerobic conditions. Tunni- cliff ^ has cultivated the organisms anaerobically upon slants of ascitic agar at 37.5° C. This observer found that in such cultures, before the fifth day, bacilli only could be found, that after this time, however, spirilla gradually appeared and finally constituted the majority of the organisms in the culture. It appeared to TunnicHff from this study that the spirilla might be developed out of the fusiform microorgan- isms representing the adult form. The microorganisms of Vincent's angina, when occurring in the throat, are rarely present alone, being usually accompanied by other microorganisms, such as staphylococci, streptococci, and not infre- quently diphtheria bacilli. When occurring together with diphtheria, they are said, by some German observers, to aggravate the latter condition considerably. This frequent association with other micro- organisms renders it impossible to decide conclusively that the fusi- form bacilli and spirilla are the primary etiological factors in these inflammations. It has been frequently suggested that they may be present as secondary invaders upon the soil prepared for them by other microorganisms. Animal inoculation with these microorganisms has led to little result. Fusiform Bacilli other than those in Vincent's Angina. — ^Fusiform bacilli morphologically indistinguishable from those found in the angina of Vincent may frequently be found in smears taken from the gums, * Babes, in KoUe und Wassermann, 1 . Erganzungsband, 1907. 2 TunnicHff, Jour, of Infec. Dis., 3, 1906. 614 PATHOGENIC MICROORGANISMS from carious teeth, and occasionally among the microorganisms in the pus from old sinuses. Several varieties of these bacilli have been de- scribed in connection with definite pathological conditions. Babes/ in 1893, observed spindle-shaped bacilli not unlike those described above, but somewhat shorter, in histological sections prepared from tissues from the gums of individuals suffering from scurvy. He found similar bacilli in rabbits intravenously inoculated with material from the patients and was able to cultivate the bacilli for several genera- tions. His descriptions, however, of the microorganisms as found in the secondary cultures vary considerably from those of the original findings in the gums of the patients. His results are not convincing. In noma, a gangrenous disease of the gums and cheeks, occurring occasionally in individuals who have been severely run down by acute infectious diseases or great hardship. Weaver and Tunnicliff have found spirilla and fusiform bacilli in large numbers. The organisms were pres- ent not only in smears from the surface, but were also found by histo- logical methods, in large numbers, lying in the tissues beyond the area of necrosis. Here again it is not entirely certain whether these microorganisms were the primary etiological factors or whether they are to be regarded merely as secondary invaders of a necrotic focus. Fusiform bacilli are cultivated with greater ease than formerly sup- posed; we have found it relatively simple to grow them together with Gram positive cocci in symbiosis in simple broth tubes covered with paraffin oil without the addition of any enriching substance and in similar symbiotic conditions on infusion agar plates under incomplete anaerobic conditions. In such plates they form curious colonies in which the fusiform bacilli and micrococci are intimately commingled. Krumwiede ^ has had no difficulty in cultivating them in pure culture in anaerobic plates. SPIROCHUETA PERTENUIS In a disease known as ^'Framboesia tropica," or popularly "Yaws," occurring in tropical and subtropical countries and much resembling syphilis, Castellani,^ in 1905, was able to demonstrate a species of spirochsete which has a close morphological resemblance to Spirochseta pallida. The microorganism was found in a large percentage of the cases 1 Babes, Deut. med. Woch., xliii, 1893. ^Krumwiede, Jour. Inf. Dis., 1913. 3 CasteUani, Brit. Med. Jour., 1905, and Deut. med. Woch., 1906. DISEASES CAUSED BY SPIROCHETES 615 examined both in the cutaneous papules and in ulcerations. Confirm- atory investigations on a larger series of cases were later carried out by von dem Borne. ^ The microorganism is from 7 to 20 micra in length with numerous undulations and pointed ends. Examined in fresh preparations, it has an active motility similar to that of Spirochaeta paUida. In smears it is easily stained by means of the Giemsa method. Both the clinical similarity between yaws and syphilis, as well as the similarity between the microorganisms causing the diseases, has opened the question as to the identity of the two microorganisms. According to most clinical observers, however, yaws, which is a disease characterized chiefly by a generalized papular eruption, is unquestion- ably distinct, clinically, from lues, and experiments of Neisser, Baermann, and Halberstadter,2as well as of Castellani ^ himself, have tended to show that there is a distinct difference between the immunity produced by attacks of the two diseases. The disease is transmissible to monkeys, as is syphihs, but it has been satisfactorily shown that monkeys inocu- lated with syphilitic material, while no longer susceptible to infection with Spirochaeta pallida, may still be successfully inoculated with Spirochaeta pertenuis. I SPIROCHAETA GALLINARUM An acute infectious disease occurring among chickens, chiefly in South America, has been shown by Marchoux and Salimbeni^ to be caused by a spirochaete which has much morphological similarity to the spirochaete of Obermeier. The disease comes on rather suddenly with fever, diarrhea, and great exhaustion, and often ends fatally. The spirochaete is easily demon- strated in the circulating blood of the animals by staining blood-smears with Giemsa's stain or with dilute carbol-fuchsin. Artificial cultivation of the microorganism has not yet been ac- complished. Experimental transmission from animal to animal is easily carried out by the subcutaneous injection of blood. Other birds, such as geese, ducks, and pigeons, are susceptible; mammals have, so far, not been successfully inoculated. According to the investigations of 1 Von dem Borne, Jour. Trop. Med., 10, 1907. 2 Neisser, Baermann, und Halherstadter, Miinch. med. Woch., xxviii, 1906. 3 Castellani, Jour, of Hyg., 7, 1907. * Marchovx et Salimbeni, Ann. de Tinst. Pasteur, 1903. 616 PATHOGENIC MICROORGANISMS Levaditi and Manouelian/ the spirochsetes are found not only in the blood but thickly distributed throughout the various organs. Under natural conditions, infection of chickens seems to depend upon a species of tick which acts as an intermediate host and causes infection by its bite. The spirochsete, according to Marchoux and Sal- imbeni, may be found in the intes- tinal canal of the ticks for as long as five months after their infection from a diseased fowl. In the blood of animals which have survived an infection, agglutin- ating substances appear and active immunization of animals may be carried out by the injection of in- fected blood in which the spirochsetes have been killed, either by moderate heat or by preservation at room temperature. The serum of immune animals, furthermore, has a pro- tective action upon other birds. It is not impossible that the Spiro- chseta gallinarum may be identical with the Spirochseta anserina previ- ously discovered by Sacharoff .^ This last-named microorganism causes a disease in geese, observed espe- cially in Russia and Northern Africa, which both clinically and in its pathological lesions corresponds closely to the disease above described as occurring in chickens. The spirochsete is found during the febrile period of the disease in the circulating blood, is morphologically indis- tinguishable from the spirochsete of chickens, and can not be cultivated artificially. The similarity is further strengthened by the fact that Spirochaeta anserina is pathogenic for other birds, but not for animals of other genera. Noguchi has succeeded in cultivating Spirochaeta gallinarum by the same method by which he has cultivated the or- ganisms of relapsing fever. Ascitic fluid tubes with a piece of sterile rabbit kidney were inoculated with a few drops of blood containing the spirochsetes and cultivated at 37.5° C. under anaerobic conditions. Spirochaeta phagedenis. — This is an organism cultivated by Noguchi 1 Levaditi et Manouelian, Ann. de I'inst. Pasteur, 1906. 2 Sacharoff, Ann. de I'inst. Pasteur, 1891. Fig. 138. — Spiroch^ta gallina- rum . (From preparation furnished by Dr. G. N. Calkins.) DISEASES CAUSED BY SPIROCHiETES 617 by his ascitic-fluid-tissue method from phagedenic lesions on human external genitals. It is probably a new species. Spirochseta macrodentium. — Cultivated by Noguchi; * is believed by him to be identical with the spirochsete found in Vincent's angina. Spirochseta microdentium. — ^A similar organism with wide con- volutions, cultivated by Noguchi from the tooth deposits chiefly in children. It was grown on mixtures of sheep serimi water and sterile tissue in a way similar to that employed by him for other organisms of this group. SpirochsBta calligyrum. — Cultivated by Noguchi ^ from condylomata — ^is probably a new species. Weil's Disease. — Weil's disease is a malady which has been known for a long time, in which there is a moderate febrile movement, with jaundice, enlarged liver, and a hemorrhagic eruption. In this disease, Inada, Yutaka, Hoki, Kaneko, and Ito ^ have described the Spiro- chaete icterohaemoragica. They have found the organisms in the liver by the Levaditi method, the adrenal glands and the kidneys, and have transmitted it, with the blood of cases, to guinea-pigs by intraperi- toneal injection. They succeeded in cultivating it by the Noguchi method. Rat-Bite Fever. — Rat-bite fever is a peculiar disease, which, after an incubation period of ten or more days, is characterized by fever, head- ache and inflammation at the site of the bite, swollen lymph glands, skin eruption and pains. After three to six days the fever ceases and an afebrile period of two or three days ensues. After this the fever again occurs. Recently Futaki, Takaki, Taniguchi, and Osumi * have described a treponema which they have called Treponema morsus muris: It is a spiral organism, somewhat larger than the Treponema pallidum, and is found in the skin, the lymph nodes, and in the blood. They have succeeded in inoculating rats and have cultivated it in Schi- mamine medium, which consists of 100 c.c. of horse serum in which 0.5 to 0.75 gram of sodium nucleate is dissolved and carbon dioxide passed through the solution until the serrmi becomes transparent. It is then heated for three days at 60°, and on the fourth day at 65° until it coagulates. This medium is deeply inoculated, but no other anaerobic precautions are taken. * Noguchi, Jour. Exp. Med., xv, 1912. * Noguchi, Jour. Exp. Med., xvii, 1913. ' Jour, of Exp. Med., 1916, xxiii, p, 249. * Jour. Exp. Med,, 1916, xxiii, p, 377. H; GLENN BfilL CHAPTER XLIV THE HIGHER BACTERIA {Chlamydobacteriacece, Trichomycetes) Standing midway between the true bacteria and the more complex molds or Hyphomycetes, there are a number of pathogenic micro- organisms which offer great difficulties to classification. In the classifi- cation of Migula most of these forms have been placed in a rather heterogeneous group, the Chlamydobacteriacese. By other authors, notably Lachner-Sandoval,^ Berestnew,^ and by Petruschky,^ the close relationship of these forms to the higher hyphomycetes has been em- phasized and they have been grouped as a subdivision of the true molds under the family name of Trichomycetes. Petruschky'' proposes the following clear schematization, which, even though possibly defective from a purely botanical point of view, is at least serviceable for the purposes of the bacteriologist. Hyphomycetes True molds Trichomycetes Leptothrix Cladothrix Streptothrix Actinomyces Leptothrix is used to designate those forms which appear as simple threads without branching. Cladothrix is a thread-like form in which false branching may be recognized. By false branching is meant an appearance resulting from the fragmentation of threads. The terminal cell breaks away from the main stem, is set at an angle by the elongation of the thread itself, and, ^ Lachner-Sandoval, "Ueber Strahlenpilze." Diss. Strassburg, 1898. 2 Berestnew, Ref. Cent. f. Bakt., xxiv, 1898. 3 Petnischky, in KoUe imd Wassermann, "Handbuch," etc. * Petruschky, loc. cit. 618 THE HIGHER BACTERiA* 619 as both continue dividing, the simulation of true branching is pro- duced. Streptothrix denotes forms with numerous true branches and spores which usually appear in chains. Actinomyces is of more complicated structure, characterized by the formation of club-shaped ends and the stellate arrangement of its threads. LEPTOTHRIX Members of the leptothrix group have been observed in connection with inflammations of the mouth and pharynx by Frankel,^ Michelson,^ Epstein,^ and others. In many of these cases the organism was identi- fied by morphology chiefly, pure cultures not having been obtained. The disease in none of these cases was accompanied by severe systemic symptoms and it is likely that when found in human beings the organ- isms may be regarded simply as comparatively harmless saprophytes appearing in connection with some other specific inflammation. Cultivation of the Leptothrices is not easy and has been successful only in the hands of Vignal * and Arustamoff .^ GLADOTHRIX Owing to much confusion in the differentiation of these forms from the streptothrices, it is not possible to determine whether cases of true cladothrix infection have been observed. It is likely that most cases ascribed to microorganisms of this class have really been due to strep- tothrix infection. The deciding criterion is, of course, the formation of branches and these seem to have been observed in most of the cases described. A closer differentiation, in the future, between true and false branching can alone determine whether or not cases of cladothrix infection proper may occur. STREPTOTHRIX Reports of cases of streptothrix infection of various parts of the body, in both animals and man, are abundant in the literature. The * Frdnkel, Eulenburg's " Realencycl. d. gesam. Heilkunde," 1882. 2 Michelson, Berl. klin. Woch., ix, 1889. » Epstein, Prag. med. Woch., 1900. • Vignal, Ann. de phys., viii, 1886. » Arustamoff, Quoted from Petruschky, loc. cit 620 PATHOGENIC MICROORGANISMS earliest observations were made upon microorganisms isolated from the human conjunctiva. Nocard * in 1888 described a member of this group as the etiological factor in a disease " farcies du boeuf " occurring among cattle in Guadeloupe. Eppinger ^ found streptothrices in the pus of a cerebral abscess. Petruschky,^ Berestneff,'* Flexner,^ Norris and Fig 139. — Cladothrix, Showing Fausb Bbanching. Larkin,® and a number of other observers have found these microor- ganisms in cases of pulmonary disease, simulating tuberculosis. Sup- purations of bone and of the skin and the intestinal canal have been reported. The infection, therefore, is not very rare, but the diverse experiences of workers who have attempted to cultivate these micro* » Nocard, Ann. de Finst. Pasteur, ii, 1888. a Eppinger, Wien. klin. Woch., 1890. s Petruschky, Verhandl. d. Kongr. f. innere Mediz., 1898. *Berestneff, Zeit. f. Hyg., xxix, 1898. *Flexner, Jour. Exp. Med., iii, 1896. • Norris and Larkin, Proc. of N. Y. Path. Soc, March, 1899. THE HIGHER BACTERIA 621 organisms seem to indicate that not all of the incitants described be- longed to one and the same variety, but that probably a number of different types may exist. Morphology. — Morphologically the streptothrices show considerable variation. In material from infectious lesions they have most often appeared as rods and filaments with well-marked branching. Occasion- ally the filaments are long and interwined, and branches have shown bulbous or club-shaped ends. In Norris and Larkin^s case, the young cultures in the first generations seem to have consisted chiefly of rod- shaped forms not uYi'.'ke bacilh of the diphtheria group, showing marked metachromatisn: when stained with Loeffler's methylene-blue. They are Fig. 140. — Streptothrix, Showing True Branching. easily stained with this dye or with aqueous fuchsin. In tissue sections they may be demonstrated by the Gram-Weigert method. Cultivation.— Direct cultivation upon agar and gelatin plates has occasionally been successful. At the end of four or five days grayish- white, glistening, flat colonies may appear which attain a diameter of several millimeters within two weeks. The colonic later may take on a yellowish hue and begin to liquefy the gelatin. In bouillon flocculent precipitates and surface pellicles of interwined threads may form, with- out clouding of the medium. Norris and Larkin* found much difficulty in cultivating, but finally succeeded by making smears of the infectious material upon fresh, sterile kidney-tissue of rabbits. The micro- » Norris and Larkin, loc. dt. 622 PATHOGENIC MICROORGANISMS organisms grew abundantly upon this, but failed to grow on any of the other tissues. After growth of several generations upon this medium, cultures were finally obtained upon agar plates and upon broth. Inoculation of cultures into rabbits and guinea-pigs have given rise to subcutaneous abscesses, bronchopneumonia, and suppuration, accord- ing to the mode of infection. ACTINOMYCES Among the diseases caused by the Trichomycetes or higher bacteria, the most important is actinomycosis. Occurring chiefly in some of the domestic animals, notably in cattle, the disease is observed in man with sufficient frequency to make it of great clinical importance. In cattle the specific microorganism which gives ries to the disease was first observed by Bollinger* in 1877. In the following year Israel^ dis- covered a similar microorganism in human cases. The parasites appear in the pus from discharging lesions as small granular bodies, plainly visible to the naked eye and somewhat resem- bling sulphur granules, of a grayish or of a pale yellow color. In size they measure usually a fraction of a millimeter. Ordinarily they are soft and easily crushed under a cover-slip, but occasionally, especially in old lesions, they may be quite hard, owing to calcification. Microscopically they are most easily recognized in fresh preparations prepared by crushing the granules upon the slide under a cover-slip and examining them without staining. They may be rendered more clearly visible by the addition of a drop or two of 20 per cent potassium hydrate. When the granules are calcareous, the addition of a drop of concentrated acetic acid will facilitate examination. Fresh preparations may be examined after staining with Gram's stain. Observed under the micro- scope, the granules appear as rosette-like masses, the centers of which are quite opaque and dense, appearing to be made up of a closely meshed network of filaments. Around the margins there are found radially arranged striations which in many cases end in characteristically club- shaped bodies. Inside of the central network there are often seen coccoid or spore-like bodies which have been variously interpreted as spores, as degeneration products, and as separate cocci fortuitously found in symbiosis with the actinomyces. Individually considered, » Bollinger, Deutsch. Zeit. f. Thiermed., iii, 1877. 2 Israel Virch. Arch.,, 74, 1878, and 78, 1879. THE HIGHER BACTERIA 623 the central filaments have approximately the thickness of an anthrax bacillus and are, according to Babes/ composed of a sheath within which the protoplasm contains numerous and different sized granules. About the periphery of the granules the free ends of the filaments become gradually thickened to form the so-called actinomycosis " clubs." These clubs, according to most observers, must be regarded as hyaline thickenings of the sheaths of the threads and are believed to represent a form of degeneration and not, as some of the earlier observers believed, organs of reproduction. They are homogeneous, and in the smaller and presumably younger granules are extremely fragile and soluble in water. In older lesions, especially in those of cattle, the clubs are more re- sistant and less easily destroyed. They appear only in the parasites taken from active lesions in animals or man, or, as Wright ^ has found, from cultures to which animal serum or whole blood has been added. In cultures from media to which no animal fluids have been added, such as glucose agar or gelatin, no clubs are found. In preparations stained by Gram's method the clubs give up the gentian-violet and take counter-stains, such as eosin. The coccus-like bodies found occasionally lying between the filaments of the central mass, most observers now agree, do not represent any- thing comparable to the spores of the true hyphomycetes. In many cases they are unquestionably contaminating cocci; in others again they may represent the results of degeneration of the threads. In tissue sections, the microorganisms 'may be demonstrated by Gram's method of staining or by a special method devised by Mallory.^ This is as follows for paraffin sections : 1. Stain in saturated aqueous eosin 10 minutes. 2. Wash in water. 3. Anilin gentian- violet, 5 minutes. 4. Wash with normal salt solution. 5. Gram's iodin solution 1 minute. 6. Wash in water and blot. 7. Cover with anilin oil until section is clear. 8. Xylol, several changes. 9. Mount in balsam. Cultivation. — The isolation of actinomyces from lesions may be easy or difficult according to whether the pus is free from contamination » Babes, Virch. Arch., 105, 1886. 2 J. H. Wright, Jour. Med. Res., viii, 1905. » Mallory, Method No. 1, Mallory and Wright, " Path. Technique," Phila., 1908. 41 624 PATHOGENIC MICROORGANISMS 1 or whether it contains large numbers of other bacteria. In the latter case it may be almost impossible to obtain cultures. The descriptions of methods of isolation and of cultural characteristics given by various writers have shown considerable differences. The most extensive cultural work has been done by Bostroem/ Wolff and Israel, and by J. H. Wright. Bostroem has described his cultures as aerobic, but Wolff and Israel ^ and Wright ^ agree in finding that the microorganisms iso- FiG. 141. — ^Actinomyces Granule Crushed Beneath a Cover-glass. Un- stained. Low power. Shows radial striations. (After Wright and Brown.) lated by them from actinomycotic lesions grow but sparsely under aerobic conditions and favor an environment which is entirely free from oxygen, or at least contains it only in small quantities. The method for isolation recommended by Wright is, briefly, as follows: Pus is obtained, if possible, from a closed lesion and washed in sterile water or broth. The granules are then crushed between two sterile slides and examined for 1 Bostroem, Beitr. z. path. Anat. u. z. allg. Path., ix, 1890. 