Full text of "Manual of bacteriology"

See other formats

'••.

BIOLvOGY

LIBRA iv i

MANUAL

BACTERIOLOGY

ROBERT MUIR, M.A., M.D., F.R.C.R(ED.)

I'KOFKSSOR OK PATHOLOGY, UNIVERSITY OF GLASGOW

AM»

JAMES RITCHIE, M.A., M.D., F.R.C.P.(Ea)

MI'KltlNTKNDK.XT OK TI1K ROYAL COLLKGK OK PHYSICIAN'S* LABORATORY, KDIXBrROII
H.UMKIM.V IMiolKSSOR OK 1'ATHOLOOY IX TIIK I'XIVKRSITY OK OXKORIt

FIFTH EDITION

WITH ONE HUNDRED ,(• SKVEXTY-FOUR IL7ArSTJiATIONS

7AT Till- TKXT
AND SIX COLOURED PL AT EX

NEW YORK
THE MACMILLAN COMPANY

Main Lib, I Of / r

Agric. Dept.

fi'OLOGY

LIBRARY

PREFACE TO THE FIFTH EDITION.

DURING the three years which have elapsed since the publication
of the previous edition, important additions to knowledge have
been made in nearly every department of bacteriological research.
We have again endeavoured to incorporate these as fully as
possible, having in view our primary object, namely, to supply a
Manual for students and practitioners of medicine.

It is impossible to refer in detail to the new matter introduced,
as this occurs in practically every chapter. It may, however, be
mentioned that in dealing with technique, we have introduced a
new chapter in which the more recent methods of investigating
the properties of serum and allied subjects are described. While
it is necessary for students of medicine to have a general know-
ledge of such subjects, those commencing independent bacterio-
logical research may benefit from the details given in this
department.

\Vi- have also in the present edition grouped together in a
new chapter the pathological conditions with which spirochyetes
are associated, but as the biological relationships of these organ-
isms to kindred forms are still not completely determined, we
have retained this chapter in its former position.

We have transferred the consideration of Yellow Fever to the
Appendices, which include diseases of protozoal origin and con-
• lit ions in which the nature of the infective agent is still un-
known. In further following this principle, we have added

258'809

vi PREFACE TO THE FIFTH EDITION

new Appendices dealing with Acute Poliomyelitis, Phlebotomus
Fever, and Typhus Fever.

Several new illustrations will be found in the text, and we
have also added a series of coloured plates which have been
reproduced from drawings by Mr. Richard Muir of the Depart-
ment of Pathology, University of Edinburgh. f

November 1910.

PREFACE TO THE FIRST EDITION.

THE science of Bacteriology has, within recent years, become
so extensive, that in treating the subject in a book of this size
we are necessarily restricted to some special departments, unless
the description is to be of a superficial character. Accordingly,
as this work is intended primarily for students and practitioners
of medicine, only those bacteria which are associated with
disease in the human subject have been considered. We have
made it a chief endeavour to render the work of practical utility
for beginners, and, in the account of the more important
methods, have given elementary details which our experience in
the practical teaching of the subject has shown to be necessary.

In the systematic description of the various bacteria, an
attempt has been made to bring into prominence the evidence
of their having an etiological relationship to the corresponding
diseases, to point out the general laws governing their action as
producers of disease, and to consider the effects in particular
instances of various modifying circumstances. Much research
on certain subjects is so recent that conclusions on many points
must necessarily be of a tentative character. We have, therefore,
in our statement of results aimed at drawing a distinction
between what is proved and what is only probable.

In an Aj-pemlix we have treated of four diseases; in two of
tin -so the causal organism is not a bacterium, whilst in the other
two its nature is not yet determined. These diseases have been

vii

viii PREFACE TO THE FIRST EDITION

included on account of their own importance and that of the
pathological processes which they illustrate.

Our best thanks are due to Professor Greenfield for his kind
advice in connection with certain parts of the work. We have
also great pleasure in acknowledging our indebtedness to
Dr. Patrick Manson, who kindly lent us the negatives or pre-
parations from which Figs. 163-168 have been executed.

As we are convinced that to any one engaged in practical
study, photographs and photomicrographs supply the most useful
and exact information, we have used these almost exclusively in
illustration of the systematic description. These have been
executed in the Pathological Laboratory of the University of
Edinburgh by Mr. Richard Muir. The line drawings were
prepared for us by Mr. Alfred Robinson, of the University
Museum, Oxford.

To the volume is appended a short Bibliography, which,
while having no pretension to completeness, will, we hope/ be of
use in putting those who desire further information on the track
of the principal papers which have been published on each of
the subjects considered.

June 1897.

CONTENTS.

CHAPTER I.
GENERAL MORPHOLOGY AND BIOLOGY.

PAGE

INTRODUCTORY — Terminology — Structure of the bacterial cell —
Reproduction of bacteria — Spore formation — Motility —
Minuter structure of the bacterial protoplasm — Chemical
composition of bacteria — Classification — Food supply — Re-
lation of bacteria to moisture, gaseous environment, tempera-
ture, and light — Conditions affecting bacterial motility —
Effects of bacteria in nature— Methods of bacterial action-
Variability among bacteria ..... 1

CHAPTER II.
METHODS OP CULTIVATION OF BACTERIA.

Introductory — Methods of sterilisation — Preparation of culture
media — Use of the culture media — Methods of the separation
of aerobic organisms— Principles of the culture of anaerobic-
organisms — Miscellaneous methods — General laboratory
rules ........ 26

CHAPTER III.
MICROSCOPIC METHODS.

The microscope — Examination of hanging-drop cultures — Film pre-
parations— Examination of bacteria in tissues — The cutting
of sections— Staining principles— Mordants and decolorisers
— Formula? of stains— Gram's method and its modifications
— Stain for tubercle and other acid-fast bacilli— Staining of
spores, capsules, and flagella— The Komaiiowsky stains . Ul

CONTENTS

CHAPTER IV.

EXAMINATION OF SERUM — PREPARATION OF VACCINES —
GENERAL BACTERIOLOGICAL DIAGNOSIS— INOCULATION
OF ANIMALS.

PAGK

Observation of agglutination and sedimentation — Opsonic methods
— Method of measuring the phagoeytic capacity of the leuco-
cytes— Bactericidal methods — Hfemolytic tests — Fixation and
deviation of complement — Wassermann reaction — Preparation
of vaccines — Wright's method of counting bacteria in dead cul-
tures— General bacteriological diagnosis — Routine procedure
— Inoculation of animals — Autopsies on animals . . 117

CHAPTER V.
BACTERIA IN AIR, SOIL, AND WATER. ANTISEPTICS.

Air : Methods of examination — Soil : Methods of examination —
Varieties of bacteria in soil. Water : Methods of examination
— Bacteria in water — Bacteriology of sewage — Antiseptics :
Methods of investigation — The action of antiseptics — Certain
particular antiseptics . . . . . .147

CHAPTER VI.

RELATIONS OF BACTERIA TO DISEASE — THE PRODUCTION
OF TOXINS BY BACTERIA.

Introductory — Conditions modifying pathogenicity — Modes of
bacterial action — Tissue changes produced by bacteria — Local
lesions — General lesions — Disturbances of metabolism by
bacterial action — The production of toxins by bacteria, and
the nature of these — Allied vegetable and animal poisons —
The theory of toxic action ..... 175

CHAPTER VII.
INFLAMMATORY AND SUPPURATIVE CONDITIONS.

The relations of inflammation and suppuration — The bacteria of
inflammation and suppuration— Experimental inoculation —
Lesions in the human subject — Mode of entrance and spread
of pyogeuic bacteria — -Ulcerative endocarditis — Acute suppur-

CONTENTS xi

ative periostitis— Erysipelas — Conjunctivitis — Acute rheu-
matism— Vaccination treatment of infections by the pyogenic
cocci — Methods of examination in inflammatory and suppur-
ative conditions ... . 200

CHAPTER VIII.
INFLAMMATORY AND SUPPURATIVE CONDITIONS, continued :

THE ACUTE PNEUMONIAS, EPIDEMIC CEREBRO-SPINAL
MENINGITIS.

Introductory — Historical — Bacteria in pneumonia — Fraenkel's
pncumococcus — Fried laender's pneumococcus — Distribution of
pnrumohacteria — Experimental inoculation — Pathology of
pneumococcus — Methods of examination. Epidemic cerebro-
spinal meningitis— Serum reactions— Allied diplococci . 224

CHAPTER X.

GONORRHOEA AND SOFT SORI:.

The gonococcus — Microscopical characters — Cultivation — Com-
parison with meningococcus — Relations to the disease — Its
toxin — Distribution — Gonococcus in joint affections — Methods
of diagnosis — Soft sore — Microscopical characters and culti-
vation of bacillus ...... 249

CHAPTER X.
TUBERCULOSIS.

IlMnrical — Tuberculosis in animals — Tubercle bacillus — Staining
relictions— Cultivation of tubercle bacillus— Powers of resist-
ance—Action on the tissues — Histology of tuberculous nodules
— Distribution of bacilli — Bacilli in tuberculous discharges —
Experimental inoculation — Varieties of tuberculosis — Other
acid-fast bacilli — Action of dead tubercle bacilli — Sources of
human tuberculosis— Specific reactions of the tubercle bacillus
—Phenomena of supersensitiveness — Tuberculin reactions —
Toxins of the tubercle bacillus — Koch's tuberculin — Immunity
phenomena in tuberculosis— Koch's Tuberculin-R — Therapeutic
application of the tuberculins — Active inimuni*ation associated
with opsonic observations — Antitubercular sera — Methods of
••xaiiiiiialiuii 260

xii CONTENTS

CHAPTER XL
LEPROSY.

PAGE

Pathological changes — Bacillus of leprosy — Position of the bacilli

— Relations to the disease — Methods of diagnosis . . 297

CHAPTER XII.

GLANDERS AND RHINOSCLEROMA.

Glanders : The natural disease — The glanders bacillus — Cultiva-
tion of glanders bacillus — Powers of resistance — Experimental
inoculation — Action on the tissues — Mode of spread — Serum
reactions — Mallein and its preparation — Methods of examina-
tion— Rhinoscleroma . . . . . 306

CHAPTER XIII.

ACTINOMYCOSIS AND ALLIED DISEASES.

Characters of the actinomyces — Tissue lesions —Distribution of
lesions — Cultivation of actinomyces — Varieties of actinomyces
and allied forms — Experimental inoculation — Methods of
examination and diagnosis — Madura disease . . .317

CHAPTER XIV.
ANTHRAX.

Historical summary — Bacillus anthracis — Appearances of cultures
— Biology — Sporulation — Natural anthrax in animals — Ex-
perimental anthrax — Anthrax in man— Pathology — Toxins of
the bacillus anthracis — Mode of spread in nature — Immunisa-
tion of animals against anthrax — Methods of examination . 331

CHAPTER XV.

TYPHOID FEVER — BACILLI ALLIED TO THE TYPHOID BACILLUS.

Introductory — Bacillus coli communis — Culture reactions —
Isolation and recognition of B. coli — Pathogenic properties —
Bacillus typhosus — Isolation uud appearances of cultures —

CONTENTS xiii

reactions — Pathological changes in typhoid fever —
Immunisation of animals — Etiological relationships of bacillus
typhosus — Epidemiology of typhoid fever — Typhoid carriers
— Serum diagnosis of typhoid fever — Vaccination against
typhoid — Methods of examination — Paratyphoid and food-
poisoning bacilli — The paratyphoid bacillus — Bacillus enteri-
tidis (Gaertner) — The psittacosis bacillus — Danysz's bacillus
and rat viruses— Bacillus dysenterise — Bacillus enteritidis
sporogenes — Summer diarrhoea — General review of coli-typhoid
group ........ 350

CHAPTER XVI.
DIPHTHERIA.

Historical — General facts — Bacillus diphtherire — Microscopical
characters — Distribution — Association with other organisms
— Cultivation — Powers of resistance — Inoculation experiments
— The toxins of diphtheria — Variations in virulence of bacilli —
Bacilli allied to the diphtheria bacillus — Summary of patho-
genic action— Methods of diagnosis .... 396

CHAPTER XVII.

TETANUS.

Introductory — Historical — Bacillus tetani — Isolation of bacillus
tetani — Characters of cultures — Conditions of growth — Patho-
genic effects — Experimental inoculation — Tetanus toxins —
Antitetanic serum — Methods of examination — Malignant
( edema — Characters of bacillus — Experimental inoculation —
Methods of diagnosis — Bacillus botulinus — Quarter-evil —
Bacillus aerogenes capsulatus — Fusiform anaerobic bacilli . 415

CHAPTER XVIII.
CHOLERA.

Introductory — The cholera spirillum — Distribution of the spirilla —
Cultivation — Powers of resistance — Experimental inoculation
— Toxins of cholera spirillum — Inoculation of human subject
— Immunity — Methods of diagnosis — General summary —
Other spirilla resembling the cholera organism — Metchnikoflf's
spirillum — Finkler and Prior's spirillum — Deneke's spirillum 446

xiv CONTENTS

CHAPTER XIX.

INFLUENZA, WHOOPING-COUGH, PLAGUE, MALTA FEVER.

PAGE

Influenza bacillus — Microscopical characters — Cultivation — Dis-
tribution— Experimental inoculation — Methods .of examina-
tion— Whooping-cough bacillus — Microscopical characters —
Pathogenic effects — Methods of examination — Bacillus of
plague — Microscopical characters — Cultivation — Anatomical
changes produced and distribution of bacilli — Experimental
inoculation — Paths and mode of infection — Toxins, immunity,
etc. — Preventive inoculation — Anti-plague sera — Methods of
diagnosis — Malta fever — Micrococcus melitensis — Relations to
the disease — Mode of spread of the disease — Methods of
diagnosis . . . . . . . 467

CHAPTER XX.

DISEASES DUE TO SPIROCH^TES — THE RELAPSING FEVERS,
SYPHILIS, AND FRAMBCESIA.

Relapsing fever and African tick fever — Characters of the spirochsete
— Relations to the disease — Immunity — African tick fever —
Transmission of the disease— Syphilis — Microscopic characters
of spirochsete pallida — Distribution — Cultivation — Trans-
mission of the disease --Serum Diagnosis — Wassermann
reaction — Frambcesia or Yaws . 494

CHAPTER XXI.
IMMUNITY.

Introductory — Acquired immunity — Artificial immunity —
Varieties — Active immunity — Methods of production — At-
tenuation and exaltation of virulence — Passive immunity —
Action of the serum — Antitoxic serum — Standardising of
toxins and of anti-sera — Nature of antitoxic action — Ehrlich's
theory of the constitution of toxins — Antibacterial serum —
Bactericidal and lysogenic action — Hsemolytic and other
sera — Methods of the hsemoly tic tests — Opsonic action — Ag-
glutination— Precipitins — Therapeutic effects of anti-sera —
Theories as to acquired immunity — Ehrlich's side-chain theory
— Theory of phagocytosis - — Natural immunity — Natural
bactericidal powers — Natural susceptibility to toxins —
Supersensitiveness or anaphylaxis — The serum disease in man 512

CONTENTS xv

APPENDIX A.

SMALLPOX AND VACCINATION.

PAGE

.Irnnerian vaccination — Relationship of smallpox to cowpox —
Micro-organisms associated with smallpox — The nature of
vaccination . . . . . . . 565

APPENDIX B.
HYDROPHOBIA.

Introductory— Pathology — The virus of hydrophobia— Prophylaxis

— Antirabic serum — Methods . . 573

APPENDIX C.
MALARIAL FEVER.

The malarial parasite — The cycle of the malarial parasite in man
— The cycle in the mosquito — Varieties of the malarial para-
site— General considerations — The pathology of malaria —
Methods of examination . . 585

APPENDIX D.

AMOSBIC DYSENTERY.

AniM-bic dysentery — Characters of the amoeba — Cultivation of the
aiiM'bji'- Distribution of the amoebse — Experimental inocula-
tion— Methods of examination ... . G02

APPENDIX E.

TRYPANOSOMIASIS — LEISHMANIOSLS — PIROPLASMOSIS.

The pathogenic trypanosomes — Morphology and biology of the
trypanosomata — Trypanosoma Lewisi— Nagana or tse-tse fly
disease — Trypanosome of sleeping sickness — Trypanosoma
Cruzi — Leishmaniosis — Leishmania Donovani — Leishmania
infantum — Leishmania tropica — Histoplasma capsulatum —

. . . . . . .610

xvi CONTENTS

APPENDIX F.

T7. PAGE

YELLOW FEVER .... g39

APPENDIX G.

ACUTE POLIOMYELITIS . . 544

APPENDIX H.
PHLEBOTOMUS FEVER • . . g46

APPENDIX J.

TYPHUS FEVER . . . g48

BIBLIOGRAPHY . . 649

INDEX . 673

LIST OF COLOURED PLATES.

PLATE I.
fa.

1. Film of pus, containing staphylococci and streptococci.

2. Fraenkel's pneumococcus in sputum.

.'>. Meningococcus in epidemic cerebro-spinal fever.

4. Film from a scraping of throat in Vincent's angina, showing fusiform

bacilli and spirocluetes.

5. Gouorrhoeal pus, showing gonococci and staphylococci.

PLATE II.

6. Spirochaete pallida, case of congenital syphilis.

7. Tubercle bacillus and other bacteria in sputum.

8. Leprous skin, showing clumps of bacilli in the cutis.

9. Leprous granulation tissue, showing bacilli.

PLATE III.

10. Stivptothrix actinomyces.

11. Anthrax bacilli.

12. Bacillus diphtheria;.

13. Bacillus diphtheria (involution forms).
1 l. Ilofmann's pseudo-diphtheria bacillus.

15. Typhoid bacillus, showing flagella.

xvii

xviii LIST OF COLOURED PLATES

PLATE IV.

FIG.

16. Negri bodies in nerve cells in rabies.

17. Bacillus pestis (involution forms).

18. Spirochaete of relapsing fever.

19. The cholera spirillum, showing Hagella.

20. Bacillus tetani, showing spores.

PLATE V.

21. The Parasite of Mild Tertian Malaria.

Cycle I. (Schizogony). Asexual cycle in the human blood.
Cycle II. (Sporogony). Sexual cycle in the mosquito.

22. The Parasite of Malignant Malaria.

PLATE VI.

23. Entamoeba histolytica in pus, from tropical abscess of liver.

24. Leishman-Donovau bodies, from a case of Kala-azar.

25. Trypanosoma Gambiense.

LIST OF ILLUSTRATIONS IN TEXT.

FIO. PAGE

1. Forms of bacteria . . . . .13

•_'. I lot-air steriliser ... .28

:}. Koch's steam steriliser . . . . 28

4. Autoclave . . . . . . .30

5. Steriliser for blood serum . . . . .31

6. Meat press ....... 32

7. Hot-water funnel ...... 36

8. Blood serum inspissator . . . • . .41

9. Potato jar . . . . 45

10. Cylinder of potato cut obliquely . . . .45

11. Ehrlich's tube containing piece of potato . • . .46

12. Apparatus for filling tubes '\ . . . .53

13. Tubes of media . . . . . .53

14. Platinum wires in glass handles . . . .54
I."-. Method of inoculating solid tubes .... 55

16. Rack for platinum needles ..... 56

17. Petri's capsule . . . . . . . 57

1 x. Koch's levelling apparatus for use in preparing plates . . 58

19. Koch's levelling apparatus . . . . .58

20. Esmarch's tube for roll culture . ... . .60

21. Apparatus for supplying hydrogen for anaerobic cultures . 63

22. Esmarch's roll-tube adapted for culture containing anaerobes . 64

23. Bulloch's apparatus for anaerobic plate cultures . .64

24. Flask for anaerobes in liquid media . . . .67

25. Flask arranged for culture of anaerobes which develop gas . 68

26. Tubes for anaerobic cultures on the surface of solid media . 68

27. Slides for hanging-drop cultures . . . .69

28. Apparatus for counting colonies . . . .70

29. Wright's 250 c.mm. pipette fitted with nipple . . 71

30. Geissler's vacuum pump for filtering cultures . 75

31. Chamberland's candle and flask arranged for filtration . 75
:;± < ,'h ^label-land's bougie with lamp funnel . ... 76

xx LIST OF ILLUSTRATIONS IN TEXT

FIG. PAGE

33. Bougie inserted through rubber stopper . . .76

34. Muencke's modification of Chamberland's filter . . 77

35. Flask for filtering small quantities of fluid . 78

36. Tubes for demonstrating gas-formation by bacteria . . 81

37. Geryk air-pump for drying in vacua . .85

38. Reichert's gas regulator . . . . .86

39. Hearson's incubator for use at 37° C. . . . . 87

40. Cornet's forceps for holding cover-glasses . . .94

41. Needle with square of paper on end for manipulating paraffin

sections . . . . . 99

42. Syphon wash-bottle for distilled water . 102

43. Wright's 5 c. mm. pipette ". . . . . . 118

44. Tubes used in testing agglutinating and sedimenting properties

of serum . . . . . ; 119

45. Wright's blood-capsule . . . . 124

46. Test-tube and pipette arranged for obtaining fluids .containing

bacteria . ".. . . ,: - . .136

47. Hollow needle for intraperitoneal inoculations . .. ':' . 143

48. Hesse's tube . . . . • . . j - . 1 48

49. Petal's sand filter . . . 149

50. Staphylococcus pyogenes aureus, young culture on agar.

xlOOO .... . ,203

51. Two stab cultures of Staphylococcus pyogenes aureus in gelatin 203

52. Streptococcus pyogenes, young culture on agar. x 1000 ;» 204

53. Culture of the streptococcus pyogenes on an agar plate ji^- 205

54. Bacillus pyocyaneus ; young culture on agar. x 1000 I' 208

55. Micrococcus tetragenus. x 1000 . . . -."... 209

56. Streptococci in acute suppuration, x 1000 . . . 212

57. Minute focus of commencing suppuration in brain, x 50 . 214

58. Secondary infection of a glomerulus of kidney by the Staphylo-

coccus aureus. x 300 ... . . . 215

59. Section of a vegetation in ulcerative endocarditis, x 600 . 217

60. Film preparation from a case of acute conjunctivitis, showing

the Koch-Weeks bacilli, x 1000 ..... 219

61. Film preparation of conjunctival secretion showing the diplo-

bacillus of conjunctivitis, x 1000 . . . -. 220

62. Film preparation of pneumonic sputum, showing numerous

pneumococci (Fraenkel's). x 1000 . . . ' . 227

63. Friedlander's pneumobacillus, from exudate in a case of

pneumonia, x 1000 . „ • . . . 228

64. Fraenkel's pneumococcus in serous exudation, x 1000 . 228

65. Stroke culture of Fraenkel's pneumococcus on blood agar . 229

66. Fraenkel's pneumococcus from a pure culture on blood agar.

xlOOO . . . . . . .230

67. Stab culture of Friedlander's pneumobacillus .,-.-, . 232

LIST OF ILLUSTRATIONS IN TEXT xxi

FIG. PAGE

68. Friedliinder's pneumobacillus, from a young culture on agar.

xlOOO . . 233

69. Capsulated pneumococci in blood taken from the heart of a

rabbit. x 1000 ... . 236

70. Film preparation of exudation from a case of meningitis.

xlOOO . . 242

71. Pure culture of diplococcus intracellulai is . . . 243

72. Portion of film of gonorrhceal pus. x 1000 . 250

73. Colonies of gouococcus on serum agar . . . 251

74. Gonococci, from a pure culture on blood agar. x 1000 . 251

75. Film preparations of pus from soft chancre, showing Ducrey'.s

bacillus. x!500 .... .258

76. Ducrey'.s bacillus x 1500 . . 259

77. Tubercle bacilli, from a pure culture on glycerin agar. x 1000 262

78. Tubercle bacilli in phthisical sputum, x 1000 263

79. Cultures of tubercle bacilli on glycerin agar . . . 266
"80. Tubercle bacilli in section of human lung in acute phthisis.

xlOOO ..... .270

81. Tubercle bacilli in giant-cells, x 1000 . . . 271

82. Tubercle bacilli in urine, x 1000 . . . 272

83. Hoeller's Timothy-grass bacillus, x 1000 . .279

84. Cultures of acid-fast bacilli grown at room temperature . 279

85. Smegma bacilli, x 1000 . . . .280

86. Section through leprous skin, showing the masses of cellular

granulation tissue in the cutis. x 80 . . . 298

S7. Superficial part of leprous skin, x 500 . . 300

88. High-power view of portion of leprous nodule showing the

arrangement of the bacilli within the cells of the granula-
tion tissue, x 1100 ..... 301

89. Glanders bacilli from peritoneal exudate of guinea-pig, x 1000 308

90. Glanders bacilli, x 1000 . . 309

91. Actinomycosis of human liver, x 500 . . 319

92. Actinomyces in human kidney, x 500 . . . 320

93. Colonies of actinomyces. x 60 . . . ->•-<. 321

94. Cultures of the actinomyces on glycerin agar. x 60 . . 324

95. Actinomyces, from a culture on glycerin agar. x 1000 . 325

96. Shake cultures of actinomyces in glucose agar. . • 326

97. Section of a colony of actinomyces from a culture in blood

serum, x 1500 . . . . . . 326

98. Streptothrix Madura, x 1000. . . . .329

99. Surface colony of the anthrax bacillus on an agar plate.

x30. . . . . . .333

100. Anthrax bacilli, arranged in chains, from a twenty-four

hours' culture on agar at 37° C. x 1000 . . . 334

101. Stab culture of the anthrax bacillus in peptone-gelatin . 334

xxil LIST OF ILLUSTRATIONS IN TEXT

FIG. PAGE

102. Anthrax bacilli containing spores, x 1000 . . 336

103. Scraping from spleen of guinea-pig dead of anthrax. x 1000 339

104. Portion of kidney of a guinea-pig dead of anthrax, x 300 .' 340

105. Bacillus coli communis. xlOOO . • '.-. . -... 351

106. A large clump of typhoid bacilli in a spleen, x 500 . . 357

107. Typhoid bacilli, from a young culture on agar, showing some

filamentous forms. xlOOO. .... 358

108. Typhoid bacilli, from a young culture on agar, showing

flagella. . x 1000 .... .359

109. Culture of the typhoid bacillus and of the bacillus coli . 360

110. Colonies of the typhoid bacillus in a gelatin plate, x 15 . 361

111. Film preparation from diphtheria membrane ; showing

numerous diphtheria bacilli, x 1000 . . 398

112. Section through a diphtheritic membrane in trachea, show-

ing diphtheria bacilli. xlOOO . . . 399

113. Cultures of the diphtheria bacillus on an agar plate . . . 401

114. Diphtheria colonies, two days old, on agar. x 8 . . . 401

115. Diphtheria bacilli from a twenty-four hours' culture ou

agar. x 1000 ..... . 402

116. Diphtheria bacilli, from a three days' agar culture, x 1000 . 402

117. Involution forms of the diphtheria bacillus. xlOOO. •••..'•- 403

118. Pseudo-diphtheria bacillus (Hofmann's). x 1000 . , ,, 411

119. Xerosis bacillus from a young agar culture, x 1000 . . 412

120. Film preparation of discharge from wound in a case of

tetanus, showing several tetanus bacilli of "drumstick"

form. xlOOO ...... 417

121. Tetanus bacilli, showing flagella. x 1000 . . . 418

122. Spiral composed of numerous twisted flagella of the tetanus

bacillus. xlOOO ... .419

123. Tetanus bacilli, some of which possess spores, x 1000 . 419

124. Stab culture of the tetanus bacillus in glucose gelatin . 420

125. Colonies of the tetanus bacillus on agar seven days old. x 50 421

126. Film preparation from the affected tissues in a case of

malignant oedema. x 1000 . . • 434

127. Bacillus of malignant cedcma, showing spores, x 1000 . 435

128. Stab cultures in agar — tetanus bacillus, bacillus of malignant

oedema, and bacillus of quarter-evil . . . 436

129. Bacillus of quarter-evil, showing spores. x 1000 . . 442

130. Bacillus aerogenes capsulatus ..... 443

131. Cholera spirilla, from a culture on agar of twenty-four hours'

growth. xlOOO .... .447

132. Cholera spirilla stained to show the terminal flagella.

xlOOO .... .448

133. Cholera spirilla from an old agar culture. x 1000 . . 448

134. Puncture culture of the cholera spirillum . . . 450

LIST OF ILLUSTRATIONS IN TKXT xxiii

Hi;. PACK

135. Colonies of the cholera spirillum on a gelatin plate . . 451

136. MetchnikoflTs spirillum, x 1000 . . .464

137. Puncture cultures in peptone-gelatin .... 465

138. Finkler and Prior's spirillum, x 1000 . . . 466

139. Influenza bacilli from a culture on blood agar. x 1000 . 467

140. Film preparation from a twenty-four hours' culture of the

whooping-cough bacillus. x 1000 .... 473

141. Film preparation from a plague bubo. x 1000 . . 476

142. Bacillus of plague from a young culture on agar. x 1000 . 477

143. Bacillus of plague in chains, x 1000 . . . . 477

144. Culture of the bacillus of plague on 4 per cent, salt agar.

xlOOO ....... 478

145. Section of a human lymphatic gland in plague, x 50 . 480

146. Film preparation of spleen of rat after inoculation with the

bacillus of plague, x 1000 . .... 482

! 17. Mirrococrus melitensis. x 1000 . . . . 490

148. Spirilla of relapsing fever in human blood, x about 1000 . 496

149. Spirillum Obermeieri in blood of infected mouse, x 1000 . 197

150. Film of human blood containing spirillum of tick fever.

x 1000 ....... 500

151. Spirillum of human tick fever (Spirillum Duttoni) in blood

of infected mouse, x 1000 . . . . .501

152 and 153. Film preparation from juice of hard chancre showing

spirochaete pallida. x 1000 .... 504
151. Film preparation from juice of hard chancre showing

spirochaete pallida. x 2000 . . . .505

s.'Ctii.n of spleen from a case of congenital syphilis, showing

^pirochaete pallida. x 1000 .... 506

15*1. Spiro"h;i-r«< refringens. x 1000 .... 507

157-162. Various phases of the benign tertian parasite . . 589

163-168. Exemplifying phases of the malignant parasite . . 590

169. Anueba? .of dysentery ...... 603

170. Section of wall of liver abscess, showing an amoeba of spherical

form with vacuolated protoplasm, x 1000 . . 607

171. Trypanosoma Brucei from blood of infected rat. Note in two

of the organisms commencing division of micronucleus and

undulating membrane, x 1000 .... 620

1 72. Trypanosoma gambiense from blood of guinea-pig, x 1000 . 623

173. Leishmau-Donovan bodies from spleen smear, x 1000 . 632
174 Leishman- Donovan bodies within endothelial cell in spleen.

xlOOO ... 633

• 0 * "' *J

PLATE I.

FIG. 1. Film of pus, containing staphylococci and streptococci. Stained
^ .by Gram's method. x 1000 diameters.

FIG. 2. Fraenkel's pneumococcus in sputum, from a case of acute
pneumonia. Rd. Muir's method of capsule staining.

x 1000 diameters.

Fid. 3. Meningococcus in epidemic cerebro- spinal fever, from lumbar
puncture fluid, showing some involution forms. Leishman's
stain. x 1000 diameters.

FIG. 4. Film from a scraping of throat in Vincent's angina, showing
fusiform bacilli and spirochaetes. x 1000 diameters.

FIG. 5. Gonorrhceal pus, showing gonococci (stained red) and staphylo-
cocci. Gram's method. x 1000 diameters.

.} tfT/.J'I

r, 1o 98JSO & rnoi

oiuaqflo I

•
-uf^ilq.f

PLATE I.

r.,

P ,<

rtsr .£

•• k

A
\

FIG. 1.

FIG. 2.

FIG. 4.

FIG. 5.

PLATE II.

FIG. 6. Spirochaete pallida in section of spleen of child; case of con-
genital syphilis. Levaditi's stain. x 1000 diameters.

FIG. 7. Tubercle bacillus and other bacteria in sputum ; case of chronic
phthisis. Ziehl-Neelsen stain. x 1000 diameters.

FIG. 8. Section of leprous skin, showing numerous clumps of bacilli
(stained red) in the cutis. Carbol-fuchsin and methylene-
blue. x 80 diameters.

FIG. 9. Section of leprous granulation tissue, showing large numbers of
bacilli, chiefly contained within cells. Carbol-fuchsin and
methylene-blue. x 1000 diameters.

FIG. 8. Fio. 9.

.II HTAJ'I

-noo lo saeo ; blrrlo lo aaalfp. ^<> n«\: ''ii

0001 x .m-fite a'idib^vaJ . 'i ^nag

oinoirfo lo ea^o ; mw^uqa nr fiheto&if isriJo bnc «rj[fioed oIoioduT .7 .
,r;ib 0001 x .nw:'

lo «qnrufo guoiomnn gniv .8 .oil

ij^-lodieO .aijuu edi at

('i3 x

.^j srroiqal .8 .aiU

;ri-l<jdijs0 .p.IIao nirii/// b

.ankf

PLATE II.

FIG. 6

FIG. 7.

FIG. 8.

FIG. 9.

••

• 1

PLATE III.

FIG. 10. Streptothrix actinomyces, from agar culture. Gram's method.

x 1000 diameters.

FIG. 11. Anthrax bacilli, from 4-days' agar culture, showing spores.
Carbol-fuchsin and methylene-blue. x 1000 diameters.

FIG. 12. Bacillus diphtherias, from a 12-houre' blood serum culture.
Neisser's stain modified. x 1000 diameters.

N\V"

FIG. 13. Bacillus diphtheria;, from a 5-days' blood serum culture, show-
ing involution forms. Neisser's stain modified.

x 1000 diameters.

FIG. 14. Pseudo-diphtheria bacillus (Hofmann's), from young agar
culture. Neisser's stain modified. x 1000 diameters.

FIG. 15. Typhoid bacillus, from a 24-hours' agar culture, showing
flagella. Rd. Muir's method. x 1000 diameters.

!»

+' \

Fir,. 15.

.Ill 3TAJ1

.bodtem g'mjsir) .aiuiltio ifigB mail ,890^moniio« zhtfoodqeiitB .01 .oil

.e.8-£oqa §niwods piuiiuo ia%s (ey&b-l moil ,'rflioad xeidinA .ft .oil
fb 0001 x .auld-anafyfoem baa fli

.9«rtluo inow?. boold 'eiwod-SI * moil .aBnadidqil) 80IUo«S .SI .
sib 0001 x .bsftibooi areda

-woria ,9iuiluo minsa faoofcf 'e^fi{)-s

.baftibom nr«i« s'ldaaisll .amiol noidulovni gni
ooiBib 0001 x

moil ,(8'nn«raloH) aaHioad B
x .baftibora nifiie • a^aaeisVI ,

jiniwoda piuiluo IB§* ^uod-^S fi raoil ,ei/IIioud biodcftT .d[ .
tjiib 0001 x .boriiara a'liwM .MI

PLATP] III.

- •

"•

• • % • f v :

• » • • • 9 1

FIG. 10.

FIG. 11.

. /.

• . >•'

»..*.» *
• *» *

* r • *

!• * •

FIG. 12.

Fir;. 13.

:-4);^

' V '•'-•%?>-'
',; N-. - 1 '(31/ ,

' '..'v .0— T

FIG. 14.

FIG. 15.

PLATE IV.

FIG. 16. Negri bodies in nerve cells in rabies (hippocampus of dog).
Alcoholic eoein and methylene-blue. x 1000 diameters.

FIG. 17. Bacillus pestis, showing involution forms, from a salt-agar
culture. x 1000 diameters.

FIG. 18. Blood film, showing the spirochaete of relapsing fever.
Irishman's stain. x 1000 diameters.

FIG. 19. The cholera spirillum, from a 12-hours' agar culture, showing
flagella. x 1000 diameters.

FIG. 20. Bacillus tetani, showing spores. x 1000 diameters.

Km. 19.

.VI ttT/.I'i

.(gob to auqraaooqqrd) aaidfii ni allao svtsa ni aeibod iiga^l .81 .oil
ib 0001 x .sufd-anatyrftem bna niaoa oiforiooIA

moil ,8miol noidulovai §aiworfp. tai

.19Y91 ^nieqfilai to sdasffooiiqa 9fld gniworfa (inlft b«x)fS[ .81 .oi'
.ai*tem«ib 0001 x .nicia 8'nBcari«i9J

,9iuiloo i*^« '«iuorf-^I « moil ,ranllhiq8 fiialorfo odT ,61 .01*?
ib 0001 x

.aieianwib 0001 x .gb'ioqa ^niworie ,in«l9i aullio«a .OS .01

PLATE IV.

Fio. 16.

<J^

FIG. -18.

< '
f

J_
FIG. 19.

FIG. 17.

_--y

\ OX \

JO 1

- ^ S

N /

FIG. 20.

ATE V.

PLATE V.

FIG. 21. THE PARASITE OP MILD TERTIAN MALARIA.
Cycle I. (Schizogony). Asexual cycle in the human blood.

a. Sporozoite entering red blood corpuscle and forming

young trophozoite.
6. Young trophozoite in red blood corpuscle.

c. Young trophozoite in red blood corpuscle, with accumulation

of pigment.

d. Large pigmented trophozoite.
' . Mature schizont.

/. Commencing segmentation of schizont.

g. Further stage of segmentation.

h. Segmented schizont ; formation of merozoites.

i. Disintegration of red blood corpuscle, setting free the

merozoites.

j. Young merozoite entering red blood corpuscle.

k. Macrogametocyte, or female sporont.

I. Microgametocyte, or male sporont.

Cycle II. (Sporogony). Sexual cycle in the mosquito,

'ra. Microgametocyte.

n. Macrogametocyte.

Formation of microgametes from the microgametocyte.

p. Free microgamete.

q. Microgamete entering the macrogametocyte.

jr. Zygote or ookinete.

s. Sporocyst.

t. Formation of sporoblasts in the sporocyst.

u. Formation of sporozoites from sporoblasts.

v. Rupture of sporocyst, setting free the sporozoites.

to. Free sporozoites in the body fluid.

i. Accumulation of sporozoites in the salivary gland.

y. Sporozoites passing from gland duct into the blood of man.

FIG. 22. THE PARASITE OF MALIGNANT MALARIA.

a. Young trophozoite entering red blood corpuscle.
6. Do. in red corpuscle.

c. Multiple infection of red corpuscle.

f/. Multiple infection with chromatic stippling in cellular proto-
plasm ; a similar cell is seen lying beneath a, — it contains
a pigmented trophozoite.

d. Pigmented trophozoite.

e. Segmented schizont, cluster of merozoites.
/. Macrogametocyte, " female crescent."

g. Microgametocyte, " male crescent."

A. Red blood corpuscle with chromatic stippling.

i. Large mono-nucleated phagocyte containing malarial pigment

.AI&AJ&M viAixaaT aaiM *o HTIHAHAI auT .IS .
.boold a&mud ad* m alo^o Jtor/xseA
ytinnol bn* alosi/q-ioo boold bai gnhsJna

.eiioxodqoiit

.sloeuqioo boold bsi ni aJioxoriqoi;*
,9loeuqTOo boold bei ni atiosodqoTi

.inosidoa to noritfiJnam^aB gnionaraaioO .\

to noiteanoi ; inoxidoe boin^jiryy^ .A
91 io n( 'L '*

sdi gahsiaa 9^ara«yrwiM .?
10 aio^iS .*»

ni o»a«ldoioqa to noiteonol .S
moil gsiiosoioqa to noii«flno^ .u
edi aai^ ^niiiaa ^a^ooioqe to artrtqw.fi .«

bai §flho*£ia aiiosoiaoi

10 ,

.*flOioqa alam 10
.oJiupeoia ad* ni alo^0 I*ff»8 .(^nogoioqfc:

" '£
a*'£>o*i
moil •ttMMBMobi to noiiaanol .o

£• !• ^

s*6" ^

.bflfils ^wvilea ad* ni aa*ioso'ioqa to nohtalimiuosA .x

.item to boold ad* o*ar *oub bnalg moi! ^aiaaaq aa*iosoioq8 .yj

.AIHAJAM TWAMOIJAM TO STiaAflAl SHT .SS .Ol'

.afoaifqioo boold bai gniiaixia e^iosodqoi* gnnox .»

.aloai/qioo bai ai -^.

.aloe/jqioo bai to n<' -Ii/M .a

-oioiq lalullao ai gnilqqiia oiJamoirio ri*iw a- r~ • 'J

ii (0 d*aanad ^ni^I naaa ei flao lalioiie a ; aiafiiq

.ajiosodqoi* ba*aaui^iq a

,ajios(;dqoi* baiaaoi^i*! .b
tinosidoa h9*aama98 .^>
J' ,a*^oo*aina^oio«M .\

ta*^U' « ^1

Iqqi*8 ocJamoirio d*tw afoeuqioo «A
sfli^iq Iciialam §nini**noo a^xoogailq ba*aaio0fl-onom agiaJ .1

PLATE V.
FIG. 21.

-fc & ^

FIG. 22.

PLATE VI.

FIG. 23. Entamceba histolytica in pus, from tropical abscess of liver.

»Wet fixed film. Stained by Benda's method,
x 1000 diameters.

Fio. 24. Leishman-Donovan bodies, from the spleen of a case of
ir. x 1000 diameters.

FIG. 25. Blood film, showing Trypanosoma Gambiense. Irishman's stain.

x 1000 diameters.

.IV

.lav il to 839.owte koiqoiJ 11101} ,*rrq ni aoiityloieuf «da>raj8^na .8S .oil

.b
rh 000 1 x

lo 9fMjo « to n
iib 0001 x

.ni*ig g'luoiffaisJ .saasidmAf) AIHOK- • woria .ralS b<x)ia .SS .oil

OOOf x

PLATE ;yi

Fia. 23.

FIG. 24.

FIG. 25.

MANUAL OF BACTERIOLOGY

CHAPTER I.

GENERAL MORPHOLOGY AND BIOLOGY.

Introductory. — At the bottom of the scale of living things there
exists a group of organisms to which the name of bacteria is
usually applied. These are apparently of very simple structure,
and may l>e subdivided into two sub-groups, a lower and'simpler
and a higher and better-developed.

The lower forms are the more numerous, and consist of
minute unicellular masses of protoplasm devoid of chlorophyll,
which multiply by simple fission. Some are motile, others non-
motile. Their minuteness may be judged of by the fact that in
one direction at least they usually do not measure more than
1 n, (^-Tfjffir inch). These forms can be classified according to
their shapes into three main groups — (1) A group in which the
shape is globular. The members of this are called cocci. (2) A
group in which the shape is that of a straight rod — the pro-
portion of the length to the breadth of the rod varying greatly
among the different members. These are called bacilli. (3) A
group in which the shape is that of a curved or spiral rod.
These are called spirilla. The full description of the characters
of these groups will be more conveniently taken later (p. 12).
In some cases, especially among the bacilli, there may occur
under certain circumstances changes in the protoplasm whereby
a resting stage or spore is formed.

The higher forms show advance on the lower along two lines.
(1) On the one hand, they consist of filaments made up of
simple elements such as occur in the lower forms. These
filaments may be more or less septate, may be provided with a

2- ^: GENERA*. ^MORPHOLOGY AND BIOLOGY

>aili,. and ^may.-sjiojv^ branching either true or false. The
minute structure of the elements comprising these filaments is
analogous to that of the lower forms. Their size, however, is
often somewhat greater. The lower forms sometimes occur in
filaments, but here every member of the filament is independent,
while in the higher forms there seems to be a certain inter-
dependence among the individual elements. For instance,
growth may occur only at one end of a filament, the other
forming an attachment to some fixed object. (2) The higher
forms, moreover, present this further development, that in certain
cases some of the elements may be set apart for the reproduction
of new individuals.

Terminology. — The term bacterium of course in strictness
only refers to the rod-shaped varieties of the group, but as it
has given the name bacteriology to the science which deals with
the whole group, it is convenient to apply it to all the members
of the latter, and to reserve the term bacillus for the rod-shaped
varieties. Other general words, such as germ, microbe, micro-
organism, are used as synonymous with bacterium, though these
are often made to include the smallest organisms of the animal
kingdom.

While no living organisms lower than the bacteria are known
(though certain facts regarding ultra-microscopic forms of life
make the occurrence of such possible), the upper limits of the
group are difficult to define, and it is further impossible in the
present state of our knowledge to give other than a provisional
classification of the forms which all recognise to be bacteria.
The division into lower and higher forms, however, is fairly well
marked, and we shall therefore refer to the former as the lower
bacteria, and to the latter as the higher bacteria.

Morphological Relations. —The relations of the bacteria to the animal
kingdom on the one hand and to the vegetable on the other constitute a
somewhat difficult question. It is best to think of there being a group
of small, unicellular organisms, which may represent the most primitive
forms of life before differentiation into animal and vegetable types had
occurred. This would include the flagellata and infusoria, the myxo-
mycetes, the lower algre. and the bacteria. To the lower algae the bacteria
show many similarities. These algae are unicellular masses of protoplasm,
having generally the same shapes as the bacteria, and largely multiply by
fission. Endogenous sporulation, however, does not occur, nor is motility
necessarily associated with the possession of flagella. Also their proto-
plasm differs from that of the bacteria in containing chlorophyll and
another blue-green pigment called phycocyan. From the morphological
resemblances, however, between these algse and the bacteria, and from
the fact that fission plays a predominant part in the multiplication of
both, they have been grouped together in one class as the Schizophyta

THE STRUCTURE OF THE BACTERIAL CELL 3

or splitting plants (German, Spaltpflanzen). And of the two divisions
forming these Schizophyta the splitting algae are denominated the
sdiizophycoe (German, Spaltalgen), while the bacteria or splitting fungi
are called the schizomycetes (German, Spaltpilzen). The bacteria are,
thrivf'oi-e, oftm spoken of as the schizomycetes. Certain bacteria which
have been described as containing chlorophyll ought probably to be
grouped among the schizophycese.

GENERAL MORPHOLOGY OF THE BACTERIA.

The Structure of the Bacterial Cell. — On account of the
minuteness of bacteria the investigation of their structure is
attended with great difficulty. When examined under the
microscope, in their natural condition, e.g. in water, they appear
merely as colourless refractile bodies of the different shapes
named. Spore formation and motility, when these exist, can
also be observed, but little else can be made out. For their
propei- investigation advantage is always taken of the fact of
their attinities for various dyes, especially those which are usually
chosen as good stains for the nuclei of animal cells. Certain
points have thus been determined. The bacterial cell consists
of a sharply contoured mass of protoplasm which reacts to,
especially basic, aniline dyes like the nucleus of an animal cell
— though from this fact we cannot deduce that the two are
identical in composition. A healthy bacterium when thus
stained presents the appearance of a finely granular or almost
homogeneous structure. The protoplasm is surrounded by an
envelope which can in some cases be demonstrated by over-
staining a specimen with a strong aniline dye, when it will appear
s halo round the bacterium. This envelope may sometimes
oe seen to be of considerable thickness. Its innermost layer is
probably of a denser consistence, and sharply contours the
contained protoplasm, giving the latter the appearance of being
surrounded by a membrane. It is only, however, in some of
the higher forms that a definite membrane occurs. Sometimes
the outer margin of the envelope is sharply defined, in which case
the l.acterinm appears to have a distinct capsule, and is known
as a capsulated bacterium (vide Fig. 1, No. 4; and Fig. 62).
The cohesion of bacteria into masses depends largely on the
character of the envelope. If the latter is glutinous, then a
l:ir^e mass of the same species may occur, formed of individual
Bacteria embedded in what appears to be a mass of jelly. When
this occurs, it is known as a zoogloea mass. On the other hand,
it the envelope has not this cohesive property the separation of
individuals may easily take place, especially in a fluid medium

4 GENERAL MORPHOLOGY AND BIOLOGY

in which they may float entirely free from one another. Many
of the higher bacteria possess a sheath which has a much more
definite structure than is found among the lower forms. It
resists external influences, possesses elasticity, and serves to bind
the elements of the organism together.

Reproduction among the Lower Bacteria. — When a bacterial
cell is placed in favourable surroundings, it multiplies ; as has
been said, this, in the great majority of cases, takes place by
simple fission. In the process a constriction appears in the
middle and a transverse unstained line develops across the
protoplasm at that point. The process goes on till two
individuals can be recognised, which may remain for a time
attached to one another, or become separate, according to the
character of the envelope, as already explained. In most
bacteria growth and multiplication go on with great rapidity.
A bacterium may reach maturity and divide in from twenty
minutes to half an hour. If division takes places only every
hour, from one individual after twenty -four hours 17,000,000
similar individuals will be produced. As shown by the results
of artificial cultivation, others, such as the tubercle bacillus,
multiply much more slowly. Sometimes division proceeds so
rapidly that the young individuals do not reach the adult size
before multiplication again occurs. This may give rise to
anomalous appearances.

From investigations by Graham- Smith and others, it appears that
the consistence of the envelope may have an importance in modifying
the naked-eye and low-power appearances presented by bacterial colonies
which constitute a feature in the identification of species (see p. 137).
Graham-Smith, working with bacilli, differentiates four groups — a "loop-
forming," in which the envelope is so tough that, after division, rupture
but rarely occurs (b. anthracis) ; a " folding " group, in which the envelope
is so flexible and extensile that the members of a chain can be folded on
one another as successive divisions take place (b. pestis) ; a "snapping"
group, in which partial rupture of the envelope occurs on division
(b. diphtherias); and a "slipping" group, where the envelope readily
breaks, and successively developed bacilli slip past each other (v. cholerse).

When bacteria are placed in unfavourable conditions as
regards food, etc., growth and multiplication take place with
difficulty. In the great majority of cases this is evidenced by
changes in the appearance of the protoplasm. Instead of its
maintaining the regularity of shape seen in healthy bacteria,
various aberrant appearances are presented. This occurs especially
in the rod-shaped varieties, where flask-shaped or dumb-bell -
shaped individuals may be seen. The regularity in structure

SPORE FORMATION 5

ami size is quite lost. The appearance of the protoplasm
also is often altered. Instead of, as formerly, staining well, it
does not stain readily, and may have a uniformly pale homo-
geneous appearance, while in an old culture only a small
proportion of the bacteria may stain at all. Sometimes, on the
other hand, a degenerated bacterium contains intensely stained
irrannlt's or globules which may be of large size. Such aberrant
;ni«l degenerate appearances are referred to as involution forms.
That these forms really betoken degenerative changes is shown
by the fact that, on their being again transferred to favourable
conditions, only slight growth at first takes place. Many
individuals have undoubtedly died, and the remainder which
live and develop into typical forms may sometimes have lost
some of their properties.

Reproduction among the Higher Bacteria. —Most of the higher bacteria
consist of thread-like structures more or less septate and often surrounded
by a sheath. The organism is frequently attached at one end to some
object or to another individual. It grows to a certain length and then
at tin' lice end certain cells, called gonidia, are cast oil' from which new
individuals are formed. These gonidia may be formed by a division
taking place in the terminal element of the filament such as has occurred
in the growth of the latter. In some cases, however, division takes
j'lai •(• in three dimensions of space. The gonidia have a free existence
fur a certain time before becoming attached, and in this stage are
sometimes motile. They are usually rod-like in shape, sometimes
pyriform. They do not possess any special powers of resistance.

Spore Forma,tion. — In certain species of the lower bacteria,
under certain circumstances, changes take place in the protoplasm
which result in the formation of bodies called spores, to which
the vital activities of the original bacteria are transferred.
Spore formation occurs chiefly among the bacilli and in some
spirilla. Its commencement in a bacterium is indicated by the
appearance in the protoplasm of a minute highly refractile
granule unstained by the ordinary methods. This increases in
-i/i-, and assumes a round, oval, or short rod-shaped form, always
shorter but often broader than the original bacterium. In the
process of spore formation the rest of the bacterial protoplasm
may remain unchanged in appearance and staining power for a
considerable time (e.g. b. tetani), or, on the other hand, it may
soon lose its power of staining and ultimately disappear, leaving
the spoiv in the remains of the envelope (?.</. b. anthracis).
This method of spore formation is called cn<l<><i< nuns. Bacterial
spores are always non-motile. The spore may appear in the
centre of the bacterium, or it may be at one extremity, or a
short distance from one extremity (Fig. 1, No. 11). In structure

6 GENERAL MORPHOLOGY AND BIOLOGY

the spore consists of a mass of protoplasm surrounded by a dense
membrane. This can be demonstrated by methods which will
be described, the underlying principle of which is the prolonged
application of a powerful stain. The membrane is supposed to
confer on the spore its characteristic feature, namely, great
capacity of resistance to external influences such as heat or
noxious chemicals. Koch, for instance, in one series of experi-
ments, found that while the bacillus anthracis in the unspored
form was killed by a two minutes' exposure to 1 per cent carbolic
acid, spores of the same organism resisted an exposure of from
one to fifteen days.

When a spore is placed in suitable surroundings for growth,
it again assumes the original bacillary or spiral form. The
capsule dehisces either longitudinally, or terminally, or trans-
versely. In the last case the dehiscence may be partial, and the
new individual may remain for a time attached by its ends to
the hinged spore-case, or the dehiscence may be complete and
the bacillus grow with a cap at each end consisting of half the
spore-case. Sometimes the spore-case does not dehisce, but is
simply absorbed by the developing bacterium.

It is important to note that in the bacteria spore formation
is rarely, if ever, to be considered as a method of multiplication.
In at least the great majority of cases only one spore is formed
from one bacterium, and only one bacterium in the first instance
from one spore. Sporulation is to be looked upon as ^jresting
sJ«|M_jQf--^Jba£leriuin, and is to be contrasted with the stage
when active multiplication takes place. The latter is usually
referred to as the vegetative . stage^ of_lh&_bacterium. Regarding
the signification of spore formation in bacteria, there has been
some difference of opinion. According to one view, it may be
regarded as representing the highest stage in the vital activity
of a bacterium. There is thus an alternation between the
vegetative and spore stage, the occurrence of the latter being
necessary to the maintenance of the species in its greatest
vitality. Such a rejuvenescence, as it were, through sporulation,
is known in many algae. In support of this view there are
certain facts. In many cases, for instance, spore formation only
occurs at temperatures specially favourable for growth and
multiplication. There is often a temperature below which,
while vegetative growth still takes place, sporulation will not
occur ; and in the case of b. anthracis, if the organism be kept
at a temperature above the limit at which it grows best, not
only are no spores formed, but the strain may lose the power
of sporulation. Furthermore, in the case of bacteria preferring

SPORE FORMATION 7

the presence of oxygen for their growth, an abundant supply
of this gas ma^r favour sporulation. It is probable that even
among bacteria preferring the absence of oxygen for vegetative
uiowth, the presence of this gas favours sporulation. Some
facts relating to tlwi.1. cases in which two spores are formed in
one bacterium have been adduced to support the view that
sporulation may represent a degenerate sexual process. Here a
partial fission of a cell has been observed followed by a re-
fusion of the protoplasmic moieties and the formation of a spore
at each end of the rod. The second view with regard to
sporulation is that a bacterium only forms a spore when its
surroundings, especially its food supply, become unfavourable
for vegetative growth ; it then remains in this condition until it
is placed in more suitable surroundings. Such an occurrence
would be analogous to what takes place under similar conditions
in many of the protozoa. Often sporulation can be prevented
from taking place for an indefinite time if a bacterium is
constantly supplied with fresh food (the other conditions of life
being equal). The presence of substances excreted by the
bacteria themselves plays, however, a more important part in
making the surroundings unfavourable than the mere exhaustion
of the food supply. A living spore will always develop into a
vegetative form if placed in a fresh food supply. With regard to
the rapid formation of spores when the conditions are favourable
for vegetative growth, it must be borne in mind that in such
circumstances the conditions may really very quickly become
unfavourable for a continuance of growth, since not only will the
food supply around the growing bacteria be rapidly exhausted,
but the excretion of effete and inimical matters will be all the
more rapid.

\\V must note that the usually applied tests of a body
developed within a bacterium being a spore are (1) its staining
reaction, namely, resistance to ordinary staining fluids, but
capacity of being stained by the special methods devised for
the purpose (vide p. 109) ; (2) the fact that the bacterium
containing the spore has higher .powers of resistance against
inimical conditions than a vegetative form. It is important to
bear these tests in mind, as, in some of the smaller bacteria
especially, it is very difficult to say whether they spore or not.
There may appear in such organisms small unstained spots, the
significance of which it is very difficult to determine.

The Question of Arthrosporous Bacteria.— It is stated by Hueppe that
aiming certain organi.Nins, <\<j. some streptococci, certain individuals may,
without endogenous sporulation, take on a resting stage. These become

8 GENERAL MORPHOLOGY AND BIOLOGY

swollen, stain well with ordinary stains, and they are stated to have
higher power of resistance than the other forms ; further, when vegetative
life again occurs, it is from them that multiplication is said to take place.
From the fact that there is no new formation within the protoplasm,
but that it is the whole of the latter which participates in the change,
these individuals have been called arthrospores. The existence of such
special individuals amongst the lower bacteria is extremely problematical.
They have no distinct capsule, and they present no special staining
reactions, nor any microscopic features by which they can be certainly
recognised, while their alleged increased powers of resistance are very
doubtful. All the phenomena noted can be explained by the undoubted
fact that in an ordinary growth there is very great variation among
the individual organisms in their powers of resistance to external
conditions.

Motility. — As has been stated, many bacteria are motile.
Motility can be studied by means of hanging-drop preparations
(vide p. 69). The movements are of a darting, rolling, or
vibratile character. The degree of motility depends on the
species, the temperature, the age of the growth, and on the
medium in which the bacteria are growing. Sometimes the
movements are most active just after the cell has multiplied,
sometimes it goes on all through the life of the bacterium,
sometimes it ceases when sporulation is about to occur. Motility
is associated with the possession of fine wavy thread-like
appendages called flagella, which for their demonstration require
the application of special staining methods (vide Fig. 1, No. 12-;
and Fig. 108). They have been shown to occur in many bacilli
and spirilla, but only in a few species of cocci. They vary in
length, but may be several times the length of the bacterium,
and may be at one or both extremities or all round. When
terminal they may occur singly or there may be several. The
nature of these fiageHa has been much disputed. Some have
held that, unlike what occurs in many algae, they are not actual
prolongations of the bacterial protoplasm, but merely appendages
of the envelope, and have doubted whether they are really organs
of locomotion. There is now, however, little doubt that they
belong to the protoplasm. By appropriate means the central
parts of the latter can be made to shrink away from the peripheral
(vide infra, " plasmolysis "). In such a case movement goes on
as before, and in stained' preparations the flagella can be seen
to be attached to the peripheral zone. It is to be noted that
flagella have never been demonstrated in non-motile bacteria,
while, on the other hand, they have been observed in nearly all
motile forms. There is little doubt, however, that all cases of
motility among the bacteria are not dependent on the possession
of flagella, for in some of the special spiral forms, and in most

STRUCTURE OF BACTERIAL PROTOPLASM 9

of the higher bacteria, motility is probably due to contractility
of the protoplasm itself.

The Minuter Structure of the Bacterial Protoplasm. — Many attempts
have been made to obtain deeper information as to the structure of the
bacterial cell, and especially as to its behaviour in division. These
ha vi- largely turned on the interpretation to 1m put on certain appear-
ances which have been observed. These appearances are of two kinds.
First, under certain circumstances irregular deeply-stained grannies are
• •liM-m-d in the protoplasm, often, when they occur in a bacillus, giving
the latter the appearance of a short chain of cocci. They are often
called metachromatic granules (vide Fig. 1, No. 16) from the fact that
by appropriate procedure they can be stained with one dye, and the
protoplasm in which they lie with another ; sometimes, when a single
stain is used, such as methylene blue, they assume a slightly different
tint from the protoplasm.

For the demonstration of the metachromatic granules two methods
have been advanced. Ernst recommends that a few drops of Lofflers
methylene blue (vide p. 104) be placed on a cover-glass preparation and
the latter passed backwards and forwards over a Bunsen flame for half
a minute after steam begins to rise. The preparation is then washed
in water and counter-stained for one to two minutes in watery Bismarck-
brown. The granules are here stained blue, the protoplasm brown.
Neisser stains a similar preparation in warm carbol-fuchsin, washes
with 1 per cent, sulphuric acid, and counter-stains with Loffler's blue.
Here the granules are magenta, the protoplasm blue. The general
character of the granules thus is that they retain the first stain more
intensely than the rest of the protoplasm does.

A second appearance which can sometimes be seen in specimens
stained in ordinary ways is the occurrence of a concentration of the

Protoplasm at each end of a bacterium, indicated by these parts being
eeply stained. These deeply stained parts are sometimes called polar
granules (vide Fig. 1, No. 16, the bacillus most to the right) (German,
Polkiirnchen or Polkorner).

With regard to the significance that is to be attached to such
appearances, much depends on whether they are constantly present
under all circumstances, or only occasionally, when the organism is
grown in special media or under special growth conditions. Some
bacteria, however stained, show evidence of having the protoplasm
somewhat granular, e.g. the diphtheria bacillus. In other cases this
granular condition is only seen when the organism has been grown under
bad conditions, or where the food supply is becoming exhausted. Some
have thought that the appearances might be due to a process allied to
mitosis and might signify approaching division, but of this there is no
evidence.

In perfect healthy and young bacteria, appearances of granule
formation and of vacuolation may be accidentally produced by physical
means in the occurrence of what is known as pltumotyffa To speak
generally, when a mass of protoplasm surrounded by a fairly linn
envelope of a colloidal nature is placed in a solution containing salts in
greater concentration than that in which it has previously been living,
then by a process of osmosis the water held in the protoplasm passes
out through the membrane, and, the protoplasm retracting from the
latt- T, the appearance of vacuolation is presented. Now, in making a

10 GENERAL MORPHOLOGY AND BIOLOGY

dried film for the microscopic examination of bacteria, the conditions
necessary for the occurrence of this process may be produced, and the
appearances of vacuolation and, in certain cases, of Polkorner may thus
be brought about. Plasmolysis in bacteria has been extensively
investigated,1 and has been found to occur in some species more readily
than in others. Furthermore, it is often most readily observed in old or
otherwise enfeebled cultures.

Biitschli, from a study of some large sulphur-containing forms, con-
cludes that the greater part of the bacterial cell may correspond to a
nucleus, and that this is surrounded by a thin layer of protoplasm which
in the smaller bacteria escapes notice, unless when, as in the bacilli, it
can be made out at the ends of the cells. Fischer, it may be said, looks
on the appearances seen in Biitschli's preparations as due to plasmolysis.

The Chemical Composition of Bacteria. — In the bodies of
bacteria many definite substances occur. Some bacteria have
been described as containing chlorophyll, but these are properly
to be classed with the schizophyceae. Sulphur is found in some
of the higher forms, and starch_gramiles areTalso described as
occurring. Many species of bacteria, when growing in masses,
are brilliantly coloured, though few bacteria associated with the
production of disease give rise to pigments. In some of the
organisms classed as bacteria a pigment named bacterio-purpurin
has been observed in the protoplasm, and similar intracellular
pigments probably occur in some of the larger forms of the
lower bacteria and may occur in the smaller ;(but it is usually
impossible to determine whether the pigment occurs inside or
outside the protoplasm/) In many cases, for the free production
of pigment abundant oxygen supply is necessary ; but sometimes,
as in the case of spirillum rubrum, the pigment is best formed
in the absence of oxygen. Sometimes the faculty of forming it
may be lost by an organism for a time, if not permanently, by
the conditions of its growth being altered. Thus, for example,
if the b. pyocyaneus be exposed to the temperature of 42° C.
for a certain time, it loses its power of producing its bluish
pigment. Pigments formed by bacteria often diffuse out into,
and colour, the medium for a considerable distance around.

Comparatively little is known of the nature of bacterial pigments.
Zopf, however, has found that many of them belong to a group of
colouring matters which occur widely in the vegetable and animal
kingdoms, namely, the lipochromes. These lipochromes, which get their
name from the colouring matter of animal fat, include the colouring
matter in the petals of Ranunculacese, the yellow pigments of serum and
of the yolks of eggs, and many bacterial pigments. The lipochromes are
characterised by their solubility in chlorolorm, alcohol, ether, and

1 Consult Fischer, " Untersuchungen iiber Bakterien," Berlin, 1894;
" Ueber den Bau der Cyauophyceen mid Bakterieu," Jena, 1897.

THE CLASSIFICATION OF BACTERIA 11

petroleum, and by their giving indigo-blue crystals with strong sulphuric
arid, and a green colour with iodine dissolved in potassium iodide.
Though cry stal line compounds of these have been obtained, their
cht inical constitution is entirely unknown, and even their percentage
i (imposition is disputed.

Some observations have been made on the chemical structure
of bacterial protoplasm. Nencki isolated from the bodies of
certain putrefactive bacteria proteid bodies which, according to
Kuppel, appear to have been allied to peptone, and which
differed from nucleo-proteids in not containing phosphorus,
but many of the proteids isolated by other chemists have
been allied in their nature to the protoplasm of the nuclei
of cells. Buchner in certain researches obtained bodies of this
nature allied to the vegetable caseins, and he adduces evidence
to show that it is to these that the characteristic staining
properties are due. Various observers have isolated similar
phosphorus-containing proteids from different bacteria. Besides
proteids, however, substances of a different nature have been
isolated. Thus cellulose, fatty material, chitin, wax-like bodies,
and other substances have been observed. There are also found
various mineral salts, especially those of sodium, potassium, and
magnesium. The amount of different constituents varies ac-
cording to the age of the culture and the medium used for
m<>\\tli, and certainly great variation takes place in the com-
position of different species.

The Classification of Bacteria. — There have been numerous
schemes set forth for the classification of bacteria, the fuixla-
nii'iital principle running through all of which has been the
recognition of the two sub-groups and the type forms mentioned
in the opening paragraph above. In the attempts to still
furl her subdivide the group, scarcely two systematists are agreed
a- to the characters on which sub-classes are to be based. Our
(•resent knowledge of the essential morphology and relations of
bacteria is as yet too limited for a really natural classification
to be attempted. To prepare for the elaboration of the latter,
Marshall Ward suggested that in every species there should be
studied the habitat, best food supply, condition as to gaseous
environment, range of growth temperature, morphology, life
hUtoiy. >pecial properties, and pathogen icity. Some recent
at! i 'in j.ts to carry out such a plan will be referred to in
• -onnectioii with the principles of general bacteriological
diagnosis (p. 135).

\\V must thus be, content with a provisional and incomplete
We have said that the division into lower and

12 GENERAL MORPHOLOGY AND BIOLOGY

higher bacteria is recognised by all, though, as in every other
classification, transitional forms have to be accounted for. In
subdividing the bacteria further, the forms they assume con-
stitute at present the only practicable basis of classification.
The lower bacteria thus naturally fall into the three groups
mentioned, the cocci, bacilli, and spirilla, though the higher
are more difficult to deal with. Subsidiary, though important,
points in still further subdivision are the planes in which fission
takes place and the presence or absence of spores. The recogni-
tion of actual species is often a matter of great difficulty. The
points to be observed in this will be discussed later (p. 137).

I. The Lower Bacteria.1 — These, as we have seen, are
minute unicellular masses of protoplasm surrounded by an
envelope, the total vital capacities of a species being represented
in every cell. They present three distinct type forms, the
coccus, the bacillus, and the spirillum ; endogenous sporulation
may occur. They may also be motile.

1. The Cocci. — In this group the cells range in different
species from '5 /u, to 2 /x, in diameter, but most measure about 1 //.
Before division they may increase in size in all directions. The
species are usually classified according to the method of division.
If the cells divide only in one axis, and through the consistency
of their envelopes remain attached, then a chain of cocci will be
formed. A species in which this occurs is known as a strepto-
coccus. If division takes place irregularly, the resultant mass may
be compared to a bunch of grapes, and the species is often called
a staphylococcus. Division may take place in two axes at right
angles to one another, in which case cocci adherent to each other
in packets of four (called tetrads) or sixteen may be found,
the former number being the more frequent. To all these forms
the word micrococcus is often generally applied. The individuals
in a growth of micrococci often show a tendency to remain
united in twos. These are spoken of as diplococci, but this is
not a distinctive character, since every coccus as a result of
division becomes a diplococcus, though in some species the
tendency to remain in pairs is well marked. The adhesion of
cocci to one another depends on the character of the capsule.
Often this has a well-marked outer limit (micrococcus tetrayenus),
sometimes it is of great extent, its diameter being many times
that of the coccus (streptococcus mcsenteriodes). It is especially
among the streptococci and staphylococci that the phenomenon
of the formation of arthrospores is said to occur. In none of

1 For the illustration of this and the succeeding systematic paragraphs,
vide Fig. 1.

THE LOWER BACTERIA

156

Fio. 1.— 1. Coccus. 2. Streptococcus. 3. Staphylococcus. 4. Capsulated diplococcus.
5. " Biscuit "-shaped coccua. 6. Tetrads. 7. Sarcina fonn. 8. Types of bacilli
(1-8 are diagrammatic). 9. Non-septate spirillum x 1000. 10. Ordinary spirillum —
(a) comma-shaped element ; (t>) formation of spiral by comma-shaped elements
x 1000. 11. Types of spore formation. 12. Flagellated bacteria. 13. Changes in
I.;K trria produced by plasmolysis (after Fischer). 14. Bacilli with terminal proto-
plasm (liutschli). 15. (a) Bacillus composed of five protoplasmic meshes; (6)proto-
plMmio network in micrococcus (Biitschli). 10. Bacteria containing metachromatic
granules (Ernst, Neisser)— some contain polar granules. 17. Beggiatoaalba. Both
filaments contain sulphur granules — one is septate. 18. Thiothrix tennis (\Vino-
ki). 19. Leptothrix innominate (Miller). 20. Cladothrix dichotoma (Zopf).
•.M. Stn-jttuthrix a<-tiiii)mvces(Bostr6m), (a) colony under low power; (b) filament
>li(.\vin«; true branching ;' («) filament containing coccus-like bodies; (d) filament
with chili at end.

14 GENERAL MORPHOLOGY AND BIOLOGY

the cocci have endogenous spores been certainly observed. The
species of the streptococci and staphylococci differentiated
number several hundreds. Usually included in this group are
coccus-like organisms which divide in three axes at right angles
to one another. These are usually referred to as sarcince. If
the cells are lying single they are round, but usually they are
seen in cubes of eight with the sides which are in contact
slightly flattened. Large numbers of such cubes may be lying
together. The sarcinse are, as a rule, rather larger than the
other members of the group. Most of the cocci are non-motile,
but a few motile species possessing flagella have been described.

2. Bacilli. — These consist of long or short cylindrical cells,
with rounded or sharply rectangular ends, usually not more than
1 [A broad, but varying very greatly in length. They may be
motile or non-motile. Where flagella occur, these may be
distributed all round the organism, or only at one or both of
the poles (pseudomonas). Several species are provided with
sharply-marked capsules (b. pneumonias). In many species
endogenous sporulation occurs. The spores may be central or
terminal, round, oval, or spindle-shaped.

Great confusion in nomenclature has arisen in this group in con-
sequence of the different artificial meanings assigned to the essentially
synonymous terms bacterium and bacillus. Migula, for instance, applies
the former term to non-motile species, the latter to the motile. Hueppe,
on the other hand, calls those in which endogenous sporulation does
not occur, bacteria, and those where it does, bacilli. In the ordinary
terminology of systematic bacteriology the word bacterium has been
almost dropped, and is reserved, as we have done, as a general term for
the whole group. It is usual to call all the rod-shaped varieties bacilli.

3. Spirilla. — These consist of cylindrical cells more or less
spiral or wavy. Of such there are two main types. In one there
is a long non-septate, usually slender, wavy or spiral thread
(Fig. 1, No. 9). In the other type the unit is a short curved
rod (often referred to as of a "comma" shape). When two
or more of the latter occur, as they often do, end to end
with their curves alternating, then a wavy or spiral thread
results. An example of this is the cholera microbe (Fig. 1,
No. 10). This latter type is of much more frequent occurrence,
Among the first group motility is often not associated, as far
as is known, with the possession of flagella. The cells here
apparently move by an undulating or screw-like contraction of
the protoplasm. Most of the motile spirilla, however, possess
flagella. Of the latter there may be one or two, or a bunch
containing as many as twenty, at one or both poles. Division

THE HIGHER BACTERIA 15

takes place as among the bacilli, but in some of the non-
-cptate forms a longitudinal fission may occur. In some species
endogenous sporulation has been observed.

Three terras are used in dividing this group, to which different authors
have given different meanings. These terms are spirillum, spirochaete,
vibrio. Migula makes " vibrio" synonymous with " microspira," which
la- applies to members of the group which possess only one or two polar
fla^clla ; "spirillum " he applies to similar species which have bunches
of polar Hagella, while " spirochsete " is reserved for the long unHagellated
spiral cells. Hueppe applies the term " spirochsete " to forms without
endospores, " vibrio " to those with endospores in which during sporula-
tion the organism changes its form, and "spirillum" to the latter
when no change of form takes place in sporulation. Flugge, another
systematist, applies " spirochsete " ancl "spirillum" indiscriminately to
any wavy or corkscrew form, and "vibrio" to forms where the undula-
tions are not so well marked. It is thus necessary, in denominating such
a bacterium by a specific name, to give the authority from whom the

name is taken.

Within recent years great doubt has arisen as to whether many
of the non-septate spirillary forms, e.g. Spirochcete pallida, are
to be looked on as bacteria at all, — the view being taken that
in, it may be, many cases they represent a stage in the life
history of what are really protozoa. The ultimate classification
of this group of bacteria must thus be left an open question,
and at present it is convenient to denominate the non-septate
spiral rods Spirochcete, and those whose vital unit is a single
curved rod Spirilla,

II. The Higher Bacteria. — These show advance on the lower
in consisting of definite filaments branched or unbranched. In
most cases the filaments at more or less regular intervals are
cut by septa into short rod-shaped or curved elements. Such
elements are more or less interdependent on one another, and
special staining methods are often necessary to demonstrate the
septa which demarcate the individuals of a filament. There is
further often a definite membrane or sheath common to all the
elements in a filament. Not only, however, is there this close
organic relationship between the elements of the higher bacteria,
but there is also interdependence of function ; for example, one
end of a filament is frequently concerned merely in attaching
the organism to some other object. The greatest advance, how-
ever, consists in the setting apart among most of the higher
bacteria of the free terminations of the filaments for the produc-
tion of new individuals, as has been described (p. 5). There
are various classes under which the species of the higher bacteria
are grouped; but our knowledge of them is still somewhat

16 GENERAL MORPHOLOGY AND BIOLOGY

limited, as many of the members have not yet been artificially
cultivated. The beggiatoa group consists of free swimming
forms, motile by undulating contractions of their protoplasm.
For the demonstration of the rod-like elements of the filaments
special staining is necessary. The filaments have no special
sheath, and the protoplasm contains sulphur granules. The
method of reproduction is doubtful. The thiothrix group re-
sembles the last in structure, and the protoplasm also contains
sulphur granules; but the filaments are attached at one end,
and at the other form gonidia. A leptothrix group is usually
described which closely resembles the thiothrix group, except that
the protoplasm does not contain sulphur granules. It cannot,
however, be with certainty said whether such organisms can be
sufficiently differentiated from the bacilli to warrant their being
placed among the higher bacteria. In the cladothrix group
there is the appearance of branching, which, however, is of a
false kind. Whaf happens is that a terminal cell divides, and
on dividing again, it pushes the product of its first division to
one side. There are thus two terminal cells lying side by side,
and as each goes on dividing, the appearance of branching is
given. Here, again, there is gonidium formation; and while
the parent organism is in some of its elements motile, the gonidia
move by means of flagella. The highest development is in the
streptothrix group, to which belongs the streptothrix actinomyces,
or the actinomyces bovis, and several other important pathogenic
agents. Here the organism consists of a felted mass of non-
septate filaments, in which true dichotomous branching occurs.
Under certain circumstances threads grow out, and produce
chains of coccus-like bodies from which new individuals can be
reproduced. Such bodies are often referred to as spores, but
they have not the same staining reaction's nor resisting powers
of so high a degree as ordinary bacterial spores. Sometimes, too,
the protoplasm of the filaments breaks up into bacillus-like
elements, which may also have the capacity of originating new
individuals. In the streptothrix actinomyces there may appear
a club-shaped swelling of the membrane at the end of the fila-
ment, which has by some been looked on as an organ of
fructification, but which is most probably a product of a
degenerative change. The streptothrix group, though its
morphology and relationships are much disputed, may be looked
on as a link between the bacteria on the one hand, and the
lower fungi on the other. Like the latter, the streptothrix forms
show the felted mass of non-septate branching filaments, which
is usually called a mycelium. On the other hand, the breaking

FOOD SUPPLY 17

up of the protoplasm of the streptothrix into coccus- and bacillus-
like forms, links it to the other bacteria.

GENERAL BIOLOGY OF THE BACTERIA.

There are five prime factors in the growth of bacteria which
must be considered, namely, food supply, moisture, relation to
gaseous environment, temperature, and light.

Food Supply. — The bacteria are chiefly found living on the
complicated organic substances which form the bodies of dead
plants and animals, or which are excreted by the latter while
i li'-y are yet alive. Seeing that, as a general rule, many bacteria
grow side by side, the food supply of any particular variety is,
relatively to it, altered by the growth of the other varieties
present. It is thus impossible to imitate the complexity of the
natural food environment of any species. The artificial media
used in bacteriological work may therefore be poor substitutes
for the natural supply. In certain cases, however, the conditions
under which wre grow cultures may be better than the natural
conditions. For while one of two species of bacteria growing
si<lu by side may favour the growth of the other, it may also
in certain cases hinder it, and therefore, when the latter is
grown alone it may grow better. Most bacteria seem to
produce exeretions which are unfavourable to their own
vitality, for, when a species is sown on a mass of artificial
food medium, it does not in the great majority of cases go on
growing till the food supply is exhausted, but soon ceases to
grow. Effete products diffuse out into the medium and prevent
growth. Such diffusion may be seen when the organism pro-
duces pigment, e.y. b. pyocyaneus growing on gelatin. In
supplying artificial food for bacterial growth, the general principle
ought to be to imitate as nearly as possible the natural surround-
iuu-s, though it is found that there exists a considerable adapt-
ability among organisms. With the pathogenic varieties it is
usually found expedient to use media derived from the fluids of
tin- animal body, and in cases where bacteria growing on plants
are being studied, infusions of the plants on which they grow
are frequently used. Some bacteria can exist on inorganic food,
but most require organic material to be supplied. Of the latter,
><>me require proteid to be present for their proper nourishment,
while others can derive their nitrogen from a non-proteid such
as asparagin. All bacteria requre nitrogen to be present in
SMiiu- form, and many require to derive their carbon from
carl. •ihydrat»i>. Mineral salts, especially sulphates, chlorides, and

18 GENERAL MORPHOLOGY AND BIOLOGY

phosphates, and also salts of iron are necessary. Occasionally
special substances are needed to support life. Thus some
species, in the protoplasm of which sulphur granules occur,
require sulphuretted hydrogen to be present. In nature the
latter is usually provided by the growth of other bacteria. When
the food supply of a bacterium fails, it degenerates and dies.
The proof of death lies in the fact that when it is transferred
to fresh and good food supply it does not multiply. If the
bacterium forms spores, it may then survive the want of food
for a very long time. It may here be stated that the reaction
of the food medium is a matter of great importance. Most
bacteria prefer a slightly alkaline medium, and some, e.g. the
cholera spirillum, will not grow in the presence of the smallest
amount of free acid.

Moisture. — The presence of water is necessary for the con-
tinued growth of all bacteria. The amount of drying which
bacteria in the vegetative stage will resist varies very much in
different species. Thus the cholera spirillum is killed by two or
three hours' drying, while the staphylococcus pyogenes aureus
will survive ten days' drying, and the bacillus diphtheriae still
more. In the case of spores the periods are much longer.
Anthrax spores will survive drying for several years, but here
again moisture enables them to resist longer than when they are
quite dry. When organisms have been subjected to»such hostile
influences, even though they survive, it by no means follows that
they retain all their vital properties.

Relation to Gaseous Environment. — The relation of bacteria
to the oxygen of the air is such an important factor in the life
of bacteria that it enables a biological division to be made among
them. Some bacteria will only live and grow when oxygen is
present. To these the title of obligatory aerobes is given. Other
bacteria will only grow when no oxygen is present. These are
called obligatory anaerobes. In still other bacteria the presence
or absence of oxygen is a matter of indifference. This group
might theoretically be divided into those which are preferably
aerobes, but can be anaerobes, and those which are preferably
anaerobes, but can be aerobes. As a matter of fact such
differences are manifested to a slight degree, but all such
organisms are usually grouped as facultative anaerobes, i.e. pre-
ferably aerobic but capable of existing without oxygen. Ex-
amples of obligatory aerobes are b. proteus vulgaris, b. subtilis ; of
obligatory anaerobes, b. tetani, b. oedematis maligni, wThile the
great majority of pathogenic bacteria are facultative anaerobes.
With regard to anaerobes, hydrogen and nitrogen are indifferent

TEMPERATURE AND EFFECT OF LIGHT 19

gases. Many anaerobes, however, do not flourish well in an
•srttnosphere of carbon dioxide. Very few experiments have
been made to investigate the action on bacteria of gas under
pressure. A great pressure of carbon dioxide is said to make
the b. anthracis lose its power of sporing, but it seems to have
no effect on its vitality or on that of the b. typhosus. In
the case of the bacillus pyocyaneus, however, it is said to
destroy life.

Temperature. — For every species of bacterium there is a
temperature at which it grows best. This is called the'
" optimum temperature." ' There is also in each case a
maximum temperature above which growth does not take
place, and a minimum temperature below which growth does
not take place. As a general rule the optimum temperature is
about the temperature of the natural habitat of the organism.
For organisms taking part in the ordinary processes of putrefac-
tion the temperature of warm summer weather (20° to 24° C.)
may l>u taken as the average optimum, while for organisms
normally inhabiting animal tissues 35° to 39° C. is a fair
average. The lowest limit of ordinary growth is from 12° to
14 C., and the upper is from 42° to 44° C. In exceptional
> growth may take place as low as 5° C., and as high as
70° C. Some organisms which grow best at a temperature of from
60° to 70° C. have been isolated from dung, the intestinal tract,
etc. These have been called thermophUic bacteria. It is to
be noted that while growth does not take place below or above
;i certain limit, it by no means follows that death takes place
• Hitside such limits. Organisms can resist cooling below their
minimum or heating beyond their maximum without being
killed. Their vital activity is merely paralysed. Especially is
this true of the effect of cold on bacteria. The results of
different olisrrvers vary; but if we take as an example the
cholera vibrio, Koch found that while the minimum temperature
of growth was 16° C., a culture might be cooled to -32° C.
without being killed. With regard to the upper limit, few
ordinary organisms in a spore-free condition will survive a
temjKirature of 57° C., if long enough applied. Many organisms
lose some of their properties when grown at unnatural temper-
atures. Thus many pathogenic organisms lose their virulence
it grown aliovc their optimum temperature, and some chromogenic
tonns must <>!' \\hich prefer rather low temperatures, lose their
capacity of producing pigment, </.//. spirillum rabmnL

Effect of Light. — Of recent years much attention has been
paid to this factor in the life of bacteria. Direct sunlight is

20 GENERAL MORPHOLOGY AND BIOLOGY

found to have a very inimical effect. It has been found that
an exposure of dry anthrax spores for one and a half hours
to sunlight kills them. When they are .moist, a much longer
exposure is necessary. Typhoid bacilli are killed in about one
and a half hours, and similar results have been obtained with
many other organisms. In such experiments the thickness of
the medium surrounding the growth is an important point.
Death takes place more readily if the medium is scanty or if
the organisms are suspended in water. Any fallacy which
might arise from the effect of the heat rays of the sun has been
excluded, though light plus heat is more fatal than light alone.
In direct sunlight it is chiefly the green, violet, and, it may be,
the ultra-violet rays which are fatal. Diffuse daylight has also
a bad effect upon bacteria, though it takes a much longer ex-
posure to do serious harm. A powerful electric light is as
fatal as sunlight. Here, as with other factors, the results vary
very much with the species under observation, and a distinction
must be drawn between a mere cessation of growth and the
condition of actual death. Some bacteria, especially occurring
on the dead bodies of fresh fish, are phosphorescent.

Conditions affecting the Movements of Bacteria. — In some
cases differences are observed in the behaviour of motile bacteria,
contemporaneous with changes in their life history. Thus, in
the case of bacillus subtilis, movement ceases when sporulation
is about to take place. On the other hand, in the bacillus of
symptomatic anthrax, movement continues while sporulation is
progressing. Under ordinary circumstances motile bacteria
appear not to be constantly moving, but occasionally to rest. In
every case the movements become more active if the temperature
be raised. Most interest, however, attaches to the fact that
bacilli may be attracted to certain substances and repelled by
others. Schenk, for instance, observed that motile bacteria
were attracted to a warm point in a way which did not occur
when the bacteria were dead and therefore only subject to
physical conditions. Most important observations have been
made on the attraction and repulsion exercised on bacteria by
chemical agents, which have been denominated respectively
positive and ' negative chemiotaxis. Pfeffer investigated this
subject in many lowly organisms, including bacterium termo
and spirillum undula. The method was to fill with the agent
a fine capillary tube, closed at one end, to introduce this into
a drop of fluid containing the bacteria under a cover-glass, and
to watch the effect through the microscope. The general result
was to indicate that motile bacteria may be either attracted or

THE PARTS PLAYED BY BACTERIA IN NATURE 21

repelled by the fluid in the tube. The effect of a given fluid
differs in different organisms, and a fluid chemiotactic for one
organism may not act on another. Degree of concentration is
important, but the nature of the fluid is more so. Of inorganic
bodies salts of potassium are the most powerfully attracting
bodies, and in comparing organic bodies the important factor
is the molecular constitution. These observations have been
confirmed by Ali-Cohen, who found that while the vibrio of
cholera and the typhoid bacillus were scarcely attracted by
chloride of potassium, they were powerfully influenced by
potato juice. Further, the filtered products of the growth of
many bacteria have been found to have powerful chemiotactic
properties. It is evident that all these observations have a
most important bearing on the action of bacteria, though we
do not yet know their true significance. Corresponding chemio-
tactic phenomena are shown also by certain animal cells, e.g.
leucocytes, to which reference is made below.

The Parts played by Bacteria in Nature. — As has been said,
tlic chief effect of bacterial action in nature is to break up into
more simple combinations the. complex molecules of the organic
substances which form the bodies of plants and animals, or
which are. excreted by them. In some cases we know some of
the stages of disintegration, but in most cases we know only
general principles and sometimes only results. In the case of
milk, for instance, we know that lastjc acid is produced from
t,frfi lap.f-Qfip, by the action of the bacillus acidi lactici and of
other bacteria, and that from urea ammonium carbonate is
produced by the micrococcus ureae. That the very complicated
process of putrefaction is due to bacteria is absolutely proved,
for any organic substance . can be preserved indefinitely from
ordinary putrefaction by the adoption of some method of
killing all bacteria present in it, as will be afterwards described.
This statement, however, does not exclude the fact that
molecular changes take place spontaneously in the passing of
tlit- organic body from life to death. Many processes not
usually referred to as putrefactive are also bacterial in their
origin. The souring of milk, already referred to, the becoming
rancid of butter, the ripening of cream and of cheese, are all
due to bacteria.

A certain comparatively small number of bacteria have been
I -loved to be the causal agents in some disease processes
occurring in man, animals, and plants. This means that the
fluids and tissues of living bodies are, under certain circum-
>tauces, a suitable pabulum for the bacteria involved. The

22 GENERAL MORPHOLOGY AND BIOLOGY

effects of the action of these bacteria are analogous to those
taking place in the action of the same or other bacteria on dead
animal or vegetable matter. The complex organic molecules
are broken up into simpler products. We shall study these
processes more in detail later. Meantime we may note that
the disease-producing effects of bacteria form the basis of
another biological division of the group. Some bacteria are
harmless to animals and plants, and apparently under no
circumstances give rise to disease in either. These are known
as saprophytes. They are normally engaged in breaking up
dead animal and vegetable matter. Others normally live on
or in the bodies of plants and animals and produce disease.
These are known as parasitic bacteria. Sometimes an attempt
is made to draw a hard-and-fast line between the saprophytes
and the parasites, and obligatory saprophytes or parasites are
spoken of. This is an erroneous distinction. Some bacteria
which are normally saprophytes can produce pathogenic effects
(e.g. bacillus oedematis maligni), and it is consistent with our
knowledge that the best-known parasites may have been derived
from saprophytes. On the other hand, the fact that most
bacteria associated with disease processes, and proved to be
the cause of the latter, can be growTi in artificial media, show's
that for a time at least such parasites can be saprophytic. As
to how far such a saprophytic existence of disease-producing
bacteria occurs in nature, we are in many instances still
ignorant.

The Methods of Bacterial Action. — The processes which
bodies undergo in being split up by bacteria depend, first, on
the chemical nature of the bodies involved, and, secondly, on
the varieties of the bacteria which are acting. The destruction
of albuminous bodies which is mostly involved in the wide and
varied process of putrefaction can be undertaken by whole
groups of different varieties of bacteria. The action of the
latter on such substances is analogous to what takes place when
albumins are subjected to ordinary gastric and intestinal
digestion. In these circumstances, therefore, the production
of albumoses, peptones, etc., similar to those of ordinary
digestion, can be recognised in putrefying solutions, though
the process of destruction always goes further, and still simpler
substances, e.g. indol, and, it may be, crystalline bodies of an
alkaloidal nature, are the ultimate results. The process is
an exceedingly complicated one when it takes place in nature,
and different bacteria are probably concerned in the different
stages. Many other bacteria, e.g. some pathogenic forms,

THE METHODS OF BACTERIAL ACTION 23

though not concerned in ordinary putrefactive processes, have
a similar digestive rapacity. When carbohydrates are being
split up, then various alcohols, ethers, and acids are produced.
l)uring bacterial growth there is not infrequently the abundant
production of such gases as sulphuretted hydrogen, carbon
dioxide, methane, etc.. For an exact knowledge of the de-
structive capacities of any particular bacterium there must be
an accurate chemical examination of its effects when it has
been grown in artificial media the nature of which is known.
The precise substances it is capable of forming can thus be
found out. Many substances, however, are produced by
bacteria, of the exact nature of which we are still ignorant,
for example, the toxic bodies which play such an important
part in the action of many pathogenic species.

Many of the actions of bacteria depend on the production by
them si ferments of a very varied nature and complicated action.
Thus the digestive action on albumins probably depends on the
production of a peptic ferment analogous to that produced in the
animal stomach. Ferments which invert sugar, which .split
up sugars into alcohols or acids, which coagulate casein, which
split up mva into ammonium carbonate, also occur.

Such ferments may be diffused into the surrounding fluid, or
be retained in the cells where they are formed. Sometimes the
breaking down of the organic matter appears to take place
within, or in the immediate proximity of, the bacteria, some-
times wherever the soluble ferments reach the organic substances.
And in certain cases the ferments diffusing out into the surround-
ing medium probably break down the constituents of the latter
t-> some extent, and prepare them for a further, probably
intracellular, disintegration. Thus, in certain putrefactions of
fibrin, if the process be allowed to go on naturally, the fibrin
dissolves and ultimately great gaseous evolution of carbon
di«>xidf and ammonia takes place, but if the bacteria, shortly
after the process has begun, are killed or paralysed by chloro-
form, then only a peptonisation of the fibrin occurs, without
the further splitting up and gaseous production. That a
purely intracellular digestion may take place is illustrated by
what has been shown to occur in the case of the micrccoccus
urea-, which from urea forms ammonium carbonate by adding
water to the urea molecule. Here, if after the action has
commenced the bacteria are filtered off, no further production
of ammonium carbonate takes place, which shows that no
ferment has been dissolved out into the urine. If now the
bodies of the bacteria be extracted with absolute alcohol or ether,

24 GENERAL MORPHOLOGY AND BIOLOGY

which of course destroy their vitality, a substance is obtained
of the nature of a ferment, which, when added to sterile urine,
rapidly causes the production of ammonium carbonate. This
ferment has evidently been contained within the bacterial cells.
In the investigation of the phenomena of the ferment action of
bacteria, it has been noted in certain cases that the ferments
formed depend on the food supply offered to the bacterium.
Thus in one case a bacterium growing in starch forms diastase,
which it does not do when grown on sugar.

In considering the effects of bacteria in nature, it must be recognised
that some species are capable of building up complex substances out of
simple chemical compounds. Examples of these are found in the bacteria
which in the soil make nitrogen more available for plant nutrition by
converting ammonia into nitrites and nitrates. Winogradski, by using
media containing non-nitrogenous salts of magnesium, potassium, and
ammonium, and free of organic matter, has demonstrated the existence
of forms which convert, by oxidation, ammonia into nitrites, and of other
forms which convert these nitrites into nitrates. Both can derive their
necessary carbon from alkaline carbonates. Other bacteria, or organisms
allied to the bacteria, exist which can actually take up and combine into
new compounds the free nitrogen of the air. These are found in the
tubercles which develop on the rootlets of the leguminosre. Without
such organisms the tubercles do not develop, and without the development
of the tubercles the plants are poor and stunted. Bacteria thus play an
important part in the enrichment and fertilisation of the soil.

The Occurrence of Variability among Bacteria. — The question of the
division of the group of bacteria into definite species has given rise to
much discussion among vegetable and animal morphologists, and at one
time very divergent views were held. Some even thought that the
same species might at one time give rise to one disease, — at another time
to another. There is, however, now practical unanimity that bacteria
show as distinct species as the other lower plants and animals, though,
of course, the difficulty of defining the concept of a species is as great in
them as it is in the latter. Still, we can say that among the bacteria AVC
see exhibited (to use the words of De Bary) "the same periodically
repeated course of development within certain empirically determined
limits of variation " which justifies, among higher forms of life, a species
to be recognised. What at first raised doubts as to the occurrence of
species among the bacteria was the observation in certain cases of what is
known as pleomorpliism. By this is meant that one species may assume
at different times different forms, e.g. appear as a coccus, a bacillus, or
a leptothrix. Undoubtedly, many of the cases where this was alleged
to have been observed occurred before the elaboration of the modern
technique for the obtaining of pure cultures, but at the present day there
are cases where evidence appears to exist of the occurrence of pleomorphism.
This is especially the case with certain bacilli, and it may lead to such
forms being classed among the higher bacteria. Pleomorphism is,
however, a rare condition, and with regard to the bacteria as a whole we
may say that each variety tends to conform to a definite type of structure
and function which is peculiar to it and to it alone. On the other hand,
slight variations from such type can occur in each. The size may vary

VARIABILITY AMONG BACTERIA 25

a little with the medium in which the organism is growing, and under
certain similar conditions the adhesion of bacteria to each other may also
vary. Thus cocci, which are ordinarily seen in short chains, may grow
in long chains. The capacity to form spores may be altered, and such
properties as the elaboration of certain ferments or of certain pigments
may be impaired. Also the characters of the growths on various media
may undergo variations. As has been remarked, variation as observed
consists largely in a tendency in a bacterium to lose properties ordinarily
possessed, and all attempts to transform one bacterium into an apparently
closely allied variety (such as the b. coli into the b. typhosus) have
failed. This of course does not preclude the possibility of one species
having been originally derived from another, or of both having descended
from a common ancestor, but we can say that only variations of an
unimportant order have been observed to take place, and here it must be
remembered that in many cases we can have forty-eight or more
generations under observation within twenty-four hours.

CHAPTER IT.

METHODS OF CULTIVATION OF BACTERIA.

Introductory. — In order to study the characters of any species
of bacterium, it is necessary to have it growing apart from every
other species. In the great majority of cases where bacteria
occur in nature, this condition is not fulfilled. Only in the
blood and tissues in some diseases do particular species occur
singly and alone. We usually have, therefore, to remove a
bacterium from its natural surroundings and grow it on an
artificial food medium. When we have succeeded in separating
it, and have got it to grow on a medium which suits it, wre are
said to have obtained a pure culture. The recognition of
different species of bacteria depends, in fact, far more on the
characters presented by pure cultures and their behaviour in
different food media, than on microscopic examination. The
latter in most cases only enables us to refer a given bacterium
to its class. Again, in inquiring as to the possible possession of
pathogenic properties by a bacterium, the obtaining of pure
cultures is absolutely essential.

To obtain pure cultures, then, is the first requisite of bacterio-
logical research. Now, as bacteria are practically omnipresent,
we must first of all have means of destroying all extraneous
organisms which may be present in the food media to be used
in the vessels in which the food media are contained, and on all
instruments which are to come in contact with our cultures.
The technique of this destructive process is called sterilisation.
We must therefore study the methods of sterilisation. The
growth of bacteria in other than their natural surroundings
involves further the preparation of sterile artificial food media,
and when we have such media prepared we have still to look
at the technique of the separation of micro-organisms from
mixtures of these, and the maintaining of pure cultures when the
latter have been obtained. We shall here find that different
methods are necessary according as we are dealing with aerobes

STERILISATION BY DRY HEAT 27

or anaerobes. Each of these methods will l>u considered in

turn.

THE METHODS OF STERILISATION.

To exclude extraneous organisms, all food materials, glass
vessels containing them, wires used in transferring bacteria from
one culture medium to another, instruments used in making
autopsies, etc., must be sterilised. These objects being so
different, various methods are necessary, but underlying these
methods is the general principle that all bacteria are destroyed
by heat. The temperature necessary varies with different
bacteria, and the vehicle of heat is also of great importance.
The two vehicles employed are hot air and hot water or steam.
The former is usually referred to as " dry heat," the latter as
"moist heat." As showing the different effects of the two
vehicles, Koch found, for instance, that the spores of bacillus
anthraois, which were killed by moist heat at 100° C., in one
hour, required three hours' dry heat at 140° C. to effect death.
Both forms of heat may be applied at different temperatures —
in the case of moist heat above 100° C., a pressure higher than
that of the atmosphere must of course be present.

A. Sterilisation by Dry Heat.

A. (1) Red Heat or Dull Red Heat. — Red heat is used for
the sterilisation of the platinum needles which, it will be found,
are so constantly in use. A dull heat is used for cauteries, the
I mints of forceps, and maybe used for the incidental sterilisation
of small glass objects (cover-slips, slides, occasionally when
necessary even test-tubes), care of course being taken not to
melt the glass. The heat is obtained by an ordinary Bunsen
burner.

A. ('!) Sterilisation by Dry Heat in a Hot- Air Chamber.—
The chamber (Fig. 2) consists of an outer and inner case of
sheet iron. In the bottom of the outer there is a large hole.
A Bunsen is lit beneath this, and thus plays on the bottom of
the inner case, round all the sides of which the hot air rises
and escapes through holes in the top of the outer case. A
thermometer passes down into the interior of the chamber, half-
way up which its bulb should be situated. It is found, as a
matter of experience, that an exposure in such a chamber for
one hour to a temperature of 160° C., is sufficient to kill all the
organisms which usually pollute articles in a bacteriological
laboratory, though circumstances might arise where this would

28 METHODS OF CULTIVATION OF BACTERIA

be insufficient.

This means of sterilisation is used for the glass
flasks, test-tubes, plates, Petri's
dishes, the use of which will
be described. Such pieces of
apparatus are thus obtained
sterile and dry. It is advisable
to put glass vessels into the
chamber before heating it, and
to allow them to stand in it
after sterilisation till the tem-
perature falls. Sudden heating
or cooling is apt to cause
glass to crack. The method is
manifestly unsuitable for food
media.

B. Sterilisation by Moist
Heat.

FIG. 2.— Hot-air steriliser. B. (1) By Boiling. — The

boiling of a liquid for five

minutes is sufficient to kill ordinary germs if no spores be
present, and this method is useful for sterilising distilled or tap
water which may be required in various manipulations. It is
best to sterilise knives and instruments
used in autopsies by boiling in water to
which a little sodium carbonate has been
added to prevent rusting. Twenty minutes'
boiling will here be sufficient. The boiling
of any fluid at 100° C. for one and a half
hours will ensure sterilisation under almost
any circumstances.

B. (2) By Steam at 100° C.— This is by
far the most useful means of sterilisation.
It may be accomplished in an ordinary
potato steamer placed on a kitchen pot.
The apparatus ordinarily used is "Koch's
steam steriliser " (Fig. 3). This consists of
a tall metal cylinder on legs, provided with
a lid, and covered externally by some bad
conductor of heat, such as felt or asbestos.
A perforated tin diaphragm is fitted in

the interior at a little distance above the ™T ^ , ,

T, , ,, . , FIG. 3. — Koch s steam

bottom, and there is a tap at the bottom steriliser.

STERILISATION BY STEAM 29

by which water may be supplied or withdrawn. If water
to the depth of 3 inches be placed in the interior and heat
applied, it will quickly boil, and the steam streaming up will
surround any flask or other object standing on the diaphragm.
Here no evaporation takes place from any medium, as it is sur-
rounded during sterilisation by an atmosphere saturated with
\\atrr vapour. It is convenient to have the cylinder tall enough
to hold a litre flask with a funnel 7 inches in diameter standing
in its neck. The funnel may be supported by passing its tube
through a second perforated diaphragm placed in the upper part
of the steam chamber. With such a " Koch " in the laboratory
a hot- water filter is not needed. As has been said, one and a
half hour's steaming will sterilise any medium, but in the case
of media containing gelatin such an exposure is not practicable,
as, with long boiling, gelatin tends to lose its physical property
of solidification. The method adopted in this case is to steam
for twenty minutes on each of three succeeding days.

This is ;i modification of what is known as "Tyndall's intermittent
>t< rilisation." The fundamental principle of this method is that all
ba'-tt'ria in a non-spored form are killed by the temperature of boiling
\\ater. while if in a spored form they may not be thus killed. Thus by
the sterilisation mi the first day all the non-spored forms are destroyed —
the spores remaining alive. During the twenty-four hours which intervene
before the next heating, these spores, being in a favourable medium, are
likely to assume the non-spored form. The next heating kills these. In
case any may still not have changed their spored form, the process is
repeated on a third day. Experience shows that usually the medium
can now be kept indefinitely in a sterile condition.

Steam at 100° C. is therefore available for the sterilisation of
all ordinary media. In using the Koch's steriliser, especially
wlu-n a large bulk is to be sterilised, it is best to put the medium
in while the apparatus is cold, in order to make certain that the
whole of the food mass reaches the temperature of 100° C. The
I'fiiod of exposure is reckoned from the time boiling commences
in the water in the steriliser. At any rate allowance must
always In- made for the time required to raise the temperature
of the medium to that of the steam surrounding it.

B. (:\) Sterilisation by Steam at High Pressure. — This is
the most rapid and effective means of sterilisation. It is effected
in an autoclave (Fig. 4). This is a gun-metal cylinder supported
in a cylindrical sheet-iron case ; its top is fastened down with
screws and nuts, and is furnished with a safety valve, pressure-
^aiige, and a hole for thermometer. As in the Koch's steriliser,
the contents are supported on a perforated diaphragm. The

30 METHODS OF CULTIVATION OF BACTERIA

0 0 o

oooo

source of heat is a large Bunsen beneath. The temperature
employed is usually 115° C. or 120° C. To boil at 115° C.,
water requires a pressure of about 23 Ibs. to the square inch
(i.e. 8 Ibs. plus the 15 Ibs. of ordinary atmo-
spheric pressure). To boil at 120° C., a
pressure of about 30 Ibs. (i.e. 15 Ibs. plus
the usual pressure) is necessary. In such
an apparatus the desired temperature is
maintained by adjusting the safety-valve so
as to blow off at the corresponding pressure.
One exposure of media to such temperatures
for a quarter of an hour is amply sufficient
to kill all organisms or spores. Here, again,
care must be taken when gelatin is to be
sterilised. It must not be exposed to a
temperature above 105° C., and is best
sterilised by the intermittent method.
Certain precautions are necessary in using
the autoclave. In all cases it is necessary
to allow the apparatus to cool well below
100° C. before opening it or allowing steam
to blow off, otherwise there will be a sudden
development of steam when the pressure is
removed, and fluid media will be blown out
of the flasks. Sometimes the instrument is
not fitted with a thermometer. In this case care must be
taken to expel all the air initially present, otherwise, a mixture
of air and steam being present, the pressure read off the gauge
cannot be accepted as an indication of the temperature. Further,
care must be taken to ensure the presence of a residuum of
water when steam is fully up, otherwise the steam is super-
heated, and the pressure on the gauge again does not indicate
the temperature correctly.

B. (4) Sterilisation at Low Temperatures. — Most organisms
in a non-spored form are killed by a prolonged exposure to a
temperature of 57° C. This fact has been taken advantage of
for the sterilisation of blood serum, which will coagulate if
exposed to a temperature above that point. Such a medium is
sterilised on Tyndall's principle by exposing it for an hour at
57° C. for eight consecutive days, it being allowed to cool in the
interval to the room temperature. The apparatus shown in
Fig. 5 is a small hot-water jacket heated by a Bunsen placed
beneath it, the temperature being controlled by a gas regulator.
To ensure that the temperature all around shall be the same,

FIG. 4. — Autoclave.

a. Safety-valve.
6. Blow-off pipe.
c. Gauge.

PREPARATION OF ORDINARY CULTURE MEDIA 31

the lid also is hollow and filled with water, and there is a
special gas burner at the side to heat it. This is the form
originally used, but serum sterilisers are now constructed in
which the test-tubes are placed iu

the

position, and in which

inspissatiori (vide p. 40) can after-
wards be performed at a higher
temperature.

THE PREPARATION OF ORDINARY
CULTURE MEDIA.

The general principle to be observed
in the artificial culture of bacteria is
that the medium used should approxi-
mate as closely as possible to that on
which the bacterium growrs naturally.
In the case of pathogenic bacteria the
medium therefore should resemble the
juices of the body. The serum of
the blood satisfies this condition, and
is often used, but its application is

limited by the difficulties in its F]G. 5._steriliser for blood
preparation and preservation. Other serum.

media have been found which can

support the life of all the pathogenic bacteria isolated. These
consist of proteids or carbohydrates in a fluid, semi-solid,
or solid form, in a transparent or opaque condition. The
advantage of having a variety of media lies in the fact that
growth characters on particular media, non-growth on some
and growth on others, etc., constitute specific differences
which are valuable in the identification of bacteria. The
most commonly used media have as their basis a watery
extract of meat. Most bacteria in growing in such an
extract cause only a grey turbidity. A great advance resulted
when Koch, by adding to it gelatin, provided a transparent
solid medium in which growth characteristics of particular
bacteria become evident. Many organisms, however, grow best
at a temperature at which this nutrient gelatin is fluid, and
therefore another gelatinous substance called agar, which does
not melt below 98° C., was substituted. Bouillon made from
Mirat extract, gelatin, and agar media, and the modifications
of these, constitute the chief materials in which bacteria are
grown.

32 METHODS OF CULTIVATION OF BACTERIA

Preparation of Meat Extract.

The flesh of the ox, calf, or horse is usually employed.
Horse-flesh has the advantage of being cheaper and containing
less fat than the others ; though generally quite suitable, it has
the disadvantage for certain purposes of containing a larger
proportion of fermentable sugar. The flesh must be freed from
fat, and finely minced. To a pound of mince add 1000 c.c.
distilled water, and mix thoroughly in a shallow dish. Set
aside in a cool place for twenty-four hours. Skim off any fat
present, removing the last traces by stroking the surface of the
fluid with pieces of filter paper. Place a clean linen cloth over
the mouth of a large filter funnel, and strain the fluid through
it into a flask. Pour the minced meat
into the cloth, and, gathering up the
edges of the latter in the left hand,
squeeze out the juice still held back in
the contained meat. Finish this expres-
sion by putting the cloth and its contents
into a meat press (Fig. 6), similar to
that used by pharmacists in preparing
extracts ; thus squeeze out the last drops.
The resulting sanguineous fluid contains
the soluble albumins of the meat, the
soluble salts, extractives, and colouring
matter, chiefly haemoglobin. It is now
FIG. 6.— Meat press. boiled thoroughly for two hours, by
which process the albumins coagulable

by heat are coagulated. Strain now through a clean cloth,
boil for another half-hour, and filter through white Swedish
filter paper (best, C. Schleicher . u. Schull, No. 595). Make
up to 1000 c.c. with distilled water. The resulting fluid
ought to be quite transparent, of a yellowish colour without
any red tint. If there is any redness, the fluid must be
reboiled and filtered till this colour disappears, otherwise in
the later stages it will become opalescent. A large quantity
of the extract may be made at a time, and what is not
immediately required is put into a large flask, the neck plugged
with cotton wool, and the whole sterilised by methods B (2) or
(3). This extract contains very little albuminous matter, and
consists chiefly of the soluble salts of the muscle, certain
extractives, and altered colouring matters, along with any slight
traces of soluble proteid not coagulated by heat. It is of acid
reaction. We have now to see how, by the addition of proteid

BOUILLON MEDIA 33

and other matter, it may be transformed into proper culture
media.

1. Bouillon Media.— These consist of meat extract with the
addition of certain substances to render them suitable for the
growth of bacteria.

(1) (a). Peptone Broth or Bouillon. — This has the com-
position : —

Meat extract1 .... 1000 c.c.
Sodium chloride ... 5 grms.

Peptone albumin . . . 10 „

Boil till the ingredients are quite dissolved, and neutralise
with a saturated solution of sodium hydrate. Add the latter
drop by drop, shaking thoroughly between each drop and testing
the reaction by means of litmus paper. Go on till the reaction
is slightly but distinctly alkaline. Neutralisation must be
practised with great care, as under certain circumstances,
depending on the relative proportions of the different phosphates
of sodium and potassium, what is known as the amphoteric
reaction is obtained, i.e. red litmus is turned blue, and blue red,
by the same solution. The sodium hydrate must be added till
red litmus is turned slightly but distinctly blue, and blue litmus
is not at all tinted red. After alkalinisation, allow the fluid to
become cold, filter through Swedish filter paper into flasks,
make up to original volume with distilled water, plug the flasks
with cotton wool, and sterilise by methods B (2) or (3) (pp. 28,
29). This method of neutralisation is to be recommended for
all ordinary work.

In tliis medium the place of the original albumins of the meat is taken
by peptone, a soluble proteid not coagulated by heat. Here it may be
remarked that the commercial peptone albumin is not pure peptone, but
a mixture of albmnoses (see footnote, p. 193) with a variable amount of
] m re peptone. Tin- addition of the sodium chloride is necessitated by
the fact that alkalinisation precipitates some of the phosphates and
carbon.it es present. Experience has shown that sodium chloride can
quite well be substituted. The reason for the alkaliuisation is that it is
found that most bacteria grow best on a medium slightly alkaline to
litmus. Some, e.g. the cholera vibrio, will not grow at all on even a
slightly acid medium.

Standardisation of Reaction of Media. — While the above
procedure of dealing with the reaction of a medium is sufficient
for ordinary work, it has been thought advisable to have a more

1 Some workers, instead of meat extract as made above, use Liebig's
extract of beef, 2 grammes to the litre.

34 METHODS OF CULTIVATION OF BACTERIA

exact method for making media to be used in growing organisms,
the growth characteristics of which are to be described for
systematic purposes. Such a method should also be used in
studying the changes in reaction produced in a medium by the
growth of bacteria. It, however, involves considerable difficulty,
and should not be undertaken by the beginner. It entails the
preparation of solutions of acid and alkali which may be used
for determining the original reaction of the medium, and for
accurately making it of a definite degree of alkalinity. Normal l
and decinormal solutions of sodium hydrate and hydrochloric
acid are used.

Preparation of Standard Solutions. — The first requisites here are
normal solutions of acid and alkali. The latter is prepared as follows :
85 grammes of pure sodium bicarbonate are heated to dull redness for
ten minutes in a platinum vessel and allowed to cool in an exsiccator ;
just over 54 grammes of sodium carbonate should now be present. Any
excess is quickly removed, and the rest being dissolved in one litre of
distilled water, a normal solution is obtained. A measured quantity is
placed in a porcelain dish, and a few drops of a '5 per cent, solution of
phenol-phthaleine in neutral methylated spirit is added to act as
indicator. The alkali produces in the latter a brilliant rose-pink, which,
however, disappears on the least excess of acid being present. The
mixture is boiled and a solution of hydrochloric acid of unknown strength
is run into the dish from a burette till the colour goes and does not
return after very thorough stirring. The strength of the acid can then
be calculated, and a normal solution can be obtained. From these two
solutions any strength of acid or alkali (such as the decinormal solution
of NaOH mentioned below) may be derived.

As Eyre has suggested, the reaction of a medium may be
conveniently expressed by the sign + or — to indicate acid or
alkaline respectively, and a number to indicate the number of
cubic centimetres of normal acid or alkaline solution necessary
to make a litre of the medium neutral to phenol-phthaleine.
Thus, for example, "reaction = -15," will mean that the
medium is alkaline, and requires 15 c.c. of normal HC1 to make
a litre neutral. It has been found that when a medium such
as bouillon reacts neutral to litmus, its reaction to phenol-

*A "normal" solution of any salt is prepared by dissolving an
' ' equivalent " weight in grammes of that salt in a litre of distilled
water. If the metal of the salt be monovalent, i.e. if it be replaceable in
a compound by one atom of hydrogen (e.g. sodium), an equivalent is the
molecular weight in grammes. In the case of NaCl, it would be 58*5
grammes (atomic Aveight of Na = 23, of Cl = 35'5). If the metal be
bivalent, i.e. requiring two atoms of H for its replacement in a compound
(e.g. calcium), an equivalent is the molecular weight in grammes divided
by two. Thus in the case of CaCl2 an equivalent would be 55 '5 grammes
(atomic weight of Ca = 40, of C12 = 71 ).

STANDARDISING THE REACTION OF MEDIA 35

I'luhaleine, according to the above standard, is on the average
+ 25. Now, as litmus was originally introduced by Koch, and
as nearly all bacterial research has been done with media tested
by litmus, it is evidently difficult to say exactly what precise
degree of alkalinity is the optimum for bacterial growth. It is
probably safe to say, however, that when a medium has been
rendered neutral to phenol-phthaleine by the addition of NaOH,
the optimum degree is generally attained by the addition of
from 10 to 15 c.c. of normal HC1 per litre, i.e. the optimum
reaction is from +10 to +15. In other words, the optimum
reaction for bacterial growth lies, as Fuller has pointed out,
about midway between the neutral point indicated by phenol-
plithaleine and the neutral point indicated by litmus.

The only objection to the use of phenol-phthaleine is that
its action is somewhat vitiated if free CO2 be present. This
can be obviated by boiling any 'medium, before it is tested,
in the porcelain dish into which titration takes place. The soda
solutions are best stored in bottles such as that shown in Fig. 42,
1 laving on the air inlet a little bottle filled with soda lime and
fitted with tubes as in the large one. The CO.2 of the air which
passes through is thus removed.

Method. — The following procedure includes most of the
improvements introduced by Eyre. The medium with all its
constituents dissolved is filtered and then heated for about forty-
five minutes in the steamer, the maximum acidity being reached
after this time. Of the warm medium take 25 c.c. and put in
a porcelain dish, add 25 c.c. distilled water, and 1 c.c. phenol-
phthaleine solution. Run in decinormal soda till neutral point
is reached, indicated by the first trace of pink colour, the
mixture being kept hot.1 Repeat process thrice, and take the
mean; this divided by 10 will give the amount (x) of normal
soda required to neutralise 25 c.c. of medium ; then 40 x =
amount necessary to neutralise a litre ; and 40 x — 10 = amount
of normal soda necessary to give a litre its optimum reaction.
Then measure the amount of medium to be dealt with, and add
the requisite amount of soda solution.

Eyre uses a soda solution of ten times normal strength, which
is delivered out of a 1 c.c. pipette divided into hundred ths ; this

1 The beginner may find considerable difficulty in recognising the first
tint of pink in the yellow bouillon. A good way of getting over this is
to take two samples of the medium, adding the indicator to one only ;
then to run the soda into these from separate burettes ; for each few
drnjis run into the medium containing the indicator the same amount is
run into the other. Thus the recognition of the first permanent change
in tint will be at once recognised by comparing the two lots of solution.

36 METHODS OF CULTIVATION OF BACTERIA

obviates, to a large extent, the error introduced by increasing
the bulk of the medium if a weaker neutralising solution be
used.

1 (b). Glucose Broth. — -To the other constituents of 1 (a)
there is added 1 or 2 per cent, of grape sugar. The steps in the
preparation are the same. Glucose being a reducing agent, no
free oxygen can exist in a medium containing it, and therefore
glucose broth is used as a culture fluid for anaerobic organisms.

1 (c). Glycerin Broth. — The initial steps are the same as in
1 (a), but after filtration 6 to 8 per cent, of glycerin (sp. grav.
1'25) is added. This medium is especially used for growing the
tubercle bacillus when the products of the growth of the latter
are required.

2. Gelatin Media. — These are simply the above broths, with
gelatin added as a solidifying body.

2 (a). Peptone Gelatin : —

Meat extract
Sodium chloride .
Peptone albumin .
Gelatin

. 1000 c.c.

5 grms.
10 „
100-150

(The "gold label " gelatin of Coignet et Cie, Paris, is the best.)
The gelatin is cut into small pieces, and added with the other
constituents to the extract ; they are then thoroughly melted on
a sand bath, or in the " Koch." The
fluid medium is then rendered slightly
alkaline, as in 1 (a), and filtered
through filter paper. As the medium
must not be allowed to solidify
during the process, it must be kept
warm. This is effected by putting
the flask and funnel into a tall
Koch's steriliser, in which case the
funnel must be supported on a
tripod or diaphragm, as there is great
danger of the neck of the flask break-
ing if it has to support the funnel
and its contents. The filtration may
also be carried out in a funnel
with water-jacket which is heated, as

shown in Fig. 7. Whichever instrument be used, before filtering
shake up the melted medium, as it is apt while melting to have
settled into layers of different density. Sometimes what first
comes through is turbid. If so, replace it in the unfiltered

FIG. 7. — Hot-water funnel.

AGAR MEDIA 37

part : often the subsequent filtrate in such circumstances is
quite clear. A litre flask of the finished product ought to be
quite transparent. If, however, it is partially opaque, add
the white of an egg, shake up well, and boil thoroughly over
the sand bath. The consequent coagulation of the album in
carries down the opalescent material, and, on making up with
distilled water to the original quantity and refiltering, it will be
found to be clear. The flask containing it is then plugged with
cotton wool and sterilised, best by method B (2), p. 28. If the
autoclave be used the temperature employed must not be above
105° C., and exposure not more than a quarter of an hour on
three successive days. Too much boiling, or boiling at too high
a temperature, as has been said, causes a gelatin medium to lose
its property of solidification. The exact percentage of gelatin
used in its preparation depends on the temperature at which
growth is to take place. Its firmness is its most valuable
characteristic, and to maintain this in hot summer weather, 15
parts per 100 are necessary. A limit is placed on higher per-
centages by the fact that, if the gelatin be too stiff, it will split
on the perforation of its substance by the platinum needle used
in inoculating it with a bacterial growth ; 1 5 per cent, gelatin
melts at about 24° C. For ordinary use in British laboratories
10 per cent, gelatin is a sufficient strength.

2 (b). Glucose Gelatin. — The constituents are the same as
2 (a), with the addition of 1 to 2 per cent, of grape sugar. The
method of preparation is identical. This medium is used for
growing anaerobic organisms at the ordinary temperatures.

3. Agar Media (French, " ge'lose ").— The disadvantage of
gelatin is that at the blood temperature (38° C.), at which most
pathogenic organisms grow best, it is liquid. To get a medium
which will be solid at this temperature, agar is used as the
stiffening agent instead of gelatin. Unlike the latter, which
is a proteid, agar is a carbohydrate. It is derived from the
stems of various seaweeds growing in the Chinese seas, com-
mercially classed together as " Ceylon Moss." For bacteriological
purposes the dried stems of the seaweed may be used, but there
is in the market a purified product in the form of a powder,
which is preferable.

3 (a). " Ordinary " Agar. — This has the following composi-
tion : —

Mi-at extract 1000 c.c.

Sodium chloride .... 5 grms

Peptone albumin . . . . 10 „
ir 15

38 METHODS OF CULTIVATION OF BACTERIA

Cut up the agar into very fine fragments (in fact till it is as
nearly as possible dust), add to the meat extract with the other
ingredients, and preferably allow to stand all night. Then boil
gently in a water bath for two or three hours, till the agar is
thoroughly melted. The process of melting may be hastened
by boiling the medium in a sand bath and passing through it a
stream of steam generated in another flask ; the steam is led
from the second flask by a bent glass tube passing from just
beneath the cork to beneath the surface of the medium (Eyre).
After melting, render slightly alkaline with sodium hydrate
solution, make up to original volume with distilled water, and
filter. Filtration here is a very slow process, and is best carried
out in a tall Koch's steriliser. In doing this, it is well to put a
glass plate over the filter funnel to prevent condensation water
from dropping oft7 the lid of the steriliser into the medium. If
a slight degree of turbidity may be tolerated, it is sufficient to
filter through a felt bag or jelly strainer. Plug the flask con-
taining the filtrate, and sterilise either in autoclave for fifteen
minutes or in Koch's steriliser for one and a half hours.
Agar melts just below 100° C., and on cooling solidifies about
39° C.

3 (b). Glycerin Agar.— To 3 (a) after filtration add 6 to 8
per cent, of glycerin and sterilise as above. This is used
especially for growing the tubercle bacillus.

3 (c). Glucose Agar. — Prepare as in 3 (a), but add 1 to 2
per cent, of grape sugar along with agar. This medium is used
for the culture of anaerobic organisms at temperatures above the
melting-point of gelatin. It is also an excellent culture medium
for some aerobes, e.g. the b. diphtherise.

These bouillon, gelatin, and agar preparations constitute
the most frequently used media. Growths in bouillon do not
usually show any characteristic appearances which facilitate
classification, but such a medium is of great use in investigating
the soluble toxic products of bacteria. The most characteristic
developments of organisms take place on the gelatin media.
These have, however, the disadvantage of not being available
when growth is to take place at any temperature above 24° C.
For higher temperatures agar must be employed. Agar is, how-
ever, never so transparent. Though quite clear when fluid, on
solidifying it always becomes slightly opaque. Further, growths
upon it are never so characteristic as those on gelatin. It is,
for instance, never liquefied, whereas some organisms, by their
growth, liquefy gelatin and others do not — a fact of prime
importance.

SPECIAL CULTURE MEDIA 39

SPECIAL CULTURE MEDIA.

An enormous variety of different media has been brought
forward for use in cases either where special difficulty is ex-
perienced in getting an organism to grow, or where some special
growth characteristic is to be studied. It is impossible to do
more than give the chief of these.

Peptone Solution.

A simple solution of peptone (Witte) constitutes a suitable
culture medium for many bacteria. The peptone in the propor-
tion of 1 to 2 per cent., along with '5 per cent. NaCl, is dissolved
in distilled water by heating. The fluid is then filtered, placed
in tubes, and sterilised. The reaction is usually distinctly
alkaline, which condition is suitable for most purposes. For
special purposes the reaction may be standardised. In such a
solution the cholera vibrio grows with remarkable rapidity. It
is also much used for testing the formation of indol by a
particular bacterium ; and by the addition of one of the sugars
to it the fermentative powers of an organism may be tested
(p. 80). Litmus may be added to show any change in reaction.

Media containing an Indicator.

Litmus Media. — To any of the ordinary media litmus (French,
tournesol) may be added to show change in reaction during
bacterial growth. The litmus is added, before sterilisation, as
a strong watery solution (e.g. the Kubel-Tiemann solution, vide
p. 48) in sufficient quantity to give the medium a distinctly
bluish tint. During the development of an acid reaction the
colour changes to a pink, and may subsequently be dis-
charged.

Neutral Red Media. — This dye has been introduced as an aid
in determining the presence or absence of members of the b. coli
group, especially in the examination of water. The media found
most suitable are agar or bouillon containing '5 per cent, of
lactose, to which '5 per cent, of a 1 per cent, watery solution
of neutral red is added. The alkaline medium is of "a yellowish
brown colour which on the presence of acid passes into a deep
rose red. Sometimes there subsequently occurs a change to a
fluorescent green, caused apparently by a change in the com-
position of the dye, as the fluorescence is not discharged by
addition of alkali.

40 METHODS OF CULTIVATION OF BACTERIA

Blood Serum Media.

Koch's Blood Serum. — Koch introduced this medium, and it
is prepared as follows : Plug the mouth of a tall cylindrical glass
vessel (say of 1000 c.c. capacity) with cotton wool, and sterilise
by steaming it in a Koch's steriliser for one and a half hours.
Take it to the place where a horse, ox, or sheep is to be killed.
When the artery or vein of the animal is opened, allow the first
blood which flows, and which may be contaminated from the
hair, etc., to escape; fill the vessel with the blood subsequently
shed. Carry carefully back to the laboratory without shaking,
and place for twenty-four hours in a cool place, preferably an ice
chest. The clear serum will separate from the clotted blood.
If a centrifuge is available, a large yield of serum may be obtained
by centrifugalising the freshly drawn blood. If coagulation has
occurred, the clot must first be thoroughly broken up. With a
sterile 10 c.c. pipette, transfer this quantity of serum to each of
a series of test-tubes which must previously have been sterilised
by dry heat. The serum may, with all precautions, have been
contaminated during, the manipulations, and must be sterilised.
As it will coagulate if heated above 68° C., advantage must be
taken of the intermittent process of sterilisation at 57° C.
[method B (4)]. It is therefore kept for one hour at this
temperature on each of eight successive days. It is always
well to incubate it for a day at 37° C. before use, to see that
the result is successful. After sterilisation it is "inspissated,"
by which process a clear solid medium is obtained. " Inspissa-
tion " is probably an initial stage of coagulation, and is effected
by keeping the serum at 65° C. till it stiffens. This temperature
is just below the coagulation point of the serum. The more
slowly the operation is performed the clearer will be the serum.
The apparatus used for the purpose is one of the various forms
of serum steriliser (e.g. Fig. 8), generally a chamber with water-
jacket heated with a Bunsen below. The temperature is con-
trolled by a gas regulator, and such an apparatus can, by altering
the temperature, be used either for sterilisation or inspissation.
As is evident, the preparation of this medium is tedious, but its
use is necessary for the observation of particular characteristics
in several pathogenic bacteria, notably the tubercle bacillus.
Pleuritic and other effusions may be prepared in the same way,
and used as media, but care must be taken in their use, as we
have no right to say that pathological effusions have the same
chemical composition as normal serum.

If blood be collected with strict aseptic precautions, then

BLOOD SERUM MEDIA

sterilisation of the serum is unnecessary. To this end the mouth

of the cylinder used for collecting the blood, instead of being

plugged with wool, has an indiarubber bung inserted in it

through which two bent

glass tubes pass. The

outer end of one of these

is of convenient length,

and, before sterilisation, a

large cap of cotton wool

is tied over it ; the other

tube is plugged with a

piece of cotton wool. In

the slaughter - house the

cap is removed and the

tube is inserted into the

blood-vessel as a cannula.

The cylinder is thus easily

filled. Another method is

to conduct the blood to

the cylinder by means

of a sterilised cannula

and indiarubber tube, the

former being inserted in

the blood-vessel. In every

case the serum must be

incubated before use, to

make sure that it is

sterile.

Coagulated Blood
Serum. — If fresh serum
be placed in sterile tubes
and be steamed in the
sloped position for an
hour, it coagulates, and

FIG. 8. — Blood serum inspissator.

there is thus obtained a solid medium very useful for the growth
of the diphtheria bacillus for diagnostic purposes.

Loffler's Blood Serum. — This is the best medium for the
growth of tin- b. diphtheria?, and may be used for other organisms.
It has the following composition : Three parts of calf's or lamb's
blood serum are mixed with one part ordinary neutral peptone
bouillon nijulo from veal with 1 }>er cent, of grape sugar added
to it. Though this is the original formula, it can be made from
ox or -Invp scrum and beef bouillon without its qualities being
markedly impaired. Sterilise by method B (4) as above (p. 30).

42 METHODS OF CULTIVATION OF BACTEEIA

Alkaline Blood Serum (Lorrain Smith's Method). — To each
100 c.c. of the serum obtained as before, add 1 to 1*5 c.c. of a
10 per cent, solution of sodium hydrate and shake gently. Put
sufficient of the mixture into each of a series of test-tubes, and,
laying them on their sides, sterilise by method B (2). If the
process of sterilisation be carried out too quickly, bubbles of gas
are apt to form before the serum is solid, and these interfere with
the usefulness of the medium. Dr. Lorrain Smith informs us
that this can be obviated if the serum be solidified high up in
the Koch's steriliser, in which the water is allowed only to
simmer. In this case sterilisation ought to go on for one and
a half hours. A clear solid medium (consisting practically of
alkali-albumin) is thus obtained, and he has found it of value
for the growth of the organisms for which Koch's serum is used,
and especially for the growth of the b. diphtheriae. Its great
advantage is that aseptic precautions in obtaining blood from the
animal are not necessary, and it is easily sterilised.

Marmorek's Serum Media. — There has always been a diffi-
culty in maintaining the virulence of cultures of the pyogenic
streptococci, but Marmorek has succeeded in doing so by growing
them on the following media, which are arranged in the order of
their utility : —

1. Human serum 2 parts, bouillon 1 part.

2. Pleuritic or ascitic serum 1 part, bouillon 2 parts.

3. Asses' or mules' serum 2 parts, bouillon 1 part.

4. Horse serum 2 parts, bouillon 1 part.

Human serum can be obtained from the blood shed in
venesection, the usual aseptic precautions being taken. In the
case of these media, sterilisation is effected by method B (4), and
they are used fluid.

Serum Media for Gonococcus. — The two following media
will be found suitable. Wertheim's medium consists of one part
of sterile human serum (conveniently obtained from placental
blood) and two parts of agar. The agar is sterilised, and fluid
is allowed to cool to 40° C. ; the serum is then added, and the
mixture is allowed to solidify in the sloped position.

Gurd's medium is a 2 per cent, agar with acid reaction + 6
to phenol-phthaleine (p. 34), with defibrinated human blood added
in the proportion of about 5 drops to 5 c.c. of agar ; the blood
is added to the melted agar as in Wertheim's medium.
W. B. M. Martin recommends the substitution of sodium
phosphate ('5 per cent.) for sodium chloride in the preparation
of the agar, and uses fluid human serum sterilised at 57° C. in

BLOOD MEDIA 43

place of blood. He also finds that the same agar medium
allowed to solidify and then smeared on the surface with a drop
or two of human serum gives excellent results.

Any of these media may be used for plate cultures, the agar
being melted and cooled to 40" C. as for agar plates ; the serum
or blood is then added ; the mixture is inoculated in the usual
way and poured out in Petri dishes.

"Nasgar."— This is a serum medium introduced by Gordon for the
isolation of the meningococcus. It is prepared as follows : —

Ascitic fluid . . . . . . 15 c.c.

Distilled water . . . . . 35 c.c.

Nutrose1 1 gramme.

Put in a flask, bring to boil, constantly shaking till ebullition occurs ;
filter. Of the resultant fluid take one part and add two parts of ordinary
iM.'l'tuiie agar. Steam for half an hour and place in tubes.

Blood Media.

Blood - Smeared Agar. — This medium was introduced by
Pfeili'er tor growing the influenza bacillus, and it has been used
for the organisms which are not easily grown on the ordinary
media, e.y. the gonococcus and the pneumococcus. Human
Mood or the blood of animals maybe used. "Sloped tubes"
(vide p. 53) of agar are employed (glycerin agar is not so
suitable). Purify a finger first with 1-1000 corrosive sublimate,
dry, and then wash with absolute alcohol to remove the sub-
limate. Allow the alcohol to evaporate. Prick with a needle
sterilised by heat, and, catching a drop of blood in the loop of a
Merile platinum wire (vide p. 54), smear it on the surface of the
agar. The excess of the blood runs down and leaves a film
on the surface. Cover the tubes with indiarubber caps, and
yicubate them for one or two days at 38° C. before use, to
make certain that they are sterile. Agar poured out in a thin
layer in a Petri dish may be smeared with blood in the same
\\ay and used for cultures.

Serum Agar is prepared in a similar way by smearing the
surface of the agar with blood serum, or by adding a few drops
of serum to the tube and then allowing it to flow over the
si ir face.

Blood Agar. — For many purposes (e.g. the growth of the

whooping-cough bacillus, the bacillus of soft sore, the cultivation

of trypanosomes and Leishmaniae), the use of agar containing

detibrinated blood, especially rabbit blood, is desirable. The

1 Nutrose is an alkaline preparation of casein.

44 METHODS OF CULTIVATION OF BACTERIA

blood may be obtained in several ways, preferably by bleeding
from the carotid. For this purpose the vessel is exposed and
as long a portion as possible is cleaned. This is ligatured high
up, and a ligature is loosely applied round the lower part of
the vessel in such a way as not to constrict it, The vessel is
clamped above this ligature, and with scissors an oblique
opening is made in its side. The clamp being removed, the
stream of blood is directed by means of the ligature into the
mouth of a stout sterile flask, which ought to contain some
fragments of broken glass rod. During the bleeding the flask
should be gently agitated, and when filled should be shaken in
a bath of water just below blood-heat. We have found that
sterile blood can be obtained from the ear vein of the rabbit by
the method of bleeding to be subsequently described. The ear
is well washed with lysol, the lysol dried off with sterile wool,
absolute alcohol dropped on and allowed to evaporate, arid the
blood withdrawn. The first c.c. or so is rejected.

However the blood is obtained, after defibrination it is warmed
to 45° C., and added to agar of the same temperature in the
proportion of about one-third of blood and two-thirds of agar.
Needless to say, such media must be incubated before use to
ensure that bacteria have not gained access during preparation.

Bordet and Gengou's Medium for Bacillus of Whooping-cough.—

An extract of potato is first prepared by adding two parts of water contain-
ing 4 per cent, of glycerin to one part of potato chips ; the mixture is
then boiled and- the fluid is separated off. An agar medium is then
prepared of the following composition : potato extract, 50 c.c.; '6 per cent,
solution of sodium chloride, 150 c.c.; and agar, Sgrrns. Of this medium,
2-3 c.c. is placed in each of a series of test- tubes, and then to each there
is added, by the method described in the preceding paragraph, an equal
part of defibrinated rabbit's (or better, human) blood, obtained by aseptic
precautions. The mixture is then allowed to solidify in the sloped
position. This medium is also very suitable for the growth of the
gonococcus, meningococcus, and influenza bacillus.

Blood- Alkali- Agar (Dieudonne). — This medium, introduced
for the. culture of the cholera spirillum, for which purpose it has
been found extremely suitable, has the property of inhibiting
the growth of most of the intestinal bacteria ; for example, the
b. coli does not grow on it, or does so very slightly. A blood-
alkali solution is prepared by adding equal parts of defibrinated
ox blood and of normal caustic soda solution ; the solution may
then be sterilised in the steam steriliser. Of this solution three
parts are added to seven parts of ordinary peptone-agar rendered
neutral to litmus, and the mixture is disposed in test-tubes.

POTATOES AS CULTURE MATERIAL 45

Novy and MacNeal's Medium for Culture of Trypanosomes. — 1 25

grammes rabbit or ox flesh is treated with 1000 c.c. distilled water,
;is in making ordinary bouillon, and there are added to the meat
extract 20 ^rms. Witt<-'s peptone, 5 grms. sodium chloride, 20 grms. agar,
ami in O.C, iiuniial sodium carbonate. The medium is placed in tubes
and sterilised in the autoclave at 110° C. for thirty minutes. It is
cooled to 50° C., and there is added to the medium in each tube twice
its volume of defibrinated rabbit blood, which has been prepared with
all asrptic precautions ; the tubes are allowed to set in the inclined
position. In inoculating such tubes they are placed in an upright
position for a few minutes, and then the infective material is introduced.

1'otatoes as Culture Material.

('/) In Potato Jars. — The jar consists of a round, shallow,

glass vessel with a similar cover (vide Fig. 9). It is washed

with 1-1000 corrosive sublimate,

and a piece of circular filter paper,

moistened with the same, is laid

in its bottom. On this latter are

placed four sterile watch-glasses.

Two firm, healthy, small, round

potatoes as free from eyes as

possible, and with the skin whole,

are scrubbed well with a brush

under the tap arid steeped for two FIG. 9.— Potato jar.

or three hours in 1-1000 corrosive

sublimate.' They are steamed in the Koch's steriliser for thirty

minutes or longer, or in the autoclave for a quarter of an hour.

When cold, each is grasped between the left thumb and forefinger

(which have been sterilised with sublimate) and cut through the

middle with a sterile knife. It is best to have the cover of the

jar raised by an assistant, and to perform the cutting beneath it.

Each half is put in one of the watch-glasses, the cut surfaces,
which are then ready for inoculation
with a bacterial growth, being upper-
most. Smaller jars, each of which
holds half of a potato, are also used
in the same way and are very con-

FIG. 10. -Cylinder of potato «

cut obliquely. (6) By Slices in Tubes. — This

method, introduced by Ehrlich, is the

best means of utilising potatoes as a medium. A large, long
potato is well washed and scrubbed, and peeled with a clean
knife. A cylinder is then bored from its interior with an apple
corer or a large cork borer, and is cut obliquely, as in Fig 10.

46 METHODS OF CULTIVATION OF BACTERIA

Two wedges are thus obtained, each of which is placed broad
end downward in a test-tube of special form (see Fig. 11). In
the wide part at the bottom of this tube is placed a piece of
cotton wool, which catches any condensation water
which may form. The wedge rests on the con-
striction above this bulbous portion. The tubes,
washed, dried, and with cotton wool in the bottom
and in the mouth, are sterilised before the slices of
potato are introduced. After the latter are in-
serted, the tubes are sterilised in the Koch steam
steriliser for one hour, or in the autoclave for
fifteen minutes, at 115° C. An ordinary test-tube
may be used with a piece of sterile absorbent wool
in its bottom, on which the potato may rest.

Glycerin potato, suitable for the growth of the
tubercle bacillus, may be prepared by covering the
slices in the tubes with 6 per cent, solution of
glycerin in water, and steaming for half an hour.
The fluid is then poured off and the sterilisation
continued for another half-hour.
FIG. 11.— Potatoes ought not to be prepared long before

Ehrlich's being used, as the surface is apt to become dry
taini'ng piece an(^ discoloured. It is well to take the reaction of
of potato. the potato with litmus before sterilisation, as this
varies ; normally in young potatoes it is weakly acid.
The reaction of the potato may be more accurately estimated by
steaming the potato slices for a quarter of an hour in a known
quantity of distilled water, and then estimating the reaction of
the water by phenol-phthaleine. The required degree of acidity
or alkalinity is obtained by adding the necessary quantity of
HC1 or NaOH solution (p. 35), and steaming for other fifteen
minutes. The water is then poured off and sterilisation con-
tinued for another half-hour. Potatoes before being inoculated
ought always to be incubated at 37° C. for a night, to make
sure that their sterilisation has been successful.

Milk as a Culture Medium.

This is a convenient medium for observing the effects of
bacterial growth in changing the reaction, in coagulating the
soluble albumin, and in fermenting the lactose. It is prepared
as follows : Fresh milk is taken, preferably after having had the
cream " separated " by centrifugalisation, as is practised in the
best dairies, and is steamed for fifteen minutes in the Koch ; it

MEDIA FOR SEPARATING BACTERIAL GROUPS 47

is then set aside in an ice chest or cool place over night to
facilitate further separation of cream. The milk is siphoned off
from beneath the cream. The reaction of fresh milk is alkaline.
If great accuracy is necessary, any required degree of reaction
may be obtained by the titration methoc|. It is then placed in
tubes, and sterilised by methods B (2) or B (3).

Bread Paste.

This is useful for growing torulse, moulds, etc. Some
ordinary bread is cut into slices, and then dried in an oven till
it is so dry that it can be pounded to a fine powder in a mortar,
or rubbed down with the fingers and passed through a sieve.
Some 100 c.c. flasks are washed, dried, and sterilised, and a
layer of the powder half an inch thick placed on the bottom.
Distilled water, sufficient to cover the whole of it, is then run in
with a pipette held close to the surface of the bread, and, the
cotton-wool plugs being replaced, the flasks are sterilised in the
Koch's steriliser by method B (2). The reaction is slightly
acid.

Jfedia used for separating the Members of Jlacterial Groups.

A great number of media have been devised for use in
differentiating the members of the coli-typhoid and other
bacterial groups. The general feature of these media is that
they contain certain substances, often sugars, which tend to
bring out the special characters of the organism under investiga-
tion. Sometimes also substances are present which inhibit the
growth of bacteria other than those belonging to the group.
The following are the media which here deserve most attention : —

Hiss's Serum Water Media.— These are composed of one part of ox's
serum and three parts of distilled water with 1 per cent, litmus ; various
sugars in a pure condition are added in the proportion of 1 per cent.
The development of acid by fermentation is shown by the alteration of
the colour and by coagulation of the medium. These media do not
coagulate at 100° C., and thus can be sterilised in the steam steriliser.
They have been extensively used by American workers in studying the
fermentative properties of the b. dysenteriae, b. coli, etc.

Drigalski and Conradi's Medium. — This is one of the media used for
tin- study of intestinal bacteria, and especially for the isolation of the
typhoid group of organisms, (a) Three pounds of meat are treated with
two litres of water overnight ; the fluid is separated as usual, boiled for
an hour, filtered, and there are added 20 grammes Witte's peptone, 20
grammes nutrose, 10 grammes sodium chloride ; the mixture is then
boiled for an hour, 60 grammes finest agar are added, and it is placed in

48 METHODS OF CULTIVATION OF BACTERIA

the autoclave till melted (usually one hour) ; it is then rendered slightly
alkaline to litmus, filtered, and boiled for half an hour, (b) 260 c.c.
Kubel-Tiemann litmus1 solution is boiled for ten minutes, 30 grammes
milk sugar (chemically pure) are added, and the mixture is boiled for
fifteen minutes ; (a) and (b) are then mixed hot, well shaken, and, if
necessary, the slightly alkaline reaction restored. There are then added
4 c.c. of a 10 per cent, sterile solution of water- free sodium carbonate and
20 c.c. of a freshly prepared solution made by dissolving °1 gramme
crystal-violet B, Hoechst, in 100 c.c. hot sterile distilled water. This
is the finished medium, and great care must be taken not to overheat it
or to heat it too long, as changes in the lactose may be originated. It
is convenient to distribute the medium in 80 c.c. flasks.

The principle of the medium is that while there is a food supply very,
favourable to the b. typhosus and the b. coli, the antiseptic action of the
crystal- violet tends to inhibit the growth of other bacteria likely to
occur in material which has been subjected to intestinal contamination.
In examining fteces, a little is rubbed up in from ten to twenty times its
volume of sterile bouillon ; in the case of urine or water, the fluid is
centrifugalised and the deposit or lower portion is used for the inocula-
tion procedures.

For use the medium is distributed in Petri capsules in a rather thicker
layer than is customary in an ordinary plate. This sheet of medium
must be transparent, but must not be less than 2 mm. in thickness — in
fact, ought to be about 4 mm. After being poured, the capsules are left
with the covers off for an hour or so, to allow the superficial layers of
the medium to become set hard. The effect of this is that during in-
cubation no water of condensation forms on the lid of the capsule, and
thus the danger of this fluid dropping on to the developing colonies is
avoided. The antiseptic nature of the crystal-violet is sufficient to
prevent the growth of any aerial organisms falling on the agar during
its exposure to the air. The plates are usually inoculated by means of
a glass spatula made by bending 3 inches of a piece of glass rod at
right angles to the rest of the rod. This part is dipped in the infective
material, and smeared in all directions over the surfaces of three or four
plates successively without any intervening sterilisation. The plates are
again exposed to the air after inoculation for half an hour, and then
incubated for twenty-four hours. At the end of such a period b. coli
colonies are 2 to 6 mm. in diameter, stained distinctly red, and are non-
transparent. Colonies of the b. typhosus are seldom larger than 2 mm.,
they are blue or bluish-violet in colour, are glassy and dew-like in
character, and have a single contour. Sometimes in the plates b. subtilis
and its congeners appear, and colonies of these organisms have a blue
colour. Their growth is, however, more exuberant than that of the
typhoid bacillus, — being often heaped up in the centre, — and the
contour of the colony is often double.

J The litmus solution is made as follows : Solid commercial litmus is
digested with pure spirit at 30° C. till on adding fresh alcohol the latter
becomes only of a light violet colour. A saturated solution of the residue is
then made in distilled water aud filtered. When this is diluted with a little
distilled water it is of a violet colour, which further dilution turns to a pure
blue. To such a blue solution very weak sulphuric acid (made by adding
two drops of dilute sulphuric acid to 200 c.c. water) is added till the blue
colour is turned to a wine-red. Then the saturated solution of the dye is
added till the blue colour returns.

MEDIA FOR SEPARATING BACTERIAL GROUPS 49

Conradi's Picric Acid Brilliant Green Method.— Applying his principle
of seeking for anilin bodies which while inhibiting the action of ordinary
intestinal bacteria rather favour the growth of b. typhosus and b. para-
typhosus, Conrad i in 1908 used for this purpose crystalline brilliant
green (Hoechst, extra pure), acting along with picric acid (Griibler). The
medium is made as follows : 900 c.c. water, 20 grins. Liebig's meat extract
and 100 c.c. of a 10 per cent, watery solution of Witte's peptone are
mixed and filtered ; 30 grms. agar in threads are dissolved in the Huid, and
the whole filtered. The reaction is then adjusted with normal sodium
hydrate or normal phosphoric acid to an acid content of 3 per cent,
(phenol-phthaleine being the indicator), i.e. the finished medium is
such that to make it neutral would require the addition to each 100
c.c. of 3 c.c. of normal sodium hydrate. The acid medium is then
sterilised, and kept in bulk in this form. For use the remaining
substances are added in the proportions of 10 c.c. of 1-1000 watery
solution of the brilliant green and 10 c.c. of 1 per cent, watery picric
acid to 1^ litres of the peptone-agar, and the finished medium is poured
in large Petris and allowed to stand at 37° C. till the surface is firm.
The capsules are inoculated in the wajr already described. Typhoid
colonies appear sharp-edged, round, flat-surfaced but slightly thicker
in the middle, transparent, and of light green colour. Colonies of the
paratyphoid bacillus are similar, but tend at the same age to be slightly
larger and have a somewhat yellowish green tint.

Fawcus's Picric Acid and Brilliant Green Medium.— This is a

modification of Conradi's medium which has been used with great
success at the Royal Army Medical College in the investigation of
typhoid carriers. It is made as follows : To 900 c.c. tap water add
5 grms. sodium taurocholate (which is commercially prepared from ox
bile), 30 grms. powdered agar, 30 grms. Witte's peptone, 5 grms. sodium
chloride ; steam for three hours, clear with white of egg, filter through
cotton wool, and bring to a reaction of + 15 with normal lactic acid or
caustic soda, and sterilise. Dissolve 10 grms. lactose in 100 c.c. sterile
distilled water, and add to melted agar. Mix and filter through Chardin
paper, sterilise carefully, and store in 100 c.c. flasks. For use, add to
each 100 c.c. flask 2 c.c. of a 1-1000 watery solution of brilliant green
and '2 c.c. of a 1 per cent, watery solution of picric acid. Pour into
large Petri dishes, and leave these to stand inverted at 37° C. till the
surface hardens. Inoculate as usual. Colonies of b. typhosus of twenty-
four hours' growth are of about 1 mm. in diameter, transparent and re-
fracting - those of b. coli, on the other hand, have a deep green centre,
though later typhoid colonies may also present a pale green centre.

In the case of several of the special media used for the isolation of
typhoid bacilli under circumstances where other bacteria are present, a
difficulty arises from the fact that the agglutinability of the strains
isolated appears to be affected by substances present in the media.
The application of this important confirmatory diagnostic method is thus
interfered with. This is said not to occur with the Conradi brilliant
given method, and we have found with this medium that, if typhoid
colonies do not at once clump with typhoid serum, daily sub-culture on
ordinary agar yields in a few days a culture to which the agglutination
test can be applied.

Endo's Medium. — This is another of the modern media introduced for
facilitating the separation of the b. typhosus from stools, etc. It is
made as follows: A litre of 3 per cent, agar is prepared with the usual

50 METHODS OF CULTIVATION OF BACTERIA

constituents, and is boiled, filtered, and rendered neutral. It is then
made alkaline by the addition of 10 c.c. of a 10 per cent, solution of
sodium hydrate, and there are added 10 grammes of chemically pure milk
sugar (free from cane sugar) and 5 c.c. of a filtered saturated alcoholic
solution of basic fuchsin. After thorough mixing there is added 25 c.c.
of a freshly prepared 10 per cent, solution of sodium sulphite, the effect
of this step being to remove the colour of the fuchsin so that the finished
medium when cool is quite colourless. Of the medium 15 c.c. are placed
in each of a number of tubes, these are steamed for fifteen minutes and
must then be kept in the dark. For use the contents of a tube are
poured into a sterile Petri capsule, allowed to set in a still, dustless
atmosphere, and are then inoculated as in the other methods described.
After twenty-four hours' growth colonies of b. coli appear red, while those
of b. typhosus are colourless. Endo also claims for his medium that
typhoid bacilli isolated by its means are agglutinable by a typhoid
serum. The rationale of the colour reaction appears to be that fuchsin,
which is rosanilin hydrochlorate (C20H19"N3HC1), is reduced to rosanilin
(a colourless substance) by the sodium sulphite. This colourless base
produces a red colour with acids, such as the lactic acid formed by the
b. coli in its fermentation of lactose.

MacConkey's Bile-Salt Media.— These media were introduced for the
purpose of differentiating the intestinal bacteria, and have been exten-
sively used for the study of the b. coli, b. typhosus, b. dysenteric, etc.
The characteristic ingredients are bile salts and various sugars. The
stock solution is the following: Commercial sodium taurocholate, '5
gramme ; Witte's peptone, 2'0 grammes; tap water, 100 c.c. (if distilled
water be used, '03 per cent, of calcium chloride should be added). The
solution is steamed for two hours, filtered when hot, allowed to stand for
twenty-four hours or till sedimentation has occurred, and filtered again. For
a liquid medium there is added to this "25 per cent, of a freshly prepared
1 per cent, solution of neutral red J and the sugar, — when glucose, dulcite,
or adonite is used, '5 per cent, is added, in the case of other sugars
1 per cent. The fluid is distributed in Durham's fermentation tubes and
sterilised in the steamer for ten minutes on two successive days, care
being taken not to overheat the medium.

For bile-salt agar 1*5 to 2 per cent, agar is dissolved in the stock
solution in the autoclave, if necessary cleared with white of egg and
filtered. Neutral red and a sugar are added, as in the case of the liquid
medium. As with Drigalski's medium, it is well to sterilise it in flasks
containing 80 c.c., this being an amount sufficient for three Petri
capsules. When this medium is used for examining urine or faeces,
plates are inoculated as with Drigalski's medium (supra} ; for its use in
water examinations, see p. 157.

With reference to their behaviour in MacConkey's fluid medium with
glucose, organisms are divided into (1) those which produce both acid and
gas ; (2) those producing acid only ; (3) those growing but not producing
either acid or gas ; (4) those incapable of growing. B. coli belongs to the
first group and b. typhosus to the second, and to these groups also belong
most ordinary organisms growing in faeces, practically none of which are
found in the third and fourth classes. Further, a number of ordinary non-

1 The use of neutral red in a bile-salt medium was first introduced by
Griinbaum and Hume. It gives a deep crimson with acids and a yellow-brown
with alkalies.

MEDIA FOR SEPARATING BACTERIAL GROUPS 51

pathogenic organisms and also some that are pathogenic have their free
growth inhibited in bile-salt media. Thus, if any growth takes place
on this medium when inoculated with, say, water, the probability is that
the bacteria have been derived from faices, but of course further procedures
for their identification must be undertaken.

When growth of a bacterium producing acid and gas occurs in neutral-
red fluid media the latter turns a rose colour, and gas appears in the
Durham's tube. Sometimes a fluorescent appearance is also observed,
the significance of which will be discussed in the chapter on b. coli.
With the neutral-red solid media the colonies of any organism giving
rise to acid will be of a rose-red colour.

Petruschky's Litmus Whey. — The preparation of this medium, which
is somewhat difficult, is as follows : Fresh milk is slightly warmed,
and sufficient very dilute hydrochloric acid is added to cause precipita-
tion of the casein, which is now filtered off. Dilute sodium carbonate
solution is added up to, but not beyond, the point of neutralisation, and
the fluid steamed for one to two hours, by which procedure any casein
which has been converted into acid albumin by the hydrochloric acid
is precipitated. This is filtered off, and a clear, colourless, perfectly
neutral fluid should result. Its chief constituent, of course, will be
lactose. To this sufficient Kubel-Tiemann solution of litmus is added,
the medium is put into tubes and then sterilised. (This is the original
method, but it is better, after the casein has been precipitated, to make
the medium slightly alkaline with the sodium carbonate and bring to
the boiling-point; then filter, neutralise, add the litmus, and sterilise.)
After growth has taken place, the amount of acid formed can be estimated
liy dropping in standardised soda solution till the tint of an uninoculated
tube is readied.

Eisner's Medium. — This is another of the media introduced in the
study of the comparative reactions of the typhoid bacillus and the b. coli.
The preparation is as follows : 500 grammes potato arc grated up ill a
litre of water, allowed to stand over night, then strained, and added to
an equal quantity of ordinary 15 per cent, peptone gelatin which has not
been neutralised. Normal sodium hydrate solution is added till the
reaction is feebly acid to litmus, the whole boiled together, filtered, and
sterilised. Just before use potassium iodide is added so as to constitute
1 per cent, of the medium. Moore has used a similar agar preparation.
Here 500 grammes potato are scraped up in one litre of water, allowed to
stand for three hours, strained, and put aside over night. The clear
fluid is poured oil', made up to one litre, rendered slightly alkaline, 20
grammes agar are added, and the whole is treated as in making ordinary
agar. The medium is distributed in test-tubes — 10 c.c. to each — and
immediately before use, to each is added *5 c.c. of a solution of 10
gramnif> j.ntassium iodide to 50 c.c. water.

Any one of these media in the hands of a worker accustomed
to its use will yield good results. MacConkey's medium is that
most used by British workers, and it has the merit of being
i-;i>ily prepared. As the result of a considerable experience
we have found it most useful and reliable. Next to it \\«
would place r'a\vi-us's modification of Conradi's brilliant green
method.

52 METHODS OF CULTIVATION OF BACTERIA

Media for growing Trichophyta, Moulds, etc.

1. Beer Wort Agar.— Take beer wort as obtainable from the brewery,
and dilute it till it has an s.g. of 1100. Add 1'5 per cent, of powdered
agar, and heat in the Koch till it is dissolved (usually about two hours
are necessary). Filter rapidly and fill into tubes. Sterilise in the Koch
for twenty minutes on three successive days. If the medium is heated too
long it loses the capacity of solidifying.

2. Sabouraud's Media. — Sabouraud recommends the following
media : —

(1) Distilled water 1000 c.c.

Maltose ("brute de Chanut") . . . 40 grms.

Peptone ("granulee de Chassaing") . . 10 ,,

Agar . . 18 ,,

(2) Distilled water 1000 c.c.

Glucose ("massee de Chanut") . . . 40 grms.
Peptone ("granule de Chassaing") . . 10 ,,
Agar . . . . . . ' . . 18 ,,

In each case the ingredients are mixed and dissolved by gradually
raising to 120° C. in an autoclave. The medium is then rapidly filtered
through papier chardin (Cogit, 36 Boulevard Saint Michel, Paris) ;
when the filtrate begins only to pass in drops the fluid is transferred to
another filter, and this is repeated as often as is necessary. The medium
is distributed in wide test-tubes or Erlerimeyer's flasks, plugged with non-
absorbent cotton wool, and sterilised by slowly raising the temperature
to 120° C.

To use these for isolating, say, the Tinea tonsurans, pick out an
infected hair, wash in absolute alcohol for a few seconds, then wash in
changes of sterile waterj; cut into small pieces and place these on the
surface of the medium ; incubate at 24° C. Usually, however, it is un-
necessary to disinfect .hair or skin scales from which dermophyta are to
be isolated.

THE USE OF THE ORDINARY CULTURE MEDIA.

The culture of bacteria is usually carried on in test-tubes
conveniently 6 x f in. These ought to be very thoroughly
washed and dripped, and their mouths plugged with plain
cotton wool. They are then sterilised for one hour at 170° C.
If the tubes be new, the glass, being usually packed in straw,
may be contaminated with the extremely resisting spores of
the b. subtilis. Cotton-wool plugs are universally used for
protecting the sterile contents of flasks and tubes from con-
tamination with the bacteria of the air. A medium thus
protected will remain sterile for years. Whenever a protecting
plug is removed for even a short time, the sterility of the
contents may be endangered. It is well to place the bouillon,
gelatin, and agar media in the test-tubes directly after filtration.
The media can then be sterilised in the test-tubes.

THE USE OF THE ORDINARY CULTURE MEDIA 53

In filling tubes, care must be taken to run the liquid down
the centre, so that none of it drops on the inside of the upper
part of the tube with which the cotton-wool plug will be in
contact, otherwise the. latter will subsequently stick to
the glass and its removal will be difficult. In the case of
liquid media, test-tubes are filled about one-third full. With
the solid media the amount varies. In the case of gelatin

A e

FIG. 13.— Tubes of media.

The apparatus explains itself.
The indiarubber stopper with
its tubes ought to be steril-
ised before use.

media, tubes filled one-third full and allowed to solidify
while standing upright, are those commonly used. With
organisms needing an abundant supply of oxygen the best
growth takes place on the surface of the medium, and for
practical purposes the surface ought thus to be as large as
possible. To this <end " sloped " agar and gelatin tubes are
used. To prepare these, tubes are filled only about one-sixth
full, and after sterilisation are allowed to solidify, lying on their
sides with their necks supported so that the contents extend

54 METHODS OF CULTIVATION OF BACTERIA

3 to 4 inches up, giving an oblique surface after solidification.
Thus agar is commonly used in such tubes (less frequently
gelatin is also " sloped "), and this is the position in which blood
serum is inspissated. Tubes, especially those of the less commonly
used media, should be placed in large jars provided with stoppers,
otherwise the contents are apt to evaporate. A tube of medium
which has been inoculated with a bacterium, and on which
growth has taken place, is called a " culture." A " pure culture "
is one in which only one organism is present. The methods of
obtaining pure cultures will presently be described. When a
fresh tube of medium is inoculated from an already existing
culture, the resulting growth is said to be a " sub-culture " of the
first. All manipulations involving the transference of small
portions of growth either from one medium to another, as in the

FIG. 14. — Platinum wires in glass handles.

a. Straight needle for ordinary puncture inoculations, b. "Platinum loop."
c. Long needle for inoculating " deep " tubes.

inoculation of tubes, or, as will be seen later, to cover-glasses for
microscopic examination, are effected by pieces of platinum wire
(Nos. 24 or 27 Birmingham wire gauge — the former being the
thicker) fixed in glass rods 8 inches long.1 Every worker should
have three such wires. Two are 2J inches long, one of these
being straight (Fig. 14, a), and the other having a loop turned
upon it (Fig. 14, b). The latter is referred to as the platinum
"loop" or platinum "eyelet," and is used for many purposes.
" Taking a loopful " is a phrase constantly used. The third wire
(Fig. 14, c) ought to be 4^ inches long and straight. It is used
for making anaerobic cultures. It is also very useful to have
at hand a platinum-iridium spud. This consists of a piece of
platinum-iridium about 1J inches long, 2 mm. broad, and of
sufficient thickness to give it a firm consistence ; its distal end is

1 Aluminium rods are made which are very convenient. The end is split
with a knife, the platinum wire is inserted and fixed by pinching the
aluminium on it in a vice,

THE USE OF THE ORDINARY CULTURE MEDIA 55

expanded into a diamond shape, and its proximal is screwed
into an aluminium rod. It is very useful for making scrapings
from organs and for disintegrating felted bacterial cultures ; in
such manipulations the ordinary platinum wire is awkward to
work with, as it bends so easily. Cultures on a solid medium
are referred to (1) as "puncture" or "stab" cultures (German,
Stichkultur), or (2) as "stroke" or " slant " cultures (Strichkultur),
according as they are made (1) on tubes solidified in the upright
position, or (2) on sloped tubes.

To inoculate,1 say, one ordinary upright gelatin tube from
another, the two tubes are held in an inverted position between
the forefinger and thumb of the left hand with their mouths
towards the j>erson holding them ; the plugs are twisted round
once or twice, to make sure they are not adhering to the glass.
The short, straight platinum wire is then heated to redness from
point to insertion, and 2 to 3 inches of the glass rod are also
passed two or three times through the Bunsen flame. It is held
between the right fore and middle fingers, with -the needle pro-
jecting backwards, i.e. away from the right palm. Remove plug
from culture tube with right forefinger and thumb, and continue
to hold it between the same fingers by the part which projected
beyond the mouth of the tube. Now touch the culture with the
platinum needle, and, withdrawing it, replace plug. In the
same way remove plug
from tube to be in-
oculated, and plunge
platinum wire down
the centre of the
gelatin to within half
an inch of the bottom.
It must on no account
touch the glass above
the medium. The wire
is then immediately
sterilised. A variation
in detail of this method
is to hold the plug of
the tube next the
thumb between the fore
and middle fingers, and

the plug of the other between the middle and ring fingers, then
to make the inoculation (Fig. 15). If a tube contain a liquid
medium, it must be held in a sloping position between the same
fingers, as above. For a stroke culture the platinum loop is

FIG. 15. — Another method of inoculating
solid tubes.

56 METHODS OF CULTIVATION OF BACTERIA

used, and a little of the culture is smeared in a line along the
surface of the medium from below upwards. In inoculating
tubes, it is always well, on removing the plugs, to make sure

that no strands of cotton
fibre are adhering to the
inside of the necks. As
these might be touched with
the charged needle and the
plug thus be contaminated,
they must be removed by
heating the inoculating
needle red-hot and scorch-
FIG. 16.— Rack for platinum needles. ing them off with it. When

the platinum wires are not

in use they may be laid in a rack made by bending up the ends
of a piece of tin, as in Fig. 16. To prevent contamination of
cultures by bacteria falling on the plugs while these are exposed
to the air during inoculation manipulations, some bacteriologists
singe the plugs in the flame before replacing. This is, however,
in most cases a needless precaution. If the top of a plug be
dusty it is best to singe it before extraction.

THE METHODS OF THE SEPARATION OF AEROBIC ORGANISMS.
PLATE CULTURES.

The general principle underlying the methods of separation
is the distribution of the bacteria in one of the solid media
liquefied by heat and the dilution of the mixture so that the
growths produced by the individual bacteria — called colonies —
shall be suitably apart. In order to render the colonies easily
accessible, the medium is made to solidify in as thin a layer as
possible, by being poured out on glass plates — hence the term
"plate cultures."

As the optimum temperature varies with different bacteria,
it is necessary to use both gelatin and agar media. Many
pathogenic organisms, e.g. pneumococcus, b. diphtheriae, etc.,
grow too slowly on gelatin to allow its ready use. On the other
hand, many organisms, e.g. some occurring in water, do not
develop on agar incubated at 37° C.

Separation by Gelatin Media. — As the naked-eye and micro-
scopic appearances of colonies are often very characteristic,
plate cultures, besides use in separation, are often taken
advantage of in the description of individual organisms. The
plate-culture method can also be used to test whether a tube

SEPARATION BY GELATIN MEDIA 57

culture is or is not pure. The suspected culture is plated (three
plates being prepared, as will be described). If all the colonies
are the same, then the culture may be held to be pure.

Either simple plates of glass 4 inches by 3 inches are used,
or, what are more convenient,
circular glass cells with similar
overlapping covers. The latter
are known as Petri's dishes or
capsules (Fig. 17). They are
usually 3 inches in diameter and
half an inch deep. The advant-
age of these is that they do not FIG. 17. -^tri's capsule,
require to be kept level by a (Cover shown partially raised.)
.special apparatus while the medium

is solidifying, and can be readily handled afterwards without
admitting impurities. Whether plates or capsules are used,
they are washed, dried with a clean cloth, and sterilised for one
hour in dry air at 170° C., the plates being packed in sheet-iron
boxes made for the purpose (see Fig. 18).

1. Glass Capsules. — While in certain circumstances, as when
the number of colonies has to be counted, it is best to use plates
of glass, Petri's capsules are to be preferred in- the usual labora-
tory routine for the above reasons.

The contents of three gelatin tubes, marked a, b, c,1 are
liquefied by placing in a beaker of water at any temperature
between 25° C. and 38° C. Inoculate a with the bacterial
mixture. The amount of the latter to be taken varies, and can
only be regulated by experience. If the microscope shows
enormous numbers of different kinds of bacteria present, just as
much as adheres to the point of a straight platinum needle is
suilicient. If the number of bacilli is small, one to three loops
of the mixture may be transferred to the medium. Shake a
wrll, but not so as to cause many fine air-bubbles to form.
Transfer two loops of gelatin from a to b. Shake b and transfer
five loops to c. The plugs of the tubes are in each case replaced
and the tubes returned to the beaker. The contents of the
three tubes are then poured out into three capsules. In doing
so the plug of each tube is removed and the mouth of the tube
passed two or three times through the Bunsen flame, the tube
being meantime rotated round a longitudinal axis. Any organ-
isms on its rim are thus killed. The capsules are labelled and
set aside till growth takes place.

1 For marking glass vessels it is convenient to use the red, blue, or yellow
oil pencils made for the purpose by Faber.

58 METHODS OF CULTIVATION OF BACTERIA

For accurate work it will be found convenient to carry out
the dilutions in definite proportions. The following is the pro-

FlG. 18. — Koch's levelling apparatus for use in preparing plates.
Hands shown in first position for transferring sterile plate from iron
box to beneath bell jar, where it subsequently has the medium poured
out upon it.

cedure which we .have found very serviceable : In a number of
small sterile test-tubes '95 c.c. sterile water is put. To the first

FIG. 19.- — Koch's levelling apparatus. Hands shown in second
position just as the plate is lowered on to the ground glass surface.
By executing the transference of the plate from the box in this way,
the surface which was undermost in the latter is uppermost in the
leveller, and thus never meets a current of air which might con-
taminate it.

tube we add '05 c.c. of the bacterial mixture. The contents of
the tube are well shaken up, and the pipette is sterilised by

SEPARATION BY GELATIN MEDIA 59

being washed out with boiling water. It is allowed to cool, and
•05 c.c. of fluid is transferred from the first tube to the second.
By a similar procedure '05 c.c. is transferred from the second to
the third, and so on. There is thus effected a twenty-fold
dilution in each successive tube. After these steps have been
carried out, a definite amount, say, '05 c.c., is transferred from
each tube to a tube of melted gelatin, — the gelatin being after-
wards plated and the colonies counted when growth occurs.
The number of tubes required will vary according to the
number of bacteria in the original mixture, but usually four or
five will be sufficient. It is quite evident that this method not
only enables us to separate bacteria, but if necessary gives us a
means of estimating exactly the number in the original mixture.
The colonies appear as minute rounded points, whitish or
variously coloured. Their characters can be more minutely
studied by means of a hand-lens or by inverting the capsule on
the stage of a microscope and examining with a low power
through the bottom. From their characters, colour, shape,
contour, appearance of surface, liquefaction or non-liquefaction
of the gelatin, etc., the colonies can be classified into groups.
Further aid in the grouping of the varieties is obtained by
making film preparations and examining them microscopically.
Gelatin or agar tubes may then be inoculated from a colony of
each variety, and the growths obtained are then examined both
as to their purity and as to their special characters, with a view
to their indentification (p. 137).

2. Glass Plates (Koch). — When plates of glass are to be used, an
apparatus on which they may be kept level while the medium is solidi-
fying is, as has been said, necessary. An apparatus devised by Koch is
used (Figs. 18, 19). This consists of a circular plate of glass "(with the
upper surface ground, the lower polished), on which the plate used for
pouring out the medium is placed. The latter is protected from the air
daring solidification by a bell jar. The circular plate and bell jar rest
<>ii the tlat rim of a circular glass trough, which is filled quite full with
a mixture of ice and water, to facilitate the lowering of the temperature
of whatever is placed beneath the bell jar. The glass trough rests on
corks on the bottom of a large circular trough, which catches any water
that may be spilled. This trough in turn rests on a wooden triangle
with a foot at each corner, the height of which can be adjusted, and
which thus constitutes the levelling apparatus. A spirit level is placed
where the plate is to go, and the level of the ground glass plate thus
assured. There is also prepared a "damp chamber," in which the
plates are to be stored after being made. This consists of a circular
glass trough with a similar cover. It is sterilised by being washed out-
side' and inside with perchloride of mercury 1-1000, and a circle of filter-
papcr nioi.xtened with the same is laid on its bottom. Glass benches on
which the plates may be laid are similarly purified.

60 METHODS OF CULTIVATION OF BACTERIA

To separate organisms by this method, three tubes, a, b, c, are inocu-
lated as in using Petri's capsules (p. 57). The hands having been
washed in perchloride of mercury 1-1000 and dried, the plate box is
opened, and a plate lifted by its opposite edges and transferred to the
levelled ground glass (as in Figs. 18, 19). The bell jar of the leveller
being now lifted a little, the gelatin in tube a is poured out on the
surface of the sterile plate, and, while still fluid, is spread by stroking
with the rim of the tube. After the medium solidifies, the plate is
transferred to the moist chamber as rapidly as possible, so as to avoid
atmospheric contamination. In doing this, it is advisable to have an
assistant to raise the glass covers. Tubes b and c are similarly treated,
and the resulting plates stacked in series on the top of a. The chamber
is labelled and set aside for a few days till the colonies appear on the
gelatin plates. The further procedure is of the same nature as with
Petri's capsules.

3. Esmarctis Roll Tubes. — Here the principle is that of
dilution as before. In each of three test-tubes 1J or 1J inch in
diameter, gelatin to the depth of three-quarters of an inch is placed.
These are sterilised. The gelatin is melted and
inoculated in series with the bacterial mixture as
in making plate cultures, but instead of being
poured out it is rolled in a nearly horizontal
position under a cold tap or on a block of ice
till it solidifies as a uniformly thin layer on the
inside of the tube. Practically we deal with a
cylindrical sheet of gelatin instead of a flat one.
A convenient form of tube for this method is
one with a constriction a short distance below
'the plug of cotton wool (Fig. 20). The great
disadvantage of the method is, that if organisms
liquefying the gelatin be present, the liquefied
gelatin contaminates the rest of the medium.

Separation by Agar Media. — 1. Agar Plates.
— The only difference between the technique here
and that with gelatin depends on the difference
in the melting-points of the two media. Agar,
we have said, melts at 98° C., and becomes
E FlG'h?tube a§am s°lid a little under 40° C. As it is danger-
for roll culture, ous to expose organisms to a temperature much
above 42° C., it is necessary in preparing tubes
of agar to be used in plate cultures first to melt the agar, by
boiling in a vessel of water for a few minutes, and then to
cool it to about 42° C. before inoculating. The manipulation
must be rapidly carried out, as the margin of time, before
solidification occurs, is narrow; otherwise the details are the
same as for gelatin. Esmarch's tubes are not suitable for use

SEPARATION BY AGAR MEDIA 61

here, as the agar does not adhere well to the sides. If to the
agar 2 per cent, of a strong watery solution of pure gum arabic
is added, Esmarch's tubes may, however, be used.

2. Scjmration I>y Stroking Mixture on Surface of Agar
Mulia. — The bacterial mixture, instead of being mixed in the
medium, i* spread out on its surface. The method may be used
both when the bacteria to be separated are in a fluid, and when
contained in a fairly solid tissue or substance, such as a piece
of diphtheritic membrane. In the case of a tissue, for example,
a small portion entangled in the loop of a platinum needle is
stroked in successive parallel longitudinal strokes on sloped
agar, the same asj)ect being brought in contact with the agar in
all the strokes. Three strokes may be made in each tube, and
three tubes are usually sufficient. In this process the organisms
on the surface of the tissue are gradually rubbed off, and when
growth has taken place it will be found that in the later strokes
the colonies are less numerous than in the earlier, and sufficiently
far apart to enable parts of them to be picked off without the
needle touching any but one colony. When, as in the case of
diphtheritic membrane, putrefactive organisms may be present
on the surface of the tissue, these can be in great part removed
by washing it well in cold water previously sterilised (vide
Diphtheria). In the case of liquids, the loop is charged and
similarly stroked. Tubes thus inoculated must be put in the
incubator in the upright position and must be handled carefully,
so that the condensation water, which is always present in
incubated agar tubes, may not run over the surface. Agar,
poured out in a Petri's capsule and allowed to stand till firm,
may be used instead of successive tubes. Here a sufficient
number of strokes can be made in one capsule. Sloped blood-
serum tubes may IK- used instead of agar. The method is rapid
and easy, and gives good results.