2 Wolff und Israel, Virch. Arch., 126, 1891. »/. H. Wright, Jour. Med. Res., viii, 1905. THE HIGHER BACTERIA 625 the presence of filaments. If these are present in reasonable abundance, the material is distributed in tubes of glucose agar, which are then allowed to solidify. If these first cultivations show a large number of contaminations, Wright recommends the preservation of other washed granules in test tubes for several weeks, in the hope that contaminating microorganisms may thus be killed by drying before the actinomyces lose their viability. If cultivation is successful colonies will ap- pear, after two to four days at 37.5° C, as minute white specks, which, in Wright^s cultures, ap- peared most abundantly within a zone situated 5 to 10 millimeters below the surface of the medium. Above and below this zone they are less numerous, indicating that a small amount of oxygen furnishes the best cultural environment. Upon the surface of agar slants, growth, if it takes place at all, is not luxuriant. In alkaline meat-infusion broth growth takes place in the form of heavy, flocculent masses which appear at the bottom of the tubes. Surface growth and clouding do not take place. Milk and potato have been used as culture media but are not particularly favorable. Pathogenicity. — As stated above, actinomy- cosis occurs spontaneously most frequently among cattle and human beings. It may also occur in sheep, dogs, cats, and horses. Its loca- tions of predilection are the various parts adjacent to the mouth and pharynx. It occurs also, however, in the lungs, in the intestinal canal, and upon the skin. When occurring in its most frequent location, the lower jaw, the disease presents, at first, a hard nodular swell- ing which later becomes soft because of central necrosis. It often involves the bone, causing a rarefying osteitis. As the sweUings break down, sinuses are formed from which the granular pus is discharged. The neighboring lymph nodes show painless, hard swell- ings. Histologically, about the filamentous knobs or granules, there is a formation of epithelioid cells and a small round-cell infiltration. In older cases there may be an encapsulation in connective tissue and a Fig. 142. — Actino- myces Granule Crushed Beneath a Cover-glass. .U n - ' stained. The prepara- tion shows the margin of the granule and the "clubs." (After Wright and Brown.) 626 PATHOGENIC klCROORGANISMS calcification of the necrotic masses, leading to spontaneous cure. As a rule, this process is extremely chronic. Infection in the lungs or in the intra-abdominal organs is, of course, far more serious. When death occurs acutely, it is often due to secondary infection. The disease is acquired probably by the agency of hay, straw, and grain. Berestnew ^ has succeeded in isolating actinomyces from straw and hay which he covered with sterile water in a potato jar and placed in the incubator. After a few days small white specks looking like chalk powder appeared upon the stalks, which, upon further cultivation, he was able to identify as the organism in question. Animal inoculation, carried out extensively both with pus and with pure cultures by several observers, has yielded little result. Progressive Fig. 143. — ^Branching Filaments of Actinomyces. (After Wright and Brown.) actinomycotic lesions were never obtained, although occasionally small knolls containing colonies surrounded by epithelioid cells and connective tissue were observed, showing that the invading microorganisms were able to survive and grow for a short time, but were not sufficiently virulent to give rise to an extensive disease process. Transmission from animal to animal, or from animal to man directly, has not been satis- factorily proven. Whether or not there are various forms of actinomyces must as yet be regarded as an open question. The investigations of Wolff and Israel, however, together with those of Wright, who alone observed thirteen different strains, seem to indicate that most, if not all, of the cases clinically observed are due to one and the same microorganism. » Berestnew, Ref. Cent. f. Bakt., 24, 1898. THE HIGHER BACTERIA 627 MYCETOMA (MADURA FOOT) The disease known by this name is not unhke actinomycosis. It is more or less strictly limited to warmer climates and was first recog- nized as a clinical entity, in India, by Carter.^ Clinically it consists of a chronic productive inflammation most frequently attacking the foot, less often the hand, very infrequently other parts of the body. Nodular swellings occur, which break down in their centers, leading to the formation of abscesses, later of sinuses. Often the bones are in- volved and a progressive rarefying osteitis results. From the sinuses a purulent fluid exudes, in which are found characteristic granular bodies. These may be hard, brittle, and black, resembling grains of gunpowder, or may be grayish-white or yellow and soft and grumous. According to the appearance of these granules, two varieties of the disease are dis- tinguished, the "melanoid" and the "ochroid.^' Many observers believe that the yellow or ochroid variety is, in fact, actinomycosis. The black variety, which is certainly a distinct disease, is caused by a member of the hyphomycetes group. The parasite has been carefully studied by Wright,^ from whose description the following points are taken : The small, brittle granules observed under the microscope show a dark, almost opaque center along the edges of which, filaments, or hyphse, may be seen in a thickly matted mass. By crushing the granules Ainder a cover-slip in a drop of sodium hypochlorite or of strong sodium hydrate, the black amorphous pigment is dissolved and the structural elements of the fungus may be observed. They seem to be composed of a dense meshwork of mycelial threads which are thick and often swollen, and show many branches. Transverse partitions are placed at short distances and the individual filaments may be very long. Spores were not observed by Wright. In a series of over fifty cultiva- tions on artificial media from the original lesion, Wright obtained growth in a large percentage. In broth, he obtained at first a rapid growth of long hyphse which eventually formed a structure which he compares in appearance to a powder-puff. On agar, growth appeared within less than a week and spread over the surface of the medium as a thick meshwork of spreading hyphae 1 Carter on My- ^ ifi- ^ - ^ c3 + + + I I 1= = = . 03 Heinemann, Jour, of Inf. Dis., 3, 1906. 702 BACTERIA IN Attl, ySOIL, WATER, AND MiLiC In consequence, they are able to persist and develop when cultural con. ditions are absolutely unsuited to other bacteria. Consequent upon acidification of the milk by lactic-acid formation, there is coagulation of casein. Casein precipitation, however, may also be due to a non-acid coagulation caused by a bacterial ferment. Casein precipitated in this way may be redissolved by a bacterial trypsin or casease, produced by the same or other bacteria, the milk again, becoming entirely liquid, transparent, and of a yellowish color. The casein precipitated by lactic-acid formation, however, is never thus redissolved, because the high acidity does not permit the pro- teolytic ferments to act.^ Butyric-acid fermentation in milk, a common phenomenon, is also an evidence of bacterial growth. As a rule, it is produced by the anaerobic bacteria, and is a process developing much more slowly than other fer- mentations. A large number of bacteria have been described which are capable of producing such changes, the chemical process by which they are produced being, as yet, not entirely understood. It is probable that the process takes place after hydrolysis of the disaccharid some- what according to the following formula: Ce B.,2 Oe = C4 Hs O2 + 2 CO2 + 2 H2. Special bacteria have been described in connection with this form of milk fermentation,^ most of them non-pathogenic. It is unquestionable, however, that many of the well-known pathogenic bacteria, such as Bacillus aerogenes capsulatus. Bacillus oedematis • maligni, possess the power of similar butyric-acid formation. While less commonly observed in milk, because milk is rarely kept long enough to permit of the action or development of these enzymes, the butyric-acid fermentation is of importance in connection with butter, where it is one of the causes pro- ducing rancidity. Alcoholic fermentation may take place in milk as a result of the ac- tivities of certain yeasts. Upon the occurrence of such fermentations depends the production of kefyr, koumys, and other beverages which have been in common use for many years, especially in the region of the Caucasus. The characteristic quality of these beverages is contrib- uted by the feeble alcoholic fermentation produced by members of the saccharomyces group, but side by side with this process lactio-acid forma- » Conn, Exper. Stat. Rep., 1892. a Schattenfroh uud Grasberger, Arch. f. Hyg., 37, 1900. BACTERIA IN MILK 703 tion also takes place. Beijerinck/ who has carefully studied the so- called kefyr seeds, used for the production of kefyr in the East, has isolated from them a forrn of yeast similar in many respects to the ordinary beer yeast, and a large bacillus to which he attributes the lactic-acid formation. Occasional but uncommon changes which occur in milk lead to the formation of the so-called " slimy milk," yellow and green milk, and bitter milk. These may be due to a number of bacteria. A microorgan- ism which is commonly found in connection with the sHmy changes in milk is the so-called Bacillus lactis viscosus. According to the researches of Ward,^ this microorganism is frequently derived from water and it is the water supply which should attract attention whenever such trouble occurs in dairies. The so-called blue, green, and yellow changes are usually due to chromogenic bacteria, such as Bacillus cyanogenes. Bacillus prodigiosus, and others. "Bitter milk," a condition which has occasionally been observed epi- demically, is also the consequence of the growth of microorganisms. Conn,^in 1891, isolated from a specimen of bitter cream a diplococcus which occasionally forms chains and which in sterilized milk develops rapidly, producing an extremely bitter taste. The organism of Conn differs from a similar diplococcus described by Wagmann ^ in that it possesses the ability of producing butyric acid. Milk in Relation to Infectious Disease. — As a source of direct in- fection, milk is second only to water, and deserves close hygienic at- tention. A large number of infectious diseases have been traced to milk, although the actual proof of the etiological part played by it in such cases has often been difficult to adduce and has necessarily been indirect. Nevertheless, even when indirect proof only has been brought, it has been sufficiently convincing to necessitate the most careful investigation into milk supplies whenever epidemics of certain infectious maladies occur. Typhoid-fever epidemics have been frequently traced to milk in- fection, and, in this disease, milk is, next to water, the most frequent etiological factor. Schiider,^ in an analysis of six hundred and fifty typhoid epidemics, found four hundred and sixty-two attributed to > Beijerinck, Cent. f. Bakt., vi, 1889. 2 Ward, Bull. 165, Cornell Univ. Agri. Exp. Stat., 1899, » C(mn, Cent. f. Bakt., ix, 1891. » Wagmann, Milchztg., 1890. * Schuder, Zeit. f. Hyg., xxxviii, 1901. 46 704 ' BACTERIA IN AIR, SOIL, WATER, AND MILK water, one hundred and ten to milk, and seventy-eight to aii other causes. Trask ^ compiled statistics of one hundred" and seventy-nine typhoid epidemics supposed to have been caused by milk, in various parts of the world. In all such epidemics the origin of infection was generally trace- able to diseased or convalescent persons employed in dairies, to con- taminated well water used in washing milk utensils, or to the use of cans and bottles returned from dwellings where typhoid fever had existed. Actual bacteriological proof of the infectiousness of milk by the isolation of Bacillus typhosus is rare, but has been accomplished in isolated in- stances. In the case of one epidemic, Conradi ^ isolated the bacillus from the milk on sale at a bakery at which a large number of the infected individuals had purchased their milk. The examination of market milk at Chicago, through a period of eight years, revealed the presence of typhoid bacilli but three times. In spite of the few cases in which actual bacteriological proof has been brought, it is not unlikely that careful and systematic researches would reveal a far greater number, since many writers have shown that typhoid bacilli may remain alive in raw milk for as long as thirty day^,^ and may actively proliferate in the milk during this time. One peculiarity of epidemics which may aid in arousing the suspicion that they have originated in milk is that, in such cases, women and children are far more frequently attacked than men.^ A feature which adds considerably to the dangers of milk infection is the unfortunate absence of any gross changes, such as coagulation, by the growth of typhoid bacilli. Scarlet fever,^ though as yet of unknown etiology, has in many cases been traced indirectly to milk infection. Trask has collected fifty-one epidemics of scarlet fever presumably due to, milk. In one epidemic occurring in Norwalk, Conn.,® twenty-nine cases were distributed among twenty-five families living in twenty-four different houses. The indi- viduals affected did not attend the same school, and were of entirely different social standing, the only factor common to all of them being the milk supply. 1 Trask, Bull. No. 41, U. S. Pub. Health and Mar. Hosp. Serv., Wash. 2 Conradi, Cent. f. Bakt., I, xl, 1905. » Heim, Arb, a. d. kais. Gesundheitsamt, v. • Wilckens, Zeit. f. Hyg., xxvii, 1898. • Trask, loc. cit. • Herbert E. Smith, Rep. Conn. State Bd. of Health, 1897. BACTERIA IN MILK 705 Diphtheria has been frequently traced to the use of infected milk. In most of the epidemics reported as originating in this way, the proof has been necessarily indirect. In two out of twenty-three epidemics reported by Trask, however, Bacillus diphtherise was isolated from the milk directly. The ability of the Klebs-Loeffler bacillus to proliferate and remain aHve for a long while in raw milk has been demonstrated by Eyre ^ and others. Whether or not cholera asiatica may be transmitted by means of milk has been a disputed question. Hesse ^ claims that cholera spirilla die out in raw milk within twelve hours. This statement, however, has not been borne out by other observers.^ Unquestionable cases of direct transmission of cholera by means of milk have been reported by a num- ber of writers, notably by Simpson."* The relation of milk to the diarrheal diseases of infants has, of late years, received a great deal of attention. In large cities, during the summer months, numerous cases of infantile diarrhea among bottle- fed babies occur, which, in many instances, are attributed to feeding with contaminated milk. Park and Holt,^ who have made extensive re- searches upon this question in New York City, have come to the con- clusion that the harmful effects of contaminated milk upon babies can not be ascribed to any given single microorganism in the milk. Specifically toxic properties were found by these writers for none of the one hundred and thirty-nine different species of bacteria isolated from unsterilized milk. It is unlikely, therefore, that the diarrheal diseases among babies have a uniform bacteriological cause. Whether or not these diarrheal conditions depend entirely upon the bacterial contents of milk or, in a large number of cases at least, upon the inabiHty of the child to digest cows' milk because of chemical conditions, must be left undecided. Park and Holt, in analyzing their extensive data, conclude that milk containing " over one million bacteria to the cubic centimeter is certainly harmful to the average infant." The significance of the presence of streptococci in milk, as an element of danger, has recently received much attention in the literature. Heine- mann,® who has made a careful comparison of Streptococcus lacticus ' Eyre, Brit. Med. Jour., 1899. ^ Hesse, Zeit. f. Hyg., xvii, 1894. \Basenau, Arch. f. Hyg., xxiii, 1895. * Simpson, Indian Med. Gaz., 1887. » Park and Holt, Arch, of Fed., Dec, 1903. * Heinemann, Jour. Inf. Dis., 3, 1906. 706 BACTERIA IN AIR, SOIL, WATER, AND MILK (formerly spoken of as Bacillus acidi lactici [Kruse]), with other, strep- tococci, has shown that, essentially, this streptococcus does not differ from streptococci from other sources, and is practically indistinguish- able by cultural methods from Streptococcus pyogenes. Similar com- parisons made by Schottmiiller,^ Muller,^ and others have led to like re- sults. Since streptococci may be found in milk from perfectly normal cows and are almost regularly associated with lactic-acid fermenta- tion, it is unlikely that these microorganisms hold ordinarily any specific relationship to disease. Recently, however, a number of epidemics of sore throat caused by streptococci have been traced to milk upon reasonably reliable evidence. Accounts of such epidemics in Chicago and in Baltimore have been published by Capps and Miller ^ and by Hamburger.^ The presence of pus cells and leucocytes in milk, together with streptococci, was also formerly regarded as of great importance. Enumerations of leucocytes in milk were first made by Stokes and Weggefarth.^ Their method of enumeration consisted in centrifugaliz- ing a definite volume of milk, spreading the entire sediment over a definite area on a slide, and counting the leucocytes found in a number of fields. Calculations from this may then be made as to the number of leucocytes per cubic centimeter. This method, and modifications of it, have been used by a large number of observers, but the value of the con- clusions drawn from them has been much exaggerated. Normal milk may contain leucocytes in moderate numbers, and importance may be attached to such leucocyte counts only when their number largely ex- ceeds that present in other specimens of perfectly normal milk. When- ever such high leucocyte counts are found, of course, a careful veteri- nary inspection and examination for pyogenic disease should be made. Foot-and-mouth disease, an infectious condition prevailing among cattle, characterized by a vesicular rash on the mouth and about the hoofs, has, in a number of cases, been definitely shown to be transmitted to man through the agency of milk. Notter and Firth^ reported an epidemic occurring among persons supplied with milk from a single dairy in which foot-and-mouth disease prevailed among the cows. In this 1 Schottmuller, Mtinch. med. Woch., 1903. 2 Milller, Arch. f. Hyg., Ivi, 1906. 3 Cap-ps and Miller, Jour. A. M. A., June, 1912, p. 1848. ^ Hamburger, Bull, of the Johns Hopk. Hosp., xxiv, Jan., 1913. 5 Stokes and Weggefarth, Med. News, 91, 1897. ^ Notter and Firth, quoted from Harrington, "Theory and Practice of Hygiene." BACTERIA IN MILK 707 epidemic, two hundred and five individuals were affected with vesic- lar eruptions of the throat, with tonsilHtis and swellings of the cervical lymph nodes. Similar cases have been reported by Pott.^ > Although anthrax has never been definitely shown to have been conveyed by milk, Boschetti ^ succeeded in isolating hving anthrax bacilli from a sample of milk two weeks after its withdrawal from the cow. Milk and Tuberculosis. — The question of the conveyance of tuber- culosis by means of milk is a subject which, because of its great im- portance, has been extensively investigated by bacteriologists. A large number of observers have succeeded in proving the presence of tubercle bacilli in the milk of tuberculous cows by intraperitoneal in- oculation of rabbits and guinea-pigs with samples of milk. Such posi- tive results have been obtained by Bang,^ Hirschberger,"^ Ernst, ^ and many others. A number of these observers, notably Ernst, have shown that tubercle bacilli may be present in the milk without tuberculous dis- ease of the udders. In an examination of the milk supply of Washington, D. C.,^ 6.72 per cent of the samples contained tubercle baciUi. The path of entrance of the bacilli from the cow into the milk has long been a subject of controversy. That the bacilli may easily enter the milk, when tuberculous disease of the udder is present, stands to reason and is universally conceded. It is now believed also, on the basis of much experimentation, that in systemically infected cows tubercle bacilli may pass through the mammary glands into the milk, without evidence of local disease in the secreting gland. An experi- ment performed by the Royal British Tuberculosis Commission ^ illus- trates this point. A cow, injected subcutaneously with tubercle bacilli behind the shoulder, began to discharge tubercle bacilli in the milk within seven days after inoculation and continued to do so until death from generalized tuberculosis. Milk may become indirectly contaminated, furthermore, with tubercle bacilli emanating from the feces of cows. It has been shown that tubercle bacilli are present in the feces of cattle so early in the disease that diagnosis can be made only by a tubercuUn test.^ Whether or not contaminated milk is common as an etiological 1 Pott, Munch, med. Woch., 1899. ^ Boschetti, Giorn. med. vet., 1891. 3 Bang, Deut. Zeit. f . Tierchem., xi, 1884. ^ Hirschberger, Deut. Arch, f . klin. Med., xliv, 1889. 5 Ernst, H. C, Amer. Jour. Med. Sci., xcviii, 1890. 6 Anderson, BuU. No. 41, U. S. Pub. Health and Mar. Hosp. Serv., Wash., 1908. 7 Quoted from Mohler, P. H., and Mar. Hosp. Serv. Bull. 41, 1908. • Schroeder and Cotton, Bull. Bureau Animal Industry, Wash., 1907. 70^ BACTERIA m AIR, SOIL, WATER, AND MlLK factor in human tuberculosis, must be con^dered at present as an ui*- settled question. Behring, at the Congress of Veterinary Medicine, at Cassel, in 1903, advanced the view that pulmonary tuberculosis in adults may be a late manifestation of a milk infection contracted dur- ing infancy. He stated as his own opinion, moreover, that most cases of tuberculosis in man are traceable to this origin. The problem is as difficult of solution as it is important. In bottle-fed infants, infection by means of milk unquestionably occurs with considerable frequency. Smith,^ Kossel, Weber, and Huess,^ and others, have isolated tubercle bacilli of the bovine type from the mesenteric lymph nodes of many infected children. Animal experimentation has, furthermore, revealed that lesions in the mesenteric nodes, as well as later in the bronchial lymph nodes, may occur as a consequence of feeding tubercle bacilli, without any demonstrable lesions in the intestinal mucosa. It is thus certain that infection by the ingestion of tuberculous milk may occur, especially among young children who, as is well-known, are com- paratively susceptible to bacilli of the bovine type. Whether or not such infection will account for many cases of tuberculosis in adults is a ques- tion which, for final solution, will require much more investigation. The sole reliable method of approaching it lies in determining the type, human or bovine, of the bacilli present in a large number of cases. Ex- perience thus far seems to indicate that the bovine type is comparatively rare in the pulmonary disease of adults. The value of the tuberculin reaction for diagnosis, and the elimination of all cattle showing a positive reaction, for the prevention of tubercu- , losis, can not be overestimated. The failure of the test in diseased animals is rare, and an accurate diagnosis can be established in over 90 per cent of diseased animals.