Separation of Pathogenic Bacteria by Inoculation of
Animals. — It is found difficult and often impossible to separate
l»y ordinary plate methods certain pathogenic organisms, such
as b. tuberculosis, 1>. mallei, and the pneumococcus, when such
occur in conjunction with other bacteria. These grow best on
special media, and the first two (especially the tubercle bacillus)
grow so slowly that the other organisms present outgrow them,
cover the whole plates, and make separation impossible. The
method adopted in such cases is to inoculate an animal with
the mixture of bacilli, wait until the particular disease develops,
kill the animal, and with all aseptic precautions (vide p. 1 •!">)
inoculate tubes of suitable media from characteristic lesion-

62 METHODS OF CULTIVATION OF BACTERIA

situated away from the seat of inoculation, e.g. from spleen
in the case of b. tuberculosis, spleen or liver in the case of
b. mallei, and heart blood in the case of pneumococcus.

Separation by killing Non-spored Forms by Heat. — This is
a method which has a limited application. As has been said,
the spores of a bacterium resist heat more than the vegetative
forms. When a mixture contains spores of one bacterium and
vegetative forms of this and other bacteria, then if the mixture
be boiled for a few minutes all the vegetative forms will be
killed, while the spores will remain alive and will develop
subsequently. This method can be easily tested in the case of
cultivating b. subtilis from hay infusion. A little chopped-up
hay is placed in a flask, of water, which is boiled for about ten
minutes. On this being allowed to cool and stand, in a day or
two a scum forms on the surface, which is found to be a pure
culture of the bacillus subtilis. The method is also often used
to aid in the separation of b. tetani, vide infra.

THE PRINCIPLES OF THE CULTURE OF ANAEROBIC
ORGANISMS.

All ordinary media, after preparation, may contain traces of
free oxygen, and will absorb more from the air on standing.
(1) For the growth of anaerobes this oxygen may be expelled by
the prolonged passing of an inert gas, such as hydrogen, through
the medium (liquefied if necessary). Further, the medium must
be kept in an atmosphere of the same gas while growth is going
on. (2) Media for anaerobes may be kept in contact with the
air, if they contain a reducing agent which does not interfere
with bacterial growth. Such an agent takes up any oxygen
which may already be in the medium, and prevents further
absorption. The reducing body used is generally glucose, though
formate of sodium may be similarly employed. The preparation
of such media has already been described (pp. 37, 38). In this
case the medium ought to be of considerable thickness.

The Supply of Hydrogen for Anaerobic Cultures. — The gas is generated
in a large Kipp's apparatus from pure sulphuric acid and pure xi.nc. It
is passed through three wash-bottles, as in Fig. 21. In the first is
placed a solution of lead acetate (1 in 10 of water) to remove any traces-
of sulphuretted hydrogen. In the second is placed a 1 in 10 solution of
silver nitrate to remove any arsenietted hydrogen which may be present
if the zinc is not quite pure. In the third is a 10 per cent, solution
of pyrogallic acid in caustic potash solution (1 : 10) to remove any
traces of oxygen. The tube leading from the last bottle to the vessel
containing the medium ought to be sterilised by passing through a

SEPARATION OF ANAEROBIC ORGANISMS 63

Bunsen flame, and should have a small plug of cotton wool in it to filter
the hydrogen ^enn-l'ree.

1'i/fOf/allate of Potassium for Anaerobic Cultures. — In arranging for the
absorption of oxygen by this substance the proportions used in Bulloch's
-r] .nation method may be employed. Here 109 grans, solid caustic
potash are dissolved in 145 c.c. water, and to this 2-4 grms. pyrogallol
art- added.

Separation of Anaerobic Organisms. — (a) By Roll-tubes. —
A 1 j inch test-tube has as much gelatin put into it as would be
used in the Esmarch roll-tube method. It is corked with an
indiarubber stopper having two tubes passing through it, as in
Fig. "22. The ends of the tubes are partly drawn out as shown,

Fit;. 2L— Apparatus for supplying hydnuren for anaerobic cultures.

(t. Ki)>i>'> apparatus for manufacture of Imlro.uvn. l>. Wash-bottle con-

taininu- 1 -It) solution ot lead :n-etate. <•. Wa^h-bolt I< ulainiii^ 1-10 solution

of silver nitrate. </. Wash-bottle containing: 1-10 solution of pyro^allic acid.
(l>, c, and </ are intentionally drawn to a larger scale than n to show details.)

and covered witli plugs of cotton wool. Three such test-tubes
HIV prepared, and they are sterilised in the steam steriliser (p. 28).
At't'-r >terilisation the gelatin is melted and one tube inoculated
with tin- mixture containing the anaerobes; the second is inocu-
lated from the first, and the third from the second, as in making
ordinary LTriatin plates. After inoculation the gelatin is kept
liquid by the lower ends of the tubes being placed in water at
about •''»<.) ('., and hydrogen is passed in through tube ./• f'-r
t unity minutes. Tin- i:as-supply tubes are then completely
Sealed ot!' at X and /, and rarli test-tube is rolled as in Ksmareh's
method till the gelatin solidities as a thin layer on the internal
surface. A little hard parallin may be run between tin- rim of the

64 METHODS OF CULTIVATION OF BACTERIA

containing anaerobes.

test-tube and the stopper and round the .perforations for the gas-

supply tubes, to ensure that the apparatus is airtight. The

gelatin is thus in an atmosphere of hydrogen in which the
colonies may develop. The latter may
be examined and isolated in a way
which will be presently described. The
method is admirably suited for all
anaerobes which grow at the ordinary
temperature.

(b) Bulloch's Apparatus for An-
aerobic Culture. — This can be recom-
mended for plating out mixtures
containing anaerobes, and for obtaining
growths (especially surface growths) of
the latter. It consists (Fig. 23) of a
glass plate as base on which a bell
jar can be firmly luted down with
unguentum resina, In the upper part
of the bell jar are two apertures fur-
nished with ground stoppers, and

through each of the latter passes a glass tube on which is a

stop-cock. One tube, bent slightly just after passing through

the stopper, extends nearly to the bottom of the chamber;

the other terminates immediately

below the stopper. In using the

apparatus there is set on the

base -plate a shallow dish, of

slightly less diameter than that

of the bell jar, and having a little

heap of from 2 to 4 grammes of

dry pyrogallic acid placed in it

towards one side. Culture plates

made in the usual way can be

stacked on a frame of glass rod's

resting on the edges of the dish,

or a beaker containing culture

tubes can be placed in it. The

bell jar is then placed in position

so that the longer glass tube is

situated over that part of the

bottom of the shallow dish far-

thest away from the pyrogallic

acid, and the bottom and stoppers are luted. The air in the

bell jar is now expelled by passing a current of hydrogen through

. 23.-BullocVs apparatus for
anaerobic plate cultures.

CULTURES OF ANAEROBES 65

the short glass tube, and both stoppers are closed. A partial
vacuum is then effected in the jar by connecting up the short
tube with an air-pump, opening the tap, and giving a few strokes
of the latter. A solution of 109 grms. solid caustic potash dis-
solved in 145 c.c. water is made, and into the vessel containing
it a rubber tube connected with the long glass tube is made to
dip, and the stopper of the latter being opened, the fluid is forced
into the chamber and spreads over the bottom of the shallow
dish ; potassium pyrogallate is thus formed, which absorbs any
free oxygen still present. Before the whole of the fluid is forced
in the rubber tube is placed in a little boiled water, and this,
passing through the glass tubes, washes out the potash and
prevents erosion of the glass. The whole apparatus may be
placed in the incubator till growth occurs.

It is often advisable in dealing with material suspected to
contain anaerobes to inoculate an ordinary deep glucose agar
tube with it, and, incubating for 24 or 48 hours, to then apply an
anaerobic separation method to the resultant growth. Sometimes
the high powers of resistance of spores to heat may be taken
advantage of in aiding the separation (vide Tetanus).

Cultures of Anaerobes. — When by one or other of the above
methods separate colonies have been obtained, growth may be
maintained on media in contact with ordinary air. The media
generally used are those which contain reducing agents, and the
test-tubes containing the medium must be filled to a depth of
4 inches. They are sterilised as usual, and are called " deep "
tubes. The long straight platinum wire is used for inoculating,
and it is plunged well down into the " deep " tube. A little air
gets into the upper part of the needle track, and no growth takes
place there, but in the lower part of the needle track growth
occurs. From such " deep " cultures growths may be maintained
indefinitely by successive sub-cultures in similar tubes. Even
ordinary gelatin and agar can be used in the same way if the
medium is heated to boiling-point before use to expel any
absorbed oxygen.

Carroll's Method for Anaerobic Cultures. — This may be used
with culture tubes containing any of the media suitable for
anaerobes, with Esmarch's roll-tubes, or with fermentation tubes.
There is required a dry tube of the same diameter as the culture
tube, a short U-shaped glass tube, and two pieces of rubber tub-
ing all of like diameter. The culture tube having been inoculated,
the plug is pushed home below the lip of the tube. The ends
of the U-tube are smeared with vaseline and a rubber tube
slipped over each ; the end of the culture tube being similarly

66 METHODS OF CULTIVATION OF BACTERIA

treated, the free end of one of the rubber tubes is pushed over it
till the glass of the U-tube is in contact with the glass of the
culture tube. In the dry tube 1 or 2 grammes of pyrogallic
acid are placed, and the powder is packed down with a layer of
filter paper. Ten or twenty cubic centimetres of a 10 per cent,
solution of sodium hydrate are then poured in, and the tube is
quickly connected up by the rubber tubing with the other end
of the U-tube. In this apparatus the oxygen is absorbed by the
sodium pyrogallate, and the conditions for anaerobic growth
are fulfilled.

Buchner's Anaerobic Tube. — This may be used either for
maintaining surface growths of anaerobes or for keeping free
from oxygen sloped culture media which are being used for
separating anaerobes from mixtures. Dry pyrogallol is placed
in a cylindrical jar of diameter sufficient to contain the tube
or tubes of media. The tubes are v then inserted, potassium
hydrate solution (p. 65) is poured into the jar, and its mouth
quickly stoppered with a rubber or glass stopper. The stopper
is made airtight by sealing with paraffin. The pyrogallol
absorbs the oxygen in the jar, and thus the cultures are kept in
oxygen-free surroundings.

Growth in Tubes with Pyrogallol-saturated Plug. — Sloped
cultures can be maintained oxygen-free as follows : — The medium
is placed in a long test-tube and inoculated. The plug of the
tube (which ought to be rather tight) is pushed down into the
tube, and a little dry pyrogallol placed on the top of it. A few
drops of the potassium hydrate solution are dropped on the
crystals, and a second plug is inserted in the mouth of the tube.
This is pushed home, and melted paraffin run on to the top to
prevent access of outside air.

Cultures of Anaerobes in Liquid Media. — It is necessary to
employ such in order to obtain the toxic products of the growth
of anaerobes. Glucose broth is most convenient. It is placed
either (1) in a conical flask with a lateral opening and a per-
forated indiarubber stopper, through which a bent glass tube
passes, as in Fig. 24, a, by which hydrogen may be delivered,.
or (2) in a conical flask with a rubber stopper furnished with
two holes, as in Fig. 24, b, through a tube in one of which
hydrogen is delivered, while through the tube in the other the
gas escapes. The inner end of the gas delivery tube must in
either case be below the surface of the liquid ; the inner end of
the lateral nozzle in the one case, and the inner end of the
escape tube in the other, must of course be above the surface of
the liquid. The single tube in the one case and the two tubes

CULTURES OF ANAEROBES IN LIQUID MEDIA 67

in the other ought to be partially drawn out in a flame to
facilitate subsequent complete sealing. The ends of the tubes
through which the gas is to pass are previously protected by
pieces of cotton wool tied on them. It is well previously to
place in the tube, through which the hydrogen is to be delivered,
a little plug of cotton wool. The flask being thus prepared, it
is sterilised by methods B (2) or B (3). On cooling it is ready
for inoculation. In the case of the flask with the lateral nozzle,
the cotton-wool covering having been momentarily removed, a
wire charged with the organism is passed down to the bouillon.
In the other kind of flask the stopper must be removed for an
instant to admit the wire. The flask is then connected with

FIG. 24.

a. Flask for anaerobes in liquid media. Lateral nozzle and stopper fitted
for hydrogen supply, b. A stopper arranged for a flask without lateral nozzle.

the hydrogen apparatus by means of a short piece of sterile
indiarubber tubing, and hydrogen is passed through for half an
hour. In the case of flask (1), the lateral nozzle is plugged
with melted paraffin and covered with alternate layers of cotton
wool and paraffin, the whole being tightly bound on with string.
The entrance tube is now completely drawn off in the flame
before being disconnected from the hydrogen apparatus. In
the case of flask (2), first the exit tube and then the entrance
tube are sealed off in the flame before the flask is disconnected
from the hydrogen apparatus. It is well in the case of both
flasks to run some melted paraffin all over the rubber stopper.
Sometimes much gas is evolved by anaerobes, and in dealing
with an organism where this will occur, provision must be made
for its escape. This is conveniently done by leading down the

68 METHODS OF CULTIVATION OF BACTERIA

exit tube, and letting the end just dip into a trough of mercury
(Fig. 25), or into mercury in a little bottle tied on to the end
of the exit tube. The pressure of gas within causes an escape

at the mercury contact,
which at the same time acts
as an efficient valve. The
method of culture in fluid
media is used to obtain the
soluble products of such
anaerobes as the tetanus
bacillus.

The Method of Tarozzi.—
This observer has found that
if small pieces of fresh
sterile organs are added to
ordinary bouillon, growth of
anaerobes takes place under
ordinary atmospheric con-
ditions. For this purpose,
portions of liver, spleen, or
kidney are most suitable.
If after the piece of tissue has been added the medium is boiled
for a few minutes it loses its property of growing anaerobes,
but the temperature may be raised for a short time almost to
boiling-point without this occurring. The tissue of the organs
gives off something into the medium which favours the growth
of anaerobes, as can be shown by placing the tissue for some
time in the medium and then removing it; thereafter the
medium is suitable for anaerobic growth.

When it is desired to grow anaerobes on the surface of a

FIG. 25. — Flask arranged for culture of
anaerobes which develop gas.

b is a trough of mercury into which
exit tube dips.

FIG. 26. — Tubes for anaerobic cultures on the surface of solid media.

solid medium such as agar, tubes of the form shown in Fig. 26,
a and b, may be used. A stroke culture having been made, the
air is replaced by hydrogen as just described, and the tubes are

HANGING-DROP CULTURES 69

fused at the constrictions. Such a method is of great value
when it is required to get the bacteria free from admixture of
medium, as in the case of staining flagella.

MISCELLANEOUS METHODS.

Hanging-drop Cultures. — It is often necessary to observe
micro-organisms alive, either to watch the method and rate of
their multiplication, or to investigate whether or not they are
motile. This is effected by making hanging-drop cultures. The
method in the form to be described is only suitable for aerobes.
For this special slides are necessary. Two forms are in use, and

FIG. 27.

A. Hollow-ground slide for hanging-drop cultures shown in plan and section.
B. Another form of slide for similar cultures.

are shown in Fig. 27. In A there is ground out on one surface
a hollow having a diameter of about half an inch. That shown
in B explains itself. The slide to be used and a cover-glass are
sterilised by hot air in a Petri's dish, or simply by being heated
in a Bunsen and laid in a sterile Petri to cool. In the case of
A, one or other of two manipulation methods may be employed.
(1) If the organism be growing in a liquid culture, a loop of
the liquid is placed on the middle of the under surface of the
sterile cover-glass, which is held in forceps, the points of which
ha\v IKVH sterilised in a Bunsen flame. If the organism be
growing in a solid medium, a loopful of sterile bouillon is
placed on the cover-glass in the same position, and a very small
quantity of the culture (picked up with a platinum needle) is

70 METHODS OF CULTIVATION OF BACTERIA

rubbed up in the bouillon. The cover is then carefully lowered
over the cell on the slide, the drop not being allowed to touch
the wall or the edge of the cell. The edge of the cover-glass is
covered with vaseline, and the preparation is then complete and
may be placed under the microscope. If necessary, it may be first
incubated and then examined on a warm stage. (2) The sterile
cover-glass is placed on a sterile plate (an ordinary glass plate
used for plate cultures is convenient). The drop is then placed
on its upper surface, the details being the same as in the last
case. The edge of the cell in the slide is then painted with
vaseline, and the slide, held with the hollow surface downwards,
is lowered on to the cover-glass, to the rim of which it of course
adheres. The slide with the cover attached is then quickly
turned right side up, and the preparation is complete.

In the case of B, the drop of fluid is placed on the centre of
the table x. The drop must be thick enough to come in contact
with the cover-glass when the latter is lowered on the slide, and
not large enough to run over into the surrounding trench y.
The cover-glass is then lowered on to the drop, and vaseline is
painted along the margin of the cover-glass. The method of
microscopic examination is described on page 91.

The Counting of Colonies. — An approximate estimate of the
number of bacteria present in a given amount of a fluid (say,
water) can be arrived at by counting the number of colonies
which develop when that amount is added to a tube of suitable
medium, and the latter plated and incubated. An ordinary
plate should be used in such a case, and the medium poured

out in as rectangular a
shape as possible. For
the counting, an appa-
ratus such as is shown
in; Fig. 28 is employed.
This consists of a sheet
of glass ruled into
squares as indicated,
and supported by its
corners on wooden
blocks. The table to

which these blocks are
FIG. 28. -Apparatus for counting colonies.

face. The plate-culture

containing the colonies is laid on the top of the ruled
glass. The numbers of colonies in, say, twenty of the smaller
squares are then counted, and an average struck. The total

METHOD OF COUNTING BACTERIA

number of squares covered by the medium is then taken, and
by a simple calculation the total number of colonies present can
be obtained. Plate-cultures in Petri's dishes are sometimes
employed for purposes of counting. The bottoms of such
dishes are, however, never flat, and the thickness of the medium
thus varies in different parts. If these dishes are to be used,
a circle of the same size as the dish can be drawn with Chinese
white on a black card, the circumference divided into equal arcs,
and radii drawn. The dish is then laid on the card, the number
of colonies in a few of the sectors counted, and an average
struck as before. In counting colonies it is always best to aid
the eye with a small hand-lens.

Method of Counting Living Bacteria in a Culture. — This
is accomplished by putting into practice a dilution method

such as that described on p. 58.
Measured amounts of high dilutions
are plated, and the numbers of
colonies which subsequently develop
are counted. In applying such a
method it is necessary to have pipettes
capable of measuring small quantities of
fluid. Those discharging '05 and '1 c.c. will
be found convenient, and such pipettes can
have subdivisions which enable them to be
used for measuring still smaller fractions of
a cubic centimetre. Pipettes of this kind
can be obtained at the instrument makers.
Wright has described a method by which
a pipette (Fig. 29) for measuring small
quantities of fluid can be made from ordinary
quill tubing. The method is as follows : —
A piece of quill tubing about 15 cm. long is
drawn out to a capillary stem. A standard
5 c.mm. pipette (such as that of the Gower's
hsemocytometer), or the pipette described
later on p. 118, is filled with mercury and
the metal transferred to the capillary stem
and run down to near its extremity; the
upper and lower limits of the mercury are
marked with an oil pencil; the mercury is
then displaced up the tube till its previously
distal end is at the proximal of the two
marks, and a third mark is made at the new position of
the upper cud of the droplet; the manipulation is repeated

250

225

2-5

FlG. 29.— Wright's
260 c.mm. pipette
fitted with nipple.

72 METHODS OF CULTIVATION OF BACTERIA

three more times, and finally the tip of the tube beyond
the lowest mark is broken off. Thus on the capillary part
of the pipette we have five divisions, each capable of
holding 5 c.mm. of fluid. The rest of the pipette is now
calibrated so as to determine that part capable of containing
225 c.mm. and 250 c.mm. This is done by placing a rubber
nipple on the wide end of the pipette and sucking up some
water tinted with, say, methylene-blue till the 25 c.mm. mark
is reached ; a small air-bubble is then allowed to enter the
pipette, then other 25 c.mm. of fluid, then another bubble, and
so on till nine volumes each of 25 c.mm. have been sucked up.
A mark is then made on the tube at the upper level of this
amount, other 25 c.mm. are sucked up, and another mark made.
The fluid is expelled, the tube dried, and that part containing
the 225 and 250 marks is drawn out into an almost capillary
diameter, the manipulation by which the marks were originally
arrived at is repeated, and thus in the new marks made a more
accurate calibration for these amounts is attained. In order to
form a safety chamber a second bulb is formed by drawing out
the tube a little higher up, as in the figure, and finally the upper
inch or two are bent at right angles to the calibrated limb. In
doing this a loop may be thrown on the plastic melted capillary
tube exactly in the way in which a similar loop may be thrown
on a piece of cord. With such a pipette any required dilution
of a culture can be made on the principles already described.

The Bacteriological Examination of the Blood. — (a) This
may be done by taking a small drop from the skin surface, e.g.
the lobe of the ear. The part should be thoroughly washed
with 1-1000 corrosive sublimate and dried with sterile cotton
wool. It is then washed with absolute alcohol to remove the
antiseptic, drying being allowed to take place by evaporation.
A prick is then made with a sterile surgical needle ; the drop of
blood is caught with a sterile platinum loop and smeared on the
surface of agar or blood serum. Film preparations for micro-
scopic examination may be made at the same time. It is rare
to obtain growths from the blood of the human subject by this
method (vide special chapters), and if colonies appear the pro-
cedure should be repeated to exclude the possibility of accidental
contamination.

(b) A larger quantity of blood may be obtained by puncture
of a vein ; this is the only satisfactory method, and should be
that followed whenever practicable. The skin over a vein in
the forearm or on the dorsum of the foot having been sterilised,
the vein is made turgid by pressure, and the needle of a syringe

EXAMINATION OF CEREBRO-SPINAL FLUID 73

of 10-15 c.c. capacity, carefully sterilised, is then plunged
obliquely through the skin into the lumen of the vessel. Several
cubic centimetres of blood can thus be withdrawn into the
syringe. Some of the blood (e.g. 1 c.c.) should be added to
.small flasks containing 50 c.c. of bouillon ; the rest may be
used for smearing the surface of agar tubes, or may be added to
melted agar at 42° C., which is then plated. The flasks, etc.
are then incubated. By this method cultures can often be
obtained where the former method fails, especially in severe
conditions such as ulcerative endocarditis, streptococcus infection,
etc. Part of the blood may be incubated by itself for twenty-
four hours, and cultures then made. Needless to say, the in-
oculations of media must be done at the bedside, as of course
the blood quickly coagulates in the syringe. Coagulation can
be prevented by drawing up into the syringe before it is used a
quantity of 2 per cent, sterile sodium citrate equivalent to the
amount of blood it is intended to withdraw.

In examining the blood of the spleen a portion of the skin
over the organ is sterilised in the same way, a few drops are
withdrawn from the organ by a sterile hypodermic syringe, and
cultures made. (For microscopic methods, vide p. 94.)

Bacteriological Examination of the Cerebro-spinal Fluid
— Lumbar Puncture.— This diagnostic procedure, which is
often called for in cases of meningitis, can be carried out with
a sterilised " antitoxin needle " as follows : — The patient should
lie on the right side, with knees somewhat drawn up and left
shoulder tilted somewhat forward, so that the back is fully
exposed. The skin over the lumbar region is then carefully
sterilised, as above described, and the hands of the operator
should be similarly treated. The spines of the lumbar vertebrae
having been counted, the left thumb or forefinger is pressed
into the space between the third and fourth spines in the middle
line ; the needle is then inserted about half an inch to the
right of the middle line at this level and pushed through the
tissues, its course being directed slightly inwards and upwards,
till it enters the subdural space. When this occurs, fluid passes
along the needle, sometimes actually spurting out, and should be
received in a sterile test-tube. Several cubic centimetres of
fluid can thus usually be obtained, no suction being required ;
thereafter it can be examined bacteriological ly by the usual
methods. The depth of the subdural space from the surface
varies from a little over an inch in children to 3 inches, or
even more, in adults — the length of the needle must be suited
accordingly. In making the puncture it is convenient to have

74 METHODS OF CULTIVATION OF BACTERIA

either a sterile syringe attached, or to have the thick end of
the needle covered with a pad of sterile wool, which is of course
removed at once when the fluid begins to flow. It is advisable
to use the platinum needles which are specially made for the
purpose, as a sudden movement of the patient may snap an
ordinary steel needle.

The Bacteriological Examination of Urine. — In such an
examination care must be taken to prevent the contamination
of the urine by extraneous organisms. In the male, specimens
withdrawn by a sterile catheter into a sterile vessel are pre-
ferable, but it is often sufficient to wash thoroughly the glans
penis and the meatus with 1-1000 corrosive sublimate — the
lips of the meatus being everted for more thorough cleansing ;
the urine is then passed into a series of sterile flasks, the first
of which is rejected in case contamination has occurred. In
the female, after similar precautions as regards external
cleansing, the catheter must be used. The latter must be
boiled for half an hour, and anointed with olive oil sterilised
by half an hour's exposure in a plugged flask to a temperature
of 120° C. Here, again, it is well to reject the urine first
passed. It is often advisable to allow the urine to stand in a
cool place for some hours, to then withdraw the lower portion
with a sterile pipette, to centrifugalise this, and to use the
urine in the lower parts of the centrifuge tubes for microscopic
examination or culture.

Filtration of Cultures. — For many purposes it is necessary
to filter all the organisms from fluids in which they may have
been growing. This is done especially in obtaining the soluble
toxic products of bacteria. The only filter capable of keeping
back such minute bodies as bacteria is that formed from a tube
of unglazed earthenware as introduced by Chamberland. The
efficiency of such a filter depends- on the fineness of the grain
of the clay from which it is made ; the finest is the Kitasato
filter and the Chamberland " B " pattern ; the next finest is
the Chamberland " F " pattern, which is quite good enough for
ordinary work. There are several filters, differing slightly in
detail, all possessing the common principle. Sometimes the
fluid is forced through the porcelain tube. In one form the
filter consists practically of an ordinary tap screwed into
the top of a porcelain tube. Through the latter the fluid is
forced, and passes into a chamber formed by a metal cylinder
which surrounds the porcelain tube. The fluid escapes by an
aperture at the bottom. Such a filter is very suitable for
domestic use, or for use in surgical operating-theatres. As

FILTRATION OF CULTURES

considerable pressure is necessary, it is evident it must be put
on a pipe leading directly from the main. Sometimes, when
fluids to be filtered are
very albuminous, they
are forced through a
porcelain cylinder by
compressed carbonic
acid gas. The filtra-
tion of albuminous
fluids may sometimes
be facilitated by keep-
ing them near blood-
heat during the pro-
cess. For ordinary
bacteriological work,
filters of various kinds
are in the market
(such as those of Klein
and others), but the
most generally con-
venient is that in
which the fluid is FIG. 30.— Geissler's vacuum pump arranged with
sucked through the manometer for filtering cultures. (The tap
i • r i and pump are intentionally drawn to a larger

porcelain by exhaust- ^ &£ the manometer board to show
ing the air in the details.)
receptacle into which

it is to flow. This is conveniently done by means of a
Geissler's water-exhaust pump (Fig. 30, g\ which must be
fixed to a tap leading directly from the main. The connection

with the tap must be effected by
means of a piece of thick-walled
rubber-tubing as short as possible,
wired on to tap and pump, and
firmly lashed externally with many
turns of strong tape. Before lashing
with the tape the tube may be
strengthened by fixing round it
with rubber solution strips of the
rubbered canvas used for mending
punctures in the outer case of a
bicycle tyre. A manometer tube
(b) and a receptacle (c) (the latter

to catch any back flow of water from the pump if the filter
breaks) are intercepted between the filter and the

FIG. 31.— Chamberland's candle
and Hask arranged for filtra-
tion.

76 METHODS OF CULTIVATION OF BACTERIA

pump. These are usually arranged on a board a, as in Fig. 30.
Between the tube / and the pump g, and between the tube d
and the filter, it is convenient to insert lengths of flexible
lead-tubing connected up at each end with short, stout-walled
rubber-tubing.

Filters are arranged in various ways, (a) An apparatus is
arranged as in Fig. 31. The fluid to be filtered is placed in
the cylindrical vessel a. Into this a " candle " or " bougie "
of porcelain dips. From the upper end of the bougie a glass
tube with thick rubber connections, as in Fig. 31, proceeds to
flask b, and passes through one of the two perforations with
which the rubber stopper of the flask is furnished. Through

FIG. 32. — Cliamberland's bougie
arranged with lamp funnel for
filtering a small quantity of
riuid.

FIG. 33.— Bougie inserted
through rubber stopper
for same purpose as in
Fig. 32.

the other opening a similar tube proceeds to the exhaust-
pump. When the latter is put into action the fluid is sucked
through the porcelain and passes over into flask b. This
apparatus is very good, but not suitable for small quantities of
fluid.

(b) A very good apparatus can be arranged with a lamp
funnel and the porcelain bougie. These may be fitted up in
two ways. (1) An indiarubber washer is placed round the
bougie c at its glazed end (vide Fig. 32). On this the narrow
end of the funnel d, which must, of course, be of an appropriate
size, rests. A broad band of sheet rubber is then wrapped
round the lower end of the funnel, and the projecting part of
the bougie. It is firmly wired to the funnel above and to the

FILTRATION OF CULTURES

bougie below. The extreme point of the latter is left exposed,
and the whole apparatus, being supported on a stand, is con-
nected by a glass tube with the lateral tube of the flask b ; the
tube a is connected with the exhaust-pump. The fluid to be
filtered is placed between the funnel and the bougie in the
space e, and is sucked through into the flask b. The efficiency
of such a filter, especially when small amounts of fluid are being
dealt with, is much increased if when the level of the fluid falls
below the upper end of the candle a closely fitting test-tube is
slipped over the latter. By this device the leakage of air
through the exposed part of the candle is prevented. There
are now in the market candles with glass sheaths cemented into
a nickle-plated fitting from the lower part of which a metal
tube emerges; the latter
can be passed through a
rubber stopper into a filter
tlask. (2) This modifica-
tion is shown in Fig. 33.
Into the narrow part of
the funnel an indiarubber
bung is fitted, with a per-
foration in it sufficiently
large to receive the candle,
which it should grasp
tightly.

(o) Muencke's modifica-
tion of the Chamberland
filter is seen in Fig. 34.
It consists of a thick-
\\alird tlask a, the lower

part conical, the upper cylindrical, with a strong flange on the
lip. There are two lateral tubes, one horizontal to connect with
exhaust-pi] >e, and one sloping, by which the contents may be
j toured out. Passing into the upper cylindrical part of the
tlask is a hollow porcelain cylinder 6, of less diameter than the
cylindrical part of flask a. It is closed below, open above, and
rests by a projecting rim on the flange of the flask, an asbestos
\\asher, c, being interposed. The fluid to be filtered is placed
in the porcelain cylinder, and the whole top covered, as shown
at /, with an indiarubber cap with a central perforation ; the
tube d is connected with the exhaust-pump, and the tube e
plugged with a rubber stopper. For filtering small quantities
of fluid the apparatus shown in Fig. 35 may be used. It
consists of a small Chamberland bougie fitted by a rubber tube

Fi({. 34. — Muencke's modification of
Chamberland's filter.

78 METHODS OF CULTIVATION OF BACTERIA

to a funnel, the stem of which has been passed through a rubber
cork; this cork fits into a conical flask with side arm for
connection with exhaust.

Before any one of the above apparatus is used it ought to
be connected up as far as possible and sterilised in the Koch's
steriliser. The ends of any important
unconnected parts ought to have pieces of
cotton wool tied over them. After use
the bougie is to be sterilised in the auto-
clave, and after being dried is to be passed
carefully through a Bunsen name to burn
off all organic matter. If the latter is
allowed to accumulate, the pores become
filled up.

The success of filtration must be tested
by inoculating tubes of media from the
filtrate, and observing if growth takes
place, as there may be minute perforations
in the candles sufficiently large to allow
bacteria to pass through. Filtered fluids
keep for a long time if the openings of
the glass vessels in which they are placed
are kept thoroughly closed, and if these
vessels be kept in a cool place in the dark.
A layer of sterile toluol about half an inch
thick ought to be run on to the top of the
filtered fluid to protect the latter from the
atmospheric oxygen.

Instead of being filtered off, the bacteria
may be killed by various antiseptics,
chiefly volatile oils, such as oil of mustard

(Roux). These oils are stated to have no injurious effect on the
chemical substances in the fluid, and they may be subsequently
removed by evaporation. It is not practicable to kill the.
bacteria by heat when their soluble products are to be studied,
as many of the latter are destroyed by a lower temperature than
is required to kill the bacteria themselves.

Bacteria can be almost entirely removed from fluid cultures
by spinning the latter in a centrifuge of very high speed (e.g.
C. J. Martin's turbine centrifuge), and this method is sometimes
adopted in practice.

The Observation of Bacterial Fermentation of Sugars, etc.
—The capacity of certain species of bacteria to originate fermenta-
tions in sugars constitutes an important biological factor. It

FIG. 35. — Flask for
filtering small quanti-
ties of fluid.

BACTERIAL FERMENTATION OF SUGARS 79

is well to consider this factor in relation to the chemical con-
stitution of the sugars. These bodies are now known to be (to
use the definition of Holleman) aldehyde or ketoue alcohols
containing one or more hydroxyl groups, one of which is directly
linked to a carbon atom in union with car bony 1. The group
characteristic of a sugar is thus — CHOH - CO - . The sugars
are divided into monosaccharides or monoses, disaccharides
(dioses), and polysaccharides (polyoses). The members of the
last two groups may be looked on as derived from the combina-
tion of two or more molecules of a monosaccharide with the
elimination of water (e.g. 2C6Hr2O6 = C1?H.22OU + H?O).

Monosaccharides. — These are classified according to the
number of C atoms they contain. The pentoses ordinarily used
are arabinose (obtained from gum arabic), xylose (from wood),
and rhamnose (which is really a methylpentose). Among the
hexoses are glucose (dextrose) with dextro-rotatory properties.
Glucose is an aldehyde alcohol (aldose). In fruit there is also
a ketone alcohol (ketose) called fructose, which from its laevo-
rotatory properties is also known as laevulose. Other hexosee
are mannose (from the vegetable ivory nut) and galactose (a
hydrolytic derivative of lactose).

Disaccharides (C12H2.2On). — The ordinary members of this
group are maltose (derived from starch), lactose, and cane sugar
(sucrose, saccharose).

Polysaccharides. — Examples are starch, raffinose, inulin (from
dahlia roots), dextrin, arabin, glycogen, cellulose.

If we consider sugars generally from the point of view of
the capacity of yeast to originate alcoholic fermentation in them,
we may say that the simpler the constitution of the sugar the
more easily is it fermented. Thus the monosaccharides are
more easily acted on by yeast than the di- or poly-saccharides.
Usually an independent process resulting in the splitting of the
higher into the lower is preliminary to the alcoholic fermentation.
Thus yeast first inverts cane sugar into glucose and fructose, and
then acts on these products. From what is known it is probable
that similar facts hold with regard to the action of bacteria.

Besides sugars, other alcohols with large molecules may be
broken down by bacterial action, and these bodies have been
used for differentiating the properties of allied bacteria. Among
these substances may be mentioned the trihydric alcohol glycerol
(glycerin), the tetrahydric erythritol and the hexahydric dulcitol
(dulcite), mannitol (mannite), and sorbitol (sorbite).

Similarly certain glucosides, such as salicin, coniferin, etc.,
have been used for testing the fermentative properties of

80 METHODS OF CULTIVATION OF BACTERIA

bacteria. Other substances allied to sugars (e.g. inosite) have
also been used.

The end products of bacterial fermentations may be various.
They differ according to the sugar employed and according to
the species of bacterium under observation, and frequently a
species which will ferment one sugar has no effect on another.
The substances finally produced, speaking roughly, may be
alcohols, acids, or gaseous bodies (chiefly carbon dioxide,
hydrogen, and methane). For the estimation of the first groups
complicated chemical procedure may be necessary. The tests
usually employed for the detection of ordinary fermentative
processes depend on two kinds of changes, namely, (a) the evolution
of gases and (6) the formation of acids. Generally speaking, we
may say that such tests are reliable, and the methods to be
pursued are simple. Besides such gases as those named, some
organisms give rise to sulphuretted hydrogen by breaking up the
proteid. The formation of this gas can be detected by the
blackening of lead acetate when it is added to the gas-containing
medium.

In testing the effect of a bacterium on a given sugar it is
essential that this sugar alone be present; the basis of the
medium ought therefore to be either peptone solution (vide p. 39)
or a dextrose-free bouillon (vide infra). The sugar or other
substance is added in the proportion of from a half to one
per cent., and care is taken not to overheat during sterilisation.

It is preferable that the addition should Le made in the form of a sterile
solution. If the sugar in solid form be placed in the bouillon and this
then sterilised, there is danger that chemical changes may take place
in the sugar, in consequence of its being heated in the presence of
substances (such as alkalies) which may act deleteriously upon it ; in any
case sterilisation should not be at a temperature above 100° C.

To obtain a " dextrose- free" bouillon it is usual to inoculate ordinary
bouillon with some organism, such as b. coli, which is known to ferment
dextrose, and allow it to act for forty-eight hours. The bouillon is
then filtered and re-sterilised. A sample is tested for another period
of forty-eight hours with b. coli, to make certain that all the dextrose
has been removed. If no fresh gas-formation is observed, then to the
remainder of the bouillon the sugar to be investigated may be added.

For the observation of gas-formation either of the following
methods may be employed : —

(1) Durham's Tubes (Fig. 36, b).— The plug of a tube which
contains about one-third more than usual of a liquid medium is
removed, and a small test-tube is slipped into the latter, mouth
downwards. The plug is replaced and the tube sterilised thrice
for ten minutes at 100° C. The air remaining in the smaller

BACTERIAL FERMENTATION OF SUGARS 81

tube is thereby expelled. The tube is then inoculated with the
bacterium to be tested. Any gas developed collects in the upper
part of the inner tube. As some of the sugars now used for
fermentation tests are rather expensive, it is well to arrange the
Durham apparatus with very small tubes ; with these a satis-
factory result can be obtained with only 1 c.c. of medium.

(2) T/te Fermentation Tube (Fig. 36, c). — This consists of a
tube of the form shown, and the figure also indicates the extent
to which it ought to be filled. It is inoculated in the bend with
the gas-forming organism, and when growth occurs the gas

FIG. 36. — Tubes for demonstrating gas-fonaatiou by bacteria.

", tube with "shake" culture.

b, Durham's fermentation tube.

c, ordinary form of fermentation tube.

collects in the upper part of the closed limit, the medium being
displaced into the bulb.

For the observation of the effect of an organism on glucose,
the following method may be employed : —

Gelatin Shake Cultures (Fig. 36, a). — The gelatin in the tube
is melted as for making plates ; while liquid it is inoculated
witli the growth to be observed, and shaken to distribute the
organisms throughout the jelly. It is then allowed to solidify,
;tn 1 i- set a>ide at a suitable temperature. If the bacterium used
is a gas-forming one, then, as growth occurs, little bubbles
appear round the colonies.
6

82 METHODS OF CULTIVATION OF BACTERIA

In this method the gas-formation results from fermentation
of the glucose naturally present in the medium from transforma-
tion of the carbo-hydrates of muscle. The amount of glucose
naturally present, however, varies much, and therefore glucose
should be added to the medium if the effects on this sugar are
to be observed with certainty. The shake culture method may
be utilised for observing fermentation in other sugars by adding
to peptone solution containing the sugar 10 to 15 per cent, of
gelatin.

The development of an acid reaction is demonstrated by the
addition of an indicator to the medium, litmus being generally
used. The details of the composition of such media have already
been given. In Hiss's serum water media the production of
acid also leads to coagulation of the medium. Sometimes acid
is formed very slowly from sugars, so that it is well to keep the
cultures under observation for several days.

Acid and gas-formation may be simultaneously tested for, by
placing the fluid medium containing the indicator in Durham's
tubes.

In all tests in which sugars are used, a control uninoculated
tube ought to be incubated along with the bacterial cultures, as
changes in reaction sometimes spontaneously occur in media
containing unstable sugars.

The capacity of an organism to produce acid may be measured
by taking a standard amount of a fluid medium and allowing
growth to take place for a standard time, and then adding an
amount of, say, decinormal soda solution sufficient to bring the
litmus back to the tint of the original medium.

The Observation of Indol-fonnation by Bacteria. — The
formation of indol from albumin by a bacterium sometimes
constitutes an important specific characteristic. To observe
indol production the bacterium is grown, preferably at incubation
temperature, in a fluid medium containing peptone. The latter
may either be sugar-free bouillon or preferably peptone solution
(see p. 39). Any medium containing sugars must be avoided,, as
the presence of these substances may inhibit the production of
indol. Two methods are in use for the detection of this body.

(1) The Nitroso-indol Method. — Indol is here recognised by
the fact that when it is acted on by nitric acid in the presence of
nitrites, a nitroso-indol compound is produced, which has a rosy
red colour. Some bacteria (e.g. the cholera vibrio) produce
nitrites as well as indol, but usually in making the test (e.g. in
the case of b. coli) the nitrites must be added. This is effected
by adding to an ordinary tube of medium 1 c.c. of a '02 per

INDOL-FORMATION BY BACTERIA 83

cent, solution of potassium nitrite, and testing with pure nitric
or sulphuric acid. In any case only a drop of the acid need be
added to, say, 10 c.c. of medium. If no result be obtained at
once it is well to allow the tube to stand for an hour, as some-
times the reaction is very slowly produced. In many instances
incubation at 37° C. for several days may be necessary before
tlir presence of indol is demonstrable. The amount of indol
produced by a bacterium seems to vary very much with certain
unknown qualities of the peptone. It is well, therefore, to test
a series of peptones with an organism (such as the b. coli)
known to produce indol, and, noting the sample with which the
best reaction is obtained, to reserve it for making media to be
used for the detection of this product. This method has for
long been felt not to be satisfactory, and the following at present
bids fair to replace it : —

(2) Ehrlictis Rosindol Reaction-: — The adaptation of this to
bacteriological purposes was brought forward by Bohme in 1906.
For ease of application and delicacy of effect the reaction
possesses great advantages. It depends on the fact that
paradimethylamidobenzaldehyde unites with indol to form a
rosindol body whose colour is readily developed, especially in
presence of an oxidising substance such as potassium per-
sulphate (K2SoO8). Two solutions are required : —

(1) Paradimethylamidobenzaldehyde (Grubler) 4 grms.
Absolute alcohol (96 per cent.) . . 380 c.c.
Concentrated hydrochloric acid . . 80 c.c.

(•2) Potassium persulphate . Saturated watery solution.

To a 10 c.c. bouillon culture of the organism add 5 c.c. of (1)
and then 5 c.c. of (2), and shake well (according to MacConkey
1 c.c. of each solution is sufficient) ; if indol be present a rose-
red colour will appear in a few minutes. Sometimes the rose
colour appears on the addition of solution (1), and the addition
of a special oxidising agent is unnecessary. The rosindol com-
pound can be separated from the culture by shaking the latter
up with amyl alcohol, and MacConkey recommends that this
should be done in cases of a doubtful reaction, as sometimes
when a faint pink colour appears in the culture tube the
extracting alcohol remains colourless, showing that no real
reaction has occurred. Marshall has pointed out that by means
of the reaction a quantitative estimate of the amount of indol
formation can be obtained. To do this a large culture, say
100 c.c., is distilled, and the colour obtained by applying the

84 METHODS OF CULTIVATION OF BACTERIA

test to the distillate in a Nessler's tube is matched against that
obtained with different amounts of a standard solution of indol
(prepared by dissolving 1 gr. indol in 5 c.c. absolute alcohol, and
making up to 500 c.c. with distilled "water).

There is no doubt that the Ehrlich test is from five to ten
times more delicate than the ordinary nitroso-indol reaction, and
it is of especial value in dealing with organisms of the coli-
typhoid group. With strains of b. coli it can often be obtained
in from twenty-four to forty-eight hours, but in the case of a nega-
tive result a culture of from six to seven days ought to be used.
The reaction is also obtainable with the cholera vibrio, but further
investigation is here necessary, as Marshall states that under
certain circumstances the nitrites formed by this bacterium may
have an inhibitory effect on the production of the rose colour.

The Drying of Substances in vacuo. — As many substances,
for example toxins and antitoxins, with which bacteriology is
concerned would be destroyed by drying with heat as is done in
ordinary chemical work, it is necessary to remove the water at
the ordinary room temperature. This is most quickly effected
by drying in vacuo in the presence of some substance such as
strong sulphuric acid, which readily takes up water vapour. The
vacuum produced by a water-pump is here not available, as in
such a vacuum there must always be water vapour present. An
air-pump is therefore to be employed. Here we have found the
Geryk pump most efficient, and it has this further advantage,
that its internal parts are lubricated with an oil of very low
vapour density, so that almost a perfect vacuum is obtainable.
The apparatus is shown in Fig. 37. The vacuum chamber
consists of a bell-jar set on a brass plate. A perforation in the
centre of the latter leads into the pipe a, W7hich can be connected
by strong- walled rubber-tubing with the air-pump, and wThich
can be cut off from the latter by a stop-cock I. In using the
apparatus the substance to be dried is poured out in flat dishes
(one-half of a Petri capsule does very well), and these are stacked
alternately with similar dishes of strong sulphuric acid on a
stand which rests on the brass plate. The edge of the bell-jar
is well luted with unguentum resinse and placed in position and
the chamber exhausted. In a few hours, if, as is always advis-
able, each dish have contained only a thin layer of fluid, the
drying will be complete. The vacuum is then broken by
admitting air very slowly through a by-pass c, and the bell-jar
is removed. In such an apparatus it is always advisable, as is
shown in the figure, to have interposed between the pump and
the vacuum chamber a Wolff's bottle containing sulphuric acid,

STORING AND INCUBATION OF CULTURES 85

This protects the oil of the pump from contamination with
water vapour. Whenever the vacuum is produced, the rubber-
tube should be at once disconnected from a, the cock b being
shut. It is advisable when the apparatus is exhausted to cover
the vacuum chamber and the Wolff's bottle with wire guards
covered with strong cloth, in case, under the external pressure,
the glass vessels give way. The connecting and disconnecting
of rubber-tubing of sufficient thickness to withstand collapse
when exhausted is difficult. Ordinary stout rubber-tubing can
be used if through it there is passed a. length of narrow7 flexible

FIG. 37. — Geryk air-pump for drying in varno.

metal-tubing, the ends of which project beyond the rubber-tubing
so as to enter the parts of the apparatus to which the latter is
fitted.

The Storing and Incubation of Cultures. — Gelatin cultures
must be grown at a temperature below their melting-point, i.e.
for 10 per cent, gelatin, below 22° C. They are usually kept in
ordinary rooms, which vary, of course, in temperature at different
times, but which have usually a range of from about 12° C. to
18° C. Agar and serum media are employed to grow bacteria
at a higher temperature, corresponding to that at which
the organisms grow best, usually 37° C. in the case of patho-
genic organisms. For the purpose of maintaining a uniform
temperature incubators are used. These vary much in the

86 METHODS OF CULTIVATION OF BACTERIA

details of their structure, but all consist of a chamber with
double walls between which some fluid (water or glycerin and
wrater) is placed, which, when raised to a certain temperature,
ensures a fairly constant distribution of
& the heat round the chamber. The latter

is also furnished with double doors, the
inner being usually of glass. Heat is sup-
plied from a burner fixed below. These
burners vary much in design. Sometimes
a mechanism devised in Koch's laboratory
is affixed, which automatically turns off the
gas if the light be accidentally extinguished.
Between the tap supplying the gas, and the
burner, is interposed a gas regulator. Such
regulators vary in design, but for ordinary
chambers which require to be kept at a
constant temperature, Reichert's is as good
and simple as any, and is not expensive.
It is shown in Fig. 38.

It consists of a long tube / closed at the lower
end, open at the upper, and furnished with two
lateral tubes. The lower part is filled with mer-
cury up to a point above tlie level of the lower
lateral tube. The end of the latter is closed by
a brass cap through which a screw d passes, the
FIG. 38. — Reichert's inner end of which lies free in the mercury. The
gas regulator. height of the latter in the perpendicular tube

can thus be varied by increasing or decreasing the

capacity of the lateral tube by turning the screw a few turns out of or
into it. Into the upper open end of the perpendicular tube fits accurately
a bent tube g, drawn out below to a comparatively small open point c,
and having in its side a little above the point "a minute needle-hole
called the peephole or by-pass e. To fix the apparatus the long
mercury bulb is placed in the jacket of the chamber to be controlled,
tube a is connected to gas supply, tube b with the burner. The upper
level of the mercury should be some distance below the lower open end
of tube c. The burner is now lit. The gas passes in at a through c
and e and out at b to the burner. When the thermometer in the
interior of the chamber indicates that the desired temperature has been
reached, the screw d is turned till the mercury reaches the end of the
tube c. Gas can only now pass through the peephole e, and the flame
goes down. The contents of the jacket cool, the mercury contracts off
the end of tube c, and the flame rises. This alternation going on, the
temperature of the chamber is kept very nearly constant. If the mercury
cuts off the gas supply before the desired temperature is reached, and
the screw d is as far out as it will go, then some of the mercury must be
removed. Similarly, if when the desired temperature is reached and the
screw d is as far in as it can go, the mercury does not reach c, some more
must be introduced. If the amount of gas which passes through the

STORING AND INCUBATION OF CULTURES 87

peephole is sufficient still to raise the temperature of the chamber when
c is closed by the rise of the mercury, then the peephole is too large. Tube
7 must be unshipped and e plastered over with sealing-wax, which is
pricked, while still soft, with a very line needle. The gas flame, when
only the peephole is supplying gas, ought to be sufficiently large not to
be blown out by small currents of air. If the pressure of gas supplied to
a regulator varies much in the twenty-four hours, a pressure regulator
ought to be interposed between the gas tap and the instrument. Several
varieties of these can be obtained. In all cases g ought to be fixed to b
with a turn of wire.

The varieties of incubators are, as we have said, numerous.
The most complicated and expensive are made by German

FIG. 39. — Hearsou's incubator for use at 37° C.

manufacturers. Many of these are unsatisfactory, as they easily
get out of order and are difficult to repair. We have found
those of Hearson of London extremely good, and in proportion
to their size much cheaper than the German articles. They are
fitted with an admirable regulator. It is preferable in using an
iinMibator to connect the regulator with the gas supply and with
the Bunsen by flexible metal-tubing. It is necessary to see that
there is not too much evaporation from the surface of cultures
placed within incubators, otherwise they may quickly dry up.
It is thus advisable to raise the amount of water vapour in the
interior by having in the bottom of the incubator a flat dish full
« if water from which evaporation may take place. With tubes

88 METHODS OF CULTIVATION OF BACTERIA

which will require to be long in the incubator, the plugs should
be pushed a little way into the tube and a few drops of
melted paraffin dropped on the top of the wool, or the plugs
should be covered either by indiarubber caps or by pieces of
sheet rubber tied over them. These caps should be previously
sterilised in 1-1000 corrosive sublimate and then dried. Before
they are placed on the tube the cotton-wool plug ought to be
well singed in a flame. " Cool " incubators are often used for
incubating gelatin at 21° to 22° C. An incubator of this kind
fitted with a low-temperature Hearson's regulator is in the
market.

Method of Mounting Bacterial Cultures as Permanent
Museum Specimens (Richard Muir). — (a) Stab or Stroke
Cultures in Nutrient Gelatin or Agar Media. — When the culture
shows typical characters, further growth is arrested by placing the
tube in a formol vapour chamber, or by saturating the cotton-
wool plug with strong formalin. Then leave for a day or two.
Make up the following : —

(1) Thymol water (saturated in cold) . . . 100 c.c.
Glycerin . . . . . . . 20 c.c.

Acetate of potash ...... 5 grms.

Coignet's (gold label) gelatin .... 10 grms.

Render the mixture acid to litmus with acetic acid ; clear with white
of egg and filter.

Warm to about 40° C., and removing cotton-wTool plug from
culture take a little of the preserving fluid in a pipette and
allow to run gently over surface of medium in tube. Place in
such a position that a thin layer of the preserving medium
remains completely covering the growth and the surface of
culture medium. The gelatin is now allowed to solidify. Add
three or four drops of strong formalin to the tube, and fill up to
within a quarter of an inch of the top of the tube with the
following fluid : —

(2) Thymol water (saturated in cold) . . . 100 c.c.
Glycerin . . . . . . . 20 c.c.

Acetate of potash ...... 5 grms.

Cover top of tube with a small piece of paper so as to keep out
dust, allow to stand for a day or two so that small air-bells may
rise to the surface.

To seal tube, pour melted paraffin gently on to the surface
of fluid to near the top of tube; allow to solidify. Cover
paraffin with layer of alcoholic orange shellac cement; allow

GENERAL LABORATORY RULES 89

this to set, and repeat until the cement becomes level with top
of test-tube. When set, a few drops of black lacquer are put on,
and a circular cover-glass of about the same diameter as the
mouth of tube is placed so as completely to seal it.

(/;) The following method is useful for preparing plate cultures :
Instead of making the cultures in Petri's capsules, use ordinary
watch-glasses. The watch-glass is sterilised in a Petri's capsule,
and the inoculated medium is poured out into the watch-glass,
allowed to solidify in the usual way, and left in the Petri's
capsule until the colonies of growth have developed. The
watch-glass is now removed from capsule, and a layer of the
preserving gelatin medium (1), to which have been added a few
drops of strong formalin, is allowed to spread over the surface
of the culture medium. When the layer is solidified the watch-
glass is filled up with the same, and a clean square or oblong
piece of glass (which of course should be of slightly larger
diameter than the watch-glass) is now carefully placed over
watch-glass, care being taken that no air-bells are formed. The
edge of watch-glass should be closely applied to the glass cover,
an<l left in position until the gelatin has solidified. The super-
fluous gelatin is now removed, and the glasses sealed first with
the orange shellac cement, then with black lacquer. It is now
lii i i -lied off by using a circular mask of suitable size.

The various kinds of solid media used in the cultivation of
bacteria, such as blqod serum, potato, bread paste, etc., can be
nvated in the same manner with excellent results.

General Laboratory Rules. — On the working bench of every
bacteriologist there should be a large dish of 1-1000 solution of
mercuric chloride in water. Into this all tubes, vessels, plates,
hanging-drop cultures, etc., which have contained bacteria and
with which he has finished, ought to be at once plunged (in the
case of tubes, the tube and plug should be put in separately).
On no account whatever are such infected articles to be left
lying about the laboratory. The basin is to be repeatedly
cleaned out. All the glass is carefully washed in repeated
changes of tap water to remove the last trace of perchloride of
mercury, a very minute quantity of which is sufficient to inhibit
growth. Old cultures which have been stored for a time, and
from which fresh sub-cultures have been made, ought to be
steamed in the Koch's steriliser for two or three hours, or in the
autoclave for a shorter period, and the tubes thoroughly washed
out. Besides a basin of mercuric chloride solution for infected
apparatus, etc., there ought to be a second reserved for the
worker's hands in case of any accidental contamination. When,

90 METHODS OF CULTIVATION OF BACTERIA

as in public-health work, a large number of tubes are being daily
put out of use, they may be placed in an enamelled slop-pail,
and this when full is placed in the steam steriliser.

A white glazed tile on which a bell-jar can be set is very
convenient to have on a bench. Infective material in watch-
glasses can be placed thus under cover while investigation is
going on, and if anything is spilled the whole can be easily
disinfected. In making examinations of organs containing
virulent bacteria, the hands should be previously dipped in
1-1000 mercuric chloride and allowed to remain wet with this
solution. No food ought to be partaken of in the laboratory,
and pipes, etc., are not to be laid with their mouth-pieces on
the bench. No label is to be licked with the tongue. Before
leaving the laboratory the bacteriologist ought to wash the
hands and forearms with 1-1000 mercuric chloride and then
with yellow soap. In the case of any fluid containing bacteria
being accidentally spilt on the bench or floor, 1-1000 mercuric
chloride is to be at once poured on the spot. The air of the
laboratory ought to be kept as quiet as possible.

CHAPTER III.

MICROSCOPIC METHODS.

The Microscope. — For ordinary bacteriological work a good
microscope is essential. It ought to have a heavy stand, with
rack and pinion and fine adjustment, a double mirror (flat on
one side, concave on the other), a good condenser, with an iris
diaphragm, and a triple nose-piece. It is best to have three
objectives, either Zeiss A, D, and y^-inch oil immersion, or the
lenses of other makers corresponding to these. The oil immer-
sion lens is essential. It is well to have two eye-pieces, say
Nos. 2 and 4 of Zeiss or lenses of corresponding strengths.
The student must be thoroughly familiar with the focussing of
the light on the lens by means of the condenser, and also with
the use of the immersion lens. It may here be remarked that
when it is desired to bring out in sharp relief the margins of
unstained objects, e.y. living bacteria in a fluid, a narrow
aperture of the diaphragm should be used, whereas, in the case
of stained bacteria, when a pure coloured picture is desired, the
diaphragm ought to be widely oj>ened. The flat side of the
mirror ought to be used along with the condenser. When the
observer has finished for the time being with the immersion
lens he ought to wipe off the oil with a piece of silk or
very fine washed linen. If the oil has dried on the lens
it may be moistened with xylol — never with alcohol, which
will dissolve the material by which the lens is fixed in its metal
carrier.

Microscopic Examination of Bacteria.— 1. Hanging-drop
Preparations. — Micro-organisms may be examined : (1) alive or
dead in fluids ; (2) in film preparations ; (3) in sections of
tissues. In the two last cases advantage is always taken of the
affinity of bacteria for certain stains. When they are to be
examined in fluids a drop of the liquid may be placed on a slide

92 MICROSCOPIC METHODS

and covered with a cover-glass.1 It is more usual, however, to
employ hanging-drop preparations. The technique of making
these has already been described (p. 69). In examining them
microscopically, it is necessary to use a very small diaphragm.
It is best to focus the edge of the drop with a low-power
objective, and, arranging the slide so that part of the edge
crosses the centre of the field, to clamp the preparation in this
position. A high-power lens is then turned into position, and
lowered by the coarse adjustment to a short distance above its
focal distance ; it is now carefully screwed down by the fine
adjustment, the eye being kept at the tube meanwhile. The
shadow of the edge will be first recognised, and then the bacteria
must be carefully looked for. Often a dry lens is sufficient, but
for some purposes the oil immersion is required. If the bacteria
are small and motile, a beginner may have great difficulty in
seeing them, and it is well to practise at first on some large non-
motile form, such as anthrax. In fluid preparations the natural
appearance of bacteria may be studied, and their rate of growth
determined. The great use of such preparations, however, is to
find whether or not the bacteria are motile, and for determining
this point it is advisable to use either broth or agar cultures not
more than twenty-four hours old. In the latter case a small
fragment of growth is broken down in broth or in sterile wrater.
Sometimes it is an advantage to colour the solution in which
the hanging-drop is made up with a minute quantity of an
aniline dye, say a small crystal of gentian violet to 100 c.c. of
bouillon. Such a degree of dilution wdll not have any effect on
the vitality of the bacteria. Ordinarily, living bacteria will not
take up a stain, but even though they do not, the contrast
between the unstained bacteria and the tinted fluid will enable
the observer more easily to recognise them. In determining
whether or not a bacterium is motile, great difficulty is often
experienced in distinguishing between true motion and Brownian
movement, especially if the organism be small. The essential
criterion to be fulfilled is that the bacteria shall be moving in
all directions, the observation of individuals lying close together
starting to move in opposite directions being important. The
observation of hanging-drop preparations must be correlated
with the results of staining for the presence of flagella which, so
far as is known, are present in all motile forms.

Within recent years the method of observing living micro-

1 In bacteriological work it is essential that cover-glasses of No. 1 thickness
(i.e. '14 mm. thick) should be used, as those of greater thickness are not
suitable for a jViuch lens.

FILM PREPARATIONS 93

organism* by oblique illumination has been much practised, and
a number of substage condensers are in the market, by means of
which this is effected. The general principle involved in these
instruments is to stop out the rays passing directly towards the
tube of the microscope, and to arrange for light being thrown
obliquely on bacteria mounted in a drop of fluid between a slide
and cover-glass. The bacteria disperse these rays in all direc-
tions, and some passing up through the lens are focussed by it*
The organisms thus appear as brightly illumined objects on a
dark background. The method has been employed for bacteria
in general, and especially for the demonstration of the spirochcbte
pallida in secretions as a means of diagnosis. Generally speak-
ing, the internal structure of the organisms under observation is
well brought out.

2. Film Preparations.— (a) Dry Method. — This is the most
extensively applicable method of microscopically examining
bacteria. Fluids containing bacteria, such as blood, pus,
scrapings of organs, can be thus investigated, as also cultures
in fluid and solid media. The first requisite is a perfectly clean
cover-glass. Many methods are recommended for obtaining
.such. The test of this being accomplished is that, when the
drop of fluid containing the bacteria is placed upon the glass, it
can be uniformly spread with the platinum needle all over the
surface without showing any tendency to retract into droplets.
The best method is that recommended by Van Ermengem. The
cover-glasses are placed for some time in a mixture of con-
centrated sulphuric acid 6 parts, potassium bichromate 6 parts,
water 100 parts, then washed thoroughly in water and stored in
absolute alcohol. For use, a cover-glass is either dried by
wiping with a clean duster or is simply allowed to dry. This
method will amply repay the trouble, and really saves time in
the end. A clean cover having been obtained, the film pre-
paration can now be made. If a fluid is to be examined a
loopful may be placed on the cover-glass, and either spread
out over the surface with the needle, or another clean cover
may be placed on the top of the first, the drop thus spread
out between them and the two then drawn apart. When
a culture on a solid medium is to be examined, a loopful of
distilled water is placed on the cover-glass, and a minute particle
of growth rubbed up in it and spread over the glass. The great
mistake made by beginners is to take too much of the growth.
The point of the straight needle should just touch the surface
<>t' the culture, and when this is rubbed up in the droplet of
water and the film dried, there should be an opaque cloud just

94 MICROSCOPIC METHODS

visible on the cover-glass. When the film has been spread, it
must next be dried by being waved backwards and forwards at
arm's-length above a Bunsen flame. The film must then be fixed
on the glass by being passed three or four times slowly through
the flame. In doing this a good plan is to hold the cover-glass

between the right forefinger
and thumb ; if the fingers just
escape being burned no harm
will accrue to the bacteria in

. FIG. 40. — Cornet's forceps for holding ',. p, r ,-L- i

cover-glasses. ^n making films 01 a thick

fluid such as pus, it is best to

spread it out on one cover with the needle. The result will be
a film of irregular thickness, but sufficiently thin at many parts
for proper examination. Scrapings of organs may be smeared
directly on the cover-glasses.

In the case of blood, a fairly large drop should be allowed to
spread itself between two clean cover-glasses, which are then to
be slipped apart, and being held between the forefinger and
thumb are to be dried by a rapid to-and-fro movement in the
air. A film prepared in this way may be too thick at one edge,
but at the other is beautifully thin. If it is desired to preserve
the red blood corpuscles in such a film it may be fixed by one
of the following methods : by being placed (a) in a hot-air
chamber at 1 20° C. for half an hour ; (b) in a mixture of equal
parts of alcohol and ether for half an hour, then washed and
dried ; (c) in formol-alcohol (Gulland) (formalin 1 part, absolute
alcohol 9 parts) for five minutes, then washed and dried ; or (d)
in a saturated solution of corrosive sublimate for two or three
minutes, then washed well in running water and dried. (Fig. 69
shows a film prepared by the last method.) In using the
Romanowsky stains no previous fixation is necessary (vide infra).
In the case of urine, the specimen must be allowed to stand, and
films made from any deposit which occurs ; or, what is still
better, the urine is centrifugalised, and films made from the
deposit which forms. After dried films are thus made from
urine it is an advantage to place a drop of distilled water on the
film and heat gently to dissolve the deposit of salts ; then w^ash
in water and dry. In this way a much clearer picture is
obtained when the preparation is stained.

Within recent years it has become common to make blood
films on ordinary microscopic slides instead of upon cover-
glasses. Here the slides must be clean. This can be effected by
washing thoroughly first with weak alkali and then with water

FILM PREPAKATIONS 95

and storing in alcohol. For use, a slide is taken from the
alcohol and the fluid adhering to it set on fire and allowed to
burn off, a dry clean slide being thus obtained. To make a film
on such, a small drop of blood is placed near one end, the edge
of a second clean slide is lowered through the drop on to the
surface of the glass on which the blood has been placed. This
second slide is held at an angle to the first, and the droplet of
blood by capillarity spreads itself in the angle between the two
slides. The edge of the second slide is then stroked along the
surface of the first slide, and the blood is spread out in a film
whose thickness can be regulated by the angle formed by the
second slide. Large-sized films can thus be obtained, and when
these are stained they are often examined without any cover-
glass being placed upon them. A drop of cedar oil is placed on
the preparation, and after use this can be removed by the careful
application of xylol.

Films dried and fixed by the above methods are now ready to
be stained by the methods to be described below.

(/;) Wet- M'tlioil. — If it is desired to examine the fine
histological structure of the cells of a discharge as well as to
investigate the bacteria present, it is advisable to substitute
" wet " films for the " dried " films, the preparation of which has
been described. The nuclear structure, mitotic figures, etc., are
by this method well preserved, whereas these are considerably
distorted in dried films. The initial stages in the preparation
of wet films are the same as above, but instead of being dried
in air they are placed, while still wet, film downwards in
the fixative. The following are some of the best fixing
methods : —

(a) A saturated solution of perohloride of mercury in '75 per cent,
sodium chloride ; fix for live minutes. Then place the films for half an
hour, with occasional gentle shaking, in '75 per cent, sodium chloride
solution to wash out the corrosive sublimate ; they are thereafter washed
in successive strengths of methylated spirit. After this treatment the
films are stained and treated as it' they were sections.

(b) Formol-alcohol — formalin 1 part, absolute alcohol 9. Fix films
for three minutes ; then wash well in methylated spirit. This is an
excellent and very rapid method.

(c) Another excellent method of fixing has been devised by Gulland.
The fixing solution has the composition — absolute alcohol 25 c.c., pure
ether 25 c.c., alcoholic solution of corrosive sublimate (2 grins, in 10 c.c.
o( alcohol)' about 5 drops. The films are placed in this solution for five
minutes or longer. They are then washed well in water, and are ready
for staining. A contrast stain can be applied at the same time as the
fixing solution, by saturating the 25 c.c. of alcohol with eosin before
mixing. Thereafter the bacteria, etc., may be stained with methylene-
blue or otlu-i >t;dn, as described below. This method has the advantage

96 MICROSCOPIC METHODS

over (a) that, as a small amount of corrosive sublimate is used, less
washing is necessary to remove it from the preparation, and deposits are
less liable to occur.

3. Examination of Bacteria in Tissues. — For the examina-
tion of bacteria in the tissues, the latter must be fixed and
hardened, in preparation for being cut with a microtome.
Fixation consists in so treating a tissue that it shall permanently
maintain, as far as possible, the condition it was in when re-
moved from the body. Hardening consists in giving such a
fixed tissue sufficient consistence to enable a thin section of it
to be cut. A tissue, after being hardened, may be cut in a
freezing microtome (e.g. Cathcart's or one of the newer instru-
ments in which the freezing is accomplished by compressed
carbonic acid gas), but far finer results can be obtained by
embedding the tissue in solid paraffin and cutting with some of
the more delicate microtomes of which, for pathological purposes,
the small Cambridge rocker is by far the best. For bacterio-
logical purposes embedding in celloidin is not advisable, as the
celloidin takes on the aniline dyes which are used for staining
bacteria, and is apt thus to spoil the preparation, and besides,
thinner sections can be obtained by the paraffin method.

The Fixation and Hardening of Tissues. — The following are
amongst the best methods for bacteriological purposes : —

(a) Absolute alcohol may be used for the double purpose of fixing and
hardening. If the piece of tissue is not more than £ inch in thickness, it
is sufficient to keep it in this reagent for a few hours. If the pieces
are thicker a longer exposure is necessary, and in such cases it is better
to change the alcohol at the end of the first twenty-four hours. The
tissue must be tough without being hard, and the necessary consistence,
as estimated by feeling with the fingers, can only be judged of after
some experience. If the tissues are not to be cut at once, they may be
preserved in 50 per cent, spirit.

(b) Formol-alcohol — formalin 1, absolute alcohol 9. Fix for not more
than twenty- four hours ; then place in absolute alcohol if the tissue is
to be embedded at once, in 50 per cent, spirit if it is to be kept for some
time. For small pieces of tissue fixation for twelve hours or even less is
sufficient. The method is a rapid and very satisfactory one.

(c) Corrosive sublimate is an excellent fixing agent. It is best used
as a saturated solution in '75 per cent, sodium chloride solution. Dis-
solve the sublimate in the salt solution by heat ; the separation of
crystals on cooling shows that the solution is saturated. For small
pieces of tissue | inch in thickness, twelve hours' immersion is 'sufficient.
If the pieces are larger, twenty-four hours is necessary. They should
then be tied up in a piece of gauze, and placed in a stream of running
water for from twelve to twenty-four hours, according to the size of the
pieces, to wash out the excess of sublimate. They are then placed for
twenty-four hours in each of the following strengths of methylated

THE CUTTING OF SECTIONS 97

spirit (free from naphtha1) : 30 per cent., 60 per cent., and 90 per cent.
Finally they are placed in absolute alcohol for twenty- four hours and
are then ready to be prepared for cutting.

If the tissue is very small, as in the case of minute pieces removed
for diagnosis, the stages may be all compressed into twenty-four hours.
In fact, after fixation in corrosive the tissue may be transferred directly
to absolute alcohol, the perchloride of mercury being removed after the
sections are cut, as will be afterwards described.

(d) Methylated spirit.— Small pieces of tissue may be placed in
methylated spirit, which is to be changed after the first day. In from
six to seven days they will be hardened. If the pieces are large, a
longer time is necessary.

The Cutting of Sections. — 1. By Means of the Freezing
Microtome. — Pieces of tissue hardened by any of the above
methods must have all the alcohol removed from them by wash-
ing in running water for twenty-four hours. They are then
placed for from twelve to twenty-four hours (according to their
size) in a thick syrupy solution containing two parts of gum
arabic and one part of sugar. They are then cut on a freezing
microtome and placed for a few hours in a bowl of water so that
the gum and syrup may dissolve out. They are then stained, or
they may be stored in methylated spirit.

2. Embedding and Cutting in Solid Paraffin. — This method
gives by far the finest results, and should always be adopted
when practicable. The principle is the impregnation of the
tissue with paraffin in the melted state. This paraffin when it
solidifies gives support to all the -tissue elements. The method
involves that, after hardening, the tissue shall be thoroughly
dehydrated, and then thoroughly permeated by some solvent
of paraffin which will expel the dehydrating fluid and prepare
for the entrance of the paraffin. The solvents most in use are
chloroform, cedar oil, xylol, and turpentine ; of these, chloroform
and cedar oil are the best, the former being preferred, as it per-
meates the tissue more rapidly. The more gradually the tissues
are changed from reagent to reagent in the processes to be gone
through, the more successful is the result. A necessity of the
process is an oven with hot- water jacket, in which the paraffin
can be kept at a constant temperature just above its melting-
point, a gas regulator, e.g. Reichert's, being of course necessary.
The tissues occurring in pathological work have a tendency to

1 In Britain ordinary commercial methylated spirit has mineral naphtha added
to it to discourage its being used as a beverage. The naphtha being insoluble
in water a milky fluid results from the dilution of the spirit. By law, chemists
can only sell 8 ounces of pure spirit at a time. Most pathological laboratories
are, however, permitted by the Excise to buy "industrial spirit," which
contains only one-nineteenth of naphtha.

98 MICROSCOPIC METHODS

become brittle if overheated, and therefore the best results are
obtained by using paraffin melting at a somewhat low tempera-
ture. We have used for some years a mixture of one part of
paraffin, melting at 48°, and two parts of paraffin melting at
54° C. This mixture has a melting-point between 52° and
53° C., and it serves all ordinary purposes well. An excellent
quality of paraffin is that known as the " Cambridge paraffin,"
but many scientific-instrument makers supply paraffins which, for
ordinary purposes, are quite as good, and much cheaper. The
successive steps in the process of paraffin embedding are as
follows : l —

1. Pieces of tissue, however hardened, are placed in fresh absolute
alcohol for twenty-four hours in order to their complete dehydration.

2. Transfer now to a mixture of equal parts of absolute alcohol and
chloroform for twenty- four hours.

3. Transfer to pure chloroform for twenty-four hours or longer. At
the end of this time the tissues should sink or float heavily.

4. Transfer now to a mixture of equal parts of chloroform and paraffin
and place on the top of the oven for from twelve to twenty-four hours.
If the temperature there is not sufficient to keep the mixture melted
then they must be put inside.

5. Place in pure melted paraffin in the oven for twenty-four hours.
For holding the paraffin containing the tissues, small tin dishes such as are
used by pastry-cooks will be found very suitable. There must be a
considerable excess of paraffin over the bulk of tissue present, otherwise
sufficient chloroform will be present to vitiate the final result and not
give the perfectly hard block obtained with pure paraffin. With ex-
perience, the persistence of the . slightest trace of chloroform can be
recognised by smell.

In the case of very small pieces of tissue the time given for each stage
may be much shortened, and where haste is desirable Nos. 2 and 4 may
be omitted. Otherwise it is better to carry out the process as described.

6. Cast the tissues in blocks of paraffin as follows : Pairs of L-shaped
pieces of metal made for the purpose by instrument makers must be at
hand. By laying two of these together on a glass plate, a rectangular
trough is formed. This is filled with melted paraffin taken from a stock
in a separate dish. In it is immersed the piece of tissue, which is lifted
out of its pure paraffin bath with heated forceps. The direction in
which it is to be cut must be noted before the paraffin becomes opaque.
When the paraffin has begun to set, the 'glass plate and trough have
cold water run over them. When the block is cold, the metal L's are
broken off, and, its edges having been pared, it is stored in a pill-box.

The Cutting of Paraffin Sections. — Sections must be cut as

1 While the method given is sufficient for ordinary purposes, a more elaborate
technique is necessary if it is desired that no changes shall take place in the
tissue. Thus after fixation the tissue must be taken up to absolute alcohol
through successive dilutions of spirit, but differing from each other by more
than 10 per cent. Again, when alcohol has been replaced by chloroform the
latter must be saturated with chips of paraffin, first at even temperature, then
at 37° C., and must be kept at 55° C. as short a time as possible.

THE CUTTING OF SECTIONS 99

thin as possible, the Cambridge rocking microtome being, on
the whole, most suitable. They should not exceed 8 /* in thick-
ness, and ought, if possible, to be about 4 /x. For their mani-
pulation it is best to have two needles on handles, two camel's-
liair brushes on handles, and a needle with a rectangle of stiff

Fin. 41. — Needle with square of paper on end for manipulating
paraffin sections.

writing paper fixed on it as in the diagram (Fig. 41). When
cut, sections are floated on the surface of a beaker of water kept
at a temperature about 10° C. below the melting-point of the
paraffin. On the surface of the warm water they become
perfectly flat.

Fixation on Ordinary Slides, (a) Gulland's Method.— A supply of
slides well cleaned being at hand, one of them is thrust obliquely into
the water below the section, a corner of the section is fixed on it with a
needle and the slide withdrawn. The surplus of water being wiped off
witli a cloth, the slide is placed on a support, with the section down-
wauls, and allowed to remain on the top of the paraffin oven or in a
bacteriological incubator for from twelve to twenty- four hours. It will
then be sufficiently fixed on the slide to withstand all the manipulations
necessary during staining and mounting.

(V) Fixation by Mann's Method. — This has the advantage of being
more rapid than the previous one. A solution of albumin is prepared
by mixing the white of a fresh egg with ten parts of distilled water and
filtering. Slides are made perfectly clean with alcohol. One is dipped
into the solution and its edge is then drawn over one surface of another
slide so as to leave on it a thin film of albumin. This is repeated with
the others. As each is thus coated it is leant, with the film down-
wards, on a ledge till dry, and then the slides are stored in a wide
stoppered jar till needea. The floating out is performed as before.
The albuminised side of the slide is easily recognised by the fact that
if it is breathed on, the breath does not condense on it. The great
advantage of this method is that the section is fixed after twenty to
thirty minutes' drying at 37° C. If the tissue has been hardened in any
of the bichromate solutions and embedded in paraffin, this or some
corresponding method of fixing the sections on the slide must be used.

Preparation of Paraffin Sections for Staining. — Before stain-
ing, the paraffin must be removed from the section. This is
best done by dropping on xylol out of a drop bottle. When the
paraffin is dissolved out, the superfluous xylol is wiped off with
a cloth and a little absolute alcohol dropped on. When the
xylol is removed, the superfluous alcohol is wiped off and a
little 50 per cent, methylated spirit dropped on. During these

100 MICROSCOPIC METHODS

procedures sections must on no account be allowed to dry.
The sections are now ready to be stained. Deposits of crystals
of corrosive sublimate often occur in sections which have been
fixed by this reagent. These can be removed by placing the
sections, before staining, for a few minutes in equal parts of
Gram's iodine solution (p. 106) and water, and then washing out
the iodine with methylated spirit.

To save repetition, we shall in treating of stains suppose that,
with paraffin sections, the above preliminary steps have already
been taken, and further, that sections cut by a freezing microtome
are also in spirit and water.

Dehydration and Clearing. — It is convenient, first of all, to
indicate the final steps to be taken after a specimen is stained.
Dry films after being stained are washed in water, dried and
mounted in xylol balsam; wet films and sections must be
dehydrated, cleared, and then mounted in xylol balsam.

Dehydration is most commonly effected with absolute alcohol.
Alcohol, however, sometimes decolorises the stained organisms
more than is desirable, and therefore Weigert devised the
following method of dehydrating and clearing by aniline oil.
which, though it may decolorise somewhat, does not do so to the
same extent as alcohol. As much as possible of the water being
removed, the section placed on a slide is partially dried by
draining with fine blotting-paper. Some aniline oil is placed on
the section and the slide moved to and fro. The section is
dehydrated and becomes clear. The process may be accelerated
by heating gently. The preparation is then treated with a
mixture of two parts of aniline oil and one part of xylol, and
then with xylol alone, after which it is mounted in xylol balsam.
Balsam as ordinarily supplied has often an acid reaction, and
preparations stained with aniline dyes are apt to fade when
mounted in it. It is accordingly a great advantage to use the
acid-free balsam supplied by Griibler. Paraffin sections can
usually be dehydrated and cleared by the mixture of aniline oil
and xylol alone.

Sections stained for bacteria should always be cleared, at
least finally, in xylol, as it dissolves out aniline dyes less readily
than such clearing reagents as clove oil, etc. Xylol, however,
requires the previous dehydration to have been more complete
than clove oil, which will clear a section readily when the
dehydration has been only partially effected by, say, methylated
spirit. If a little decolorisation of a section is still required
before mounting, clove oil may be used to commence the
clearing, the process being finished with xylol. With a little

THE STAINING OF BACTERIA 101

experience the process of decolorisatioii can be judged of •>/
observing the appearances under a low objective.

THE STAINING OF BACTERIA.

Staining Principles. — To speak generally, the protoplasm of
bacteria reacts to stains in a manner similar to the nuclear
cliiomatin, though sometimes more and sometimes less actively.
The bacterial stains par excellence are the basic aniline dyes.
These dyes are more or less complicated compounds derived
from the coal-tar product aniline (C6H5 . NH2). Many of them
have the constitution of salts. Such compounds are divided
into two groups according as the staining action depends on the
basic or the acid portion of the molecule. Thus the acetate of
rosaniline derives its staining action from the rosaniline. It
is therefore called a basic aniline dye. On the other hand,
ammonium picrate owes its action to the picric acid part of the
molecule. It is therefore termed an acid aniline dye. These
t \\ « » groups have affinities for different parts of the .animal cell.
The basic stains have a special affinity for the nuclear chromatin,
the acid for the protoplasm and various formed elements. Thus
it is that the former — the basic aniline dyes — are especially the
bacterial stains.

The number of basic aniline stains is very large. The following are
the most commonly used : —

i Stains. — Methyl- violet, R-5R (synonyms: Hoffmann's violet,
dahlia).

' ;<'iitian-violet (synonyms : benzyl-violet, Pyoktanin).
Crystal violet.

Blue Stains. — Methylene-blue T (synonym : phenylene-blue).
Victoria-blue.
Thionin-blue.
Red Stains.— Basic fuchsin (synonyms : basic rubin, magenta).

Safraiiin (synonyms : fuchsia, Girofle).

Broivn Stain. — Bismarck - brown (synonyms : vesuvin, phenylene-
brown).

It is of the greatest importance that the stains used by the
bacteriologist should be good, and therefore it is advisable to
obtain those prepared by Griibler of Leipzig.

Of the stains specified, the violets and reds are the most
intense in action, especially the former. It is thus easy in using
tin-in to overstain a specimen. Of the blues, methylene-blue
probably gives the best differentiation of structure, and it is

1 This is to lie distiiiguishc'd from methyl-blue, which is a different com-
pound.

102

MICROSCOPIC METHODS

difficult to overstain with it. Thionin-blue also gives good dif-
ferentiation and does not readily overstain. Its tone is deeper
than that of methylene-blue, and it approaches the violets in tint.
Bismarck-brown is a weak stain, but is useful for some purposes.
Formerly it was much used in photomicrographic work, as it was
less actinic than the other stains. It is not, however, needed
now, on account of the improved sensitiveness of plates.

It is most convenient to keep saturated alcoholic solutions
of the stains made up, and for use to filter a little into about
ten times its bulk of distilled water in a watch-glass. A solution
of good body is thus obtained. Most
bacteria (except those of tubercle,
leprosy, and a few others) will stain in
a short time in such a fluid. Watery
solutions may also be made up, e.g. a
saturated watery solution of methylene-
blue or a 1 per cent, solution of
gentian-violet. Stains must always be
filtered before use ; otherwise there
may be deposited on the preparation
granules which it is impossible to wash
off. The violet stains in solution in
water have a great tendency to decom-
pose. Only small quantities should
therefore be prepared at a time.

The Staining of Cover-glass Films.
— Films are made from cultures as
described above. The cover-glass may
be floated on the surface of the stain
in a watch-glass, or the cover-glass held
in Cornet's forceps with film side
uppermost may have as much stain
poured on it as it will hold. When
the preparation has been exposed for
the requisite time, usually a few
minutes, it is well washed in tap water

in a bowl, or with distilled water with such a simple siphon
arrangement as that figured (Fig. 42). The figure explains itself.
When the film has been washed the surplus of water is drawn off
with a piece of filter-paper, the preparation is carefully dried
high over a flame, a drop of xylol balsam is applied, and the
cover-glass mounted on a slide. It is sometimes advantageous
to examine films in a drop of water in place of balsam. The
films can be subsequently dried and mounted permanently.

FIG. 42. — Siphon wash-
bottle for distilled water
used in washing prepara-
tions.

MORDANTS AND DECOLORISING AGENTS 103

Films of fluids from the body (blood, pus, etc.) can be
U'Mierally stained in the same way, and this is often quite
sufficient for diagnostic purposes. The blue dyes are here
preferable, as they do not readily overstain. In the case of such
fluids, if the histological elements also claim attention it is best
first to stain the cellular protoplasm with 1-2 per cent, watery
solution of eosiu (which is an acid dye), and then to use a blue
which will stain the bacteria and the nuclei of the cells. The
Romano wsky stains (vide p. 113) are here most useful, as by these
the preparations are fixed as well as stained. Fixation by heat,
which is apt to injure delicate cellular structures, is thus avoided.
In the case of films made from urine, where there is little or
no albuminous matter present, the bacteria may be imperfectly
fixed on the slide, and are thus apt to be washed off. In such
a case it is well to modify the staining method. A drop of
stain is placed on a slide, and the cover-glass, film-side down,
lowered upon it. After the lapse of the time necessary for
staining, a drop of water is placed at one side 'of the cover-glass
and a little piece of filter-paper at the other side. The result is
that the stain is sucked out by the filter-paper. By adding
fresh drops of water and using fresh pieces of filter-paper, the
specimen is washed without any violent application of water,
and the bacteria are not displaced.

For the general staining of films a saturated watery solution
of methylene-blue will be found to be the best stain to com-
mence with, the Gram method (vide infra) is also used, and
subsequently any special stains which may appear advisable.

The Use of Mordants and Decolorising Agents. — In films
of blood and pus, and still more so in sections of tissues, if the
above methods are used, the tissue elements may be stained to
such an extent as to quite obscure the bacteria. Hence many
methods have been devised in which the general principle may
be said to be (a) the use of substances which, while increasing
the staining power, tend to fix the stain in the bacteria, and
(6) the subsequent treatment by substances which decolorise the
overstained tissues to a greater or less extent, while they leave
the bacteria coloured. The staining capacity of a solution may
be increased —

(a) By the addition of substances such as carbolic acid,
aniline oil, or metallic salts.

(b) By the addition of alkalies, such as caustic potash or
ammonium carbonate, in weak solution.

(d) By long duration of the staining process.

104 MICROSCOPIC METHODS

As decolorising agents we use chiefly mineral acids (hydro-
chloric, nitric, sulphuric), vegetable acids (especially acetic acid),
alcohol (either methylated spirit or absolute alcohol), or a com-
bination of spirit and acid, e.g. methylated spirit with a drop or
two of hydrochloric acid added, also various oils, e.g. aniline,
clove, etc. In most cases about thirty drops of acetic acid in
a bowl of water will be sufficient to remove the excess of stain
from over-stained films and sections. More of the acid may, of
course, be added if necessary.

Hot water also decolorises to a certain extent ; over-stained
films can often be readily decolorised by placing a drop of water
on the film and heating gently over a flame.

When preparations have been sufficiently decolorised by an
acid, they should be well washed in tap water, or in distilled
water with a little lithium carbonate added.

Different organisms take up and retain the stains with various
degrees of intensity, and thus duration of staining and decoloris-
ing must be modified accordingly. We sometimes have to deal
with bacteria which show a special tendency to be decolorised.
This tendency can be obviated by adding a little of the stain to
the alcohol, or aniline oil, employed in dehydration. In the
latter case a little of the stain is rubbed down in the oil. The
mixture is allowed to stand. After a little time a clear layer
forms on the top with stain in solution, and this can be drawn
off with a pipette.

When methylene-blue, methyl-violet, or gentian-violet is used,
the stain can, after the proper degree of decolorisation has been
reached, be fixed in the tissues by treating for a minute with
ammonium molybdate (2J per cent, in water).

The Formulae of some of the more commonly used Stain Combinations.

1. Lojfler's Methylene-blue.

Saturated solution of methylene-blue in alcohol . . . 30 o.c.
Solution of potassium hydrate in distilled water (1-10,000) . 100 ,,

(This dilute solution maybe conveniently made by adding 1 c.c. of a
1 per cent, solution to 99 c.c. of water. )

Sections may be stained in this mixture for from a quarter of an hour
to several hours. They do not readily overstain. The tissue containing
the bacteria is then decolorised if necessary with ^-1 per cent, acetic acid,
till it is a pale blue-green. The section is washed in water, rapidly
dehydrated with alcohol or aniline oil, cleared in xylol, and mounted.

The tissue may be contrast-stained with eosin. If this is desired,
after decolorisation wash with water, place for a few seconds in 1 per
cent, solution of eosin in absolute alcohol, rapidly complete dehydration
with pure absolute alcohol, and proceed as before.

GRAM'S STAIN 105

Films may l>e stained with Loflier's blue by five minutes' exposure or
longer in the cold. They usually do not require decolorisation, as the
tissue elements are not overstained.

2. Kiihnts Methylene-bluc.

Methylene-blue .... l'5grm.
Absolute alcohol .... 10 c.c.
Carbolic acid solution (1-20) . . 100 ,,

Stain and decolorise as with Loffler's blue, or decolorise with very weak
hydrochloric acid (a few drops in a bowl of water).

3. Carbol'Thionin-blue.—^lakc up a stock solution consisting of 1
gramme of thionin-blue dissolved in 100 c.c. carbolic acid solution (1-40).
For use, dilute one volume with three of water, and filter. Stain sections
for five minutes or upwards. Wash very thoroughly with water, other-
wise a deposit of crystals may occur in the subsequent stages. Decolorise
with very weak acetic acid. A few drops of the acid added to a bowl
of water are quite sufficient. Wash again thoroughly with water.
Dehydrate with absolute alcohol. Thionin-blue stains more deeply
than methylene-blue, and gives equally good differentiation. It is very
suitable for staining typhoid and glanders bacilli in sections. Cover-
glass preparations stained by this method do not usually require
decolorisation. As a contrast stain, 1 per cent, watery solution of eosin
may be used before staining with the thionin.

4. Gentian-violet in Aniline Oil Water. — Two solutions have here to
be made up. (a) Aniline oil water. Add about 5 c.c. aniline oil to
100 c.c. distilled water in a flask, and shake violently till as much as
possible of the oil has dissolved. Filter and keep in a covered bottle
to prevent access of light. (6) Make a saturated solution of gentian-
vi'.l.-t in alcohol. When the stain is to be used, 1 part of (b) is added
to 10 parts of (a), and the mixture filtered. The mixture should be made
not more than twenty- four hours before use. Stain sections for a few
minutes ; then decolorise with methylated spirit. Sometimes it is
advantageous to add to the methylated spirit a little hydrochloric acid
(2-3 minims to 100 c.c.). This staining solution is not so much used
by itself as in < J ram's method, which is presently to be described.

5. Carbol-Gentian-Violet. — 1 part of saturated alcoholic solution of
gentian-violet is mixed with 10 parts of 5 per cent, solution of carbolic
acid. It is used as No. 4.

6. Carbol-Fuchsin (see p. 1C8). — This is a very powerful stain, and,
when used in the undiluted condition, £-1 minute's staining is usually
sufficient. It is better, however, to dilute with from five to ten times
its volume of water and stain for a few minutes. In this form it has a
very wide application. Methylated spirit with or without a few drops
of acetic acia is the most convenient decolorising agent. Then dehydrate
thoroughly, clear, and mount.

Gram's Method and its Modifications. — In the methods
alivady described, the tissues, and more especially the nuclei,
ivt;iiu some stain when decolorisation has reached the }>oint to
which it can safely go without the bacteria themselves bein^
affected. In the method of Gram, now to be detailed, this does
not occur, for the stain can here be removed completely from

106 MICROSCOPIC METHODS

the ordinary tissues, and left only in the bacteria. All kinds
of bacteria, however, do not retain the stain in this method,
and therefore in the systematic description of any species it is
customary to state whether it is, or is not, stained by Gram's
method — by this is meant, as will be understood from what
has been said, whether the particular organism retains the
colour after the latter has been completely removed from the
tissues. It must, however, be remarked that some tissue
elements may retain the stain as firmly as any bacteria, e.g.
keratinised epithelium, calcified particles, the granules of mast
cells, and sometimes altered red blood corpuscles, etc.

In Gram's method the essential feature is the treating of the
tissue, after staining, with a solution of iodine. This solution
is spoken of as Gram's solution, and has the following com-
position : —

Iodine . . . . . 1 part.
Potassium iodide ... 2 parts.

Distilled water .... 300 „

•

The following is the method : —

1. Stain in aniline oil gentian -violet or in carbol-gentian- violet (vide
supra), for about five minutes.

2. Without washing in water, now treat the section or film with
repeated doses of Gram's solution till its colour becomes a purplish
black, and allow the solution to net for one minute.

3. Again without washing with water, decolorise with absolute alcohol
or methylated spirit till the colour has almost entirely disappeared, the
tissues having only a faint violet tint. Tlie period of time tor which the
alcohol is allowed to act varies in different laboratories. The best period
is probably about three minutes.

4. Dehydrate completely, clear with xylol, and mount. In the case
of film preparations, the specimen is simply washed in water, dried, and
mounted.

In stage (3) the process of decolorisation is more satisfactorily per-
formed by using clove oil after sufficient dehydration with alcohol, the
clove oil being afterwards removed by xylol.

As a contrast stain for the tissues, carmalum or lithia carmine is used
before staining with gentian-violet (1). As a contrast stain for other
bacteria which are decolorised by Gram's method, carbol-fuchsin diluted
with twenty volumes of water or a saturated watery solution of Bismarck-
brown may be used before stage (4) ; the former should not be applied
for longer than a few seconds.

The following modifications of Gram's method may be given : —

1. Weigert's Modification. — The contrast staining of the tissues and
stages (1) and (2) are performed as above.

(3) After using the iodine solution the preparation is dried by blotting
and then decolorised by aniline-xylol (aniline-oil 2, xylol 1).

TUBERCLE STAINS 107

(4) Wash well in xylol, and mount in xylol balsam. Film preparations
after being washed in xylol may be dried, and thereafter dilute carbol-
fuchsin may be used to stain bacteria which have been decolorised.

This modification probably gives the most uniformly successful results.

2. Nicolle's Modification. — Carbol-gentian-violet is used as the stain.
Treatment with iodine is carried out as above, and decolorisation is
effected with a mixture of acetone (1 part) and alcohol (2 parts).

3. Kiihnes Modification. — (1) Stain for five minutes in a solution
made up of equal parts of saturated alcoholic solution of crystal-violet
(" Krystall- violet ") and 1 per cent, solution of ammonium carbonate.

(2) Wash in water.

(3) Place for two to three minutes in Gram's iodine solution, or in
the following modification by Kiihne : —

Iodine 2 parts.

Potassium iodide . . . . 4 ,,

Distilled water . . . . . 100 ,,

For use, dilute with water to make a sherry- coloured solution.

(4) Wash in water.

(5) Decolorise in a saturated alcoholic solution of fluorescein (a
saturated solution in methylated spirit does equally well).

(6) Dehydrate, clear, and mount.

There is great variability in the avidity with which organisms stained
by Gram retain the dye when washed with alcohol, and sometimes
difficulty is experienced in saying whether an organism does or does not
stain by this method.

Most bacteria are either frankly Gram-positive or frankly
Gram-negative, but cases occur when an organism, usually Gram-
positive or Gram-negative, tends when grown on certain media to
show an opposite tendency, and sometimes an organism is met
with in which the individuals in a film show slightly different
reactions to the Gram stains.

Stain for Tubercle and other Acid-fast Bacilli. — These
bacilli cannot be well stained with a simple watery solution of
a basic aniline dye. This fact can easily be tested by at-
tempting to stain a film of a tubercle culture with such a
solution ; with the Gram method, however, a partial staining
is sometimes effected. Such bacteria require a powerful stain
containing a mordant, and must be exposed to the stain for a
long time, or its action may be aided by a short application of
heat. When once stained, however, they resist decolorising
even with very powerful acids; they are therefore called "acid-
fast." The smegma bacillus also resists decolorising with
strong acids (p. 280), and a considerable number of other acid-
fast bacilli are now known (p. 278). Any combination of
gentian-violet or fuchsin with aniline oil or carbolic acid or

108 MICROSCOPIC METHODS

other mordant will stain the bacilli named, but the following
methods are most commonly used : —

Ziehl-Neelsen Carbol-Fuchsin Stain.

Basic fuchsin . . . 1 part.

Absolute alcohol . . . . 10 parts.
Solution of carbolic acid (1 : 20) . 100 „

1. Place the specimen in this fluid, and having heated it till steam
rises, allow it to remain there for five minutes, or allow it to remain in
the cold stain for from twelve to twenty-four hours. (Films and paraffin
sections are usually stained with hot stain, loose sections with cold ; in
hot stain the latter shrink.)

2. Decolorise with 20 per cent, solution of strong sulphuric acid, nitric
acid, or hydrochloric acid, in water. In this the tissues become yellow.

3. Wash well with water. The tissues will regain a faint pink tint.
If the colour is distinctly red, the decolorisation is insufficient, and the
specimen must be returned to the acid. As a matter of practice, it is
best to remove the preparation from the acid every few seconds and
wash in water, replacing the specimen in the acid and re-washing till
the proper pale pink tint is obtained. Then wash in alcohol for half a
minute, and replace in water.

4. Contrast stain with a saturated watery solution of methyleue-blue
for half a minute, or with saturated watery Bismarck-brown for from
two to three minutes.

5. Wash well with water. In the case of films, dry and mount. In
the case of sections, dehydrate, clear, and mount.

Fraenkel's Modification of the Ziehl-Neelsen Stain.

Here the process is shortened by using a mixture containing
both the decolorising agent and the contrast stain.

The sections or films are stained with the carbol-fuchsin as above
described, and then placed in the following solution : —

Distilled water ...... 50 parts.

Absolute alcohol 30 ,,

Nitric acid . . . . . 20 ,,

Methylene-blue in crystals to saturation.

They are treated with this till the red colour has quite disappeared and
been replaced by blue. The subsequent stages are the same as in No. 5,
supra.

Leprosy bacilli are stained in the same way, but are rather
more easily decolorised than tubercle bacilli, and it is better
to use only 5 per cent, sulphuric acid in decolorising.

In the case of specimens stained either by the original Ziehl-
Neelsen method, or by Fraenkel's modification, the tubercle or

STAINING OF SPORES AND CAPSULES 109

leprosy bacilli ought to be bright red, and the tissue blue or
In-own, according to the contrast stain used. Other bacteria
which may be present are also coloured with the contrast stain.

The Staining of Spores. — If bacilli containing spores are
stained with a watery solution of a basic aniline dye the spores
remain unstained. The spores either take up the stain less
readily than the protoplasm of the bacilli, or they have a resisting
envelope which prevents the stain penetrating to the protoplasm.
Like the tubercle bacilli, when once stained they retain the
colour with considerable tenacity. The following is the simplest
method for staining spores : —

1. Stain cover-glass films as for tubercle bacilli.

2. Decolorise with 1 per cent, sulphuric acid in water or with methy-
lated spirit. This removes the stain from the bacilli.

3. Wash in water.

4. Stain with saturated watery methylene-blue for half a minute.

5. Wash in water, dry, and mount in balsam.

The result is that the spores are stained red, the protoplasm of the
bacilli blue.

The spores of some organisms lose the stain more readily than those
of others, and for some, methylated spirit is a sufficiently strong
decolorising agent for use. If sulphuric acid stronger than 1 per cent,
is used, the spores of many bacilli are readily decolorised.

Mt,1t<-rx M'tluul.— The following method, recommended by Moller, is
much more satisfactory than the previous. Before being stained, the
films are placed in chloroform for two minutes, and then in a 5 per cent,
solution of chromic acid for ^-2 minutes, the preparation being well
washed after each reagent. Thereafter they are stained and decolorised
as above.

The Staining of Capsules. — The two following methods may
be recommended in the case of capsulated bacteria : —

(a) Welch's Method. — This depends on the fact that in many cases
the capsules can be fixed with glacial acetic acid.

Films when still wet are placed in this acid for a few seconds.

The superfluous acid is removed with filter-paper, and the preparation
is treated with gentian-violet in aniline oil water repeatedly till all the
acetic acid is removed.

Then wash with 1-2 per cent, solution of sodium chloride, and examine
in the same solution.

The capsule appears as a pale violet halo around the deeply stained
bacterium.

(b) Hiss's Method.— The staining solution consists of 1 part of a
saturated alcoholic solution of fuchsin or gentian-violet and 19 parts of
distilled water. A few drops of the stain are placed on a film, previously
<lri«'(l and fixed by heat, and the preparation is steamed for a few seconds over
a flame. The staining solution is washed off with a 20 per cent, solution
of copper sulphate, the preparation (without being washed in water) is dried
between filter-papers, and when thoroughly dry is mounted in balsam. The
capsules of pneumococci growing in a fluid serum medium can be readily

110 MICROSCOPIC METHODS

demonstrated by this method ; in the case of solid cultures films should
be made without any diluent, or a drop of fluid serum should be used.
The method is easily applied, and gives excellent results.
(c) Richard Muir's Method (as recently modified).

1. The film containing the bacteria must be very thin. It is dried
and stained in filtered carbol-fuchsin for half a minute, the preparation
being gently heated.

2. Wash slightly with spirit and then well in water.

3. Place in following mordant for a few seconds : —

Saturated solution of corrosive sublimate . . .2 parts.
Tannic acid solution — 20 per cent. . . . . 2 , ,
Saturated solution of potash alum . . . . 5 , ,

4. Wash well in water.

5. Treat with methylated spirit for about a minute.
The preparation has a pale reddish appearance.

6. Wash well in water.

7. Counterstain with watery solution of ordinary methyl ene-blue for
half a minute.

8. Dehydrate in alcohol, clear in xylol, and mount in balsam.

The bacteria are a deep crimson, and the capsules of a blue tint. The
capsules of bacteria in certain culture media may be demonstrated by
this method.

The Staining of Flagella. — The staining of the flagella of
bacteria is the most difficult of all bacteriological procedures,
and it requires considerable practice to ensure that good results
shall be obtained. Many methods have been introduced, of
which the two following are the most satisfactory : —

Preparation of Films. — In all the methods of staining
flagella, young cultures on agar should be used, say a culture
incubated for from ten to eighteen hours at 37° C. A very
small portion of the growth is taken on the point of a platinum
needle, and carefully mixed in a little water in a watch-glass ;
the amount should be such as to produce scarcely any turbidity
in the water. A film is then made by placing a drop on a
clean cover-glass and carefully spreading it out with the needle.
It is allowed to dry in the air, and is then passed twice or thrice
through a flame, care being taken not to over-heat it. The
cover-glasses used should always be cleaned in the mixture of
sulphuric acid and potassium bichromate described on page 93.

1. PitfieWs Method as modified by Richard Muir.

Prepare the following solutions : —
A. The Mordant.

Tannic acid, 10 per cent, watery solution, filtered . 10 c.c.

Corrosive sublimate, saturated watery solution . 5 ,,

Alum, saturated watery solution . . . 5 ,,

Carbol-fuchsin (vide p. 108) 5 ,,

STAINING OF FLAGELLA 111

Mix thoroughly. A precipitate forms, which must be allowed to
deposit, either by centrifugalising or simply by allowing to stand.
Remove the clear fluid with a pipette, and transfer to a clean bottle.
The mordant keeps well for one or two weeks.

B. The Stain.

Alum, saturated watery solution . . . . 10 c.c.
Gentian- violet, saturated alcoholic solution . . 2 ,,

The stain should not be more than two or three days old when used.
It may be substituted in the mordant in place of the carbol-fuchsin.

The film having been prepared as above described, pour over it as
much of the mordant as the cover-glass will hold. Heat gently over a
Hume till steam begins to rise, allow to steam for about a minute, and
then wash well in a stream of running water for about two minutes.
Then dry carefully over the flame, and when thoroughly dry pour on
some of the stain. Heat as before, allowing to steam for about a minute,
wash well in water, dry, and mount in a drop of xylol balsam.

This method has yielded the best results in our hands.

2. Van Ermengeni's Method for Staining Flagella.

The films are prepared as above described. Three solutions are here
necessary : —

Solution A. (Bain fixatcur)—

Osmir acid, 2 per cent, solution ... 1 part.
Tannin, 10-25 per cent, solution ... 2 parts.

Place the films in this for one hour at room temperature, or heat over
a flame till steam rises and keep in the hot stain for five minutes.
Wash with distilled water, then with absolute alcohol for three to four
minutes, and again in distilled water, and treat with

Solution B. (Bain sensibilisaleur] —

•5 per cent, solution of nitrate of silver in distilled water. Allow
films to be in this a few seconds. Then without washing transfer to

Solution C. (Bain reducteur et reinforfateur) —

Gallic acid ....... 5 grms.

Tannin ........ 3 ,,

Fused potassium acetate 10 ,,

Distilled water 350 c.c.

Keep in this for a few seconds. Then treat again with Solution B till
the preparation begins to turn black. Wash, dry, and mount.

It is better, as Mervyn Gordon recommends, to leave the specimen in
B for two minutes, and then to transfer to C for one and a half to two
minutes, and not to transfer again to B. It will also be found an
advantage to use a fresh supply of C for each preparation, a small
quantity being sufficient. The beginner will find the typhoid bacillus
or the bacillus coli communis very suitable organisms to stain by this
method.

112 MICROSCOPIC METHODS

Although the results obtained by this method are sometimes excellent,
they vary considerably. Frequently both the organisms and flagella
appear of abnormal thickness. This is due to the fact that the process
on which the method depends is a precipitation rather than a true
staining. The pictures on the whole are less faithful than in the first
method.

Staining of Spirochsetes in Sections. — The following im-
pregnation methods have been applied for this purpose by
Levaditi, and give excellent results : —

(a) Levaditi's Original Method.

(1) The tissues, which ought to be in thin slices, about 1 mm. in
thickness, are best fixed in 10 per cent, formalin solution for twenty-four
hours.

(2) They are washed for an hour in water, and then brought into 96 per
cent, alcohol for twenty-four hours.

(3) They are then placed in 1 -5 per cent, solution of nitrate of silver in
a dark bottle, and are kept in an incubator at 37° C. for three days.

(4) They are washed in water for about twenty minutes, and are
thereafter placed in the following mixture, namely : —

Pyrogallic acid, 4 parts.

Formalin, 5 parts.

Distilled water up to 100 parts.

They are kept in this mixture in a dark bottle for forty-eight hours at
room temperature.

(5) They are then washed in water for a few minutes, taken through
increasing strengths of alcohol, and embedded in paraffin in the usual
way. The sections ought to be as thin as possible. In satisfactory
preparations the spirochsetes appear of an almost black colour against the
pale yellow background of the tissues. The latter can be contrast-
stained by weak carbol- fuchsin or by toluidin blue.

(b) Levaditi's Newer Pyridin Method.

(1) The tissues are fixed in formalin as in the previous method, are
hardened in alcohol for twelve to sixteen hours, and then washed in water.

(2) They are then impregnated with a 1 per cent, solution of silver
nitrate, to which 10 per cent, of pyridin puriss. is added at the time of
use. The tissues are placed in the solution in a well-stoppered bottle, and
are kept for two to three hours at room temperature and four to six hours
at about 50° C. They are thereafter washed quickly in 10 per cent,
pyridin solution.

(3) Reduction is then carried out in the following mixture, namely, a
4 per cent, solution of pyrogallic acid to which are added, at the time of
use, 10 per cent, pure acetone and 15 per cent, pyridin.

(4) The tissues are then put through alcohol and xylol, and embedded
in paraffin. The sections can be stained with toluidin blue or Unna's
polychrome blue.

(For the staining of spirochoetes in films, see p. 115.)

THE ROMANOWSKY STAINS 113

The Romanowsky Stains. — Within recent years the numerous
mollifications of the Romanowsky stain have been extensively
used. The dye concerned is the compound which is formed
when watery solutions of medicinal methylene-blue and water-
soluble eosin are brought together. This compound is insoluble
in water but soluble in alcohol — the alcohol employed being
methyl alcohol.- The stain was originally used by Romanowsky
for the malarial parasite, and its special quality is that it
imparts to certain elements, such as the chromatin of this
oruiiiiHin, ;i reddish-purple hue. This was at first thought to be
simply due to the combination of the methylene-blue and the
eosin, but it is now recognised that certain changes, such as
occur in methylene-blue solutions with age, are necessary. In
the modern formula; these changes are brought about by
treatment with alkalies, especially alkaline carbonates, as was
first practised by Unna in the preparation of his polychrome
methylene-blue. The stains in use thus contain a mixture of
methylene-blue and its derivatives in combination with eosin ;
the differences in these bodies and the different proportions in
which they occur in individual stains account for the different
effects produced on the various constituents of a cell. The
underlying chemical reactions are complicated and as yet not
fully understood. Thus it is not certainly known to what partic-
ular new body the reddish hue produced in chromatin is due,
but the active constituent may be methyl-violet or methyl-azure
or thionin, all of which result from the 'action of alkali on
methylene-blue. The stains are much used in staining blood-
lilms (in which the characters of both nucleus and cytoplasm
in leucocytes are beautifully brought out), in staining bacteria
in tissues or exudates, the malaria parasite, trypanosomes, the
pathogenic spirochajtes (such as the spirochaete pallida), and
protozoa generally.

The following are the chief formulae in use : —

1. Jcnner's Stain. — This is an excellent blood stain, hut is not so good
for the study of parasites as the others to be mentioned. In its
preparation n«> alkali is used. It is made by mixing equal parts of (a)
a 1'2 to 1*25 per cent, solution of Griibler's water-soluble eosin (yellow
shade) in distilled water and (b) 1 per cent. Griibler's medicinal methy-
leue-blue (also a watery solution). The mixture is allowed to stand
twenty- four hours, is filtered, and the residue is dried at 55° C. ; the
powder is shaken up in distilled water, filtered, washed with distilled
wat«-r, and dried. Of the powder, '5 grin, is dissolved in 100 c.c. Merck's
methyl alcohol. For use a few drops are placed on the dried uniixed
liliu for one to three minutes, the dye is poured off, and the pre pa ration
wa.-Oied with distilled water till it presents a pink colour; it is then
dried between (liter-paper and mounted in xylol balsam.

114 MICROSCOPIC METHODS

2. Leishman's Stain. — The following solutions are prepared : (a) to
a 1 per cent, solution of medicinal methylene-blue is added "5 per cent,
sodium carbonate ; the mixture is kept at 65° C. for twelve hours, and
then for ten days at room temperature ('25 per cent, formalin may be
added as a preservative) ; (&) 1-1000 solution of eosin, extra B.A., in
distilled water. Equal volumes of the two solutions are mixed and
allowed to stand for six to twelve hours with occasional stirring, the
precipitate is collected, filtered, washed with distilled water, and dried.
For use, '15 per cent, is dissolved in Merck's methyl alcohol ("for
analysis, acetone free") as follows: The powder is placed in a clean
mortar, a little of the alcohol is added and well rubbed up with a
pestle ; the undissolved powder is allowed to settle and the fluid
decanted into a dry bottle ; the process is repeated with fresh fractions
of the solvent till practically all the stain is dissolved, and the bottle
is well stoppered. The stain will keep for a long period. For the
staining of films a few drops of the stain are placed on the unfixed
preparation for fifteen to thirty seconds so as to cover it with a
shallow layer (the stain may be conveniently spread over the film
with a glass rod), and the film is tilted to and fro so as to prevent
drying. This treatment efficiently fixes the film by the action of the
methyl alcohol. About double the quantity of distilled water is now
dropped on the film, and the stain and diluent are quickly mixed with
the rod. Five minutes are now allowed for staining, and the stain is
then gently washed off with distilled water. A little of the water is
kept on the film for half a minute to intensify the colour contrasts in
the various cells. For certain special structures, such as Schuffner's dots
or Maurer's dots in the malarial parasite, a longer staining (up to one
hour) may be necessary, and in any case it is well to practise being able
to control the depth of the staining effect by observation with a low-
power objective. If a preparation is to be stained for a long time it
must be kept covered, and if in such cases a granular deposit is formed
this may be got rid of by a quick wash with absolute alcohol. If in blood
films the red corpuscles appear bluish instead of pink, the colour may
be restored by washing the film with acetic acid, 1-1500. The film is
dried between filter-paper and mounted.

For staining sections a little modification is necessary. A paraffin
section is taken into distilled water as usual, the excess of water is drained
off, and a mixture of one part of stain and two parts of distilled water
is placed on it. The stain is allowed to act for five to ten minutes till the
tissue appears a deep Oxford blue ; it is then decolorised with 1-1500
acetic acid — the effect being watched under a low-power lens. The blue
begins to come out, and the process is allowed to go on till only the
nuclei remain blue. The section is then washed with distilled water,
rapidly dehydrated with alcohol, cleared, and mounted. If, as some-
times happens, the eosin tint be too well marked, it can be lightened
by the action of 1-7000 solution of caustic soda, this being washed off
whenever the desired colour has been attained.

In certain cases, e.g. for the staining of old films or of trypanosomes
or Leishmanife in sections, Leishrnan recommends an initial treat-
ment of the preparation with serum. This modification is described in
Appendix E.

3. J. II. Wright's Stain. — In this modification 1 per cent, methylene-
blue (BX or Ehrlich's rectified) and ^ per cent, sodium carbonate (both
in water) are mixed and placed in a Koch's steriliser for an hour. When

THE ROMANOWSKY STAINS 115

the fluid is cold, 1-1000 solution of extra B. A. eosiii is added till the mix-
ture becomes purplish and a finely granular black precipitate appears in
suspension (about 500 c.c. eosin to 100 c.c. methylene-blue solution are
required) ; the precipitate is filtered off and dried without being washed.
A saturated solution of this is made in the pure methyl alcohol ; this is
filtered and diluted by adding to 80 c.c. of the saturated solution 20 c.c.
of methyl alcohol. The application of the stain is almost the same as
with Leishman's. A few drops are placed on the preparation for a
minute for fixation ; water is then dropped on till a green iridescent
scum appears on the top of the fluid, and staining goes on for about
two minutes ; the stain is then washed off with distilled water, and
a little is allowed to remain on the film till differentiation is com-
plete ; the preparation is carefully dried with filter - paper, and
mounted.

4. Giemsa's Stain. — Giemsa believes that the reddish-blue hue
characteristic of the Romanowsky stain is due to the formation of
methyl-azure, and he has prepared this by a method of his own under
the name "Azur I." From this, by the addition of equal parts of
medicinal methylene-blue, he prepares what he calls "Azur II.," and
from this again by the addition of eosin he prepares "Azur II. -eosin."
The latest formula for the finished stain is as follows : Azur II. -eosin,
3gr. ; Azur II., 8 gr. ; glycerin (Merck, chemically pure), 250 gr. ; methyl
alcohol (Kahll.aum, I.), 250 gr. This stain has been extensively used
for demonstrating the spirochsete pallida, but it can be used for any
other purpose to which the Romanowsky stains are applicable. For the
spirochiete the following are Giemsa's directions : —

(1) Fix films in absolute alcohol for fifteen to twenty minutes, dry
with filter-paper. (2) Dilute stain with distilled water — one drop of
stain to 1 c.c. water (the mixture being well shaken). (Sometimes the
water is made alkaline by the addition of one drop of 1 per cent, potassium
carbonate to 10 c.c. water.) (3) Stain for fifteen minutes. (4) Wash in
brisk stream of distilled water. (5) Drain with filter-paper, dry, and
mount in Canada balsam.

With regard to the Jenner and Giemsa stains it is best to obtain the
solutions from Griibler ready for use ; the powder for Leishman's stain
may be obtained from the same source, and the solution made up by
the worker himself. Cabot states that Wright's stain can be obtained
from the Harvard Co-operative Society, Boylston Street, Boston,
U.S.A.

Neisser's Stain.— (a) The following is the original method introduced
by Neisser as an aid to the diagnosis of the diphtheria bacillus. Two
solutions are used as follows : (a) I grm. methylene-blue (Griibler) is
dissolved in 20 c.c. of 96 per cent, alcohol, and to the solution are added
950 c.c. of distilled water and 50 c.c. of glacial acetic acid ; (ft) 2 grins.
Bismarck-brown (vesuvin) dissolved in a litre of distilled water. Films
are stained in (a) for 1-3 seconds or a little longer, washed in water,
stained for 3-5 seconds in (6), dried, and mounted. The protoplasm of
the diphtheria bacillus is stained a faint brown colour, the granules a blue
colour. Neisser considers that this reaction is characteristic of the
organism, provided that culture..* on Loffler's serum are used and examined
after 9-24 hours' incubation at 34° to 35° C. Satisfactory results are not
always obtained in the case of films pn-pan-d from membrane, etc., but
there is no doubt that here also the method is one of considerable
value.

116 MICROSCOPIC METHODS

(b) The following is Neisser's modified cresoidin method : —

1. Stain films for a few seconds in a mixture of solutions A and B,

two parts of the former to one of the latter.

A. Methylene-blue . . . . 1 part.
Absolute alcohol . '. . . 50 parts.
Glacial acetic acid . .. '". . 50 ,,
Distilled water .; . . . . 1000 „

B. Crystal-violet (Hochst) . ; ,• 1 part.
Absolute alcohol . . . 10 parts.
Distilled water -, . .. . 300 ,,

2. Wash in water, and

3. Stain in cresoidin solution (1 : 300) for a few seconds (the cresoidin

should be dissolved in warm water and the solution then
filtered).

4. Wash in water, dry, and mount.

Sabouraud's Method for Staining Trichophyta. — Remove the fat from
the hair or epithelial squames witli chloroform. Place in a test-tube
with 10 per cent, formol, and warm for two or three minutes till ebullition
commences. Wash well in distilled water, and stain for one minute in
Sahli's blue, which is made up as follows : —

Distilled water . . . . . .40 parts.

' Saturated watery methylene blue . . . 24 ,,
5 per cent, solution of borax in water . 16 ,,

Mix the constituents. Allow to stand for a day, and filter. After
staining, wash in water, dehydrate with absolute alcohol, clear in xylol,
and mount in balsam.

CHAPTEK IV.

METHODS OF EXAMINING THE PROPERTIES' OF
SERUM— PREPARATION OF VACCINES -
GENERAL BACTERIOLOGICAL DIAGNOSIS— IN-
OCULATION OF ANIMALS.

THE TESTING OF AGGLUTINATIVE AND SEDIMENTING
PROPERTIES OF SERUM.

Wright's Method of measuring Small Amounts of Fluids.—
It is convenient here to describe this method. In ordinary work
fine calibrated pipettes may be used for measuring small
quantities of fluids, but such pipettes are not always available,
and by Wright's technique if a Gower's 5 c.mm. haemocytometer
pipette be at hand any measurements may be undertaken — in
fact, once the pipette now to be described (see Fig. 43) is made
we are independent of other means of measurement. A piece
of quill tubing is drawn out to capillary dimensions, and the
extreme tip of it is heated in a peep name and then drawn out
till it is of the thickness of a hair, though still possessing a bore.
If the point be broken off this hair, and mercury be run into the
tube, the metal will be caught where the tube narrows and will
pass no further — in fact, though air will pass, mercury will not.
Into the wide end of this tube 5 c.mm. of mercury, measured
from a Gower's pipette, is run down till it will go no further.
A mark is made on the tube at the proximal end of the mercury,
which is now allowed to run out, and the tube is carefully cut
through at the mark. A piece of ordinary quill tubing is drawn
out and broken off just below where its narrowing has begun,
the hair end of the capillary tube is slipped through the broken-
off end, and the tube is fixed in position with wax as shown in
the figure. A rubber nipple placed on the end of the pipette
completes the apparatus. If by pressing the nipple the air be
expelled from the pipette, and the end dipped under mercury,
exactly 5 c.mm. will be taken up. Thus, when pressure on the

117

118

METHODS OF EXAMINING SERUM

nipple is relaxed, other tubes can be very readily calibrated by
the mercury being expelled into them, and its limits marked on
their bores.

For measuring equal parts of different fluids, the pipette
shown in Fig. 44, d, in connection with
agglutination is very useful.

Methods of testing for Simple Ag-
glutination.— By agglutination is meant
the aggregation into clumps of uniformly
disposed bacteria in a fluid ; by sedimenta-
tion the formation of a deposit composed
of such clumps when the fluid is allowed
to stand. Sedimentation is thus the
naked-eye evidence of agglutination. The
blood serum may acquire this clumping
power towards a particular organism under
certain conditions ; these being chiefly met
with when the individual is suffering from
the disease produced by the organism, or
has recovered from it, or when a certain
degree of immunity has been produced
artificially by injections of the organism.
The nature of this property will be dis-
cussed later. Here we shall only give the
technique by which the presence or absence

of the p.r°perty may be tested- There are

Casing of quill tubing; two chief methods, a microscopic and a
B, rubber nipple ; c, naked-eye, corresponding to the effects
Hilary0 *Ube o°f F5 mentioned above. In both, the essential
c.mm. capacity ; D to process is the bringing of the diluted
E, hair capillary. serum into contact with the bacteria

uniformly disposed in a fluid. In the

former this is done on a glass slide, and the result is watched
under the microscope ; the occurrence of the phenomenon is
shown by the aggregation of the bacteria into clumps, and if the
organism is motile this change ' is preceded or accompanied by
more or less complete loss of motility. In the latter method
the mixture is placed in an upright thin glass tube ; sedimenta-
tion is shown by the formation within a given time (say from two
hours at 37° C. to twenty-four hours at room temperature) of a
somewhat flocculent layer at the bottom, the fluid above being
clear. Two points should be attended to : (a) controls should
always be made with normal serum, and (6) the serum to be
tested should never be brought in the undiluted condition into

METHODS OF TESTING FOR AGGLUTINATION 119

contact with
following : —

the bacteria. The stages of procedure are the

1. Blood is conveniently obtained by pricking the lobe of the ear,
which should previously have been washed with a mixture of alcohol
and ether, and allowed to

dry. The blood is drawn /^
up into a Wright's blood-
capsule (Fig. 45) or into
the bulbous portion of a
<-apillary pipette, such as
in Fig. 44, a. (These pip-
ettes can be readily made
by drawing out quill glass-
tubing in a flame. It is
convenient always to have
several ready for use.)
The pipette is kept in the
upright position, one end
liein^ closed. For purposes
of transit, break off the
bulb at the constriction
and seal the ends. After
the . serum has separated
from the coagulum the
bulb is broken through
near its upper end, and tlie
serum removed by means
of another capillary pip-
ette. The serum is then
to be diluted.

2. The serum may be
diluted (a) by means of a
graduated pipette — either
a leucocytorneter pipette
(Fig. 44, 6) or some cor-
responding form. In this
way successive dilutions
of 1 : 10, 1 : 20, 1 : 100,
etc. can be rapidly made.
This is the best method.
(b) By means of a capillary
pipette with a mark on the
tube, the serum is drawn
up to the mark and then
blown out into a glass
capsule ; equal quantities
of bouillon are successively

measured in the same way, and added till the requisite dilution is
obtained, (c) By means of a platinum needle with a loop at the end
(Delepine's method). A loopful of serum is placed on a slide, and the
dfsiwl number of similar loopfuls of bouillon are separately placed
around on the slide. The drops are then mixed.

A very convenient and rapid method of combining the steps 1 and 2

FIG. 44. — Tubes used in testing agglutinating
and sediinenting properties of serum.

120 METHODS OF EXAMINING SERUM

is to draw a drop of blood up to the mark 1 or '5 on a leucocytometer
pipette, and draw the bouillon after it till the bulb is filled. A dilution
of 10 or 20 times is thus obtained. Then blow the mixture into a
U-shaped tube (Fig. 44, c), and ceutrifugalise or simply allow the red
corpuscles to separate by standing. (In this method, of course, the
dilution is really greater than if pure serum were used, and allowance
must therefore be made in comparing results.) The presence of red
corpuscles is no drawback in the case of the microscopic method, but when
sedimentation tubes are used the corpuscles should be separated first.

3. The bacteria to be tested should be taken from young cultures,
preferably not more than twenty-four hours old, incubated at 37° C.
They may be used either as a bouillon culture or as an emulsion made
by adding a small portion of an agar culture to bouillon or '8 per cent,
solution of sodium chloride. In tlie latter case the mass of bacteria on a
platinum loop should be gently broken down at the margin of the fluid in
a watch-glass. When a thick turbidity is thus obtained, any remaining
fragments should first be removed, and then the organisms should be uni-
formly mixed with the rest of the fluid. The bacterial emulsion ought to
have a faint but distinct turbidity. (When the exact degree of sedimenting
power of a serum is to be tested — expressed as the highest dilution in which
it produces complete sedimentation within twenty-tour hours — a standard
quantity (by weight) of bacteria must be added to a given quantity of
bouillon. This is not necessary for clinical diagnosis. )

4. To test microscopically, mix equal quantities (measured by a
marked capillary pipette) of the diluted serum and the bacterial
emulsion on a glass slide, cover with a cover-glass, and examine under
the microscope. The form of glass slide used for hang-drop cultures
(Fig. 27) will be found very suitable. The ultimate dilution of the
serum will, of course, be double the original dilution.

To observe sedimentation, mix equal parts of diluted serum and of
bacterial emulsion, and place in a thin glass tube — a simple tube with
closed end or a U-tube. Keep in upright position for twenty-four
hours. One of Wright's sedimentation tubes is shown in Fig. 44, d.
Diluted serum is drawn up to fill the space mn, a small quantity of air
is sucked up after it to separate it from the bacterial emulsion, which
is then drawn up in the same quantity ; the diluted serum will then
occupy the position Id. The fluids are then drawn several times up
into the bulb, and returned to the capillary tube so as to mix, and finally
blown carefully down close to the lower end, which is then sealed off.
The sediment collects at the lower extremity.

It is often important to observe not merely the fact that agglutination
occurs, but also the weakest concentration of the serum with which the
reaction can be obtained.

Measurement of Group Agglutinins. — In the case of certain
groups of allied organisms, — notably the b. coli and its allies,—
it has been found that when a serum clumps one member of the
group it frequently also clumps the allied forms. If the greatest
dilution with which agglutination is obtained be estimated, the
end-points for the different strains affected are usually found to
differ. The determination of the end-point is important, as the
disease condition from which the serum is derived is generally
caused by the organism which is clumped in highest dilution.

METHOD OF TESTING AGGLUTININS 121

Tn comparing the effect of a serum on different bacteria, the
sedimentation method is usually employed. A series of
emulsions of the clitic- rent bacteria to be tested is prepared by
scraping off the growth on an agar tube, and suspending in
bouillon. Each of these should approximately contain the same
number of bacteria per unit volume. This is attained by using
emulsions of equal opacity, as judged of by noting the point at
which transparency to some arbitrary standard such as a
particular type or set of parallel lines ceases. A given amount
of each emulsion is now mixed with different dilutions of the
serum to be tested, the mixtures are all made up to the same
volume, say 1 c.c., and the tubes placed at 37° C. for two or
three hours. The results are then read, the tubes are set aside
at room temperature for twenty-four hours, and read again;
usually the two readings correspond.

The Absorption Method of testing Agglutinins. — This
method is applied under circumstances similar to those of the
last, namely, when several agglutinins acting on allied organisms
are present in a serum. The principle is to remove all the
agglutinins acting on one organism, and to study the properties
of those which remain. In practice the method consists in
adding to the serum a mass of one of the bacteria of the group
under study (the organisms being scraped off an agar slope),
allowing the mixture to stand at 37° C. for two or three hours,
and then separating the bacteria with the centrifuge. The
supernatant clear fluid is now pipetted off, and its agglutinating
pro] Arties studied on the other members of the bacterial group
either by sedimentation or by the microscopic method. The
use of the method is to aid in differentiating which member of
a bacterial group is causally related to the condition from which
the serum is obtained, and an example of its application for this
purpose will be found in the chapter on typhoid fever (p. 375).
It has also been used by Park and Collins and by Bainbridge
for identifying strains of organisms of the typhoid-coli group.
Here the principle is that, when an unknown strain belonging
to such a bacterial group is under investigation, if its capacities
for absorbing agglutinins from a serum containing a mixture
of such are the same as those of an already recognised strain,
then the two are probably identical.

OPSONIC METHODS.

Method of measuring the Phagocytic Capacity of the
Leucocytes. — This was first done by Leishmau by a very

122 METHODS OF EXAMINING SERUM

simple method, as follows : A piece of quill tubing is drawn out
to a capillary diameter so as to make a pipette about 6 inches
long. The point is broken off, and a rubber nipple adjusted to
the wide end ; a mark is made with an oil pencil about three-
quarters of an inch above the orifice. Blood is drawn from the
finger up to the mark, then an air-bubble is allowed to pass in.
A thin emulsion of the bacterium to be tested having been pre-
pared, a quantity of this is also drawn up to the mark. The
two fluids are thoroughly mixed by being first blown out on
to a sterile slide and then being drawn back into the pipette
and expelled, — this being repeated several times. A cover-glass
is placed over the drop, and the slide is placed in the incubator
at 37° C. for fifteen minutes. The cover-glass is then slipped
off so as to make a film preparation, which in the case of
ordinary bacteria may be stained by Leishman's method. The
number of bacteria present in, say, fifty polymorphonuclear cells
successively examined is determined, and an average struck. The
method was first used for showing that in cases of staphylococcus
infection the average number of bacteria taken up was less than
in a control in which the same bacterial emulsion was exposed
to the blood of a healthy individual. In making such an
observation, drops from the two mixtures are placed on the same
slide under separate cover-glasses, and the preparation incubated.
One cover is then slipped to one end of the slide, and the other
to the other, — the two films being then stained as one.

Leishman's method gives what may be called the total phago-
cytic capacity of the blood, but according to Wright's view the
process of phagocytosis in blood outside the body is not a
simple one, and before a leucocyte takes up a' bacterium the
latter must be acted on in some way by substances present in
the serum, which Wright calls opsonins (see Immunity). The
technique by which the actions of these opsonins is studied
has been elaborated by Wright and his co-workers in connec-
tion with bacterial vaccines, especially in relation to infection
by the pyogenic cocci and the tubercle bacillus. This technique
involves (1) the preparation of the bacterial emulsion, (2) the pre-
paration of the leucocytes, (3) the preparation of samples of (a)
serum from a normal person, and (b) serum from the infected
person.

(1) Preparation of Bacterial Emulsion. — In the case of the
pyogenic cocci, a little of a twenty-four hour living culture off
a sloped agar tube is taken and rubbed up in a watch-glass with
•85 per cent, saline. The mixture is placed in a tube and centri-
fugalised, so as to deposit any masses of bacteria which may be

PREPARATION OF LEUCOCYTES 123

present. Only by experience can a knowledge be gained of
the amount of culture to be used in the first instance, but the
resultant emulsion usually should exhibit only the merest trace
of cloudiness to the naked eye. Wright states it will then con-
tain from 7000 to 10,000 million bacteria per c.cm. If too
strong an emulsion be used, the leucocytes may take up so many
organisms that these cannot be accurately enumerated. In the
case of the tubercle bacillus, as short a variety of the organism
as possible should be selected, and a mass of growth off a solid
medium is taken (bacilli in mass can be obtained in the market
from wholesale chemists) and is well washed with changes of
distilled water, drained on filter paper in a Petri dish, and
thoroughly rubbed up with a little 1'5 per cent, saline in an
agate mortar, so as to disintegrate the bacterial masses and get
an emulsion composed as far as possible of individual bacilli. It
is extremely difficult to obtain a good emulsion of tubercle bacilli,
i.e. one that shall consist as far as possible of separate bacilli
— on the one hand without clumps, and on the other without
portions of disintegrated bacilli. The rubbing-up in the mortar,
which usually occupies many hours, must be done very slowly
and gently with a very light pestle, and the manipulation must
be frequently controlled by microscopic observation. A thick
cream should be obtained, and this should be sterilised by
steaming for half an hour on three successive days. Before
sterilisation it is convenient to seal up the stock emulsion in
small quantities in a number of pieces of quill tubing, so that
in the subsequent procedures only small portions of the emulsion
are exposed to aerial contamination at one time. For actual
use, one of those tubes is opened, a little is withdrawn with a
sterile pipette, and a weak emulsion made in the same way as
with the staphylococcus, except that 1 '5 per cent, saline is used.
The stock tube may be sealed with wax and kept for use again.
A fresh emulsion ought to be made up for each day's work.

(2) Preparation of Leucocytes. — Here the observer uses his
own blood cells. A 1*5 per cent, solution of sodium citrate in
•85 per cent, sodium chloride is prepared. This is placed in a
glass tube 3 inches long, made by drawing out a piece of
half-inch tubing to a point, the tube being filled nearly to the
brim. A handkerchief being bound round the. finger, this is
now pricked, and the blood allowed to flow directly into the
fluid, to the bottom of which it sinks. The tube ought to be
inverted between the addition of every few drops of blood, so as
to bringc'the blood in contact with the citrate and^'prevent
coagulation. The equivalent of about ten to twenty drops of

124

METHODS OF EXAMINING SERUM

blood should be obtained. The diluted blood is then centri-
f ugalised, and when the corpuscles are separated the supernatant
fluid is removed, '85 per cent, saline is substituted, and the
centrifugalisation repeated. The fluid is again removed, care
being taken not to disturb the layer of white cells lying on the
top of the red corpuscles. This layer is then pipetted off into
a watch-glass or tube, and the leucocytes required are thus
obtained.

(3) Preparation of the' Sera. — The serum whose sensitising
effect on the bacteria it is desired to test is obtained by Wright
as follows : A " blood-capsule " is made by drawing a piece of
No. 3 quill tubing into the shape shown in Fig. 45, the part not
drawn out being about 1 inch in length. It is convenient to
make a number of these capsules at
one time, and to draw off their
extremities and seal them in the
flame. For use, the tips of both
extremities are broken off, the finger
is pricked, and blood allowed to pass
into the capsule through the bent
limb till the capsule is about half
full. The air remaining in the
capsule is rarefied by passing the
straight end through a flame and
then sealing it off. By this manipul-

FIG. 45. -Wright's blood-cap- ation the blood is sucked over the
sule, and method of filling bend into the straight part of the
same, tube, and the bent end is now also

sealed off or closed with wax. It is

well to shake the blood down towards the closed straight end,
care being taken to previously allow the glass to cool sufficiently.
The capsule is now hung by the bend on the edge of a centri-
fuge tube, and the serum separated by spinning the instrument.
In any particular case a capsule of serum from the infected
person and one from a normal individual are prepared.

The emulsion, corpuscles, and serum being thus prepared,
the next step is to mix them. This is done by taking a piece
of quill tubing and drawing it out to a capillary point so as to
make a pipette about 8 inches long; on the thick end of
this a rubber teat is fixed, and about 1 inch from the capillary
point a mark is made with an oil pencil. From the watch-glass
containing the separated leucocytes a portion is sucked up to
the mark, and then an air-bubble is allowed to pass in. A
similar portion of the serum is drawn up, and then another

PREPARATION OF THE SERA 125

air-bubble, and finally a similar portion of the bacterial
emulsion. The three droplets are carefully blown on to a slide,
and are thoroughly mixed with one another by being alternately
drawn up into the tube and expelled ten times. The mixture
is then drawn into the tube, and the end sealed off in the flame.
The rubber nipple is removed, and the tube placed in the
incubator at 37° for fifteen minutes. A slide is now prepared
by rubbing it once or twice with very fine emery paper (No. 000)
and thoroughly wiping it. This is a procedure adopted by
Wright to cause an evenly distributed film to be made. The
tube being removed from the incubator and the end broken off,
its contents are again mixed by expelling and drawing up into
the tube. A minute droplet is placed on the prepared slide,
and by means of the edge of the end of another slide a film
is made, which is then dried and is ready for staining. The
spreader should be slightly narrower than the slide on which
the film is made ; in this way the film has two definite edges —
a fact of importance, as the leucocytes are usually in greatest
abundance near these edges. Films containing staphylococci
are stained either by Leishman's stain (</.v.) or with carbol-
thionin blue. In the former case no fixation is necessary, in
the latter it i.s usual to fix in saturated perchloride of mercury
for one and a half minutes, wash in water, and then stain. With
tubercle films the following is the procedure : The film is fixed for
twoininutfs in | »erehloride of mercury, washed thoroughly, stained
with carbol-fuchsin as usual, decolorised with 2*5 per cent,
sulphuric acid, cleared with 4 per cent, acetic acid, counter-
stained with watery solution of methylene-blue, and dried.

In applying the technique two preparations are made, in both
of which the same emulsion and the same leucocytes are em-
ployed ; but in one the bacteria have been exposed to the serum
of the infected individual under observation, and in the other
to that of a normal person, — usually the observer himself, — or
better still, to a mixture of sera from several normal persons.
Each of these preparations is now examined microscopically with
a movable stage, the number of bacteria in the protoplasm of at
least fifty polymorphonucleated leucocytes is counted, and an
average per leucocyte struck (this is often called the " phagocytic
index ") ; the proportion which this average in the case of the
abnormal serum bears to the average in the preparation in which
the healthy serum was used constitutes the opsonic index — that
of healthy serum being reckoned as unity.

The reliability of the opsonic method, of course, depends on
whether or not the phagocytic activity of the cells counted

126 METHODS OF EXAMINING SERUM

represents the phagocytic activity of the cells in the preparation.
Considerable controversy has arisen on this point. The general
result may be said to be that where such organisms as the
pyogenic cocci are concerned, the ordinary opsonic technique
gives on the whole reliable results. In the case of the tubercle
bacillus, there is considerable difference of opinion. Generally
speaking, it may be said that indices varying between '8 and
1 *2 are to be reckoned as unity — that is to say, that no deduction
can be drawn from indices falling between these limits. In the
case of such organisms as those of the coli-typhoid group a.nd
cholera, which are susceptible to bacteriolytic influences in the
serum, it may be necessary to heat the sera of the patient and
observer for half an hour at 55° C. This destroys any com-
plement present and prevents bacteriolysis occurring. In the
case of the b. typhosus the virulence of the strain employed has
been shown to be an important factor.

Several modifications of Wright's technique have been
suggested. Thus Klien, instead of enumerating the bacteria
ingested, takes a series of dilutions of the serum and estimates
the dilution with which capacity for opsonising bacteria dis-
appears (or at any rate the dilution with which the phagocytic
index falls below '5). The content of the patient's serum and
of that of the observer may be thus compared, or the course of
an immunisation may be followed by making daily observations
of the content in opsonin. In another modification of Wright's
technique Simon compares not the numbers of bacteria ingested,
but the percentages of cells containing bacteria to those not
containing bacteria. This he calls the " percentage index," and
he states that the figure thus obtained corresponds very closely
to the ordinary opsonic index; he claims that the method
eliminates some of the errors which may arise in the use of the
ordinary technique if only a relatively small number of phago-
cyting cells, such as 50, be examined.

BACTERICIDAL METHODS— DEVIATION OF COMPLEMENT.

The Estimation of the Bactericidal Action of Serum. — This
may be carried out by various methods, of which those of
Neisser and Wechsberg and of Wright may be given as examples.
In the former, the effects of varying amounts of serum on the
same amounts of bacteria are observed by means of plate
cultures ; in the latter, the number of bacteria which can be com-
pletely killed off by a given quantity of serum is ascertained.
In carrying out experiments of this kind it is convenient to have

BACTERICIDAL METHODS 127

a number of small test-tubes sterilised and plugged with cotton-
wool. We can then make any required dilution of a young
bacterial culture in bouillon as follows : To each of a number
of tubes we add '9 c.c. of '8 per cent, solution of sodium chloride.
To the first tube (a) we add '1 c.c. of the bacterial culture, and
thoroughly shake up the mixture ; to the second (6) we add
•1 c.c. of the contents of (a), and shake up ; to the third tube
(c) we add '1 c.c. of the contents of (6), and so on. It is thus
evident that '1 c.c. of the contents of (a) will correspond to
'01 c.c., and '1 c.c. of (b) to '001 c.c. of the original culture ; any
minimi fraction can thus be readily obtained. In the making
of all mixtures of serum and bacteria it is essential that none of
the latter shall escape the action of the former, e.y. by remaining
on a part of the mixing vessel with which the serum does not
come in contact.

(a) Method of Neisser and Wechsbery. — A series of small
plugged sterile tubes is taken, and to each we add '5 c.c. of
•8 per cent, sodium chloride solution, and a given quantity, say
•5^5. c.c., of a young bouillon culture to be tested. To the
several tubes in series we then add varying amounts of the
fresh serum whose action is to be observed, e.y. '2 c.c., '1 c.c.,
•05 c.c., '025 c.c., etc. The contents of each tube are then
made up to 1 c.c. with salt solution, and a few drops of sterile
bouillon are added to each tube. The tubes are then well shaken
and placed in the incubator at 37° C. for three hours, to allow
the serum to act. (Of course several series of such tubes may be
prepared and placed in the incubator for varying periods of time ;
we can thus observe when the bactericidal effect reaches the
maximum.) At the end of the given period of time a small
quantity, say '05 c.c., of the contents of each tube is added to a
tube of melted agar (cooled to about 40° C.) ; each agar tube is
then shaken, and the contents are poured out into a sterile Petri
capsule. The other tubes are similarly treated, and the Petri
capsules are placed in the incubator for a suitable period of time.
The number of colonies in each can then be noted. Of
course gelatine can be substituted for the agar in the plates if
desired.

(b) Wright's Method. — A twenty-four hours1 bouillon culture is
used, and various dilutions with sterile bouillon are made according
to the method described on p. 58 : thus 5-, 10-, 20-, 50-, 100-,
1000-, etc., fold dilutions may be prepared. A small quantity,
.say 1 c.mm., of the fresh serum to be tested is mixed with an
equal amount of the bacterial culture, and the mixture is placed
in a small capillary tube which is sealed at the ends ; similar

128 METHODS OF EXAMINING SERUM

mixtures of equal parts of serum and of each of the dilutions of
culture are prepared and treated in the same way. The tubes
are then placed in the incubator for eighteen to twenty-four
hours at 37° C., and at the end of that time the contents of
each are tested as regards sterility by means of cultures. In
this way the greatest dilution in which the bacteria are com-
pletely killed off is ascertained. The number of bacteria in
the original culture per c.mm. can be counted by the method
given on p. 70, and thus the total number of bacteria killed
off by the quantity of serum used can readily be calculated.

As will afterwards (see chapter on Immunity) be described in
greater detail, when an animal is immunised against a particular
bacterium the bactericidal action of its serum may be greatly
increased, and this depends on the development of a particular sub-
stance called an immune-body, which is comparatively thermo-
stable and is not destroyed at 55° C. To analyse the bactericidal
properties of such a serum, it should in the first place be heated
in order to destroy the normal complement. Then to each of a
series of sterile tubes we add (a) a quantity of normal unheated
serum insufficient of itself to destroy the bacteria, (b) a given
amount of the bacterial culture, and (c) varying amounts of the
heated immune-serum — •]., '01, '001, etc. c.c. In this way we
can find the quantity of the immune-serum which gives the
maximum bactericidal action.

In some cases, however, when an animal is immunised against
a given bacterium, or when a patient is infected with the
organism, the serum may not have increased bactericidal action,
but nevertheless contains an immune-body which leads to the
absorption or fixation of complement. In other wrords, the
immune-body is a substance which, along with the corresponding
or homologous bacterium, binds complement (p. 130). In order,
however, to explain the methods by which the fixation of com-
plement may be demonstrated, we must first of all give some
facts with regard to hsemolytic sera.

Methods of Haemolytic Tests.— A hamiolytic serum is usually
prepared by injecting the red corpuscles of an animal into the
peritoneum of an animal of different species — the corpuscles of
the ox are most frequently used, and the rabbit is the most
suitable animal for injection. The corpuscles ought to be com-
pletely freed of serum by repeatedly washing them in sterile
salt solution, and centrifugalising. An injection of the corpuscles
of 5 c.c. of ox's blood, followed by two injections, each of 10 c.c.,
at intervals of eight days, will usually give an active serum. The
animal should be killed by bleeding it, aseptically as far as poss-

METHODS OF ELEMOLYTIC TESTS 129

ible, seven to ten days after the last injection ; the serum which
separates may be collected in suitable lengths of quill glass-
tubing drawn out at the ends, which are afterwards sealed in
the flame. To ensure sterility when tbe serum is to be kept
some time, it is advisable to heat it for an hour at 55° C. on
three successive days; we have always found that serum treated
in this way remains sterile. It is, of course, devoid of comple-
ment. The test amount of corpuscles is usually 1 c.c. of a
5 per cent, suspension of corpuscles in '8 per cent, sodium
chloride solution : that is, the corpuscles of 5 c.c. blood are
completely freed of serum by repeatedly washing in salt solution,
and then salt solution is added to make up 100 c.c. In any
investigation it is necessary to obtain the minimum hajmolytic
dose (M.H.D.) of the immune-body and of the complement to
be used. (It is to be noted that as complement does not
increase during immunisation, the haemolytic dose of the fresh
scrum will come far short of representing the amount of
immune-body present.) In testing the dose of immune-body,
the fresh serum to be used as complement must be devoid of
htemolytic action (in the present instance rabbit's serum will be
found suitable), and more than sufficient to produce lysis with
immune-body is added to each of a series of tubes. Varying
amounts of immune-body are added to the tubes, the contents
are shaken, made up to 1*5 c.c., and incubated for two hours.
The amount of lysis is then noted, and the tubes are placed in
a cool chamber till next morning, when a final reading is taken.
The smallest amount of immune-body which gives complete
lysis is, of course, the M.H.D. : sometimes this may be as low
as '001 c.c. for the test amount of corpuscles. When further
observations are to be continued on the same day, the reading
after incubation must be taken as the working standard. To
estimate the M.H.D. of complement, proceed in a corresponding
manner; to each of a series of tubes add several doses of
immune-body, and then to the several tubes different amounts
of complement. The activity of a serum as complement varies
considerably, and each sample must be separately tested.1 The
above will serve as an indication of the fundamental methods ;
for further details, special papers on the subject must be
consulted. Corpuscles treated with sufficient immune-body to

1 Complement is a substance which rapidly (often within twenty-four hours)
loses its strength when kept at room temperature. It can, however, be pre-
•-.•rved for a considerable time at or near its original strength if it be kept frozen.
Even if this be done, however, the strength of the complementary serum must
l>e titred at the commencement of every experiment in which it is employed.

130 METHODS OF EXAMINING SERUM

produce complete lysis on the addition of complement are
usually spoken of as sensitised corpuscles.

The Removal of Blood-Samples from Rabbits, etc. — In such work as
that just described, it is often convenient to watch the progress of an
immunisation procedure by removing a sample of blood without the
animal being killed. With proper care any amount of blood up to one-
third of that contained in the body can be removed from the ear vein of
a rabbit. The animal, which must not be flurried, is placed on a bench,
and its body kept warm by being covered with a cloth. The root of the
ear should be shaved over the marginal vein, the hairs on the edge of the
ear should also be clipped short. It is best to have the ear dry, as the
evaporation of a fluid causes contraction of the vessels. In a great deal
of hsemolytic work absolute sterility of the sample is not necessary, so
that washing the ear is not required. When sterile blood is desired, the
precautions detailed on p. 44 may be applied. A frosted incandescent
electric lamp, such as is used for microscopic illumination, is placed lighted
an inch or two from the ear. The left hand of the operator should cover
the animal's head in front of the ears, the thumb and index finger being
left free to compress the vein at the root of the ear. In this way not
only is the animal's eye protected from the glare of the lamp, but the
distance of the latter from the ear can be regulated so as to keep it at
what to the operator's hand is a pleasant warmth. In a minute or two
the ear vessels will dilate, and the vein, being compressed at the root, a
lateral opening is made with a bayonet-pointed surgical needle (the
triangular-pointed needles supplied with the Gowers-Haldane hsemo-
globinometer are also very suitable), and the blood allowed to drop into a
sterile test-tube. Usually waves of contraction of the ear vessels will be
observed to occur, the passing off of which must be waited for, and from
time to time the clot must be gently squeezed out of the opening in the
vein with the flat side of the needle, or it may be necessary slightly to
enlarge the opening. The blood should be allowed to clot completely,
and then, by means of a sterile platinum needle, the clot should be loosened
from the sides of the tube in order that it may freely contract. The tube
should be placed in the ice-chest till the following morning, when the
serum can be pipetted off with a sterile, nippled pipette.

Daily samples can thus be obtained from an animal. If care be taken
not to make ragged openings in the vein, often the simple removal
of the previous scab will be followed by a free blood flow.

A worker associated with one of us has shown that this method can be
applied in guinea-pigs, provided these be of fair size. Here successive
samples of 2 c.c. can be obtained from the ear veins.

Fixation of Complement or Complement Deviation. — From
the facts given above it follows that sensitised corpuscles, i.e.
corpuscles treated with immune-body, may be made to serve as an
indicator for the presence of complement. If an immune-body is
present in a serum heated at 55° C., the serum when added to the
corresponding bacterium leads to the fixation of complement, and
thus prevents haemolysis when the sensitised corpuscles are added.
If we represent the bacteria, or rather the receptors in the bacteria,
by X, the immune-body by anti-X, and the complement by C

THE SERUM DIAGNOSIS OF SYPHILIS 131

(normal serum, say of a guinea-pig), we may represent the method
of experiment by the following scheme : —

X + anti-X + C

+ sensitised corpuscles

(The vertical dotted line represents a period of incubation for
one and a half hours at 37° C.)

If lysis of the sensitised corpuscles does not occur after incuba-
tion at -37° C., then the complement has been fixed and an
immune-body has been shown to be present, provided that a
suitable control shows that the bacteria alone, without immune-
body, do not fix sufficient complement to interfere with lysis.

This method has now been extensively used for demonstrating
the presence of immune-bodies in the blood of patients suffering
from a particular bacterial infection. It has also been applied
to determine whether a suspected bacterium is really the cause
of a disease, for if the bacterium gives with the serum of the
patient deviation of complement, then there is a strong pre-
sumption that it is the infective agent (vide Immunity).

The Serum Diagnosis of Syphilis, Wassermann Reaction.—
Wassernmim, Neisser and Bruck, proceeding in accordance with
the facts established with regard to the deviation of complement,
tested whether a similar phenomenon might not be obtained in
the case of syphilis. For this purpose they mixed together a
watery extract of syphilitic liver, rich in spirochaetes (antigen),
and serum from a syphilitic case (supposed to contain anti-sub-
stances), and found that a relatively large amount of complement
was fixed. On the other hand, when the serum from a non-
syphilitic case was substituted for the syphilitic serum, little or
no fixation of complement occurred. The result was thus in
accordance with expectations on theoretical grounds. Marie
and Levaditi, however, found that an extract of normal guinea-
pig's liver along with syphilitic serum fixed complement, and
subsequent observations showed that extracts of other tissues are
also more or less efficient, as are also certain definite substances,
such as sodium oleate, sodium glycocholate, lecithin, mixtures
of such and especially mixtures of lecithin and cholesterin, etc.
Although abundant observations have established the validity
of the test as a means of diagnosis, the reaction which led to its
discovery is no longer sufficient to explain it, and the nature of
the reaction is not yet understood.

In order to carry out the test, we require (a) serum from the
suspected case, (6) an extract of liver or other organ, and
(c) the fresh serum of an animal to act as complement. The fol-
lowing are the details, arranged in two stages : —

132 METHODS OF EXAMINING SERUM

M
(Jr

V^iJ (t
(y^ v

w>*(
fid*** act

1. We add to a small test-tube —

(a) '05 c.c. of serum from the suspected case, heated for half
an hour at 55° C. to destroy the human complement, and '5 c.c.
of "8 per cent, salt solution ;

(b) "I c.c. of an alcoholic extract of guinea-pig's or ox's liver
(this can be prepared by extracting finely minced liver with four
volumes of alcohol for 3 to 4 days and then filtering) ;

A certain amount of guinea-pig's serum, usually '1 c.c., to
act as complement.

The mixture is then placed in the incubator for one and a half
hours, to allow fixation of complement to occur.

2. We then add to the tube 1 c.c. of a 5 per cent, suspension
of sensitised corpuscles (usually sheep's or ox's), i.e. corpuscles
to which there has been added a sufficient quantity of immune
serum to produce lysis on the addition of complement.

The mixture is then placed in the incubator for another hour.
If lysis of the corpuscles does not occur, the complement has
been fixed in the first stage by the mixture of serum and liver
extract. This is a positive result, and indicates the presence of
syphilis. If the corpuscles undergo lysis, all the complement
has not been fixed — the result is negative. When the amount
of serum to be tested is small, the amounts given may all be
proportionately reduced.

Such is the test as usually performed, and in this form it
usually gives satisfactory results. It is to be noted, however,
on the one hand, that the liver extract alone may fix a certain
amount of complement, rarely more than three doses, and, on
the other hand, that the hsemolytic value of fresh serum varies,
i.e. the amount of complement is not always indicated by the
volume of serum. It is accordingly better, and in a laboratory
this can be readily done, to estimate the hamiolytic dose of the
guinea-pig's serum, and to prepare a series of tubes, each con-
taining the same amounts of serum and of liver extract, but
with a different number of doses of complement in each tube. In
this way we can find the number of doses of complement deviated
in each case. As controls, the effect of the extract alone and of the
serum alone can be tested at the same time. With the amounts
of extract and serum mentioned, a positive result indicating the
presence of syphilis may be accepted when five or more doses of
complement are deviated in addition to the amount absorbed in
the controls. Some observers use the same amount of comple-
ment in each tube, but vary the amounts of suspected serum, and
in this way some idea of the deviating power of the serum is
obtained, but we consider that the method given is to be preferred.

WRIGHT'S METHOD OF COUNTING BACTERIA 133

THE PREPARATION OF VACCINES.

During recent years, in consequence of the work of Sir
Almroth Wright, the principle of treating bacterial disease by
vaccines has been very much developed. The general principle
is to inject into the infected individual an emulsion of dead
bacteria. In certain cases the bacteria are subjected to dis-
integrating processes before being used, but most frequently the
vaccines simply contain killed bacterial cells, and the preparation
is comparatively simple.

In the case of pyoyenic cocci, either bouillon cultures or a
growth off sloped agar emulsified in normal saline is taken and
killed by heat. The temperature employed should be the
minimum at which death occurs, say 65° C., applied for half an
hour. In the case of certain staphylococci, we have found,
however, that a higher temperature is necessary. After any
sterilisation procedure, tubes of agar must be inoculated from
the presumably dead vaccine, and incubated for twenty-four
hours in order to ascertain if the sterilisation has been effective.
As the dosage of a vaccine is of great importance, it is necessary
to count the bacteria present. This is done by one of the
methods given below. Appropriate doses (see Chapter VII.) are
then with all aseptic precautions measured by means of a sterile
graduated pipette, and placed, along with an equal volume of
•5 per cent, lysol, in little glass bulbs drawn out to a capillary
tube at one end. These when charged are sealed off, and for
use the sealed end is broken off, the contents are sucked up into
a sterile hypodermic needle, and injected fairly deeply into the
-kin, usually in the region of the flank.

In the case of the typhoid bacillut, organisms are used of such
virulence that a quarter of a twenty-four hours' old sloped agar
culture, when administered hypodermically, will kill a guinea-
pig of from 350 to 400 grams. Flasks of bouillon are inoculated
with such a culture for forty-two hours at 37° C. The bacteria are
then killed by the flask being put into a water bath at 62° C.
for fifteen minutes ; '5 per cent, lysol is added, and the bacteria
in the vaccine are counted. By such methods, vaccines against
any of the pyogenic cocci and against any members of the coli-
typhoid group can be made.

The vaccines used in tuberculosis, cholera, and plague will be
described in the chapters on these diseases.

Wright's Method of counting the Bacteria in Dead
Cultures. — In the making of vaccines it is, as indicated above,
necessary to know the total number of bacterial cells, whether

134 THE PREPARATION OF VACCINES

dead or living, present in a culture, for the dead as well as the
living contain the toxins which may stimulate the therapeutic
capacities of the body. The method consists in making a
mixture of blood (whose content in red blood corpuscles is
known) with the bacterial culture, and comparing the number of
bacteria with the number of corpuscles. The observer first
estimates the red cells in his blood ; a capillary pipette with a
rubber nipple and with a mark near its capillary extremity is
then taken, blood is sucked up to the mark, then an air-bubble,
and then an equal volume of the bacterial emulsion diluted
according to the empirical estimate the observer forms of its
strength. The blood and bacterial emulsion are then thoroughly
mixed by being drawn backwards and forwards in the wide
part of the pipette, a drop is blown out on to a slide, and a
blood film is spread which may be stained by Leishman's
method. The bacteria and blood corpuscles are now separately
enumerated in a series of fields in different parts of the
preparation. If a dilution has been taken in which a large
number of bacteria are present, an artificial field may be used,
made by drawing with the oil pencil a small square on a circular
cover-glass, and dropping the latter on to the diaphragm of the
microscope eye-piece. Suppose, now, that the observer's blood
contained 5,000,000 red cells per c.mm., that to the bacterial
emulsion three volumes of diluent had been added, and that in
the fields examined there were 500 red cells and 600 bacteria.
It is evident that in the undiluted culture for 500 red cells there
would have been 2400 bacteria. Now 500 : 2400 : ; 5,000,000 :
24,000,000, which last figure is the number of bacteria per
c.mm. of the emulsion.

It has been found in the case of certain bacteria, e.g. the
members of the coli-typhoid and cholera groups, that when an
emulsion of these is mixed with whole blood, the serum of the
latter may have a bacteriolytic or an agglutinating action on the
organisms, which interferes with the counting. To obviate the
inaccuracies or difficulties thus introduced, Harrison has modified
Wright's method by substituting, in a given quantity of blood,
normal saline for the serum. The method is as follows : — A
capillary pipette has a mark made upon it, to which blood is
sucked up and quickly expelled into a small tube containing a
little "75 per cent, sodium citrate solution ; any remaining blood is
washed out of the pipette with the same fluid. The tube is then
centrifuged to deposit the corpuscles, the supernatant fluid
carefully removed, and the corpuscles are washed by centrifuging
twice or thrice with normal saline, care being always taken not

WRIGHT'S METHOD OF COUNTING BACTERIA 135

to lose any of the corpuscles in the successive washings. After
the last washing the corpuscles are sucked up into the pipette,
and the n saline up to the mark which indicated the volume of
the original blood. Such a mixture is taken, and, to prevent
loss of corpuscles, the pipette and tubes are washed with a
definite number of equal volumes of broth or saline. Thus
there can be obtained in a watch-glass a mixture of, say, one
volume of corpuscles and saline, and two volumes of the diluting
fluid. To this mixture is now added an appropriate number of
volumes, again measured in the same pipette, of the bacterial
emulsion to be counted, the amount, of course, depending upon
a rough judgment which with experience can be made of the
probable numbers present. A drop of the mixture is put under
a cover-glass, and the numbers of corpuscles on the one hand and
of bacteria on the other present in a number of fields are
counted. It is not necessary to stain the bacteria, but in the
case of motile organisms it is recommended that they be
rendered motionless by using as a diluent saline to which formol
has been added in the proportion of two or three drops to 10 c.c.
If the number of red blood corpuscles in the observer's blood be
known, it is evident that the amount of blood corresponding to
a certain number of blood corpuscles in a microscopic field can
be calculated, and the number of bacteria present in the same
amount of the mixture will be the number corresponding to the
number of corpuscles. Thus it is now only necessary to allow
for the dilution to obtain the number of bacteria in the original
emulsion.

GENERAL BACTERIOLOGICAL DIAGNOSIS.

Under this heading we have to consider the general routine
which is to be observed by the bacteriologist when any material
is submitted to him for examination. The object of such
examination may be to determine whether any organisms are
present, and if so, what organisms ; or the bacteriologist may
simply be asked whether a particular organism is or is not
present. In any case, his inquiry must consist (1) of a micro-
scopic examination of the material submitted ; (2) of an attempt
to isolate the organisms present ; and (3) of the identification of
the organisms isolated. We must, however, before considering
these points, look at a matter often neglected by those who seek
a bacteriological opinion, namely, the proper methods of ob-
toiniii'i and transferring to t/ie bacteriologist the material u'hich
I i' /.s to be asked to examine. The general principles here are

136 GENERAL BACTERIOLOGICAL DIAGNOSIS

(1) that every precaution must be adopted to prevent the
material from being contaminated with extraneous organisms;

(2) that nothing be done which may kill any organisms which
may be proper to the inquiry ; and (3) that the bacteriologist
obtain the material as soon as possible after it has been removed
from its natural surroundings.

The sources of materials to be examined, even in patho-
logical bacteriology alone, are, of course, so varied that we can
but mention a few examples. It is, for instance, often necessary
to examine the contents of an abscess. Here the skin must be
carefully purified by the usual surgical methods • the knife used
for the incision is preferably to be sterilised by boiling ; the first
part of the pus which escapes is allowed to flow away (as it might
be spoilt by containing some of the antiseptics used in the
purification), and a little of what subsequently escapes allowed
to flow into a sterile test-tube. If test-tubes sterilised in a
laboratory are not at hand, an ordinary test-
tube may be quarter-filled with water and
vigorously boiled over a spirit-lamp. The tube
is then emptied and plugged with a plug of
cotton wool, the outside of which has been
singed in a flame. Small stoppered bottles
may be sterilised and used in the same way.
A discharge to be examined may be so small
in quantity as to make the procedure described
impracticable. It may be caught on a piece
of sterile plain gauze, or of plain absorbent
wool, which is then placed in a sterile vessel.
Wool or gauze used for this purpose, or for
swabbing out, say the throat, to obtain shreds
of suspicious matter, must have no antiseptic
impregnated in it, as the latter may kill the
bacteria present and make the obtaining of
cultures impossible.

Fluids from the body cavities, urine, etc.,

FIG. 46.— Test-tube mav j^ secured with sterile pipettes. To make

and pipette ar- J £ , . _ . . rf ,. .,,

ranged for obtain- one o* these, take 9 inches of ordinary quill

ing fluids contain- glass-tubing, draw out one end to a capillary

diameter, and place a little plug of cotton wool

in the other end. Insert this tube through

the cotton plug of an ordinary test-tube, and sterilise by heat.

To use it, remove test-tube plug with the quill tube in its centre,

suck up some of the fluid into the latter, and replace in its

former position in the test-tube (Fig. 46). Another method

KOUTINE EXAMINATION OF MATERIAL 137

very convenient for transport is to make two constrictions on
the glass tube at suitable distances, according to the amount of
fluid to be taken. The fluid is drawn up into the part between
the constrictions, but so as not to fill it completely. The tube
is then broken through at both constrictions, and the thin ends
an- sealed by heating in a flame.

Solid organs to be examined should, if possible, be obtained
whole. They may be treated in one of two ways. (1) The
surface over one part about an inch broad is seared with a
cautery heated to dull red heat. All superficial organisms are
thus killed. An incision is made in this seared zone with a
sterile scalpel, and small quantities of the juice are removed by
a platinum spud to make cover-glass preparations and plate
or smear cultures. (2) An alternative method is as follows : —
The surface is sterilised by soaking it well with 1 to 1000
corrosive sublimate for half an hour. It is then dried, and the
capsule of the organ is cut through with a sterile knife, the
incision being further deepened by tearing. In this way a
perfectly uncontaminated surface is obtained. Hints are often
obtained from the clinical history of the case as to what the
procedure ought to be in examination. Thus, as a matter of
practice, cultures of tubercle and often of glanders bacilli can
be easily obtained only by inoculation experiments. Typhoid
bacilli need hardly be looked for in the faeces after the first ten
days of the disease, and so on.

Eoutine Procedure in Bacteriological Examination of
Material. — In the case of a discharge regarding which nothing
is known, the following procedure should be adopted : —
(1) Several cover-glass preparations should be made. One ought
to be stained with saturated watery methylene-blue, one with
a stain containing a mordant such as Ziehl-Neelsen carbol-
fuchsin, one by Gram's method. (2) a. Gelatin plates should
be made and kept at room temperature ; b. a series of agar
plates or successive strokes on agar tubes (p. 60) should be made
and incubated at 37° C. Method b of course gives results
more quickly. In every case when an unknown disease is
being investigated, some of the material should be subjected to
methods suitable to the growth of anaerobic bacteria. If micro-
scopic investigation reveals the presence of bacteria, it is well
to keep the material in a cool place till next day, when, if no
growth has appeared in the incubated agar, some other culture
medium (e.y. blood serum or agar smeared with blood) may be
employed. If growth has taken place, say in the agar plates,
one with about two hundred or fewer colonies should be made

138 GENERAL BACTERIOLOGICAL DIAGNOSIS

the chief basis for research. In such a plate the first question
to be cleared up is : Do all the colonies present consist of the
same bacterium 1 The shape of the colony, its size, the appear-
ance of the margin, the graining of the substance, its colour,
etc., are all to be noted. One precaution is necessary, namely,
it must be noted whether the colony is on the surface of the
medium or in its substance, as colonies of the same bacterium
may exhibit differences according to their position. The
arrangement of the bacteria in a surface colony may be still
more minutely studied by means of impression preparations.
A cover-glass is carefully cleaned and sterilised by passing
quickly several times through a Bunsen flame. It is then placed
on the surface of the medium, and gently pressed down on the
colony. The edge is then raised by a sterile needle, it is seized
with forceps, dried high over the flame, and treated as an
ordinary cover-glass preparation. In this way very characteristic
appearances may sometimes be noted and preserved, as in the
case of the anthrax bacillus. The colonies on a plate having
been classified, a microscopic examination of each group may
be made by means of cover-glass preparations, and tubes of
gelatin and agar are inoculated from each representative colony.
Each of the colonies used must be marked for future reference,
preferably by drawing a circle round it on the under surface of
the plate or capsule with one of Faber's pencils for marking on
glass, a number or letter being added for easy reference.

The general lines along which observation is to be made
in the case of a particular bacterium may be indicated as
follows : —

1. Microscopic Appearances. — For ordinary descriptive pur-
poses, young cultures, say of twenty-four hours' growth, on agar
should be used, though appearances in older cultures, such as
involution forms, etc., may also require attention. Note,— (1)
the form ; (2) the size ; (3) the appearance of the protoplasmic
contents, especially as regards uniformity or irregularity of
staining ; (4) the method of grouping ; (5) the staining reactions.
Has it a capsule 1 Does the bacterium stain with simple watery
solutions'? Does it require the use of stains containing
mordants ? How does it behave towards Gram's method ? It
is important to investigate the first four points, both wrhen the
organism is in the fluids or tissues of the body and when growing
in artificial media, as slight variations occur. It must also be
borne in mind that slight variations are observed according to
the kind and consistence of the medium in which the organism
is growing. (6) Is it motile, and has it fiagella? If so, how

GROWTH CHARACTERISTICS 139

are they arranged 1 (7) Does it form spores, and if so, under
what conditions as to temperature, etc.1?

'2. fi'roii'tk Characteristics. — Here the most important points
on which information is to be asked are, What are the characters
of UTO \\th and what are the relations of growth (l)to tempera-
ture ; (2) to oxygen ? These can be answered from some of the
folloNsin- experiments : —

A. (Jrowth on gelatin. (1) Stab culture. Note, — (a) rate of
growth ; (/>) form of growth, (a) on surface, (/?) in substance; (c)
presence or absence of liquefaction ; (d) colour; (e) presence or
absence of gas formation and of characteristic smell ; (/) relation
to reaction of medium. (2) Streak culture. (3) Shake culture.
( \ ) Plate cultures. Note ap}>earances of colonies, (a) superficial,
(b) deep. (5) Growth in fluid gelatin at 37° C.

U. (Jrowth on agar at 37° C. (1) Stab. (2) Streak. Also
on glycerin-agar, blood-agar, etc. Appearances of colonies in
airar plates.

C. Growth in bouillon, (a) character of growth, (£) smell,
('•) reaction.

D. Growth on social media. (1) Solidified blood serum.
(2) Potatoes. (3) Lactose and other sugar media. Does
fermentation occur, and is gas formed ? (4) Milk. Is it curdled
or turned sour ? (5) Litmus media. Note changes in colour.
(6) Peptone solution. Is indol formed 1

K. What is the viability of organism on artificial medial
3. Result* <>>' inoculation experiment* on animals.
By attention to such points as these a considerable knowledge
is attained regarding the bacterium, which will lead to its
identification. In the case of many well-known organisms,
however, a few of the above points taken together will often
be sufficient for the recognition of the species, and experience
teaches what are the essential points as regards any individual
organism. In the course of the systematic description of the
pathogenic organisms, it will be found that all the above points
will be referred to, though not in every case.

The methods hy wliicli the morphological and biological characteristics
of any growth may be observed have already been fully described. It
need cnly l.r pointed out here that in giving descriptions of bacteria the
greatest care must be taken to state every detail of investigation. Thus
in any description of microscopic appearances the age of the growth from
which tin- preparation was made, the medium employed, the temperature
at \\liidi development took place, must be noted, along with the stain
which was used ; and with regard to the latter it is always preferable to
employ one of the well-known staining combinations, such as Loffler's
methylene-blue. Especial care is necessary in stating the size of a

140 GENERAL BACTERIOLOGICAL DIAGNOSIS

bacterium. The apparent size often shows slight variations dependent
on the stain used and the growth conditions of the culture. Accurate
measurements of bacteria can only be made by preparing microphoto-
graphs of a definite magnification, and measuring the sixes on the
negatives. From these the actual sizes can easily be calculated. A
rough method of estimating the size of an organism is to mix a little
with a drop of the observer's blood and make a blood film. As the size
of a normal red blood corpuscle is about 7 '5 IJL, an idea of the size of a
bacterium can be obtained by comparing it with this as a standard. In
describing bacterial cultures it must be borne in mind that the appearances
often vary with the age. It is suggested that in the case of cultures
grown at from 36° to 37° C. the appearances between twenty-four and
forty-eight hours should be made the basis of description, and in the
case of cultures grown between 18° and 22° C. the appearances between
forty-eight and seventy-two hours should be employed. The culture
fluids used must be made up and neutralised by the precise methods
already described. The investigator must give every detail of the
methods he has employed, in order that his observations may be capable
of repetition.

In the case of a number of pathogenic organisms, identification
is a comparatively easy matter. In some cases, however, great
difficulties arise in consequence of the existence of groups of
organisms presenting closely allied characters, and the difficulty
and importance of identification is enhanced by the fact that
the same group may include both harmful and innocent members.
Examples of this occurrence are found in the pyogenic cocci and
their allies, in the coli-typhoid group of bacilli, and in the group
of cholera vibrios. In such cases it is usually necessary to take
into account all the morphological and cultural reactions of an
organism before it can be adequately classified. Within recent
years attempts have been made to apply the statistical method
to the solution of the difficulties of the situation, and here the
results appear to be promising. The method has been applied
to the coccaceae by Winslow and Rogers, who have investigated
500 strains of cocci isolated from the tissues in disease, from
the outer surfaces of the normal human body, from water, earth,
and air. A great variety of properties was studied, and while
in each test applied wide variation was exhibited in such bacteria,
there usually emerged a type property to which individual
strains tended to approach. Thus, while the size varies from
"1 to 2'0 fji, out of about 350 strains examined about 115
measured '3 ^ and the remaining strains tended to be a little
below or a little above this figure. When similar lines of
inquiry were pursued with regard to other characteristics of
the organisms, it was found that important correlations could
be noted. Thus capacity for staining by Gram's method was
found especially amongst the staphylococci and streptococci as

INOCULATION OF ANIMALS HI

contrasted with forms tending to grow in sarcinal packets, and
the Gram-staining forms were chiefly parasitic in habitat.
Looking at their results as a whole, Winslow and Rogers divide
the cocci into two great groups, the Paracoccaceae and the Meta-
coccaceae. The former comprise most of the forms derived from
the body, show a staphylococcal or streptococcal tendency, stain
by Gram, yield only moderate surface growths, form acid in
carbohydrates, and produce no pigment or a white or orange
colour. The latter come chiefly from air and water, often are
suviniform, decolorise by Gram, grow well on the surface of
media, do not ferment carbohydrates, and produce red or yellow
pigment. On similar lines, further subdivision of the groups
could be effected. It is manifest that important means of
differentiating allied bacteria may be available by the extended
application of this method.

INOCULATION OF ANIMALS. l

The animals generally chosen for inoculation are the mouse,
the rat, the guinea-pig, the rabbit, and the pigeon. Great caution
must be shown in drawing conclusions from isolated experiments
on rabbits, as these animals often manifest exceptional symptoms,
and are very easily killed. Dogs are, as a rule, rather insusceptible
to microbic disease, and the larger animals are too expensive for
ordinary laboratory purposes. In the case of the mouse and rat
the variety must be carefully noted, as there are differences in
>iisreptibility between the wild and tame varieties, and between
the white and In-own varieties of the latter. In the case of the
wild varieties, these must be kept in the laboratory for a week or
two before use, as in captivity they are apt to die from very slight
causes ; and, further, each individual should be kept in a separate
cage, as they show great tendencies to cannibalism. Of all the
ordinary animals the most susceptible to microbic disease is the
guinea-pig. Practically all inoculations are performed by means
of the hypodermic syringe. The best variety is made on the
ordinary model with metal mountings, asbestos washers, and
preferably furnished with platinum-indium needles. Before use,
the syringe and the needle are sterilised by boiling for five
minutes. The materials used for inoculation are cultures, animal
• •\ IK lat ions, or the juice of organs. If the bacteria already exist
in a fluid there is no difficulty. The syringe is most conveniently
filled out of a shallow conical test-glass, which ought previously

1 Experiments on animals, of course, cannot, in Britain, be performed with-
out a licence granted by the Home Secretary.

142 INOCULATION OF ANIMALS

to have been covered with a cover of filter paper and sterilised.
If an inoculation is to be made from organisms growing on the
surface of a solid medium, either a little ought to be scraped
off and shaken up in sterile bouillon or '85 , per cent, salt
solution to make an emulsion, or a little- sterile fluid is poured
on the growth, and the latter scraped off into it. This fluid is
then filtered into the test-glass through a plug of sterile glass
wool. This is easily effected by taking a piece of f-inch glass-
tubing 3 inches long, drawing one end out to a fairly narrow point,
plugging the tube with glass wool above the point where the
narrowing commences, and sterilising by heat. By filtering an
emulsion through such a pipette, flocculi which might block the
needle are removed. If a solid organ or an old culture is used
for inoculation, it ought to be rubbed up in a sterile porcelain or
metal crucible with a little sterile distilled water, by means of a
sterile glass rod, and the emulsion filtered as in the last case.

The methods of inoculation generally used are: (1) by scari-
fication of the skin ; (2) by subcutaneous injection ; (3) by
intraperitoneal injection ; (4) by intravenous injection ; (5) by
injections into special regions, such as the anterior chamber of
the eye, the substance of the lung, etc. Of these (2) and (3)
are most frequently used. When an anaesthetic is to be ad-
ministered, this is conveniently done by placing the animal,
along with a piece of cotton wool or sponge soaked in chloroform,
under a bell-jar or inverted glass beaker of suitable size.