^ The assertion that the cattle are permanently injured by tuberculin injections is without scientific basis. If this test were conscientiously carried out, and infected cattle elim- inated, the dangers from bovine bacillus infection would be practically eliminated, for there are but few instances in which science has been able to furnish such definite information for absolute protection. It is need- less to say, however, that the carrying out of such precautions is subject to great expense and great difficulties of organization. Dairy inspection is practiced in the vicinity of many of our larger 1 Smith, Trans. Assn. Amer. Physic, 18, 1903. ^Kossel, Weber, and Huess, Tuberkul. Arb. a. d. kais. Gesundheitsamt, 1904, 1905, Hft. 1 and 3. » MoMer, loc. cit. BACTERIA IN MILK 709 cities, and the movement is daily gaining ground. Until fully estab- lished, however, upon a financial basis which brings the best products within the means of the poorer classes, other inexpensive measures to render milk safe must often be resorted to. Sterilization by high temperatures is objected to by pediatricians because of the physical and chemical changes produced in the milk which are said to detract from its nutritive value. The development of scurvy and rickets in infants has often been attributed to the use of such milk. These objections, however, do not apply to the use of milk which has been subjected to the process of "pasteurization." By this term is meant the heating of any substance to 60° C. for twenty to thirty minutes. The process, first devised by Pasteur for the purpose of destroying germs in wine and beer in which excessive heating was supposed to injure flavor, brings about the death of all microorganisms which do not form spores — in other words, of all the bacteria likely to be found in milk which can give rise to infection per OS. At the same time the chemical and physical constitution of the milk is not appreciably changed, at least not to an extent which renders it less valuable as a food. Statistics by Park and Holt ^ have shown strikingly the advantages of pasteurized over raw milk in infant feed- ing. Of fifty-one children fed with raw milk during the summer months, thirty-three had diarrhea, two died, and only seventeen remained entirely well. Of forty-one receiving pasteurized milk, but ten had diarrhea, one died, and thirty-one remained entirely well throughout the summer. The actual diminution of the living bacterial contents of milk by pasteurization is enormous, the milk so treated often con- taining not more than one thousand, usually less than fifteen thou- sand, living bacteria to each cubic centimeter. Methods of Estimating the Number of Bacteria in Milk. — In estimating the number of bacteria in milk, colony counting in agar or gelatin plates is resorted to. Great care must be exercised in obtaining the specimens. If taken from a can, the contents of the can should be thoroughly mixed, since the cream usually contains many more bacteria than the rest of the milk. The specimen io then taken into a sterile test tube or flask. If the milk is supplied in an ordinary milk bottle, this should be thoroughly shaken before being opened, and the specimen for exam- ination taken out with a sterile pipette. Dilutions of the specimen can then be made in sterile broth or salt solution. If an initial dilution » Park and Holt, loc. cit. 710 BACTERIA IN AIR, SOIL, WATER, AND MILK of 1 : 100 is made, quantities ranging from 1 c.c. to 0.1 c.c. of this will furnish 0.01 c.c. to 0.001 c.c. of the milk, respectively. Inoculation of properly cooled tubes of melted neutral agar and gelatin, with varying quantities of these dilutions, are then made and plates poured. After twenty-four to forty-eight hours at room temperature or in the in- cubator, colony counting is done as described upon page 161, and the proper calculation is made. In samples in which few bacteria are ex- pected, direct transference of 1/20 or 1/40 of a c.c. of the whole milk into the agar may be made. This method saves time but is less accurate. Direct Methods of Counting Bacteria. — Direct methods of counting bacteria in milk have recently been advised, the one most extensively tried being that of Prescott and Breed. By this method a capillary tube is marked to measure accurately 0.01 c.c. This amount of the milk is spread over a square cm. on a microscope slide. It is dried in the air and fixed with methyl alcohol, after which the fatty constituents can be dissolved with xylol. It can then be stained lightly with the Jen- ner stain. The bacteria are counted under an oil immersion lens, the tube length and magnification being so arranged that the microscopic field covers 1/50 sq. mm. A standardized eyepiece micrometer may be used. The average number of bacteria found in such fields may be multiplied by 5,000 to give the number of bacteria contained in 0.01 c.c. of milk. This method has not yet displaced the one of plating and does not promise to do so for some time. For the isolation of special pathogenic bacteria from milk, no rules can be laid down, since, in every case, the method adapted to the. par- ticular organism sought for must be chosen. Tubercle bacilli can be isolated from milk with success only by guinea-pig injection. The milk is centrifugalized and 5 c.c. of the sediment, together with some of the cream that has risen to the top, is intraperitoneally or subcutaneously injected. The control of milk in the market depends upon careful legulations, which must include care of cattle, dairy inspection and bacteriological control of the delivered milk. This is a subject which is too extensive to touch upon in a book of this kind. However, a general idea of the methods employed may be obtained by studying the accompanying table, which is taken from the New York City Department of Health Regulation for the Sale of Milk and Cream. Bacteria and Butter. — Butter is made from cream separated from milk either by standing or by centrifugalization. After this, the cream is agitated by churning, which brings the small fat-globules into mutual BACTERIA IN MILK 711 contact, allows them to adhere to each other and form clumps of butter. It has been a matter of common experience, however, that unless the cream is allowed to ** ripen" for a considerable period before churning, the resulting butter lacks the particular quality of flavor which gives it its market value. The interval of ripening, at first a necessity upon small farms where cream must be collected and allowed to accumulate, has now been recognized as an essential for the production of the best grades of butter, and it has been shown that the changes taking place in the cream during this period are referable to the action of bacteria. Cream, which before the ripening process contains but 50,000 bacteria to each cubic centimeter, at the end of a period of "ripening" will often contain many millions of microorganisms. At the same time, the cream becomes thick and often sour. The species of bacteria which take part in this process and which, therefore, must determine to a large extent the quality of the end prod- uct, are various and, as yet, incompletely known. Usually some variety of lactic-acid bacilli is present and these, as in milk, outgrow other species and, according to Conn,^ are probably essential for "ripening." It would be of great practical value, therefore, if definite pure cul- tures of the bacteria which favor the production of agreeable flavors could be distributed among dairies. In Denmark this has been attempt- ed by first pasteurizing the cream and then adding a culture of bacteria isolated from "favorable" cream. These cultures, delivered to the dairyman, are planted in sterilized milk, in order to increase their quan- tity, and this culture is then poured into the pasteurized cream. In most cases, these so-called "starters" are not pure cultures, but mix- tures of three or more species derived from the original cream. Adverse accidents in the course of butter-making, such as "souring" or "bittering" of butter, -are due to the presence of contaminating, probably proteolytic, microorganisms in the cream during the process of "ripening." As a means of transmitting infectious disease, butter is of importance only in relation to tuberculosis. Obermuller,^ Rabinowitch,^ Boyce,'^ and others, have repeatedly found tubercle bacilli in market butter, and Mohler,^ Washburn, and Rogers have recently shown that these bacilli 1 Crnin, "Agricultural Bacteriology," Phila., 1901. 2 Obermiiller, Hyg. Rundschau, 14, 1897. ' Rabinotoitch, Zeit. f. Hyg., xxvi, 1897. * Boyce and Woodhead, Brit. Med. Jour., 2, 1897. » Mohler, U. S. P. H. and Mar. Hosp. Serv. Bull. 41, 1908. 712 BACTERIA IN AIR, SOIL, WATER, AND MILK REGULATIONS GOVERNING THE GRADES AND DESIGNATION OF MILK The following classifications apply to milk and cream. The regulations regarding m GRADES OF MILK OR CREAM WHICH MAY BE SOLD IN THE CITY OF NEW.YORK DEFINITION TUBERCULIN TEST AND PHYSICAL CONDITION BACTERIAL CONTENTS GRADE A Milk or Cream (Raw) Grade A milk or cream (raw) is milk or cream produced and handled in accord- ance with the minimum requirements, rules and regulations as herein set forth. 1. Only such cows shall be ad- mitted to the herd as have not reacted to a diagnostic injec- tion of tubercu- lin and are in good physical condition. 2. All cows shall be tested annually with tuberculin and all reacting ani- mals shall be ex- cluded from the herd. Grade A milk (Raw) shall not con- tain more than 60,000 bacteria per c.c. and cream more than 300,000 bacteria per c.c. when delivered to the consumer or at any time prior to such delivery. Milk or Cream (Pasteurized) Grade A milk or cream (pasteurized) is milk or cream handled and sold by dealers holding permits therefor frpm the Board of Health, and produced and handled in accordance with the requirements, rules and regulations as herein set forth. No tuberculin test required but cows must be healthy as dis- closed by physi- cal examination made annually. Grade A milk (pasteurized) shall not contain more than 30,000 bac- teria per c.c. and Cream (pasteurized) more than 150,000 bacteria per c.c. when delivered to the consumer or at any time after pasteurization and prior to such delivery. No milk supply averaging more than 200,000 bacteria per c.c. shall be pasteurized for sale under this designation. GRADE B Milk or Cream (Pasteurized) Grade B milk or cream (pasteurized) is milk or cream produced and handled in accordance with the minimum require- ments, rules and regulations herein set forth and which has been pasteurized in accordance with the requirements and rules and regulations of the Department of Health for pasteurization. No tuberculin test required but cows must be healthy as dis- closed by physi- cal examination made annually. No milk under this grade shall contain more than 100,000 bacteria per c.c. and no cream shall contain more than 500,000 bacteria per c.c. whep delivered to the consumer or at any time after pasteurization and prior to such delivery. * No milk supply averaging more than 1,500,000 bacteria per c.c. shall be pasteurized in this city for sale under this designation. No milk supply averaging more than 300,000 bacteria per c.c. shall be pasteiu-ized outside of this city for sale under this designation. GRADE C Milk or Cream (Pasteurized) (For cooking and manu- facturing pur- poses only). Grade C milk or cream is milk or cream not conforming to the requirements of any of the subdivisions of Grade A or Grade B and which has been pasteiu-ized according to the requirements and rules and regulations of the Board of Health or boiled for at least two (2) minutes. No tuberculin test required but cows must be healthy as dis- closed by physi- cal examination made annually. No milk of this grade shall contain more than 300.000 bacteria per c.c. and no cream of this grade shall con- tain more than 1,500,000 bacteria per c.c. after pasteurization. NOTE — Sour milk, buttermilk, sour cream, kumyss, matzoon, zoolac and similar products shall not be made the process of souring. Sour cream shall not contain a less percentage of fats than that designated for cream. No other words than those designated herein shall appear on the label of any container containing milk or cream The term " certified " milk is usually defined for each region by a special commission of the County Milk Commission of Kings County, N. Y.: Certified Milk must have every characteristic of pure, clean, fresh, wholesome cow's milk. The or preservatives. Nothing must be added to the milk and nothing taken away. Certified Milk shall not contain less than 4 per cent of butter fat. » Table taken from Rules and Regulations of N. Y. City Department of Health, 1914, applying to BACTERIA IN MILK AND CREAM WHICH MAY BE SOLD IN THE CITY OF NEW YORRi bacterial content and time of delivery shall not apply to sour cream 713 NECES- SARY SCORES FOR DAIRIES PRODUC- ING TIME OF DELIVERY BOTTLING LABELING PASTEUR. IZATION Equip. 25 Meth. 50 Total 75 Shall be de- livered within 36 hours after production. Unless other- wise specified in the permit this milk or cream shall be delivered to the consumer only in bottles. Outer caps of bottles shall be white and shall contain the words Grade A, Raw, in black letters in large type, and shall state the name and address of the dealer. Equip. 25 Meth. 43 Total 68 Shall be de- livered within 36 hours after pasteurization. Unless other- wise specified in the permit this milk or cream shall be delivered to the consumer only in bottles. Outer caps of bottles shall be white and shall contain the words Grade A in black letters in large type, date and hours be- tween which pasteurization was com- pleted; place where pasteurization was performed; name of the person, firm or corporation offering for sale, selling or delivering same. Only such milk or cream shall be re- garded as pas- teurized as has been subjected to a tempera- ture averaging 145° Fahr. for not less than 30 minutes. Equip. 20 Meth. 35 Total 55 Milk shall be delivered within 36 hours and cream within 48 hours after pas- teurization. May be deliv- ered in cans or bottles. Outer caps of bottles containing milk and tags affixed to cans containing milk or cream shall be white and marked "Grade B" in bright green letters in large type, date pasteurization was com- pleted, place where pasteurization was performed, name of the person, firm or corporation offering for sale, selling or de- livering same. Bottles containing creams shall be labeled with caps marked "Grade B" in bright green letters, in large type and shall give the place and dat« of bot- tling and shall give the name of person, firm or corporation offering for sale, sell- ing or delivering same. Only such milk or cream shall be re- garded as pas- teurized as has been subjected to a tempera- ture averaging 145° Fahr. for not less than 30 minutes. Score 40 • Shall be de- livered within 48 hours after pasteurization. May be deliv- ered in cans only. Tags affixed to cans shall be white and shall be marked in red with the words "Grade C" in large type and "for cook- ing" in plainly visible type, and cans shall have properly sealed metal collars, paint- ed red on necks. Only such milk or cream shall be re- garded as pas- teurized as has been subjected to a tempera- ture averaging 145° Fahr. for not less than 30 minutes. from any milk of a less grade than that designated for " Grade B " and shall be pasteurized before being put through or milk or cream products except the word "certified" when authorized under the State laws. Med. Soc, sanctioned by State Law. The following is the definition of certified milk given by the milk must be in its natural state, not having been heated and without the addition of coloring matter sale of milk and cream. 714 BACTERIA IN AIR, SOIL, WATER, AND MILK could remain alive and virulent for as long as five months in butter kept at refrigerator temperature. The acid-fast butter bacillus, described by Rabinowitch as similar to the true Bacillus tuberculosis, shows decided cultural and morphological differences from the latter. Bacteria and Cheese. — The conversion of milk products into cheese consists in a process of proteid decomposition which, by its end products, leucin, tyrosin, and ammonia compounds, largely determines the cheese- flavors. The production of cheese, therefore, is due to the action of proteolytic bacterial enzymes ^ and the variety of a cheese is largely determined by the microorganisms which are present and by the cul- tural conditions prevailing. The sterilization of cream, or the addition of antiseptics, absolutely prevents cheese production. The organisms which are concerned in such processes have been ex- tensively studied and attempts have been made, with moderate success, to produce a definite flavor with pure cultures. In the production of cheese the two varieties, hard and soft cheeses, depend not so much upon the bacterial varieties as upon the differences in the treatment of the curds before bacterial action has begun. In the former case, a complete freeing of the curds from the whey furnishes a culture medium which is comparatively dry and of almost exclusively proteid composition; in the latter, retention of whey gives rise to cul- tural conditions in which more rapid and complete bacterial action may take place. The holes, which are so often observed in some of the hard cheeses, are due to gas production during the process of ''ripening." As to the varieties of microorganisms present in various cheeses, much careful work has been done. Duclaux ^ attributed the ''ripening" of some of the soft cheeses to a microorganism closely related to Bacillus subtilis. V. Freudenreich ^ in part substantiated this, but laid particular stress upon the action of Oi'dium lactis, a mold, and upon several vari- eties of yeast. Conn,^ more recently, in a bacteriological study of Cam- embert cheese, has demonstrated that the production of this cheese depends upon the united action of two microorganisms, one an oidium, like the Oidium lactis of Freudenreich, which is found chiefly in the interior softened areas, the other a mold belonging to the penicillium variety, found in a matted felt-work over the surface and penetrating but a short distance. In spite of the scientific basis upon which the 1 Freudenreich, Koch's Jahresbericht, etc., 135, 1891. 2 Duclaux, "Le Lait," Paris, 1887. 3 V. Freudenreich, Cent. f. Bakt., II, i, 1895. * Conn, BuU. Statis. Agri. Exp. Stat. 35, 1905. LACTIC-ACID BACILLI 7lS work of these men and of others has seemed to place cheese production, attempts at uniformity in cheese production have met with almost in- superable obstacles because of the presence of a variety of adventitious microorganisms which, depending in species and proportion upon the local conditions under which the various cheeses have been produced, have added minor characteristics of flavor which have determined mar- ket value. Occasional failure of good results in cheese production ^ is due to contamination with other chromogenic or putrefactive bacteria. In its relationship to the spread of infectious disease, cheese is rela- tively unimportant except in regard to tuberculosis. Typhoid and other non-spore forming pathogenic germs can not survive the condi- tions existing during cheese-ripening for any length of time. Tubercle bacilli, both of the human and bovine types, have been found in cheese by Harrison ^ and others, and Galtier has shown experimentally that tubercle bacilli may remain alive and virulent in both salted and un- salted cheese for as long as ten days. THE LACTIC-ACID BACILLI AND METCHNIKOPP'S BACTERIO- THERAPY A problem which has occupied clinical investigation for many years is that of gastrointestinal autointoxication. There are a number of conditions occurring in man, in which symptoms profoundly affecting the nervous system, the circulation, and, in a variety of ways, the entire body, can be clinically traced to the intestines, and can, in many cases, be relieved by thorough purgation and careful diet. In some of these conditions, specific microorganisms can be held accountable for the diseases (B. enteritidis, B. botulinus, etc.). In other cases, however, etiological investigations have met with but partial success because of the large variety of microorganisms present in the intestinal tract and because of the complicated symbiotic conditions thereby produced. Intestinal putrefaction, recognized as the cardinal feature of such maladies, has been attributed to Bacillus proteus vulgaris,^ to Bacillus aerogenes capsulatus, to Bacillus putrificus,^ and to a number of other bacteria, but definite and satisfactory proof as to the etiological im- portance of any of these germs has not yet been advanced. The fact 1 Beijerinck, Koch's Jahresber, etc., 82, 189, 2 Harrison and Galtier, quoted from Mohler, U. S. Pub. H. and Mar. Hosp. Serv., Hygiene Lab. Bull. 41, 1908. ' Lesage, Rev. de m^d., 1887. * Tissier. Ann. de Tinst. Pasteur, 1905. 716 BACTERIA IN AIR, SOIL, WATER, AND MILK remains, however, that, whatever may be the specific cause, the disease itself, a grave and often fatal affliction, may be clinically traced, in a number of cases, to the absorption of poisons from the intestinal canal, and it is more than likely that these poisons are the products of bacterial activity. Reason dictates, furthermore, that the bacteria primarily responsible for the production of these toxic substances do not belong to the varieties which attack carbohydrates only, but must belong to that class of aerobic and anaerobic germs which possess the power of breaking up proteids — in other words, the bacteria of putrefaction. On the basis of the mutual antagonism existing in culture between many acid-producing bacteria and those of putrefaction — a phenomenon recognized by some of the earliest workers in this field, many investiga- tors have suggested the possibility of combating intestinal putrefaction by adding acid-forming bacteria together with carbohydrates to the diet of patients suffering from this condition. The first to suggest this therapy was Escherich ^ who proposed the use, in this way, of Bacillus lactos aerogenes; with the same end in view, Quincke,^ a little later, suggested the use of yeasts — Oidiiun lactis. The reasoning underlying these attempts was meanwhile upheld by experiments carried out both in vitro and upon the living patient. Thus Brudzinski ^ was able to demonstrate that Bacillus lactis aerogenes, in culture, inhibited the development of certain races of the proteus species and succeeded in obtaining markedly favorable results by feeding pure cultures of Bacillus lactis aerogenes to infants suffering from fetid diarrhea. Similar ex- periments ^ carried out with the Welch bacillus (aerogenes capsulatus) and Bacillus coli, however, had no such corroboratory results, since this anaerobe possesses a considerable resistance against an acid reaction. In considering the difficulties of the problems involved in this question, it occurred to Metchnikoff ^ that much of the practical failure of therapy, based upon the principles stated above, might be referred to insufficient powers of acid production on the part of Bacillus coli. Bacillus lactis aerogenes, and other germs previously used. In searching for more pow- erful acid producers, his attention was attracted to Bacillus bulgaricus, ^ Escherich, Therapeut. Monatshefte., Oct., 1887. 2 Quincke, Verhandl. des Congress f . Inn. Med., Wiesbaden, 1898. • ^ BrudJinski, Jahrbuch f. Kinderheilkunde, 52, 1900 (Erganzungshef t) . ^ Tissier and Martelly, Ann. de I'inst. Pasteur, 1906. ^ Metchnikoff, "Prolongation of Life," G. P. Putnam's Sons, N. Y.; also in "Bac- t^riotherapie," etc. " Bibliotheque de th^rapeutique," Gilbert and Carnot, Paris, 1909. BACTERIA IN THE INDUSTRIES 717 isolated from milk by Massol ^ and Cohendy ^ in 1905. This bacillus, according to the researches of Bertrand and Weisweiller,^ produces as much as 25 grams of lactic acid per liter of milk. In addition to this, it manufactures, from the same quantity of milk, about 50 centigrams of acetic and succinic acids and exerts no putrefactive action upon pro- FiG. 155. — Bacillus bulgarictjs. teids. Added to these characters, it is especially adapted to therapeutic appUcation by its complete lack of pathogenicity. The administration of the bacillus to patients suffering from intestinal putrefaction, first suggested by Metchnikoff in 1906, has, since that time, been extensively practiced and often with remarkable success. In spite of sharp criticism, especially by Luersen and Kiihn,^ who deny much of 1 Massol, Revue medicale de la Suisse romande, 1905. 2 Cohendy, Comptes rend, de la soc. de biol., 60, 1906. ' Bertrand and Weisweiller, Ann. de I'inst. Pasteur, 1906. * Luersen and Kilhn, Cent, f . Bakt., II, xx, 1908. 718 BACTERIA IN AIR, SOIL, WATER, AND MILK the antiputrefactive activity of the bacillus, the treatment of Metch- nikoff has found many adherents, upon the basis of purely clinical ex- periment. It is not possible to review completely the already extensive literature. Among the more valuable contributions may be mentioned the articles by Grekoff,^ by Wegele,^ and by Klotz.^ In Metchnikoff's experinients and in the work of his immediate successors, the bacillus was used either in milk culture or in broth in which it was induced to grow in symbiosis with other microorganisms. Recently, North ^ has suggested'^^he use of Bacillus bulgaricus in parts of the body other than the digestive tract. His work was made feasible by the discovery that the bacillus could be cultivated in dex- trose-pepton broth to which calcium carbonate has been added, after the manner recommended by Hiss. With such cultures, applied in the form of a spray, inflammations of the ear, nose, throat, genitourinary tract, etc., have been treated, many of them with success. BACTERIOLOGICAL EXAMINATION OF OYSTERS On accoimt of the danger of the transmission of typhoid fever by. oysters which have been bred or stored in contaminated water, stand- ard methods ^ have been devised for the estimation of the bacterial con- tent of oysters. These are similar in principle and method to those used for the examination of water, and a most important index of sewage contamination and consequent danger of typhoid infection is *the number of colon bacilli present in the shell fish. The shell liquor is used for examination, and in examining oysters in the shell the fol- lowing procedure is followed: Five oysters having deep bowls and closed shells are selected. Lips of the shell are sterilized in the flame or by burning with alcohol. The liquor is obtained by opening the shell with a sterilized knife, or better, by drilling a hole through the flame surface with a sterilized gimlet. For determining the total num- ber of bacteria the shell liquor is withdrawn with a sterilized pipette, diluted with 1 per cent salt solution, and placed in agar, as described on page 693. More important, however, is the presumptive colony test, which is carried out by inoculating three lactose bile tubes with ^Grekoff, "Observations cliniques sur I'effet du lact. agri.," etc., St. Petersburg, 1507. ^ Wegele, Deut. med. Woch., xxxiv, 1908. ' Klotz, Zentralbl. f. innere Med., 1908. * North, Med. Record, March, 1909. » Amer. Jour. Pub. Health, 1913, ii, 34. 1 EXAMINATION FOR BACTERIA IN OYSTERS 719 1.0 c.c, 0.1 c.c, and 0.01 c.c, respectively, from each of the five oysters. The tubes are incubated for three days, and the development of over 10 per cent of gas in the closed arm is considered a positive reaction. The score is recorded as the approximate number of colon bacilli con- tained in the 5.55 c.c. of shell liquor from the five oysters, and is esti- mated in the following way : A positive reaction in a tube inoculated in 1 c.c. is recorded as 1.0, a positive reaction in 0.1 c.c, is 10, and in 0.01 is recorded as 100. The sum of these figures is the score for the batch of oysters from which the five have been taken. In examin- ing shucked oysters a well-mixed sample of oysters and the surround- ing fluid are put in a sterilized vessel and lactose bile tubes inoculated in triplicate with 1.0 c.c, 0.1 c.c, 0.01 c.c, 0.001 c.c of the liquor. No definite standard score has been adopted, but the United States Pure Food Board ^ has condemned unshucked stock having a score of 32 or higher. BACTERIA IN THE INDUSTRIES Bacteria and Tobacco. — In the manufacture of tobacco, the har- vested leaves are first dried and then heaped up in large masses, in which the tobacco undergoes fermentation. During this fermentation, which goes on at temperatures varying from 50° C. to 60° C, carbohy- drates are split up and much nicotin is destroyed.^ The end products consist largely of CO2 and various organic acids, butyric, formic, succinic, etc During the fermentation, bacteria of many varieties are found in the heaps of tobacco leaves and many attempts have been made to determine flavors artificially by inoculating tobacco leaves of a poorer quality with cultures obtained from the finer Havana grades. Suchsland ^ and others, who have attempted this, claim to have ob- tained marked improvements in domestic products by this method. The bacteria found in tobacco fermentation belong to many varieties. Some of these are closely related to the proteus and subtilis groups. Others are distinctly thermophilic, an attribute required by the high temperatures attained in the fermenting tobacco leaves. It is probable that the tobacco flavors cannot be regulated by bacteriological methods alone, since it has been shown by Loew * that an important factor in the 1 Gorham, U. S. Pure Food, Amer. Jour. Pub. Health, 1913, ii, 32. ^ Behrens, quoted from Flugge, "Die Mikroorganismen," Bd. 1, Leipzig, 1896, ' Suchsland, Ber. der Deut. botan. Ges., ix. < Loew, Rep. U. S. Dep. Agriculture, 59, 1899, 720 BACTERIA IN AIR, SOIL, WATER, AND MILK tobacco fermentation is contributed by the leaf-enzymes, which, of course, depend intimately upon soil and climatic conditions. Opium Productions. — In the preparation of opium f oi* smoking pur- poses, the raw product is subjected to a prolonged period of fermenta- tion by which the carbohydrates in the material are destroyed. Accord- ing to various observers, the process is carried out in most cases by a species of aspergillus. Indigo Production. — Indigo, which is obtained from the plants ''Isatis tinctoria*' and ''Indigofer tinctoria," is not present, as such, in the plants. In some of these it is found in the form of indican, in others, as indoxyl. It has been shown by Alvarez that the oxidation of indican and indoxyl into indigo-blue is carried out largely by bacterial oxydases. Sterilized indigo plants do not produce the blue color. Alva- rez ^ has isolated a bacillus closely related to the Bacillus mucosus capsulatus group, to the action of which he attributes this oxidation. Bacteria in the Tanning of Hides. — Raw animal hides are subject to decomposition until treated by a process known as tanning. This consists first in the depilation of the dried and salted skins, either by partial putrefaction in an atmosphere saturated with water vapor or by chemical treatment with solutions of milk of lime. After this, the tanning proper consists in subjecting the skins to prolonged immersion in solutions made up according to a large variety of formulae — the principle of all of which, however, seems to be found in the mixing of various organic ingredients, such as bran-mash, oak-bark, and often dried animal excreta, in which fermentation and acid production oc- cur. According to Haenlein,^ this acidification is the essential by which the leather is sterilized and rendered soft. This author has described the Bacillus corticalis, which he found in fir-tree bark and to which he ascribes the acid fermentation of tanning liquors in which this ingredient is employed. Wood,^ who has worked extensively upon the subject, has attempted to substitute pure cultures for the old un- certain chance mixtures employed. In spite of these investigations, however, while we must acknowledge the probable importance of bac- teria in the tanning process, the subject is by no means on a scientific or exact basis. 1 Alvarez, Comptes rend, dc I'acad. des sci., vol. 105. 2 Haenlein, Cent, f . Bakt., II, i, 1895. » Wood, Jour. Soc. Chem. Industry, 1895, 1899, SECTION VI PATHOGENIC PROTOZOA Frederick F. Russell, M.D. INTRODUCTION In the practice of his profession the physician requires a knowl- edge of the pathogenic protozoa found in man and the domestic ani- mals and of their closely related non-pathogenic forms. Quite com- monly in the diagnosis of fevers it is necessary to examine the blood of the same patient for both malaria and bacteria, therefore a work- ing knowledge of the principal pathogenic protozoa is essential. In this work it will be possible to describe the forms only of medical interest, and the reader is referred to other works for further infor- mation. The protozoa are unicellular animal organisms that occur singly or in temporary colonies. The functions of the animal are carried out by the protoplasm of the single cell, parts of which may be differentiated for special purposes and are then called organellae. CLASSIFICATION OF THE PROTOZOA Class I. Sarcodina (Bhizopoda) . — The body is naked or encased and the animal moves by means of protruding temporary processes of the body called pseudopods. They possess one or many nuclei and reproduce by fission or multiplication in a cyst. Order I, Amehce (Lohosa). — Naked or with a simple" shell, the pseu- dopia are lobose or finger-shaped, the nucleus is usually single and there is sometimes a contractile vacuole. Example, the amehce. Class II. Mastigophora (Flagellata). — They possess flagella for locomotion and for obtaining food ; they may be naked or fur- nished with a membrane ; many forms possess nucleus, contractile vacuole and a small groove spoken of as the cytostome. Ex- amples, the trypanosomes and intestinal flagellates. 72J 722 PATHOGENIC PROTOZOA Class III. Sporozoa. — They live parasitically in the tissues o£ other animals, ingesting food by osmosis ; they have no cilia in the adult stage but may form pseudopodia, one or more nuclei, no contractile vacuole, reproduction by spores. They are divided into two sub- classes, telosporidia and neosporidia. Examples, gregarinida, coc- cidiidea, hemosporidia, sarcosporidia, etc. Class IV. Infusoria (Ciliata), — The body is generally uniform in shape, with cilia and contractile vacuole, and usually with macro- and micronucleus. Examples, parameciiim, halantidium. CHAPTER LV CLASS I— SARCODINA (RHIZOPODA) THE AMEB-ffi These organisms belong to the order Amehina (Ehrenberg). They are characterized during the vegetative stage by a semifluid consis- tence, permitting rapid changes of form, ameboid movements, and progression by means of pseudopods. There is no internal skeleton and the protoplasm is naked and may be differentiated into endo- and ectoplasm, and in some cases a contractile vacuole is present. All forms possess one or more nuclei. Multiplication takes place by divi- sion into two or more daughter cells. Fertilization takes place by the conjugation of two merozoites and possibly by autogamy. Since some flagellates possess an ameboid stage, it is necessary to know most of the life cycle of an organism before classifying it as an ameba. The protoplasm varies greatly in its consistency, depending on the species as well as the stage of the life cycle, and the environ- ment and food supply. Most amebag, including all the parasitic forms (endamebge), possess a single nucleus, yet Ameba diploidea and Ameha hinucleata always have two, and the other species may show more. The nucleus of all types possesses a karyosome. The nucleus is well developed and in it may be followed either a simple or typical mitosis. The cytoplasm is usually at some stage divided into a granular endo- plasm and a clear or hyaline ectoplasm, the latter forming the pseu- dopods by which the animal moves from place to place. Until recent years all ameboid organisms were placed in the genus Ameha, but Schaudinn revived a genus originated by Leidy of Phila- delphia, the Endameha, for the parasitic species which have many points of difference from the free living varieties. Of the free living forms, the easiest to study is the Ameha proteus (Pallas), a very large organism, 200 microns in diameter, found frequently in stagnant water; it, however, has no direct importance in medicine. Another group of free living amebae is of some interest, because of the con- fusion they have caused in the study of parasitic ameba ; they are the so-called **limax amebae,*' which have been cultivated on agar, and for this genus Chatton (1912) has proposed the name Vahlkampfia. They are small organisms, 5 to 30 microns in diameter, provided with 723 724 PATHOGENIC PROTOZOA finger like or spinous pseudopodia, and characterized by a nucleus with a large karyosome and a single nucleated resistance cyst in which no multiplication occurs. They have repeatedly been cultivated from hu- man dysenteric stools, from the air, and apparently from liver abscess pus. It has been shown, beyond doubt, that they are harmless to man, and tha't they pass through the intestinal tract with food and water in the cyst form. While they will develop in cultures at body tempera- ture, a better growth is obtained at the temperature of the room. Since f % 0 r *• • Fig. 156. — Endameba histolytica. Vegetative form showing histolytica type of nucleus. (Army Medical School Collection, Washington, D. C.) the true parasitic amebae have never been cultivated on artificial media, the Vahlkampfia may be dismissed with the statement that they are not pathogenic. Leidy 's genus Endameba includes all parasitic forms, and is char- acterized, among other things, by the absence of a contractile vacuole, which is always present in Ameba and Vahlkampfia. It was proposed for the large ameba parasitic in the cockroach, Endameba hlattce, and the genus is a natural one and well established. The species of impor- tance to physicians are Endameba histolytica, Endameba coli and Endameba gingivalis. SARCODINA 725 ENDAMEBA HISTOLYTICA (Endameha tetragena [Viereck], Endameha africana [Hartmann], Endameha nipponica [Koidzumi, pro parte], Endameha tropicalis [Lesage, pro parte] ) Amebae as a cause of disease were first described by Lambl of Prague, in 1860, who found them present in the stools from a case of severe diarrhea in a child. In 1870 Lewis and Cunningham found amebae in 20 per cent of the stools of cholera patients, but attached no pathogenic importance to them. The first accurate description we owe to Loesch of Petrograd, who in 1875 studied an undoubted case of i' IG. 157. — Endameba histolytica. Vegetative form, simple division. Tetragena type of nucleus. (X 1300.) (Army Medical School Collection, Washington, D. C.) amebic dysentery with relapses, and he named the organism Ameha coll. He was further successful in reproducing the disease in a dog, and thus began its experimental investigation. Not much progress was made until Kartulis in Egypt began, in 1886, the publication of a long series of studies which has continued up to the present time, and because of the rich clinical and pathological material at his dis- posal his work has been of the greatest value. In 1890 Osier published the first paper in America. He was followed by Musser and Stengel and, in 1891, by Dock, and Councilman and Lafleur. The work of the 726 PATHOGENIC PROTOZOA last two authors was especially complete and firmly established the entity of this disease in America. In 1902 Jiirgens differentiated the pathogenic ameba from the harmless, and in 1903 the epoch-making work of Schaudinn appeared. This author, who was a zoologist by training, showed clearly that there were two forms of parasitic amebse and he followed out most of the details in their life history, renaming them Endameha histolytica and Endameba coli. Schaudinn 's work has been generally confirmed, in this country by Craig, Whitmore, Walker, Darling and others. CLINICAL DYSENTERY Dysentery as a disease has been known from the earliest times and references are found to it in Sanscrit and Egyptian literature and in early Greek and Roman writings. Until recent years its etiology was obscure, but we now recognize two separate forms, bacillary and amebic; the former has already been described under the dysentery bacilli. Amebic dysentery is a distinct clinical entity, and runs a course quite different from the bacillary form. It begins gradually, and in some cases is chronic in character from the start. Usually there is no rise in temperature nor any great change in weight or health until the disease has existed some time. The bowel movements become gradually more frequent and the fecal matter is accompanied by larger and larger amounts of mucus and blood. As the disease pro- gresses and more and more of the colon is involved the amount of blood and mucus increases until the stool contains little else. The colicky pains increase in frequency and severity and there is added tenesmus and finally nausea and vomiting. The patient loses flesh and strength and when the stools increase to 20 and 30 daily, becomes bed-ridden. The abdomen is concave and tender on pressure, especially over the colon. The course of the disease, if untreated, tends to progress with periods of remission, and spontaneous cure probably does not occur. Bacillary dysentery, it will be remembered, is a disease with a short in- cubation period and an acute onset ; after two or three days ' illness the bacillary case is confined to bed, is pale, weak, emaciated and presents every evidence of profound toxemia; an amebic case, sick the same length of time, will be up and about and perhaps will not have applied for treatment. Complications. — A common and most dangerous complication is abscess of the liver. The amebae travel from the ulcers in the colon by SARCODINA 727 way of the lymphatics to the liver and there set up a liquefying ne- crosis of the parenchyma. The liquefied portion contains a reddish or chocolate-colored fluid, which is not pus in the ordinary sense, althouj>h it may become a pus-containing abscess if secondary bacterial infection occurs. Liver abscesses may be single, but are much more often mul- tiple, and at times the whole liver may be riddled with large and small abscess cavities; both right and left lobes may be involved. If sur- gical interference be withheld, the abscess increases in size, approaches the surface, and finally ruptures into the lung through the diaphragm or into the peritoneal cavity. At autopsy the lesions are found in the colon, principally at the sigmoid flexure and in the cecum, though in chronic cases the whole colon is involved, showing ulcers with undermined edges, swollen soli- tary follicles and a hemorrhagic-catarrhal inflammation of the mucous membrane. The ulcers, readily differentiated from those caused by the tubercle bacillus, are of all sizes, shallow or deep, and are charac- terized by irregular margins and undermined edges. Fresh smears made at autopsy will show vegetative amebae. In chronic cases, the colon is a mass of scars and ulcers and acutely inflamed, swollen and thickened mucous membrane resting on a hypertrophied submucosa. The severe and chronic forms of the disease are now as rare as they were formerly common as a result of the present specific treatment with emetin. Geographical Distribution. — Although amebic dysentery is classed among the tropical diseases, it is by no means confined to the tropics. In the United States, for example, it is endemic as far north as Balti- more and Washington, and cases are not very infrequent in the north- ern tier of states; hence one must examine the stools for amebae in dysenteric cases regardless of the location of the patient's home. Diagnosis. — While the history of the case may suggest amebic infec- tion, the diagnosis can onTy be made with certainty by microscopic examination of the stool. For this purpose the examination should be made as soon after the stool is passed as possible; and in this disease it is usually practicable to have the patient come to the hospital, clinic or office and pass a stool there. It may then be examined imme- diately. If this is impracticable, the stool may be kept warm and sent to the laboratory in a small glass jar inside a tin pail partly filled with water at body heat; a little cloth or absorbent cotton will hold the hot water and prevent splashing during transit. The stool will show bloody mucous masses, and small drops of this are placed on 728 PATHOGENIC PROTOZOA slides, covered with a glass and ringed with warm vaseline to prevent evaporation. The preparation, to be of value, must be thin, and the bloody mucus may be diluted with salt solution if necessary. Except in hot weather, the slide should be examined on a warm stage, or the slide may be warmed by placing heated coins on it, near the cover glass. At least half a dozen slides should be examined before reporting a negative result. Stained preparations are not difficult to prepare, although the process requires some time and care. As in most zoological work, wet, rather than dry, fixation is used. Thin smears are made on cover glasses or slides and before they can dry are covered with or immersed in Schaudinn's fluid. This is a mixture of two parts of a saturated solution of bichlorid of mercury in normal salt solution and one part of absolute alcohol. The mercuric solution is prepared by adding to boiling normal salt solution a little more mercury than will dissolve;' on cooling, some of the bichlorid crystallizes out. At no stage of the process must the preparation become dry or the smear is worthless. 