1. Scarification. — A few parallel scratches are made in the
skin of the abdomen previously cleansed, just sufficiently deep
to draw blood, and the infective material is rubbed in with a
platinum eyelet. The disadvantage of this method is that the
inoculation is easily contaminated. The method is only occasion-
ally used.

2. Subcutaneous Injection. — A hypodermic syringe is charged
with the fluid to be inoculated. The hair is cut off the part to
be inoculated, and the skin purified with 1 to 1000 corrosive
sublimate, or by dropping upon it some strong solution of iodine.
The skin is then pinched up, and, the needle being inserted, the
requisite dose is administered. The wound is then sealed with a
little collodion.

3. Intraperitoneal Injection. — This may be performed by
means of a special form of needle. The needle is curved, and
has its opening not at the point, but in the side in the middle
of the arch (Fig. 47). The hair over the lower part of the
abdomen is cut, and the skin purified with an antiseptic. The
whole thickness of the abdominal walls is then pinched up

METHODS OF INOCULATION

143

FIG. 47. — Hollow
needle with
lateral aperture
(at a) for intra-
jifi'iloneal in-
oculatious.

between the forefingers and thumbs of the two hands, and
the needle is plunged through the fold thus formed. The
result is that the hole in the side of the needle is within the
abdominal cavity, and the inoculation can thus
In- made. Intraperitoneal inoculation can also
be practised with an ordinary needle. The
mode of procedure is similar, but, after the
needle is plunged through the abdominal fold,
it is partially withdrawn till the point is felt
to be free in the peritoneal cavity, when the
injection is made. There is little risk of injur-
ing the intestines by either method.

4. Intravenous Injection. — The vein most
usually chosen is one of the auricular veins.
Tli<- part has the hair removed, the skin is
purified, and the vein made prominent by
pressing on it between the point of inoculation
and the heart. The needle is then plunged into
the vein, and the fluid injected. That it has
perforated the vessel will be shown by the
escape of a little blood ; and that the injection
has taken place into the lumen of the vessel will be known by
the absence of the small swelling which occurs in subcutaneous
injections. If preferred, the vein may be first laid bare by
snipping the skin over it. The needle is then introduced.

"». inoculation into the Anterior Gliamber of the Eye. — Local
anaesthesia is established by applying a few drops of 2 per cent,
solution of hydrochlorate of cocaine. The eye is fixed by pinch-
ing up the orbital conjunctiva with a pair of fine forceps, and,
the edge of the cornea being perforated by the hypodermic needle,
the injection is easily accomplished.

Sometimes inoculations are made by planting small pieces of
pathological tissues in the subcutaneous tissue. This is especially
dune in the case of glanders and tubercle. The skin over the
back is purified, and the hair cut. A small incision is made with
a sterile knife, and the skin being separated from the subjacent
tissues by means of the ends of a blunt pair of forceps, a little
pocket is formed into which a piece of the suspected tissue is
inserted. The wound is then closed with a suture, and collodion
is applied. In the case of guinea-pigs, the abdominal wall is to
br preferred as the site of inoculation, as the skin over the back
is extremely thick.

Injections are sometimes made into other parts of the body,
'.'/. the pleurae, the cranium, the spinal canal. With regard to

144 INOCULATION OF ANIMALS

the last, Ford-Robertson has pointed out that in the rabbit it
can be easily practised through the space between the seventh
lumbar and first sacral vertebrae. The spine of the former
lies in a line with the iliac crests. With regard to operative
procedures in special regions of the body, it is unnecessary to
describe these, as the application of the general principles
employed above, together with those of modern aseptic surgery,
will sufficiently guide the investigator as to the technique which
is requisite.

After inoculation, the animals ought to be kept in comfortable
cages, which must be capable of easy and thorough disinfection
subsequently. For this purpose galvanised iron wire cages are
the best. They can easily be sterilised by boiling them in the
large fish-kettle which it is useful to have in a bacteriological
laboratory for such a purpose. It is preferable to have the
cages opening from above. Otherwise material which may be
infective may be scratched out of the cage by the animal. The
general condition of the animal is to be observed, how far it
differs from the normal, whether there is increased rapidity of
breathing, etc. The temperature is usually to be taken. This
is generally done per rectum. The thermometer (the ordinary
clinical variety) is smeared with vaselin, and the bulb inserted
just within the sphincter, where it is allowed to remain for a
minute ; it is then pushed well into the rectum for five minutes.
If this precaution be not adopted a reflex contraction of the
vessels may take place, which is likely to vitiate the result by
giving too low a reading.

Collodion Capsules. — These have been used to allow the
sojourn of bacteria within the animal body without their coming
into contact with the cells of the tissues. Various substances
in solution can pass in either direction through the wall by
diffusion, but the wall is impermeable alike to bacteria and
leucocytes. The following method of preparing such capsules is
that of M'Rae modified by Harris : — A gelatin capsule, such as
is used by veterinary surgeons, is taken, and in one end there
is fixed a small piece of thin glass tubing by gently heating the
glass and inserting it. The tube becomes fixed when quite cold,
and the junction is then painted round with collodion, which
is allowed to dry thoroughly. The bore of the tubing is cleared
of any obstructing gelatin, and the whole capsule is dipped into
a solution of collodion so as to coat it completely. The collodion
is allowed to dry, and the coating is repeated ; it is also advis-
able to «j strengthen the layer by further painting it at the
extremity and at the junction. The interior of the capsule is

AUTOPSIES ON ANIMALS 145

then filled with water by a fine capillary pipette, and the capsule
is placed in hot water in order to liquefy the gelatin, which
can be removed from the interior by means of the fine pipette.
The sac is filled with bouillon and is placed in a tube of
bouillon. It is then sterilised in the autoclave. A small
quantity of the bouillon is removed, and the contents are
inoculated with the particular bacterium to be studied, or an
emulsion of the bacterium is added. The glass tubing is seized
in sterile forceps, and is sealed off in a small flame a short
distance above the junction. The closed sac ought then to be
placed in a tube of sterile bouillon to test its impermeability.
The result is satisfactory if no growth occurs in the surrounding
medium. The sac with its contents can now be transferred to
the peritoneal cavity of an animal.

Autopsies on Animals dead or killed after Inoculation.—
These should be made as soon as possible after death — in fact,
it is preferable to kill the animal when it shows serious signs of
illness. It is necessary to have some shallow troughs, con-
structed either of metal or of wood covered with metal, conveni-
ently with sheet lead, and having a perforation at each corner
to admit a tape or strong cord. The animal is tightly stretched
out in the trough and tied in position. The size of the trough
will therefore have to vary with the size of the outstretched
body of the animal to be examined. In certain cases it is well
to soak the surface of the animal in carbolic acid solution
(1 to 20) or in corrosive sublimate (1 to 1000) before it is tied
out. This not only to a certain extent disinfects the skin, but,
what is more important, prevents hairs which might be affected
with pathogenic products from getting into the air of the
laboratory. The instruments necessary are scalpels (preferably
with metal handles), dissecting forceps, and scissors. They are
to be sterilised by boiling for five minutes. This is conveniently
done in one of the small portable sterilisers used by surgeons.
Two sets at least ought to be used in an autopsy, and they may
l»f placed, after boiling, on a sterile glass plate covered by a
bell-jar. It is also necessary to have a medium-sized hatchet-
si uiped cautery, or other similar piece of metal. It is well to
have prepared a few freshly- drawn-out capillary tubes stored in
a sterile cylindrical glass vessel, and also some larger sterile glass
pipettes. The hair of the abdomen of the animal is removed.
If some of the peritoneal fluid is wanted, a band should be
cauterised down the linea alba from the sternum to the pubes,
and another at right angles to the upper end of this ; an incision
should be made in the middle of these bands, and the abdominal
10

146 INOCULATION OF ANIMALS

walls thrown to each side. One or more capillary tubes should
then be filled with the fluid collected in the flanks, the fluid
being allowed to run up the tube and the point sealed off; or a
larger quantity, if desired, is taken in a sterile pipette. If
peritoneal fluid be not wanted, then an incision may be made
from the episternum to the pubes, and the thorax and abdomen
opened in the usual way. The organs ought to be removed with
another set of instruments, and it is convenient to place them
pending examination in deep Petri's capsules (sterile). It is
generally advisable to make cultures and film preparations from
the heart's blood. To do this, open the pericardium, sear the
front of the right ventricle with a cautery, make an incision in
the middle of the part seared, and remove some of the blood
with a capillary tube for future examination ; or, introducing a
platinum eyelet, inoculate tubes and make cover-glass prepara-
tions at once. To examine any organ, sear the surface with a
cautery, cut into it, and inoculate tubes and make film prepara-
tions with a platinum loop. For removing small parts of organs
for making inoculations on tubes, a small platinum spud is very
useful, as the ordinary wires are apt to become bent. Place
pieces of the organs in some preservative fluid for miscroscopic
examination. The organs ought not to be touched with the
fingers. When the examination is concluded, the body should
have corrosive sublimate or carbolic acid solution poured over it,
and be forthwith burned. The dissecting trough and all the
instruments ought to be boiled for half an hour. The amount
of precaution to be taken will, of course, depend on the character
of the bacterium under investigation, but as a general rule every
care should be used.

CHAPTER V.

BACTERIA IN AIR, SOIL, AND WATER.
ANTISEPTICS.

IT is impossible here to do more than indicate the chief methods
which are employed by bacteriologists in the investigation of the
bacteria present in air, soil, and water, and to add an outline of
the chief results obtained. In dealing with the latter, the subject
lias been approached mainly from the standpoint of the bearings
which the results have towards human pathology. In dealing
with antiseptics, so far as possible the effects of the various
agents on the chief pathogenic bacteria have been given, though
in many cases our information is very imperfect.

AIR.

Very little information of value can be obtained from the
examination of the air, but the following are the chief methods
used, along with the results obtained. More can be learned
from the examination of atmospheres experimentally contamin-
ated than by the investigation of the air as it exists under
natural conditions.

Methods of Examination. — The methods employed vary with the
objects in view. If it be sought to compare the relative richness of
different atmospheres in organisms, and if the atmospheres in question
be fairly quiescent, then it is sufficient to expose gelatin plates for
definite times in the rooms to be examined. Bacteria, or the particles of
dust carrying them, fall on the plates, and from the number of colonies
which develop a rough idea of the richness of the air in bacteria can be
obtained. Petri states that in five minutes the bacteria present in
10 litres of air are deposited on 100 square centimetres of a gelatin plate.

More complete results are available when some method is employed by
which the bacteria in a given quantity of air are examined. The oldest
in.'tlin 1 employed, and one which is still used, is that of Hesse. The
apparatus is shown in Fig. 48. It consists of a cylindrical tube a about
20 inches long and 2 inches in diameter. At one end this is closed by a
rubber cork having a piece of quill tubing, /, passing through it and

148

BACTERIA IN AIR

projecting some distance into the interior. For use the tube is sterilised
in a tall "Koch," and then a quantity of peptone gelatin, sufficient to
cover the whole interior to the thickness of an ordinary gelatin plate, is
poured in. This gelatin is kept from escaping by the projection of the
quill tubing into the lumen of the large tube. A plug of cotton wool is
now placed in the outer end of the quill tubing. Over the other end of
the large tube is tied a sheet of rubber having a hole about a quarter of
an inch in diameter in its centre, and over this again is tied a piece
of similar but unperforated sheet rubber. The tube is then sterilised

in the tall "Koch." On
removal from this it is
rolled, after the manner
of an Esmarch's tube
(q.v.), till the gelatin is
set as a layer over its
interior, and it is then
placed horizontally on the
tripod as shown. The
other part of the appa-
ratus is an aspirator, by
means of which a known
quantity of air can be
brought in contact with
the gelatin. It consists
of two conical glass flasks
connected by means of a
tube which passes through
the cork of each down to
the bottom of the flask.
When this tube is filled
with water it, of course,
can act as a syphon tube
between volumes of water
in the flasks. Such a
syphon system being es-
tablished, the levels of
the water are marked on
the flasks, and to one a
litre of water is added ;
by depressing flask b the
whole litre can be got
into it, and the connect-
ing tube c is then clamped. The two flasks are now connected by a
rubber tube with the tube/, the clamp on c is opened, and the passing of
a litre of water into d will draw a litre of air through the gelatin tube,
when the outer rubber sheet is removed from the end and the clamp h
opened. By disconnecting at g and reversing the syphon flasks, another
litre can be sucked through, and so any desired quantity of air can be
brought in contact with the gelatin. The speed ought not to be more
than one litre in two minutes, and in such a case practically all the
organisms will be found to have fallen out of the air on to the gelatin
in the course of their transit. This fact can be tested by interposing
between the tube a and the aspirator a second tube prepared in the same
way, which ought, of course, to show no growth. When forty-eight
hours at 20° C. or four days at lower temperature have elapsed, the

FIG. 48. — Hesse's tube, mounted for use.

PETRT'S SAND-FILTER METHOD

149

i which develop in a may be counted. The disadvantage of the
method is that if particles of dust carrying more than one bacterium
alight on the gelatin, these bacteria develop in one colony, and thus the
numeration results may be too low ; difficulties may also arise from
liquefying colonies developing in the upper parts of the tube and running
over the gelatin.

Petri's Sand-Filter Method.— A glass tube open at both ends, and
about 3i inches long and half an inch wide, is taken, and in its centre is
placed a transverse diaphragm of very fine iron gauze
(Fig. 49, e.} ; on each side of this is placed some fine
quartz sand which has been well washed, dried, and
burned to remove all impurities, and this is kept in
position by cotton plugs. The whole is sterilised by
dry heat. One plug is removed, and a sterile rubber
cork, r, inserted, through which a tube, d, passes to
an exhausting apparatus. The tube is then clamped
in an upright position in the atmosphere to be ex-
amined, with the remaining plug, /, uppermost.
The latter is removed and the air sucked through.
Difficulty may be experienced from the resistance of
the sand if quick filtration be attempted. The best
means to adopt is to use an air-pump — the amount
of air drawn per stroke of which is accurately known
— and to have a manometer (as in Fig. 30) interposed
between the tube and the pump. Between each two
strokes of the air-pump the mercury is allowed to
return to zero. After the required amount of air has
passed, the sand a is removed, and is distributed
among a number of sterile gelatin tubes which arc
well shaken ; plate cultures are then made, and
when growth has occurred the colonies are enumer-
ated ; the sand b is similarly treated, and acts as a
control.

When it is necessary to examine air for particular
organisms, special methods must often be adopted.
Thus in the case of the suspected presence of tubercle bacilli a given
quantity of air is drawn through a small quantity of water and then
injected into a guinea-pig.

FIG. 49.— Petri's
sand filter.

It must be admitted that comparatively little information
bearing on the harmlessness or harmfulness of the air is obtain-
able by the mere enumeration of the living organisms present,
for under certain conditions the number may be increased by
the presence of many individuals of a purely non-pathogenic
character. The organisms found in the air belong to two
groups — firstly, a great variety of bacteria ; secondly, yeasts and
the spores of moulds and of the lower fungi. With regard to
the spores, the organisms from which they are derived often
consist of felted masses of threads, from which are thrust into
the air special filaments, and in connection with these the spores
are formed. By currents of air these latter can easily be detached.

150 BACTERIA IN AIR

and may float about in a free condition. With the bacteria, on
the other hand, the case is different. Usually these are growing
together in little masses on organic materials, or in fluids, and
it is very much by the detachment of minute particles of the
substratum that the organisms become free. The entrance of
bacteria into the air, therefore, is associated with conditions
which favour the presence of dust, minute droplets of fluid, etc.
The presence of dust, in particular, would specially favour a large
number of bacteria being observed, and this is the case with the
air in many industrial conditions, where the bacteria, though
numerous, may be quite innocuous. Great numbers of bacteria
thus may not indicate any condition likely to injure health, and
this may be true also even when the bacteria come from the
crowding together of a number of healthy human beings. On
the other hand, there is no doubt that disease germs can be
disseminated by means of the air. The possibility of this
has been shown experimentally by infecting the mouth with the
b. prodigiosus, which is easily recognised by its brilliantly
coloured colonies, and then studying its subsequent distribution.
Most important here is the infection of the air from sick persons.
The actions of coughing, sneezing, speaking, and even of deep
breathing, distribute, often to a considerable distance, minute
droplets of secretions from the mouth, throat, and nose, and these
may float in the air for a considerable time. Even live hours
after an atmosphere has been thus infected evidence may be
found of bacteria still floating free. Before this time, however,
most of the bacteria have settled upon various objects, where
they rapidly dry, and are no longer displaceable by ordinary air
currents. The diseases of known etiology where infection can
thus take place are diphtheria, influenza, pneumonia, and phthisis ;
and here also possibly typhus fever and measles are to be
added, though the morbific agents are unknown. In the case of
phthisis, the deposition of tubercle bacilli has been demonstrated
on cover-glasses held before the mouths of patients while talking,
and animals made to breathe directly in front of the mouths of
such patients have become infected with tuberculosis. Apart
from direct infection from individuals, however, pathogenic
bacteria may be spread in some cases from the splashing of
infected water, as from a sewage outfall. This possibility has to
be recognised especially in the cases of typhoid and cholera.
Besides infection through fluid particles, infection can be caused
in the air by dust coming from infected skin or clothes,
etc. Fliigge, in dealing with this subject in an experimental
inquiry, distinguishes between large particles of dust which

DISTRIBUTION OF BACTERIA BY AIR 151

nM|iihv ;ui air current moving at the rate of 1 centimetre per
second in l«-rp them suspended, and the finer dust which can In-
kept in suspension by currents moving at from 1 to 4 milli-
metres per second. In the former case, when once the particles
settle they cannot be displaced by currents of air except when
these are moving at, at least, 5 metres per second, but the
brushing, shaking, or beating of objects may, of course, distribute
tin-in. In the case of the liner dust the particles will remain for
long suspended, and when they have settled can ' be more easily
displaced, as by the waving of an arm, breathing, etc. With re-
gard to infection by dust, a most important factor, howrever, is
whether or not the infecting agent can preserve its vitality in
a dry condition. In the case of a sporing organism such as
anthrax, vitality is preserved for long periods of time, and great
resistance to drying is also possessed by the tubercle and
diphtheria bacilli ; but apart from such cases there is little
doubt that infection is usually necessarily associated with the
transport of moist particles, and is thus confined to a limited
area around a sick person. Among diseases which may occasion-
ally be thus spread, cholera and typhoid have been classed.
Considerable controversy has arisen with regard to certain out-
breaks of the latter disease, which have apparently been spread
by dusty winds, although we have the fact that the typhoid
bacillus does not survive being dried even for a short time.
It appears, however, that in such epidemics the transport of
infection by means of insects carried by the wind has not been
entirely excluded.

As in the cases of the soil and of water, presently to be described,
attempts have been made to obtain indirect evidence of the contamination
ni the air from human sources. Thus Gordon has shown that certain
streptococci are common in the saliva ; these usually correspond to the
f'ococcus salirarius of Andrewes »«d Horder (q.v.) in that they grow
at 37° C., form acid and clot in litmus milk, reduce neutral-red, and fer-
ment saccharose, lactose, and raffinose. Andrewes and Horder also describe
another group, — sir. cquinm, — as common in London air, which they
think is there derived from horse dung. Thus the finding of streptococci
of the first group in plates exposed to air would indicate that a human
source was probable, and, if the observation were made on air from the
neighbourhood of a sick person, that risk of the dissemination of disease
g«Tins was present. The value of this as a practical method has yet to he
determined,

Son..

The investigation of the bacteria which may be found in the
soil is undertaken from various points of view. Information

152 BACTERIA IN SOIL

may be desired as to the change its composition undergoes by
a bacterial action, the result of which may be an increase
in fertility and thus in economic value. Under this head may
be grouped inquiries relating to the bacteria which convert
ammonia and its salts into nitrates and nitrites, and to the
organisms concerned in the fixation of the free nitrogen of the
air. The discussion of the questions involved in such inquiries
is outside the scope of the present chapter, which is more con-
cerned with the 'relation of the bacteriology of the soil to questions
of public health. So far as this narrower view is concerned, soil
bacteria are chiefly of importance in so far as they can be washed
out of the soils into potable water supplies. An important aspect
of this question thus is as to the significance of certain bacterio-
logical appearances in a water in relation to the soil from which
it has come or over which it has flowed. In this country these
questions have been chiefly investigated by Houston, and it is
from his papers that the following account is largely taken.

Methods of Examination. — For examination of soil on surface or not
far from surface, Houston recommends tin troughs 10 in. by 3 in., and
pointed at one extremity, to be wrapped in layers of paper and sterilised
by dry heat. If several of these be provided, then the soil can be well
rubbed up and a sample secured and placed in a sterile test-tube for
examination as soon as convenient after collection. If samples are to
be taken at some depth beneath the surface, then a special instrument
of which many varieties have been devised must be used. The general
form of these is that of a gigantic gimlet stoutly made of steel. Just
above the point of the instrument the shaft has in it a hollow chamber,
and a sliding lateral door in this can be opened and shut by a mechanism
controlled at the handle. The chamber being sterilised and closed, the
instrument is bored to the required depth, the door is slid back, and by
varying devices it is effected that the chamber is filled with earth ; the
door is reclosed and the instrument withdrawn.

In any soil the two important lines of inquiry are first, as to the total
number of organisms (usually reckoned per gramme of the fresh sample) ;
and secondly, as to the varieties of organisms present. The number of
organisms present in a soil is often, however, so enormous that it is con-
venient to submit only a fraction of a gramme to examination. The
method employed is to weigh the tube containing the soil, shake out an
amount of about the size of a bean into a litre of distilled water, and
revveigh the tube. The amount placed in the water is distributed as
thoroughly as possible by shaking, and, if necessary, by rubbing down
with a sterile glass rod, and small quantities .measured from a graduated
pipette are used for the investigation. For estimating the total number of
organisms present in the portion of soil used, small quantities, say "1 c.c.
and 1 c.c., of the fluid are added to melted tubes of ordinary alkaline
peptone gelatin ; after being shaken, the gelatin is plated, incubated at
22° C., and the colonies are counted as late as the liquefaction, which
always occurs round some of them, will allow. From these numbers the
total number of organisms, which grow in gelatin, in a given amount of
soil can be calculated.

BACTERIA IN SOIL 153

The numbers of bacteria in the soil vary very much. Accord-
ing to Houston's results, fewest occur in uncultivated sandy soils,
these containing on an average 100,000 per gramme. Peaty soils,
though rich in organic matter, also give low results, it being
possible that the acidity of such soils inhibits free bacterial
growth. Garden soils yield usually about 1,500,000 bacteria
per gramme, but the greatest numbers are found in soils which
have been polluted by sewage, when the figures may rise to
several millions. In addition to the enumeration of the numbers
of bacteria present, it is a question whether something may not
be gained from a knowledge of the number of spores present in
a soil relative to the total number of bacteria. This is a point
which demands further inquiry, especially by the periodic investi-
gation of examples of different classes of soils. The method is to
take 1 c.c. of such a soil emulsion as that just described, add it
to 10 c.c. of gelatin, heat for ten minutes at 80° C. to destroy
the non-spored bacteria, plate, incubate, and count as before.

Besides the enumeration of the numbers of bacteria present in
a soil, an important question in its bacteriological examination
lies in inquiring what kinds of bacteria are present in any par-
ticular case. Practically this resolves itself into studying the
most common bacteria present, for the complete examination of
the bacterial flora of any one sample would occupy far too much
time. Of these common bacteria the most important are those
from whose presence indications can be gathered of the con-
tamination of the soil by sewage, for from the public health
standpoint this is by far the most important question on which
bacteriology can shed light.

Bacillus tnycoides. — This bacillus is 1'6 to2'4 /u. in length, and about
•9 JJL in breadth. It grows in long threads which often show motility.
It can be readily stained by such a combination as carbol-thionin, and
retains the dye in Gram's method. All ordinary media will support its
growth, and, in surface growths on agar or potato, spore formation is
readily produced. Its optimum temperature is about 18° C. On gelatin
plates it shows a very characteristic appearance. At first under a low
power it shows a felted mass of filaments throwing out irregular shoots
from the centre, and later to the naked eye these appear to be in the
form of thick threads like the growth of a mould. They rapidly spread
over the surface of the medium, and the whole resembles a piece of wet
teased-out cotton wool. The gelatin is liquefied.

Cladothrices. — Of these several kinds are common in the soil. The
ordinary cladothrix dichotoma, is among them. This organism appears
as a colourless flocculent growth with an opaque centre, and can be seen
under the microscope to send out into the medium apparently branched
threads which vary in thickness, being sometimes 2 /* across. They
consist of rods enclosed in a sheath. These rods may divide at any
point, and thus the terminal elements may be pushed along the sheath.

154 BACTERIA IN SOIL

Sometimes the sheath ruptures, and thus by the extrusion of these
dividing .cells and their further division the branching appearance is
originated. Reproduction takes place by the formation ot'gonidia in the
interior of the terminal cells. These gonidia acquire at one end a bundle
of flagella, and for some time swim free before becoming attached and
forming a new colony. Houston describes as occurring in the soil another
variety, which with similar microscopic characters appears as a brownish
growth with a pitted surface and diffuses a Bismarck-brown pigment
into the gelatin which it liquefies.

A few experiments made with an ordinary field soil will, however,
familiarise the worker with the non-pathogenic bacteria usually present.
We have referred to these two because of their importance. In regard
to pathogenic organisms, especially in relation to possible sewage con-
tamination, attention is to be directed to three groups of organisms,
those resembling the b. coli, the bacillus enteritidis sporogenes, and the
streptococcus pyogenes. The characters of the first two of these will
be found in the chapter on Typhoid Fever ; of the third in Chapter VII.
For the detection of these bacteria Houston recommends the following
procedure : —

(a) The B. coli Group. — A third of a gramme of soil is added to
10 c.c. broth containing '2 per cent, of phenol and incubated at 37° C.
In this medium very few if any other bacteria except those of the b. coli
group will grow, so that if after twenty-four hours a turbidit}r appears,
some of the latter may be suspected to be present. In such a case a
loopful of the broth is shaken up in 5 c.c. sterile distilled water, and of
this one or two loopfuls are spread over the surface of a solid plate of
phenol gelatin in a Petri capsule either by means of the loop or of a
small platinum spatula, and the plate is incubated at 20° C. Any
colonies which resemble b. coli are then examined by the culture methods
detailed under that organism. Further, all organisms having the micro-
scopic appearances of b. coli, and which generally conform to its culture
reactions, are to be reckoned in the coli group. The media of MacConkey
and Drigalski are very useful in connection with the plating and separa-
tion of such soil organisms (vide pp. 50, 47).

(b) The Bacillus enleritidis sporogenes. — To search for this organism
1 gramme of the soil is thoroughly distributed in 100 c.c. sterile distilled
water, and of this 1 c.c., '1 c.c., and '01 c.c. is added to each of three
sterile milk tubes. These are heated to 80° C. for ten minutes, and then
cultivated anaerobically at 37° C. for twenty-four hours. If the charac-
teristic appearances seen in such cultures of the b. enteritidis (q.v.} are
developed, then it may fairly safely be deduced that it is this organism
which has produced them.

(c) FcKcal Streptococci. — The method here is to pour out a tube of agar
into a Petri capsule, and when it has solidified to spread out *1 c.c. of
the emulsion of soil over it and incubate at 37° C. for twenty-four hours.
At this temperature many of the non-pathogenic bacteria grow with
difficulty, and thus the number of colonies which develop is relatively
small. Colonies having appearances resembling those of the streptococcus
pyogenes (q.v.} can thus be investigated.

Another method is that of Prescott and Winslow modified by Mair.
This depends on the fact that when b. coli and streptococci are growing
together in glucose broth, as the medium becomes acid the streptococci
tend to outgrow the b. coli. If lactose agar plates be made at this stage,
the colonies of streptococci, being small and intensely red, can be distin-
guished from the larger and less acid colonies of the b. coli. They can

BACTERIA IN SOIL 155

then be picked off for investigation. It is evident lli.it here tlie method
must be adopted of taking as a measure of the number of streptococci
present the least quantity of the original fluid in which evidence of their

presence can be detected.

\\V may now ^ive in brief the results at which Houston has
arrived by the application of these methods. First of all, un-
cultivated soils contain very few, if any, representatives of the
b. mycoides, and this is also true to a less extent of the
dadothrices. Cultivated soils, on the other hand, do practically
always contain thrsr organisms. With regard to the b. coli,
its presence in a soil must be looked on as indicative of
recent pollution with excremental matter. The presence of
b. eiiti'ritidis is also evidence of such pollution, but from the fact
that this is a sporing organism the pollution may not have been
recent. With regard to the streptococci, on the other hand, the
opinion is advanced that their presence is, on account of their
tWble viability outside the animal body, to be looked on as
evidence of extremely recent excremental |x)llution. The very-
great importance of these results in relation to the bacterio-
logical examination of water supplies will be at once apparent,
and will be referred to again in connection with this subject.

While such means have been advanced for the obtaining of
indirect evidence of excremental pollution of soil, and therefore
of a pollution dangerous to health from the possible presence of
pathogenic organisms in excreta, investigations have also been
conducted with regard to the viability in the soil of pathogenic
bacteria, especially of those likely to be present in excreta, namely,
the typhoid and cholera organisms. The solution of this problem
is attended with difficulty, as it is not easy to identify these
• ir^anisms when they are present in such bacterial mixtures as
naturally occur in the soil. Now there is evidence that bacteria
when growing together often influence each other's growth in an
unfavourable wray, so that it is only by studying the organisms in
<iuestion when growing in unsterilised soils that information can
l>e obtained as to what occurs in nature. For instance, it has
been found that the b. typhosus, when grown in an organically
polluted soil which has been sterilised, can maintain its vitality
for fifteen weeks, but if the conditions occurring naturally be so
far imitated by growing it in soil in the presence of a pure
culture of a soil bacterium, it is found that sometimes the
typhoid bacillus, sometimes the soil bacterium in the course
of a few weeks, or even in a few days, disappears. Further, the
character of the soil exercises an important effect on the results ;
for instance, the typhoid bacillus soon dies out in a virgin sandy

156 BACTERIA IN WATER

soil, even when it is the only organism present. In experiments
made by sowing cultures of cholera and diphtheria in plots in a
field, it was found that after, at the longest, forty days they were
no longer recognisable. Further, it is a question whether
ordinary disease organisms, even if they remain alive, can
multiply to any great extent in soil under natural conditions.
If we are dealing with a sporing organism such as the b.
anthracis, the capacity for remaining in a quiescent condition of
potential pathogenicity is, of course, much greater. The most
important principle to be deduced from these experiments is that
the ordinary conditions of soil rather tend to be unfavourable
to the continued existence of pathogenic bacteria, so that by
natural processes soil tends to purify itself. It must, however,
be noted that such an organism as the typhoid bacillus can exist
long enough in soil to be a serious source of danger.

WATER.

In the bacteriological examination of water three lines of
inquiry may have to be followed. First, the number of bacteria
per cubic centimetre may be estimated. Second, the kinds of
bacteria present may be investigated. Third, it may be necessary
to ask if a particular organism is present, and, if so, in what
number per c.c. it occurs.

Methods. — Collection of Samples. — In all water examinations it is pre-
ferable that the primary culture media (i.e. those to which the water is
actually to be added) should be inoculated at the spot at which the sample
is collected. When this is not possible, the samples should be packed in
sawdust and ice and the primary inoculations made as soon as possible.
Otherwise the bacteria will multiply, and an erroneous idea of the number
present will be obtained. Immediately after collection a slight diminution
in numbers may be observed, but at any rate after six hours an increase
over the initial numbers is manifest.

When samples have to be taken for transport to the laboratory, these
are best collected in 8-ounce, wide-mouthed, stoppered bottles, which
are to be sterilised by dry heat (the stopper must be sterilised separately
from the bottle and not inserted in the latter till both are cold, otherwise
it will be so tightly held as to make removal very difficult).

In the case of water taken from a house tap, the water should be allowed
to run for some time before the sample is taken, as water standing in
pipes in a house is under very favourable conditions for multiplication of
bacteria taking place, and if this precaution be not adopted an altogether
erroneous idea of the number present may be obtained.

With river waters it is best to immerse the sampling bottle and then
remove the stopper with forceps. Care must be taken not to touch the
river bed, as the vegetable matter covering it contains many organisms.
When water has to be taken from below the surface of a well or lako, a
weighted sample bottle must be used. Several special bottles have been

BACTERIA IN WATER 157

devised for such a purpose. Quite good results are obtained by tying
two short lengths of string to the neck and stopper of an ordinary bottle
respectively, winding them round the neck and enveloping in cotton
wool ; any required length of string can afterwards be knotted on these.
A piece of lead can be attached to the bottom of the bottle by wires
passing round the neck. The whole is then wrapped in paper and
sterilised. For use the bottle is carefully lowered to the required depth
by the string attached to the neck, the stopper is jerked out, and the
bottle filled. If the bottle and stopper be rapidly jerked through the
topmost layers, contamination with surface bacteria does not appear as a
serious factor.

Counting of Bacteria in Water. — This is done by adding a given quantity
of water to 10 c.c. of liquefied gelatin or agar, plating, and counting the
colonies which develop. The amount of water added depends on its
source, and varies from '1 c.c. of a water likely to have a high bacterial
content to 5 c.c. of a purer water. It is usual to inoculate both gelatin
and agar tubes. The former, incubated at 20° C., gives an idea of the
numbers of bacteria present which grow at summer heat ; the latter,
incubated at 37Q C., those which grow at blood-heat. As the pathogenic
and intestinal bacteria grow at this temperature, the determination of the
numbers of blood-heat bacteria is important. The counts on the two
media usually differ as each is favourable to the growth of its own group
of organisms. With regard to the summer-heat bacteria it is important
to note that gelatine of a slightly greater alkalinity than that ordinarily
prepared — such an increased degree as is caused by the addition of
•01 grm. of Na2C03 to 10 c.c. peptone gelatin — will give a greater yield
of colonies. In the case of both gelatin and agar plates usually forty-
••ight hours' incubation is allowed before the colonies are counted, but, with
the former, difficulties may arise in consequence of the presence of rapidly
liquefying colonies, and it may thus be necessary to count after twenty-
four hours.

Probably no one medium will support the growth of all the organisms
present in a given sample of water, and under certain circumstances special
media must therefore be used. Thus Hansen found that in testing
waters to be used in brewing it was advisable to have in the medium
employed some sterile wort or beer, so that the organisms in the test
experiments should be provided with the food materials which would be
present in the commercial use of the water. Manifestly this principle
applies generally in the bacteriological examination of waters to be used
for industrial purj>oses.

Detection of the Presence of Special Organisms. — (a) The B. coli Group. —
In ordinary public health work, it may be taken that the most frequent and
important inquiry with regard to a water is directed to the investigation
of the presence or absence of the b. coli and its congeners. Here the
method adopted is to determine the smallest quantity of a water which
gives evidence of containing organisms of this type. In applying any
method with this object in view it is, we consider, absolutely necessary
that it shall be carried out at the spot at which samples are collected.

The usual method is to use as the primary culture medium one of the
bile-salt preparations, of which the best is MacConkey's bile-salt glucose
bouillon to which litmus has been added — glucose being used in prefer-
ence to lactose in order to bring out b. enteritidis of Gaertner if this be
present. In this medium the members of the b. coli group cause changes
n suiting in the formation of acid and gas. It is thus convenient to put
the nu-dium into Durham's fermentation tubes. In practice we employ

158 BACTERIA IN WATER

2-ouuce cylindrical mediciue bottles, 4£ in. high by 1£ in. in diameter.
The medium, along with the inverted test-tube, is placed in these ;
rubber stoppers are inserted in the mouths, and they are sterilised. It is
customary to test for the presence of the organisms in any sample by
adding to a series of such tubes the following quantities of the, water
— 50 c.c., 20c. c., 10 c.c., 5 c.c., 1 c.c., and, it may be, in specially sus-
picious waters, '5 c.c., -1 c.c., and even -01 c.c. The result is estimated
in terms of the smallest amount of water with which the occurrence of
acid and gas formation is observed. By starting with a. concentrated
MacConkey's mixture, it is arranged that, when the sample is added, the
resulting fluid shall be of the concentration of MacConkey's medium as
ordinarily prepared. Thus, in the bottle to which the 50 c.c. sample is
to be added, there are placed 10 c.c. of a sixfold concentration of
MacConkey's medium. In the 20 c.c. tube, there are present 20 c.c. of a
medium of double strength ; in the 10 c.c. tube, 10 c.c. of a mixture
of double strength ; and in the 5 c.c. tube, 5 c.c. of a mixture of double
strength. With smaller samples, we simply use the ordinary MacConkey's
medium.

For the taking of the samples, sterile 8-ounce stoppered bottles are
convenient, and for each sample it is necessary to have sterile 25 c.c.,
10 c.c. (graduated to tenths), and 1 c.c. (graduated to hundredths)
pipettes. The armamentarium being thus simple, there is. no difficulty
in carrying out the necessary manipulations at the spot where the sample
is collected.

The tubes are incubated for forty-eight hours, and it is well to read the
results at the end of the first twenty-four hours also. The formation of
acid and gas in the tube is usually recognised as " presumptive evidence "
of the presence of members of the b. coli group, but it is necessary to
further investigate the bacteria giving rise to this change to determine
whether they are "typical" or "atypical" b. coli. With this end in
view, each bottle in which acid and gas is present is well shaken up, two
or three loopfuls are placed on a plate of MacConkey's neutral-red bile-salt
lactose agar. These loopfuls are spread over the surface by means of a
sterile spreader, made by taking a piece of glass rod and turning a portion
about 2 inches long at right angles to the shaft. The plates are incubated
for twenty-four hours. As typical b. coli produces acid in lactose, any
colonies of such an organism are of a rosy red colour. These are then
picked off, sloped agar tubes are inoculated and used for the further
investigation of the properties of the bacterium isolated.

The media inoculated should be gelatin stab, litmus milk, neutral-red
lactose bouillon, glucose broth, peptone water, dulcite peptone water,
adonite peptone water, inuline peptone water, saccharose peptone water,
and potato.

It is well in dealing with the neutral-red lactose agar plates to inoculate
a lactose peptone water tube from all the kinds of colonies present,
whether these are red or not, as MacConkey rightly points out that some-
times an organism which is really a lactose fermenter does not produce
a red colour on the solid medium. There is another point to be noted
here, namely, that the naked-eye appearances of colonies on lactose agar
are not of value in identifying the kind of organism present.

The object of growing suspicious colonies on a range of media such as
that given, is to enable typical b. coli to be recognised when present. At
the present time it cannot be said that bacteriologists are in agreement
as to what characters determine the type of organism most frequently
found in the human intestine — this, of course, being the important point

BACTERIA IN WATER 159

in judging of the contamination of a water supply. The subject will be
more fully discussed in the chapter on Typhoid Fever. Here it may be
said that for work on water two attitudes are taken up in this country.
I ii>t. that of Houston, who recognises as typical qualities the following :
fluorescence in neutral red broth, production of acid and gas in lactose
peptone water, production of indol, production of acid and clot in litmus
milk (so-called " flaginac " reaction). Secondly, that of the English
Committee of 1904, which, on the one hand, laid stress on the additional
factor of non-liquefaction of gelatin, and on the other, attached less
importance to the production of indol and the occurrence of fluorescence
(see p. 355).

With regard to saccharose fermentation, different strains of coli of
undoubted intestinal origin behave differently towards saccharose, but
when saccharose is fermented the occurrence is significant, as indicating
a great probability that the organism is intestinal in origin.

(6) B. enteritidis sporogcncs and streptococci. — As in the case of
sewage, the presence of these in a water may be sought for. The methods
are those which have already been given (p. 154).

Much work has been devoted to the question of these faical streptococci
presenting specific characters by which they could be differentiated
from other streptococci. Houston has found that the prevailing type
of organism here is one which produces acid and clot in milk, reduces
neutral-red, and ferments saccharose, lactose, and salicin. It corresponds
to the streptococcus fcecalis of Andrewes and Horder. The important
point in this connection is to recognise that streptococci of such a type
exist in great numbers in human faeces, and that when in any circum-
stances faecal contamination is suspected, the isolation of streptococci
.strengthens the suspicion.

With regard to the objects with which the bacteriological
examination of water may be undertaken, though these may
be of a purely scientific character, they usually aim at contribut-
ing to the settlement of questions relating to the potability of
waters, to their use in commerce, and to the efficiency of
processes undertaken for the purification of waters which have
undergone pollution. The last of these objects is often closely
associated with the first two, as the question so often arises
whether a purification process is so efficient as to make the
water again fit for use.

Water derived from any natural source contains bacteria,
though, as in the case of some artesian wells and some springs,-
the numbers may be very small, e.g. 4 to 100 per c.c. In rain,
snow, and ice there are often great numbers, those in the first
two being derived from the air. Great attention has been paid
to the bacterial content of wells and rivers. With regard to
the former, precautions are necessary in arriving at a judgment.
If the water in a well has been standing for some time,
multiplication of bacteria may give a high value. To meet this
• litlieulty the well ought, if practicable, to be pumped dry and

160 BACTERIA IN WATER

then allowed to fill, in order to get at what is really the im-
portant point, namely, the bacterial content of the water
entering the well. Again, if the sediment of the well has
been stirred up, a high value is obtained. Ordinary wells of
medium depth contain from 100 to 2000 per c.c. With regard
to rivers very varied results are obtained. Moorland streams
are usually fairly pure. In an ordinary river the numbers
present vary at different seasons of the year, whilst the pre-
vailing temperature, the presence or absence of decaying
vegetation, or of washings from land, and dilution with large
quantities of pure spring water, are other important features.
Thus the Franklands found the rivers Thames and Lea purest
in summer, and this they attributed to the fact that in this
season there is most spring water entering, and very little water
as washings off land. In the case of other rivers the bacteria
have been found to be fewest in winter. A great many circum-
stances must therefore be taken into account in dealing with
mere enumerations of water bacteria, and such enumerations
are only useful when they are taken simultaneously over a
stretch of river, with special reference to the sources of the
water entering the river. Thus it is usually found that im-
mediately below a sewage effluent the bacterial content rises,
though in a comparatively short distance the numbers may
markedly decrease, and it may be that the river as far as
numbers are concerned may appear to return to its previous
bacterial content. The numbers of bacteria present in rivers
vary so greatly that there is little use in quoting figures, most
information being obtainable by comparative enumerations before
and after a given event has occurred to a particular water.
Such a method is thus of great use in estimating the efficacy
of the filter-beds of a town water supply. These usually
remove from 95 to 98 per cent, of the bacteria present, and
a town supply as it issues from the filter-beds should not
contain more than 100 bacteria per c.c. Again, it is found
that the storage of water effects a very marked bacterial purifica-
tion. Thus Houston has shown in one series of observations
that while 93 per cent, of samples of raw river Lea water
contained b coli. in 1 c.c. or less, in the stored water 62 per
cent, of the samples showed no b. coli to be present in 100 c.c.
The highest counts of bacteria per c.c. are observed with sewage ;
for example, in the London sewage the numbers range from
six to twelve millions.

Much more important than the mere enumeration of the
bacteria present in a water is the question whether these include

BACTERIA IN WATER 161

forms pathogenic to man. The chief interest here, so far as
Europe is concerned, lies in the fact that typhoid fever is so
frequently water-borne, but cholera and certain other intestinal
diseases have a similar source. The search in waters for the
organisms concerned in these diseases is a matter of the greatest
difficulty, for each belongs to a group of organisms morpho-
logically similar, very widespread in nature, and many of which
have little or no pathogenic action. The biological characters
of these organisms will be given in the chapters devoted to
the diseases in question, but here it may be said that from
the public health standpoint the making of their being found
a criterion for the condemning of a water is impracticable. There
is no doubt that the typhoid and cholera bacteria can exist
for some time in water — at least this has been found to be the
case when sterile water has been inoculated with these bacteria.
But to what extent the same is true when they are placed in
natural conditions, which involve their living in the presence
of other organisms, is unknown, for by no known method can
the presence of either be with certainty demonstrated in the
complex mixtures which occur in nature. With regard to
the typhoid bacillus, of late the tendency has been to seek for
the presence of indirect bacteriological evidence which might
point in the direction of the possibility of the presence of this
organism. The whole question turns on the possibility of
recognising bacteriologically the contamination of water with
<«• wage. Klein and Houston here insist on the fact that in
crude sewage the b. coli -or the members of the coli group are
practically never fewer than 100,000 per c.c., and therefore if
in a water this organism forms a considerable proportion of the
total number of organisms present, then there is great reason
for suspecting sewage pollution. In these circumstances, all
modern work tends to taking the presence of b. coli in a water
as the best indirect evidence of the possibility of disease
organisms of intestinal origin being likely to gain access to
that water. It must, however, be at once clearly recognised that
the presence of members of the coli group is only an indication,
and so far as the potability of any water is concerned, there
is no evidence that these organisms, however undesirable, are
under ordinary circumstances actually harmful to man. In all
inquiries there is the difficulty that at present no means exist
of differentiating between b. coli as derived from the human
inti-stine on the one hand, and from the intestine of animals
on the other. It is thus necessary in reporting upon a water to
havr had an opportunity of inspecting the locality. We have
1 1

162 BACTERIA IN WATER

known cases where a moorland water had a very high content
in b. coli, without there being the remotest possibility
that such came from man. With this proviso, we must inquire
as to what criteria are to be adopted in determining the
significance of the presence of different members of the group
in a water, and here reliance is chiefly to be placed on the
presence of the typical forms of b. coli.

If a sufficient quantity of practically any water be taken,
except, perhaps, that coming from artesian wells, organisms
of the coli group will be found to be present. Therefore, the
question resolves itself into setting up some standards of
relative purity which may be followed in dealing with waters
coming from different sources. These standards are at present
empirical, and different bacteriologists have different views on
the subject. There is, however, a general agreement that deep
well water and the filtered water supplied to urban communities
should be entirely free from b. coli in quantities of 100 c.c. or
less. The great difficulty lies in dealing with river water and
water from shallow and surface wells. Here the usual view
is that the presence of b. coli in 10 c.c. or less is sufficient to
condemn the water. It may be said that under ordinary circum-
stances an inspection of the surroundings and an unfavourable
chemical analysis are sufficient to condemn such a water, for even
if a bacteriological examination showed the absence of b. coli
in large samples, yet the water ought to be condemned ; and
further, if in a suspicious locality the bacteriological analysis
yielded a bad result, the water ought to be condemned even if
from the chemical analysis it could be passed. The difficult
cases are those where the inspection of the locality is satisfactory,
and yet b. coli is present in large numbers. 'Here contamination
is often of animal origin, and the water can after careful inquiry
be passed.

Great care is often necessary in interpreting bacteriological
analysis in consequence of the delicacy of the method. Thus
in examining raw waters, especially those derived from
moorland catchment areas to be used for urban supplies,
bacteriological examinations are relatively of little value, as
storage and filtration will completely alter the bacterial content.
Bacteriological methods are, however, of the greatest value —
much more than mere chemical analysis — in determining the
efficiency of filtration processes.

As the b. coli is fairly widespread in nature, Klein and
Houston hold that valuable supporting evidence is found in
the presence of the b. enteritidis sporogenes and of strepto-

BACTERIOLOGY OF SEWAGE 163

cocci, both of which are probably constant inhabitants of the
human intestine. The spores of the former usually number
100 per c.c. in sewage, and the presence of the latter can always
be recognised in '001 grm. of human faeces. The deductions
to be drawn from the presence of these in water are the same
as those to be drawn from their presence in soil.

It may be said that in water artificially polluted with sewage
containing intestinal bacteria, these can be detected by bacterio-
logical methods in mixtures from ten to a hundred times more
dilute than those in which the pollution can be detected by
purely chemical methods.

Bacteriology of Sewage. — It is sometimes necessary to
examine the bacterial content of sewage, especially in connection
with the efficiency of purification works. The main lines of
inquiry are here the same as for water, and the general methods
are identical, the only modification necessary being that, in
consequence of the high bacterial content, much smaller
quantities of the raw material must be worked with. With
regard to the numbers of bacteria in sewage, these may run
from a million to ten millions or even more per c.c., and here
of course the question of the presence of intestinal organisms
of the coli group is of great importance. The numbers of these
are large, and members of the group may be detected in a
•000001 c.c. or less. The numbers present are frequently
considerably reduced by purification methods, but it is to be
noted that, even when such methods are most successful, b. coli
may yet be present in considerable quantities. This is especially
true in Britain, where sewage is much more concentrated than
it apparently is in America. In the latter country, purification
may yield effluents in which b. coli can be detected in only
'001 c.c. By no purification method has the production of a
potable water been attained, and the high content of effluents
in b. coli makes the passage of typhoid bacilli through a purifica-
tion system possible although the organism has perhaps never
been certainly demonstrated.

The part which bacteria play in the purification of sewage
constitutes a question of great interest, to which much attention
has been directed. The methods adopted for sewage purification
may be divided into two groups. In the first of these, the
sewage coming from the mains is run on to a bed of gravel,
clinker, or coke, on which it is allowed to stand for some hours.
The effluent is then run out through the bottom of the bed,
which is then allowed to rest for some hours before being
recharged. In a modification of this method the sewage is

164 BACTERIA IN WATER

allowed to percolate slowly through a bed consisting of large
porous objects, such as broken bricks or large pieces of coke, and
here the percolation may be constant, no interval of rest being
given. The bacterial processes which take place in these two
methods are, however, probably closely similar. In the second,
the essential feature is a preliminary treatment of the
sewage in more or less closed tanks ("septic tanks"), where
the conditions are supposed to be largely anaerobic. This
method has been adopted at Exeter, Sutton, and Yeovil in this
country, and very fully worked at in America by the State
Board of Health of Massachusetts. In the explanation given
of the rationale of this process, sewage is looked on as exist-
ing in three stages. (1) First of all, fresh sewage — the newly
mixed and very varied material as it enters the main sewers.

(2) Secondly, stale seivage — the ordinary contents of the main
sewers. Here there is abundant oxygen, and as the sewage flows
along there occurs by bacterial action a certain formation of
carbon dioxide and ammonia, which combine to form ammonium
carbonate. This is the sewage as it reaches the purification works.
Here a preliminary mechanical screening may be adopted, after
which it is run into an airtight tank — the septic tank.

(3) It remains there for from twenty-four to thirty-six hours, and
becomes a foul-smelling fluid — the septic sewage. The chemical
changes which take place in the septic tank are of a most complex
nature. The sewage entering it contains little free oxygen, and
therefore the bacteria in the tank are probably largely anaerobic,
and the changes which they originate consist of the formation
of comparatively simple compounds of hydrogen with carbon,
sulphur, and phosphorus. As a result, there is a great reduction
in the amount of organic nitrogen, of albuminoid ammonia, and
of carbonaceous matter. The last is important, as the clogging
of ordinary filter-beds is largely due to the accumulation of such
material, and of matters generally consisting of cellulose. One
further important effect is that the size of the particles of the
deposited matter is decreased, and therefore it is more easily broken
up in the next stage of the process. This consists of running the
effluent from the septic tank on to filter-beds, preferably of coke,
where a further purification process takes place. By this method
there is first an anaerobic treatment, succeeded by an aerobic ;
in the latter the process of nitrification occurs by means of the
special bacteria concerned. The results are of a satisfactory
nature, there being often a marked diminution in the number of
coli organisms present.

In the earlier stages of any sewage purification, there is little

BACTERIOLOGY OF SEWAGE 165

doubt that the albuminous material present is being split up by
ordinary putrefactive bacteria. In the mains and where open
systems of purification are at work, aerobic forms play the chief
part, while in the closed methods anaerobic organisms are those
chiefly concerned. In contact and percolating systems there is
evidence that at first the purifying action of bacteria is materially
furthered by physical processes. Thus Dunbar has shown that
when such a substance as coke is used in a sewage filter-bed a
considerable amount of the albuminous material is removed in
a very few minutes by adsorption, for, albumin, being of a
colloidal nature, is readily deposited under such circumstances
in the pores of the coke in the form of films. After a time such
a filter-bed becomes clogged, but on access of oxygen being
allowed, it regains its adsorptive properties — probably from the
oxidation of the material adsorbed.

During this stage, as in the whole purification process, four,
and it may be five, processes are at work : — First, the action of
ordinary bacteria splitting up the higher albuminous molecules ;
secondly, the action of nitrifying bacteria building up nitrites
and nitrates from ammoniacal products ; thirdly, the action of
denitrifying bacteria which reduce nitrates to lower gaseous
oxides and to free nitrogen (the presence of which in filter beds
can be demonstrated) ; fourthly, the action of higher forms of
vegetable and animal life ; fifthly, it is possible that direct
chemical oxidation of the earlier products of bacterial action
may occur, and in any case the access of an abundant oxygen
supply to adsorbed material hastens its destruction. It is
possible, as is indicated by the work of Lorrain Smith and of
Mair, that perhaps too little weight has been attached to the
parts played by the two last processes specified, for in the later
.stages of the purification process there is a very marked
diminution in the number of bacteria present in the filter.
Much further work, however, is necessary before the part to be
assigned to each factor in operation can be properly estimated.

Further, the details of the essentially bacterial part of the
process are obscure, and the relative parts played, even in an
open purification process, by aerobes on the one hand, and
anaerobes on the other, is little understood. When sewage is
drained off to rest a filter-bed, great quantities of oxygen are
sucked in, but as to how long the bed thus remains aerated,
authorities differ — some maintaining that oxidation processes per-
sist even after the bed has been recharged, while others state that
the oxygen in the resting bed is consumed, and its place
by carbon dioxide and nitrogen. Certainly, at certain

166 ANTISEPTICS

stages of the purification process, large amounts of free nitrogen
come off the bed, but whether at such periods anaerobic bacteria
are or are not in the ascendant, is not known. It is probable
that, from the practical standpoint, the later stages of purification
should take place with free oxidation, as when anaerobic bacteria
are active at this point a very offensive effluent is produced.

Often the effluent from a sewage purification system contains
as many bacteria as the sewage entering, but there is often a
marked diminution. It is said by some that pathogenic
bacteria do not live in sewage. The typhoid bacillus has been
found to die out when placed in sewage, but it certainly can
live in this fluid for a much longer period than that embraced
by any purification method. Thus the constant presence of
b. coli, b. enteritidis, and streptococci which has been observed
in sewage effluents must here still be looked on as indicating a
possible infection with the typhoid bacillus, and it is only by
great dilution and prolonged exposure to the conditions present
in running water that such an effluent can become suitable for
forming a part of a potable water.

ANTISEPTICS.

The death of bacteria is judged of by the fact that, when
they are placed on a suitable food medium, no development
takes place. Microscopically it would be observed that division
no longer occurred, and that in the case of motile species move-
ment would have ceased, but such an observation has only
scientific interest. From the importance of being able to kill
bacteria, an enormous amount of work has been done in the way
of investigating the means of doing so by chemical means, and
the bodies having such a capacity are called antiseptics. It is
now known that the activity of these agents is limited to the
killing of bacteria outside the animal body, but still even this is
of high importance.

Methods. — These vary very much. In early inquiries a great point
was made of the prevention of putrefaction, and work was done in the
way of finding how much of an agent must be added to a given solution
such as beef extract, urine, etc., in order that the bacteria accidentally
present might not develop ; but as bacteria vary in their powers of re-
sistance, the method was unsatisfactory, and now an antiseptic is usually
judged of by its effects on pure cultures of definite pathogenic microbes,
and in the case of a sporing bacterium, the effect on both the vegetative
and spore forms is investigated. The organisms most used are the
staphylococcus pyogenes, streptococcus pyogenes, and the organisms of
typhoid, cholera, diphtheria, and anthrax— the latter being most used

ANTISEPTICS 167

for testing the action on spores. The best method to employ is to take
sloped agar cultures of the test organism, scrape off the growth, and mix
it up with a small amount of distilled water, and filter this emulsion
through a plug of sterile glass wool held in a .small sterile glass funnel,
add a measured quantity of this fluid to a given quantity of a solution
of the antiseptic in distilled water, then after the lapse of the period of
observation to remove one or two loopfuls of the mixture and place them
in a great excess of culture medium. Here it is preferable to use fluid
agar, which is then plated and incubated ; such a procedure is preferable
to the use of bouillon tubes, as any colonies developing can easily be
recognised as belonging to the species of bacterium used. In dealing
with strong solutions of chemical agents it is necessary to be sure that
the culture fluid is in great excess, so that the small amount of the
antiseptic which is transferred with the bacteria may be diluted far
beyond the strength at which it still can have any noxious influence.
Sometimes it is possible at the end of the period of observation to
change the antiseptic into inert bodies by the addition of some other
substance, and then test the condition of the bacteria, and if the inert
substances are fluid there is no objection to this proceeding ; but if in
the process a precipitate results, then it is better not to have recourse
to such a method, as sometimes the bacteria are carried down with the
precipitate and may escape the culture test. The advisability of, when
possible, thus chemically changing the antiseptic was first brought to
notice by the criticism of Koch's statements as to the efficacy of
mercuric chloride in killing the spores of the b. anthracis. The method
he employed in his experiments was to soak silk threads in an emulsion
of anthrax spores and dry them. These were then subjected to the
action of the antiseptic, well washed in water, and laid on the surface of
agar. It was found, however, that, with threads exposed to a far higher
concentration of the corrosive sublimate than Koch had stated was
sufficient to prevent growth, if the salt were broken up by the action of
ammonium sulphide and this washed off, growth of anthrax still occurred
when the threads were laid on agar. The. explanation given was that
the antiseptic had formed an albuminate with the case of each spore, and
that this prevented the antiseptic from acting upon the contained
protoplasm. Such an occurrence only takes place with spores, and the
method given above, in which the small amount of antiseptic adhering
to the bacteria is swamped in an excess of culture fluid, can safely be
followed, especially when a series of antiseptics is being compared.
Kro'nig and Paul introduced what is known as the Garnet method for
testing antiseptics. In this, small garnets of equal size are carefully
cleaned, dipped in an emulsion of anthrax spores, and allowed to dry.
They are then placed in mercuric chloride, and from time to time some
are removed, gently washed, and treated with ammonium sulphide to
decompose the chloride. They are then well shaken in a measured
quantity of water. This is plated, and the number of anthrax colonies
developing is counted.

Ponder and Woodhead have introduced an ingenious apparatus by
which the effects of different concentrations of an antiseptic on the
vitality of such an organism as the b. coli can be automatically
recorded.

Much attention has been paid to the standardisation of antiseptics,
and a watery solution of carbolic acid is now generally taken as the
standard with which other antiseptics are compared. Rideal and
Walker point out that 110 parts by weight of B.P. carbolic acid equal

168 ANTISEPTICS

100 parts by weight of phenol, and they recommend the following method
of standardising: To 5 c.c. of a particular dilution of the disinfectant
add 5 drops of a 24-hour-old bouillon culture of the organism (usually
b. typhosus), which has been incubated at 37° C. Shake the mixture
and make subcultures every 2£ minutes to 15 minutes. Perform a
parallel series of experiments with carbolic acid, and express the
comparative result in multiples of the carbolic acid doing the same
work.

The Action of Antiseptics. — In inquiries into the actions
of antiseptics attention to a great variety of factors is necessary,
especially when the object is not to compare different antiseptics
with one another, but when the absolute value of any body is
being investigated. Thus the medium in which the bacteria to
be killed are situated is important ; the more albuminous it
is, the greater degree of concentration is required. Again,
the higher the temperature at which the action is to take
place, the more dilute may the antiseptic be, or the shorter the
exposure necessary for a given effect to take place. The most
important factor, however, to be considered is the chemical
nature of the substances employed. Chick has shown that the
action of a disinfectant upon a bacterium presents close
analogies with the interaction of simple chemical substances,
such as an acid and an alkali. In the case of anthrax spores,
during the first few minutes a great fatality occurs, after which
the action of the antiseptic gradually tails off. With certain
other organisms, however, such as the paratyphoid bacillus,
the presence in a culture — especially in a young culture — of
highly resistant forms renders the initial action of an antiseptic
less marked. The action of an antiseptic, like the action of an
acid and an alkali, is very much increased by raising the
temperature ; from which follows the practical conclusion that,
in any disinfection, the use of warm solutions is advisable.
Chick and C. J. Martin have further investigated the fact that
the presence of albuminous material in a mixture of disinfectant
and bacteria decreases the action of the disinfectant, and
consider that the latter is adsorbed by the albumin. They have
shown grounds for believing that a disinfectant in an emulsion-
ised form is more efficient than a similar disinfectant in actual
solution, because of a similar phenomenon occurring ; for,
just as a disinfectant may be put out of action by being
adsorbed by organic particles, so when these organic particles
happen to be bacteria, the adsorption process causes a greater
concentration of the antiseptic round the bacterial protoplasm,
and thus hastens its death.

Though nearly every substance which is not a food to the

THE ACTION OF ANTISEPTICS 169

animal or vegetable body is more or less harmful to bacterial
life, yet certain bodies have a more marked action than others.
Thus it may be said that the most important antiseptics are the
salts of the heavy metals, certain acids, especially mineral acids,
certain oxidising and reducing agents, a great variety of sub-
stances belonging to the aromatic series, and volatile oils generally.
In comparing different bodies belonging to any one of these
groups the chemical composition or constitution is very important,
and if such comparisons are to be made, the solutions compared
must be equimolecular ; in other words, the action of a molecule
of one body must be compared with the action of a molecule of
another body. This can be done by dissolving the molecular
weight in grammes in, say, a litre of water (see p. 34). When
this is done, important facts emerge. Thus, generally speaking,
the compounds of a metal of high atomic weight are more
powerful antiseptics than those of one belonging to the same
series, but of a lower atomic weight. Among organic bodies,
;iur tin, substances with high molecular weight are more powerful
than those of low molecular weight — thus butyric alcohol is more
powerful than ethylic alcohol — and important differences among
the aromatic bodies are associated with their chemical constitu-
tion. Thus among the cresols the ortho- and para-bodies re-
semble each other in general chemical properties, and stand apart
from metacresol ; they also are similar in antiseptic action, and
are much stronger than the meta-body. The same may be
observed in other groups of ortho-, meta-, and para-bodies.
Again, such a proj>erty as acidity is important in the action of a
substance, and, generally speaking, the greater the avidity of an
acid to combine with an alkali, the more powerful an antiseptic
it is. With regard to oxidising agents and reducing agents,
probably the possession of such properties has been overrated as
increasing bactericidal potency. Thus in the case of such re-
ducers as sulphurous acid and formic acid, the effect is apparently
chiefly due to the fact that these substances are acids. Formic
acid is much more efficient than formate of sodium. In the case
of permanganate of potassium, which is usually taken as the
type of oxidising agents in this connection, it can be shown that
the greater amount of the oxidation which takes place when this
agent is brought into contact with bacteria occurs after the
organisms are killed. Such an observation is, however, not
conclusive as to the non-efficiency of the oxidation process, t < »
the death of the bacteria might be due to the oxidation of a
\<TY small part of the bacterial protoplasm. Apart from the
chemical nature of antiseptic agents, the physical factors con-

170 ANTISEPTICS

cerned in their solution, especially when they are electrolytes,
probably play a part in their action. The part played by such
factors is exemplified in the important fact that a strong solution
acting for a short time will have the same effect as a weaker
solution acting for a longer time. From what has been said it
will be realised that the real causes of a material being an
antiseptic are very obscure, and at present we can only have a
remote idea of the factors at work.

The Effects of certain Antiseptics. — Here we can only
briefly indicate certain results obtained with the more common
members of the group.

Chlorine. — All the halogens have been found to be powerful
antiseptics, but from the cheapness with which it can be produced
chlorine has been most used ; not only is it the chief active
agent in the somewhat complex action of bleaching powder, but
it is also the chief constituent of several proprietary substances,
of which " Electrozone " is a good example. This last substance
is made from electrolysing sea- water, when magnesia, and chlorine
being liberated, magnesium hypochlorite and magnesium chloride
are formed. In the action of this substance free hypochlorous
acid is formed, and the effect produced is thus similar to that
of bleaching powder. Nissen, investigating the action of the
latter, found that 1J per cent, killed typhoid bacilli in faeces ;
and Rideal found that 1 part to 400-500 disinfected sewage in
fourteen minutes, and Delepine's results show that 1 part to 50
(equal to '66 per cent, of chlorine) rapidly kills the tubercle
bacillus, and 1 part to 10 (equal to 3*3 per cent.) killed anthrax
spores. Klein found that '05 per cent, of chlorine killed most
bacterial spores in five minutes.

Iodine Terchloride. — This is a very unstable compound of
iodine and chlorine, and, seeing that the substance only remains
as IC13 in an atmosphere of chlorine gas, it is open to doubt
whether the antiseptic effects attributed to it are not due to a
very complicated action of free hydrochloric acid, hydriodic acid,
of oxyacids of chlorine and iodine produced by its decomposition,
and also, in certain cases, of organic iodine compounds formed
from its contact with albuminous material. It is stated that the
action is very potent : a 1 per cent, solution is said instantly to
kill even anthrax spores, but if the spores be in bouillon, death
occurs after from ten to twelve minutes. In serum the necessary
exposure is from thirty to forty minutes. A solution of 1-1000
will kill the typhoid, cholera, and diphtheria organisms in five
minutes.

Nascent Oxygen. — This is chiefly available in two ways — firstly,

THE EFFECTS OF CERTAIN ANTISEPTICS 171

when in the breaking up of ozone the free third atom of the
ozone molecule is seeking to unite with another similar atom ;
secondly, when peroxide of hydrogen is broken up into water
and an oxygen atom is thereby liberated. In commerce the
activity of "Sanitas" compounds is due to the formation of
ozone by the slow oxidation of the resin, camphor, and thymol
they contain.

Perchloride of Mercury. — Of all the salts of the heavy metals
this has been most widely employed, and must be regarded as
one of the most powerful and useful of known antiseptics. In
testing its action on anthrax spores there is no doubt that in the
earlier results its potency was overrated from a neglect of the
fact already alluded to, that in the spore-case an albuminate of
mercury was formed which prevented the contained protoplasm
from developing, while not depriving it of life. It has been
found, however, that this salt in a strength of 1-100 will kill the
spores in twenty minutes, although an hour's exposure to 1-1000
has no effect. The best results are obtained by the addition to
the corrosive sublimate solution of '5 per cent, of sulphuric acid
or hydrochloric acid ; the spores will then be killed by a seventy-
minute exposure to a 1-200 solution. When, however, organisms
in the vegetative condition are being dealt with, much weaker
solutions are sufficient; thus anthrax bacilli in blood will be
killed in a few minutes by 1-2000, in bouillon by 1-40,000, and
in water by 1-500,000. Plague bacilli are killed by one to two
minutes' exposure to 1-3000. Generally speaking, it may be said
that a 1-2000 solution must be used for the practically instan-
taneous killing of vegetative organisms.

Perchloride of mercury is one of the substances which have
been used for disinfecting rooms by distributing it from a spray
producer, of which the Equifex may be taken as a type. With
such a machine it is calculated that 1 oz. of perchloride of
mercury used in a solution of 1-1000 will probably disinfect 3000
square feet of surface. Such a procedure has been extensively
used in the disinfection of plague houses, but the use of a stronger
solution (1-500 acidulated) is probably preferable.

Formalin as a commercial article is a 40 per cent, solution of
formaldehyde in water. This is a substance which of late years
has come much into vogue, and it is undoubtedly a valuable
antiseptic. A disadvantage, however, to its use is that, when
diluted and exposed to air, amongst other changes which it
undergoes it may be transformed, under little understood
conditions, into trioxymethylene and paraformaldehyde, these
being polymers of formaldehyde. The bactericidal values of these

172 ANTISEPTICS

mixtures are thus indefinite. Formalin may be used either by
applying it in its liquid form or as a spray, or the gas which
evaporates at ordinary temperatures from the solution may be
utilised. To disinfect such an organic mixture as pus containing
pyogenic organisms, a 10 per cent, solution acting for half an
hour is necessary. In the case of pure cultures, a 5 per cent,
solution will kill the cholera organism in three minutes, anthrax
bacilli in a quarter of an hour,- and the spores in five hours.
When such organisms as pyogenic cocci, cholera spirillum, and
anthrax bacillus infect clothing, an exposure to the full strength
of formalin for two hours is necessary, and in the case of anthrax
spores, for twenty-four hours. Silk threads impregnated with
the plague bacillus were found to be sterile after two minutes'
exposure to formalin.

The action of formalin vapour has been much studied, as its
use constitutes a cheap method of treating infected rooms, in
which case some spray-producing machine is employed. It is
stated that a mixture of 8 c.c. of formalin with 48 c.c. of water
is sufficient when vaporised to disinfect one cubic metre, so far
as non-sporing organisms are concerned. It is also stated that

I part formalin in 10,000 of air will kill the cholera vibrio in
one hour, diphtheria bacillus in three hours, the staphylococcus
pyogenes in six hours, and anthrax spores in thirteen hours. In
the .case of organisms which have become dry it is probable,
however, that much longer exposures are necessary, but on this
point we have not definite information.

Formalin gas has only a limited application ; it has little
effect on dry organisms, and in the case of wet organisms, in
order to be effective, probably must become dissolved so as to
give the moisture a proportion analogous to the strengths stated
above with regard to the vapour.

Sulphurous Acid. — This substance has long been in use,
largely from the cheapness with which it can be produced by
burning sulphur in the air. An atmosphere containing -98 per
cent, will kill the pyogenic cocci in two minutes if they are wet,
and in twenty minutes if they are dry ; and anthrax bacilli are
killed by thirty minutes' exposure, but to kill anthrax spores an
exposure of from one to two hours to an atmosphere containing

I 1 per cent, is necessary. For a small room the burning of about
a pound and a half (most easily accomplished by moistening the
sulphur with methylated spirit) is usually considered sufficient.
It has been found that if bacteria are protected, e.g. when they
are in the middle of small bundles of clothes, no effect is
produced even by an atmosphere containing a large proportion

THE EFFECTS OF CERTAIN ANTISEPTICS 173

of the sulphurous acid gas. The practical applications of this
agent are therefore limited.

Potassium Permanganate. — The action of this agent very
much depends on whether it can obtain free access to the
bacteria to be killed or whether these are present in a solution
containing much organic matter. In the latter case the oxidation
of the organic material throws so much of the salt out of action
that there may be little left to attack the organisms. Koch
found that to kill anthrax spores a 5 per cent, solution required
to act for about a day ; for most organisms a similar solution
acting for shorter periods has been found sufficient, and in the
cases of the pyogenic cocci a 1 per cent, solution will kill in ten
minutes. There is little doubt that such weaker solutions are of
value in disinfecting the throat on account of their non-irritating
properties, and good results in this connection have been obtained
in cases of diphtheria. A solution of 1 in 10,000 has been found
to kill plague bacilli in five minutes.

Carbolic Acid. — Of all the aromatic series this is the most
extensively employed antiseptic. All ordinary bacteria in the
vegetative condition, and of these the staphylococcus pyogenes
is the most resistant, are killed in less than five minutes by a
2-3 per cent, solution in water, so that the 5 per cent, solution
usually employed in surgery leaves a margin of safety. But for
the killing of such organisms as anthrax spores a very much
longer exposure is necessary ; thus Koch found it necessary to
expose these spores for four days to ensure disinfection. The
risk of such spores being present in ordinary surgical procedure
may be overlooked, but there might be risk of tetanus spores
not being killed, as these will withstand fifteen hours' exposure
to a 5 per cent, solution.

In the products of the distillation of coal there occur, besides
carbolic acid, many bodies of a similar chemical constitution, and
many mixtures of these are in the market — the chief being cyllin,
izal, and lysol, all of which are agents of value. Of these lysol
is perhaps the most noticeable, as from its nature it acts as a
soap, and thus can remove fat and dirt from the hands. A one-
third per cent, solution is said to destroy the typhoid and cholera
organisms in twenty minutes. A one per cent, solution is sufficient
for ordinary surgical procedures.

lodoform. — This is an agent regarding the efficacy of which
there has been much dispute. There is little doubt that it owes
its efficiency to its capacity for being broken up by bacterial
action in such a way as to set free iodine, which acts as a
powerful disinfectant. The substance is therefore of value in the

174 ANTISEPTICS

treatment of foul wounds, such as those of the mouth and
rectum, where reducing bacteria are abundantly present. It acts
more slightly where there are only pyogenic cocci, and it seems
to have a specially beneficial effect in tubercular affections. In
certain cases its action may apparently be aided by the presence
of the products of tissue degeneration.

From the results which have been given it will easily be
recognised that the choice of an antiseptic and the precise
manner in which it is to be employed depend entirely on the
environment of the bacteria which are to be killed. In many
cases it will be quite impossible, without original inquiry, to say
what course is likely to be attended with most success.

CHAPTER VI.

RELATIONS OF BACTERIA TO DISEASE— THE
PRODUCTION OF TOXINS BY BACTERIA.

Introductory. — It has already been stated that a strict division
of micro organisms into saprophytes and true parasites cannot be
made. No doubt there are organisms, such as the bacillus of
leprosy, which as yet have not been cultivated outside the animal
body, and others, such as the gonococcus, which are in natural
conditions always parasites associated with disease. But these
latter can lead a saprophytic existence in specially prepared
conditions, and there are many of the disease - producing
organisms, such as the organisms of typhoid and cholera, which
can nourish readily outside the body, even in ordinary con-
ditions. The conditions of growth are, however, of very great
importance in the study of the modes of infection in the various
diseases, though they do not form the basis of a scientific
division.

A similar statement applies to the terms pathogenic and
saprophytic, and even to the terms pathogenic and non-patho-
genic. By the term pathogenic is meant the power which an
organism has of producing morbid changes or effects in the
animal body, either under natural conditions or in conditions
artificially arranged, as in direct experiment. Now we know of
no organisms which will in all circumstances produce disease in
all animals, and, on the other hand, many bacteria described as
harmless saprophytes will produce pathological changes if intro-
duced in sufficient quantity. When, therefore, we speak of a
pathogenic organism, the term is merely a relative one, and
indicates that in certain circumstances the organism will produce
disease, though in the science of human pathology it is often
used for convenience as implying that the organism produces
disease in man in natural conditions.

Modifying Conditions. — In studying the pathogenic effects in
any instance, both the micro-organisms and the animal affected

175

176 RELATIONS OF BACTERIA TO DISEASE

must be considered, and not only the species of each, but also
its exact condition at the time of infection. In other words,
the resulting disease is the product of the sum-total of the
characters of the infecting agent, on the one hand, and of the
subject of infection, on the other. We may, therefore, state
some of the chief circumstances which modify each of these
two factors involved, and, consequently, the diseased condition
produced.

1. The Infecting Agent. — In the case of a particular species
of bacterium its effect will depend chiefly upon (a) its virulence,
and (b) the number introduced into the body. To these may
be added (c) the path of infection.

The virulence, i.e. the power of multiplying in the body and
producing disease, varies greatly in different conditions, and the
methods by which it can be diminished or increased will be
afterwards described (vide Chapter XXI.). One important
point is that when a bacterium has been enabled to invade
and multiply in the tissues of an animal, its virulence for that
species is often increased. This is well seen in the case of
certain bacteria which are normally present on the skin or
mucous surfaces. Thus it has been repeatedly proved that the
bacillus coli cultivated from a septic peritonitis is much more
virulent than that taken from the bowel of the same animal.
The virulence may be still more increased by inoculating from
one animal to another in series — the method of passage. Widely
different effects are, of course, produced on the virulence being
altered. For example, a streptococcus which produces merely
a local inflammation or suppuration, may produce a rapidly
fatal septicaemia when its virulence is raised. Virulence also
has a relation to the animal employed, as occasionally on being
increased for one species of animal it is diminished for another.
For example, streptococci, on being inoculated in series through
a number of mice, acquire increased virulence for these animals,
but become less virulent for rabbits (Knorr). The theoretical
consideration of virulence must be reserved for a later chapter
(see Immunity).

The number of the organisms introduced, i.e. the dose of the
infecting agent, is another point of importance. The healthy
tissues can usually resist a certain number of pathogenic
organisms of given virulence, and it is only in a few instances
that one or two organisms introduced will produce a fatal
disease, e.g. the case of anthrax in white mice. The healthy
peritoneum of a rabbit can resist and destroy a considerable
number of pyogenic micrococci without any serious result, but

CONDITIONS MODIFYING PATHOGENICITY 177

if a larger dose be introduced, a fatal peritonitis may follow.
Again, a certain quantity of a particular organism injected
subcutaneously may produce only a local inflammatory change,
but in the case of a larger dose the organisms may gain entrance
to the blood stream and produce septicaemia. There is, there-
fore, for a particular animal, a minimum lethal dose which can
be determined by experiment only ; a dose, moreover, which is
modified by various circumstances difficult to control.

The path of infection may alter the result, serious effects often
following a direct entrance into the blood stream. Staphylo-
cocci injected subcutaneously in a rabbit may produce only a
local abscess, whilst on intravenous injection multiple abscesses
in certain organs may result and death may follow. Local
inflammatory reaction with subsequent destruction of the
organisms may be restricted to the site of infection or may
occur also in the related lymphatic glands. The latter
therefore act as a second barrier of defence, or as a filtering
mechanism which aids in protecting against blood infection.
This is well illustrated in the case of "poisoned wounds." In
some other cases, however, the organisms are very rapidly
destroyed in the blood stream, and Klemperer has found that,
in the dog, subcutaneous injection of the pneumococcus produces
death more readily than intravenous injection.

'2. The. Xnhject of Infection. — Amongst healthy individuals
susceptibility and, in inverse ratio, resistance to a particular
microbe may vary according to (a) species, (b) race .and in-
dividual peculiarities, (c) age. Different species of the lower
animals show the widest variation in this respect, some being
extremely susceptible, others highly resistant. Then there are
diseases, such as leprosy, gonorrhoea, etc., which appear to be
peculiar to the human subject and have not yet been trans-
mitted to animals. And further, there are others, such as
cholera and typhoid, which do not naturally affect animals,
and the typical lesions of which cannot be experimentally
reproduced in them, or appear only imperfectly, although
pathogenic effects follow inoculation with the organisms. In
the case of the human subject, differences in susceptibility to a
certain disease are found amongst different races, and also amongst
individuals of the same race, as is well seen in the case of tubercle
and other diseases. Age also plays an important part, young
subjects being more liable to certain diseases, e.g. to diphtheria.
Kurt her. at different periods of life certain parts of the body are
more susceptible, for example, in early life, the bones and joints
to tubercular and acute suppurative affections.

178 RELATIONS OF BACTERIA TO DISEASE

In increasing the susceptibility of a given individual, con-
ditions of local or general diminished vitality play the most
important part. It has been experimentally proved that
conditions such as exposure to cold, fatigue, starvation, etc.,
all diminish the natural resistance to bacterial infection. Rats
naturally immune to glanders can be rendered susceptible by
being fed with phloridzin, which produces a sort of diabetes,
a large amount of sugar being excreted in the urine (Leo).
Guinea-pigs may resist subcutaneous injection of a certain dose
of the typhoid bacillus, but if at the same time a sterilised
culture of the bacillus coli be injected into the peritoneum, they
quickly die of a general infection. Also a local susceptibility
may be produced by injuring or diminishing the vitality of a
part. If, for example, previous to an intravenous injection of
staphylococci, the aortic cusps of a rabbit be injured, the
organisms may settle there and set up an ulcerative endocarditis ;
or if a bone be injured, they may produce suppuration at the
part, whereas in ordinary circumstances these lesions would not
take place.

Such facts, established by experiment (and many others
might be given), illustrate the important part which local or
general conditions of diminished vitality may play in the pro-
duction of disease in the human subject. This has long been
known by clinical observation. In normal conditions the blood
and tissues of the body, with the exception of the skin and
certain of the mucous surfaces, are bacterium-free, and if a few
organisms gain entrance, they are destroyed. But if the vitality
becomes lowered, their entrance becomes easier and the possibility
of their multiplying and producing disease greatly increased.
In this way the favouring part played by fatigue, cold, etc.,
in the production of diseases of which the direct cause is a
bacterium, may be understood. It is important to keep in view
in this connection that many of the inflammation-producing
and pyogenic organisms .are normally present on the skin and
various mucous surfaces. The action of a certain organism
may devitalise the tissues to such an extent as to pave the
way for the entrance of other bacteria ; we may mention the
liability of the occurrence of pneumonia, erysipelas, and various
suppurative conditions in the course of or following infective
fevers. In some cases the specific organism may produce lesions
through which the other organisms gain entrance, e.g. in typhoid,
diphtheria, etc. A notable example of diminished resistance to
bacterial infection is seen in the case of diabetes ; tuberculosis
and infection with pyogenic organisms are prone to occur in

MODES OF BACTERIAL ACTION 179

this disease, and are of a severe character. It is not uncommon
to find in the bodies of those who have died from chronic
wasting disease, collections of uiicrococci or bacilli in the
capillaries of various organs, which have entered in the later
hours of life; that is to say, the bacterium-free condition of
the blood has been lust in the period of prostration preceding
death.

The methods by which the natural resistance may be speci-
fically increased belong to the subject of immunity, and are
dc-MTibed in the chapter on that subject.

Modes of Bacterial Action. — In the production of disease by
micro-organisms there are two main factors involved, namely,
(a) the multiplication of the living organisms after they have
entered tin- body, and (6) the production by them of poisons
\\hirh may act both upon the tissues around and upon the body
generally. The former corresponds to infection, the latter is of
the nature of intoxication or poisoning. In different diseases
one of these is usually the more prominent feature, but both are
always more or less concerned.

1. Inflation and Distribution of the Bacteria in the Body. —
After pathogenic bacteria have invaded the tissues, or in other
words, after infection by bacteria has taken place, their further
behaviour varies greatly in different cases. In certain cases
they may reach and multiply in the blood stream, producing a
fatal septicaemia. In the lower animals this multiplication of
the organisms in the blood throughout the body may be very
extensive (for example, the septicaemia 'produced by the pneunio-
coccus in rabbits) ; but in septicaemia in man it very seldom, if
ever, occurs to so great a degree, the organisms rarely remain in
large numbers in the circulating blood, and their detection in it
during life by microscopic examination is rare, and even culture
methods may give negative results unless a large amount of blood
is used. In such cases, however, the organisms may be found
}><>.<t mortem lying in large numbers within the capillaries of
various organs, e.g. in cases of septicaemia produced by strepto-
cocci. In the human subject more frequently one of two things
happens. In the first place, the organisms may remain local,
producing little reaction around them, as in tetanus, or a well-
marked lesion, as in diphtheria, etc. Or in the second place,
they may pa<s by the lymph or blood stream toother parts or
organs in which they settle, multiply, and produce lesions, as in
tubercle.

L'. /'/•'"///••//"// of CJwmical Poisons. — In all these cases the
growth of the organisms is accompanied by the formation of

180 RELATIONS OF BACTERIA TO DISEASE

chemical products, which act generally or locally in varying degree
as toxic substances. The toxic substances become diffused
throughout the system, and their effects are manifested chiefly
by symptoms such as the occurrence of fever, disturbances of
the circulatory, respiratory, and nervous systems, etc. In some
cases corresponding changes in the tissues are found, for example,
the changes in the nervous system in diphtheria, to be afterwards
described. The general toxic effects may be so slight as to be
of no importance, as in the case of a local suppuration ; or they
may be very intense, as in tetanus; or again, less severe but
producing cachexia by their long continuance, as in tuberculosis.

The occurrence of local tissue changes or lesions produced in
the neighbourhood of the bacteria, as already mentioned, is one
of the most striking results of bacterial action, but these also
must be traced to chemical substances formed in or around the
bacteria, and either directly or through the medium of ferments.
In this case it is more difficult to demonstrate the mode of
action, for in the tissues the chemical products are formed by
the bacteria slowly, continuously, and in a certain degree of
concentration, and these conditions cannot be exactly repro-
duced by experiment. It is also to be noted that more than one
poison may be produced by a given bacterium, e.g. the tetanus
bacillus (p. 424). Further, it is very doubtful whether all the
chemical substances formed by a certain bacillus growing in the
tissues are also formed by it in cultures outside the body (vide
p. 191). The separated toxin of diphtheria, like various vegetable
and animal toxins, however, possesses a local toxic action of
very intense character, evidenced often by extensive necrotic
change.

The injection of large quantities of many different pathogenic
organisms in the dead condition results in the production of a
local inflammatory change which may be followed by suppura-
tion, this effect being possibly brought about by certain sub-
stances in the bacterial protoplasm common to various species,
or at least possessing a common physiological action (Buchner
and others). When dead tubercle bacilli, however, are intro-
duced into the blood stream, nodules do result in certain parts
which have a resemblance to ordinary tubercles. In this case
the bodies of the bacilli evidently contain a highly resistant and
slowly acting substance which gradually diffuses around and
produces effects (vide Tuberculosis).

Summary. — We may say, then, that the action of bacteria as
disease-producers, as in fact their power to exist and multiply in
the living body, depends upon the chemical products formed

TISSUE CHANGES PRODUCED BY BACTERIA 181

directly or indirectly by them. This action is shown by tissue
'•//'////AN produced in the vicinity of the bacteria or throughout
thr system, and l»y /o.r/V *// mjitunis of great variety of degree and
character.

We shall first consider the effects of bacteria on the body
generally, and afterwards the nature of the chemical products.

KI-TKCTS OF BACTERIAL ACTION.

These may l»e for convenience arranged in a tabular form as
follows :

A. Tissue Changes.

(1) Local changes, i.e. changes produced in the neigh-

bourhood of the bacteria.

Position (a) At primary lesion.
(6) At secondary foci.

Character (a) Tissue reactions \ Acute or

(b) Degeneration and necrosis/ chronic.

(2) Produced at a distance from the bacteria, directly or

indirectly, by the absorption of toxins.

(a) In special tissues —

(a) as the result of damage, e.g. nerve cells
and fibres, secreting cells, vessel walls, or

((3) changes of a reactive nature in the blood-
forming organs.

(b) General anatomical changes, the effects of

malnutrition or of increased waste.

B. Symptoms and Cfianyes in Metabolism.

The occurrence of fever, of errors of assimilation and
elimination, etc.