1. Fix in hot (60° C.) Schaudinn's fluid, 5 to 10 minutes. 2. Harden in 70 per cent alcohol 10 to 30 minutes, then wash in 70 per cent alcohol to which a few drops of tincture of iodin have been added until it is distinctly colored — 10 minutes ; store in 70 or 80 per cent alcohol until ready to stain. 3. The Rosenbusch hematoxylin is quite satisfactory. Transfer the slides to distilled water and change several times until they are free from alcohol, then immerse in 3.5 per cent iron-alum solution for from half an hour to over night. 4. Stain in the following solution, after rapid washing in distilled water: (a) 1 per cent hematoxylin in 95 per cent alcohol. (&) Saturated aqueous solution of lithium carbonate. Solution (h) is added to solution (a) until the mixture is a cherry red, four or five drops of lithium to 10 c.c. of hematoxylin is sufficient. The solution is either pipetted onto the slides or .they are im- mersed in it. Stain from 20 minutes to over night. 5. Wash thoroughly in distilled water. 6. Differentiate with a weak iron-alum solution (three parts of distilled water to one of the iron-alum solution is satisfactory), until SAHCODINA 729 the slide under the microscope shows the structure of the nucleus ; the examination is made in water under a cover glass. 7. When the differentiation is complete the slide is washed in dis- tilled water and passed through graded alcohols, 80, 95 and absolute into xylol and xylol-balsatm. This stain is permanent. Romanowski stains on dried smears may be used, but are not so good. In fresh specimens Endameba histolytica presents the following^ appearance: the vegetative forms are pale, unstained with bile, and are seen to be large bodies, 20 to 30 microns in diameter, consisting of endo- and ectoplasm, and often showing a delicate nucleus and also Fig. 158. — Endameba histolytica. (Army Medical School Collection, Washington, D. C.) many inclusions in the digestive vacuoles, principall}^ red blood cells. The organisms for several hours after the stool is passed remain ac- tively motile, pushing out clear, glass-like pseudopods, into which the granular endoplasm pours as the ameba progresses across the field. Even when there is no progression the pseudopods are protruded or retracted first in one then in another direction. There is always, dur- ing motion, a distinct separation of the clear ectoplasm from the granular endoplasm, and the latter, in acute cases especially, contains many red blood cells, occasional examples showing as many as twenty or thirty. The presence of red blood cells either entire or partly di- gested is characteristic of Endameba histolytica, since they are pres- ent in Endameba coli in only rare instances. The ameba is sometimes greenish, and it is supposed that this color is due to hemaglobin liberated from the ingested red cells. The pseudopods of this species 730 PATHOGENIC PROTOZOA are clear, glassy and evidently viscid and dense and have given it its name ''histolytica," since Schaudinn states that he saw the ameba penetrate the mucous membrane, the pseudopods dissecting apart the epithelial cells. The nucleus, when the endoplasm is packed with in- clusions, may not be visible, but further search will reveal amebse showing a nucleus. It is vesicular, with a delicate limiting membrane,, and as it is highly refractile, may appear as a clear bright spot. As the specimen grows older the amebse lose much of their motility and the nucleus may become clearly visible, revealing small chromatic dots or masses adherent to its inner surface and a small central karyosome. Fig. 159. — End ameba histolytica. (Army Med. School Collection, Wash- ington, D. C.) Fig. 160. — End ameba histolytica (X 1150.) (Army Med. School Collection, Washington, D. C.) The cysts are round, quite small, about 10 microns, and show four small ring-like nuclei ; the wall is distinct and often double-contoured. The motile amebae cannot be confused with anything else, but when in the resting stage they have been mistaken for swollen and edematous epithelial cells. A little attention to the nucleus will prevent this error, since the tissue-cell nucleus is large, distinct, and entirely dif- ferent from the nucleus of an ameba. In specimens stained with hematoxylin the finer details, especially in the nucleus, may be studied, but stained preparations are never nec- essary for clinical diagnosis. In smears from fresh cases vegetative forms only are found, later many degenerative forms appear and dur- ing convalescence only cysts may be seen. In stained specimens there SARCODINA 731 is rarely any separation of ectoplasm and endoplasm, but the nucleus is always visible. The cytoplasm is granular and has a coarse honey- combed" appearance. The nucleus shows a distinct, though delicate, limiting membrane, on the inner surface of which are few or many chromatin dots. In the center is a small karyosome, which may show a central body or centriole. The outer zone of the nucleus has a honey- comb structure, in which are imbedded granules of chromatin. Multiplication in the vegetative stage is by division into two daugh- ter cells, and dividing cells are common after the disease has existed for some time and the amebge are preparing to encyst. At such time's, due probably to rapid division, the amebae are small and the nucleus shows various stages of a simple mitosis; at the poles of the delicate spindle may be seen the two new centrioles ; this may be detected even in unstained specimens. ^ Degenerative Forms. — These are extremely common in stale stools, in cases during convalescence, or under active treatment, and also in experimental dysentery in the cat, and they have led to much con- fusion in the past. The nucleus breaks up into fragments and chro- matin masses are extruded into the cytoplasm in irregular forms, and parts of the cells are apparently budded off. At one time the budding process was looked upon as normal by Schaudinn and his followers, but there is now little doubt JJiat both spores and buds are degenera- tive changes and that the animal multiplies only by binary fission in the vegetative forms or by the development of four nuclei in the cysts. Cyst Formation. — The cyst of Endameba histolytica w^as first de- scribed by Viereck as Endameha tetragena, and for a time was believed to be a new species. In fact, Hartmann described a vegetative stage of Endameba tetragena as Endameha africana, afterwards accepting the name " tetragena, ' ' but it is now apparent that tetragena is merely the end, or cyst stage, of Endameba histolytica, which had formerly been overlooked. Cysts are not easily found in all cases, and it is possible that when treatment is vigorous they never develop. They are, without doubt, the form in which the parasite leaves the body to infect new victims ; because of their heavy cyst wall they are quite re- sistant to drying and other harmful influences. It is also possible that they may be retained in the wall of the colon and so be the starting- point of the relapse, which is so characteristic a feature of amebic dysentery. The protoplasm of the cyst and the precystic stage is granular, but shows no vacuoles nor cell inclusions. The nucleus un- dergoes division by mitosis first into two, and then four small ring-like 732 PATHOGENIC PROTOZOA 4 iff" ^j. nuclei, and the presence of these four nucleated cysts is pathognomonic of the disease. They may be found most abundantly, not in the small amount of mucus which may adhere to the formed feces, but in surface scrapings from the fe- Pcal mass. In addition to the four small ring- ^ like nuclei, the cysts contain few or many clumps of chromatin; these in total mass may be many times greater than the nucleus, and it is impossible, there- fore, that they are simply extruded from the nucleus; evidently, the chromatin grains, while in the cytoplasm, increase in size and number. In hematoxy- lin stains no structure in these masses is dis- cernible and their func- tion is unknown; after a time they disappear and one finds cysts quite free of them. The presence, however, of many large chromatin masses in the cysts is characteristic of Enda- meba histolytica. Fertilization inside the cyst has not been demonstrated, but it is possible that the four young amebae, liberated from the cyst when in- gested by a new host, are gamets, and that conjugation takes place between them, as in the case with Endameba blattae. The treatment of amebic dysentery, to be effective, must be radical and persistent, and may be compared to the treatment of malaria with quinine. For many years the English in India, with a few followers in other parts of the world, had treated dysentery with ipecac in mas- sive doses, with wonderful results in some cases and failure in others. The treatment was quite disagreeable and not entirely satisfactory. Vedder, in 1911, examined the various alkaloids of ipecac and found that emetin was strongly amebacidal, and he recommended its use for 0, m i i Fig. 161. — Endameba coli. (Army Med. Collection, Washington, D. C.) SARCODINA 733 this disease. Rogers, in India, following out this suggestion, soon re- ported excellent results, and the drug is now accepted as a specific. It is administered hypodermically in ^-grain doses three times a day at first, then twice and later once daily until a total of ten grains has been administered (Vedder). In addition, the patient is put to bed and placed on a milk diet. During convalescence large doses of bis- muth subnitrate, a heaping teaspoonful suspended in water or milk, may be given (Eieeks). Relapses may be prevented by occasional examination of the stools for cysts and a course of emetin if they are Fig. 162. — ^Endameba coli. Typical nucleus. (Army Med. School Collection, Washington, D. C.) found. As a result of the emetin treatment and exact diagnosis the clinical picture of amebic dysentery has completely changed, and we no longer see the weak and emaciated dysenteries who formerly crowded the wards of tropical hospitals. Prevention. — One significant fact appears in the epidemiology of the disease — ^it always occurs sporadically and never in explosive epi- demics such as we see in water-borne diseases, like typhoid and chol- era; house epidemics are, on the contrary, not uncommon. This fact 734 PATHOGENIC PROTOZOA points to the importance of contact, and perhaps flies, as the chief agents in its spread. Extreme cleanliness among the servants and in the kitchen will prevent the transfer of histolytica cysts from the ill to the well. The disease has disappeared from the Panama Canal Zone, where it formerly was common, since the introduction of good water and sewer systems and better hygienic conditions. ENDAMEBA COLI (Loesch emend. Schaudinn) {Endameha nipponica [Koidzumi, pro parte] and possibly Endameha minuta [Elmassian]) This is a harmless parasite of man, and its presence in stools, at one time, gave rise to much confusion, and in the minds of many, threw doubt upon the ex- istence of a form of dys- entery due to ameba, since it was found not infre- quently in healthy individ- uals. Schaudinn found it present in the stools of fifty per cent of the per- sons examined in East Prussia, in Berlin in twenty per cent, and in Istria in sixty per cent. Craig, and Craig and Ash- burn found it present in 176, or fifty-eight per cent, of 307 examinations of healthy American soldiers. Craig was able to follow* some individuals for four to six years, during which time they constantly showed Endameha coli in the feces, yet never developed dysentery. The organism seems to be found in all countries, regardless of climate. Its recognition and separation from histolytica we owe to Schaudinn, Jiirgens, Craig and others. In size it varies from ten to forty microns, the average being be- tween twenty and forty. The ectoplasm is never seen except during movement, and it is then hyaline, and only slightly refractile, and Yid. 1G3.^Endameba coli. Small precystic form. (Army Med. School Collection, Wash- ington, D. C.) SARCODINA 735 much more fluid than in histolytica. The digestive vacuoles rarely contain red blood cells, but are filled with cocci and bacilli, a form of food rarely seen in histolytica. In general, the vacuoles are larger and more numerous in coli than in histolytica, and the motility is feebler. In fresh specimens the nucleus is rather easier to find than in histolytica, and is distinctly outlined by a heavy, double-contoured membrane. The nucleus, as in all ameba, is vesicular, and shows a small karyosome and dots of chromatin on the nuclear membrane and imbedded in the nuclear network. Fig. 164. — ^Endameba coli Cyst. (Army Med. School Collection, Washington, D. C.) Multiplication in the vegetative stage is- by binary fission of the nucleus and the cytoplasm, resulting in two daughter cells, or the nucleus may continue to divide into four or eight daughter nuclei before the cytoplasmic division begins, producing, in the end, two, four, or eight daughter cells. Cyst Formation. — This is characteristic of the species, and it fur- nished one of the principal reasons for Schaudinn 's separation of coli and histolytica. Before encysting the animal frees itself of all in- 736 PATHOGENIC PROTOZOA elusions and becomes clear, transparent, and assumes a spherical form, and secretes a cyst wall. The nucleus divides first into two, then four, and finally eight daughter nuclei; there is usually a large vacuole found in the cyst during this division, but its function is uncertain. Schaudinn described a complicated autogamy in the cyst, yet later researches by Hartmann and Whitmore show nothing more than re- peated binary division of the nucleus. The normal number of nuclei in a coli cyst is eight, yet occasionally cysts are seen in which division has gone on until there are as many as sixteen. Cats or human beings may be parasitized by feeding material con- taining coli cysts, and in nature, as the cysts are the resistant forms of the parasite, the infection is probably transmitted from one host to another by means of them. No disease, however, results, though the amebae con- tinue to be present in the stools for years. It is possible that fertilization-takes place J between the young amebae (gametes?), which are liberated when the cyst dissolves , in a new host, as is the case with Enda- meba blattse. -^_--' ENDAMEBA GINGIVALIS Fig. 165.-ENDAMEBA coli. (^^^^ 1849, emend, von Prowazek 1904) Cyst showing eight nu- i . n -, • ,, , clei. (Arch, fiir Protisten- Ihis ameba is lound m the human kunde, 1912, xxiv.) mouth both in health and disease. It has been described at different times under various names {huccalis, dent alls) by Gros, Steinberg, von Prowazek, Lewald, Smith and Barrett, Chiavaro and Craig, and quite recently has been suggested as the cause of pyorrhea alveolaris by Smith and Barrett and Bass and Johns. It is widely distributed, and has been reported from all quarters of the world. The organism is easily found in the tartar at the base of the teeth, in cavities in the teeth, and even at the gum margin in healthy mouths. It varies in size from seven to thirty-five microns, averaging between twelve and twenty (Craig). Motility is well marked, though it is not so active an organism as histolytica, the pseudopods, mostly short and blunt, being formed of the clear, slightly refractile ectoplasm. The endoplasm is granular, contains the nucleus and many food vacuoles containing nuclei of leucocytes and granular matter, and rarely a few SARCODINA 737 red blood cells. The nucleus is small, and in fresh specimens is usually invisible. Before encysting the parasites are much reduced in size, and the cytoplasm frees itself from all inclusions and becomes clear, spherical and immobile. The cysts are small, eight to ten microns, circular and definitely outlined, sometimes with a double contour, and in stained specimens the nucleus is always visible. It is small, averaging only three microns (Craig), making it smaller than in histolytica and coli. The limiting membrane, whil-e not heavy, is distinct, and encloses a nuclear body having very little' chromatin other than the small cen- trally located karyosome. Multiplication occurs only in the vegetative state and by binary fission. Cyst Formation. — Cysts are rarely observed, and then in small num- bers ; the cyst wall is not heavy, but may show a double contour ; the protoplasm is clear, free from all inclusions and vacuoles and shows a single small nucleus, but without any signs of multiplication. It is apparent, therefore, that the cyst is a protective stage and has nothing to do with reproduction, which occurs in the vegetative state only. In this respect it resembles the Vcthlkampfia. Although the organism is almost constantly present in pyorrhea alveolaris it is also found in healthy mouths, and in the absence of all experimental proof, it is doubtful if the organism is of pathological importance. Emetin has a decided effect upon many cases of pyorrhea alveolaris, and under that treatment alone the disease may disappear; the nature of its therapeutic action is not yet clear, and does not neces- sarily indicate any etiological relationship. CHAPTER LVI CLASS II— MASTIGOPHORA (DIESING) SUB-CLASS— FLAGELLATA (COHN EMEND. BtJTSCHLI) ORDER I— POLYMASTIGINA (Blochmann) These are flagellates, possessing three to eight flagella. Genus 1. — Trichomonas (Donne, 1837). — These have pyroform (pear-shaped) bodies, rounded in front and tapering to a point behind, provided with three long flagella, often matted together at the anterior end. An internal supporting structure, known as the axial filament or axostyle, is present. There is an undulat- ing membrane bordered by a trailing flagellum that begins anteriorly and runs obliquely backwards. Trichomonas vaginalis (Donne). — The organism is fifteen to twenty-five microns long and seven to twelve wide; it is pro- vided with three flagella and an undulat- ing membrane. It is found in the vaginal secretion only when it is acid, and in three instances it has been transmitted to the male. Trichomonas intestinalis (R. Leuckard, 1879). — This parasite is practically indis- tinguishable from Ttichomonas vaginalis. It occurs in the small intestine and appears in the stools during diarrheal attacks, but is probably non-pathogenic. It is readily found in the intestine and colon of mice and guinea-pigs. In fresh specimens (protected with a cover glass and vaseline) it is ac- tively motile, but the undulating membrane is difficult to detect until the movement has slowed down. 738 Fig. 166. — Trichomonas in- TESTiNALis. (After Brumpt, " Precis de Para- sitologie," 1914 ed.) MASTIGOPHORA 739 The presence of cystic forms has been questioned, and two quite different forms have been called resistance or dauer cysts. The earlier one, described by Ucke (1908), Bohne and von Prowazek (1908), and Benson (1910), is a fairly large body^ showing a double contour and a central homogeneous mass, perhaps food material, and an outer ring- like body containing two or more nuclei. Brumpt and Alexieff believe this form to be a fungus, having no relation to the trichomonad, and have called it ' ' Blastocystis hominis." Lynch ^ agrees with these au- thors, and describes an altogether different body as the resistant form. It is six by eight microns in size and perfectly symmetrical in shape. The wall is distinct, and there is a clear space between it and the body of the parasite. The nucleus, undulating membrane and flagella re- main visible in the cyst, but Lynch was unable to detect any change in the parasite indicating intracystic multiplication. Infection takes place probably by contact, and, as in typhoid fever, food, fingers and flies carry the resistant forms from one individual to another. Among the natives of tropical countries infection is almost universal, but the parasites are rarely seen in the large cities of the North. Genus 2. — Tetramitis mesnili (Wenyon, 1910). — Macrostoma mes- nili, Chilomastix mesnili, Fanapapea intestinalis. This organism, first described by Wenyon, from a native of the Bahamas, differs from tri- chomonas by the possession of a deep groove or cystotome, in which is found the undulat- ing membrane. It is present in diarrheal dis- charges, but its pathogenicity is doubtful. Fig. 167. — Lamblia intestinalis. Cyst formation. (After Doflein, "Lehrbuchder Protozoenkunde. ' ' ) Genus 3. — Lamblia intestinalis (Lambl, 1859). — The lambia are peculiar, bilaterally symmetrical, pear-shaped organisms, provided 1 Lynch, Kenneth M., Jour. Parasitol., Urbana, 1916, iii, 28. 740 PATHOGENIC PROTOZOA with a sucking disk anteriorly. There are eight pairs of flagella, the two posterior ones being continuations of longitudinal axostyles. The nucleus is first dumb-bell shaped and later divided into two separate nuclei. Cysts are found, and according to Schaudinn conjugation occurs in them with the development of four nuclei. The young para- sites attach themselves to the surface of epithelial cells of the small intestine by the sucking disk, but even when present in large numbers do not produce any characteristic symptoms. The same or like para- sites are present in mice, rats, dogs, cats, and sheep. Transmission is by contact, as in trichomonad infections. ORDER II— PROTOMONADINA The Protomonadina, another order of the Flagellata, have less than three flagella, and are divided into the Cercomonadidce, Bodonidce and the Trypanosomidce. Genus 1. — Cercomonadidse — Cercomonas hominis (Davaine, 1854). — As originally described, this organism has a pear-shaped body, drawn out to a point posteriorly, is armed with a single flagellum in front, and has no undulating membrane. It is a doubtful species and of no present importance. Genus 2. — Bodonidae — Prowazekia (Hartmann and Chagas). — These organisms, the only examples of the Bodonidce of medical inter- est, are of some importance, since they have been cultivated from hu- man feces on agar plates. The genus was founded for Prowazekia cruzi, a species discovered in human feces in Brazil. Other species are urinaria, asiatica, parva, weinhergi and javanensis. There are two flagella, arranged in the heteromastigote manner, that is, one flagellum projects forward and one trails behind. There is no undulating mem- brane, but in stained specimens a second nucleus is seen, the kineto- nucleus or blepharoplast. They are also found in water, and are probably not the cause of any disease. Genus 3. — Trypanosomidae. — History of the genus. — In 1841 Val- entine discovered the first hemoflagellate in the blood of a trout, and the following year Gruby described a flagellate in frog's blood and named it a ''trypanosome.'^ It was not until 1878 that Lewis discov- ered the rat trypanosome, Trypanosoma lewisi. The first pathogenic member of the genus was noted by Evans in 1889 in the blood of In- dian horses sick with surra, Trypanosoma evansi. Bruce in 1894 de- scribed the trypanosome of Nagana, a horse disease of Zululand, and MASTIGOPHORA 741 also demonstrated its transmission by the tsetse fly, Trypanosoma hrucei. In 1894 to 1899, Rouget, Schneider and Buffard found the trypanosome of Dourine, or *'mal de coit," among Algerian horses. Elmassian described, in 1901, the South American horse disease, ''mal de caderas,'' and discovered the parasite, Trypanosoma equinum. Since this time a large number of new species have been discovered, the more important of which will be described. Morphology. — The morphology of the trypanosomes, while subject to many variations in detail, is still uniform as to the characteristics of the genus, so that there is little difficulty in immediately recognizing the parasite. The body is long and sinuous, tapering anteriorly to a fine point called the flagellum ; the posterior end is never so delicate and is often quite blunt. All contain two nuclei, the larger being called the trophonucleus and the smaller the kinetonucleus. The trophonu- cleus is usually located midway in the length of the body, and the kinetonucleus behind it, often at the posterior extremity. The flagel- lum arises from a centriole (blepharoplast), which is located close to or in the kinetonucleus, and quickly reaches the surface of the body, when it turns forward and forms the border of the undulating mem- brane, a thin fold of periblast running the entire length of the body, and is often continued further forward as delicate filament. During life the undulating membrane has a constant wave-like motion. Transmission from one animal to another is usually by means of some blood-sucking invertebrate. Two possible forms of transmission have been recognized, the direct and indirect or cyclical; the direct form is used in the laboratory when transferring blood with a hypo- dermic syringe from an infected animal to a healthy one, and it also occurs in nature, although not so frequently as the second. Dourine, or mal de coit, is the best example of the natural direct method. The cyclical method is exemplified in the transmission of Trypanosoma lewisi by the rat flea, Ceratophyllus fasciatus, in which insect the tryp- anosome passes through a complicated life cycle. Whether the para- site in the insect ever passes from parent to offspring is still doubtful. Among fishes, reptiles and amphibians the parasites are carried by leeches, in whose intestinal tract they undergo a cycle of development. Just as in malaria, there is usually an alternation of hosts, from invertebrate to vertebrate, a part of the life cycle being passed in each. In the blood of the vertebrate is found the fully developed trypano- some, and in the intestinal tract of the invertebrate, crithidial and trypanomonad types, whicii are characterized by having the kineto- 742 PATHOGENIC PROTOZOA nucleus placed in front of or close beside the trophonucleus and by having a rudimentary undulating membrane. Cultivation. — In 1903, Novy and MacNeal ^ first obtained pure cul- tures of trypanosomes on artificial media. The medium devisee^ by them is prepared by equal parts of nutrient agar and defibrinated rab- bit blood. After the agar has been melted and cooled to about 50° C. an equal quantity of rabbit blood is added, mixed and allowed to cool. ''The tubes thus prepared are allowed to set in an inclined position, after which they are at once inoculated. It is essential that the sur- face of the medium be moist and soft, and if this is not the case, the tubes should be placed in an upright position until some water of condensation accumulates at the bottom. The initial culture usually requires a week or more, although not infrequently fairly rich growths may be obtained in three or four days" (Novy). Trypanosoma rotatorium. — Gruby described and named this hemoflagellate in 1843, and it is, therefore, the type species of the genus. The organism is widely distributed throughout the world, and is found in Bana esculenta, Rana temporaria and Hyla arhorea; the organisms are, however, not very nu- merous in any single frog. It is most often found dur- ing the spring and summer months, rarely in winter. Morphology. — Both body and undulating mem- brane are broad, the cyto- plasm is granular, and to- w^ard the straight side shows striae, probably indicating the presence of myonemes. The tro- phonucleus is large, lies near the middle of the body and near the undulating membrane; the kinetonucleus is smaller, lies posteriorly and stains deeply; the flagellum which originates near the kinetonu- cleus turns forward, forming the border of the undulating membrane, so/t Fig. 168. — Trypanosoma rotatorium in Blood OF Frog. (After MacNeal, '' Pathogenic Micro- oro;anisms," published by P. Blakiston's Sons & Co.) 1 Novy and MacNeal, Contrib. Med. Research (Vaughan), Ann Arbor, 1903, p. 549. MASTIGOPHORA 743 and is continued forward as a short flagellum. The posterior end is usually drawn out to a stubby point. The fully developed organism is large, being 40 to 80 microns long by 5 to 40 wide. One striking thing about this parasite is its tendency to pleomorphism. Multiplication in the blood stream of the frog is by binary fission ; in addition, a form of multiple division occurs in the viscera, preceded, according to Machado, by conjugation of sexually differentiated forms. The merozoites liberated from the mother cell are small trypanosomes, which in turn grow to large size, thus explaining the pleomorphism of the parasite. Cultures have beon obtained by Lewis and Williams on the blood agar of Novy and MacNeal in which a great variety of forms may be seen; the method of transmission is unknown, but the infection is probably conveyed by leeches. Many other trypanosomes have been found in fishes, frogs, and reptiles all over the world. Trypanosoina lewisi (Kent). — This, one of the longest known and commonest forms, has been studied more completely than any other organism of its class. It occurs in a large proportion of rats through- out the world, twenty-five to one hundred per cent being infected, and since it is non-pathogenic, it is a convenient organism for research. It may be passed from wild to white rats without difficulty, by in- oculating the latter with a small quantity of citrated blood contain- ing the organisms. At first the parasites are few, but after the lapse of three or four days, large numbers may be found ; the condi- tion of rapid multiplication lasts from eight to fourteen days, and is succeeded by a period of a month or more, during which time the parasites gradually diminish in number, finally disappearing com- pletely, rendering the animal immune from further infection, the immunity being complete. The serum of an immune rat has a certain protective power, and when inoculated simultaneously with blood con- FiG. 169. — Trypanosoma lewisi. (After Doflein and Minchin. Mac- Neal, "Pathogenic Microorganisms," published by P. Blakiston's Sons & Go.) 744 PATHOGENIC PROTOZOA taining trypanosomes, may prevent the infection. No other animals are susceptible. The blood should be examined in both fresh and stained specimens. In fresh specimens, because of the rapid, lashing movements of the parasite, the organisms are particularly easy to find. The details of structure, however, do not appear except in spreads stained with some of the modifications of the Romanowski stain, such as Wright's or MacNeal 's. In the adult stage the organisms are quite uniform in size and shape, being 27 or 28 microns long and 1.5 to 2.0 microns broad; the posterior end is long, tapering and pointed; tl^e kinetonucleus oval and flattened ; the trophonucleus is located near the anterior end, and the undulating membrane, while distinct, is relatively narrow. The endoplasm is finely granular, and by careful focusing the body wall or periblast may be seen. Multiplication in the rat is rapid, and many young forms are seen ; these are smaller, stain more deeply, and vary much more in size than the adults. Dividing forms are common, the division being longi- tudinal and unequal, the parent retaining the flagellum. Multiple division also occurs, resulting in the production of rosettes, whose structure suggests that repeated longitudinal division has occurred without the separation of the daughter cells. The insect hosts are two : the rat flea, Ceratophyllus fasciatus, and the rat louse, Hcematopinus spinulosus; the former being the right host and the latter the wrong one, since in it development is incomplete. Minchin and Thompson ^ have studi-ed the cycle in the flea, which is briefly as follows: When the injected blood and parasites reach the midgut of the flea, the trj^panosomes lose their flexibility and become more or less rigid, and are able to penetrate the outer wall of the epithelial cells of the stomach. Once inside the cell, the parasite folds upon itself and grows to large size ; the nuclei multiply, the body be- comes spherical and divides up within its own periblast into six or eight daughter cellsj all actively moving within their common en- velope. This becomes tense and finally bursts, liberating the young trypanosomes within the epithelial cell, through whose wall they soon escape into the lumen of the stomach. This form of multiplication may be several times repeated, after which the young trypanosomes pass down the intestine to the lower end to begin the rectal phase. 1 Minchin and Thompson, Quarter. Jour. Micr. Sc, Lond., 1915, Ix, 463. MASTIGOPHORA 745 There the parasites in large numbers are found attached to the epithelial cells by their flagella. Rapid multiplication takes place by repeated fission and the parasite becomes crithidial in form, that is, it loses its undulating membrane, becomes short and stubby, and the kinetonucleus moveS forward close to or in front of the trophonucleus. Ultimately some change back to minute trypanosomes, and these, when regurgitated or passed in the feces, serve to infect the next vic- tim. The rectal phase, when once established, lasts for several months or perhaps indefinitely, making every infected flea a chronic carrier. Trypanosoma evansi. — Surra is a disease of horses and mules, camels, elephants, buffaloes, and dogs, which prevails in India and other parts of Asia, and also in the Philippines and Northern Aus- tralia. The Philippine outbreak was traced to animals returned from China after the Boxer outbreak; for at that time American troops came into contact with native Indian troops and their animals. The trypanosome causing the disease w^as discovered by Evans in 1880. The clinical course of the disease is marked by an irregular recurring fever, with many remissions, during which the parasite can- not be demonstrated in the blood, although it is not difficult to find during the febrile period. The animal is anemic, weak, emaciated, and may show an ecchymotic eruption on the abdomen. The course of the disease may be either short or long, but leads almost invariably to death. In camels it lasts from two to four years, often without symp- toms until near the end, and these animals probably act as chronic carriers. Morphology. — ^Morphologically, the parasite is very like the Tryp- anosoma brucci of Nagana, yet, as a rule, the trophonucleus lies- nearer the anterior end than in brucei, although it may be impossible to distinguish in smears between the two. The disease is carried by biting flies, tahanidce and' stomoxys, and also by fleas. Trypanosoma brucei. — ^Nagana is a well-known horse and animal disease of Africa, which causes an enormous economic loss and has greatly interfered with the development of the country. The para- site was discovered by Bruce in 1895. Among the natives it is known as tsetse fly disease, and investigation has incriminated Glossina mor- sitans as the carrier. Clinically, the disease in horses is much like the Surra of India; the native name for the disease, nagana, means weak- ness. Nearly all the larger animals are susceptible to either natural or artificial infection, yet man is apparently immune. , 746 PATHOGENIC PROTOZOA Morphology. — Morphologically, it resembles closely most of the other pathogenic trypanosomes, and Minchin makes it the. type of a group of pathogenic trypanosomes, all closely resembling one another and possibly descended from one common ancestor : the group consists of hrucei, gambiense, evansi, equiperdum, rhodesUnse, and hippicum. The organism is less slender than lewisi and has a wider undulating membrane. The posterior end is relatively short, the trophonucleus lies in the middle of the body and the kinetonucleus at the extreme pos- B «^ Fig. 170. — The Most Important Trypanosomes Parasitic in Vertebrates. A, Tr. lewisi; B, Tr. evansi (India); C, Tr. evansi (Mauritius); D, Tr. brucei; E, Tr. equiperdum; F, Tr. equinum; G, Tr. dimorphon; H, Tr. gambiense. (X 1500.) (From Doflein after Novy. MacNeal, ''Pathogenic Microorganisms," published by P. Blakiston's Sons & Co.) terior end; a vacuole is placed just in front of the latter. In length the parasite measures twenty-five to thirty-five microns and is one and a half to two and a half microns in width ; multiplication in the blood stream is by binarj'- fission. Transmission is by means of the tsetse fly, Glossina morsitans, and perhaps Glossina pallidipes and Glossina fusca. The fly may transmit the disease directly after infection, acting as a mere mechanical car- rier, but it is more probable that a cyclical development of the para- site takes place in the fly, after which it remains infectious for a long MAStlGOPHORA U7 period. It has been shown that after the first few hours the fly is not infectious again until the lapse of eighteen days, when its bite once more conveys the disease, and trypanosomes may be found in the intes- tinal canal, the body cavity, the salivary glands and in the proboscis. Studies of the cycle in the fly show that only about five per cent of the flies permitted to feed on sick animals become chronic carriers. The work of Bruce and others has shown that the trypanosomes are more or less harmless parasites of the big game animals of Africa, w^hich therefore are believed to act as a reservoir, from which the dis- ease is transferred to the domestic animals by the tsetse fly. The dis- tribution of Glossina is not uniform, as they are only present in cer- tain definite areas called fly belts. Since the disease does not spread in the absence of the larger wild animals, it has been proposed that all big game be exterminated as a prophylactic measure. Mice and rats are susceptible and die in six to fourteen days after inoculation; guinea-pigs are more resistant, and may show one or more relapses within two to ten weeks. It has not been possible to immunize larger animals, although a certain degree of success has been obtained with the smaller animals used in the laboratory. Cultures have been grown on artificial media, yet not so readily as with lewisi and avian trypanosomes. The medium recommended by MacNeal contains the extractives of one hundred and twenty-five grams of meat, ten of peptone, five of salt, and twenty -five of agar to the liter; to this is added twice its volume of warm, defibrinated rabbit's blood. The blood agar slants should be soft and moist when inocu- lated. Filtrates from cultures are not toxic, the toxin apparently being liberated, according to MacNeal, from the body of the disinte- grating trypanosome. Trypanosoma hippicum (Darling). — The disease caused by this trypanosome in horses and mules has been known in Panama for many years under the name of '^Murrina de caderas" or " Derrengadera de caderas,'' the latter term being used when paralysis of the posterior extremities is the dominant symptom ; both names indicate a weakness of the hind quarters. The symptoms are weakness, emaciation, and, sooner or later, conjunctivitis and subconjunctival ecchymosis, and anemia. The horses and mules affected are obviously weak, and while in the stall, pull back on the halter, or stand with straddling hind legs. The incubation period in animals used for experiment is less than a week, a few animals lose weight rapidly and die within a few days, others live for several weeks. 748 PATHOGENIC PROTOZOA Treatment, including the use of arsenical preparations, is without effect, and all infected animals should be destroyed. The disease is apparently transmitted directly by flies, which carry blood and serum from ulcers and abrasions on infected to healthy animals. Morphology of the Parasite. — The trypanosome is sixteen to eight- een microns long and two microns wide. The kinetonucleus is about two microns from the posterior end, the trophonucleus about eight to ten microns from the same point. The posterior end is blunt and the cytoplasm usually contains numerous basophile granules; the undu- lating membrane is well developed and a chromatin filament runs from the kinetonucleus to the tip of the flagellum. The large kinetonucleus distinguishes this organism from Trypanosoma equinum. Pathological Anatomy. — Aside from the emaciation, edema of the belly wall, conjunctivitis and subconjunctival ecchymosis, there is usu- ally excessive fluid in the body cavities, an enlarged spleen, and, what is more characteristic, small petechial spots on the capsule of the spleen and in the cortex of the kidney and in the endo- and pericar- dium and occasionally on the pleural surfaces. Prophylaxis consists in the destruction of all infected animals ; the protection of wounds and ulcers in otherwise healthy animals by dress- ings and, wherever possible, the use of fly screens about the stables. Trypanosoma eqiiiperdum (Dourine). — This organism is the cause of dourine, a disease of horses and donkeys, which is usually trans- mitted by coitus, but may be carried by biting fliies, stomoxys. The organism was first described by Rouget in 1894 ; it resembles hrucei in many ways and produces a progressive, fatal disease of great economic importance. Formerly it was present throughout the greater part of Europe, but is now almost limited to the shores of the Mediterranean. From time to time it has been introduced into the United States and Canada by blooded French stallions and has spread into parts of the Northwest. The clinical course may be divided into a stage of edema, lasting about a month, during which there is a painless, soft swelling, limited to the genitalia and the belly wall. This is followed by the stage of eruption, during which plaques, or round edematous areas, are found under the hide on the flanks and hind quarters, and sometimes on thighs, shoulders and neck; this stage is short, lasting about a week. It is followed by the third stage of paralysis and anemia ; the animal loses flesh and strength, develops superficial ulcers, conjunctivitis, MASTIGOPHORA 749 keratitis, and ultimately paralysis, leading to death in two to eighteen months. The trypanosome is found most readily in the serous exudate from the ulcers, as it is infrequent in the peripheral circulation; in this respect it resembles the treponema of lues. The organism is about twenty-five microns in length and possesses a clear cytoplasm, free from granules, except when propagated in white mice, when they are plentiful. Fig. 171. — Dourine. Showing swelling of genitalia and plaques on the skin. (After KoUe and Wassermann, "Handbuch der Pathogenen Mikro-organismen," 2te Aufl., 1913.) Diagnosis hy Complement Fixation. — E. A. Watson,^ of Canada, has shown that it is possible not only to diagnose the disease when the clinical signs are clear, but also to determine the existence of its non- clinical, obscure and latent forms. Horses may tolerate an infection for one to three years, during which time they are capable of convey- ing the disease and yet remain normal in health and general appear- ance, and this method of diagnosis is, therefore, invaluable. Watson obtains the antigen by inoculating a large number of white rats with Trypanosoma equiperdum, collecting their blood when ^E. A. Watson, Parasitology^ Cambridgs, Enp., 1915, VIII, 156. 