A. Tissue Changes produced by Bacteria. — The effects of
I 'arterial action are so various as to include almost all known
pathological changes. However varied in character, they may
be classified under two main headings : (a) those of a degenera-
tive or necrotic nature, the direct result of damage ; and (b) those
of reactive nature, defensive or reparative. The former are the
r.\l res<i<m of the necessary vulnerability of the tissues, the latter
of protective powers evolved for the benefit of the organism. In
the means of defence both leucocytes and the fixed cells of the

182 RELATIONS OF BACTERIA TO DISEASE

tissues are concerned. Both show phagocytic properties, i.e.
have the power of taking up bacteria into their protoplasm. The
cells are guided towards the focus of infection by chemiotaxis,
and thus we find that different bacteria attract different cells.
The most rapid and abundant supply of phagocytes is seen in
the case of suppurative conditions where the neutrophile leuco-
cytes of the blood are chiefly concerned. When the local lesion
is of some extent there is usually an increase of these cells
in the bloo'd — a neutrophile leucocytosis. And further, observa-
tion has shown that associated with this there is in the
bone-marrow an increased number of the mother-cells of these
leucocytes — the neutrophile myelocytes. The passage of the
neutrophile leucocytes from the marrow into the blood, with the
resulting leucocytosis, is also apparently due to the absorbed
bacterial toxins acting chemiotactically on the marrow. These
facts abundantly show that the means of defence is not a mere
local mechanism, but that increased proliferative activity in
distant tissues is called into play. In addition to direct phago-
cytosis by these leucocytes, there is now abundant evidence that
an important function is the production in the body of bactericidal
and other antagonistic substances. In other cases the cells
chiefly involved are the mononuclear hyaline leucocytes, and
with them the endothelial cells, e.g. of serous membranes, often
play an important part in the defence ; this is well seen in
typhoid fever, where the specific bacillus appears to have little
or no action on the neutrophile leucocytes. In other cases,
again, the reaction is chiefly on the part of the connective
cells, though their proliferation is always associated with some
variety of leucocytic infiltration and usually also with the forma-
tion of new blood vessels. Such a connective tissue reaction
occurs especially in slow infections or in the later stages of an
acute infection. The tissue changes resulting from cellular
activity in the presence of bacterial invasion are naturally very
varied, — examples of this will be found in subsequent chapters, —
but they may be said to be manifestations of the two funda-
mental processes of (a) increased functional activity — movement,
phagocytosis, secretion, etc.— and (b) increased formative activity
— cell growth and division. The exudation from the blood
vessels has been variously interpreted. There is no doubt that
the exudate has bactericidal properties and also acts as a diluting
agent, but it must still be held as uncertain whether the process
of exudation ought to be regarded as primarily defensive or as
the direct result of damage to the endothelium of the vessels.
It may also be pointed out that the various changes referred to

LOCAL LESIONS 183

are none of them peculiar to bacterial invasion ; they are examples
of the general laws of tissue change under abnormal conditions,
and they can all be reproduced by chemical substances in solution
or in a particulate state. What constitutes their special feature
is their progressive or spreading nature, due to the bacterial
multiplication.

(1) Local Lesions. — In some diseases the lesion has a special
site; for example, the lesion of typhoid fever and, to a less
extent, that of diphtheria. In other cases it depends entirely
upon the point of entrance, e.g. malignant pustule and the con-
ditions known as wound infections. In others again, there is a
special tendency for certain parts to be affected, as the upper
parts of the lungs in tubercle. In some cases the site has a
mechanical explanation.

When organisms gain an entrance to the blood from a primary
lesion, the organs specially liable to be affected vary greatly in
different diseases. Pyogenic cocci show a special tendency to
settle in the capillaries of the kidneys and produce miliary
abscesses, whilst these lesions rarely occur in the spleen. On
the other hand, the nodules in disseminated tubercle or glanders
are much more numerous in the spleen than in the kidneys,
which in the latter disease are usually free from them. The
important point is that the position of the disseminated lesions
is not to be explained by a mechanical process, such as embolism,
but depends upon a special relation between the organisms and
the tissues, which may be spoken of either as a selective power
on the part of the organisms or a special susceptibility of tissues,
possibly in part due to their affording to the organisms more
suitable conditions of nutriment. Even in the case of the
lesions produced by dead tubercle bacilli, a certain selective
character is observed.

Acute Local Lesions. — The local inflammatory reaction presents
different characters in different conditions. It may be accom-
panied by abundant fibrinous exudation, or by great catarrh (in
the case of an epithelial surface), or by haemorrhage, or by
redema ; it may be localised or spreading in character ; it may
be followed by suppuration, and may lead up to necrosis of
the tissues of the part, a good example of the latter event
being found in a boil. Examples will be given in subsequent
chapters. The necrotic or degenerative changes affecting
especially the more highly developed elements of tissues are
chiefly produced by the direct action of the bacterial poisons,
though aided by the disturbances of nutrition involved in the
vascular phenomena. It may here be pointed out that a well-

184 RELATIONS OF BACTERIA TO DISEASE

marked inflammatory reaction is often found in animals which
occupy a medium position in the scale of susceptibility, and that
an organism which causes a general infection in a certain animal
may produce only a local inflammation when its virulence is
lessened.

Chronic Local Lesions. — In a considerable number of diseases
produced by bacteria the local tissue reaction is a more chronic
process than those described ; there is less vascular disturbance
and a greater preponderance of the proliferative processes, lead-
ing to new formation of connective tissue. This formation
may occur in foci here and there, so that nodules result, or it may
be more diffuse. Such changes especially occur in the diseases
often known as the infective granulomata, of which tubercle,
leprosy, glanders, actinomycosis, syphilis, etc., are examples.
A hard-and-fast line, however, cannot be drawn between such
conditions and those described above as acute. In glanders, for
example, especially in the human subject, the lesion often
approaches very nearly to an acute suppurative change, and
sometimes actually is of this nature. Whilst in these diseases
the fundamental change is the same — namely, a reaction to an
irritant of minor intensity — the exact structural characters and
arrangement vary in different diseases. In some cases the
disease may be identified by the histological changes alone, but
on the other hand, this is often impossible.

(2) General Lesions produced by Toxins. — In the various in-
fective conditions produced by bacteria, changes commonly
occur in certain organs unassociated with the presence of the
bacteria ; these are produced by the action of bacterial products
circulating in the blood. Many such lesions can be produced
experimentally. The secreting cells of various organs, especially
the kidney and liver, are specially liable to change of this kind.
Cloudy swelling, which may be followed by fatty change or
by actual necrosis with granular disintegration, is common.
Hyaline change in the walls of arterioles may occur, and in
certain chronic conditions amyloid change is brought about in
a similar manner. The latter has been produced in animals
by repeated injections of the staphylococcus aureus. Capillary
haemorrhages are not uncommon, and are in many cases due to
an increased permeability of the vessel walls, aided by changes
in the blood plasma, as evidenced sometimes by diminished
coagulability. Similar haemorrhages may follow the injection of
some bacterial toxins, e.g. of diphtheria, and also of vegetable
poisons, e.g. ricin and abrin. Skin eruptions occurring in the
exanthemata are probably produced in the same way, though in

DISTURBANCES OF METABOLISM, ETC. 185

many of the«r diseases the causal organism has not yet been
isolated. \\ e have, however, the important fact that corre-
sponding skin eruptions may be produced by poisoning with
certain drugs. In the nervous system degenerative changes
have been found in diphtheria, both in the spinal cord and in
the peripheral nerves, and have been reproduced experimentally
by the products of the diphtheria bacilli. There is also experi-
mental evidence that the bacillus coli communis and the strepto-
coccus pyogenes may, by means of their products, produce areas
of softening in the spinal cord, and this may furnish an ex-
planation of some of the lesions found clinically. It is also
p »-ilile that some serous inflammations may be produced in the
same way.

B. Disturbances of Metabolism, etc. — It will easily be
realised that such profound tissue changes as have been detailed
cannot occur without great interference with the normal bodily
metabolism. General malnutrition and cachexia are of common
occurrence, and it is a striking fact found by experiment that
alter injection of bacterial products, e.g. of the diphtheria
bacillus, a marked loss of body weight often occurs which may
be progressive, leading to the death of the animal. In bacterial
disease assimilation is often imperfect, for the digestive glands
are affected, it may be, by actual poisoning by bacterial products, it
may be by the occurrence of fever, and excretion is interfered with
by the damage done to the excretory cells. But of all the changes
in metabolism the most difficult to understand is the occurrence
of that interference with the heat-regulating mechanism which
results in fever. The degree and course of the latter vary,
sometimes conforming to a more or less definite type, where the
bacilli are selective in their field of operation, as in croupous
pneumonia or typhoid, sometimes being of a very irregular kind,
especially when the bacteria from time to time invade fresh
areas of the body, as in pyaemic affections. The main point of
interest regarding the development of fever is as to whether it is
a direct effect of the circulation of bacterial toxins, or if it is to
be looked on as part of the reaction of the body against the
irritant. This question has still to be settled, and all that we
can do is to adduce certain facts bearing on it. Thus in diph-
theiia and tetanus, where toxic action leading to degeneration
plays such an important part, fever may be a very subsidiary
feature, except in the terminal stage of the latter disease ; and
in fact in diphtheria profoundly toxic effects may be produced
with little or no interference with heat regulation. On the
other hand, in bacterial disease, where defensive and reparative

186 EELATIONS OF BACTERIA TO DISEASE

processes predominate, fever is rarely absent, and it is nearly
always present when there is an active leucocytosis going on.
In this connection it may be remarked that several observers
have found that, when a relatively small amount of the dead
bodies of certain bacteria are injected into an animal, fever
occurs ; while the injection of a large amount of the same is
followed by subnormal temperatures and rapidly fatal collapse.
It might appear as if this indicated that the occurrence of fever
had a beneficial effect, but this is one of the points at issue.
Certainly such an effect is not due to the bacteria being unable
to multiply at the higher degrees of temperature occurring
in fever, for this has been shown not to be the case. Whether
the increase of bodily temperature indicates the occurrence of
changes resulting in the production of bactericidal bodies, etc.,
is very doubtful ; a production of antagonistic substances may
be effected without the occurrence of fever or of any apparent
disturbance of health. If we consider the site of the heat
production in fever we again are in difficulties. It might appear
as if the tissue destruction, indicated by the occurrence of fatty
degeneration, would lead to heat development, but frequently
excessive heat production with increased proteid metabolism
occurs without any discoverable changes in the tissues ; and
further, in phosphorus poisoning there is little fever with great
tissue destruction. The increased work performed by the heart
in most bacterial infections no doubt contributes to the rise of
bodily temperature. But we must bear in mind that in fever
there is more than mere increase of heat production — there is
also a diminished loss of heat from interference with the nervous
mechanism of the sweat apparatus. The known facts would
indicate that in fever there is a factor involving the nervous
system to be taken into account. The whole subject is thus
very obscure.

Symptoms. — Many of the symptoms occurring in bacterial
infections are produced by the histological changes mentioned,
as can be readily understood ; whilst in the case of others, corre-
sponding changes have not yet been discovered. Of the latter
those associated with fever, with its disturbances of metabolism
and manifold affections of the various systems, are the most
important. The nervous system is especially liable to be
affected — convulsions, spasms, coma, paralysis, etc., being
common. The symptoms due to disturbance or abolition of the
functions of secretory glands also constitute an important group,
forming, as they do, a striking analogy to what is found in the
action of various drugs.

THE TOXINS PRODUCED BY BACTERIA 187

These tissue changes and symptoms are given only as illus-
trative examples, and the list might easily be greatly amplified.
The important fact, however, is that nearly all, if not quite all,
tli> changes found throughout the organs (without the actual
presence of bacteria), and also the symptoms occurring in infec-
tive diseases, can either be experimentally reproduced by the in-
jection of bacterial poisons or have an analogy in the action of
drugs.

THE TOXINS PRODUCED BY BACTERIA.

Early Work on Toxins. — We know that bacteria are capable
of giving rise to poisonous bodies within the animal body and
also in artificial media. We know, however, comparatively little
of the actual nature of such bodies, and therefore we apply to
them as a class the general term toxins. The .necessity for
accounting for the general pathogenic effects of certain bacteria,
which in the corresponding diseases were not distributed through-
out the body, directed attention to the probable existence of
such toxins ; and the first to systematically study the production
of such poisonous bodies was Brieger. This observer isolated
from putrefying substances, and also from bacterial cultures,
nitrogen-containing bodies, which he called ptomaines. Similar
bodies occurring in the ordinary metabolic processes of the
body had previously been described and called leucomaines.
I'toinaiii'.'s isolated from pathogenic bacteria in no case re-
produced the symptoms of the disease, except perhaps in tetanus,
and this only owing to their impurity. The methods by which
they were isolated were faulty, and they have therefore only a
historic interest.

The introduction of the principle of rendering fluid cultures
bacteria-free by filtration through unglazed porcelain, and its
application by Roux and Yersin to obtain, in the case of the
1). diphtherias, a solution containing a toxin which reproduced
the symptoms of this disease (vide Chapter XVI.), -encouraged the
further inquiry as to the nature of this toxin. An attempt on
ihf part of Brieger and Fraenkel'to obtain a purified diphtheria
toxin by precipitating bouillon cultures by alcohol (the product
being denominated a toxalbumin) did not greatly advance
knowledge on the subject, and further investigation soon showed
that specific toxins can be isolated from but few bacteria.

General Facts regarding Bacterial Toxins. — The following
may be regarded as the chief facts regarding bacterial toxins
which have been revealed by the study, partly of the bodily
tissues of animals infected by the bacteria concerned, partly of

188 THE TOXINS PRODUCED BY BACTERIA

artificial cultures of these bacteria. In dealing with these it is
necessary to distinguish between the effects produced by the
actual constituents of the bacterial protoplasm (intracellular
toxins) and those which in a few bacteria are traceable to
soluble substances passing out into the media in which these
bacteria may be growing (extracellular toxins). The former
are concerned in the action of by far the greater number of
pathogenic bacteria; the latter account for the pathogenic
processes originated in a limited number of cases of which
diphtheria and tetanus are the most important. This dis-
tinction is important as, in consequence of these last two
diseases having had much attention directed towards them early
in the history of research on the subject, there has hitherto
been too much tendency to take for granted that poisons of
a similar constitution are concerned in all cases of bacterial
intoxication. At present such an assumption is not justified
by facts, and we do not even know whether the intracellular
and extracellular toxins belong to the same group of chemical
bodies. At present, however, the terms are used as a con-
venient means of accentuating a difference in solubility between
the two groups of toxic bodies.

The dead bodies of certain bacteria have been found to be
very toxic. When, for instance, tubercle bacilli are killed by
heat and injected into the body tissues of a susceptible animal,
tubercular nodules are found to develop round the sites where
they have lodged. From this it is inferred that they must have
contained characteristic toxins, seeing that characteristic lesions
result. The bodies of such organisms as the pyogenic cocci, the
b. typhosus, and the v. cholerse likewise give rise to pathogenic
effects. Such intracellular toxins may appear in the fluids in
which the bacteria are living (1) by excretion in an unaltered
or altered condition, (2) by the disintegration of the bodies
of the organisms which we know are always dying in any
bacterial growth. The death of bacteria occurs also in the
body of an infected animal, and the disintegration of these
dead bacteria constitutes an important means by which the
poisons they contain are absorbed. There is some evidence
that often bacteria originate during growth poisons which are
hurtful to their own vitality, and also that ferments are produced
by them which have a solvent effect on the poisoned members
of the colony. Such a process of autolysis, as it has been called,
may have an important effect in liberating intracellular toxins.
It is impossible, at present, to obtain intracellular toxins apart
from other derivatives of the bacterial protoplasm, and thus

FACTS REGARDING BACTERIAL TOXINS 189

our chief knowledge concerning their effects is derived from the
study of what happens when the bodies of bacteria killed by
chloroform vapour or by heat are injected into animals. When
effects are produced by such injections they do not present in
any particular case specific characters. They are of the nature
• •!' u>'i irr;il disturbances of metabolism, as manifested by fever,
or by depression of temperature, loss of weight, etc., often of
such serious degree as to result in death. It is important to
note that when pathogenic effects are produced these usually
appear very soon, it may be in a few hours after injection of the
toxic material; there is not the definite period of incubation
which with other toxins often elapses before symptoms appear.

In certain cases there is difficulty in understanding the action
of bacteria which neither form soluble toxins in a fluid medium
nor possess a highly toxic protoplasm, and yet with which we
often see effects produced at a distance from the focus of
infection, e.g. b. anthracis. To explain such occurrences it has
long been regarded as a possibility that some bacteria are
only capable of producing toxins within the animal tissues,
and it has further been thought possible that bacteria, such as,
for example, the typhoid bacillus, which do distribute into
media intracellular toxins, might either produce these toxins
more readily in the tissues or might produce in addition other
toxins of a different nature. During recent years such toxins
have been much studied, and the name aygressins has been
given to them. The evidence adduced for the existence of
these aggressins as a separate group of bacterial poisons is of
the following kind : An animal is killed by a dose of the
typhoid, dysentery, cholera, or tubercle bacillus, or by a staphy-
lococcus, the organism being introduced into one of the serous
cavities. After death the serous exudation, which in all these
cases is present, is taken, and centrifugalised to remove the
bacteria so far as this can be done by such a procedure ; the
bacteria which are left are killed by shaking the fluid up with
toluol and leaving it to stand for some days. It is stated that
such a fluid is of itself without pathogenic effect, but has the
property of transforming a non-lethal dose of the bacterium used
into one having fatal effect. Further, the effects of the com-
bined actions of the bacteria and aggressins are often of a much
more acute character than can be obtained with toxic products
developed in vitro. Thus, in the case of the action of a non-
lethal dose of the tubercle bacillus plus its aggressin, it is
>;ii« I that death may occur in twenty hours, a result never
obtained with artificial cultures of the organism. The effects

190 THE TOXINS PEODUCED BY BACTERIA

produced by aggressins are attributed to a paralysing action
on the phagocytic functions of the leucocytes. The subject is
full of difficulties, and in the case of certain of the organisms
employed it is stated that results similar to those attributed to
aggressin action have been observed with macerated cultures, —
the deduction being that in the aggressins we are merely dealing
with concentrated intracellular toxins. On the other hand,
as evidence of the existence of a special group of toxins, it has
been stated that a special type of immunity against the
aggressins can be originated. Perhaps the most important
aspect of the controversy is the recognition of the existence of
toxins having an action on the leucocytes. A poison causing
death of these cells in connection with the pus-forming action of
the pyogenic cocci has been described under the name of'
leucocidin, and Eisenberg records that in in vitro mixtures of
leucocytes and cultures of the bacillus of symptomatic anthrax
loss of inotility and degeneration of the cells may be observed.
The investigation of such poisons must be of the highest
importance in view of the part played by the blood-cells in the
protection of the body against infection, and it is possible that
toxins having a fatal effect in strong concentrations may, when
dilute, be responsible for the phenomena of attraction or repulsion
of leucocytes which we know occur round a focus of bacterial
growth in the body.

Sometimes the media in which bacteria are growing become
extremely toxic. This is more marked in some cases than in
others. The two best examples of bacteria thus producing
soluble toxins are the diphtheria and tetanus bacilli. In these
and similar cases when bouillon cultures are filtered bacterium-
free by means of a porcelain filter, toxic fluids are obtained,
which on injection into animals reproduce the highly character-
istic symptoms of the corresponding diseases. In the case of
the b. anthracis and of many others, at any rate when growing
in artificial media, such toxin production is much less marked,
a filtered bouillon culture being relatively non-toxic. Poisons
appearing in culture media have been called extracellular toxins,
but we cannot as yet say whether they are excreted by the
bacteria or whether they are produced by the bacteria acting on
the constituents of the media. The extracellular toxins are
easily obtainable in large quantities, and it is their nature and
effects which are best known. No method has been discovered
of obtaining them in a pure form, and our knowledge of their
properties is exclusively derived from the study of the toxic
nitrates of bouillon cultures — these filtrates being usually re-

FACTS REGARDING BACTERIAL TOXINS 191

ferrecl to simply as the toxins. These toxins differ in their
ell'i.'ctN from the intracellular poisons in that s^cific actions on
certain ti-> :i--s are often manifested. Thus the toxins of the
diphtheria, the tetanus, and the botulismus bacilli all act on
the ner\o i- system ; with some of the pyogenic bacteria, on the
other haiul, poisons, probably of similar nature, produce solution
of red blood corpuscles (this last might be thought to explain
the aiuemias so common in. the associated diseases, but it is to
be noted that, in cultures at least, these htemolytic toxins are
developed in very small amounts). In the action of many of
these toxins the occurrence of a period of incubation between
the introduction of the poison into the animal tissues and the
appearance of symptoms is often a feature.

The whole question of the parts played by toxins in bacterial
action is manifestly very complex. On the one hand, we have
a few processes, for example, diphtheria and tetanus, in which
UTV characteristic effects are produced on special tissues, these
being accounted for by the formation of soluble toxins which
are capable of being separated from the bacterial growths in vitro.
On the other hand, we have the great mass of bacterial infec-
tions. With regard to these, the distribution of the bacteria
in the bodies of infected animals makes it necessary for us
to take for granted that a toxic action is at work. All that
we know, however, regarding a possible explanation of such
toxicity is that the bodies of the bacteria or substances directly
derived from them are capable of producing pathogenic effects.
These effects are of a non-specific character in the sense that
they are not the result of an action on any particular tissue in
the body, but on the vital processes of the organism as a whole.
We are at present entirely ignorant of the interpretation to be
put, for instance, on the lowering of bodily temperature on the
one hand and of the occurrence of fever on the other, both of
which may be produced by the injection of the so-called intra-
cellular toxins in varying doses, and we are ignorant of the
relations which either event may have to the bringing into play
of the defensive mechanisms of the body. At the same time we
must admit the possibility that with any one species of organism
different effects may be produced by, it may be, different elements
in the protoplasm of the invading bacterial cell. Some of these
elements may act on certain groups of specialised cells of the
body, such as those of the nervous system, liver, or kidneys,
giving rise to what we are forced to describe in general terms as
disturbances of metabolism. Other poisonous elements may
mainly act on the defensive cells of the body, of which the

192 THE TOXINS PRODUCED BY BACTERIA

leucocytes may be taken as the type. Here a small dose of
toxin may stimulate these cells to an activity which results in
the infection being thrown off, either by the poison being neutra-
lised, or by the supply of toxin being cut off by the killing of
the bacterium producing it. A large dose of such a toxin, may,
on the other hand, altogether break down the defensive mechanism
of the invaded body. A possible complexity in toxic action
may occur even in such an apparently simple case as diphtheria.
As will be seen later, the special neuro-toxin excreted by the
diphtheria bacillus can be neutralised by an antitoxic substance,
but the action of this does not necessarily cause the death of
the bacteria in the throat whose capacity for multiplication may
be dependent on a vital activity of the protoplasm distinct
from neurotoxin production, and therefore requiring another
mechanism for its neutralisation. The complexity of the toxic
process is also illustrated by the facts known regarding the
cholera vibrio. In man, this organism is confined in its habitat
to the intestinal tract, and its serious effects are attributed to
the absorption of toxins therefrom. On the other hand, in
animals, not susceptible to such intestinal infection, death can
be readily produced by the injection intraperitoneally of a com-
paratively small amount of dead cholera vibrios, and it will be
seen in the chapter on Cholera that the possibility has to be
faced of the toxins acting in the two conditions being different.
Thus it is possible that the toxic element in an organism which
enables it to effect its initial multiplication in or on the tissues
is not necessarily bound up with the toxicity which is respons-
ible for the manifestation of specific disease effects. This is
borne out by the work of Grassberger and Schattenfroh on the
bacillus of symptomatic anthrax. In this case an organism,
which in vitro has lost to a large extent its capacity of producing
soluble toxins, may show great capacity for multiplying when
introduced into a susceptible animal.

There is another point which must be kept in view, namely,
that some of the phenomena which have been regarded as
dependent upon the activity of bacterial toxins may possibly
be related to the little-understood process of anaphylaxis (see
Immunity). Anaphylaxis essentially consists in the develop-
ment under certain circumstances in an animal of a hypersensi-
tiveness to foreign albuminous materials which in themselves
are not essentially toxic. Effects of the gravest kind may be
produced during this period of hypersensitiveness, and it has
been thought that some of the phenomena of an infectious
disease, e.g. the occurrence of an incubation period, may be

THE NATURE OF TOXINS 193

accounted for by the development of hypersensitiveuess to the
protoplasm of the invading bacteria. It may be said here that
the effect seen when horse serum is injected into a rabbit
i luring its hypersensitive stage to this substance bears a striking
resemblance to what is seen in natural infection in man by the
cholera vibrio.

The phenomena of any bacterial disease may thus in reality
be due to very different and complex causes.

The Nature of Toxins. — There is still comparatively little
known regarding this subject, and it chiefly relates to the extra-
cellular toxins. The earlier investigations upon toxins suggested
that analogies exist between the modes of bacterial action and
what takes place in ordinary gastric digestion, and the idea was
worked out for anthrax, diphtheria, tetanus, and ulcerative
endocarditis by Sidney Martin. This observer found that
albumoses l and peptones were formed by the action of the
pathogenic bacteria studied, and further, that the precipitate
containing these albumoses was toxic. A similar digestive
action has been traced in the case of the tubercle bacillus by
Kiihne.

Further evidence that bacterial toxins are either albumoses
or bodies having a still smaller molecule is adduced by C. J.
Martin. This worker, by filling the pores of a Chamberland
bougie with gelatin, has obtained what is practically a strongly
supported colloid membrane through which dialysis can be made
to take place under great pressure, say, of compressed oxygen.
He finds that in such an apparatus toxins — at least two kinds
tried — will pass through just as an albumose will.

Brieger and Boer, working with bouillon cultures of diphtheria
and tetanus, separated, by precipitation with zinc chloride,

1 In the digestion of albumins by the gastric and pancreatic juices, the
albumoses are a group of bodies formed preliminarily to the production of
peptone. Like the latter they differ from the albumins in their not being
coagulated by heat, and in being slightly dialysable. They differ from
the peptones in being precipitated by dilute acetic acid in presence of
much sodium chloride, and also by neutral saturated sulphate of ammonia.
Both are precipitated by alcohol. The first albumoses formed in digestion
are proto-alburnose and hetero-albumose, which ditfer in the insolubility
of the latter in hot and cold water (insolubility and coagulability are
ipiite different properties). They have been called the primary albumoses.
Hv further digestion both pass into the secondary albumose, deutero-
.illiumose, which differs slightly in chemical reactions from the parent
bodies, e.g. it cannot be precipitated from watery -solutions by saturated
sodium chloride unless a trace of acetic acid be present. Dysalbumose is
probably merely a temporary modification of hetero-albumose. Further
digestion of deutero-albumose results in the formation of peptone.

194 THE TOXINS PRODUCED BY BACTERIA

bodies which show characteristic toxic properties, but which had
the reactions neither of peptone, albumose, nor albuminate, and
the nature of which is unknown. It has also been found that
the bacteria of tubercle, tetanus, diphtheria, and ( cholera can
produce toxins when growing in proteid-free fluids In the case
of diphtheria, when the toxin is produced in such a fluid a proteid
reaction appears. Of course this need not necessarily be caused
by the toxin. Further investigation is here required, for
Uschinsky, applying Brieger and Boer's method to a toxin so
produced, states that the toxic body is not precipitated by zinc
salts, but remains free in the medium. If the toxins are really
non-proteid they may, on the one hand, be the final product of
a digestive action, or they may be the manifestation of a separate
vital activity on the part of the bacteria. On the latter theory
the toxicity of the toxic albumoses of Sidney Martin may be due to
the precipitation of the true toxins along with these other bodies.
From the chemical standpoint this is quite possible. When we
take into account the extraordinary potency of these poisons (in
the case of tetanus the fatal dose of the pure poison for a
guinea-pig must often be less than '00000 1 grm ), we can under-
stand how attempts by present chemical methods to isolate them
in a pure condition are not likely to be successful, and of their
real nature we know nothing. Friedberger and Moreschi have
shown that the intravenous injection in the human subject of
a fraction of a loopful of a dead typhoid culture gives rise
to toxic symptoms, including marked febrile reaction. Such
injections are followed by the appearance of agglutinating and
bacteriolytic substances in the serum. These results show that
intracellular toxins may be comparable with extracellular toxins
so far as concerns the extremely small dose sufficient to produce
toxic effects.

Amongst the properties of the extracellular toxins are
the following : They are apparently all uncrystallisable ; they
are soluble in water and they are dialysable ; they are pre-
cipitated along with proteids by concentrated alcohol, and also
by ammonium sulphate; if they are proteids they are either
albumoses or allied to the albumoses ; they are often relatively un-
stable, having their toxicity diminished or destroyed by heat (the
degree of heat which is destructive varies much in different cases),
light, and by certain chemical agents. Their potency is often
altered in the precipitations practised to obtain them in a pure
or concentrated condition, but among the precipitants ammonium
sulphate has little if any harmful effect. Regarding the toxins
which are more intimately associated with the bacterial proto-

THE NATURE OF TOXINS 195

plasm we know much less, but it is probable that, chemically,
their nature is similar, though some of them at least are not so
ra-ily injured by heat, e.g. those of the tubercle bacillus, already
mentioned. In the case of all toxins the fatal dose for an
animal varies with the species, body weight, age, and previous
conditions as to food, temperature, etc. In estimating the
minimal lethal dose of a toxin these factors must be carefully
considered.

The following is the best method of obtaining concentrated extra-
cellular toxins : The toxic fluid is placed in a shallow dish, and ammonium
sulphate crystals are well stirred in till no more dissolve. Fresh crystals
to form a bulk nearly equal to that of the whole fluid are added, and the
dish set in an incubator at 37° C. overnight. Next day a brown scum
of precipitate will be found floating on the surface. This contains the
toxin. It is skimmed off with a spoon, placed in watch-glasses ; these
are dried in vocno and stored in the dark, also in, vacuo, or in an exsiccator
< nntaining strong sulphuric acid. For use the contents of one are
dissolved up in a little normal saline solution.

The comparison of the action of bacteria in the tissues in
the production of these toxins to what takes place in the gastric
digestion,, has raised the question of the possibility of the elabora-
tion by these bacteria of ferments by which the process may
be started. Thus Sidney Martin puts forward the view that
ferments may be produced which we may look on as the
primary toxic agents, and which act by digesting surrounding
material and producing albumoses — these poisons being, as it
were, secondary poisons. Hitherto all attempts at the isolation
of bacterial ferments of such a nature have failed.

But apart from the fact that with such bacteria as those of
tetanus and diphtheria, a digestive action may occur, analogies have
In-en drawn between ferment and toxic action. The chief facts
upon which such analogies have been founded are as follows :
Thus the toxic products of these and other bacteria lose their
tuxicity by exposure to a temperature which puts an end to the
activity of such an undoubted ferment as that of the gastric
juice. If a bouillon containing diphtheria toxin be heated at
G5° C. for one hour, it is found to have lost much of its toxic
effect, and in the case of b. tetani all the toxicity is lost by
exposure at this temperature. In both diseases there is a still
further fact which is adduced in favour of the toxic substances
1 icing of the nature of ferments, namely, the existence of a
driinitr period of incubation between the injection of the toxic
bodies and the appearance of symptoms. This may be inter-
preted as showing that after the introduction of, say, affiltered

196 THE TOXINS PRODUCED BY BACTERIA

bouillon culture, further chemical substances are formed in the
body before the actual toxic effect is produced. Too much
reliance must not be placed on such an argument, for in the
case of tetanus, at least, the delay may be explained by the fact
that the poison apparently has to travel up the nerve trunks
before the real poisonous action is developed. Further, with
some poisons presently to be mentioned which are closely allied
to the bacterial toxins, an ' incubation period may not exist.
It would not be prudent to dogmatise as to whether the toxins
do or do not belong to such an ill-defined group of substances
as the ferments. It may be pointed out, however, that the
essential concept of a ferment is that of a body which can
originate change without itself being changed, and no evidence
has been adduced that toxins fulfil this condition. Another
property of ferments is that so long as the products of fermenta-
tion are removed, the action of a given amount of ferment is
indefinite. Again, in the case of toxins no evidence of such an
occurrence has been found. A certain amount of a toxin is
always associated with a given amount of disease effect, though
a process of elimination of waste products must be all the time
going on in the animal's body. Again, too much importance
must not be attached to loss of toxicity by toxins at relatively
low temperatures. This is not true of all toxins, and further-
more many proteids show a tendency to change at such
temperatures ; for instance, if egg albumin be kept long
enough at 55° C. nearly the whole of it will be coagulated.
We must therefore maintain an open mind on this subject.

Similar Vegetable and Animal Poisons. — It has been found that
the bacterial poisons belong to a group of toxic bodies all present-
ing very similar properties, other members of which occur widely
in the vegetable and animal kingdoms. Among plants the best-
known examples are the ricin and abrin poisons obtained by making
watery emulsions of the seeds of the Ricinus communis and the Abrus
precatorius (jequirity) respectively. From the JRobinia pseudacacia
another poison — robin — belonging to the same group is obtained. The
chemical reactions of ricin and abrin correspond to those of the bacterial
toxins. They are soluble in water, they are precipitable by alcohol, but
being less easily dialysable than the albumoses they have been called
toxalbumins. Their toxicity is seriously impaired by boiling, and they
also gradually become less toxic on being kept. Both are among the
most active poisons known — ricin being the more powerful. When they
are injected subcutaneously a period of twenty-four hours usually elapses
— whatever be the dose — before symptoms set in. Both tend to produce
great inflammation at the seat of inoculation, which in the case of ricin
may end in an acute necrosis ; in fatal cases hsemorrhagic enteritis and
nephritis may be found. Both act as irritants to mucous membranes,
abrin especially being capable of setting up most acute conjunctivitis.

VEGETABLE AND ANIMAL TOXINS 197

In the actiou of a poisonous fungus, Amanita phalloides, a similar
toxin is at work. After an incubation period of some hours, symptoms of
abdominal pain, diarrhoea with bloody stools, and later jaundice occur. In
vitro the toxin has a haemolytic action. Like other poisons of this class,
an antitoxin can be produced towards the fungus poison.

It is also certain that the poisons of scorpions and of poisonous snakes
belong to the same group. The poisons derived from the latter are
usually called venins, ana a very representative group of such venins
derived from different species has been studied. To speak generally,
there is derivable from the natural secretions of the poison glands a
series of venins which have all the reactions of the bodies previously
considered. Like ricin and abrin, they are not so easily dialysable as
bacterial toxins, and therefore have also been classed as toxalbumins.
Their properties are also similar ; many of them are destroyed by heat,
but the degree necessary here also varies much, and some will stand
boiling. There is also evidence that in a crude venin there may be several
poisons differently sensitive to heat. All the venius are very powerful
poisons, but here there is practically no period of incubation — the effects
are almost immediate. An outstanding feature of the venins is the
complexity of the crude poison secreted by any particular species of
snake. C. J. Martin, in summing up the results of many observers, has
pointed out that different venoms have been found to contain one or
more of the following poisons : a neurotoxiu acting on the respiratory
centre, a neurotoxin acting on the nerve-endings in muscle, a toxin
causing haemolysis, toxins acting on other cells, e.g. the endothelium of
blood vessels (this from its effects has been named ha-morrhagin),
leucocytes, nerve-cells, a toxin causing thrombosis, a toxin having an
opposite effect and preventing coagulation, a toxin neutralising the
I'.K tericidal qualities of the body fluids and thus favouring putrefaction,
a toxin causing agglutination of the red blood corpuscles, a proteolytic
ferment, a toxin causing systolic standstill of the excited heart. Any
particular venom contains a mixture in varying proportions of such
toxins, and the different effects produced by the bites of different snakes
largely depend on this variability of composition. The neurotoxic, the
thrombotic, and the haemolytic toxins are very important constituents
of any venom. The toxicity of different venoms varies much, and no
general statement can be made with regard to the toxicity of different
poisons towards man. Lamb has calculated that the fatal dose of crude
cobra venom for man is probably about '015 of a gramme, and that
if such a snake bites with full glands many times this dose would
probably be injected, but, of course, the amount emitted depends largely
on the period which has elapsed since the animal last emptied its glands.
When a dose of a venom not sufficient to cause immediate death from
general effects be given, very rapid and widespread necrosis often may
occur in a few hours round the site of inoculation.

An extremely important fact was discovered by Flexner and Noguchi,
namely, that the haemolytic toxin of cobra venom in certain cases has no
action by itself, but produces rapid solution of red corpuscles when some
normal serum is added, the latter containing a labile complement-like
body, which activates the venom. In this there is a close analogy to
what holds in the case of a haemolytic serum deprived of complement by
heat at 55° C. (p. 130). Kyes and Sachs further showed that in addition
to serum-complement a substance with definitely known constitution,
namely lecithin, had the property of activating the haemolytic substance
in cobra venom, the two apparently uniting to form an actively toxic

198 THE TOXINS PRODUCED BY BACTERIA

substance. So far no example of the activation of a bacterial toxin is
known, but the results mentioned point to the possibility of this occurring
in some cases in the tissues of the body.

There is another group of toxic manifestations which present some
analogies to those of the bacterial toxins, but concerning which very little
is known. The best example of these is found in the toxic properties of
the serum of the eel. If a small quantity of such serum, say '25 of a c.c.,
be injected into a rabbit subcutaneously, death occurs in a few minutes.
Although nothing is known of the substances giving rise to such effects,
the phenomenon is to be considered in relation, on the one hand, to
the action of bacterial toxins, and on the other to the phenomenon of
anaphylaxis. (See Chapter on Immunity.)

The Theory of Toxic Action. — While we know little of the
chemical nature of any toxins, we may, from our knowledge of
their properties, group together the tetanus and diphtheria
poisons, ricin, abrin, snake poisons, and scorpion poisons.
Besides the points of agreement already noted, all possess the
further property that, as will 'be afterwards described, when
introduced into the bodies of susceptible animals they stimu-
late the production of substances called antitoxins. The
nature of the antagonism between toxin and antitoxin will
be discussed later. Here, to explain what follows, it may be
stated (1) that the molecule of toxin forms directly a combina-
tion with the molecule of antitoxin, and (2) that it has been
shown that toxin molecules may lose much of their toxic power
and still be capable of uniting with exactly the same proportion
of antitoxin molecules. From these and other circumstances
Ehrlich has advanced the view that the toxin molecule has .a
very complicated structure, and contains two atom groups. One
of these, the haptophorous (aTrreu/, to bind to), is that by
which combination takes place with the antitoxin molecule, and
also with presumably corresponding molecules naturally existing
in the tissues. . The other atom group he calls the toxophorous,
and it is to this that the toxic effects are due. This atom group
is bound to the cell elements, e.y. the nerve cells in tetanus, by
the haptophorous group. Ehrlich explains the loss of toxicity
which with time occurs in, say, diphtheria toxin, on the theory
that the toxophorous group undergoes disintegration. And if we
suppose that the haptophorous group remains unaffected we can
then understand how a toxin may have its toxicity diminished
and still require the same proportion of antitoxin molecules for
its neutralisation. To the bodies whose toxophorous atom
groups have become degenerated, Ehrlich gives the name toxoids.
The theory may afford an explanation of what has been sus-
pected, namely, that in some instances toxins derived from

THE THEORY OF TOXIC ACTION 199

different sources may be related to one another. For example,
Ehrlich has pointed out that ricin produces in a susceptible
animal body an antitoxin which corresponds almost completely
with that produced by another vegetable poison, robin (vide
supra), though ricin and robin are certainly different. This may
be explained on the supposition that robin is a toxoid of
ricin, i.e. their haptophorous groups correspond, while their
toxophorous differ. The evidence on which Ehrlich's deductions
are based is of a very weighty character, but another view of
toxic action is that the relation between a toxin and the cell
on which it acts is an example of the physical phenomenon of
adsorption. The whole subject will be again referred to in the
chapter on Immunity.

With regard to the intracellular toxins we shall see it is
difficult to determine whether or not they share with the extra-
cellular poisons the property of stimulating antitoxin formation,
—if they do not, then they may belong to an entirely different
class of substances. It is certain that a tolerance against such
poisons is difficult to establish and is not of a lasting character.
We thus cannot say what the mechanism is by which these
poisons act. It may be said that Macfadyen, by grinding up
typhoid bacilli frozen by liquid air, claimed that on thawing he
obtained the intracellular toxins in liquid form, and he further
stated that by using this fluid he could immunise animals not
only against the toxins but also against the living bacteria.

\\V have already pointed out that those who claim for the
aggressins a special character hold that the activity of these
bodies has as its effect the interference with the phagocytic
functions of the leucocytes. They also hold that a special type
of immunity can be developed against the aggressins.

CHAPTER VII.

INFLAMMATORY AND SUPPURATIVE CONDITIONS.

THIS subject is an exceedingly wide one, and embraces a great
many pathological conditions which in their general characters
and results are widely different. Thus, in addition to suppura-
tion, various inflammations, ulcerative endocarditis, septicaemia
and pyaemia, will come up for consideration. With regard to
these, the two following general statements, established by
bacteriological research, may be made in introducing the subject.
In the first place, there is no one specific organism for any one
of these conditions; various organisms may produce them,
and not infrequently more than one organism may be
present. In the second place, the same organism may produce
widely varying results under different circumstances, — at one
time a local inflammation or abscess, at another multiple sup-
purations or a general septicaemia. The principles on which
this diversity in results depends have already been explained
(p. 177). Furthermore, there are conditions like acute pneu-
monia, epidemic meningitis, acute rheumatism, etc., which have
practically the character of specific diseases, and yet which, as
regards their essential pathology, belong to the same class.
The arrangement followed is to a certain extent one of
convenience.

It may be wrell to emphasise some of the chief points in the
pathology of these conditions. In suppuration the two main
phenomena are — (a) a progressive immigration of leucocytes,
chiefly of the polymorpho-nuclear (neutrophile) variety, and
(b) a liquefaction or digestion of the supporting elements of the
tissue along with necrosis of the cells of the part. The result
is that the tissue affected becomes replaced by the cream-like
fluid called pus. A suppurative inflammation is thus to be
distinguished on the one hand from an inflammation without
destruction of tissue, and on the other from necrosis or death
en masse, where the tissue is not liquefied, and leucocyte

200

NATURE OF SUPPURATION 201

accumulation may be slight. When, however, suppuration is
taking place in a very dense fibrous tissue, liquefaction may be
incomplete, and a portion of dead tissue or slough may remain
in the centre, as is the case in boils. In the case of suppuration
in a serous cavity the two chief factors are the progressive
leucocytic accumulation and the disappearance of any fibrin
which may be present.

Many experiments have been performed to determine whether
suppuration can be produced in the absence of micro-organisms
by various chemical substances, such as croton oil, nitrate of
silver, turpentine, etc. — care, of course, being taken to ensure
the absence of bacteria. The general result obtained by inde-
I>endent observers is that as a rule suppuration does not follow,
but that in certain animals and with certain substances it may,
the pus being free from bacteria. Buchner showed that sup-
puration may be produced by the injection of dead bacteria, e.g.
sterilised cultures of bacillus pyocyaneus, etc. The subject has
now more a scientific than a practical interest, and the general
statement may be made that practically all cases of true sup-
puration met with clinically are due to the action of living
micro-organisms.

The term septicaemia, is applied to conditions in which the
organisms multiply within the blood and give rise to symptoms
of general poisoning, without, however, producing abscesses in
the organs. The organisms are usually more numerous in the
capillaries of internal organs than in the peripheral circulation,
but the application of the newer methods of cultivation has
shown that they can be detected in the peripheral blood much
more frequently than was formerly supposed to be the case.
The essential fact in pycemia, on the other hand, is the occur-
rence of multiple abscesses in internal organs and other parts of
the body. In most of the cases of typical pyaemia, common in
pre-antiseptic days, the starting-point of the disease was a septic
wound with bacterial invasion of a vein leading to thrombosis
and secondary embolism. Multiple foci of suppuration may be
produced, however, in other ways, as will be described below
(p. 213). If the term "pyaemia" be used to embrace all such
conditions, their method of production should always be dis-
tinguished.

BACTERIA AS CAUSES OF INFLAMMATION AND SUPPURATION.

A considerable number of species of bacteria have been found
in acute inflammatory and suppurative conditions, and of these

202 INFLAMMATION AND SUPPURATION

many have been proved to be causally related, whilst of some
others the exact action has not yet been fully determined.

Ogston, who was one of the first to study this question (in
1881), found that the organisms most frequently present were
micrococci, of which some were arranged irregularly in clusters
(staphylococci), whilst others formed chains (streptococci). He
found that the former were more common in circumscribed
acute abscesses, the latter in spreading suppurative conditions.
Rosenbach shortly afterwards (1884), by means of cultures,
differentiated several varieties of micrococci, to which he gave
the following special names : staphylococcus pyogenes aureus,
staphylococcus pyogenes albus, streptococcus pyogenes, micrococcus
pyogenes tennis. Other organisms are met with in suppuration,
such as staphylococcus pyogenes citreus, staphylococcus cereus
albus, staphylococcus cereus Jlavus, pneumococcus, pneumobacillus,
(Friedlander), bacillus pyogenes foetidus (Passet), bacillus coli
communis, bacillus lactis aerogenes, bacillus aerogenes encapsul-
atus, bacillus pyocyaneus, micrococcus tetragenus, pneumococcus,
pneumobacillus, diplococcus intracellularis meningitidis, and
others. Various anaerobic bacteria are also concerned in the
production of inflammation, which is often associated with
redema, haemorrhage, or necrosis (vide Chap. XVII.).

In secondary inflammations and suppurations following acute
diseases, the corresponding organisms have been found in some
cases, such as gonococcus, typhoid bacillus, influenza bacillus,
etc. Suppuration is also produced by the actinomyces and the
glanders bacillus, and sometimes chronic tubercular lesions have
a suppurative character.

Staphylococcus Pyogenes Aureus. — Microscopical Characters.
— This organism is a spherical coccus about '9 //, in diameter,
which grows irregularly in clusters or masses (Fig. 50). It
stains readily with all the basic aniline dyes, and retains the
colour in Gram's method (Plate I., Fig. 1).

Cultivation. — It grows readily in all the ordinary media at
the room temperature, though much more rapidly at the
temperature of the body. In stab cultures in peptone gelatin
a streak of growth is visible on the day after inoculation, and
on the second or third day liquefaction commences at the top.
As liquefaction proceeds, the growth falls to the bottom as a
flocculent deposit, which soon assumes a bright yellow colour,
while a yellowish film may form on the surface, the fluid portion
still remaining turbid. Ultimately liquefaction extends out to
the wall of the tube (Fig. 51). In gelatin plates colonies may
be seen with the low power of the microscope in twenty-four

STAPHYLOCOCCnS PYOGENES AUREUS 203

hours, as little balls somewhat granular on the surface and of
brownish colour. On the second day they are visible to the
naked eye as whitish yellow points, which afterwards become

Ki(i. r>0.- Staphylococeoa pyogenes auveus,
young culture on agar, showing clumps
of cocci.

Stained with weak carbol-fuchsin. x 1000.

more distinctly yellow. Liquefac-
tion occurs around these, and little
• -ii] is are formed, at the bottom
<>t which the colonies form little
yellowish masses. On ayar, a
stroke culture forms a line of
abundant yellowish growth, with
smooth, shining surface, well
formed after twenty-four hours at
37° C. Later it becomes bright
orange in colour, and resembles
;i -tivjik of oil paint. Single
colonies on the surface of agar are circular discs of similar
appearance, which may reach 2 mm. or more in diameter.
On potatoes it grows well at ordinary temperature, forming a
somewhat abundant layer of orange colour. In bouillon it
produces a uniform turbidity, which afterwards settles to the
bottom as an abundant layer and assumes a brownish yellow
tint. In tin- various media it renders the reaction acid, and it
n tabulates milk, in which it readily grows. The cultures have
a siiiiii-wliat sour odour. It lia< <-<>nsi<lrral>lr tenacity of life

FKI. :")!.— Two stab cultures
of staphylococcus pyogenes
aureus in gelatin, (a) 10 days
old, (6) 3 weeks old, showing
liquefaction of the medium
and characters of growth.
Natural size.

204 INFLAMMATION AND SUPPURATION

outside the body, cultures in gelatin often being alive after
having been kept for several months.

The staphylococcus pyogenes albus is similar in character,
with the exception that its growth on all the media is white.
The colour of the staphylococcus aureus may become less dis-
tinctly yellow after being kept for some time in culture, but it
never assumes the white colour of the staphylococcus albus, and
it has not been found possible to transform the one organism
into the other. A micrococcus called by Welch staphylococcus
epidermidis albus is practically always present in the skin
epithelium ; it is distinguished by its relatively non-pathogenic
properties and by liquefying gelatin somewhat slowly. It is
probably an attenuated variety of the staphylococcus albus.

The staphylococcus pyogenes citreus, which is less frequently
met with, differs in the colour of the cultures, being a lemon
yellow, and is less virulent than the other two.

The staf)hylococcus cereus albus and staphylococcus cere us
flavus are of much less importance. They produce a wax-like
growth on gelatin without liquefaction ; hence their name.

Streptococcus pyogenes. — This organism (Plate I., Fig. 1) is a

coccus of slightly larger
size than the staphylo-
coccus aureus, about 1 /m
in diameter, and forms
chains which may contain
a large number of mem-
bers, especially when it is
growing in fluids (Fig.
52). The chains vary
somewhat in length in
different specimens, and
on this ground varieties
have been distinguished,
e.g. the streptococcus
brevis and streptococcus
longus (vide infra}. As
FIG. 52. -Streptococcus pyogenes, young cul- division may take place
ture on agar, showing chains of cocci. in many of the COCci in

Stained with weak carbol-fuchsiii. x 1000. a cnajn afc the same

time, the appearance of

a chain of diplococci is often met with. In young cultures the
cocci are fairly uniform in size, but after a time they present
considerable variations, many swelling up to twice their normal
diameter. These are to be regarded as involution forms. In its

STREPTOCOCCUS PYOGENES

205

staining reactions the streptococcus resembles the staphylococci
described, being readily coloured by Gram's method.

t 'ni 'ft' r,' if ion. — In cultures outside the body the streptococcus
pyogenes grows much more slowly than the staphylococci, and
also «lies out more readily, being in every respect a more delicate
organism.

In peptone gelatin a stab culture shows, about the second day,
a thin line, which in its subsequent growth is seen to be formed of
a row of minute rounded colonies of whitish colour, which may be
separate at the lower part of the
puncture. They do not usually ex-
ceed the size of a small pin's head,
this size being reached about the fifth
or sixth day. The growth does not
spread on the surface, and no lique-
faction of the medium occurs. The
colonies in gelatin plates have a cor-
responding appearance, being minute
>l>lierical points of whitish colour.
A somewhat warm temperature is
necessary for growth ; even at 20° C.
-••me varieties do not grow. On the
"//"/• media, growth takes place along
the stroke as a collection of small
circular discs of semi - translucent
appearance, which show a great
tendency to remain separate (Fig.
53). The separate colonies remain
sin; ill. rarely exceeding 1 mm. in
diameter. Cultures on agar kept at
the body temperature may often be
found to be dead after ten days. On
potato, as a rule, no visible growth

takes place. In milk it produces a strongly acid reaction but no
••I'M ting of the medium. It ferments lactose, saccharose, and
salicin ( Andre wes and Horder) ; it produces no fermentation of
in ul in in Hiss's serum- water-medium, in this respect differing
from the pneumococcus. It has a strong haemolytic action, as
can be demonstrated by growing it in blood-agar plates (p. 43).
In lioiiillon, growth forms numerous minute granules which after-
wards fall to the bottom, the deposit, which is usually not very
abundant, having a sandy appearance. The api>earance in
broth, however, presents variations which have been used as an
aid to distinguish different species of streptococci. It has been

FIG. 53.— Culture of the
streptococcus pyogenes on
an agar plate, showing
numerous colonies — three
successive strokes. Twenty-
four hours' growth. Natu-
ral size.

206 INFLAMMATION AND SUPPURATION

found that those which form the longest chains grow most
distinctly in the form of spherical granules, those forming short
chains giving rise to a finer deposit. To a variety which
forms distinct spherules of minute size the term streptococcus
conglomerate has been given.

Varieties of Streptococci. — Formerly the streptococcus pyogenes
and the streptococcus erysipelatis were regarded as two distinct
species, and various points of difference between them were
given. Further study, and especially the results obtained by
modifying the virulence (p. 210), have shown that these dis-
tinctions cannot be maintained, and now practically all authorities
are agreed that the two organisms are one and the same,
erysipelas being produced when the streptococcus pyogenes of a
certain standard of virulence gains entrance to the lymphatics of
the skin. Petruschky, moreover, showed conclusively by inocu-
lation that a streptococcus cultivated from pus could cause
erysipelas in the human subject.

Streptococci have also been classified according to the length
of the chains. Thus there have been distinguished (a) strepto-
coccus longus, which occurs in long chains and is pathogenic to
rabbits and mice; (b) streptococcus brevis, which is common in
the mouth in normal conditions, and is usually non-pathogenic ;
and (c) streptococcus conglomeratus, so called from its forming in
bouillon minute granules composed of very long chains. It may
be stated that pathogenic streptococci obtained from the human
subject usually form fairly long chains on agar, whilst the short
streptococci obtained from the mouth and intestine are usually
devoid of virulence. But to these statements exceptions occur,
as short streptococci may be associated with grave lesions ; it
has also been found that the length of the chains is not a
constant feature.

As in the case of other organisms attempts have also been made to
differentiate streptococci by means of tlieir fermentative properties.
Mervyn Gordon introduced for this purpose nine tests, namely: (1) The
clotting of milk, (2) the reduction of neutral red, (3-9) the fermentation
with acid production of saccharose, lactose, raffinose, inulin, salicin,
coniferin, and mannite. Andrewes and Horder by means of these have
differentiated six varieties, of which five occur in the -human subject.
These are : (a) A short-chained form called streptococcus mitis, which
occurs chiefly in the saliva and faeces as a saprophyte. It ferments
saccharose and lactose, and sometimes the glucosides ; it produces an acid
reaction in milk but no clotting, and often reduces neutral-red. (6) The
streptococcus pyogenes, which is the most important pathogenic variety, and
has the characters described above, (c) The streptococcus salivarius, which
corresponds to the streptococcus brevis of the mouth, and which, as
regards fermentative action, seems to bear the same relation to the next

VARIETIES OF STREPTOCOCCI 207

variety as the streptococcus mitis does to the streptococcus pyogenes. It
ferments saccharose, lactose, and raffiuose, sometimes the glucosides and
ruivly inulin ; it cl<>cs milk and reduces neutral-red. (d> Tlie strepto-
coccus iiiKjiiuHtHS, wnicli corresponds with the so-called streptococcus
searlatime and the streptococcus conglomeratus. It ferments saccharose
and lactose, and sometimes ratfinose, reduces neutral-red, and is actively
hiumolytic. It us .ally clots milk and does not grow on gelatin at 20° C.
(e) The xti-i'iitiicuccns fcecalis, a. short-chained form, which abounds in the
inte-tine and which has great fermentative activity, and reacts positively
t-> all Gordon's tests with the exception of raffinose and inulin. It forms
sulphuretted i-ydropen, and is devoid «»f haemolytic action. (/) The
sixth variety is the streptococcus equiims. which is common in the air and
dust of t"wns, and appears to be derived from horse dung.1 It fernu-nts
saccharose and the two glucosides. and forms little or no ncid in milk. It
is, ho\vev.-r, to be noted that t> all these varieties variants are met with.
Schottmiiller has employed the appearance of the colonies of strepto-
cocci on blood agir as a means of separating varieties, the medium used
cou.-isting of i wo parts human blood and live parts melted agar. He
distinguishes the streptoroccu* lonyus or erysipclatis, which forms grey
colonies and has a marked luemolytic action ; a streptococcus mitior or
viridan-s, a short-chained organism, which produces small green colonies
and very little haemolysis ; and a streptococcus muco us encapsulates,
which, as its name indicates, shows wall-marked capsules and produces
colonies which have a slimy consistence. Mandelbaum adds to these the
streptococcus saprojthyticus, which is without haemoiytic action. It should
be noted that on blood agar the pneumococcus forms green colonies and
1 .K.I luces little «»r no haemolysis. Levy finds that a 2 '5 per cent, solu-
ti >n of taurocholate of sodium in bouillon produces complete bacterio-
lysis of the pneumococcus and the streptococcus mucosus, while it has
no effect on other varieties of streptococcus. He considers the strepto-
coccus mucosus to lie a variety of pneumococcus. The general statement
may be made that most of the streptococci from lesions in the human
subject have hsemolytic action, but that occasionally streptococci without
this property are found even in severe infections.

It will be thus seen from this account that the streptococcus
pyogenes as described above is the organism most frequently
associated with the pathogenic processes, and that short-chained
forms are common saprophytes in the. human body, although
they may be associated with conditions of disease ; these may
be subdivided according to their fermentative activity as
detailed. And lastly, there is the streptococcus conglomeratus
(anginosus), which is specially abundant in the throat in scarlet
fever, though it also occurs in other acute catarrhal states. No
definite statement can yet be made as to the etiological relation
of streptococci to scarlet fever ; we can only say that streptococci
are almost invariably present in the fauces, and that to them
many of the complications of the disease are due.

1 For further details, reference must be made to the original papers, Lancet,
September 1906, ii. 708, etc.

208 INFLAMMATION AND SUPPURATION

Bacillus coli communis. — The microscopic and cultural characters are
described in the chapter on typhoid fever. The bacillus lactis aerogenes
and the bacillus pyogenes fo&tidus closely resemble it ; they are either
varieties or closely related species. The former is distinguished by
producing more abundant gas formation, and by its growth on gelatin,
etc., being thicker and whiter than that of the bacillus coli.

Bacillus aerogenes encapsulatus sometimes invades the tissues before
death, and is characterised by the formation of bubbles of gas in the
infected parts. Its characters are described in Chapter XVII.

Bacillus pyocyaneus. — This organism occurs in the form of minute
rods 1*5 to 3 ^ in length and less than '5 ft in thickness (Fig. 54).

Occasionally two or three

_ ^ are found attached end to

i » ss- ^t , end. They are actively

v «A ^ , . N/ motile, and do not form

•* V% V ', * spores. They stain readily

£ * ^ . |>V*»'%* •' with the ordinary basic

^ A A1 fc / *£ stains, but are decolorised

" t & ** \ *\ •'•*•' bJ Gram's method.

' *• * . ,1$* t .* ,* t ' • Cultivation. — It grows

readily on all the ordinary
media at the room tem-
perature, the cultures being
distinguished by the for-
„» mation of a greenish pig-
•'<V - ment. In puncture cul-
* """^ **.& ^- tures in peptone-gelatin a

; '• greyish line appears in

.£ -.•*** ^ * twenty-four hours, and at

i • •*• its upper part a small cup

* of liquefaction forms with-

in forty-eight hours. At
FIG. 54.-Bacillus pyocyaneus ; young thig time a slightlv green.

<f*

gelatin. The liquefaction

extends pretty rapidly, the fluid portion being turbid and showing
masses of growth at its lower part. The green colour becomes more and
more marked, and diffuses through the gelatin. Ultimately liquefaction
reaches the wall of the tube. In plate cultures the colonies appear as
minute whitish points, those on the surface being the larger. Under a
low power of the microscope they have a brownish-yellow colour and
show a nodulated surface, the superficial colonies being thinner and
larger. Liquefaction soon occurs, the colonies on the surface forming
shallow cups with small irregular masses of growth at the bottom, the
deep colonies small spheres of liquefaction. Around the colonies a
greenish tint appears. On agar the growth forms an abundant slimy
greyish layer which afterwards becomes greenish, and a bright green
colour diffuses through the whole substance of the medium. On potatoes
the growth is an abundant reddish-brown layer resembling that of the
glanders bacillus, and the potato sometimes shows a greenish discoloration.
From the cultures there can be extracted by chloroform a coloured
body, pyocyanin, which belongs to the aromatic series, and crystallises
in the form of long, delicate bluish-green needles. On the addition of
a weak acid its colour changes to a red.

EXPERIMENTAL INOCULATION 209

This organism has distinct pathogenic action in certain animals.
Subcutaneous injection of small doses in rabbits may produce a local
suppuration, but if the dose be large, spreading haemorrhagic oedema
results, which may be attended by septicaemia. Intravenous injection
may produce, according to the dose, rapid septicaemia with nephritis, or
sometimes a more chronic condition of wasting attended by alliuminuria.

Micrococcus tetragenus. — This organism, first described by Galfky, is
characterised by the fact that it divides in two planes at right angles to
one another (Fig. 55), and is
thus generally found in the
tissues in groups of four, or
tetrads, which are often seen
to be surrounded by a cap-
sule. The cocci measure 1/4 f ^|>
in diameter. They stain J»
readily with all the ordinary C

stains, and also retain tin.-
stain in Gram's method. ^» * .

It grows readily on all the I ^ ^ 1

media at the room tempera- f
ture. In a puncture culture

on peptone-gelatin a pretty J «-.

thick whitish line forms ^

along the track of the needle, « ^

whilst on the surface there >-

is a thick rounded disc of *-

whitish colour. The gelatin
is not liquefied. On the sur-
face of a^ar and of potato *™- 5o.— Micrococcus tetragenus ; young

culture on agar, showing tetrads.
Staged with weatcarbol.f,,chsi,, x ,000.

•«
t
£ .

colour. The growth on all

the media has a peculiar viscid or tenacious character, owing to the

gelatinous character of the sheaths of the cocci.

White mice are exceedingly susceptible to this organism. Subcutaneous
injection is followed by a general septicaemia, the organism being found
in large numbers in the blood throughout the body. Guinea-pigs are
less susceptible ; sometimes only a local abscess with a good deal of
necrotic change results ; sometimes there is also septicaemia.

Experimental Inoculation. — We shall consider chiefly the
staphylococcus pyogenes aureus and the streptococcus pyogenes,
an these have been most fully studii-d.

It may be stated at the outset that the occurrence of suppura-
tion depends upon the number of organisms introduced into the
tissues, the number necessary varying not only in different
animals, but also in different parts of the same animal, — a
smaller number producing suppuration in the anterior chamber
of the eye, for example, than in the peritoneum. The virulence
of the organism also may vary, and corresponding results may
be produced. Especially is this so in the case of the strepto-
coccus pyogenr-.

210 INFLAMMATION AND SUPPURATION

The staphylococcus aureus, when injected subcutaneously in
suitable numbers, produces an acute local inflammation, which
is followed by suppuration, in the manner described above.
If a large dose is injected, the cocci may enter the blood stream
in sufficient numbers to cause secondary suppurative foci in
internal organs (cf. intravenous injection).

Intravenous injection in rabbits, for example, produces in-
teresting results, which vary according to the quantity used. If
a considerable quantity be injected, the animal may die in
twenty-four hours of a general septicaemia, numerous cocci being
present in the capillaries of the various organs, often forming
plugs. If a smaller quantity be used, the cocci gradually dis-
appear from the circulating blood; some become destroyed,
while others settle in the capillary walls in various parts and
produce minute abscesses. These are most common in the
kidneys, where they occur both in the cortex and medulla as
minute yellowish areas surrounded by a zone of intense con-
gestion and haemorrhage. Similar small abscesses may be
produced in the heart wall, in the liver, under the periosteum,
and in the interior of bones, and occasionally in the striped
muscles. Very rarely indeed, in experimental injection, do the
cocci settle on the healthy valves of the heart. If, however,
when the organisms are injected into the blood, there be any
traumatism of a valve, or of any other part of the body, they
show a special tendency to settle at these weakened points.

Experiments on the human subject have also proved the
pyogenic properties of those organisms. Garre inoculated
scratches near the root of his finger-nail with a pure culture, a
small cutaneous pustule resulting; and by rubbing a culture
over the skin of the forearm he caused a carbuncular condition
which healed only after some weeks. Confirmatory experiments
of this nature were made by Bockhart, Bumm, and others.

When tested experimentally, the staphylococcus pyogenes
albus has practically the same pathogenic effects as the staphylo-
coccus aureus.

The streptococcus pyogenes is an organism the virulence of
which varies much according to the diseased condition from
which it has been obtained, and also one which loses its virulence
rapidly in cultures. Even highly virulent cultures, if grown
under ordinary conditions, in the course of time lose practically
all pathogenic power. By passage from animal to animal, how-
ever, the virulence may be much increased, and pari passu the
effects of inoculation are correspondingly varied. Marmorek,
for example, found that the virulence of a streptococcus can be

BACILLUS COLI COMMUNIS 211

enormously increased by growing it alternately (a) in a mixture
of human blood serum and bouillon (vide p. 42), and (6) in
the body of a rabbit ; ultimately, after several passages it pos-
sesses a super-virulent character, so that even an extremely
minute dose introduced into the tissues of a rabbit produces
rapid septicaemia, with death in a few hours. It has been
proved by Marmorek's experiments, and those of others, that
the same species of streptococcus may produce at one time
merely a passing local redness, at another a local suppuration,
at another a spreading erysipelatous condition, or again a
general septicsemic infection, according as its virulence is
artificially increased. Such experiments are of extreme import-
ance as explaining to some extent the great diversity of lesions
in the human subject with which streptococci are associated.

Bacillus Coli Communis. — The virulence of this organism
also varies much, and can be increased by passage from animal
to animal. Injection into the serous cavities of rabbits pro-
duces a fibrinous inflammation which becomes purulent if the
animal lives sufficiently long. If, however, the virulence of the
organism be of a high order, death takes place before suppura-
tion is established, and there is a septicaemic condition, the
organisms occurring in large numbers in the blood. Intravenous
injection of a few drops of a virulent bouillon culture usually
produces a rapid septicaemia with scattered haemorrhages in
various organs.

Lesions in the Human Subject. — The following statement
may be made with regard to the occurrence of the chief organisms
mentioned, in the various suppurative and inflammatory con-
ditions in the human subject. The account is, however, by no
means exhaustive, as clinical bacteriology has shown that practi-
cally every part of the body may be the site of a lesion produced
by the pyogenic bacteria. It may also be noted that acute
catarrhal conditions of cavities or tubes are in many cases also
to be ascribed to their presence.

The stapkylococci are the most common causal agents in
localised abscesses, in pustules on the skin, in carbuncles, boils,
etc., in acute suppurative periostitis, in catarrhs of mucous
surfaces, in ulcerative endocarditis, and in various pyaemic
conditions. They may also be present in septicaemia.

Streptococci are especially found in spreading inflammation
with or without suppuration ; in diffuse phlegmonous and
erysipelatous conditions, suppurations in serous membranes and
in joints (Fig. 56). They also occur in acute suppurative
periostitis and ulcerative endocarditis. Secondary abscesses in

212

INFLAMMATION AND SUPPURATION

FIG. 56. — Streptococci in acute suppuration.

Corrosive film ; stained by Gram's method

and safranin. x 1000.

lymphatic glands and lymphangitis are also, we believe, more
frequently caused by streptococci than staphylococci. They also
produce fibrinous exudation on the mucous surfaces, leading

to the formation of false
membrane in many of the
cases of non-diphtheritic
inflammation of the
throat, which are met
with in scarlatina1 and
other conditions, and they
are also the organisms
most frequently present
in acute catarrhal inflam-
mations in this situation.
In puerperal peritonitis
they are frequently found
in a condition of purity,
and they also appear to
be the most frequent
cause of puerperal septi-
caemia, in which condition
they may be found after

death in the capillaries of various organs. In pyaemia they are
frequently present, though in most cases associated with other
pyogenic organisms. Some cases of enteritis in infants —
streptococcus enteritis — are also ' apparently due to a strepto-
coccus, which, however, presents in cultures certain points of
difference from the streptococcus pyogenes.

The bacillus coli communis is found in a great many inflam-
matory and suppurative conditions in connection with the ali-
mentary tract — for example, in suppuration in the peritoneum,
or in the extra peritoneal tissue with or without perforation of the
bowel, in the peritonitis following strangulation of the bowel, in
appendicitis and the lesions following it, in suppuration in and
around the bile ducts, etc. It may ateo occur in lesions in other
parts of the body, — endocarditis, pleurisy, etc., which in some
cases are associated with lesions of the intestine, though in others
such cannot be found. It is also frequently present in inflamma-
tion of the urinary passages, cystitis, pyelitis, abscesses in the
kidneys, etc., these lesions being in fact most frequently caused
by this or closely allied organisms.

In certain cases of enteritis it is probably the causal agent,

1 True diphtheria may also occasionally be associated with this diseases
usually as a sequel.

KN TRANCE AND SPREAD OF BACTERIA 213

though tliis is difficult of proof, as it is much increased in
numbers in practically all abnormal conditions of the intestine.
We may ivmurk that it has been repeatedly proved that the
bacillus coli cultivated from various lesions is more virulent than
that in the intestine, its virulence having been heightened by
growth in the tissues.

The micrococcus tetrayenus is often found in suppurations in
the region of the mouth or in the neck, and also occurs in
various lesions of the respiratory tract, in phthisical cavities,
abscesses in the lungs, etc. Sometimes it is present alone, and
probably has a pyogenic action in the human subject under
certain conditions. In other cases it is associated with other
organisms. Cases of pyaemia have been described in which this
organism was found in a state of purity in the pus in various
situations. In this latter condition the pus has been described
as possessing an oily, viscous character, and as being often
blood-stained.

The bacillus pyocyaneus is rarely found alone in pus, though
it is not infrequent along with other organisms. We have met
with it several times in cases of multiple abscesses, in association
with the staphylococcus pyogenes aureus. Lately some diseases
in children have been described in which the bacillus pyocyaneus
has been found throughout the body ; in these cases the chief
symptoms have been fever, gastro-intestinal irritation, pustular
or petechial eruptions on the skin, and general marasmus. It
has also been said to be constantly present in pemphigus, and
it certainly occurs in some cases of this disease.

Inflammatory and suppurative conditions, associated with the
organisms of special diseases, will be described in the respective
chapters.

Mode of Entrance and Spread. — Many of the organisms of
suppuration have a wide distribution in nature, and many also
are present on the skin and mucous membranes of healthy
individuals. Staphylococci are commonly present on the skin,
and also occur in the throat and other parts, and streptococci
can often be cultivated from the secretions of the mouth in
normal conditions. The pneumococcus of Fraenkel and the
pneumobacillus of Friedlander have also been found in the
mouth and in the nasal cavity, whilst the bacillus coli communis
is a normal inhabitant of the intestinal tract. The entrance of
these organisms into the deeper tissues when a surface lesion
occurs can be readily understood. Their action will, of course,
be favoured by any condition of depressed vitality. Though in
normal, conditions the blood is bacterium-free, we must suppose

214

INFLAMMATION AND SUPPURATION

that from time to time a certain number of such organisms gain
entrance to it from trifling lesions of the skin or mucous surfaces,
the possibilities of entrance from the latter being especially
numerous. In most cases they are killed by the action of the
healthy serum or cells of the body, and no harm results. If,
however, there be a local weakness, they may settle in that part
and produce suppuration, and from this other parts of the body
may be infected. Such a supposition as this is necessary to

FIG. 57. — Minute focus of commencing suppuration in brain— case
of acute ulcerative endocarditis. In the centre a small haemorrhage ;
to right side dark masses of staphylococci ; zone of leucocytes at
periphery.

Alum carmine and Gram's method, x 50.

explain many inflammatory and suppurative conditions met with
clinically. In some cases of multiple suppurations due to staphy-
lococcus infection, only an apparently unimportant surface lesion
is present ; whilst in others no lesion can be found to explain
the origin of the infection. The term cryptogenetic has been
applied by some writers to such cases in which the original point
of infection cannot be found, but its use is scarcely necessary.

The paths of secondary infection may be conveniently sum-
marised thus : First, by lymphatics ; in this way the lymphatic

ENTRANCE AND SPREAD OF BACTERIA 215

glands may be infected, and also serous sacs in relation to the
organs where the primary lesion exists. Second, by natural
channels, such as the ureters and the bile ducts, the spread
Itrin^ generally associated with an inflammatory condition of the
lining i-pithrlium. In this way the kidneys and liver respectively
may be infected. Third, by the blood vessels : (a) by a few
organisms gaining entrance to the blood from a local lesion, and

Kn:. 58.— Secondary infection of a glomerulus of kidney by the
, staphylococcus aureus, in a case of ulcerative endocarditis. The
cocci (stained darkly) are seen plugging the capillaries and also lying
free. The glomerulus is much swollen, infiltrated by leucocytes,
and partly necrosed.
Paraffin section ; stained by Gram's method and Bismarck-brown, x 300.

settling in a favourable nidus or a damaged tissue, the original
path of infection often being obscure; (6) by a septic phlebitis
with suppurative softening of the thrombus and resulting em-
bolism ; and we may add (c), by a direct extension along a vein,
producing a spreading thrombosis and suppuration within the
vein. In this way suppuration may spread along the portal vein
to the liver from a lesion in the alimentary canal, the condition
being known as pylephlebitis suppurativa.

216 INFLAMMATION AND SUPPURATION

Although many of the lesions produced by the bacteria
under consideration have already been mentioned, certain con-
ditions may be selected for further consideration on account of
their clinical importance or bacteriological interest.

Endocarditis. — There is now strong presumptive evidence
that all cases of endocarditis are due to bacterial infection. In
the simple or vegetative form, so often the result of acute
rheumatism, the micrococcus rheumaticus (p. 221) has been
cultivated from the valves in a certain number of cases, and is
probably the causal agent in most instances.

Endocarditis of the ulcerative type may be produced by
various organisms, chiefly pyogenic. Of these the staphylococci
and streptococci are most frequently found. In some cases of
ulcerative endocarditis following pneumonia the pneumococcus
(Fraenkel's) is present ; in these the vegetations often reach a
large size and have not so much tendency to break down as in
the case of staphylococcus infections. Other organisms have
been cultivated from different cases of the disease, and some of
these have received special names; for example, the diplo-
coccus endocarditis encapsulatus, bacillus endocarditidis griseus
(Weichselbaum), and others. In some cases the bacillus coli
communis has been found, and occasionally in endocarditis
following typhoid the typhoid bacillus has been described as the
organism present, but further observations on this point are
desirable. The gonococcus also has been shown to affect the
heart valves (p. 256), though this is a very rare occurrence.
Tubercle nodules on the heart valves have been found in a few
cases of acute tuberculosis, though no vegetative or ulcerative
condition is produced.

In some cases, though we believe not often, the organisms
may attack healthy valves, producing a primary ulcerative endo-
carditis, but more frequently the valves have been the seat* of
previous endocarditis, secondary ulcerative endocarditis being
thus produced. In some cases, especially when the valves have
been previously diseased, the source of the infection is quite
obscure. It is evident that as the vegetations are composed for
the most part of unorganised material, they do not offer the
same resistance to the growth of bacteria, when a few reach them,
as a healthy cellular tissue does. On microscopic examination
of the diseased valves the organisms are usually to be found
in enormous numbers (Fig. 59). By their action breaking
down of the vegetations occurs, and the emboli thus produced
act as the carriers of infection to other organs, and give rise to
secondary suppurations.

PERIOSTITIS AND OSTEOMYELITIS 217

••'iiftttal. — Occasionally ulcerative endocarditis is produced by the
simple intravenous injection of staphylococci aud streptococci into the
circulation, but this is a very rare occurrence. It often follows, however,
when tlie valves have heen previously injured. Orth and Wyssokowitsch
at a comparatively early date produced the condition by damaging the
aortic cusps hy a glass rod introduced through the carotid, and after-
wards injecting staphylococci into the circulation. Similar experiments
have since been repeated with streptococci, pneumococci, and other
organisms, with like result. Ribbert found that if a potato culture of

PlO. ~>9. — Section of a vegetation in ulcerative endocarditis showing
numerous staphylococci lying in the spaces. The lower portion is a
tia-nient in process of separation.

Stained by Gram's method and Bismarck-brown, x 600.

the staphylococcus aureus were rubbed down in salt solution so as to
form an emulsion, and then injected into the circulation, some minute
fragments became arrested at the attachment of the chordae tendinere and
produced an ulcerative endocarditis.

Acute Suppurative Periostitis and Osteomyelitis. — Special
mention is made of this condition on account of its comparative
frequency and gravity. The great majority of cases are caused
by the pyogenic cocci, of which one or two varieties may be

218 INFLAMMATION AND SUPPURATION

present, the staphylococcus aureus, however, occurring most
frequently. Pneumococci have been found alone in some cases,
and in a considerable number of cases following typhoid fever
the bacillus typhosus has been found alone. In others, again,
the bacillus coli communis is present.

The affection of the periosteum or interior of the bones by
these organisms, which is especially common in young subjects,
may take place in the course of other affections produced by
the same organisms or in the course of infective fevers, but in a
great many cases the path of entrance cannot be determined.
In the course of this disease serious secondary infections are
always very liable to follow, such as small abscesses in the
kidneys, heart- wall, lungs, liver, etc., suppurations in serous
cavities, and ulcerative endocarditis ; in fact, some cases present
the most typical examples of extreme general staphylococcus
infection. The entrance of the organisms into the blood stream
from the lesion of the bone is especially favoured by the
arrangement of the veins in the bone and marrow.

Experimental. — Multiple abscesses in the bones and under the peri-
osteum may occur in simple intravenous injection of the pyogenic cocci
into the blood, and are especially liable to be formed when young
animals are used. These abscesses are of small size, and do not spread
in the same way as in the natural disease in the human subject.

In experiments on healthy animals, however, the conditions are not
analogous to those of the natural disease. We must presume that in the
latter there is some local weakness or susceptibility which enables the
few organisms which have reached the part by the blood to settle and
multiply. Moreover, if a bone be experimentally injured, e.g. by actual
fracture or by stripping off the periosteum, before the organisms are
injected, then a much more extensive suppuration occurs at the injured
part.

Erysipelas. — A spreading inflammatory condition of the skin
may be produced by a variety of organisms, but the disease
in the human subject in its characteristic form is almost
invariably due to a streptococcus, as was shown by Fehleisen in
1884. He obtained pure cultures of the organism, and gave it
the name of streptococcus erysipelatis ; and, further, by inocu-
lations on the human subject as a therapeutic measure in
malignant disease, he was able to reproduce erysipelas. As
stated above, however, one after another of the supposed points
of difference between the streptococcus of erysipelas and that of
suppuration has broken down, and it is now generally held that
erysipelas is produced by the streptococcus pyogenes of a certain
degree of virulence. It must be noted, however, that erysipelas
passes from patient to patient as erysipelas, and purulent con-

CONJUNCTIVITIS 219

ditions due to streptococci do not appear liable to be followed
by erysipelas. On the other hand, the connection between
erysipelas and puerperal septicaemia is well established clinically.

In a case of erysipelas the streptococci are found in large
numbers in the lymphatics of the cutis and underlying tissues,
just beyond the swollen margin of the inflammatory area. As
the inflammation advances they gradually die out, and after a
time their extension at the periphery comes to an end. The
streptococci may extend to serous and synovial cavities and set
up inflammatory or suppurative change, — peritonitis, meningitis,
and synovitis may thus be produced.

Conjunctivitis. — A considerable number of organisms are
concerned in the production of conjunctivitis and its associated
lesions. Of these a mini
ber appear to be speci-
ally associated with this
region. Thus a small
organism,generally known /-
as the Koch-Weeks bacil- /
Ins, is the most common /
cause of acute contagious i
conjunctivitis, especially
prevalent in Egypt, but •
also common in this
country. This organism V
is very minute, being
little more than 1 /x in
length, and morphologic-
ally resembles the in-

tinmr/.} W-illiiQ • if« ™n FIG. 60.— Film preparation from a case of

acute conjunctivitis, showing Koch- Weeks

ditions of growth are bacilli, chiefly contained within a leucocyte.
even more restricted, as (From a preparation by Dr. Inglis Pollock.)
it rarely grows on blood x 1000<

agar, the best medium

b-.-ing serum agar. On this medium it produces minute
transparent colonies like drops of dew. The obtaining of pure
cultures is a matter of considerable difficulty, and it is nearly
always accompanied by the xerosis bacillus. It can readily be
found in the muco-purulent secretion by staining films with
weak (1:10) carbol-fuchsin, and is often to be seen in the interior
of leucocytes (Fig. 60). Another organism exceedingly like the
previous, apparently differing from it only in the rather wider
conditions of growth, is Miiller's bacillus. It was cultivated by
him in a considerable proportion of cases of trachoma, but its

220

INFLAMMATION AND SUPPURATION

relation to this condition is still matter of dispute. Another
bacillus which is now well recognised is the diplo-bacillus of
conjunctivitis first described by Morax. It is especially common
in the more subacute cases of conjunctivitis. Eyre found it in
2*5 per cent, of all cases of conjunctivitis. Its cultural characters
are given below. The xerosis bacillus, which is a small diph-
theroid organism (Fig. 119), has been found in xerosis of the
conjunctiva, in follicular conjunctivitis, and in other conditions ;
it appears to occur sometimes also in the normal conjunctiva.
It is doubtful whether it has any pathogenic action of importance.
Acute conjunctivitis is also produced by the pneumococcus,
epidemics of the disease being sometimes due to this organism,
and also by streptococci and staphylococci. True diphtheria of

the conjunctiva caused
by the Klebs - Loffler
bacillus also occurs,
whilst in gonorrhoeal
conjunctivitis, often of
an acute purulent type,
the gonococcus is pre-
sent (p. 255).

Diplo-bacillus of Con-
junctivitis. — This organ-
ism, discovered by Morax,
is a small plump bacillus,
measuring 1x2 /A, and
usually occurring in pairs,
or in short chains of pairs
(Fig. 61). It is non-motile,
does not form spores, and
is decolorised by Gram's
method. It does not grow
on the ordinary gelatin and
agar media, the addition
of blood or serum being
necessary. On serum it forms small rounded colonies which produce
small pits of liquefaction ; hence it has been called the bacillus
lacunatus. In cultures it is distinctly pleomorphous, and involution
forms also occur. It is non-pathogenic to the lower animals.

Acute Rheumatism. — There are many facts which point to
the infective nature of this disease, and investigations from this
point of view have yielded important results. Of the organisms
isolated, the one which appears to have strongest claims is a
small coccus observed by Triboulet, and by Westphal and
Wassermann, the characters and action of which were first
investigated in this country by Poynton and Paine. It is now

™ r.-, T,-VI ,. . ,

JIG. 61.— Film preparation of conjunctival

secretion, showing the Morax diplo-bacilhis
of conjunctivitis, x 1000.

ACUTE RHEUMATISM 221

usually spoken of as the micrococcus rhewmaticus. The organism
is sometimes spoken of as a diplococcus, but it is best described
as a streptococcus growing in short chains ; in the tissues, how-
ever, it usually occurs in pairs. It is rather smaller than the
streptococcus pyogenes, and although it can be stained by Gram's
method, it loses the colour more readily than the streptococcus.
In the various media it produces a large amount of acid, and
usually clots milk after incubation for two days; on blood agar
it alters the haemoglobin to a brownish colour. Its growth on
media generally is more luxuriant than that of the strepto-
coccus, and it grows well on gelatin at 20° C. Injection
of pure cultures in rabbits often produces polyarthritis and
synovitis, valvulitis and pericarditis, without any suppurative
change — lesions which it is stated are not produced by the
ordinary streptococci (Beattie). In one or two instances
choreiform movements have been observed after injection. The
organism is most easily obtained from the substance of inflamed
synovial membrane where it is invading the tissues ; a part
where there is special congestion should be selected as being
most likely to give positive results. It is only occasionally to
be obtained from the fluid in joints. It has also been cultivated
from the blood in rheumatic fever, from the vegetations on the
heart valves, and from other acute lesions ; in many cases, how-
ever, cultures from the blood give negative results. Beattie in
a recent paper has shown that in rabbits the arthritis produced
reproduces the main features of acute or subacute rheumatism
in man, namely, the rapidity with which the affection flies from
joint to joint, the tendency to relapses, the contributory effects
of exposure to cold, and the absence of gross anatomical changes
in the joints. Poynton and Paine cultivated the streptococcus
from the cerebro-spinal fluid in three cases where chorea was
present, and also detected it in the membranes of the brain.
They consider that this disease is probably of the nature of a
slight meningo-myelitis produced by the organism. The facts
already accumulated speak strongly in favour of this organism
being causally related to rheumatic fever, though this cannot
be considered completely proved. Andrewes finds that the
organism has the same cultural characters and fermentative
effects as the streptococcus fiecalis, a common inhabitant of the
intestine. Even, however, if the two organisms were the same,
it might well be j»ossible that rheumatic fever is due to an
infection of the tissues by this variety of streptococcus. The
clinical data, in fact, rather point to rheumatic fever being
due to an infection by some organism frequently present in the

222 INFLAMMATION AND SUPPURATION

body, brought about by some state of predisposition or acquired
susceptibility.

Vaccination Treatment of Infections by the Pyogenic Cocci.
— From his study of the part played by phagocytosis in the
successful combat of the pyogenic bacteria by the body, Wright
was led to advocate the treatment of such infections during their
course by active immunisation by means of dead cultures of the
infecting agent (for methods of preparation see p. 133). The
treatment is applicable when the infection is practically local,
as in acne pustules, in boils, etc. (For the theoretical questions
raised, see Immunity.) For an isolated faruncle, Wright recom-
mends a dose of 100 million staphylococci to be followed three or
four days later by the injection of 250 to 300 millions, and for an
incipient streptococcic lymphangitis a dose of 2,000,000 strepto-
cocci. In chronic infections the number of bacteria used for an
injection is from 250,000,000 to 500,000,000, but a smaller
number may give a good result, and the general principle to be
adopted is to use the smallest dose necessary for a therapeutic
effect. If repeated injections are necessary, Wright recommends
that the opsonic index should be observed every few days and
the injections only practised during a positive phase. If it is
not practicable to use the strain derived from the lesion for
the preparation of the vaccine, then laboratory cultures or the
stock vaccines which are now in the market may be used ; in
such cases it is well to use a vaccine made from a mixture of
strains ; in skin infections a mixture of staphylococcus aureus
and albus may be employed. Such means have been extensively
used in the treatment of acne, boils, sycosis, infections of the
genito-urinary tract by the b. coli, infections of joints by the
gonococcus, and in many cases considerable success has followed
the treatment. Sometimes acute affections, e.g. puerperal
septicaemia, have been treated by vaccines. It is difficult here
to judge of the results which have been attained, but in any
case only very small doses of the vaccine should at first, at any
rate, be administered.

Methods of Examination in Inflammatory and Suppurative
Conditions. — These are usually of a comparatively simple nature,
and include (1) microscopic examination, (2) the making of
cultures.

(1) The pus or other fluids should be examined microscopic-
ally, first of all by means of film preparations in order to
determine the characters of the organisms present. The films
should be stained (a) by one of the ordinary solutions, such
as carbol-thionin-blue (p. 105), or a saturated watery solution of

METHODS OF EXAMINATION 223

methylene-blue ; and (b) by Gram's method. The use of the
latter is of course of high importance as an aid in the recognition.

(2) As most of the pyogenic organisms grow readily on the
gelatin media at ordinary temperatures, pure cultures can be
readily obtained by the ordinary plate methods. But in many
cases the separation can be affected much more rapidly by the
method of successive streaks on agar tubes, which are then
incubated at 37° C. When the presence of pneumococci is
suspected, this method ought always to be used, and it is also to
l>e preferred in the case of streptococci. Inoculation experiments
may be carried out as occasion arises.

In cases of suspected blood infection the examination of the
blood is to be carried out by the methods already described
(p. 72).