750 PATHOGENIC PROTOZOA teeming with trypanosomes, and separating them from the erythro- cytes and plasma by washing and centrifuging. Each of ten to twenty rats receive 0.3 c.c. of blood rich in trypanosomes intraperitoneally, and at about the end of the third day, when the organisms are very numerous, the rats are bled into citrate solution. By repeated washing the organisms may be separated, as a pure white layer overlying the erythrocytes. This mass of organisms is killed and preserved by a formalin-glycerin mixture, after which its antigenic strength is stand- ardized by titration in the usual Avay. The test, a pure culture of trypanosomes being used as antigen, is specific and is not positive in any other disease of horses. Trypanosoma avium. — This parasite was first described by Dani- lewski in 1885. In 1905, Novy and MacNeal ^ found trypanosomes Fig. 172. — Trypanosoma avium in Blood of Common Wild Birds. (After Novy and MacNeal. MacNeal, "Pathogenic Microorganisms," published by P. Blakis- ton's Sons & Co.) in 8.8 per cent of 431 American birds. Although there are doubt- less several species, the most common is Trypanosoma avium, a para- site twenty to seventy microns long and four to seven microns wide. They are found in the blood over long periods of time and do not appear to be pathogenic. Cultures are easily made and kept alive for long periods by weekly transfers. The mode of transmission is unknown. This was the parasite which was confounded in 1904 by Schaudinn with developmental stages in the life cycle of Hemorproteus noctiice 1 Novy and MacNeal, Jour. Infect. Dis., Chicago, 190.5, ii, 256. MASTIGOPHORA 751 and Hemorproteus ziemani with resulting confusion in the study of trypanosomes and hemoeytozoa, and it is only recently that the error has been generally ackowledged. Fig. 173. — Trypanosoma avium in Culture on Blood Agar. (X 1500.) (After Novy and MacNeal. MacNeal, "Pathogenic Microorganisms," published by P. Blakiston's Sons & Co.) Trypanosoma gambiense (Sleeping Sickness). — Two names have been given to the disease caused by this parasite, both of which are now recognized as stages in one and the same infection, human tryp- anosomiasis: they were trypanosome fever, and sleeping sickness. It is a chronic infection characterized by fever, lassitude, weakness, wasting, and, in its terminal stages, by a protracted lethargy. Sleep- ing sickness and trypanosome fever had long been known in tropical Africa, and the disease at present is widespread and the cause of tremendous mortality. It is estimated that one hundred thousand deaths accurred during the ten years ending in 1910. It is endemic 752 PATHOGENIC PROTOZOA in the lake region of Central Africa, and in the Congo basin. It was early introduced into Martinique in the West Indies, but did not spread and has now died out. Dutton and Todd found the parasite in 1901 in the blood of an Englishman in Gambia, who died after a febrile illness of two years' Fig. 174. — Trypanosoma gambiense. Calkin, "Protozoology." duration ; Castellani in 1903 found the parasite in the cerebro-spinal fluid of well-marked cases of sleeping sickness occurring among natives of Uganda. Clinical Signs. — The disease begins with slight febrile attacks, headache and increasing weakness, emaciation, swelling of the eyelids and enlargement of the lymph nodes. The temperature increases, edema of the extremities appears and the spleen enlarges. During the last stages nervous symptoms predominate and the patient sleeps day and night, but may have periods of excitement or convulsions, y.et finally sinks into deep coma and dies of exhaustion. Etiology. — The disease is transmitted by the bite of the tsetse fly, Glossina palpalis, which is apparently able to transmit the infection mechanically immediately after biting an infected host, yet in most flies the trypanosomes disintegrate and disappear from the intestinal tract within four or five days. Jn from Jive .to -ten per cent of the Mastigophora ■753 Fig. 175. — Tsetse Fly (Glossind pnlpalis). (From Rosenau) "Preventive Medicine and Hygiene.") flies, however, the trypanosomes multiply in the intestinal tract, and after eighteen to fifty -three days they again become infectious and re^ main so for a long period, the parasites be- ing found regularly in the salivary glands and in the proboscis. It is possible that the disease is transmitted in. other ways than by Glossina palpalis ; blood- .sucking insects, such as stomoxys, anopheles, mansonia and perhaps fleas, may act as me- chanical carriers. It is also possible that the disease is transmitted by coitus. Without some isuch explanation it is difficult to understand certain house epidemics which have occurred outside the fly belts. The animal host of the Trypanosoma gamhiense is believed to be the big game animals, particularly the antelope. Morphology. — The organism belongs to the brucei group, and its 'differentiation on morphology is difficult, yet, on the average, the posterior end is somewhat more pointed than the hrucei. In length it varies from fifteen to thirty microns, and in thickness from one to three microns. In fresh preparations the motility is not. marked; both plump and slender forms are found in the blood, but in the eere- bro-spinal fluid slender forms only are seen. Cultures on blood agar have been made by Thompson and Sinton, yet they died out after a few weeks, and were never virulent. The pathogenicity varies somewhat with the strain used, but apes are easily infected. In white rats there may be two or three relapses before death occurs, while when inoculated with hrucei death follows within two weeks. Pathogenicity. — Although cultures vary greatly in virulence, it is possible to infect rats, dogs and monkeys with a fatal trypanosomiasis ; cattle, sheep and goats continue to show a few parasites for months after inoculation but without sickening. In no animal, however, is 754 PATHOGENIC PROTOZOA it possible to reproduce the sleeping sickness stage as it occurs in man. Diagnosis. — When the disease is well developed in an endemic area, the diagnosis is easily made. During the early stages the exami- nation of the cerebro-spinal fluid, puncture fluid from the lymph nodes and the peripheral blood may all show the trypanosome; since the parasites are scarce the use of the thick film method of Ross may be necessary. When direct examination is unsuccessful, enrichment in the blood of susceptible animals, rats and mice will establish the diagnosis. Treatment. — Treatment is based upon the observation of Bruee and Lingard, that arsenious acid is trypanocidal. The best results have been obtained with atoxyl, in half gram doses, repeated at inter- vals of ten days or more for not less than four months. Light cases: become trypanosome free and are apparently cured, yet many relapse on cessation of treatment. Well marked cases may show improvement; yet ultimately grow worse and die. Other arsenical preparations have been used but none are entirely successful. Salvarsan drives the parasite from the peripheral blood but not from the cerebro-spinal fluid. The prognosis is unfavorable. Prophylaxis. — Prophylaxis is quite complicated and is carried out along several different lines. Infected fly belts are depopulated, the inhabitants being removed to a fly-free district where they may be treated at hospital stations. The fly breeding may be greatly diminished by clearing off the forest and brush, especially along the river courses, since the glossina needs abundant moisture for its propa- gation. Since the fly bites only during the day, all traveling in infected districts is best done at night. As it is recognized that the antelope is the permanent reservoir for Trypanosoma gambiense, the obvious remedy is its extermination. Trypanosoma rhodesiense. — This parasite was established hy Stephens and Fantham.^ It is transmitted by the Glossina morsitans, a fly which is widespread over large tracts of country, independently of the presence of water. It is becoming generally recognized that there are two forms of sleeping sickness, one of which is caused by this trypanosome. This form of the disease is more acute and is unaffected by treatment; the trypanosome is also more virulent for animals and may be differentiated from gambiense on its morphology. * Stephens and Fantham, Proc. Roy. Soc, 1910, Ser. B., Ixxxiii, 28. MASTIGOPHORA 755 As both parasites are found in the antelope, the prophylaxis is the same. Bruce ^ is of the opinion that rhodesiense and brucei are identical. Schizotrypanum cruzi. — This parasite, which differs from all other trypanosomes, is the cause of a form of human trypanosomyasis occurring in Brazil. It is transmitted by a bug, Conorhinus megistus, in which the parasite passes part of its cycle of development. In the human being, multiplication takes place in endothelial cells, lympho- cytes and other parenchymatous cells of the viscera; and also in the skeletal and heart muscles. While in this stage the parasite has no flagel- lum and resembles the leishmania; only after escape into the blood does it take on the trypanosome form. Guinea-pigs, rats, mice and mon- keys are susceptible ; the bed-bug, Fig. 176.— Schizotrypanum cruzi cimex, is also capable of transmitting in Human Blood. (From Dof- the disease ^^^^ after Chagas. MacNeal, Path- n 1. 1^2. ' ;j T^ nv. ogenic Microorganisms," pub- Cultures were obtained by Chagas j.^^^ j^^ p. Blakiston's So^ & and proved virulent for animals. The Qq ) human disease is found both in chil- dren and adults and is regularly fatal. It is characterized by an irregular fever, severe anemia, swelling of the lymph nodes, edema and disturbance of the nervous system. Leishmania. — This genus was founded by Ross in 1903 for the Leish man-Donovan and Wright bodies found in kala-azar and Delhi boil, to which Nicolle added another in 1909, the parasite of infantile splenomegaly. Leishman, Donovan and Wright, working independ- ently, described the first two parasites in 1903, and, although they have received various names, leishmania is now the accepted term. Rogers, Calkins and others, however, class them a^ herpetomonads, because of the elongated, flagellated form all take in cultures on the Novy-Mac- Neal-Nicolle blood agar medium. It is, however, best to consider them as a separate genus, because of their natural parasitic habits in human beings. Leveran, Fantham and others have shown that it is possible in the laboratory to induce the herpetomonads parasitic in the intestine of various insects to become parasitic in various ver- tebrates. 1 Bruce, Bull. Trop. Dis., 1916, vii, 68. 756 PATHOGENIC PROTOZOA ^W^~ \ 2 RLShcppako. Fig. 177. — Schizotrypanum cruzi Developing in Tissues of Guinea-pig. 1, Cross-section of fibers of striated muscle containing Schizotrypanum cruzi; 2, Section of brain showing cyst in a neurogUa cell containing chiefly flagellated forms; 3, Section through suprarenal, fascicular zone; 4, Section of brain show- ing neuroglia cell filled with round forms, (After Low and Vianna. MacNeal, " Pathogenic Microorganisms, published by Blakiston's Sons & Co.) MASTIGOPHORA 757 Leishmania donovani (Kala-azar).— This parasite is the cause of kala-azar, a disease characterized by irregular fever, weakness, anemia, cachexia and a remarkable enlargement of the spleen, and occasionally of the liver. It is chronic, progressive and frequently fatal, the mortality being about 80 per cent. The disease .is com- mon in tropical Asia and in northeastern Africa. Morphology. — The parasite is intracellu- lar, and is found principally in the endothelial cells of the spleen and liver, and in the bone mar- row. It is oval, two to four microns in di- ameter, finely granu- lar and occasionally vacuolated. It con- tains a large, round nucleus and a smaller blepharoplast which is oval or rod shaped ; a third body, a slen- der short thread, may sometimes be recognized, which is presum- ably the undeveloped flagellum. Stained specimens of blood, spleen and liver pulp, and bone marrow, usually show large endothelial celk or leucocytes closely . packed with parasites, one to two hundred to a single cell. Multiplication in the body is by simple division, and incompletely divided pairs of organisms are frequently seen. Cultures have been obtained in citrated blood and on the usual N. N. N. medium. When fully grown the cultural organisms are typical herpetomonads (leptomonads) ; the cell body elongates and the rudimentary, whip develops into a true flagellum. Both dogs and monkeys are susceptible to artificial inoculations. Fig. 178. — Leishmania donovani. (Army School Collection, Washington, D. C.) Med. 758 PATHOGENIC PROTOZOA The parasite is probably transmitted by some insect, either cimex (Rogers), or by the dog flea, Ctenocephalus canis (Wenyon), or a plant-feeding bug, Conorhinus, which occasionally sucks blood. Animal Pathogenicity. — Wenyon in 1913 ^ inoculated a dog with splenic emulsion from a man who died in London of kala-azar con- tracted in Calcutta. The parasite has been successfully carried through five animals, and in 1915 an examination of the bone marrow showed not only typical leishmania, but also a few large, well-marked leptomonad forms. Similar forms were described by Escomel in 1911, from South American dermal lesions. Monkeys may also be infected. Leishmania tropica (Delhi or Aleppo boil) is the organism found in a local skin affection variously termed Delhi boil, Aleppo boil or tropical ulcer. While it is probably transmitted by some insect, there is as yet no definite proof. The incubation period is about two months, while the disease, once manifest, lasts twelve to eighteen months and is followed by immunity for life. The parasite, which was first described by J. H. Wright,^ shows minor differences from Leishmania donovani, particularly a variable morphology, all gradations, from the usual oval to elongated narrow forms with pointed ends, being found. Cultures may be obtained on the N. N. N. blood agar, which develop into leptomonads, as with Leishmania donovani. Dogs and monkeys are susceptible to artificial inoculation, and it is possible that in nature the disease is carried from dogs to human beings by some insect. Leishmania infantum (Infantile Splenomegaly) was described by Nicolle in 1909 from cases of infantile splenomegaly occurring in Northern Africa. The disease resembles kala-azar in all respects, except that the patients are young children, and it is possible they are the same disease. The parasites are found in abundance in the liver, .spleen and bone marrow at autopsy and may be cultivated in the usual way on the N. N. N. blood agar. Animal Pathogenicity. — The disease occurs naturally in African dogs, and they are probably the source of infection, the parasite being carried by a flea or some other insect. Dogs, monkeys and guinea-pigs are susceptible to artificial inoculation. The treatment of leishmaniasis is unsatisfactory since there is no known specific. 1 Wenyon, Jour. Trop. Med. and Hyg., London, 1915, xviii, 218. 2 Wnght, J. H., Jour. Med. Res. Bost. 1903, x, 472. MASTIGOPHORA 759 Two other forms of dermal leishmaniasis have been described ; the first, due to Leishmania hraziliensis, occurs in many parts of South America. The parasite is morphologically identical with Leishmania tropica. Since the disease is always contracted in the virgin forest, one name for the affection is forest yaws ; uta and espundia are prob- V f m Fig. 179. — Leishmania infantum. (Army Med. School Collection, Washington, D. C.) ably different clinical forms of the same disease. The transmitting insect cannot well belong to the household vermin or domestic insects ; sylvan insects such as the ixodides, tabanides, simulids, mosquitoes and Conorhinus are all suspected of being carriers. The second form is called Leishmania nilotica (Brumpt, 1913), and is found in non-ulcerating keloid ijodules in Egyptian negroes. Morphologically, the parasite is indistinguishable from Leishmania tropica. CHAPTER LVII CLASS III— SPOROZOAi SUB-CLASS— TELOSPORIDIA HEMOSPORIDIA The Hemosporidia and Sarcosporodia are the only members of this order of medical interest. The hemosporidia belong to the sub-class Telosporidia of the Sporozoa, because spore formation begins at the end of the life cycle. The systematists have not yet agreed upon the proper classification of this group of parasites; consequently the older arrangement will be followed. They are, like the coccidia, parasites of cells, at least during the schizogenous cycle ; all change hosts to some insect for the sporogenous cycle. As the name implies, they live in blood cells and are rapidly growing ameboid bodies, which, beginning as sporozoites, penetrate the host cells and develop into trophozoites. These grow rapidly to adult segmenting parasites, in which case they are called schizonts, or to sexual forms, or gametes, when they are termed sporonts. In the course of their development, most species produce melanin from the destruction of the hemaglobin. The nucleus, which is readily stained, is single and possesses a karyosome ; the mature schizont divides into many small forms called merozoites, and these, when freed by the rupture of the degenerated erythrocyte, escape into the blood plasma, and if not phagocyted, penetrate other erythrocytes and repeat the asexual or schizogenous cycle. The pigment and undivided portion (restkorper) of the cyto- plasm of the mother cell accumulate in the bone marrow, spleen and other viscera. After a number of cycles of asexual multiplication have been lived through, a new development takes place and sexual forms begin to appear in the circulation. These grow to large size, yet show no indi- cation of division into merozoites and were at one time considered degeneration forms. Two varieties may be distinguished, one with a 1 For classification, see page 722, 760 SPOROZOA 761 dark staining cytoplasm and fine granular melanin, and the other with light staining, hyaline cytoplasm and coarse pigment; the former, loaded with reserve food material, is the female or macrogametocyte ; the latter, the male or microgametocyte. The gametes do not develop further until taken into the digestive tract of the insect host. For purposes of study, however, the microgametocytes may be made to ex- flagellate on the slide, dampened a little by breathing upon it, to stimulate the condition in the insect host. In such a preparation, the flagella, or microgametes, may be seen actively moving inside the cell body, whose wall they ultimately rupture, and all, four to eight, escape and whip about until they come in contact with a macrogametocyte, when one microgamete enters through the micropyle and finally fuses with the female nucleus. Hemoproteus columbse (Halteridium). — This parasite of the red blood cells of doves was described in 1891 by Celli and Sanfelice. It is widely distributed in nature and has been reported from Europe, Asia and North and South America. The organism is found within the cyptoplasm of the erythrocyte ; the nucleus, which is not regularly displaced, is surrounded by the growing parasite like a halter, and for this reason it was named halteridium by Labbe. It is sluggishly ameboid and produces an abundance of melanin, and when the blood is drawn the ripe male sporonts, the microgametocytes. rupture easily, liberating the active flagella, or microgametes. Under favorable cir- cumstances the fertilization of , the macrogametocyte by the micro- gametes may be observed on the slide, and it was while working with this parasite that Macallum first followed out the whole process of fertilization in the hemosporidia and gave the proper explanation of the flagellate stage seen in the malarial parasite. In the blood of the dove this parasite is usually seen as a large or small crescent, partly encircling the nucleus ; the gametes are readily recognized by the usual marks, that is, the female, or macrogametocyte, is rich in reserve material and the stained specimen takes a deep color ; the male, or microgametocyte, being poor in reserve material stored in the cyptoplasm, appears relatively pale in stained specimens. The invertebrate host of the parasite is Lynchia maura (Bigot), or Lynchia lividocolor, a biting hippoboscid fly of louse-like habits which lives in the nest and in the plumage. The cycle in the fly has been successfully worked out by Adie,^ who has demonstrated the * Adie, Helen, Indian Jour. Med. Research, Calcutta, 1915. 762 PATHOGENIC PROTOZOA ookinetes, zygotes and oocysts in the lower portion of the midgut. As the oocyst grows, it stands out from the gut wall and finally shows the striations indicative of the presence of sporozoites; after rupture of I % ^ ^ c^jf^=5n Fig. 180. — ^ILemoprotbus columb^. la to 3a, Development of female parasite in blood of dove; lb to 3b, Development of male parasite in blood of dove; 4a, 4b, 5b, 6 to 12, Development in the digestive tube of the fly (Lynchia) ; 13 to 20, Development of the parasite inside leucocytes in the lung of the dove. (After Avagao. MacNeal, "Pathogenic Microorganisms," published by P. Blakiston's Sons & Co.) SPOROZOA 763 the mature cyst, these collect in large numbers in the salivary glands and ducts. The life history of the parasite is seen to be like that of proteosoma and malaria, except that the asexual or schizogenous cycle appears to be lacking. Proteosoma ( Plasmodium) prsecox. — This parasite is a typical rep- resentative of the sporozoa, and is interesting historically, since it was the one with which Ross worked in 1898, when he first demonstrated the part played by the mosquito in *'bird ma- laria. ' * Grassi and Feletti de- scribed the parasite in 1890 under the name of hemameha precox. It is •widely distributed geo- graphically, and is com- mon in the blood of small birds, sparrows, robins and larks. It can be propa- gated in the laboratory in the blood of canaries with- out great difficulty; spar- rows, however, do not long survive in captivity unless kept in round glass jars, where they cannot injure themselves by dashing against the walls. The blood for examination is obtained from the cephalic wing vein, close to the body, which is nicked with a razor, and the blood taken up in a capillary glass tube containing a little citrate solution. To inoculate a new bird, it is sufficient to inject a small quantity of citrated blood from an infected canary into the breast muscles of the new bird, transferring to a new host at intervals of a month or less. Because it is not difficult to keep on hand, this organ- ism may be used for class study in localities where malarial cases are infrequent. There is no apparent reason for placing it in a different genus from the malarial parasites. The entire asexual cycle, schizogony, may be studied in the periph- eral circulation, as in quartan malarial fever. In nature it is transmitted by both culex and stegomyia {Aedes calopus), and its development is briefly as follows: The bird is inocu- lated by the mosquito with spindle-shaped young forms known as A B C D Fig. 181. — Proteosoma precox in Blood op Field Lark. A, Young parasite in blood cell; B, Half-grown parasite which has pushed aside nucleus of blood cell; C, Parasite with clump of pigment and many nuclei; D, Division into many merozoites. (After Doflein and Wasie- lewski. MacNeal, "Pathogenic Microorgan- isms," published by P. Blakiston's Sons & Co.) 764 PATHOGENIC PROTOZOA sporozoites. These possess the power of ameboid motion, and rapidly penetrate into an erythrocyte, in which they grow quickly ; they con- stantly move about inside the cell until nearly full grown, and are during this stage called trophozoites. The substance of the erythrocyte is rapidly consumed by the parasite and a dark pigment, melanin or Fig. 182. — MroauT of Culex Mosquito, Covered with Oocysts of Proteosoma PRiECOX. V, Vasa malpighii. (After Doflein and Ross. MacNeal, "Pathogenic Microorgaaisms," published by P. Blakiston's Sons & Co.) hemozoin, is formed from the destroyed hemaglobin. The mature parasite divides into many small forms called merozoites, and these, when freed by the rupture of the degenerated erythrocyte, escape into the blood plasma, and if not phagocyted, penetrate other erythrocytes and repeat the asexual or schizogenous cycle. The pigment and undi- vided portion (restkorper) of the cytoplasm of the mother cell accumulate in the bone marrow, spleen and other viscera. MALARIA This is one of the most common and widespread of preventable human diseases, and in some localities is the cause of a greater mor- tality and morbidity than tuberculosis. It is caused by one or more of the three forms of the malarial Plasmodium. As a rule the infec- tions are simple, yet in the tropics it is not uncommon to find two species of Plasmodia in the same patient, and this condition is called a mixed infection. History. — The disease under various names, as chills and fever, Roman fever, Chagres fever, has been known since the greatest an- tiquity. The cause was not discovered until 1880, when Laveran, a French military surgeon stationed in Algeria, first saw the organism and described it as the cause of malaria. He saw and described not only the pigmented trophozoite, but also the crescentic gametes and flagellating microgametocytes, and, because of the activity of the flagella, called the parasite Oscillaria malariw, a name afterwards SPOROZOA 765 given up. Later, in 1885, Celli and Marchifava described the parasite with greater accuracy and named it Plasmodium malarice, a poor name, since it describes merely a condition assumed by some fungi and mycetozoa, yet, according to the rules of zoological nomenclature, it must stand. In the same year, Golgi described the quartan parasite and in the following year demonstrated the relation of the various stages of the life cycle of the tertian parasite to the temperature curve. Even in antiquity many had noted the curious distribution of malaria, and its intimate relation to swamps and marshy places. Manson, who had already shown the role played, by an infected mos- quito in transmitting filarial disease, in 1894, suggested that the epi- demiology of the disease could best be explained on the hypothesis that it was conveyed by the bite of some blood-sucking insect, probably the mosquito. For years the interpretation of the flagella was a subject of con- troversy. They were regarded as degeneration products by some and as living elements by others. In 1897, MacCallum, working with halteridium, was able to show that they were, in fact, spermatozoa, as he saw them penetrate and fertilize the macrogametes, or large spherical forms without flagella. In 1897, Ross, of the British Indian Medical Service, described the beginning of the sporogenous cycle in what he called a dapple-winged mosquito, which w^e now recognize as an anopheline. Following out further Hanson's hypothesis, he was able the same year after long and laborious research to clear up the method of transmission of bird malaria, proteosoma, an analogous disease. Grassi and Bignami and Bastianelli, in 1898, succeeded in demonstrating the complete life cycle of the human form of malaria in the anopheles mosquito. Geographical Distribution. — The disease is found in a belt round the world extending from 40 degrees S. latitude to 60 degrees N. ; it is, however, not equally distributed throughout this zone, and even in the tropics there are many malaria-free areas, principally in the regions of higher altitudes, since the special home of malaria is in the low- lying, swampy and torrid coastal districts and river basins. Islands at a distance from the main land may be entirely free. Malaria reaches its maximum intensity in the tropics, where the anopheline mosquitoes breed continuously throughout the year, and new infec- tions may occur at any time; while in the sub-tropics and temperate regions it is a seasonal disease, appearing soon after the onset of hot weather with its new crop of anophelines and continuing until the 766 PATHOGENIC PROTOZOA first cold weather. Relapses continue to occur throughout the winter season. Modern times have seen it disappear from many regions where it was formerly 'endemic, because of increased cultivation of the soil and better surface drainage, as, for example, in England and the Ohio river valley. In the registration area of the United States there were 1,565 deaths from malaria in 1913 ; in Italy, up to 1900, the average number of deaths from this cause annually was 16,000. One cannot obtain a. o^^itS' w Fig. 183. — Plasmodium vivax. (Army Med. School Collection, Washington, D. C.) Fig. 184. — Plasmodium vivax. (Gamete.) (Army Med. School Collection, Washing- ton, D. C.) true picture of the importance of the disease, however, from mortality statistics, since it is not often fatal, and the morbidity is out of pro- portion to the mortality. In many villages, where it is endemic, one- third to one-half the population may have parasites in the blood, most of them without clinical symptoms, yet they are not able to work and the children remain undeveloped and backward. Much of the illness attributed to hoolrworm infection is, in reality, due to latent malaria. The parasites belong to the class of hemosporidia, and are closely related to the coccidia, which are parasites of epithelial cells, while the Plasmodia are parasitic on red-blood cells. There are tw^o divisions of the life cycle; that which occurs in man, the endogenous, asexual or schizogenous, and that which occurs in the mosquito, the exogenous, sexual or sporogenous; for this reason the mosquito is the definitive and the man the intermediate host. SPOROZOA 767 There are three well-recognized forms of the plasmodia, (1) Plas- modium vivax (Grassi and Filetti), causing tertian fever (also called ''benign tertian"); (2) Plasmodium malarice (Laveran), causing quartan fever; (3) Plasmodium falciparum (immaculatum) (Welch), causing the tropical form of malaria, the so-called astivo-autumnal or subtertian. Since in general the life history of the three forms is alike they will be considered together as far as possible. As the ^ \ Fig. 185. — Plasmodium vivax, an Atypical Macrogametocyte. Form interpreted '' by Schaudinn as undergoing partheno- genesis. (Army Med. School Collection, Washington, D. C.) Fig. 186. — Plasmodium vivax. (Army Med. School Collection, Washington, D. C.) details of development cannot be made out easily in fresh specimens, the following decription applies to those stained with some form of the Romanowski stain. Plasmodium vivax. — The parasite of tertian fever has a life cycle lasting forty-eight hours and is easily recognized only when full grown, that is, twenty-four to forty-eight hours after the chill. While a diagnosis may be made on younger forms, it is not so readily made. As its name implies, the Plasmodium vivax is actively ameboid, and pseudopods and irregular outlines characterize the well-grown parasite ; the infected erythrocyte is swollen, often to twice its normal size, the hemoglobin is pale and, especially in spreads in which Mar son's stain has been used, it is so much paler than in the surrounding cells that the infected cell stands out clearly. The part of the cell um 76S Pathogenic protozoa occupied by the parasite is stippled, that is, dotted with reddish gran- ules called Schu,ffner's dots, and, as the swollen red cell and Schuff- ner's dots are found in no other form of malaria, their presence is pathognomonic of tertian. The youngest form, the free merozoite, is rarely seen, but young comet-like forms composed of a particle of red chromatin and a little blue cytoplasm may readily be detected at the height of the fever; that is, a few hours after the chill and sporulation. The round, young schizont as it grows develops early a central vacuole and assumes the shape of a signet ring, the red chromatin dot being the stone. This Fig. 187. — Plasmodium vivax. (Army Med. School Collection, Washington, D. C.) small tertian ring grows rapidly as the fever subsides, and at the same time the infected cell increases in size. Twenty-four hours after the chill the ring has grown so much that it is referred to as the large tertian ring, and its tendency to irregularities of shape and ameboid form becomes apparent, and fine granules of pigment, called melanin or hemozoin, begin to be visible. After thirty -six hours the rings will all have grown into large ameboid forms. After about forty hours the parasite occupies almost the entire cell and the pigment begins to col- lect in masses toward the center. Soon after the first signs of seg- mentation appear, which becomes more and more distinct until fifteen 'to twenty separate segments or merozoites are seen, each composed SPOROZOA 769 of nucleus and cyptoplasm. The pigment of the adult parasite and the unused portion of the cytoplasm are cast off after segmentation as a restkorper, which is promptly phagocyted and such masses ac- cumulate in the spleen, bone marrow and viscera. With rupture of the erythrocyte, at the time of the chill, the merozoites are set free, and if not phagocyted, immediately attack new erythrocytes and the asexual or schizogenous cycle is repeated, until treatment or increas- ing immunity halts or alters the cycle. Fig. 188. — Plasmodium vivax. (Army Med. School Collection, Washing '/>ii, D. C.) In practice it is not unusal to find parasites of different ages in the same film, as some individuals seem to develop in advance of others; in this case, however, there will not be much difference in their appearance. When extreme difference of age is noted in films it is probable that there have been several different inoculations, pro- ducing double or triple infections with quotidian or irregular fever curves, and such cases are not uncommon. As all the forms so far described belong to the schizogenous cycle, they may be called schizonts, or trophozoites of the schizogenous cycle. The sporogenous cycle begins in man and is completed in the mosquito. The earliest sexual forms noted were the so-called ' ' spheres, ' ' large adult parasites, first seen in wet preparations, which did not seg- 770 PATHOGlSNIC PROTOZOA ment with the schizonts. They are now called gametes and after the disease has lasted some time are found in films made at all stages of the fever; that is, they are incapable of further development un- til taken into the stomach of the mosquito. The possibility of parthenogenesis will be referred to latir. In appearance they are round or oval, and in this fever may be twice the size of the red cell. As a rule a narrow margin of red cell is visible after Ro- manowski stains, although the gamete may lie free in the plasma. Unlike the schizonts, the gametes have the pigment uniformly dis- tributed throughout the body and there is no indication of segmen- tation. The young sporonts are distinguished from schizonts by the absence of the vacuole, and, when a little older, by a larger amount of hemozoin. Plasmodium malariae. — The quartan parasite has a life cycle of seventy -two hours, or twenty-four hours longer than the tertian, and r ^ Fig, 189. — Plasmodium malaria. (Army Fig. 190. — Plasmodium malaria. (Army Med. School Collection, Washington, Med. School Collection, Washington^ D. C.) D. C.) the paroxysms come on every third day, or, according to the Italian method of reckoning time, on the fourth day. The young rings of the Plasmodium malarice are indistinguishable from young tertian rings, but the diagnosis may be made on older forms. The bleaching, enlarge- ment and stippling of the erythrocyte characteristic of tertian is never found in quartan fever, the infected erythrocyte being almost normal in appearance. The well-grown quartan parasite does not show ameboic changes but assumes a band form, more or less wide, stretch- ing across the red cell from b( rder to border ; with increasing age the SPOROZbA 771 band widens until the parasite is nearly square and the hemozoin ac- cumulates toward the center. Segmentation gives rise to almost sym- metrical "daisy" forms, showing six to eight or, rarely, fourteen mero- zoites. Parasites of different ages may be found, as in tertian, and it is characteristic of quartan fever that examples of all stages of the life cycle may be found at the proper time in the peripheral cir- culation. Gametes differ from tertian mainly in size, since they are never larger than the normal erythrocyte until after the latter has ruptured, but when free in the plasma it is practically impossible to distinguish them from tertians. Fig. 191. — Plasmodium malari^e. (Army Med. School Collection, Wash- ington, D. C.) Fig. 192. — Plasmodium falciparum. (X 1500.) (Army Med. School Col- lection, Washington, D. C.) Plasmodium falciparum. — The parasite of aestivo-autumnal fever, Plasmodium falciparum, differs considerably from the two forms al- ready described; the life cycle varies between twenty-four and forty- eight hours, and, at least in new itifections, only ring forms are jEound in the peripheral blood, although at a later stage crescentic gametes may be present. The youngest aestivo-autumnal rings, found at the height of the fever, are more delicate than the young tertians. As the temperature falls the rings increase in size, but without change of form ; the growth is not uniform, but occurs as a thick cres- centic swelling on the convex surface of the ring, and occasionally more than one such swelling is present. The large aestivo-autumnal ring, found after the febrile paroxysm has passed, occupies one-third to one-half the red cell, which is never swollen nor stippled, as in ter- tian, and the parasite is never band-like, as in quartan. Segmenting 772 PATHOGENIC PROTOZOA parasites are almost never seen in the peripheral blood in aestivo- autumnal fever, though in tertian they are common and in quartan numerous. If, however, films are prepared at autopsy from the spleen, liver, bone marrow and brain, enormous numbers of segment- ing forms, together with all other stages of the parasite, may be found. The full grown segmenter occupies one-third to one-half the cell and shows a collection of hemozoin in large blocks in the center, The merozoites vary in number from eight to twenty -five. In addition to the small and large rings the peripheral blood shows, after the fever has lasted sufficiently long, the sexual forms or gametes. The infected erythrocyte is never stippled nor swollen, but, on the contrary, may appear shrunken. Both the micro- and macrogametocytes in sestivo- autumnal fever are crescentic in shape, their length being about one and one-half and the width about one-half that of an erythrocyte; the pigment is collected toward the center, which is rather paler in stained specimens than the poles. At first sight the gametes appear to lie free in the plasma, yet in stained specimens a rim or rib of the pale red cell may be seen on the concave side. When liberated from the erythrocyte the gamete becomes first spindle-shaped and finally oval or round. The male crescent is short and broad, and the female relatively long and slender. The Finer Structure of the Plasmodia. — The finer details, w'nich are only hinted at in fresh specimens and in those stained with Man- son's stain, can be studied to advantage in those stained with some one of the many modifications of the Romanowski stain, such as that of Wright, Hastings, MacNeal or Giemsa. The tertian parasite, which lies in a red cell, may be seen to be divided into a blue cytoplasm and a brilliant red nucleus, and it would be well for the novice to remember that these three conditions m.ust be satisfied before the diagnosis of malaria can be made; the principal stumbling-block is the blood platelet, often found overlying a red cell, but it, although possessing a ragged blue cytoplasm, has al- FiG. 193. — Plasmodium falciparum (X 1500.) (Army Med. School Col- lection, Washington, D. C.) SPOROZOA 773 ways a relatively large purple nucleus. The chromatin of the young rings is usually present as a single dot, but two such dots are frequently seen. In older forms of the tertian and quartan parasites the various changes found in mitosis may be followed in the nucleus. The whole schizogenous cycle may be followed by taking blood smears from a single case of malaria at intervals of three or four hours for forty- eight hours for tertian and aestivo-autumnal, and for seventy-two hours for quartan. Fig. 194. — Plasmodium falciparum, Male Crescent. (Army Med. School Col- lection, Washington, D. C.) Two forms of sporonts or gametes may be seen ; in one the quantity of chromatin is large and the cytoplasm pale blue, while in the other the reverse is found, the nuclear chromatin is comparatively small in quantity and the cytoplasm, being rich in nutrient material, stains deeply. The first form, with abundant chromatin, is the male, or microgametocyte, and the latter the female, or macrogametocyte. The differentiation between schizont and sporont may be made while the parasites are still quite young; since the schizont is characterized by the presence of a nutrient vacuole, and the sporont, of equal age, while 774 - PATHOGENIC PROTOZOA lacking the vacuole, shows a greater amount of hemozoin, which is never concentrated in the. center of the parasite but is scattered equally throughout the body. The cytoplasm of the sporont is less fluid than that of the schizont and shows no tendency to ameboid motion. The chromatin is relatively large in amount and, although broken up more or less into granules and threads, shows no real ten- dency to segment or disperse, but remains a compact mass. The quartan parasite, when stained with Wright's or a similar preparation, shows quite regular and symmetrical segmentation, usu- ally into eight merozoites. The distinction between schizont and sporont and between male and female gametes may be made on the same grounds as in tertian. In aestivo-autumnal fever the chromatin dot in the young ring is often doubled, or even trebled, and in general is large and stains brilliantly. The adult and half -grown gametes may be differentiated into male and female by the criteria already given. The Examination of Fresh Blood. — Directions have already been given (Chap. LIX) for making wet preparations and if, by ringing the cover-glass with vaseline, drying be prevented, the preparations will keep and may be studied for hours. In tertian fever the young ring forms are at first difficult to detect, unless the amount of light going through the microscope be cut down to the minimum. As the parasite grows older, an increasing number of hemozoin granules ap- pear, and since they are in constant motion the parasite is readily de- tected. Its cytoplasm is delicate, and with very young parasites is difficult to distinguish from the red cell itself ; older parasites, however, develop pseudopods, which are constantly projected and retracted, and the entire organism shows active movements, rendering it easy to see. The pigment continues to increase, and in the gametes is abundant and in constant motion; the gametes, however, fail to show any ame- boid changes, and the protoplasm is stiff and rigid with a regular, un- broken margin. At times a clear refractile spot is seen, which is the nucleus. The infected erythrocyte is pale and swollen. Even in un- stained preparations the sexes may be distinguished ; the microgameto- cyte is about the size of a red cell, the cytoplasm is hyaline, and after the preparation has been made ten to twenty minutes the flagella, or microgametes, . may be seen thrashing about in the parasite. After repeated attempts four to eight microgametes rupture the cell and emerge. The macrogamete is larger than a red cell and is finely granular and no exflaggelation of microgametes occurs. SPOROZOA 775 In quartan malaria the differences already described in stained blood may be easily followed. In agstivo-autumnal fever the diagnosis with fresh blood is much more difficult in new infections because of the relative scarcity of the parasites in the peripheral blood and the exceedingly small size of the young rings, the absence of hemozoin in them, and the very slight ameboid motion. The older rings are larger, contain some pigment and are more easily seen. The infected erythrocyte is never pale nor swollen, but, on the contrary, may be shrunken and brassy in color. The crescentic gametes are readily detected, and the sexes may be differentiated by their shape and the hyaline or granular character of the cytoplasm. Incubation Period of the Malarial Fevers. — Two methods have been used to determine this point — the injection of infected malarial blood, and biting experiments with infected anophelines. By the first method the incubation period was eighteen days (the longest) for quartan, three days (the shortest) for aestivo-autumnal, and ten days for tertian. By the second method aestivo-autumnal was nine to twelve days, and tertian fourteen to nineteen days. Since aestivo- autumnal is the only parasite which can complete its cycle in twenty- four hours, the short incubation period is easily understood; on the other hand, the long life cycle of quartan, seventy -two hours, explains its slower development. Clinical descriptions of the malarial fevers may be found in the standard text-books on medicine, and it is only necessary here to refer briefly to the various forms found in practice. The classical malarial fever consists of a series of paroxysms, following one another with a definite periodicy, daily, every other day, or every third day. Each paroxysm is ushered in by a pronounced chill, which is sometimes pre- ceded by malaise, headache and lassitude. The chill lasts from ten minutes to an hour or more, and the patient wraps himself up in heavy blankets. During the chill the temperature begins to rise and within a few hours reaches its high point, 103° to 106°, and then falls slowly to normal during the next few hours. The decline of the fever is accompanied by a profuse perspiration. Successive parox- ysms may occur at exactly the same hour of the day, or may antici- pate, fehris anteponens, or be delayed an hour or more, fehris post- ponens. The sequence of events, therefore, in a typical malarial paroxysm is malaise, chill, "fever and sweat, followed by a period of apparent well-being. 776 PATHOGENIC PROTOZOA 1. Tertian malaria is distinguished by a chill and fever occurring every other day, the patient feeling quite well on fever-free days, A double tertian infection occurs not infrequently, giving a daily, or quotidian, chill and fever with no free day. 2. Quartan fever, which is relatively rare, gives a chill and fever every third day, with two fever-free days. In this disease also there may be double or even triple infections, giving a quotidian or irreg- ular type of fever. 3. ^stivo-autumnal fever (subtertian, or malignant tertian) shows an irregular temperature curve, the cycle varying from twenty-four to forty-eight hours. By some authors the disease is divided into two forms, quotidian and tertian; and as multiple infections are com- mon the resulting fever curve may be irregular or continuous and the chill entirely absent. In contrast to the regular intermittency of ter- tian and quartan this form is often remittent, the temperature curve never dropping to normal. 4. Mixed infections with any two of the above fevers are often found in bad malarial regions in the tropics. 5. Latent malaria is also not infrequent, in which the patient, hav- ing no symptoms of the disease, consults a physician for some other reason. 6. The carrier state is found among natives or persons long resi- dent in malarial regions, and, aside from the presence of a large spleen and some secondary anemia, may present no symptoms. It is particu- larly common among native children, tramps and vagabonds. It is not uncommon to find fifty to one hundred per cent of the children in a native village harboring the parasite. THE DEVELOPMENT OF THE HUMAN MALARIAL PARASITE IN THE MOSQUITO {Sexual half of the life cycle , Sporogonie) For this stage to be successful the mosquito must bite a malarial patient with gametes in his blood, for if the patient be one in the first stage of the disease, with only schizonts in his blood, no infection of the mosquito will take place, since the schizonts all perish in its stomach. On the contrary, if the mosquito takes blood from a per- son who has been ill with malaria for some time, or from an appar- ently healthy carrier, the schizonts die as usual, but the gametes find in the mosquito stomach for the first time conditions suitable for their further development. SPOROZOA 777 The various stages may be studied by causing suitable species of anophelines to bite persons with many gametes in their blood, and then dissecting the stomach and observing the changes which take place there. The development is visible in unstained specimens with high, dry lenses. Since there is no essential dift'erence in the develop- ment of the three forms of malaria in the mosquito, they will be con- sidered together. The first stage has already been described in dis- cussing the appearance and behavior of gametes in fresh blood. In the mosquito the process may be followed further; the macrogamete, freed from its enveloping red cell, projects a little mound on its sur- face, and this apparently attracts the microgametes to its neighbor- hood. Into this microphyle one, but never more, of the flagella pene- trates, following which the mound is instantly retracted. The fer- tilized macrogamete, now called a "zygote," soon develops the powei ^^