CHAPTER VIII.

INFLAMMATORY AND SUPPURATIVE CONDITIONS,
CONTINUED: THE ACUTE PNEUMONIAS, EPI-
DEMIC CEREBRO-SPINAL MENINGITIS.

Introductory. — The term Pneumonia is applied to several con-
ditions which present differences in pathological anatomy and in
origin. All of these, however, must be looked on as varieties of
inflammation in which the process is modified in different ways,
depending on the special structure of the lung or of the parts
which compose it. There is, first of all, and, in adults, the com-
monest type, the acute croupous or lobar pneumonia, in which
an inflammatory process attended by abundant fibrinous exuda-
tion affects, by continuity, the entire tissue of a lobe or of a
large portion of the lung. It departs from the course of an
ordinary inflammation in that the reaction of the connective
tissue of the lung is relatively slight, and there is usually no
tendency for organisation of the inflammatory exudation to
take place. Secondly, there is the acute catarrhal or lobular
pneumonia, where a catarrhal inflammatory process spreads from
the capillary bronchi to the air vesicles, and in these a change,
consisting largely of proliferation of the endothelium of the
alveoli, takes place which leads to consolidation of patches of
the lung tissue. Up till 1889 acute catarrhal pneumonia was
comparatively rare except in children. In adults it was chiefly
found as a secondary complication to some condition such as
diphtheria, typhoid fever, etc. Since, however, influenza in an
epidemic form has become frequent, catarrhal pneumonia has
been of much more frequent occurrence in adults, has assumed
a very fatal tendency, and has presented the formerly quite
unusual feature of being sometimes the precursor of gangrene
of the lung. Besides these two definite types other forms also
occur. Thus instead of a fibrinous material the exudation may
be of a serous, hemorrhagic, or purulent character. Cases of
mixed fibrinous and catarrhal pneumonia also occur, and

224

TYPES OF PNEUMONIA 225

in the catarrhal there may be great leucocytic emigration.
Hemorrhages also may occur here.

Besides the two chief types of pneumonia there is another
group of cases which are somewhat loosely denominated septic
pneumonias, and which may arise in two ways : (1) by the
entrance into the trachea and bronchi of discharges, blood, etc.,
which form a nidus for the growth of septic organisms ; these
often set up a purulent capillary bronchitis and lead to infection
of the air cells and also of the interstitial tissue of the lung ; (2)
from secondary pyogenic infection by means of the blood stream
from suppurative foci in other parts of the body. (See chapter
on Suppuration, etc.) In these septic pneumonias various
changes, resembling those found in the other types, are often
seen round the septic foci.

In pneumonias, therefore, there may be present a great variety
of types of inflammatory reaction. We shall see that with all of
them bacteria have been found associated. Special importance
is attached to acute croupous pneumonia on account of its course
and characters, but reference will also be made to the other
forms.

Historical. — Acute lobar pneumonia for long was supposed to be an
effect of exposure to cold ; but many observers were dissatisfied with
this view of its etiology. Not only did cases occur where no such
exposure could be traced, but it had been observed that the disease
sometimes occurred epidemically, and was occasionally contracted by
hospital patients lying in beds adjacent to those occupied by pneumonia
oases. Further, the sudden onset and definite course of the disease con-
formed to the type of an acute infective fever ; it was thus suspected by
some to be due to a specific infection. This view of its etiology was
promulgated in 1882-83 by Friedliinder, whose results were briefly as
follows. In pneumonic lungs there were cocci, adherent usually in
pairs, and possessed of a definitely contoured capsule. These cocci
could be isolated and grown on gelatin, and on inoculation in mice they
produced a kind of septicaemia with inflammation of the serous membranes.
The blood and the exudation in serous cavities contained numerous
capsulated diplococci. There is little doubt that many of the organisms
seen by Friedlander were really Fraenkel's pneumococcus, to be presently
described.

By many observers it had been found that the sputum of healthy
men, when injected into animals, sometimes caused death, with the
same symptoms as in the case of the injection of Friedliinder's coccus ;
and in the blood and serous exudations of such animals capsulated
diplococci were found. A. Fraenkel found that the sputum of pneumonic
patients was much more fatal and more constant in its effects than that
of healthy individuals. The cocci which were found in animals dead of
this "sputum septicaemia," as it was called, differed from Friedlander's
cocci in several respects, to be presently studied. Fraenkel further
investigated a few cases of pneumonia, and isolated from them cocci
identical in microscopic appearances, cultures, and pathogenic effects,

226 THE ACUTE PNEUMONIAS

with those isolated in sputum septicaemia. The most extensive investi-
gations on the whole question were those of Weichselbaum, published
in 1886. This author examined 129 cases of the disease, including
cases not only of acute croupous pneumonia, but of lobular and septic
pneumonia. From them he isolated four groups of organisms. (1)
Diplococcus pneumonia;. This he described as an oval or lancet-formed
coccus, corresponding in appearance and growth characters to Fraenkel's
coccus. (2) Streptococcus pneumonia?. This on the whole presented
similar characters to the last, but it was more vigorous in its growth,
and could grow below 20° C. , though it preferred a temperature of 37° C.
(3) Staphylococcus pyogenes aureus. (4) Bacillus pneumonice. This was
a rod-shaped organism, and was identical with Friedlander's pneumo-
coccus. Of these organisms the diplococcus pneumoniae was by far the
most frequent. It also occurred in all forms of pneumonia. Next in
frequency was the streptococcus pneumonias, and lastly the bacillus
pneumonias. Inoculation experiments were also performed by Weichsel-
baum with each of the three characteristic cocci he isolated. The
diplococcus pneumonias and the streptococcus pneumonias both gave
pathogenic effects of a similar kind in certain animals.

The general result of these earlier observations was to establish
the occurrence in connection with pneumonia of two species of
organisms, each having its distinctive characters, viz. : —

1. Fraenkel's pneumococcus^ which is recognised to be identical
with the coccus of " sputum septicremia," with Weichselbaum's
diplococcus pneumonias, and with his streptococcus pneumonias.

2. Friedlander's pneumococcus (now known as Friedlander's
pneumobacillus), which is almost certainly the bacillus pneu-
moniae of Weichselbaum.

We shall use the terms " Fraenkel's pneumococcus " and
"Friedlander's pneumobacillus," as these are now usually
applied to the two organisms.

Microscopic Characters of the Bacteria of Pneumonia.—
Methods. — The organisms present in acute pneumonia can best
be examined in film preparations made from pneumonic lung
(preferably from a part in a stage of acute congestion or early
hepatisation), or from the gelatinous parts of pneumonic sputum
(here again preferably when such sputum is either rusty or
occurs early in the disease), or in sections of pneumonic lung.
Such preparations may be stained by any of the ordinary weak
stains, such as a watery solution of methylene-blue, but Gram's
method is to be preferred, with Bismarck-brown or Ziehl-Neelsen
carbol-f uchsin (one part to ten of water) as a contrast stain ;
with the latter it is best either to stain for only a few seconds,
or to overstain and then decolorise with alcohol till the ground
of the preparation is just tinted. The capsules can also be
stained by the methods already described (p. 109). In such
preparations as the above, and even in specimens taken from

BACTERIA IN PNEUMONIA 227

the lungs immediately after death (as may be quite well done
by means of a hypodermic syringe), putrefactive and other
bacteria may be present, but those to be looked for are capsulated
organisms, which may be of either or both of the varieties
mentioned.

(1) FraenkeVs Pneumococcus. — This organism occurs in the
form of a small oval coccus, about 1 p. in longest diameter,
arranged generally in pairs (diplococci), but also in chains of
four to ten (Fig. 62). The free ends are often pointed like a
lancet, hence the term

diplococcus lanceolatus .% ...

has also been applied to .X ' , '/; *

it. These cocci, in their •.•>"•

typical form, have round /^ * * ''* '.

them a capsule, which, in /j* f t . -

films stained by ordinary ^f

methods, usually appears I , „.« , tf

as an unstained halo, but 1 ;»>/

is sometimes stained more \ ji v , t" s, s '

deeply than the ground V?»; ,* *

of the preparation. This

difference in staining de- >•/

pends, in part at least, on

the amount of decolorisa-

tiou to which the prepara- m ^ _p.lm prepftrations of pneumonic

tlOll lias been subjected. sputum, showing numerous pneumococci

The capsule is rather (Fraenkel's) with unstained capsules ;

Kr^urlor tkan tlia KrtrJv sonie are arranged in short chains. See
broader tnan tne boay algo plate j ^ F?g 2

of the coccus, and has a Stained with carbol fuchsin. x 1000.

sharply defined external

margin. Often in sputum and even in the lung no capsule can
be demonstrated. The organism takes up the basic aniline
stains with great readiness, and also retains the stain in Gram's
method. It is the organism of by far the most frequent occur-
rence in true croupous pneumonia, and in fact may be said to
be rarely absent.

(2) Friedldnder's Pneumobacillus. — As seen in the sputum
and tissues, this organism, both in its appearance and arrange-
ment, as also in the presence of a capsule, somewhat resembles
Fraenkel's pneumococcus, and it was at first described as the
" pneumococcus." The form, however, is more of a short rod-
shape, and it has blunt rounded ends ; it is also rather broader
than Fraenkel's pneumococcus. It is now classed amongst the

illi, especially in view of the fact that elongated rod forms

228

THE ACUTE PNEUMONIAS

may occur

(Fig-

FIG. 63.— Friedlander's pneumobacillus, show-
ing the variations in length, also capsules.
Film preparation from exudate in a case of

pneumona,

1000.

pneumonia than Fraenkel's
latter; very rarely it
occurs alone.

In sputum prepara-
tions the capsule of
both pneurnobacteria
may not be recognis-
able, and the same is
sometimes true of lung
preparations. This is
probably due to changes
which occur in the cap-
sule as the result of
changes in the vitality
of the- organisms. Some-
times in preparations
stained by ordinary
methods the difficulty
of recognising the cap-
sule when it is present,
is due to the refractive
index of the fluid in
which the specimen is

capsule has the same general
characters as that of
Fraenkel's organism.
Friedlander's pneumo-
bacillus stains readily
with the basic aniline
stains, but loses the
stain in Gram's method,
and is accordingly col-
oured with the contrast
stain, — fuchsin or Bis-
marck-brown, as above
recommended. A valu-
able means is thus
afforded of distinguish-
ing it from Fraenkel's
pneumococcus in micro-
scopic preparations.

Friedlander's organ-
ism is much less fre-
quently present in
sometimes it is associated with the

FIG. 64. — Fraenkel's pneumococcus in serous
exudation at site of inoculation in a rabbit,
showing capsules stained.
Stained bv Rd. Muir's method, x 1000.

mounted being almost identical with

CULTIVATION OF PNEUMOCOCCTJS 229

that of the capsule. This difficulty can always be overcome by
having the groundwork of the preparation tinted.

The Cultivation of Fraenkel's Pneumococcus. — It is usually
difficult, and sometimes impossible, to isolate this coccus directly
from pneumonic sputum. On culture media it has not a vigorous
growth, and when mixed with other bacteria it is apt to be
overgrown by the latter. To get a pure culture it is best to
insert a small piece of the sputum beneath the skin of a rabbit
or a mm isc. In about twenty-four to forty -eight hours the
animal will die, with numerous capsul-
ated pneomooocci throughout its blood.
K PI 111 the heart-blood cultures can be
easily obtained. Cultures can also be
got post mortem from the lungs of
pneumonic patients by streaking a
number of agar or blood-agar tubes with
a scraping taken from the area of acute
congestion or commencing red hepatisa
tion, and incubating them at 37° ('.
The colonies of the pneumococcus appear
as almost transparent small discs which
have been compared to drops of dew
(Fig. 65). This method is also some-
times successful in the case of sputum.

The appearances presented in cultures Fro. K._Stroke culture of
by different varieties nt the pneuino- Fraenkel's pneumococcus
coccus vary somewhat. It always grows °n Mow I agar. The
best „„ l,l,,o,l son,,,, ,, „„ l>,ViHer'S !*&££?**£
blood agar. It usually grows well on four hours' growth at
ordinary agar or in bouillon, but not :;7 r- Natural size.
BO well on glycerin agar. In a stroke

culture on hlood serum, growth appears as an almost trans-
parent pellicle along the track, with isolated colonies at the
margin. On agar media it is more manifest, but otherwise
has similar characters. On agar plates colonies are very
transparent, but under a low power of the microscojK3 appear
to have a compact finely granular centre and a pale trans-
parent periphery. The appearances arc similar to those of
a culture of streptococcus pyogenes, but the growth is less
vigorous, and is more delicate in appearance. A similar
statement also applies to cultures in f/elatin at 2'2° C., growth
in a stab culture appearing as a row of minute points which
remain of small size; there is, of course, no. liquefaction of the
medium. In bouillon, growth forms a slight turbidity, \\hich

230 THE ACUTE PNEUMONIAS

settles to the bottom of the vessel as a slight dust-like deposit.
On potatoes, as a rule, no growth appears. Cultures may be
maintained for one or two months, if fresh sub-cultures are made
every four or five days, but they tend ultimately to die out.
They also rapidly lose their virulence, so that four or five days
after isolation from an animal's body their pathogenic action
is already diminished. Eyre and Washbourn, however, have
succeeded in maintaining cultures in a condition of constant
virulence for at least three months by growing the organisms

on agar smeared with

\ rabbit blood. The agar

I * must be prepared with

A' £** *s N Witte's peptone, must

/ V v not be heated over 100°

^ ^ / C., and after neutralisa-

| s -* «^«* V» tion (rosolic acid being

i k t ;$y used as the indicator)

must have '5 per cent, of

^ ** \ normal sodium hydrate

N V^ added. The tubes when

\ v» % inoculated are to be kept

r *. \ at 37°*5 C. and sealed

i^ % to prevent evaporation.

In ordinary artificial

•* ' media pneumococci usu-

FIG. 66. — Fraenkel's pueumococcus from a pure ally appear as diplococci
culture on blood agar of twenty-four hours' w:tilont fl Par>milp but
growth, some in pairs, some in short chains. Wltnout a capsule, t
Stained with weak carbol-fuchsin. x 1000. in preparations made

from the surface of agar

or from bouillon, shorter or longer chains may be observed
(Fig. 66). After a few days' growth they lose their regular shape
and size, and involution forms appear. Usually the pneumo-
coccus does not grow below 22° C., but forms in which the
virulence has disappeared often grow well at 20° C. Its optimum
temperature is 37° C., its maximum 42° C. It is preferably an
aerobe, but can exist without oxygen. It prefers a slightly alka-
line medium to a neutral, and does not grow on an acid medium.
In ordinary media the pneumococcus does not usually appear to
develop a capsule, but according to Hiss, the absence of a capsule
is often only apparent, and if in making cover-glass preparations
off such media some ox or rabbit serum be used as the diluent,
and the films stained by his copper-sulphate method (p. 109), a
capsule can be demonstrated. Capsulation frequently appears
in fluid serum media, e.g., can be readily recognised if the

CULTIVATION OF PNEUMOCOCCUS 231

organism be grown in rabbit or human serum which has been
obtained under aseptic precautions and heated for half an hour
at 55° C.

The pneuinococcus ferments saccharose, raffinose, and lactose,
and a similar fermentative action on inuliri is important, as
ordinary streptococci do not so readily ferment this sugar.
Apparently some samples of inulin are more readily acted on
than others. Usually the test is carried out with Hiss's inulin
serum water medium, in which coagulation of the serum results
(p. 47), but some investigators have had more success with
inulin bouillon, acid production being estimated by titration
against soda with a phenolphthalein indicator.

There has been described by Eyre and Washbourn a non-
pathogenic type of the pneumococcus which may be found in
the healthy mouth, and which may also be produced during the
saprophytic growth of the virulent form. From the latter it
differs generally in its more vigorous growth, in producing a
uniform cloud in bouillon, in slowly liquefying gelatin, and in
growing on potato. The facts that in cultures the pneumococcus
often grows in chains, and that occasionally streptococci are
found to develop capsules, have raised the question of the rela-
tionship of the pneumococcus to other streptococci. When,
however, biological characters are taken along with morpho-
logical, relatively little difficulty arises in the recognition of a
true pneumococcus. Here the reaction in inulin is important.
It may be said that the capacity of a capsulated organism to
produce acid from this sugar makes its being a true pneumococcus
extremely probable. That the pneumococcus may be related to
other streptococci is, however, shown by the fact that both sets
of organisms tend to originate common group agglutinins.

Considerable attention has been devoted to a bacterium
originally described by Schottmiiller, and called by him the
Streptococcus mucosus. This organism has been isolated from
a variety of suppurative conditions and also from certain cases of
pneumonia. In culture, it diners from the pneumococcus in the
colonies being more clear, transparent, and dewdrop-like, showing
great tendency to confluence, and being more slimy than those
of the pneumococcus. It coagulates the serum in Hiss's inulin
serum water medium. It is, pathogenic to white mice, but its
pathogenicity in the rabbit seems to be less than that of the true
pneumococcus. Its agglutinative reactions are somewhat pecu-
liar. Unlike the pneumococcus, it produces in animals only a
weak agglutinating serum, but such a serum often can agglu-
tinate pneumococci. Further, antipneumococcal sera frequently

232 THE ACUTE PNEUMONIAS

agglutinate the streptococcus mucosus more readily than other

streptococci. All the facts seem to point to this organism being

closely allied to the true pneumococcus.

The Cultivation of Friedlander's Pneumobacillus. — This

organism, when present in sputum or in a pneumonic lung, can
... i~< be readily separated by making ordinary

gelatin plate cultures, or a series of suc-
cessive strokes on agar tubes. The surface
colonies always appear as white discs
which become raised from the surface so
as to resemble little knobs of ivory.
From these, pure cultures can be readily
obtained. The appearance of a stab cul-
ture in gelatin is very characteristic.
At the site of the puncture, there is on
the surface a white growth heaped up,
it may be fully one-eighth of an inch,
_,;.;,% above the level of the gelatin; along the

needle track there is a white granular
appearance, so that the whole resembles
a white round-headed nail driven into
the gelatin (Fig. 67). Hence the name
" nail-like " which has been applied.
Occasionally bubbles of gas develop along
the line of growth. There is no lique-
faction of the medium. On sloped agar
it forms a very white growth with a
shiny lustre, which, when touched with
a platinum needle, is found to be of a
viscous consistence. In cultures much
^>i i M*^ longer rods are formed than in the

FIG. 67.-Stab culture of tisSUGS °f the body (FiS' 68)' On tlle
Friedlander's pneumo- surface of potatoes it forms an abundant
bacillus in peptone moist white layer. It is non-motile.
nalSe "appearance • Friedlander's bacillus has active ferment-
ten days' growth.' ing powers on sugars, though varieties
Natural size. isolated by different observers vary in

the degree in which such powers are
possessed. It always seems capable of acting on dextrose,
lactose, maltose, dextrin, and mannite, and sometimes also on
glycerin. The substances produced by the fermentation vary
with the sugar fermented, but include ethylic alcohol, acetic
acid, laevolactic acid, succinic acid, along with hydrogen and
carbonic acid gas. The amount of acid produced from lactose

OCCURRENCE OF PNEUMOBACTERTA 233

seems only exceptionally sufficient to cause coagulation of milk.
With regard to indol formation the results of different observers
vary. Here, as with other reactions, it is to be noted that only
strains isolated from
cases <>t pneumonia are
to be taken into account.
It is said by some that
the bacillus is identical
with an organism com-
mon in sour milk, and
also a normal inhabitant
of the human intestiiu',
namely, the bacterium

The Occurrence of
the Pneumobacteria in
Pneumonia and other
Conditions. Capsulated
organisms have been FlG- 68.— Friedliinder's pueumobacillus,i
c i > • from a young culture on agar, showing

found m every variety some n.d -shaped forms.

of the disease — in acute Stained \vith thionin-blue. x 1000.

croupous pneumonia, in

broncho-pneumonia, in septic pneumonia. In the great majority
of these it is Fraenkel's pneumococcus which both microscopic-
ally and culturally has been found to be present. Fried lander's
pneumobacillus occurs in only about 5 per cent, of the cases.
It may be present alone or associated with Fraenkel's organism.
In a case of croupous pneumonia the pneumococci are found all
through the affected area in the lung, especially in the exudation
in the air-cells. They also occur in the pleural exudation and
effusion, and in the lymphatics of the lung. The greatest number
are found in the parts where the inflammatory process is most
recent, e.g. in an area of acute congestion in a case of croupous
pneumonia, and therefore such parts are preferably to be selected
for microscopic examination, and as the source of cultures.
Sometimes there occur in pneumonic consolidation areas of
suppurative softening, which may spread diffusely. In such
areas the pneuniococci occur with or without ordinary pyogenic
organisms, streptococci being the commonest concomitants. In

1 The apparent size of this organism, on account of the nature of its sheath.
varies much according to the stain used. 1 1' stained with a strong stain, e.g.
carhol-fuchsin, its thi<-kne^ appears nearly twice as great as is shown in the
tigure.

234 THE ACUTE PNEUMONIAS

other cases, especially when the condition is secondary to in-
fluenza, gangrene may supervene and lead to destruction of large
portions of the lung. In these a great variety of bacteria, both
aerobes and anaerobes are to be found.

In ordinary broncho-pneumonias also Fraenkel's pneumo-
coccus is usually present, sometimes along with pyogenic cocci ;
in the broncho-pneumonias secondary to diphtheria it may be
accompanied by the diphtheria bacillus, and also by pyogenic
cocci ; in typhoid pneumonias the typhoid bacillus or the b. coli
may be alone present or be accompanied by the pneurno-
coccus, and in influenza pneumonias the influenza bacillus may
occur. In septic pneumonias the pyogenic cocci in many cases
are the only organisms discoverable, but the pneumococcus may
also be present. Especially important, as we shall see, from the
point of view of the etiology of the disease, is the occurrence in
other parts of the body of pathological conditions associated
with the presence of the pneumococcus. By direct extension to
neighbouring parts empyema, pericarditis, and lymphatic enlarge-
ments in the mediastinum and neck may take place ; in the first
the pneumococcus may occur either alone or with pyogenic cocci.
But distant parts may be affected, and the pneumococcus may
be found in suppurations and .inflammations in various parts of
the body (subcutaneous tissue, peritoneum (especially in children),
joints, kidneys, liver, etc.), in otitis media, ulcerative endocarditis
(p. 216), and meningitis. In fact, there is practically no inflam-
matory or suppurative condition in the body in which the
pneumococcus in pure culture may not be found. These condi-
tions may take place either as complications of pneumonia, or
they may constitute the primary disease. The occurrence of
meningitis is of special importance, for next to the lungs the
meninges appear to be the parts most liable to attack by the
pneumococcus. A large number of cases have been investigated
by Netter, who gives the following tables of the relative fre-
quency of the primary infections by the pneumococcus in man : —

(1) In adults-
Pneumonia . . . . . . . 65 '95 per cent.

Broncho-pneumonia) 15 -85

Capillary bronchitis)

Meningitis . • 13'00

Empyema . . . . . . . 8'53

Otitis 2-44

Endocarditis 1'22

Liver abscess . . . . . . ,1'22

(2) In children 46 cases were investigated. In 29 the primary affection
was otitis media, in 12 broncho-pneumonia, in 2 meningitis, in 1 pneu-
monia, in 1 pleurisy in 1 pericarditis.

EXPERIMENTAL INOCULATION 235

Thus in children the primary source of infection is in a great
many cases an otitis media, and Netter concludes that infection
takes place in such conditions from the nasal cavities.

As bearing on the occurrence of pneumococcal infections
.secondary to such a local lesion as pneumonia, it is important to
note that in a large proportion of cases of the latter disease the
jineumococcus can be isolated from the blood.

Experimental Inoculation. — The pneumococcus of Fraenkel is
pathogenic to various animals, though the effects vary somewhat
\vith the virulence of the race used. The susceptibility of
different species, as Gamaleia has shown, varies to a considerable
extent. The rabbit, and especially the mouse, are very sus-
ceptible ; the guinea-pig, the rat, the dog, and the sheep
occupy an intermediate position ; the pigeon is immune. In
the more susceptible animals the general type of the disease
produced is not pneumonia, but a general septiccemia. Thus, if
a rabbit or a mouse be injected subcutaneously with pneumonic
sputum, or with a scraping from a pneumonic lung, death
occurs in from twenty-four to forty-eight hours. There is some
tibrinous infiltration at the point of inoculation, the spleen is
often enlarged and firm, and the blood contains capsulated
pneumococci in large numbers (Fig. 69). If the seat 'of inocula-
tion be in the lung, there generally results pleuritic effusion on
both sides, and in the lung there may be a process somewhat
resembling the early stage of acute croupous pneumonia in man.
There are often also pericarditis and enlargement of spleen.
We have already stated that cultures of the pneumococcus on
artificial media in a few days begin to lose their virulence.
Now, if such a partly attenuated culture be injected sub-
cutaneously into a rabbit, there is greater local reaction ;
pneumonia, with exudation of lymph on the surface of the
pit-lira, and a similar condition in the peritoneum, may occur.
In sheep greater immunity is marked by the occurrence, after
subcutaneous inoculation, of an enormous local sero-fibrinous
exudation, and by the fact that few pneumococci are found in
the blood stream. Intra-pulmonary injection in sheep is
followed by a typical pneumonia, which is generally fatal. The
dog is still more immune ; in it also intra-pulmonary injection is
followed by a fibrinous pneumonia, which is only sometimes
fatal. Inoculation by inhalation appears only to have been
performed in the susceptible mouse and rabbit; here also
septicaemia resulted.

The general conclusion to be drawn from these experiments
thus is that in highly susceptible animals virulent pneumococci

236

THE ACUTE PNEUMONIAS

produce a general septicaemia ; whereas in more immune species
there is an acute local reaction at the point of inoculation, and
if the latter be in the lung, then there may result pneumonia,
which, of course, is merely a local acute inflammation occurring
in a special tissue, but identical in essential pathology with an
inflammatory reaction in any other part of the body. When a
dose of pneumococci sufficient to kill a rabbit is injected sub-

FIG. 69. — Capsulated pueumococcus in blood taken from the heart
of a rabbit, dead after inoculation with pneumonic sputum.

Dried film, fixed with corrosive sublimate. Stained with oarbol-
fuchsin and partly decolorised, x 1000.

cutaneously in the human subject, it gives rise to a local inflam-
matory swelling with redness and slight rise of temperature, all
of which pass off in a few days. It is therefore justifiable to
suppose that man occupies an intermediate place in the scale of
susceptibility, probably between the dog and the sheep, and
that when the pneumococcus gains an entrance to his lungs the
local reaction in the form of pneumonia occurs. In this con-
nection the occurrence of manifestations of general infection
associated with pneumonia in man is of the highest import-

EXPERIMENTAL INOCULATION 237

ance. We have seen that meningitis and other inflammations
are not very rare complications of the disease, and such cases
form a link connecting the local disease in the human subject
with the general septicaemic processes which may be produced
artificially in the more susceptible representatives of the lower
animals.

A fact which at first appeared rather to militate against the
pneumococcus being the cause of pneumonia was the discovery
of this organism in the saliva of healthy men. This fact was
early pointed out by Pasteur, and also by Fraenkel, and the
observation has been confirmed by many other observers. It
can certainly be isolated from the mouths of a considerable
proportion of normal men, from their nasal cavities, etc., being
probably in any particular individual more numerous at some
times (especially, it is stated, during the winter months, i.e. a
little before the period of the greatest prevalence of pneumonia)
than at others, HI id sometimes being entirely absent. This
can be proved, of course, by inoculation of susceptible animals.
Such a fact, however, only indicates the importance of pre-
disposing causes in the etiology of the disease, and it is further
to be observed that we have corresponding facts in the case of
the diseases caused by pyogenic staphylococci, streptococci, the
bacillus coli, etc. It is probable that by various causes the
vitality and power of resistance of the lung are diminished, and
that then the pneumococcus gains an entrance. In relation to
this possibility we have the very striking fact that in the
irregular forms of pneumonia, secondary to such conditions as
typhoid and diphtheria, the pneumococcus is very frequently
present, alone or with other organisms. Apparently the effects
produce;! by such bacteria as the b. typhosus and the b.
diphtherias can devitalise the lung to such an extent that
secondary infection by the pneumococcus is more likely to occur
and set up pneumonia. We can therefore understand how
much less definite devitalising agents such as cold, alcoholic
excess, etc., can play an important part in the causation of
pneumonia. In this way also other abnormal conditions of the
respiratory tract, a slight bronchitis, etc., may play a similar
part.

It is more difficult to explain why sometimes the pneumo-
coccus is associated with a spreading inflammation, as in croupous
pneumonia, whilst at other times it is localised to the catarrhal
patches in broncho-pneumonia. It is quite likely that in the
former condition the organism is possessed of a different order
of virulence, though of this \ve have no direct proof. We have,

238 THE ACUTE PNEUMONIAS

however, a closely analogous fact in the case of erysipelas ; this
disease, we have stated reasons for believing, is produced by
a streptococcus which, when less virulent, causes only local
inflammatory and suppurative conditions.

Summary. — We may accordingly summarise the facts re-
garding the relation of Fraenkel's pneumococcus to the disease
by saying that it can be isolated from nearly all cases of acute
croupous pneumonia, and also from a considerable proportion
of other forms of pneumonia. When injected into the lungs of
moderately insusceptible animals it gives rise to pneumonia. If,
in default of the crucial experiment of intra-pulmonary injection
in the human subject, we take into account the facts we have
discussed, we are justified in holding that it is the chief factor in
causing croupous pneumonia, and also plays an important part
in other forms. Pneumonia, in the widest sense of the term, is,
however, not a specific affection, and various inflammatory con-
ditions in the lungs can be set up by the different pyogenic
organisms, by the bacilli of diphtheria, of influenza, etc.

The possibility of Friedlander's pneumobacillus having an
etiological relationship to pneumonia has been much disputed.
Its discoverer found that it was pathogenic towards mice and
guinea-pigs, and to a less extent towards dogs. Rabbits appeared
to be immune. The type of the disease was of the nature of a
septicaemia. No extended experiments, such as those performed
by Gamaleia with Fraenkel's coccus, have been done, and there-
fore we cannot say whether any similar pneumonic effects are
produced by it in partly susceptible animals. The organism
appears to be present alone in a small number of cases of
pneumonia, and the fact that it a^so appears to have been the
only organism present in certain septic£emic complications of
pneumonia, such as empyema and meningitis, render it possible
that it may be the causal agent in a few cases of the disease. It
is also stated to have been observed in certain cases of appen-
dicitis and occasionally in pysemic cases.

In the septic pneumonias the different pyogenic organisms
already described are found, and sometimes in ordinary
pneumonias, especially the catarrhal forms, other organisms,
such as the b. coli or its allies, may be the causal agents.

The Pathology of Pneumococcus Infection. — The effects of
the action of the pneumococcus, at any rate in a relatively
insusceptible animal such as man, seem to indicate that toxins
may play an important part. Pneumonia is a disease which
presents in many respects the character of an acute poisoning.
In very few cases does death take place from the functions of

PNEUMOCOCCUS INFECTION 239

the lungs being interfered with to such an extent as to cause
asphyxia. It is from cardiac failure, from grave interference
with the heat-regulating mechanism, and from general nervous
depression that death usually results. These considerations,
taken in connection with the fact that in man the organisms are
found in the greatest numbers in the lung, suggest that a toxic
action is at work. Various attempts have been made to isolate
toxins from cultures of the pneumococcus, e.g., by precipitating
bouillon cultures with alchohol or ammonium sulphate, and
poisonous effects have been produced by certain substances thus
derived ; but the effects produced are, as in so many other
similar cases, of a non-specific character, and to be classed as
interferences with general metabolism. The general conclusion
has been that the toxins at work in pneumonia are intracellular ;
but no special light has been thrown on the common effects of
the members of this group of bacterial poisons.

Immunisation against the Pneumococcus. — Animals can be
immunised against the pneumococcus by inoculation with
cultures which have become attenuated by growth on artificial
media, or" with the naturally attenuated cocci which occur in the
sputum after the crisis of the disease. Netter effected immun-
isation by injecting an emulsion of the dried spleen of .an animal
dead of pneumococcus septicaemia. Virulent cultures killed by
heating at 62° C., rusty sputum kept at 60° C. for one to two
hours and then filtered, and filtered or unfiltered bouillon
cultures similarly treated have also been used. In all cases one
or two injections, at intervals of several days, are sufficient for
immunisation, but the immunity has often been observed to
be of a fleeting character and may not last more than a few
weeks. The serum of such immunised animals protects rabbits
against subsequent inoculation with pneumococci, and if injected
within twenty-four hours after inoculation, may prevent death.
A protective serum was obtained by Washbourn, who employed
pneumococcus cultures of constant virulence. This observer
immunised a pony by using successively (1) broth cultures killed
by one hour's exposure to 60° C. ; (2) living agar cultures ; (3)
living broth cultures. From this animal there was obtained a
serum which protected susceptible animals against many times
an otherwise fatal dose, and wThich also had a limited curative
action. It is stated that the serum of patients who have
recovered from pneumonia has in a certain proportion of cases
a protective effect against the pneumococcus in rabbits, similar
to that exhibited by the serum of immune animals.

The Klemperers treated a certain number of cases of human

240 THE ACUTE PNEUMONIAS

pneumonia by serum derived from immune animals, and appar-
ently with a certain measure of success, and sera prepared by
Washbourn and by others have also been used. The results
obtained by different observers have, however, been rather con-
tradictory. The use of these sera apparently causes the tempera-
ture in some cases to. fall, and even may hasten a crisis, but
further experience is necessary before their value in therapeutics
can be properly estimated.

There has been considerable difference of opinion as to the
explanations to be given of the facts observed regarding im-
munisation against the pneumococcus and especially regarding
the properties of immune sera. At first these sera were supposed
to possess antitoxic qualities — largely on the ground that no
bactericidal effect was produced by them on the bacterium in
vitro. As no specific toxin has been proved to be concerned in
the action of the organism, the development of an antitoxin
during immunisation must, in the present state of knowledge,
be looked on as not yet proved. To explain the action of a
serum in preventing and curing pneumococcal infections, it has
been thought to have the complex character seen in anti-typhoid
sera in which two substances — immune body and complement
(see Immunity) — are concerned, and the variability in the
therapeutic results obtained has been accounted for on the view
that there might be a deficiency of complement, such as occurs
in other similar cases. The absence of bactericidal effect, how-
ever, raises several difficult points. It is stated that no such
effect is observable either in immune sera, or in the serum of
patients who have successfully come through an attack of the
natural disease. Some effect of the kind would be expected to
be present if the anti-pneumonic serum were quite comparable
to the anti-typhoid serum. Within recent times many have
turned to the opsonic property of sera to account for the facts
observed. In this connection Mennes observed that normal
leucocytes only become phagocytic towards pneumococci when
they are lying in the serum of an animal immunised against
this bacterium. Wright had in his early papers looked to the
phagocytosis of sensitised bacteria to explain their destruction
in the absence of bactericidal qualities in the serum alone, and
Neufeld and Rlmpau have described the occurrence of an
opsonic effect in the action of an anti-pneumococcic serum.
Further work may show that along these lines lies the explana-
tion of the facts observed.

In studying further the relationship of the opsonic effect to
pneumococcal infection, inquiry has been directed to the opsonic

M KTHODS OF EXAMINATION 241

qualities of the blood of pneumonic patients, especially with a
view to throwing light on the nature of the febrile crisis.
According to some results, the opsonic index as compared with
that of a healthy person is not above normal, but if the possible
phagocytic capacities of the whole blood of the sick person be
taken into account, these will probably be much above normal
in consequence of the leucocytosis which usually accompanies a
successful resistance to this infection. It has been observed,
however, that as the crisis approaches in a case which is to
recover the opsonic index rises, and after defervescence gradually
falls to normal. As bearing on the factors involved in the
successful resistance of the organism to the pneumococcus, it
has been noted that avirulent pneumococci are more readily
up^mised than more virulent strains. It is further stated that
avirulent cultures of the pneumococcus can be made to resist
phagocytosis if they are treated with the products of the
autolysis of virulent strains or with washings from such strains,
and that virulent cocci if washed with saline become capable of
1 is -ing readily phagocyted. Further observations along such
lines are to be looked for with interest, and it may be said that
Wright's vaccination methods have been applied to the treatment
of pneumonic cases, and in certain instances are said to have
been followed by favourable result. It may be noted here, in
conclusion, that in man it is probable that immunity against
pneumonia may be short-lived, as in a good many cases of
pneumonia a history of a previous attack is elicited.

Agglutination of the Pneumococcus. — If a small amount of a
culture of Fraenkel's pneumococcus be placed in an anti-pneumo-
coccic serum, an aggregation of the bacteria into clumps occurs.
Such an agglutination, as it is called, is frequently observed
under similar circumstances with other bacteria. The pheno-
menon is not invariably associated with the presence of protective
bodies in a serum, but it has been used for diagnostic purposes
in the differentiation of sore throats due to pueumococcus infec-
tion from those due to other bacteria. Whether the method is
reliable has still to be proved. It has been shown that a serum
which agglutinates the pneumococcus may also agglutinate
streptococci isolated from various sources. Such organisms are,
however, not so uniformly agglutinated by a pneumococcus
serum as are pneumococci isolated from pneumonic cases.

Methods of Examination. — These have been already

described, but may be summarised thus : (1) Microscopic.

Stain films from the densest part of the sputum or from the

area of spreading inflammation in the lung by Grain's method

242 EPIDEMIC CEREBRO-SPINAL MENINGITIS

and by carbol-fuchsin, etc. (pp. 105, 108), in the latter case
it is usually best not to decolorise the groundwork of the
preparation.

(2) By cultures, (a) FraenkeVs pneumococcus. With similar
material make successive strokes on agar, blood agar, or blood
serum. The most certain method, however, is to inject some
of the material containing the suspected cocci into a rabbit. If
the pneumococcus be present the animal will die, usually within
forty -eight hours, with numerous capsulated pneumococei in its
heart blood. With the latter inoculate tubes of the above media
and observe the growth. In some cases of severe pneumococcic
infection the organism may be cultivated from the blood obtained
by venesection (p. 72). (b) Friedlander 's pneumohacillus can
be readily isolated either by ordinary gelatin plates or by
successive strokes on agar media.

EPIDEMIC CEKEBRO-SPINAL MENINGITIS.

As the result of observations on this disease in different parts
of the world, it has been now established that the causal agent

is the diplococcus intra-
cellularis meningitidis,
first described by Weich-
selbaum, and now often
known as the meningo-
coccus. This organism is
a small coccus measuring
about 1 /ji in diameter
and usually occurs in
pairs, the adjacent sides
being somewhat flattened
against each other. In
most cases the cocci are
chiefly contained within
polymorphonuclear leuco-
cytes in the exudation

FIG. 70. — Film preparation of exudation from (Fig. 70) ; ill some cases,

a case of meningitis showing -the diplococci however) the majority

within leucocytes. See also Plate I., Fig. 3. > . J J

Stained with carbol-thionin-blue. x 1000. may be lying tree. It

stains readily with basic

aniline dyes, but loses the stain in Gram's method. Both
in appearance and in its staining reactions it is similar to
the gonococcus (vide infra}. The organism can readily be
cultivated outside the body, but the conditions of growth

THE MENINGOOOOOtfS 243

are < . >iin-\\ hat restricted — u^ar with an admixture of serum
or blood (preferably human) is most suitable (p. 43). Strains
separated in different epidemics appear to present slight
individual variations, but the following description may be
taken as summing up the common characters: — Growth takes
place best at the temj>erature of the body, and practically
(•t-ases at 25° C. On serum agar the colonies are circular discs
of somewhat transparent appearance, and possessing a smooth,
shining surface ; they have little tendency to become confluent.
When examined under a low magnification the colour is seen to
be somruliat yellowish, and the margins usually are smooth
and regular, though on

some media slight crena- • " ^

tion may appear. The '-.***

colonies may be of con- 9 *• %

siderable size, reaching •*' . "'*""* **t, »

sometimes a diameter of * " * ./, * ». »*k

2 to 3 mm. on the third * ." ' '

day. On plain agar the *t

colonies are very niudi • * j£% ** '

smaller, and sometimes *

no growth occurs; sub- -

cultures esi>ecially often

fail to give any growth • • • „* ^ * • • t

on this medium. In N * ^ *. j ..

serum bouillon the organ- •/"•*«• > *"•

ism ]>roduces a general

turbidity with formation Vu;. 71.— Pure. culture of <liplococcus intra-

of some deposit after a '-fllulsiris, sliowiiiK involution forms.

day or two. It ferments

maltose and dextrose with acid production, a pro}>erty which

distinguishes it from the micrococcus catarrhalis (vide infra).

Fermentation tests are most satisfactorily carried out by means

of solid serum media containing 1' per cent, of the sugar to be

tested (p. 80). In all cases growth occurs best when the

medium ha- a neutral or very slightly alkaline reaction. In

cultures the organism presents the same appearance as in the

body, and often shows tetrad formation. There is also a great

tendency to the production of involution forms (Fig. 71), many

of the cocci becoming much swollen, staining badly, and after-

wards undergoing disintegration. This change, according to

Flexner's observations, would api>ear to be due to the production

of an autolytic enzyme, and he has also found that this substance

has the property of producing dissolution of the bodies of other

244 EPIDEMIC CEREBRO-SPINAL MENINGITIS

bacteria. The life of the organism in cultures is a comparatively
short one ; after a few days cultures will often be found to be
dead, but, by making sub-cultures every three or four days,
strains can be maintained alive for considerable periods. The
organism is readily killed by heat at 60° C., and it is also very
sensitive to weak antiseptics ; drying for a period of a day has
been found to be fatal to it. The facts established accordingly
show it to be a somewhat delicate parasite.

As stated above, the organism occurs in the exudate in the
meninges and in the cerebro-spinal fluid, and it can usually be
obtained by lumbar puncture. In acute cases, especially in the
earlier stages, it is usually abundant ; but in the later stages of
cases of more sub-acute character, its detection may be a matter
of difficulty, and only a few examples may be found after a
prolonged search ; in extremely acute cases also the organism
may be difficult to demonstrate. In most cases the disease is
practically restricted to the nervous system, but occasionally
complications occur, and in these the organism may sometimes
be found. It has been observed, for example, in arthritis, peri-
carditis, pneumonic patches in the lung, and in other inflam-
matory conditions associated with the disease. In a certain
number of cases it has been obtained from the blood during life,
but cultures in most instances give negative results.

Experimental inoculation shows that the ordinary laboratory
animals are relatively insusceptible to this organism. An in-
flammatory condition may be produced in mice and guinea-pigs
by intra-peritoneal injection, and a fatal result may follow ; in
such cases the organism does not seem to undergo very active
multiplication, though it may sometimes be cultivated from the
blood, and none of the lesions in the nervous system are repro-
duced. Flexner and Stuart M 'Donald have shown that cerebro-
spinal meningitis may be produced in monkeys by injections of
the organism into the spinal canal, the latter observer finding
that exudate containing mehingococci was more effective than
cultures. In such experiments the organism extends upwards
to the brain, and produces meningitis within a very short time.
The resulting lesions, both as regards their distribution and
general characters, and also as regards the histological changes,
resemble the disease in the human subject. Even these animals,
however, are manifestly less susceptible than the human
subject.

Many questions of great importance with regard to the spread
of the disease still require further investigation. The organism
has been obtained by culture from the throat and nasal cavities

SERUM REACTIONS 245

of those suffering from the disease in a considerable number of
instances. It has also been obtained from the same positions in
healthy individuals, during an epidemic of the disease. In
some epidemics also a pharyngitis has been found to occur, and
the organism has been obtained from the affected fauces. The
majority of workers at this subject are inclined to believe that
the organism spreads by means of the lymphatics from the
pharynx or nose to the base of the brain, but direct evidence
that this occurs has not been supplied. On the other hand, the
facts established with regard to other infections make it quite
probable that the organism gains entrance to the blood stream
from the upper respiratory passages, and then settles in the
meninges. Infection by the alimentary canal, the organisms
thereafter reaching the spinal meninges by the lymphatics, has
been suggested as a possibility, but such a view does not appear
to have much support. Whatever may be found to be the path
by which the organism reaches the brain, the evidence at
present tends to show that the entrance of the organism into
the body is by the naso-pharynx, and that this usually results
by inhalation of the organism distributed in fine particles of
expectoration, etc. In fact, as regards the mode and conditions
of infection, an analogy would appear to hold between this
disease and influenza.

Apart from the epidemic form of the disease, cases of sporadic
nature also occur, in which the lesions are of the same nature,
and in which the meningococcus is present. The facts stated
would indicate that the origin and spread of the disease in the
epidemic form depend on certain conditions which produce an
increased virulence of the organism. We are, however, as yet
entirely ignorant as to what these conditions may be. In
simple posterior basal meningitis in children a diplococcus is
present, as described by Still, which has the same microscopic
and cultural characters as the diplococcus intracellularis ; it has
been regarded as probably an attenuated variety of the latter.
Recently, however, Houston and Ran kin have found that the
serum of a patient suffering from epidemic meningitis does not
exert the same opsonic and agglutinative effects on the diplo-
coccus of basal meningitis as on the diplococcus intracellularis ;
and this result points to the two organisms being distinct,
though closely allied, species.

Serum Reactions. — An agglutination reaction towards the
• lijiplococcus intracellularis is given by the serum of patients
suffering from the disease, where life is prolonged for a sufficient
length of time. It usually appears about the fourth day, when

246 EPIDEMIC CEREBRO-SPINAL MENINGITIS

the serum may give a positive reaction in a dilution of 1 : 50 ;
at a later stage it lias been observed in so great a dilution as
1 : 1000. Specific opsonins may appear in the blood about
the same time, and though they are not always proportional in
amount to the agglutinins, the two classes of substances have
pretty much the same significance, and may occasionally be of
use in diagnosis when lumbar puncture fails to give positive
results. Although their presence in large amounts may be said
to indicate a marked reaction, they do not supply information of
much value in relation to prognosis. Immune-bodies, as shown
by bactericidal and deviation of complement tests (p. 126, 130),
may also be developed in considerable amount in the course of
the disease.

Anti-sera for therapeutical purposes have been introduced by
various workers, and of these the one which has been most
extensively used is that of Flexner and Jobling. This serum
is prepared from the horse by repeated injections in increasing
doses of dead cultures, followed by injections of culture autolysate
and of living cultures, these two latter being best administered
by the subcutaneous method. Several strains of meningococci
are mixed together for purposes of injection, and the immunisa-
tion is continued over a period of several months. For treat-
ment of the disease the serum is injected under the spinal
dura, 30 c.c. being generally used for an injection, this being
continued for several days. This serum has been used on a
large scale in various parts of the world, and there is general
agreement as to its favourable effects — the mortality of the
disease, which is generally 70 to 80 per cent., having been
reduced to about 30 per cent, or even less. By means
of its use the tendency to the occurrence of chronic lesions
has also been markedly diminished. The action of such
anti-sera cannot as yet be fully explained. They certainly
contain opsouins, agglutinins, immune-bodies which bind com-
plement, and possibly also anti-endotoxins. After the injection
the number of meningococci becomes markedly reduced,
probably as a result of increased phagocytosis; there can
scarcely be any direct bactericidal action owing to the absence
of complement. The standardisation of such anti-sera is a
matter of some difficulty ; at first the deviation of complement
method was used (p. 130), but now the opsonic index is regarded
with more favour as an index of the potency of the serum.
Mackenzie and Martin have treated cases by the intra-spinal
injection of the fresh serum of patients suffering from the
disease or who have recovered from it, such serum being in

ALLIED DIPLOCOCCI 247

many cases rich in immune-bodies for the meningococcus, and
possessing a greatly increased bactericidal action as compared
with normal serum. Though the number of cases treated by
this method is not yet large, a distinctly favourable result has
been obtained.

Allied Diplococci. — In the naso-pharynx there occur other
Gram-negative diplococci which have j a close resemblance to
the diplococcus intracellularis. These occur in the healthy
state, but are especially abundant in catarrhal conditions. Of
these the diplococcus catarrhalis has the closest resemblance to
the diplococcus intracellularis. In addition to occurring in
health this organism has also been found in large numbers in
•epidemic catarrh. Its microscopic appearances are practically
similar to those described above, and it also occurs within leuco-
cytes. Its colonies on serum agar are more opaque than those
of the diplococcus intracellularis, and often have a somewhat
firm though friable consistence, so that they are sometimes
removed en masse by the platinum needle. The organism grows
on gelatin at 20° C. without liquefying the medium, and it has
none of the fermentative properties described above as belonging
to the diplococcus intracellularis. The diplococcus pharyngis
mccus (v. Lingelsheim) also grows at room temperature, and its
colonies are very tough and adhere to the surface of the medium ;
it can thus readily be distinguished from the meningococcus. It
has marked fermentative properties, acting on glucose, maltose,
saccharose, and laevulose. The diplococcus mucosus has colonies
of slimy consistence ; it grows at room temperature, and it forms
capsules, which can be demonstrated by the method of Hiss.
There are other Gram-negative diplococci which are chromogenic,
and thus can readily be distinguished. The points of difference
between the meningococcus and the gonococcus are given on
p. 252. A Gram-positive diplococcus called the diplococcus
crassus is also of common occurrence; it is rather larger than
the diplococcus intracellularis, and especially in sub-cultures may
tend to assume staphylococcal forms.

Apart from the epidemic form of the disease, meningitis may
be produced by almost any of the organisms described in the
previous chapter, as associated with inflammatory conditions.
A considerable number of cases, especially in children, are due
to the pneumococcus. In many instances where no other lesions
are present the extension is by the Eustachian tube to the middle
ear. In other cases the path of infection is from some other
lesion by means of the blood stream. This organism also infects
the meninges not infrequently in lobar pneumonia, and in some

248 EPIDEMIC CEREBRO-SPINAL MENINGITIS

cases with head symptoms we have found it present where there
was merely a condition of congestion. The pneumobacillus also-
has been found in a few cases. Meningitis is not infrequently
produced by streptococci, especially when middle-ear disease is
present, less frequently by one of the staphylococci ; occasionally
more than ; one organism may be concerned. In meningitis
following influenza the influenza bacillus has been found in a
few instances, but sometimes the pneumococcus is the causal
agent. Cases of meningitis occur associated with organisms
which resemble the influenza bacillus morphologically and also
in presenting haemophilic culture reactions, but which possess
pathogenic properties for rabbits and guinea-pigs. These bacilli
frequently both in the cerebro-spinal fluid and in cultures show
a tendency to produce long filamentous forms and also under
both circumstances may show a beading of the protoplasm
which gives them a diphtheroid appearance. Gram-negative
anaerobic bacilli have also been found in cases of meningitis.
An invasion of the meninges by the anthrax bacillus occurs,
but is a rare condition ; it is attended by diffuse
haemorrhage in the subarachnoid space. In tubercular
meningitis the tubercle bacillus, of course, is present, especially
in the nodules along the sheaths of the vessels.

In conclusion, here it may be stated that mixed infections
may occur in meningitis. Thus the pneumococcus has been
found associated with the tubercle bacillus and also with the
diplococcus intracellularis.

Methods of Examination. — During life these involve the microscopic
investigation of the centrifuged lumbar puncture fluid and making
cultures therefrom. For the former, Gram-stained smears make the
recognition of the meningococcus relatively easy, and the presence of
Gram-negative cocci, especially within cells, is practically diagnostic of
a case of cerebro-spinal fever. Tubes of serum-agar, nasgar (pp. 42, 43),
or agar containing 25 per cent, of ascitic or ovarian fluid, may then be
inoculated. The difficult cases are those where no bacteria can be found
microscopically in the lumbar fluid. Here the character of the exudate
may give help. A predominance of polymorphonuclear cells is usually
manifest in meningococcic, pneumococcic, and influenzal cases, whereas in
tubercular meningitis the exudate is chiefly lymphocytic. In such
circumstances, besides other media, a tube of blood-smeared agar should
be inoculated in case the pneumococcus or the influenza bacillus is the
causal organism. To speak generally, if with a polymorphonuclear exu-
date no growth occurs in the media mentioned the case is most likely to be
due to the meningococcus. In tubercular cases it is sometimes impossible
to demonstrate the bacilli microscopically in the exudate, though on
careful search they may often be found.

CHAPTER IX.

GONORRHOEA AND SOFT SORE.

GONORRH(EA.

Introductory. — The micrococcus now known to be the cause
of gonorrhoea, and called the gonococcus, was first described
by Neisser, who in 1879 gave an account of its microscopical
characters as seen in the pus of gonorrhceal affections, both of
the urethra and of the conjunctiva. He considered that this
organism was peculiar to the disease, and that its characters
were distinctive. Later it was successfully isolated and cultivated
on solidified human serum by Bumm and others. Its characters
have since been minutely studied, and by inoculations of cultures
on the human subject its causal relationship to the disease has
been conclusively established.

The Gonococcus. — Microscopical Characters. — The organism
of gonorrhoea is a small micrococcus which usually is seen in the
diplococcus form, the adjacent margins of the two cocci being
flattened, or even slightly concave, so that between them there is
a small oval interval which does not stain. An appearance is
thus presented which has been compared to that of two beans
placed side by side (vide Fig. 72). When division takes place in
the two members of a diplococcus, a tetrad is formed, which,
however, soon separates into two sets of diplococci — that is to
say, arrangement as diplococci is much commoner than as tetrads.
Cocci in process of degeneration are seen as spherical elements
of varying size, some being considerably swollen.

These organisms are found in large numbers in the pus of
acute gonorrhoea, both in the male and female, and for the most
part are contained within the leucocytes. In the earliest stage,
when the secretion is glairy, a considerable number are lying
free, or are adhering to the surface of desquamated epithelial
cells, but when it becomes purulent the large proportion within
eucocytes is a very striking feature. In the leucocytes they He

249

250

GONORRHCEA AND SOFT SORE

within the protoplasm, especially superficially, and are often so
numerous that the leucocytes appear to be filled with them, and
their nuclei are obscured. As the disease becomes more chronic,

the gonococci gradually
become diminished in
number, though even in
long-standing cases they
may still be found in con-
siderable numbers. They
are also present in the
purulent secretion of
gonorrhoeal conjunctivitis,
also in various parts of
the female genital organs
when these parts are the
seat of true gonorrhreal
infection, and they have
been found in some cases
in the secondary infections
of the joints in the dis-
ease, as will be described
below.

Stained with fuchsin. x 1000. Staining. — The gono-

coccus stains readily and

deeply with a watery solution of any of the basic aniline dyes —
methylene-blue, fuchsin, etc. It is, however, easily decolorised,
and it completely loses the stain by Gram's method — an
important point in the microscopical examination.

Cultivation of the Gonococcus. — This is attended with
some difficulty, as the conditions of growth are somewhat
restricted. The most suitable media are "blood-agar" and -the
serum media already described for the purpose (pp. 42, 43). It
is advisable to inoculate the media within half an hour after
obtaining the material from the body, and place the tubes
at once in the incubator. Growth takes place best at the
temperature of the body, and ceases altogether at 25° C. Cultures
are obtained by taking some pus on the loop of the platinum
needle and inoculating one of the media mentioned by leaving
minute quantities here and there on the surface. The medium
may be used either as ordinary " sloped tubes " or as a thin layer
in a Petri's capsule. The young colonies are usually visible
within forty-eight hours, and often within twenty-four hours ;
it is important, however, to note that sometimes growth may
not appear till the fourth day. They appear around the

FIG. 72. — Portion of film of gonorrhoeal pus,
showing the characteristic arrangement of
the gonococci within leucocytes. See also
Plate I., Fig. 5.
d wit

CULTIVATION OF GONOCOCCUS

251

points of inoculation as small semi-transparent discs of rounded
shape. The colonies vary somewhat in size, and tend to

#• ft-

.*-v \

FIG. 73. — Colonies of gonococcus on serum agar ; (a) three days' growth ;

(6) and (c) five days' growth, x 9.
From photographs l.y Dr. W. B. M. Martin.

remain more or less separate. Later, the margin tends to be
undulated and the centre more opaque ; a radial marking may
be present (Fig. 73). The first cultures die out somewhat

• I u irk ly, but in sub-
cultures, kept at 37° C.,
the organism remains
alive for a considerable
time, sometimes three /g
weeks. After a week [M
more active foci of growth / |-
may appear in some of

tin- colonies in the form j

of heaped -up opaque \£g

points, thus giving an

appearance suggestive of

contamination. In the

early stage of the disease

the organism is present

in the male urethra in Fj(; 74._Gonococci; from a pure culture

practically pure condition, On blood agar of twenty-four hours'

and if the meatus of the growth. Some already are beginning to

urethra be sterilised by ^owth^swojlen appearance cominon in

washing with weak solu- Stained with carbol-thionin blue, x 1000.
t ion of corrosive sublimate

and then with absolute alcohol, and the material for inoculation
be expressed from the deeper part of the urethra, cultures may

• •ft en be obtained which are pure from the first. In culture the

252 GONORRHOEA AND SOFT SORE

organisms have similar microscopic characters to those described
(Fig. 74), but show a remarkable tendency to undergo degenera-
tion, becoming swollen and of various sizes, and staining very
irregularly. Degenerated forms are seen even on the second
day, whilst in a culture four or five days old comparatively few
normal cocci may be found. The less suitable the medium the
more rapidly does degeneration take place. When mixed with
other organisms the gonococcus may be separated by serum agar
plates (p. 43).

On ordinary agar and on glycerin-agar some growth may
take place when the reaction is just alkaline to litmus, but these
media are quite unsuitable for ordinary purposes. The organism
does not grow on gelatin, potato, etc.

Comparison with Meningococcus. — The morphological and
cultural characters of the gonococcus and meningococcus are
in many respect closely similar; the following points are of
importance in distinguishing them. The conditions of growth
of the gonococcus are more restricted than those of the meningo-
coccus. The gonococcus usually does not grow on the ordinary
agar media, whereas the meningococcus grows well, at least
after the first sub-culture. The colonies of the latter are more
opaque and have more regular margins than those of the
gonococcus. The meningococcus grows well in bouillon, pro-
ducing a general turbidity, whereas the gonococcus does not
grow ; even in serum bouillon the latter organism flourishes
feebly, and the scanty growth falls to the bottom leaving the
medium clear, whilst the meningococcus produces abundant
growth with general turbidity. The fermentative effects have
also been studied, and the chief results obtained are that glucose
is the only sugar usually employed which is fermented by the
gonococcus, whereas the meningococcus always ferments maltose
also. (For fermentative tests in the case of the gonococcus,
solid media, as introduced by v. Lingelsheim, should be used,
the serum medium of Martin, with litmus and the particular
sugar added, being specially suitable.)

Specific serum reactions — agglutination, opsonic action,
bactericidal action, and fixation of complement — have been
studied by Torrey, Elser and Huntoon, and Martin, in the case
of the two organisms. The general results obtained are that
each organism represents a somewhat heterogeneous group
showing considerable variations as regards the tests mentioned.
An anti-gonococcus serum produced by injecting one strain of
gonococcus has the maximum effect on that strain, whilst its
Action on other strains may be much feebler • so also with an

RELATIONS TO THE DISEASE 253

aiiti-meuingococciis serum in relation to different strains of
mt'iiingococci. An anti-gonococcus serum may have some effect,
though usually very slight, on a meningococcus and vice versa ;
this indicates that there are some receptors common to the two
organisms. These results are in many ways comparable with
the facts established with regard to members of the typhoid-coli
group, and are of course quite compatible with the gonococcus
and the meniugococcus being distinct species.

Relations to the Disease. — The gouococcus is invariably
present in the urethral discharge in gonorrhoea, and also in
other parts of the genital tract when these are the seat of true
gonorrhoeal infection. Its presence in these different positions
has been demonstrated not only by microscopic examination
but also by culture. From the description of the conditions of
growth in culture it will be seen that a life outside the body
in natural conditions is practically impossible — a statement
which corresponds with the clinical fact that the disease is
always transmitted directly by contagion. Inoculations of pure
cultures on the urethra of lower animals, and even of apes, is
followed by no effect, but a similar statement can be made with
regard to inoculations of gonorrhceal pus itself. In fact,
hitherto it has been found impossible to reproduce the disease by
any means in the lower animals. On a considerable number of
occasions inoculations of pure cultures have been made on the
human urethra, both in the male and female, and the disease,
with all its characteristic symptoms, has resulted. (Such
experiments have been performed independently by Bumm,
Steinschneider, Wertheim, and others.) The causal relationship
of the organism to the disease has therefore been completely
established, and it is interesting to note how the conditions of
growth and the pathogenic effects of the organism agree with
the characters of the natural disease.

I n tra peritoneal injections of pure cultures of the gonococcus in white
mice produce a localised peritonitis with a small amount of suppuration,
the organisms being found in large numbers in the leucocytes (Wertheim).
They also penetrate the peritoneal lining and are found in the sub-
eudothelial connective tissue, but they appear to have little power of
proliferation, they soon disappear, and the inflammatory condition does
not spread. Injection of pure cultures into the joints of rabbits, dogs,
and guinea-pigs causes an acute inflammation, which, however, soon
subsides, whilst the gonococci rapidly die out ; a practically similar
result is obtained when dead cultures are used. These experiments show
that while the organism, when present in large numbers, can produce a
certain amount of inflammatory change in these animals, it has little or
no power of multiplying and spreading in their tissues.

Toxin of the Gonococcus.— De Christmas has cultivated the gonococcus

254 GONORRHOEA AND SOFT SORE

in a mixture of one part of ascitic fluid and three parts of bouillon, and
has found that the fluid after twelve days' growth has toxic properties.
At this period all the organisms are dead ; such a fluid constitutes the
"toxin." The toxic substances are precipitated along witli the proteids
by alcohol, and the precipitate after being desiccated possesses the toxic
action. In young rabbits injection of the toxin produces suppuration ;
this is well seen in the anterior chamber of the eye, where hypopyon
results. The most interesting point, however, is with regard to its
action on mucous surfaces ; for, while in the case of animals it produces
no effect, its introduction into the human urethra causes acute catarrh,
attended with purulent discharge. He found that no tolerance to the
toxin resulted after five successive injections at intervals. In a more
recent publication he points out that the toxin on intracerebral injection
has marked effects ; he also claims to have produced an antitoxin.
He states that the toxin diffuses out in the culture medium, and does
not merely result from disintegration of the organisms. This has,
however, been called in question by other investigators.

Distribution in the Tissues. — The gonococcus having been
thus shown to be the direct cause of the disease, some additional
facts may be given regarding its presence both in the primary
and secondary lesions. In the human urethra the gonococci
penetrate the mucous membrane, passing chiefly between the
epithelial cells, causing a loosening and desquamation of many
of the latter and inflammatory reaction in the tissues below,
attended with great increase of secretion. There occurs also
a gradually increasing emigration of leucocytes, which take up
a large number of the organisms. The organisms also penetrate
the subjacent connective tissue and are especially found, along
with extensive leucocytic emigration, around the lacunae. Here
also many are contained within leucocytes. Even, however,
when the gonococci have disappeared from the urethral dis-
charge, they may still be present in the deeper part of the
mucous membrane of the urethra, possibly also in the prostate,
and may thus be capable of producing infection. The prostatic
secretion may sometimes be examined by making pressure on
the prostate from the rectum when the patient has almost
emptied his bladder, the secretion being afterwards discharged
along with the remaining urine. (Foulerton.) In acute
gonorrhoea there is often a considerable degree of inflammatory
affection of the prostate and vesiculae seminales, but whether
these conditions are always due to the presence of gonococci
in the affected parts we have not at present the data for deter-
mining. A similar statement also applies to the occurrence of
orchitis and also of cystitis in the early stage of gonorrhoea.
Gonococci have, however, been obtained in pure culture from
peri-urethral abscess and from epididymitis : it is likely that

DISTRIBUTION OF GONOCOCCUS 255

the latter condition, when occurring in gonorrhoea, is usually
due to the actual presence of gonococci. During the more
chronic stages other organisms may appear in the urethra, aid
in maintaining the irritation, and may produce some of the
secondary results. The bacillus coli, the pyogenic cocci, etc.,
are often present, and may extend along the urethra to the
bladder and set up cystitis, though in .this they may be aided
by the passage of a catheter. It may be mentioned here that
Wertheim cultivated the gonococcus from a case of chronic
gonorrhea of two years' standing, and by inoculation on the
human subject proved it to be still virulent.

In the disease in the female, gonococci are almost invariably
present in the urethra, the situation affected next in frequency
being the cervix uteri. They do not appear to infect the lining
epithelium of the vagina of the adult unless some other abnormal
condition be present, but they do so in the gonorrhceal vulvo-
vaginitis of young subjects. They have also been found in
suppurations in connection with Bartholini's glands, and some-
times produce an inflammatory condition of the mucous
membrane of the body of the uterus. They may also pas.-
along the Fallopian tubes and produce inflammation of the
mucous membrane there. From the pus in cases of pyosalpinx
they have been cultivated in a considerable number of cases.
According to the results of various observers they are present
in one out of four or five cases of this condition, usually un-
as>ociated with other organisms. Further, in a large proj)ortion
of the cases in which the gonococcus has not been found, no
organisms of any kind have been obtained from the pus, and
in these cases the gonococci may have been once present and
have subsequently died out. Lastly, they may pass to the
peritoneum and produce peritonitis, which is usually of a local
character. It is chiefly to the methods of culture supplied by
\Yertheim that we owe our extended knowledge of such
conditions.

In gonorrhoeal conjunctivitis the mode in which the gonococci
spread through the epithelium to the subjacent connective
tissue is closely analogous to what obtains in the case of the
urethra. Their relation to the leucocytes in the purulent
secretion is also the same. Microscopic examination of the
secretion alone in acute cases often gives positive evidence, and
I mre cultures may be readily obtained on blood-agar. As the
condition becomes more chronic the gonococci are less numerous
and a greater proportion of other organisms may be present.

Relations to Joint-Affections, etc. — The relations of the gono-

256 GONORRHCEA AND SOFT SORE

coccus to the sequelae of gonorrhoea form a subject of great
interest and importance, and the application of recent methods
of examination shows that the organism is much more frequently
present in such conditions than the earlier results indicated.
The following statements may be made with regard to them :
First, in a large number of cases of arthritis following gonorrhoea
pure cultures of the gonococcus may be obtained. A similar
statement applies to inflammation of the sheaths of tendons
following gonorrhoea. Secondly, in a considerable proportion of
cases no organisms have been found. It is, however, possible
that in many of these the gonococci may have been present
in the synovial membrane, as it has been observed that they
may be much more numerous in that situation than in the
fluid. Thirdly, in some cases, especially in those associated
with extensive suppuration, occasionally of a pya3mic nature,
various pyogenic cocci have been found to be present. In the
instances in which the gonococcus has been found in the joints,
the fluid present has usually been described as being of a
whitish yellow tint, somewhat turbid, and containing shreds
of fibrin-like material, sometimes purulent in appearance. In
one case Bordoni-Uffreduzzi cultivated the gonococcus from a
joint - affection, and afterwards produced gonorrhoea in the
human subject by inoculating with the cultures obtained. In
another case, in which pleurisy was present along with arthritis,
the gonococcus was cultivated from the fluid in the pleural
cavity. The existence of a gonorrhoeal endocarditis has been
established by recent observations. Cases apparently of this
nature occurring in the course of gonorrhoea had been previously
described, but the complete bacteriological test has now been
satisfied in several instances. In one case Lenhartz produced
gonorrhoea in the human subject by inoculation with the
organisms obtained from the vegetations. That a true
gonorrhoeal septicaemia may occur has also been established,
cultures of the gonococcus having been obtained from the
blood during life on more than one occasion (Thayer and
Blumer, Thayer and Lazear, Ahmann).

Methods of Diagnosis. — For microscopical examination, dried
films of the suspected pus, etc., may be stained by any of the
simple solutions of the basic aniline stains. We prefer methy-
lene- or thionin-blue, as they do not overstain, and the films do
not need to be decolorised. Staining for one minute is sufficient.
It is also advisable to stain by Gram's method, and it is a good
plan to put at one margin of the cover-glass a small quantity of
culture of staphylococcus aureus if available, in order to have

SOFT SOUK 257

a standard by which to be certain that the supposed gonococci
are really decolorised. Regarding the value of microscopic
examination alone, we may say that the presence of a large
number of micrococci in a urethral discharge having the
characters, position, and staining reactions described above,
is practically conclusive that the case is one of gonorrhea.
There is no other condition in which the sum-total of the
microscopical characters is present. We consider that it is
sufficient for purposes of clinical diagnosis, and therefore
of great value ; in the acute stage a diagnosis can thus be
made earlier than by any other method. The mistake of
confusing gonorrhoea with such conditions as a urethral chancre
with urethritis, will also be avoided. Even in chronic cases
the typical picture is often well maintained, and microscopic
examination alone may give a definite positive result. When
other organisms are present, and especially when the gonococci
are few in number, it is difficult, and in some cases impossible,
to give a definite opinion, as a few gonococci mixed with other
organisms cannot be recognised with certainty. This is often
the condition in chronic gonorrhoea in the female. Microscopic
examination, therefore, though often giving positive results,
will sometimes be inconclusive. As regards lesions in other
parts of the body, microsocopic examination alone is quite
insufficient ; it is practically impossible, for example, to
distinguish by this means the gonococcus from the diplococcus
intracellularis of meningitis. Cultures alone supply the test,
and the points above detailed are to be attended to.

SOFT SOKK.

Tin; bacillus of soft sore was first described by Ducrey in
1889, who found it in the purulent discharge from the ulcerated
surface; and later, in 1892, Unna described its appearance and
distribution as seen in sections through the sores. The state-
ments of these observers regarding the presence and characters
of this organism have been fully confirmed by other observers.

Microscopical Characters. — The organism occurs in the form
of minute oval rods measuring about 1*5 ft in length, and *5 ft
in thickness (Fig. 75). It is found mixed with other organisms
in the purulent discharge from the surface, and is chiefiy arranged
in small groups or in short chains. When studied in sections
through the ulcer, it is found in the superficial part of the floor,
but more deeply situated than other organisms, and may be
present in a state of purity amongst the leucocytic infiltration.

258

GONORRHOEA AND SOFT SORE

In this position it is usually arranged in chains, which may be
of considerable length, and w^hich are often seen lying in parallel
rows between the cells. The bacilli chiefly occur in the free
condition, but occasionally a few may be contained within
leucocytes.

There is no doubt that in many cases the organism is present
in the buboes in a state of purity ; it has been found there by
microscopic examination, and cultures have also been obtained

from this source. The
negative results of some
observers are probably
due to the organism
having died off. On the
whole the evidence goes

§>* w fJJff £F\ to show tliat tlie ordinary
(flHllH^' KB J bubo associated with soft
*• i^JjJpBJr Kl fm sore is to be regarded as
^& ^jlgiKp ^ another lesion produced

*^ * by Ducrey's bacillus.

Sometimes the ordinary
pyogenic organisms be-
come superadded.

This bacillus takes up
the basic aniline stains
FIG. 75.— Film preparation of pus from soft fairly readily, but loses
chancre, showing Ducrey's bacillus, chiefly t^ Pnlnnr vprv ranirllv
arranged in pairs. Stained with carbol- l _ve!7 iaPlcll>

fuchsin and slightly decolorised, x 1500. when a decolorising agent

is applied. Accordingly,

in film preparations when dehydration is not required, it can
be readily stained by most of the ordinary combinations, though
Loffler's or Kiihne's methylene-blue solutions are preferable, as
they do not overstain. In sections, however, great care must
be taken in the process of dehydration, and the aniline-oil
method (vide p. 100) should be used for this purpose, as alcohol
decolorises the organism very readily. A little of the methylene-
blue or other stain may be with advantage added to the aniline-
oil used for dehydrating.

Cultivation. — Although for a long period of time attempts
to obtain cultures were unsuccessful, success has been attained
within recent years. Bezancon, Griffon, and Le Sourd obtained
pure cultures in four cases, the medium used being a mixture of
rabbit's blood and agar, in the proportion of one part of the
former to two of the latter. The blood is added to the agar in
the melted condition at 45° C., and the tubes are then sloped.

SOFT SORE

259

Davis confirms these results, and finds that another good medium
is freshly-drawn human blood distributed in small tubes; this
method is specially suitable, as the blood inhibits the growth
of various extraneous organisms. On the solid medium (blood-
agar) the growth appears in the form of small round globules,
which attain their complete development in forty-eight hours,
having then a diameter of 1 to 2 mm. ; the colonies do not
become confluent. Microscopic examination of these colonies,
which are dissociated
with some difficulty,
shows appearances simi-
lar to those observed
when the organism is
in the tissues (Fig. 76),
but occasionally long
undivided filaments are
observed which Davis
regards as degenerative
forms. Within a com-
paratively short period
cultures undergo marked
degenerative changes,
and great irregularities
of form arid shape are to
be found. It would ap-
pear that a comparatively
large amount of blood is

necessary for the growth of this organism, and even sub-cultures
on the ordinary media, including blood -serum media, give
negative results. Inoculation of the ordinary laboratory animals
is not attended by any result, but it has been found that some
monkeys are susceptible, small ulcerations being produced by
superficial inoculation, and in these the organism can be demon-
strated. Tomasczewski cultivated the organism for several
generations, and reproduced the disease by inoculation of the
human subject. The causal relationship of this bacillus must
therefore be considered as completely established, and the con-
ditions under which it grows show it to be a strict parasite — a
fact which is in conformity with the known facts as to the
transmission of the disease.

1 We are indebted to Dr. Davis for the use of Figs. 75 and 76,

j&.-'-v

^x$

FIG. 76. — Ducrey's bacillus from a 24-hour
culture in blood-bouillon, x 1500.1

CHAPTER X.

TUBERCULOSIS.

THE cause of tubercle was proved by Koch in 1882 to be the
organism now universally known as the tubercle bacillus.
Probably no other single discovery has had a more important
effect on medical science and pathology than this. It has not
only shown what is the real cause of the disease, but has also
supplied infallible methods for determining what are tubercular
lesions and what are not, and has also given the means of
studying the modes and paths of infection. A definite answer
has in this way been supplied to many questions which were
previously the subject of endless discussion.

Historical. — By the work of Armarmi and of Cohnheim and Salomonsen
(1870-80) it had been demonstrated that tubercle was an infective disease.
The latter observers found on inoculation of the anterior chamber of the
eye of rabbits with tubercular material, that in many cases the results of
irritation soon disappeared, but that after a period of incubation, usually
about twenty-five days, small tubercular nodules appeared in the iris ;
afterwards the disease gradually spread, leading to a tubercular disorgan-
isation of the globe of the eye. Later still, the lymphatic glands became
involved, and finally the animal died of acute tuberculosis. The question
remained as to the nature of the virus, the specific character of which
was thus established, and this question was answered by the work of
Koch.

The announcement of the discovery of the tubercle bacillus was made
by Koch in March 1882, and a full account of his researches appeared in
1884 (Mirth, a. d. K. Gfsndhtsamte., Berlin). Koch's work on this subject
will remain as a classical masterpiece of bacteriological research, both on
account of the great difficulties which he successfully overcame and the
completeness with which he demonstrated the relations of the organism
to the disease. The two chief difficulties Avere, first, the demonstration
of the bacilli in the tissues, and, secondly, the cultivation of the organism
outside the body. For, with regard to the first, the tubercle bacillus
cannot be demonstrated by a simple watery solution of a basic aniline
dye, and it was only after prolonged staining for twenty-four hours, with
a solution of methylene-blue with caustic potash added, that he was
able to reveal the presence of the organism. Then, in the second place,
all attempts to cultivate it on the ordinary media failed, and he only

260

TUBERCULOSIS IN ANIMALS 261

succeeded in obtaining growth on solidified blood serum, the method of
preparing which he himself devised, inoculations being made on this
medium from the organs of animals artificially rendered tubercular.
The fact that growth did not appear till the tenth day at the earliest,
might easily have led to the hasty conclusion that no growth took place.
All difficulties were, however, successfully overcome. He cultivated the
organism by the above method from a great variety of sources, and by
a large series of inoculation experiments on various animals, performed
by different methods, he conclusively proved that bacilli from these
different sources produced the same tubercular lesions and were really of
the same species. His work was the means of showing conclusively that
such conditions as lupus, "white swelling" of joints, scrofulous disease
of glands, etc., are really tubercular in nature.

Tuberculosis in Animals. — Tuberculosis is not only the most
widely spread of all diseases affecting the human subject, and
produces a mortality greater than any other, but there is probably
no other disease which affects the domestic animals so widely.
We need not here describe in detail the various tubercular lesions
in the human subject, but some facts regarding the disease in
the lower animals may be given, as this subject is of great
importance in relation to the infection of the human subject.

Amongst the domestic animals the disease is commonest in cattle
(bovine tuberculosis), in which animals the lesions are very various, both
in character and distribution. In most cases the lungs are affected, and
contain numerous rounded nodules, many being of considerable size ;
these may be softened in • the centre, but are usually of pretty firm
consistence and may be calcified. There may be in addition caseous
pneumonia, and also small tubercular granulations. Along with these
changes in the lungs, the pleurae are also often affected, and show
numerous nodules, some of which may be of large size, firm and pedun-
culated, the condition being known in Germany as Perlsucht, in France
as pommeliere. Lesions similar to the last may be chiefly confined to
the peritoneum and pleune. In other cases, again, the abdominal organs
are principally involved. The udder becomes affected in a certain pro-
portion of cases of tuberculosis in cows — in 3 per cent, according to Bang
— but primary affection of this gland is very rare. Tuberculosis is also
a comparatively common disease in pigs, in which animals it in many
cases affects the abdominal organs, in other cases produces a sort of
osseous pneumonia, and sometimes is met with as a chronic disease of
the lymphatic glands, the so-called "scrofula" of pigs. Tubercular
lesions in the muscles are less rare in pigs than in most other animals.
In the horse the abdominal organs are usually the primary seat of the
disease, the spleen being often enormously enlarged and crowded with
nodules of various shapes and sizes ; sometimes, however, the primary
lesions are pulmonary. In sheep and goats tuberculosis is of rare
occurrence, especially in the former animals. It may occur spontaneously
in dogs, cats, and in the large carnivora. It is also sometimes met with
in monkeys in confinement, and leads to a very rapid and widespread
affection in these animals, the nodules having a special tendency to
soften and break down into a pus-like fluid.

262 TUBERCULOSIS

Tuberculosis in fowls (avian tuberculosis) is a common and very
infectious disease, nearly all the birds in a poultry-yard being sometimes
affected.

From these statements it will be seen that the disease in
animals presents great variations in character, and may differ in
many respects from that met with in the human subject. The
relation of the different forms of tuberculosis is discussed below.
Tubercle Bacillus— Microscopical Characters. — Tubercle
bacilli are minute rods which usually measure 2 -5 to 3 -5 //, in
length, and '3 /x in thickness, i.e. in proportion to their length
they are comparatively thin organisms (Figs. 77 and 78). Some-
times, however, longer
forms, up to 5 /x or more
in length, are met with,
both in cultures and in
the tissues. They are
straight or slightly curved,
and are of uniform thick-
ness, or may show slight
swelling at their extremi-
ties; When stained they
appear uniformly colour-
ed, or may present small
uncoloured spots along
their course, with darkly
stained parts between. In

such a minute organism
FIG. It. — Tubercle bacilli, from a pure ... -, J-&, u

culture on glycerin agar. ^ 1S extremely difficult

Stained with carbol-fuchsin. x 1000. to determine the exact

nature of the unstained

points. Accordingly, we find that some observers consider
these to be spores, while others find that it is impossible to
stain them by any means whatever, and consider that they
are really of the nature of vacuoles. Against their being
spores is also the fact that many occur in one bacillus. Others
again hold that some of the condensed and highly stained
particles are spores. It is impossible to speak definitely on the
question at present. We can only say that the younger bacilli
stain uniformly, and that in the older forms inequality in stain-
ing is met with ; this latter condition is, however, not found to
be associated with greater powers of resistance.

The bacilli in the tissues occur scattered irregularly or in
little masses. They are usually single, or two are attached end
to end and often form in such a case an obtuse angle. True

THE TUBERCLE BACILLUS 263

chains~are not formed, but occasionally short filaments are met
with. In cultures the bacilli form masses in which the rods are
closely applied to one another and arranged in a more or less
parallel manner. Tubercle bacilli are quite devoid of motility.
Aberrant Forms. — Though such are the characters of the
organism as usually met with, other appearances are sometimes
found. In old cultures, for example, very much larger elements

FIG. 78. — Tubercle bacilli iu phthisical sputum ; they are longer than

is often the case. See also Plate II., Fig. 7.

Film preparation, stained with carbol-fuchsin and inethylene-blue.
xlOOO.

may occur. These may be in the form of long filaments, some-
times swollen or clubbed at their extremities, may be irregularly
beaded, and may even show the appearance of branching. Such
forms have been studied by Metchnikoff, Maffucci, Klein, and
others. Their significance has been variously interpreted, for
while some look upon them as degenerated or involution forms,
others regard them as indicating a special phase in the life
history of the organism, allying it with the higher bacteria.
Recent observations, however, go to establish the latter view,
and this is now generally accepted by authorities. It has also

264 TUBERCULOSIS

been found that under certain circumstances tubercle bacilli in
the tissues produce a radiating structure closely similar to that
of the actinomyces. This was found by Babes and also by
Lubarsch to be the case when the bacilli were injected under
the dura mater and directly into certain solid organs, such as
the kidneys in the rabbit. Club-like structures may be present
at the periphery ; these are usually not acid-fast, but they retain
the stain in the Weigert-Gram method. Similar results obtained
with other acid-fast bacilli will be mentioned below, and these
organisms would appear to form a group closely allied to the
streptothricese, the bacillary parasitic form being one stage of
the life history of the organism. This group is often spoken of
as the mycobacteria.

Staining Reactions. — The tubercle bacillus takes up the
ordinary stains very slowly and faintly, and for successful stain-
ing one of the most powerful solutions ought to be employed, e.g.
gentian-violet or fuchsin, along with aniline-oil water or solution
of carbolic acid. Further, such staining solutions require to be
applied for a long time, or the staining must be accelerated by
heat, the solution being warmed till steam arises and the
specimen allowed to remain in the hot stain for two or three
minutes. One of the best and most convenient methods is the
Ziehl-Neelsen method (see p. 108). -The bacilli present this
further peculiarity, however, that after staining has taken place
they resist decolorising by solutions which readily remove the
colour from the tissues and from other organisms which may be
present. Such decolorising agents are sulphuric or nitric acid
in 20 per cent, solution. Preparations can thus be obtained in
which the tubercle bacilli alone are coloured by the stain first
used, and the tissues can then be coloured by a contrast stain.
Within recent years certain other bacilli have been discovered
which present the same staining reactions as tubercle bacilli ;
they are therefore called " acid-fast " (vide infra). The spores
of many bacilli become decolorised more readily than tubercle
bacilli, though some retain the colour with equal tenacity.

Bullocli and Macleod, by treating tubercle bacilli with hot alcohol
and ether, extracted a wax which gave the characteristic staining
reactions of the bacilli themselves. The remains of the bacilli, further,
when extracted with caustic potash, yielded a body which was probably
a chitin, and which was acid-fast when stained for twenty-four hours
with carbol-fuchsin.

It had long been recognised that it might not be possible to
detect by microscopic methods tubercle bacilli in old tubercular

CULTIVATION OF TUBERCLE BACILLUS 265

lesions, and yet the material from such was virulent on
inoculation. This was supposed to be due either to the
paucity of the bacilli or possibly to the presence of spores.
Recently observations have been brought forward by Much
which may throw important light on this subject. Briefly put,
his conclusions are that the tubercle virus exists in three forms
— (a) the ordinary bacillary form stainable by the Ziehl method ;
(6) a fine bacillary form which is not acid-fast, often showing
granules in its interior ; and (c) free granules which also fail to
stain with the Ziehl method. The two last forms can be stained
by Gram's method when the stain is applied for a long time.
Much gives three modifications of Gram's method, the following
being one which has been found by others to be specially
suitable : —

Methyl violet B. N. 10 c. c. of a saturated alcoholic solution in 100 c.c. of
a 2 per cent, watery solution of carbolic acid ; stain by boiling over the
name for a few minutes or at 37° C. for 24-48 hours, then treat with Lugoe's
iodine for 1-5 minutes, 5 per cent, hydrochloric acid for one minute, 3
per cent, hydrochloric acid 10 seconds, and complete the decolorisation
with ;i mixture of acetone and alcohol in equal parts.

Much claims that by such a method bacilli and granules can
be found in tubercular lesions when the Ziehl method gives
a negative result. He also found that, w^ien bacilli from a
culture were added to sterilised milk and incubated, the acid-fast
forms disappeared whilst those stainable with Gram's method
remained; and that when this had occurred the milk when
injected into an animal produced tuberculosis in which acid-fast
bacilli were demonstrable. His statements have received con-
firmation by other observers, e.g. Wirths and Treuholtz, but as
yet it is not possible to give a definite pronouncement on the
whole subject. If the bacillus can pass into a form not demon-
strable by the method usually employed but still virulent, we
have manifestly to deal with a fact of the highest importance.
There seems to be no doubt that in certain conditions more
tubercle bacilli can be demonstrated in the tissues by Much's
method than by the ordinary carbol-f uchsin method.

Cultivation. — The medium first used by Koch was inspissated
blood serum (vide p. 40). If inoculations are made on this
medium with tubercular material free from other organisms,
there appear in from ten to fourteen days minute points of growth
of dull whitish colour, rather irregular, and slightly raised above
the surface (it is advisable to plant on the medium an actual
piece of the tubercular tissue and to fix it in a wound of the

266

TUBERCULOSIS

surface of the serum). Koch compared the appearance of these
to that of small dry scales. In such cultures the growths usually
reach only a comparatively small size and remain separate, be-
coming confluent only when many occur close together. In sub-
cultures, however, growth is more luxuriant and may come to
form a dull wrinkled film of whitish colour, which may cover

the greater part of the
surface of the serum and
at the bottom of the tube
may grow over the sur-
face of the condensation
water on to the glass
(Fig. 79, A). The growth
is always of a dull ap-
pearance, and has a con-
siderable degree of con-
sistence, so that it is
difficult to dissociate a
portion thoroughly in a
drop of water. In older
cultures the growth may
acquire a slightly brown-
ish or buff colour. When
the small colonies are
examined under a low
power of the microscope,
they are seen to be ex-
tending at the periphery
in the form of wavy or
sinuous streaks which
radiate outward, and
which have been com-
pared to the flourishes of
a pen. The central part
shows similar markings
closely interwoven. These
streaks are composed of masses of the bacilli arranged in a more
or less parallel manner.

On glycerin agar, which was first introduced by Nocard and
Roux as a medium for the culture of the tubercle bacillus,
growth takes place in sub-cultures at an earlier date and pro-
gresses more rapidly than on serum, but this medium is not
suitable for obtaining cultures from the tissues, inoculations
with tubercular material usually yielding a negative result.

FIG. 79. — Cultures of tubercule bacilli on
glycerin agar.

A and B. Mammalian tubercle bacilli ; A is an

old culture, B one of a few weeks' growth.

C. Avian tubercle bacilli. The growth is whiter

and smoother on the surface than the others.

CULTIVATION OF TUBERCLE BACILLUS 267

The growth has practically the same characters as on serum, but
is more luxuriant. In glycerin broth, especially when the layer
is not deep, tubercle bacilli grow readily in the form of little
white masses, which fall to the bottom and form a powdery layer.
If, however, the growth be started on the surface, it spreads
superficially as a dull whitish wrinkled pellicle which may reach
the walls of the flask ; this mode of growth is specially suitable
for the production of tuberculin (vide infra). The culture has
a peculiar fruity and not unpleasant odour. On ordinary agar
and on gelatin media no growth takes place. The use of animal
tissues in glycerine bouillon as a medium for the growth of the
tubercle bacillus has been recently introduced by Frugoni, and
is one which gives excellent results. He recommends that small
wedges of rabbit's lung should be sterilised in the autoclave, and
placed in tubes of glycerine bouillon in such a way that their
surface is kept moist by the medium, without the fragments
being submerged. The growth is probably more rapid and
luxuriant than in any other method.

Use of Egg Media. — Within recent years media containing
either the yolk or both the yolk and the white of egg have been
used for the culture of the tubercle bacillus by Dorset and others.
The following is Dorset's method : The contents of four eggs
are well beat, 25 c.c. of water are added and thoroughly mixed,
the mixture being passed through muslin to remove air bells.
The fluid is then filled into tubes, and these are heated for four
hours in the sloped position at 70° C. Before the inoculation of
a tube, two drops of sterilised water are placed on the surface.
The inoculation material is well rubbed over the surface of the
medium, the tubes are sealed with a few drops of paraffin on the
top of the plug and are incubated in the sloped position.
Vigorous growth takes place on such media, having, generally
speaking, the naked eye characters seen in blood serum cultures.

It was at one time believed that the tubercle bacillus would only grow
on media containing animal fluids, but of late years it has been found
that growth takes place also on a purely vegetable medium, as was first
shown by Pawlowsky in the case of potatoes. Sander found that the
bacillus grew readily on potato, carrot, macaroni, and on infusion of
these substances, especially when glycerin was added. He also found
that cultures from tubercular lesions could be obtained on glycerin potato
(p. 46).

The optimum tcmi>craturc tor growth is 37° to 38° C.
Growth ceases about 42° and usually below 28°, but on long-
continued cultivation outside the body and in special circum-
stances, growth may take place at a lower temperature, e.g.

268 TUBERCULOSIS

Sander found that growth took place in glycerin-potato broth
even at 22° to 23° C.

Powers of Resistance. — Tubercle bacilli have considerable
powers of resistance to external influences, and can retain their
vitality for a long time outside the body in various conditions ;
in fact, in this respect they may be said to occupy an inter-
mediate position between spores and spore-free bacilli. Dried
phthisical sputum has been found to contain still virulent bacilli
(or their spores) after two months, and similar results are obtained
when the bacilli are kept in distilled water for several weeks.
So also they resist for a long time the action of putrefaction,
which is rapidly fatal to many pathogenic organisms. Sputum
has been found to contain living tubercle bacilli even after being
allowed to putrefy for several weeks (Fraenkel, Baumgarten), and
the bacilli have been found to be alive in tubercular organs which
have been buried in the ground for a similar period. They are
not killed by being exposed to the action of the gastric juice for
six hours, or to a temperature of - 3° C. for three hours, even
when this is repeated several times. It has been found that
when completely dried they can resist a temperature of 100° C.
for an hour, but, on the other hand, exposure in the moist
condition to 70° C. for the same time is usually fatal. It may
be stated that raising the temperature to 100° C. kills the bacilli
in fluids and in tissues, but in the case of large masses of tissue
care must be taken that this temperature is reached throughout.
They are killed in less than a minute by exposure to 5 per cent,
carbolic acid, and both Koch and Straus found that they are
rapidly killed by being exposed to the action of direct sunlight.

Action on the Tissues. — The local lesion produced by the
tubercle bacillus is the well-known tubercle nodule, the
structure of which varies in different situations and according to
the intensity of the action of the bacilli. After the bacilli gain
entrance to a connective tissue such as that of the iris, their
first action appears to be on the connective-tissue cells, which
become somewhat swollen and undergo mitotic division, the
resulting cells being distinguishable by their large size and pale
nuclei — the so-called epithelioid cells. These prolif erative changes
may be well seen on the fifth day after inoculation or even
earlier. A small focus of proliferated cells is thus formed in the
neighbourhood of the bacilli, and about the same time numbers
of leucocytes — chiefly lymphocytes — begin to appear at the
periphery and gradually become more numerous. Soon, however,
the action of the bacilli as cell-poisons comes into prominence.
The epithelioid cells become swollen and somewhat hyaline, their

ACTION ON THE TISSUES 269

outlines become indistinct, whilst their nucleus stains faintly,
and ultimately loses the power of staining. The cells in the
centre, thus altered, gradually become fused into a homogeneous
substance, and this afterwards becomes somewhat granular in
appearance. If the central necrosis does not take place quickly,
then giant-cell formation may occur in the centre of the follicle,
this constituting one of the characteristic features of the tuber-
cular lesion ; or after the occurrence of caseation giant-cells may
be formed in the cellular tissue around. The centre of a giant-
cell often shows signs of degeneration, such as hyaline change
and vacuolation, or it may be more granular than the rest of
the cell. The exact mode of formation of a tubercle follicle
varies, however, in different tissues.

Though there has been a considerable amount of discussion
as to the mode of origin of the giant-cells, we think there can
be little doubt that in most cases they result from enlargement
of single epithelioid cells, the nucleus of which undergoes pro-
liferation without the protoplasm dividing. These epithelioid
cells may sometimes be the lining cells of capillaries. Some
consider that the giant-cells result from a fusion of the epithelioid
cells ; but, though there are occasionally appearances which
indicate such a mode of formation, it cannot be regarded as of
common occurrence. In some cases of acute tuberculosis, when
the bacilli become lodged in a capillary, the endothelial cells of
its wall may proliferate, and thus a ring of nuclei may be seen
round a small central thrombus. Such an occurrence gives rise
to an appearance closely resembling a typical giant-cell.
There can be no doubt that the cell necrosis and subsequent
caseation depend upon the products of the bacilli, and are not
due to the fact that the tubercle nodule is non-vascular. This
non-vascularity itself is to be explained by the circumstance
that young capillaries cannot grow into a part where tubercle
bacilli are active, and that the already existing capillaries become
thrombosed, owing to the action of the bacillary products on
their walls, and ultimately disappear. At the periphery of
tubercular lesions there may be considerable vascularity and new
formation of capillaries.

The general symptoms of tuberculosis — pyrexia, perspiration,
wasting, etc., are to be ascribed to the absorption and distribution
throughout the system of the toxic products of the bacilli ; in
the case of phthisical cavities and like conditions where other
bacteria are present, the toxins of the latter also play an im-
portant part. The occurrence of waxy change in the organs is
believed by some to be chiefly due to the products of other,

270

TUBERCULOSIS

especially pyogenic, organisms, secondarily present in the tuber-
cular lesions. This matter, however, requires further elucidation.
Presence and Distribution- of the Bacilli. — A few facts may
be stated regarding the presence of bacilli, and the numbers in
which they are likely to be found in tubercular lesions.
They are usually very few in number in chronic lesions,
whether these are tubercle nodules with much connective tissue

FIG. 80. — Tubercle bacilli in section of human lung in acute phthisis.
The bacilli are seen lying singly, and also in large masses to left of
field. The pale background is formed by caseous material.

Stained with carbol-fuchsiu and Bismarck-brown, x 1000.

formation or old caseous collections. In caseous material one
can sometimes see a few bacilli faintly stained, along with very
minute unequally stained granular points, some of which may
possibly be spores of the bacilli. Whether they are spores or
not, the important fact has been established, that tubercular
material in which no bacilli can be found microscopically, may
be proved, on experimental inoculation into -animals, to be still
virulent. In such cases the bacilli may be present in numbers so
small as to escape observation, or it may be that their spores only

ACTION ON THE TISSUES

271

*ire present. In subacute lesions, with well-formed tubercle
follicles and little caseation, the bacilli are generally scanty.
They are most numerous in acute lesions, especially where
caseation is rapidly spreading, for example, in such conditions as
case. HI* catarrlial pneumonia (Fig. 80), acute tuberculosis of the
spleen in children, which is often attended with a good deal of
rapid casemis change, etc. ; in such conditions they often form

*"•

Fi<;. 81.— Tubercle bacilli in giant-cells, showing the radiate
arrangement at the periphery of the cells. Section of tubercular
udder of cow.

Stained with carbol-fuchsin and Rismarrk-brown. x 1000.

large masses which are easily seen under a low power of the
microscope. In acute miliary tuberculosis a few bacilli can
generally be found in the centre of the follicles ; but here they
are often much more scanty than one would expect. The
tubercle bacillus is one which not only has comparatively slow
growth, but retains its form and staining power for a much
longer period than most organisms. As a rule the bacilli are
-extra-cellular in jK)sition. Occasionally they occur within the
iriant -cells, in which they may lie arranged in a somewhat radiate

272

TUBERCULOSIS

manner at the periphery, occasionally also in epithelioid cells
and in leucocytes.

The above statements, however, apply only to tuberculosis
in the human subject, and even in this case there are exceptions.
In the ox, on the other hand, the presence of tubercle bacilli
within giant-cells is a very common occurrence ; and it is
also common to find them in considerable numbers scattered
irregularly throughout the cellular connective tissue of the lesions,.
even when there is little or no caseation present (Fig. 81).

In tuberculosis in the horse and in avian tuberculosis the
numbers of bacilli may be enormous, even in lesions which are

not specially acute ; and
considerable variation
both in their number and

4N91 ^ in their site is met with

in tuberculosis of other
animals.

In discharges from
tubercular lesions which
are breaking down, tu-
bercle bacilli are usually
to be found. In the
sputum of phthisical
patients their presence
can be demonstrated al-
most invariably at some
period, and sometimes

FIG. 82.— Tubercle bacilli in urine ; showing their numbers are very

one of the characteristic clumps, in which j (for method of stain_

tney oiten occur. . • r\**\ o i

Stained with carbol-fnchsin and methylene- ing, see p. 10/). Several

blue. xlOOO.

examnatons may,
ever, require to be made ;

this should always be done before any conclusion as to the non-
tubercular nature of a case is come to. In cases of genito-urinary
tuberculosis they are usually present in the urine ; but as they
are much diluted it is difficult to find them unless a deposit is
obtained by means of the centrifuge. This deposit is examined
in the same way as the sputum. The bacilli often occur in little
clumps, as shown in Fig. 82. In tubercular ulceration of the
intestine their presence in the faeces may be demonstrated, as
was first shown by Koch ; but in this case their discovery is
usually of little importance, as the intestinal lesions, as a rule,
occur only in advanced stages when diagnosis is no longer a
matter of doubt.

EXPERIMENTAL INOCULATION 273

Experimental Inoculation. — Tuberculosis can be artificially
produced in animals by infection in a great many different ways
— by injection of the bacilli into the subcutaneous tissue, into
the peritoneum, into the anterior chamber of the eye, into the
veins; by feeding the animals with the bacilli; and, lastly, by
making them inhale the bacilli suspended in the air.

The exact result, of course, varies in different animals and
according to the method of inoculation, but we may state
generally that when introduced into the tissues of a susceptible
animal, the bacilli produce locally the lesions above described,
terminating in caseation ; that there occurs a tubercular affection
of the neighbouring lymphatic glands, and that lastly there
may be a rapid extension of the bacilli to other organs by the
blood stream and the production of general tuberculosis. Of
the animals generally used for the purpose, the guinea-pig is
most susceptible.

When a guinea-pig is inoculated subcutaneously with tubercle
bacilli from a culture, or with material containing them, such as
phthisical sputum, a local swelling gradually forms which is
usually well marked about the tenth day. This swelling becomes
softened and caseous, and may break down, leading to the
formation of an irregularly ulcerated area with caseous lining.
The lymphatic glands in relation to the parts can generally be
found to be enlarged and of somewhat firm consistence, about
the end of the second or third week. Later, in them also caseous
change occurs, and a similar condition may spread to other
groups of glands in turn, passing also to those on the other side
of the body. During the occurrence of these changes, the animal
loses weight, gradually becomes .cachectic, and ultimately dies,
sometimes within six weeks, sometimes not for two or three
months. Post mortem, in addition to the local and glandular
changes, an acute tuberculosis is usually present, the spleen
being specially affected. This organ is swollen, and is studded
throughout by numerous tubercle nodules, which may be minute
and grey, or larger and of a yellowish tint. If death has been
lono; delayed, calcification may have occurred in some of the
ii Mlules. Tubercle nodules, though rather less numerous, are
also present in the liver and in the lungs, the nodules in the
latti-r organs being usually of smaller size though occasionally in
large numbers. The extent of the general infection varies;
-"inetimes the chronic glandular changes constitute the out-
standing feature. Statements as to differences in the pathogenic
effects of bacilli from human and bovine sources will be found
below (p. 274).
18

274 TUBERCULOSIS

Varieties of Tuberculosis. 1. Human and Bovine Tuberculosis.
—Up till recent years it was generally accepted that all
mammalian tuberculosis was due to the same organism, and,
in particular, that tuberculosis could be transmitted from the
ox to the human subject. The matter became one of special
interest owing to Koch's address at the Tuberculosis Congress
in 1901, in which he stated his conclusion that human and
bovine tuberculosis are practically distinct, and that if a
susceptibility of the human subject to the latter really exists,
infection is of very rare occurrence, — so rare that it is not
necessary to take any measures against it. Previously to this,
Theobald Smith had pointed out differences between mammalian
and bovine tubercle bacilli, the most striking being that the
latter possess a much higher virulence to the guinea-pig, rabbit,
and other animals, and in particular that human tubercle bacilli,
on inoculation into oxen, produce either no disease or only local
lesions without any dissemination. Koch's conclusions were
based chiefly on the result of his inoculations of the bovine
species with human tubercle bacilli, the result being confirmatory
of Smith's, and also on the supposition that infection of the
human subject through the intestine is of very rare occurrence.

Since the time of Koch's communication an enormous amount
of work has been done on this subject, and Commissions of
inquiry have been appointed in various countries. We may
summarise the chief facts which have been established.
Practically all observers are agreed that there are two chief
types of tubercle bacilli, which differ both in their cultural
characters and in their virulence — a bovine type and a human
type. The bacilli of the bovine type, when cultivated, are shorter
and thicker and more regular in size ; whilst their growth on
various culture media is scantier than that of the human type.
From the latter character the British Royal Commission have
applied the term dysgonic to the bovine and eugonic to the
human type. As already stated, there is also a great difference
in virulence towards the lower animals, the bacillus from the ox
having a much higher virulence. This organism when injected
in suitable quantities into the ox produces a local tubercular
lesion, which is usually followed by a generalised and fatal
tuberculosis ; whereas injection of human tubercle bacilli pro-
duces no more than a local lesion, which undergoes retrogression.
(In certain experiments, e.g. those of Delepine, Hamilton, and
Young, general tuberculosis has "been produced in the bovine
species by tubercle bacilli from the human subject, but these
results are exceptional.) Corresponding differences come out

VARIETIES OF TUBERCULOSIS 275

in the case of the rabbit; in fact, intravenous injection of
suitable quantities in this animal is the readiest method of
distinguishing the two types — an acute tuberculosis resulting
with the bovine, but not with the human type. In guinea-pigs
and monkeys a generalised tuberculosis may result from sub-
cutaneous injection of bacilli of the human type, but in this
case also the -difference in favour of the greater virulence of the
bovine type is made out. With regard to the distribution of
the two tyiws of organisms, it may be stated that, so far as we
know, the bacillus obtained from bovine tuberculosis is always
of the bovine type, and the same may be said to be true of
tuberculosis in pigs ; in fact this seems to be the prevalent
<>ri:aiii>!it in animal tuberculosis. In human tuberculosis the
bacilli in a large majority of the cases are of the human type ;
but, on the other hand, in a certain proportion bacilli of the
bovine type are present, the bacilli when cultivated being
indistinguishable by any means at our disposal from those
obtained from bovine tuberculosis. The Royal Commission
found the bovine type in 14 out of 60 cases of human
tuberculosis — a somewhat higher proportion than has been
obtained by most other investigators — and in all of these,
with one exception, the bacilli were obtained either from caseous
cervical glands, or from the lesions of primary abdominal tuber-
culosis, that is from cases where there was evidence of infection
by alimentation. It is also to be noted that almost all the
tubercular lesions from which the bovine type has been obtained
have btrn in children. The general result accordingly is that
bovine tubercle bacilli are present in a certain proportion of
caaea of tuberculosis in young subjects, and that these are
r-pccially cases where infection by the alimentary canal has
occurred. It must thus be held as established that tuberculosis
is transmissible from the ox to man, and that the milk of
tubercular cows is a common vehicle of transmission.

Although most of the bacilli which have been cultivated
correspond to one of the two types, as above described, it is
also to%be noted that intermediate varieties are met with. It
has also been found that the type characters of the bacillus are
not constant. Various observers have found it possible to
modify bacilli of the human type by passing them through the
bodies of certain animals, e.y. guinea-pigs, sheep, and goats, So
that they acquire the characters of bovine bacilli. In view
of these facts, it is probable that bovine bacilli will undergo
corresponding modifications in the tissues of the human subject
— what period of time is necessary for such a change we

276 TUBERCULOSIS

cannot say. It is thus possible that the cases of human tuber-
culosis from which the bovine type has been obtained do not
represent the full number where infection from the ox has
occurred. It is quite likely that although the bovine bacilli
are more virulent to the lower animals than the human bacilli
are, this does not also hold in the case of the human subject.
In fact, the comparative chronicity of the primary abdominal
lesions in children, in the first instance, would point rather to a
low order of virulence towards the human subject. We may
also add that there are cases, notably those of Ravenel, in
which accidental inoculation of the human subject with bovine
tubercle has resulted in the production of tuberculosis.

2. Avian Tuberculosis. — In the tubercular lesions in birds
there are found bacilli which correspond in their staining re-
actions and in their morphological characters with those in
mammals, but differences are observed in cultures, and also on
experimental inoculation. These differences were first described
by Maffucci and by Rivolta, but special attention was drawn to
the subject by a paper read by Koch at the International Medical
Congress in 1890. Koch stated that he had failed to change
the one variety of tubercle bacillus into the other, though he did
not conclude therefrom that they were quite distinct species.
The following points of difference may be noted : —

On glycerin agar and on serum, the growth of tubercle bacilli from
birds is more luxuriant, has a moister appearance (Fig. 79, C), and,
moreover, takes place at a higher temperature, 43*5° C., than is the case
with mammalian tubercle bacilli. Experimental inoculation brings out
even more distinct differences. Tubercle bacilli derived from the human
subject, for example, when injected into birds, usually fail to produce
tuberculosis, whilst those of avian origin very readily do so. Birds are
also very susceptible to the disease when fed with portions of the organs
of birds containing tubercle bacilli, but they can consume enormous
quantities of phthisical sputum without becoming tubercular (Strauss,
Wurtz, Nocard). No doubt, on the other hand, there are cases on record
in which the source of infection of a poultry-yard has apparently been
the sputum of phthisical patients. Again, tubercle bacilli cultivated from
birds have not the same effect on inoculation of mammals as ordinary
tubercle bacilli have. When guinea-pigs are inoculated subcutaneously
they usually resist infection, though occasionally a fatal result' follows.
In the latter case, usually no tubercles visible to the naked eye are found,
but numerous bacilli may be present in internal organs, especially in the
spleen, which is .much swollen. Further, intravenous injection even of
large quantities of avian tubercle bacilli, in the case of dogs, leads to no
effect, whereas ordinary tubercle bacilli produce acute tuberculosis. [The
rabbit, on the other hand, is comparatively susceptible to avian tuber-
culosis (Nocard).]

There is, therefore, abundant evidence that the bacilli derived

VARIETIES OF TUBERCULOSIS 277

from the two classes of animals show important differences, and,
reasoning from analogy, we might infer that probably the human
.subject also would be little susceptible to infection from avian
tuberculosis. The question remains, are these differences of a
permanent character 1 The matter seems conclusively settled by
the experiments of Nocard, in which mammalian tubercle bacilli
have been made to acquire all the characters of those of avian
origin. The method adopted was to place bacilli from human
tuberculosis in small collodion sacs (v. p. 144) containing bouillon,
and then to insert each sac in the peritoneal cavity of a fowl.
The sacs were left in situ for periods of from four to eight
months. They were then removed, cultures were made from
their contents, fresh sacs were inoculated from these cultures
and introduced into other fowls. In such conditions the bacilli
are subjected only to the tissue juices, the wall of the sac being
impervious both to bacilli and to leucocytes, etc. After one
sojourn of this kind, and still more so after two, the bacilli are
found to have acquired some of the characters of avian tubercle
bacilli, but are still non-virulent to fowls. After the third
sojourn, however, they have acquired this property, and produce
in fowls the same lesions as bacilli derived from avian tuber-
culosis. It therefore appears that the bacilli of avian tuber-
culosis are not a distinct and permanent species, but a variety
which has been modified by growth in the tissues of the bird.
It is also interesting to note that Rabinowitch has cultivated
tubercle bacilli of the mammalian type from some cases of tuber-
culosis in parrots kept in confinement.

3. Tuberculosis in the Fish. — Bataillon, Dubard, and Terre
cultivated from a tubercle-like disease in a carp, a bacillus
which, in staining reaction and microscopic characters, closely
agrees with the tubercle bacillus. The lesion with which it
was associated was an abundant growth of granulation tissue in
which numerous giant-cells were present. It forms, however,
luxuriant growth at the room temperature, the growth, being
thick and moist like that of avian tubercle bacilli (Fig. 84, c).
Growth does not occur at the body temperature, though by
gradual acclimatisation a small amount of growth has been
obtained up to 36° C. Furthermore, the organism appears to
undergo no multiplication when injected into the tissues of
mammals, and attempts to modify this characteristic have so
far been unsuccessful. Weber and Taute have cultivated this
organism from mud, and also from organs of healthy frogs. It
is thus probably to be regarded as a saprophyte which is only
occasionally associated with disease in the fish. According to

278 TUBERCULOSIS

the results of different experimenters, it is possible to modify
human tubercle bacilli by allowing them to sojourn in the
tissues of cold-blooded animals, e.g. the frog, blind-worm, etc.,
so that they flourish at lower temperatures. These results have,
however, been recently called in question, as it has been stated
the organisms obtained were not modified tubercle bacilli, but
other acid-fast bacilli which may be found in the tissues of
normal cold-blooded animals. This question must accordingly
be considered still an open one.

All the above facts taken together indicate that tubercle
bacilli may become modified in relative virulence and in con-
ditions of growth by sojourn in the tissues of various animals.
This modification appears slight, though of definite character in
the case of bovine tuberculosis, more distinct in the case of
avian tuberculosis, and much more marked, if not permanent, in
the case of fish tuberculosis, that is, of course, in their relations
to the bacilli from the human subject.

Other Acid-fast Bacilli. — Within recent years a number of
bacilli presenting the same staining reaction as the tubercle
bacilli have been discovered. Such bacilli have a comparatively
wide distribution in nature, as they have been obtained from
various species of grass, from butter and milk, from manure,
and from the surfaces of animal bodies. Microscopically, they
agree more or less closely with tubercle bacilli, though most of
them are shorter and plumper ; many of them show filamentous
and branching forms under certain conditions of culture. More-
over, on injection, they produce granulation-tissue nodules which
may closely resemble tubercles, although on the whole there is
a greater tendency to softening and suppuration, and usually the
lesions are localised to the site of inoculation. The most im-
portant point of distinction is the fact that their multiplication
on artificial media is much more rapid, growth usually being
visible within forty-eight hours .and often within twenty-four
hours at 37° C. Furthermore, in most instances growth occurs
at the room temperature. The general character of the cultures
in this group is a somewhat irregular layer, often with wrinkled
surface, dry or moist in appearance, and varying in tint from
white to yellow or reddish brown. The number of such organ-
isms is constantly being added to, but the following may be
mentioned as examples : —

Moellers Grass Bacilli I. and II. — The former was found in infusions
of Timothy-grass (Phleum pratense). It is extremely acid-fast, morpho-
logically resembles the tubercle bacillus, and in cultures may show club
formation and branching. The lesions produced closely resemble

OTHER ACID-FAST BACILLI 279

tubercles. The colonies, visible in thirty-six hours, are scale-like and of
^ivyish-whitc colour (Fig. 84, a}. Moeller's bacillus II. was obtained
from the dust of a hay-loft. The colonies at first are moist and some-
what tenacious, but afterwards run together, and are of a dull yellowish
colour. The general results of inoculation resemble those of grass
bacillus I., but are less marked. Moeller also obtained a similar organism
from milk. He also discovered a third acid-fast bacillus, which ho
obtained from manure and therefore called the " Mistbacillus " (dung
bacillus). This organism has analogous characters, though presenting
minor differences. It also has pathogenic effects.

Petri and Rabinowitch independently cultivated an acid-fast bacillus
from butter ("butter bacillus"), in which it occurs with comparative

/* *

FIG. 83.— Moeller's Timothy-grass bacillus. c

From a culture on agar. FIG. 84.— Cultures of acid-fast bacilli

Stained with carbol-fuchsin, and treated grown at room temperature,

with 20 per cent, sulphuric acid. (rt) Moeller's Timothy-grass bacillus I.

x 1000. (&) The Petri-Rabinowitch butter bacillus,

frequency. The organism resembles the tubercle bacillus, although it is
on the whole shorter and thicker. Its lesions closely resemble tuber-
culosis, especially when injection of the organism is made into the
peritoneal cavity of guinea-pigs, along with butter,— the method usually
adopted in searching for tubercle bacilli in butter. This organism
produces pretty rapidly a wrinkled growth (Fig. 84, b) not unlike that
of Moeller's grass bacillus II. Korn has also obtained, other two bacilli
from butter which he holds to be distinct from one another and from
Rabinowitch's bacillus. The points of distinction are of a minor
character. Other more or less similar bacilli have been cultivated by
Tobler, Coggi, and others.1

Another bacillus of considerable interest is Johne's bacillus or the

i For further details on this subject, vide Potet, Etudes sur les bacilles dites
acidophiles. Paris, 1902.

280

TUBERCULOSIS

V-

bacillus of "chronic bovine pseudo-tuberculous enteritis," the lesions
produced by it being corrugated thickenings of the mucous membrane,
especially of the small intestine. The disease has now been observed in
various countries, and several cases in Britain have been recorded by
M'Fadyean. The bacilli occur in large numbers in the lesions, and can
readily be found in scrapings from the surface. They resemble the
tubercle bacillus in appearance, but on the whole are rather shorter ; they
are equally acid-fast. The organism has not yet been cultivated outside
the body.

Smegma Bacillus. — This organism is of importance, as in form and
staining reaction it somewhat resembles the tubercle bacillus and may be

mistaken for it. It occurs

^,— / • often in large numbers in

the smegma preeputiale and
in the region of the external
genitals, especially where
there is an accumulation of
fatty matter from the secre-
tions. Morphologically it is
a slender, slightly curved
organism, like the tubercle
bacillus, but usually dis-
tinctly shorter (Fig. 85).
Like the tubercle bacillus, it
stains with some difficulty
and resists decolorisation
with strong mineral acids.
Most observers ascribe the
latter fact to the fatty
matter with which it is
surrounded, and find that
if the specimen is treated
with alcohol the organ-
ism is easily decolorised.
Czaplewski, however, who

claims to have cultivated it on various media, finds that in culture it
shows resistance to decolorisation both with alcohol and with acids, and
considers, therefore, that the reaction is not due to the surrounding fatty
medium. We have found that in smegma it can be readily decolorised
by a minute's exposure to alcohol after the usual treatment with sulphuric
acid, and thus can be readily distinguished from the tubercle bacillus.
We, moreover, believe that minor points of difference in the microscopic
appearances of the two organisms are quite sufficient to make the
experienced observer suspicious if he should meet with the smegma
bacillus in urine, and lead him to apply the decolorising test. Difficulty
will only occur when a few scattered bacilli retaining the fuchsin are
found.

Its cultivation, which is attended with some difficulty, was first effected
by Czaplewski. On serum it grows in the form of yellowish -grey
irregularly rounded colonies about 1 mm. in diameter, sometimes
becoming confluent to form a comparatively thick layer. He found that
it also grew on glycerin agar and in bouillon. It is non -pathogenic to
various animals which have been tested.

Cowie has recently found that acid- fast bacilli are of common occur-

FIG. 85. — Smegma bacilli. Film preparation

of smegma.
Ziehl-Neelsen stain, x 1000.

ACTION OF DEAD TUBERCLE BACILLI 281

rence in the secretions of the external genitals, mammae, etc., in certain
of the lower animals, and that these organisms vary in appearance. He
considers that the term "smegma bacillus" probably represents a
number of allied species.

The question may be asked — do these results modify the
validity of the staining reaction of tubercle bacilli as a means of
diagnosis? The source of any acid-fast bacilli in question is
manifestly of importance, and it may be stated that when these
have been obtained from some source outside the body, or where
contamination from without has been possible, their recognition
as tubercle bacilli cannot be established by microscopic examina-
tion alone. In the case of material coming from the interior of
the body, however, — sputum, etc., — the condition must be looked
on as different, and although an acid-fast bacillus (not tubercle)
has been found by Rabinowitch in a case of pulmonary gangrene,
we have no sufficient data for saying that acid-fast bacilli other
than the tubercle bacillus flourish within the tissues of the hwnian
body, except in such rare instances as to be practically negligible.
(To this statement the case of the leprosy bacillus is of course
an exception.) Accordingly, up till now, the microscopic ex-
amination of sputum, etc., cannot be said to have its validity
shaken, and we have the results of enormous clinical experience
that such examination is of practically unvarying value. Never-
theless the facts established with regard to other acid-fast bacilli
must be kept carefully in view, and great care must be exercised
when only one or two bacilli are found, especially if they deviate
in their morphological characters from the tubercle bacillus.

Action of dead Tubercle Bacilli. — The remarkable fact has
been established by independent investigators, that tubercle
bacilli in the dead condition, when introduced into the tissues
in sufficient numbers, can produce tubercle-like nodules. Prudden
and Hodenpyl, by intravenous injection in rabbits of cultures
sterilised by heat, produced in the lungs small nodules in which
giant-cells, but no caseation, were occasionally present, and
which were characterised by more growth of fibrous tissue than
in ordinary tubercle. The subject was very fully investigated
with confirmatory results by Straus and Gamaleia, who found
that, if the number of bacilli introduced into the circulation were
large, there resulted very numerous tubercle nodules with well-
formed giant-cells, and occasionally traces of caseation. The
bacilli can be well recognised in the nodules by the ordinary
staining method. In these experiments the bacilli were killed
by exposure to a temperature of 115° C. for ten minutes before
being injected. Similar nodules can be produced by intra-

282 TUBERCULOSIS

peritoneal injection. Subcutaneous injection, on the other
hand, produces a local abscess, but in this case no secondary
tubercles are found in the internal organs. Further, in many
of the animals inoculated by the various methods, a condition of
marasmus sets in and gradually leads to a fatal result, there
being great emaciation before death. These experiments, which
have been confirmed by other observers, show that even after
the bacilli are dead they preserve their staining reactions in
the tissues for a long time, and also that there are apparently
contained in the bodies of the dead bacilli certain substances
which act locally, producing proliferative and, to a less extent,
degenerate changes, and which also markedly affect the general
nutrition. S. Stockman has found that an animal inoculated
with large numbers of dead tubercle bacilli afterwards gives the
tuberculin reaction.

Practical Conclusions. — From the facts above stated with
regard to the conditions of growth of the tubercle bacilli, their
powers of resistance, and the paths by which they can enter the
body and produce disease (as shown by experiment), the manner
by which tuberculosis is naturally transmitted can be readily
understood. Though the experiments of Sander show that
tubercle bacilli can multiply on vegetable media to a certain
extent at warm summer temperature, it is doubtful whether all
the conditions necessary for growth are provided to any extent
in nature. At any rate, the great multiplying ground of tubercle
bacilli is the animal body, and tubercular tissues and secretions
containing the bacilli are the chief, if not the only, means by
which the disease is spread. The tubercle bacilli leave the body
in large numbers in the sputum of phthisical patients, and when
the sputum becomes dried and pulverised they are set free in
the air. Their powers of resistance in this condition have already
been stated. As examples of the extent to which this takes
place, it may be said that their presence in the air of rooms
containing phthisical patients has been repeatedly demonstrated.
Williams placed glass plates covered with glycerine in the
ventilating shaft of the Brompton Hospital, and after five days
found, by microscopic examination, tubercle bacilli on the surface,
whilst Klein found that guinea-pigs kept in the ventilating shaft
became tubercular. Cornet produced tuberculosis in rabbits by
inoculating them with dust collected from the walls of a con-
sumptive ward. Tubercle bacilli are also discharged in consider-
able quantities in the urine in tubercular disease of the urinary
tract, and also by the bowel when there is tubercular ulceration ;
but, so far as the human subject is concerned, the great means

PRACTICAL CONCLUSIONS 283

of disseminating the bacilli in the outer world is dried phthisical
sputum, and the source of danger from this means can scarcely
be overestimated. Every phthisical patient ought to be looked
upon as a fruitful source of infection to those around, and should
only expectorate on pieces of rag which are afterwards to be
burnt, or into special receptacles which are to be then sterilised
either by boiling or by the addition of a 5 per cent, solution of
carbolic acid.

Another great source of infection is in all probability the
milk of cows affected with tuberculosis of the udder. In such
cases the presence of tubercle bacilli in the milk can usually be
readily detected by centrifugalising it, and then examining the
dei>osit microscopically, or by inoculating an animal with it. As
pointed out by Woodhead and others, the milk from cows thus
affected is probably the great source of tabes mesenterica, which
is so common in young subjects. In these cases there may be
tubercular ulceration of the intestine, or it may be absent.
Woodhead found that out of 127 cases of tuberculosis in children,
the mesenteric glands showed tubercular affection in 100, and
that there was ulceration of the intestine in 43. It is especially
in children that this mode of infection occurs, as in the adult
ulceration of the intestine is rare as a primary infection, though
it is common in phthisical patients as the result of infection by
the bacilli in the sputum which has been swallowed. There is
less risk of infection by means of the flesh of tubercular animals,
for, in the first place, tuberculosis of the muscles of oxen being
very rare, there is little chance of the bacilli being present in the
flesh unless the surface has been smeared with the juice of the
tubercular organs, as in the process of cutting up the parts ; and,
in the second place, even when present they will be destroyed if
the meat is thoroughly cooked.

We may state, therefore, that the two great modes of infection
are by inhalation, and by ingestion, of tubercle bacilli. By the
former method the tubercle bacilli will in most cases be derived
from the human subject ; in the latter, probably from tubercular
cows, though inhaled tubercle bacilli may also be swallowed and
contamination of food by tubercular material from the human
subject may occur. Alike when inhaled and when ingested,
tubercle bacilli may lodge about the pharynx and thus come to
infect the pharyngeal lymphoid tissue, tonsils, etc., tubercular
lesions of these parts being much more frequent than was
formerly supposed. Thence the cervical lymphatic glands may
become infected, and afterwards other groups of glands, bones,
or joints, and internal organs.

284 TUBERCULOSIS

The Specific Reactions of Tubercle Bacillus. — The tubercle
bacillus belongs to the group of organisms which do not secrete
soluble toxins into the media in which they are growing. It
shares with other members of the group the capacity to induce
serious changes in the metabolism of an animal. As, in other
similar cases, we are in the dark as to how these changes come
about, and thus can only summarise the chief effects which, by
present methods, can be detected as occurring in the bodies of
infected animals. These effects which, it may be remarked, are
of value in the diagnosis of tubercular affections, consist on the
one hand (a) of certain phenomena of supersensitiveness, and on
the other (6) of certain changes in the blood serum of tubercular
patients resulting from reactions of immunity. The former are
seen when the bacilli or substances artificially derived from their
bodies are introduced into the tissues of those suffering from
tuberculosis, and were first demonstrated by Koch in his work
on tuberculin. In recent times, examples of similar effects are
the ophthalmic reaction of Calmette and the cutaneous reaction
of von Pirquet. The changes in the blood serum of infected
persons depend on the presence of anti-substances in the blood.
These may be of the nature of (a) immune bodies which lead to
fixation of complement, and (b) precipitins, (c) agglutinins, (d)
opsonins. These may now be severally discussed in detail.

(1) Phenomena of Supersensitiveness. (a) Koctis Old Tuber-
culin.— Koch (1890-1) stated that if in a guinea-pig suffering
from the effects of a subcutaneous inoculation with tubercle
bacilli, a second subcutaneous inoculation of tubercle bacilli was
practised in another part of the body, superficial ulceration
occurred in the primary tubercular nodule, the wound healed,
and the animal did not succumb to tuberculosis. This reaction
was further studied by means of tuberculin, which consisted of
a concentrated glycerin bouillon culture of tubercle in which the
bacilli had been killed by heat. Its essential components pro-
bably were the dead and often macerated bacilli and the
substances indestructible by boiling which existed in these
bacilli, or which were formed during their growth. The in-
jection of '25 c.c. of tuberculin into a healthy man causes, in
from three to four hours, malaise, tendency to cough, laboured
breathing, and moderate pyrexia ; all of which pass off in
twenty-four hours. The subcutaneous injection, however, of
"01 c.c. into a tubercular person gives rise to similar symptoms
(now known as the tuberculin reaction), but in a much more
aggravated form, and in addition there occurs around any
tubercular focus great inflammatory reaction, resulting in necrosis

PHENOMENA OF SUPERSENSITIVENESS 285

and a casting off of the tubercular mass, when this is possible,
as for instance in the case of lupus. The bacilli are, it was
shown, not killed in the process.

The hopes which the introduction of tuberculin raised, that a
curative agent against tuberculosis had been discovered, were
soon found not to be justified. It was very difficult to see how
the necrosed material which was produced and which contained
the still living bacilli, could be got rid of either naturally, as
would be necessary in the case of a small tubercular deposit in
a lung or a lymphatic gland, or artificially, as in a complicated
joint-cavity where surgical interference could be undertaken.
Not only so, but the ulceration which might be the sequel of the
necrosis appeared to open a path for fresh infection. Soon facts
were reported which justified these criticisms. Cases where
rapid acute tubercular conditions ensued on the use of tuberculin
were reported, and in a few months the treatment was practically
abandoned.

(b) The Cutaneous Tuberculin reaction of von Pirquet and
tJie Op/tt/tal mo-reaction of Calmette. — In recent times the diagnosis
of tuberculosis has been considerably aided by the introduction
of these two tests. Both are essentially of the same nature, and
depend like the original tuberculin reaction on the sensitiveness
of the tissues of tubercular patients to tuberculin.

The cutaneous test is carried out as follows : The skin, usually
that of the flexor aspect of the forearm, is well cleansed with
ether and then allowed to dry. Two drops of tuberculin are
placed on the prepared surface about four inches apart, and then
midway between the two drops a small spot is scarified with a
small metal bore constructed for the purpose. This serves as a
control, any reaction which follows in it being merely a traumatic
one. Similar scarification is effected through the drops of
tuberculin, so that the scarified spots are exposed to its action.
Small portions of cotton wool are placed over the drops to
prevent the tuberculin from running off, and the latter is allowed
to act for ten minutes. After that time the cotton wool is
removed ; no dressing is required. In the process of scarification
only the epidermis should be injured and blood should not be
drawn. The "old" tuberculin of Koch is that used. In the
case of a positive reaction an inflammatory redness and swelling
make their appearance round the sites of tuberculin inoculation,
generally within a few hours, and at the end of twenty-four hours
there is a distinct inflammatory papule about half an inch in
diameter, with a somewhat paler centre like a spot of urticaria;
sometimes in the centre there are minute vesicles. The maximum

286 TUBERCULOSIS

effect usually occurs within forty- eight hours, and after that time
the reaction gradually recedes. Such is the typical reaction, but
of course slighter, and also more intense reactions are met with.
In a negative reaction all three points of scarification show
merely a slight traumatic redness which soon passes off.

For the ophthalmo-reaction Calmette uses a purified tuberculin.
The tuberculin is prepared as in Koch's original method, and is
precipitated with 95 per cent, alcohol; the precipitate is then
dissolved in water. This process is repeated other two times,
and the final precipitate is made up as a 1 per cent, solution in
distilled water. For use, in the case of an adult, a drop of this
solution is placed in the conjunctival sac and the fluid allowed
to spread over the surface ; for children about half this quantity
is sufficient. In the case of a positive reaction the ocular con-
junctiva is congested, the lids become somewhat swollen and
their inner surface presents a bright red colour, there is increased
secretion of tears and a varying amount of fibrinous exudation.
The reaction usually reaches its maximum in from six to ten
hours after the instillation, and commences to pass off" in from
twenty-four to thirty-six hours, — in children a little sooner.

The general results obtained by these two reactions appear to
correspond closely. A distinct positive result obtained by either
is practically conclusive as to the presence of a tubercular lesion.
In cases of latent tuberculosis the reaction is sometimes obtained,
sometimes not. Again, in very advanced cases of tuberculosis,
especially a short time before death, a negative result may be
got y in some of these cases v. Pirquet has met with a colourless
papule or a livid spot without exudation, conditions which he
describes as indicating a "cachectic reaction." The ophthalmo-
reaction is the more easily applied, at least in adults, but its use
is contra-indicated when there is any abnormal condition of the
conjunctiva. Even apart from this, however, inflammatory
symptoms of disagreeable severity sometimes supervene. It
should also be noted that a second test ought not to be applied
to the same eye, as the first may produce a condition of super-
sensibility (p. 284). V. Pirquet claims for his method that in the
case of children it can be satisfactorily carried out with greater
ease than the ophthalmic test.

It will be recognised that the processes underlying the original
tuberculin reaction on the one hand, and the cutaneous and
ophthalmic reactions on the other, are analogous. In the former
there is the occurrence of local inflammation with metabolic
changes and fever ; in the latter, of mild inflammatory effects, —
in both cases the phenomena being found only in tubercular

PHENOMENA OF SUPERSENSITIVENESS 287

subjects. The original explanation given by Koch of the
tuberculin reaction was that the essential constituent of tuber-
culin being toxic products of the tubercle bacillus, the action
of these was superadded to the toxins produced at the focus of
infection. The combined action of the toxins from these two
sources caused a rapid necrosis of the newly formed cells, and
opened the way for the dead tissue being rapidly cast off. This
explanation was, however, not generally accepted, for it was
found that other substances, such as albumoses, when injected
into animals suffering from local tuberculosis, gave rise to modified
effects of the same kind as those produced by tuberculin. This
dissatisfaction with regard to the original explanation is ac-
centuated by a consideration of the effects seen in the Calmette
and v. Pirquet tests, as these clearly indicate that the sensitive-
ness in a tubercular subject is not confined to tissues actually
affected with the tubercular process, but is also manifested in
parts of the body distant from the site of actual infection.

Further, it has been found, first, that the injection of tuber-
culin directly into a tubercular focus does not produce the
tuberculin reaction (a fact which militates against the idea of
concentrations of toxins), and, secondly, that the injection of
living or dead tubercle bacilli into healthy animals produces
anaphylactic phenomena similar to those originated by foreign
albumins generally. At the present time, therefore, although
no full explanation can be given of the tuberculin reaction and
of similar reactions, it is likely that in tuberculosis a general
hypersensitivenesH is developed, and may be the underlying
phenomenon. These reactions must therefore be considered in
the light of what will be set forth on the subject of hypersen-
sitiveness in the chapter on Immunity.

The Use of Old Tuberculin in the Diagnosis of Tuberculosis in Cattle. —
111 cattle, tuberculosis may be present without giving rise to apparent
symptoms. It is thus important from the point of view of human
infection that an early diagnosis should be made. The method is
applied as follows : — The animals are kept twenty-four hours in their
stalls, and the temperature is taken every three hours, from four hours
before the injection till twenty-four a'fter. The average temperature in
cattle is 102'2° F. ; 30 to 40 centigrammes of tuberculin are injected, and
if the animal be tubercular the temperature rises 2° or 3° F. in eight to
twelve hours, and continues elevated for ten to twelve hours. Bang, who
has worked most at the subject, lays down the principle that the more
nearly the temperature approaches 104° F. the more reason for suspicion
is there. He gives a record of 280 cases where the value of the method
was tested by subsequent post-mortem examination. He found that with
proper precautions the error was only 3'3 per cent. The method has
l>ti 11 largely practised in all parts of the world, and is of great value.

288 TUBERCULOSIS

(2) Immunity Phenomena in Tuberculosis. — Koch's Tuber-
culin-R. The study of immunity phenomena in tuberculosis
dates from the introduction by Koch in 1897 of the substance
denominated by him "Tuberculin-R." Koch's new researches
consisted (1) of an attempt to immunise animals against the
tubercle bacillus by employing its intracellular toxins ; (2) of
trying to utilise such an immunisation to aid the tissues of an
animal already attacked with tubercle the better to combat the
effects of the bacilli. The method of obtaining the intracellular
toxins was as follows : Bacilli from young virulent cultures were
dried in vacuo, and disintegrated in an agate mill, treated with
distilled water and centrifugalised. The clear fluid was decanted,
and is called by Koch "Tuberculin-O." The remaining deposit
was again dried, ground, treated with water and centrifugalised,
the clear fluid being again decanted, and this process was
repeated with successive residues till no residue remained.
These fluids put together constitute the " Tuberculin-R." From
the fact that tuberculin-O gave no cloudiness when glycerin was
added, Koch concluded that it contained the substances present
in the glycerin-bouillon extracts originaUy used by him, and he
held this was borne out by the readiness with which a tuberculin
reaction could be caused by it. Similarly, as tuberculin-R gave
a cloudiness with glycerin and did not readily originate a
reaction, he considered that it contained different products of
the bacillus. When injected into animals in repeated and
increasing doses, -^ J^ mgrm. being the initial dose, tuberculin-R
was said to produce immunity against the original extract,
against tuberculin-O, and against living and virulent tubercle
bacilli. Further research has not confirmed this last result.

Itoch's New Tuberculin. — Another preparation has also been
introduced, known as " Koch's new tuberculin " (Bazillenemul-
sion). This is an emulsion of ground tubercle bacilli in water
containing 50 per cent, of glycerin ; it thus really contains both
tuberculin-O and tuberculin-R.

Scientific enquiry into the action of these new tuberculin
preparations has resulted in attempts being made to recognise in
their effects phenomena similar to those produced by organisms
such as the typhoid and cholera bacteria, the investigation of
which has brought out the complex processes at work in the
reaction of an organism against invading bacteria. The phen-
omena manifested in such cases consist in the formation of
immune-bodies, precipitins, agglutinins, and opsonins.

(1) Immune-bodies and Precipitins. — Evidence for the
existence of these in tuberculosis has been sought by applying

IMMTXITV I'HKXOMKXA IX TriJKIK'ULO-TS -JSO

the method of complement fixation (see p. 130), e.y. the serum
of a tubercular animal being mixed with tuberculin, the mixture
is tested for its capacity of absorbing complement. Following
this line, Wasserman and others have found evidence of the
presence of an antituberculin in tubercular foci, and this is
taken as an indication of the occurrence of a vital reaction
against the poisons of an invading organism. Generally speak-
ing, such an antituberculin is absent from the blood serum of
most tubercular patients. It is present, however, in the serum
of such individuals after they have been subjected to repeated
tuberculin injections. Here it is chiefly seen when a patient is
losing the capacity for reacting to the injections. Another
immunity phenomenon which may be observed is the formation
of a precipitate when some of the serum of a tuberculous patient
is added to a solution of tuberculin, the mixture being allowed
to stand for twenty-four hours (precipitin reaction). The exact
relationship of such precipitins to immune-bodies is still doubtful ;
that it is a close one is shown by the fact that such precipitates
have the property of absorbing complement. At present it is
enough to say that there is evidence in tubercular infection of a
vital reaction resulting in the formation of antagonistic bodies,
which may include both immune-bodies and precipitins. In sup-
port of the view that immune-bodies exist against the tubercle
bacillus, it may be said that the sera of certain animals, e.g.
rabbit and ox, when mixed with tuberculin, become capable of
deviating complement from a haemolytic combination.

(2) Agglutinins. — The serum of tubercular patients has been
found to exert an agglutinating action on the tubercle bacillus.
A convenient method is to add different amounts of serum, com-
mencing with, say, 1 c.c., to quantities of a dilution of the new
tuberculin (Bazillenemuhion) equivalent to 1 part of the
bacterial bodies to 10,000 of diluent, and leave the mixture for
twenty-four hours before observing. As with other agglutinative
observations, it is difficult to correlate the degree of agglutinating
power of the serum with the degree of immunisation possessed
1 > v the individual from which it was derived. The method has
been used by some as a means of diagnosis, but its value is
doubtful and is certainly inferior to the methods depending on
supersensitivcness.

(3) Opsonins. — The serum of most normal men and of several
s|n'cies of animals normally contains opsonins to the tubercle
bacillus. The opsonic effect is also manifested in varying
degree by the serum during the course of natural infection ;
such variations are considered l>elow.

290 TUBERCULOSIS

In considering the relationships of the specific immune
reactions against the tubercle bacillus, it is to be noted that
while the existence of such reactions has been established, the
development of these to an extent likely to benefit an infected
animal is limited, and the production of such a lasting immunity
as would enable it to resist an infection or to throw off an
infection already established is extremely difficult or impossible.
There are probably factors in the pathology of the tubercular
process which militate against such an occurrence. This pro-
cess seems to differ from what occurs in more acute infections, in
that a local lesion may be in existence for a very considerable
period without other parts of the body being much or at all
concerned. This is especially marked in certain tubercular
manifestations, the outstanding example of which is lupus, in
which for years, while the bacilli are present and active in the
skin, even the adjacent lymphatic glands may show no signs of
disease. What underlies this apparent independence of the
body generally in relation to a serious condition affecting one
locality is unknown. Other examples of a similar process are
found in leprosy and also in certain chronic suppurations of the
skin. f

Therapeutic Applications of the Tuberculins. — We have
already stated that the use of the old tuberculin to mechanically
remove local foci of tuberculosis through the use of large doses
of the reagent was soon found to be impracticable, but both this
preparation and its modifications have been largely used in
what is now denominated "vaccine-therapy." It has been
already pointed out that the tubercular process is peculiar in
that the disease may exist locally without much affecting the
general health of the infected individual. The principle of
vaccine-therapy may roughly be said to be to bring into play the
potential but latent defensive mechanisms of the body with the
object of so reinforcing the cells locally attacked as to enable
them to destroy the invading bacteria. This is effected by
introducing into the body small doses of the infecting agent, and
is in reality an immunisation carried through after infection has
already taken place. For this purpose all the tuberculin pre-
parations, but especially tuberculin-R and the "new tuberculin,"
have been used. In the case of both the latter, doses commen-
cing with from TJ^ to 3^ mgrm., gradually increased, were
given every second day, and the rule originally laid down for
the regulation of the dosage was that no amount should be
given which raised the temperature more than '5° F. Opinion
varied as to the efficacy of such treatment. There was little

OPSONINS IN TUBERCULOSIS 291

doubt that in certain cases of local conditions, such as lupus,
tubercular joints, glands, and genito-urinary tuberculosis, improve-
ment followed its application ; but where febrile conditions
indicated that general disturbances were in existence, there was
little justification for its being applied, and even in many local
conditions the absence of benefit was so marked that by many
physicians the method had been abandoned.

Active Immunisation associated with Opsonic Observations. —
The credit of rehabilitating the vaccine-therapy of tuberculosis
and of defining its scope belongs to Wright, who directed atten-
tion to the possibility of controlling the use of the tuberculin by
observations of its effect on the opsonic qualities of the serum.
Early in his work he showed that tubercle bacilli when sensitised
by an appropriate serum, were readily phagocyted by the poly-
morpho-nucleate leucocytes, and the relative sensitising capacities
of serum from tubercular and non-tubercular cases has been
widely studied. According to Wright, in strictly localised tuber-
culosis, the opsonic index is persistently low, varying from '1 to '9,
while in tuberculosis with general disturbances it fluctuates
greatly from day to day, being sometimes below, sometimes
above unity. To take the former and simpler case, he holds
that if the treatment with injections of tuberculin be controlled
by noting the effect produced on the opsonic index, great
improvement in the patient's condition may result. Wright's
interpretation of what occurs is briefly as follows : For
reasons unknown the opsonic qualities of the body fluids
may become abnormally low, and the tubercle bacilli, if
they gain admission to the body, can multiply locally. This
multiplication is associated with a still further local diminution
of the opsonins. By the introduction of such a substance
as tuberculin, the bodily mechanism, whatever it is, which
produces the opsonins is stimulated, and a rise in the general
opsonic index occurs. Naturally this is accompanied by a
passing to the site of infection of fluids more rich in opsonins
than previously, the activity of the phagocytes comes into play,
and the tubercle bacilli are destroyed. But any such vaccination
process must be controlled by constant observations of the
opsonic index, and it is only by this means, not only that good
results can be obtained, but that the production of harmful
effects can be prevented. The reason of this is that in a great
many cases the injection of a bacterial vaccine is followed by a
decrease in the opsonic qualities of the serum, — the occurrence
of a negative phase. During such a period of depression there
is probably an increased susceptibility to the action of the

292 TUBERCULOSIS

bacilli. Now, in order to get permanent benefit from the vac-
cination process, repeated injections of the tuberculin must be
practised, and if an injection be given during a negative phase,
actual harm may be done. The course of a successful vaccina-
tion is that, after the passing off of the negative phase, the
opsonic index should rise to above its original level, — the
occurrence of a positive phase. It is when this positive phase
is fully developed that a fresh inoculation can be practised with
success. The new negative phase which will now occur may not
cause a drop to below the level of the original state of the serum,
and the hope is that its succeeding positive phase will carry the
opsonic index still higher and ensure a still greater resistance
to the bacterium. There are very great variations in the capa-
cities shown by tubercular patients to react to a vaccination
process. In certain cases good positive phases are readily and
quickly produced, while in others, after an inoculation the negative
phase is long continued and may even show no tendency to pass
into a positive phase. The irregularities in the opsonic index in
cases where there is a general disturbance of metabolism, Wright
explains by supposing that they result from very irregular auto-
infections of the patient's body by tubercular products from the
local lesions, — positive and negative phases being produced with-
out the purposive quality which ought to characterise a success-
ful therapeutic vaccination. Such auto-infections may come
about in various ways, and Wright is of opinion that exercise,
for instance, may disseminate both tubercular products and
tubercle bacilli, — he having noticed in certain patients a fall in
the opsonic index after muscular exertion. For ordinary cases
with low opsonic index and no evidence of constitutional dis-
turbance, an amount of tuberculin corresponding to from one-
thousandth to a six-hundredth of a milligramme of tubercle
powder is a sufficient dose, and if any dose seems to produce a
pronounced negative phase, then a smaller dose ought to be tried
at the next inoculation. For cases clinically tubercular, where
the index is about normal, then smaller doses, say, the equivalent
of a two-thousandth of a milligramme or less, ought to be used,
— the effect on the index being carefully watched. In any case,
the dose which is found to give the highest positive phase is the
optimum dose, and one which need not necessarily be increased.
Cases where there is constitutional disturbance should be as a
rule left untreated.

With regard to the results obtained, many cases have been
brought forward by Wright and others where benefit has
followed the putting into practice of the principles enunciated,

OPSONINS IN TUBERCULOSIS 293

and there is little doubt that the work done has given a fresh
start to the active immunisation method in the treatment of
tuberculosis. An outstanding event of Wright's work in this
field has been his insistance on the good effects produced by
extremely small doses of tuberculin (down to the four-thousandth
of a milligramme) given at fairly long intervals (say ten days or
more). With regard to the efficacy of the opsonic method as
affording an index to the progress of a case, it must be recognised
that the method is still on its trial, and it is doubtful if even in
the work of the most careful observers the limits of the experi-
mental error of the opsonic method have been sufficiently defined.

Great controversy has taken place as to whether it is justi-
fiable, in the treatment of tubercular cases with tuberculin,
merely to rely on the observation of the clinical effects, with-
out having recourse to the constant estimation of the opsonic
index, which Wright considers advisable. There is no doubt
that in all complicated cases of tuberculosis, such as lung
affections and cases of multiple foci in the body, the treatment
ought to be in the hands of an expert. In cases of strictly
localised tubercle, however, such as adenitis, arthritis, cystitis,
or lupus, Wright admits that in many cases, without much
risk, an uncontrolled treatment may be undertaken. The
injections ought to begin with doses of one-twenty-thousandth
of a milligramme, with ten-day intervals intervening between
each dose. If clinical improvement occurs, the dose may be
gradually increased until it- reaches one four-thousandth of a
milligramme after six months. If the treatment of any other
form of tuberculosis be undertaken along similar lines, the pre-
liminary injection should not consist of more than one-fifty-
thousandth of a milligramme.

The whole question of the immunisation treatment of tuber-
culosis presents many difficulties, and it is the merit of Wright's
work that it has shed fresh light on some of these. One great
difficulty arises from the great chronicity of the results of the
infection in the majority of human cases. It is probably true
not only of man but of many species of animals used in experi-
mental inquiries, that many individuals are on the border-line
between immunity and susceptibility. From the widespread
distribution of the bacilli in centres of human population, it is
certain that the opportunity for infection arises in a very large
proportion of the race ; in many cases no results follow infection,
and in many others small lesions occur which do not develop
further ; this has actually been shown by morbid anatomists to be
the case. The disease being thus so often characterised by transient

294 TUBERCULOSIS

local effects without constitutional disturbance, the course of an
immunisation may be expected to be rather different from that
obtaining in an ordinary acute affection, though the underlying
processes may be of the same nature. It is difficult, for instance,
on account of the slowness of tubercular processes, to define
recovery from an attack of the disease, or to speak of an animal
recovering from the effect of an inoculation during an immunisa-
tion. It follows that little is known regarding an attenuation of
the tubercle bacillus analogous to what is an important feature
in immunisations against other organisms.

Antitubercular Sera. — Several attempts have been made to
treat tuberculosis with the serum of animals immunised by the
tubercle bacillus or its products. The most successful is perhaps
that of Maragliano. This author distinguishes between the toxic
materials contained in the bodies of the bacilli (which withstand,
unchanged, a temperature of 100° C.) and those secreted into
the culture fluid (which are destroyed by heat). The substance
used by him for immunising his animals consists of three parts
of the former and one of the latter. The animals employed are
the dog, the ass, the horse. The serum obtained from these is
capable of protecting healthy animals against an otherwise fatal
dose of tuberculin, but very little importance can be attached to
this result. Maragliano does not appear to have studied the
effects of this serum on tubercular animals, but it has been tried
in a great number of cases of human tuberculosis, 2 c.c. being
injected subcutaneously every two days. Improvement is said
to have taken place in a certain proportion, especially of mild
non-febrile cases.

An antitubercular serum has also been introduced by Marmorek.
This observer considers that the tubercle bacillus cannot produce
in ordinary media the toxins which it originates when exposed
to the antagonism of the bodily cells. He tries to make good
this defect by first growing it in a serum antagonistic to some of
the phagocytic cells of the body ; for this a leucotoxic serum is
used. When the bacillus has grown in this presumably favour-
able soil, it is transferred to a medium containing a substance
which may be unfavourable ; and for this there is employed a
medium containing liver extract, the liver being an organ in
which in man tubercular lesions are comparatively rare. The
bacilli being thus accustomed to an unfavourable surrounding
are used for immunising animals, the serum of which is now
suitable for the treatment of human tuberculosis. There is con-
siderable diversity of opinion as to the efficacy of Marmorek's
serum as a therapeutic agent.

METHODS OF EXAMINATION - 295

Methods of Examination. — (1) Microscopic Examinatioii.
Tuberculosis is one of the comparatively few diseases in which
a diagnosis can usually be definitely made by microscopic
examination alone. In the case of sputum, one of the yellowish
fragments which are often present ought to be selected ; dried
films are then prepared in the usual way, and stained by the
Ziehl-Neelsen stain (p. 108). In the case of urine or other fluids,
a deposit should first be obtained by centrifugalising a quantity
in a test-tube, or by allowing the fluids to stand in a tall glass
vessel (an ordinary burette is very convenient). Film prepara-
tions are then made with the deposit and treated as before. If
a negative result is obtained in a suspected case, repeated exam-
ination should be undertaken. To avoid risk of contamination
with the smegma bacillus, the meatus of the urethra should be
cleansed and the urine first passed should be rejected, or the
urine may be drawn off with a sterile catheter. As stated above,
it is only exceptionally that difficulty will arise to the experienced
observer from this cause. (For points to be attended to, vide p.
280). The detection of tubercle bacilli by microscopical methods
in sputum, pus, faeces, and even tissues, has been greatly facilitated
by the recent introduction of a preparation called "antiformin."
This is a mixture of equal parts of liquor soda3 chlorinatae (B.P.)
and of a 15 per cent, solution of caustic soda. It has a re-
markable disintegrative and dissolving action on the tissues, etc.,
so that after it has been allowed to act on sputum, for example,
and the mixture is centrifugalised, the resulting deposit is scanty
and the tubercle bacilli, if present, are accordingly greatly
concentrated. The time necessary may be judged of by the
appearance of the mixture, but it will generally be found that
the desired result will be obtained if one part of antiformin be
added to five or six parts of sputum and allowed to act for two or
three hours.

(2) Inoculation. — The guinea-pig is the most suitable animal.
If the material to be tested is a fluid, it is injected subcutaneously
or into the peritoneum ; if solid or semi-solid, it is placed in a
small pocket in the skin, or it may be ' thoroughly broken up
in sterile water or other fluid and the emulsion injected. By
this method, material in which no tubercle bacilli can be found
microscopically may sometimes be shown to be tubercular.

(3) Cultivation. — Owing to the difficulties this is usually
quite impracticable as a means of diagnosis, and it is also
unnecessary. The best method to obtain pure cultures is
to produce tuberculosis in a guinea-pig by inoculation with
tubercular material, and then, killing the animal after four or

296 ' TUBERCULOSIS

five weeks, to inoculate tubes of solidified blood serum, under
strict aseptic precautions, with portions of a tubercular organ,
e.g. the spleen. The portions of tissue should be fairly large,
and should be well rubbed into the broken surface of the medium.
Cultures may, however, be obtained from sputum by means of
antiformin, as this substance readily kills most of the ordinary
bacteria and has comparatively slight effect on the tubercle
bacillus. Antiforrain should be allowed to act on sputum in
the proportion and for the time mentioned in paragraph (1), the
mixture should then be centrifugalised, the supernatant fluid
removed, and the deposit washed with sterile water and again
centrifugalised, these processes being repeated several times. If,
then, inoculations be made from the deposit on blood serum or
on Dorset's egg medium, pure cultures of the tubercle bacillus
may, in some instances, be obtained. The method is one which
gives good results. Another somewhat similar method is that
introduced by Twort ; in this, portions of sputum are exposed
to the action of a 2 per cent, solution of ericolin (a glucoside) for
an hour at 38° C., and thereafter cultures are made on Dorset's
medium.

(4) Reactive phenomena. — The presence of immune-substances
in the blood and the tuberculin reaction, along with the methods
of applying the respective tests, have been described above
(p. 284).

CHAPTER XL

LEPROSY.

LEPROSY is a disease of great interest, alike in its clinical and
pathological aspects; whilst from the bacteriological point of
view, also, it presents some striking peculiarities. The invariable
association of large numbers of characteristic bacilli with all
leprous lesions is a well-established fact, and yet, so far,
attempts to cultivate the bacilli outside the body, or to produce
the disease experimentally in animals, have been attended with
failure. Leprosy, so far as is known, is a disease which is con-
fined to the human subject, but it has a very wide geographical
distribution. It occurs in certain parts of Europe — Norway,
Russia, Greece, etc., but is commonest in Asia, occurring in
Syria, Persia, etc. It is prevalent in Africa, being especially
found along the coast, in the Pacific Islands, in the warmer
parts of North and South America, and also to a small extent
in the northern part of North America. In all these various
regions the disease presents the same general features, and the
• study of its pathological and bacteriological characters, wherever
such has been carried on, has yielded similar results.

Pathological Changes. — Leprosy is characteristically a chronic
disease, in which there is a great amount- of tissue change, with
comparatively little necessary impairment of the general health.
In other words, the local effects of the bacilli are well marked,
often extreme, whilst the toxic phenomena are proportionately
at a minimum.

There are two chief forms of leprosy. The one, usually called
the tubercular form, — lepra tuberosa or tulerculosa, — is character-
ised by the growth of granulation tissue in a nodular form or
as a diffuse infiltration in the skin, in mucous membranes, etc.,
great disfigurement often resulting. In the other form, the
anaesthetic, — maculo-anajsthetic of Hansen and Looft, — the out-
standing changes are in the nerves, with consequent anaesthesia,
paralysis of muscles, and trophic disturbances.

297

298

LEPROSY

In the tubercular form, the disease usually starts with the
appearance of erythematous patches attended by a small amount
of fever, and these are followed by the development of small
nodular thickenings in the skin, especially of the face, of the
backs of hands and feet, and of the extensor aspects of arms and
legs. These nodules enlarge and produce great distortion of the
surface, so that, in the case of the face, an appearance is produced

FIG. 86. — Sections through leprous skin, showing the masses of
cellular granulation tissue in the cutis ; the dark points are cells
containing bacilli deeply stained.

Paraffin section ; Ziehl-Neelsen stain, x 80.

which has been described as " leonine." The thickenings occur
chiefly in the cutis (Fig. 86), to a less extent in the subcutaneous
tissue. The epithelium often becomes stretched over them,
and an oozing surface becomes developed, or actual ulceration
may occur. The cornea and other parts of the eye, the mucous
membrane of the mouth, larynx, and pharynx, may be the seat
of similar nodular growths. Internal organs, especially the
spleen, liver, and testicles, may become secondarily affected. In
all situations the change is of the same nature, consisting in an

BACILLUS OF LEPROSY 299

abundant formation of granulation tissue, nodular or diffuse in
its arrangement. In this tissue a large proportion of the cells
are of rounded or oval shape, like hyaline leucocytes ; a number
of these may be of comparatively large size, and may show
vacuolation of their protoplasm and a vesicular type of nucleus.
These are often known as " lepra-cells." Amongst the cellular
elements there is a varying amount of stroma, which in the
earlier lesions is scanty and delicate, but in the older lesions
may be very dense. Periarteritis is a common change, and very
frequently the superficial nerves become involved in the nodules,
and undergo atrophy. The tissue in the leprous lesions is
comparatively vascular, at least when young, and, unlike
tubercular lesions, never shows caseation. Some of the lepra
cells may contain several nuclei, but we do not meet with cells
resembling in their appearance tubercle giant-cells, nor does an
arrangement like that in tubercle follicles occur.

In the anesthetic form, the lesion of the nerves is the out-
standing feature. These are the seat of diffuse infiltrations,
which lead to the destruction of the nerve fibres. In the earlier
stages, in which the chief symptoms are pains along the nerves,
there occur patches on the skin, often of considerable size, the
margins of which show a somewhat livid congestion. Later,
these patches become pale in the central parts, and the periphery
becomes pigmented. There then follow remarkable series of
trophic disturbances, in which the skin, muscles, and bones are
especially involved. The skin often becomes atrophied, parch-
ment-like, and anaesthetic ; frequently pemphigoid bulke or other
skin eruptions occur. Partly owing to injury to which the feet
and arms are liable from their anaesthetic condition, and partly
owing to trophic disturbances, necrosis and separation of parts
are liable to occur. In this way great distortion results. The
lesions in the nerves are of the same nature as those described
above, but the granulation tissue is scantier, and has a greater
tendency to undergo cicatricial contraction. This is to be
associated with the fact that the bacilli are present in fewer
numbers.

Bacillus of Leprosy. — This bacillus was first observed in
leprous tissues by Hansen in 1871, and was the subject of several
communications by him in 1874 and later. Further researches,
first by Neisser in 1879, and afterwards by observers in
various parts of the world, agreed in their main results, and
confirmed the accuracy of Hansen's observations. The bacilli, as
seen in scrapings of ulcerated leprous nodules, or in sections,
have the following characters : — They are thin rods of practically

300 LEPROSY

the same size as tubercle bacilli, which they also resemble both
in appearance and in staining reaction. They are straight or
slightly curved, and usually occur singly, or two may be attached
end to end ; but they do not form chains. When stained they
may have a uniform appearance, or the protoplasm may be
fragmented, so that they appear like short rows of cocci. They
often appear tapered at one or both extremities ; occasionally

ffi

FlG. 87. — Superficial part of leprous skin ; the cells of the granula-
tion tissue appear as dark patches, owing to the deeply stained bacilli
in their interior. In the upper part a process of epithelium is seen.

Paraffin section ; stained with carbol-fuchsin and Bismarck-brown,
x 500.

there is slight club-like swelling. Degenerated and partially
broken-down forms are also seen. They take up the basic
aniline stains more readily than tubercle bacilli, but in order
to stain them deeply, a powerful stain, such as carbol-fuchsin,
is necessary. When stained, they strongly resist decolorising,
though they are more easily decolorised than tubercle bacilli
(p. 108). The bacilli are also readily stained by Gram's method.
Regarding the presence of spores, practically nothing is known,
though some of the unstained or stained points may be of this

POSITION OF THE BACILLI 301

nature. We have, however, no means of testing their powers of
resistance. Leprosy bacilli are non-motile.

Position of the Bacilli. — They occur in enormous numbers
in the leprous lesions, especially in the tubercular form — in fact,
so numerous are they that the granulation tissue in sections,
properly stained as above, presents quite a red colour under a
low power of the microscope (Plate II., Fig! 8). The bacilli

FIG. 88. — High -power view of portion of leprous nodule, showing the
arrangement of the bacilli within the cells of the granulation tissue.

Paraffin section ; stained with carbol-fuchsiu and methylene-blue
x 1100.

occur for the most part within the protoplasm of the round
cells of the granulation tissue, and are often so numerous that
the structure of the cells is quite obscured (Fig. 87). They
are often arranged in bundles which contain several bacilli
lying parallel to one another, though the bundles lie in various
directions (Fig. 88 and Plate II., Fig. 9). The appearance thus
presented by the cells filled with bacilli is very characteristic.
Bacilli are also found free in the lymphatic spaces, but the greater
ii'imber are undoubtedly contained within the cells. They are

302 LEPROSY

also found in spindle-shaped connective-tissue cells, in endothelial
cells, and in the walls of blood vessels. They are for the most
part confined to the connective tissue, but a few may be
seen in the hair follicles and glands of the skin. Occasionally
one or two may be found in the surface epithelium, where they
probably have been carried by leucocytes, but this position is, on
the whole, exceptional. They also occur in large numbers in the
lymphatic glands associated with the affected parts. In the
internal organs, — liver, spleen, etc., — when leprous lesions are
present, the bacilli are also found, though in relatively smaller
numbers. In the nerves in the anaesthetic form they are com-
paratively few, and in the sclerosed parts it may be impossible to
find any. There are few also in the skin patches referred to
above as occurring in this form of the disease.

Their spread is chiefly by the lymphatics, though distribution
by the blood stream also occurs. They are said to have been
found in the blood during the presence of fever and the eruption
of fresh nodules, and they have also been observed in the blood
vessels post mortem, chiefly contained within leucocytes. Recent
observations (e.g. those of Doutrelepont and Wolters) show that
the bacilli may be more widely spread throughout the body than
was formerly supposed. A few may be detected in some cases
in various organs which show no structural change, especially in
the capillaries. The brain and spinal cord are almost exempt,
but in some cases bacilli have been found even within nerve
cells.

Eelations to the Disease. — Attempts to obtain pure cultures
of the leprosy bacillus have so far been unsuccessful. Clegg has
recently attempted to grow the organism in association with
amoebae and other bacteria on agar plates, and has obtained a
short acid-fast bacillus which does not grow on ordinary media,
and which has been carried through several generations in the
conditions mentioned. The identity of this organism with the
leprosy bacillus has, however, not been established. Attempts
to transfer the disease to animals, including monkeys, have been
unsuccessful. When a small portion of leprous material is
transplanted, a nodule may result in which leprosy bacilli may be
demonstrated for some time, but this probably represents merely
the reaction to a foreign body ; there is no sufficient evidence
that the bacilli undergo multiplication, and it is impossible to
continue such lesions in fresh animals. The only exception to
this statement is afforded by the experiments of Melcher and
Orthmann, who inoculated the anterior chamber of the eye of
rabbits with leprous material, the inoculation being followed by

RELATIONS TO THE DISEASE 303

an extensive growth of nodules in the lungs and internal organs,
which they affirmed contained leprosy bacilli. It has been
questioned, however, by several authorities whether the organisms
in the nodules were really leprosy bacilli, and up to the present
we cannot say that there is any satisfactory proof that the
disease can be transmitted to any of the lower animals. Diph-
theroid bacilli of more than one variety have been cultivated
from the blood and tissues of leprous patients by Babes and
others. Their presence would appear to be by no means in-
frequent, but it is not possible to say at present what their
significance is.

It is interesting to note that a disease occurs under natural
conditions in rats which presents many points of close similarity
to leprosy. It is very widespread, having been observed in
Europe, Asia, America, and Australia ; an excellent description
has been given by Dean. In this affection there are lesions
in the skin which resemble those in leprosy, and the cells con-
tain enormous numbers of an acid-fast bacillus. The disease
can be transmitted to rats by inoculation with the tissue juices
containing the bacilli, but not to animals of other species. All
attempts to cultivate the characteristic organism outside the
body have failed, but Dean has obtained a diphtheroid bacillus
— a result of interest in relation to what has been found in
leprosy. Whether this disease has any relation to leprosy in the
human subject is very doubtful, but the facts which have been
ascertained may prove of high importance in connection with
the pathology of the latter disease.

It would also appear that the disease is not readily inoculable
in the human subject. In a wrell-known case described by Arning,
a criminal in the Sandwich Islands was inoculated in several
parts of the body with leprosy tissue. Two or three years later,
well-marked tubercular leprosy appeared, and led to a fatal result.
This experiment, however, is open to the objection that the
individual before inoculation had been exposed to infection in a
natural way, having been frequently in contact with lepers. In
other cases, inoculation experiments on healthy subjects and
inoculations in other parts of leprous individuals have given
negative results. It has been supposed by some that the failure
to obtain cultures and to reproduce the disease experimentally
may be partly due to the bacilli in the tissues being dead. That
many of the leprous bacilli are in a dead condition is quite
possible, in view of the long period during which dead tubercle
bacilli introduced into the tissues of animals retain their form
and staining reaction. There is also the fact that from time to

304 LEPROSY

time in leprous subjects there occur febrile attacks, which are
followed by a fresh outbreak of nodules, and it would appear
that especially at these times multiplication of the bacilli takes
place more actively.

The facts stated with regard to cultivation and inoculation
experiments go to distinguish the leprosy bacillus all the more
strongly from other organisms. Some have supposed that leprosy
is a form of tubercle, or tubercle modified in some way, but for
this there appears to us to be no evidence. It should also be
mentioned that tubercle is a not uncommon complication in
leprous subjects, in which case it presents the ordinary characters.
It has been found that, a considerable proportion of lepers react
to tuberculin like tubercular patients. This result has been
variously interpreted, some considering that tuberculosis is also
present in such cases, whilst others maintain that the reaction
may be given in the absence of tubercle. If, as is probable, the
latter is the case, the result most likely depends on the close
relationship of the organisms of the two diseases ; it by no means
proves their identity. Another curious fact is that the Wasser-
mann reaction (p. 131) may be given by the serum of leprous
patients (in about 50 per cent., according to some observers) ;
this would seem to be quite independent of the concurrent
presence of syphilis, but it is not possible at present to give an
explanation of the phenomenon.

The mode by which leprosy is transmitted has been the subject
of great controversy, and is one on which authorities still hold
opposite opinions. Some consider that it is a hereditary disease,
or at least that it is transmitted from a parent to the offspring ;
others, again, that it is transmitted by direct contact. There
appears to be no doubt, however, that on the one hand leprous
subjects may bear children free from leprosy, and that on the
other hand healthy individuals entering a leprous district may
contract the disease, though this rarely occurs. Of the latter
occurrence there is the well-known instance of Father Damien,
who contracted leprosy after going to the Sandwich Islands. In
view of all the facts, there can be little doubt that leprosy in
certain conditions may be transmitted by direct contact, though
its contagiousness is not of a high order.

Methods of Diagnosis. — Film preparations should be made
with the discharge from any ulcerated nodule which may be
present, or from the scraping of a portion of excised tissue, and
should be stained as above described. The presence of large
numbers of bacilli situated within the cells and giving the staining
reaction of leprosy bacilli, is conclusive. It is more satisfactory,

METHODS OF DIAGNOSIS 305

however, to make microscopic sections through a portion of the
excised tissue, when the structure of the nodule and the arrange-
ment of the bacilli can be readily studied. The points of differ-
ence between leprosy and tubercle have already been stated, and
in most cases there is really no difficulty in distinguishing the
two conditions.

CHAPTER XII.

GLANDERS AND RHINOSCLEROMA.

GLANDERS.

THE bacillus of glanders (bacillus mallei ; Fr., bacille de la morve ;
Ger., Rotzbacillus) was discovered by Loffler and Schutz, the
announcement of this discovery being made towards the end
of 1882. They not only obtained pure cultures of this organism
from the tissues in the disease, but by experiments on horses
and other animals conclusively established its causal relationship.
These have been fully confirmed. The same organism has also
been cultivated from the disease in the human subject, first by
Weichselbaum in 1885, who obtained it from the pustules in a
case of acute glanders in a woman, and by inoculation of animals
obtained results similar to those of Loffler and Schutz.

Within more recent times a substance, mallein, has been
obtained from the cultures of the glanders bacillus by a method
similar to that by which tuberculin was prepared, and has been
found to produce effects in animals suffering from glanders corre-
sponding to those produced by tuberculin in tuberculous animals.

The Natural Disease. — Glanders chiefly affects the equine
species — horses, mules, and asses. Horned cattle, on the other
hand, are quite immune, whilst goats and sheep occupy an inter-
mediate position, the former being rather more susceptible and
occasionally suffering from the natural disease. It also occurs
in some of the carnivora — cats, lions and tigers in menageries,
which animals are infected from the carcasses of animals affected
with the disease. Many of the small rodents are highly sus-
ceptible to inoculation (vide infra).

Glanders is also found in man as the result of direct inocula-
tion on some wound of the skin or other part by means of the
discharges or diseased tissues of an animal affected, and hence is
commonest amongst grooms and others whose work brings them
into contact with horses ; even amongst them it is a comparatively
rare disease.

306

THE NATURAL DISEASE 307

In horseS the lesions are of two types, to which the names "glanders "
proper and "farcy" have been given, though both may exist together.
In glanders proper, the septum nasi and adjacent parts are chiefly atfected,
there occurring in the mucous membrane nodules which are at first firm
and of somewhat translucent grey appearance. The growth of these is
attended usually by inflammatory swelling and profuse catarrhal dis-
charge. Afterwards the nodules soften in the centre, break down, and
give rise to irregular liberations. Similar lesions, though in less degree,
may be found in the respiratory passages. Associated with these lesions
there is usually implication of the lymphathic glands in the neck, media-
stinum, etc. ; and there may be in the lungs, spleen, liver, etc., nodules
of the size of a pea or larger, of greyish or yellow tint, firm or somewhat
softened in the centre, and often surrounded by a congested zone. The
term "farcy" is applied to the affection of the superficial lymphatic
vessels and glands, which is specially seen where infection takes place
through an abrasion of the skin, such as is often produced by the rubbing
of the harness. The lymphatic vessels become irregularly thickened, so
as to appear like knotted cords, and the associated lymphatic glands
become enlarged and firm, though suppurative softening usually follows,
and there may be ulceration. These thickenings are often spoken of as
" farcy buds " and " farcy pipes." In farcy, also, secondary nodules may
occur in internal organs and the nasal mucous membrane. The disease
is often present in a " latent form," and its presence can only be detected
by the mallein test (vide infra). In the ass the disease runs a more
acute course than in the horse.

In man the disease is met with in two forms, an acute and a
chronic — though intermediate forms also occur, and chronic cases
may take on the characters of the acute disease. The site of
inoculation is usually on the hand or arm, — by means of some
scratch or abrasion, or possibly by infection along a hair follicle, —
sometimes on the face, and occasionally on the mucous membrane
of the mouth, nose, or eye. In the acute form there appears at
the site of inoculation inflammatory swelling, attended usually with
spreading redness, and the lymphatics in relation to the part also
become inflamed, the appearances being those of a " poisoned
wound." These local changes are soon followed by marked
constitutional disturbance, and by a local or widespread eruption
on the surface of the body, at first papular and afterwards
pustular, and later there may form in the subcutaneous tissue
and muscles larger masses which soften and suppurate, the pus
being often mixed with blood ; suppuration may occur also in the
joints. In some cases the nasal mucous membrane may be
secondarily infected, and thence inflammatory swelling may
spread to the tissues of the face. The patient usually dies in
two or three weeks, sometimes sooner, with the symptoms of
rapid pyaemia. In addition to the lesions mentioned, there may
be foci, usually suppurative, in the lungs (attended often with
pneumonic consolidation), in the spleen, liver, bone-marrow,

308 GLANDERS

salivary glands, etc. In the chronic form a local granulomatous
condition may occur, which usually breaks clown and gives rise to
the formation of an irregular ulcer with thickened margins, and
sanious, often foul, discharge. The ulceration spreads deeply as
well as superficially, and the thickened lymphatics also have a
great tendency to ulcerate, though the lymphatic system is not
so prominently affected as in the horse. Deposits may form in
the subcutaneous tissue and muscles, and the mucous membrane
may become affected. The disease may run a very chronic
course, lasting for months or even years, and recovery may

occur ; on the other
hand, such a case may
at any time take on the
characters of the acute
form of the disease and
rapidly become fatal.

/ / The Glanders Bacil-

lus. — Microscopical
Characters. — The glan-
ders bacilli are minute
rods, straight or slightly
curved, with rounded
ends, and about the same
length as tubercle bacilli,
but distinctly thicker

(Fig. 89). They show,

FIG. 89.-Glanders bacilli, -several con- however, _ considerable
tained within leucoytes,— from peritoneal variations in size and in
exudate in a guinea-pig appearance, and their

Stained with weak carbol-fuehsm. x 1000. , .,

protoplasm is otten

broken up into a num-
ber of deeply-stained portions with unstained intervals between.
These characters are seen both in the tissues and in cultures,
but, as in the case of many organisms, irregularities in form and
size are more pronounced in cultures (Fig. 90) ; short filamentous
forms 8 to 12 /x in length .are sometimes met with, but these are
on the whole rare. The organism is non-motile and does not
form spores, though some have considered certain of the non-
staining portions of the protoplasm to be of that nature.

Tn the tissues the bacilli usually occur irregularly scattered
amongst the cellular elements ; a few may be contained within
leucocytes and connective-tissue corpuscles, but the position of
most is extracellular. They are most abundant in the acute
lesions, in which they may be found in considerable numbers ;

CULTIVATION OF GLANDERS BACILLUS 309

but in the chronic nodules, especially when softening has taken
place, they are few in number, and it may be impossible to find
any in sections.

Staining. — The glanders bacillus differs widely from the
tubercle bacillus in its staining reactions. It stains with simple
watery solutions of the basic stains, but somewhat faintly (better
when an alkali or a mor-
dant, such as carbolic
acid, is added), and even
when deeply stained it
readily loses the colour
when a decolorising agent
such as alcohol is applied.
We have obtained the
best results by carbol-
thionin-blue (p. 105), and
we prefer to dehydrate
by the aniline-oil method.
In film preparations of
fresh glanders nodules
the bacilli can be readily
found by staining with
any of the ordinary com-
binations, e.g., carbol-
thionin-blue or weak car-
bol-fuchsin. By using a
stain of suitable strength

no decolorising agent is necessary, the film being simply washed
in water, dried, and mounted.

McFadyean recommends that after sections have been stained in
Lbffler's methylene-blue and slightly decolorised in weak acetic acid, they
should be treated for fifteen minutes with a saturated solution of tannic
acid ; thereafter they are washed thoroughly in water, and as a contrast
stain a 1 per cent, solution of acid fuchsin may be applied for half a
minute ; they are then dehydrated, cleared, and mounted. Gram's
method is quite inapplicable, the glanders bacilli rapidly losing the stain
in the process.

Cultivation. — (For the methods of separation, vide infra.)
The glanders bacillus grows readily on most of the ordinary
media, but a somewhat high temperature is necessary, growth
taking place most rapidly at 35° to 37° C. Though a certain
amount of growth occurs down to 21° C., a temperature above
25° C. is always desirable.

On ayar and ylycerin-ayar in stroke cultures growth appears

FIG. 90.— Glanders bacilli, from a pure
culture on glycerin agar. Stained with
car bol -fuchsin and partially decolorised
to show segmentation of protoplasm.
xlOOO.

310 GLANDERS

along the line as a uniform streak of greyish-white colour arid
somewhat transparent appearance, with moist-looking surface,
and when touched with a needle is found to be of rather slimy
consistence. Later it spreads laterally for some distance, and
the layer becomes of slightly brownish tint.

On serum the growth is somewhat similar but more transparent,
the separate colonies being in the form of round and almost clear
drops. In sub-cultures on these media at the body temperature
growth is visible within twenty -four hours, but when fresh cultures
are made from the tissues it may not be visible till the second day.
Serum or potato, however, is much more suitable for cultivating
from the tissues than the agar media ; on the latter it is some-
times difficult to obtain growth.

In broth, growth forms at first a uniform turbidity, but soon
settles to the bottom, and after a few days forms a pretty thick
flocculent deposit of slimy and somewhat tenacious consistence.

On potato at 30° to 37° C. the glanders bacillus nourishes
well and produces a characteristic appearance, incubation at a
high temperature, however, being necessary. Growth proceeds
rapidly, and on the third day has usually formed a transparent
layer of slightly yellowish tint, like clear honey in appearance.
On subsequent days, the growth still extends and becomes darker
in colour and more opaque, till about the eighth day it has a
reddish-brown or chocolate tint, while the potato at the margin
of the growth often shows a greenish-yellow staining. The
characters of the growth on potato, along with the microscopical
appearances, are quite sufficient to distinguish the glanders
bacillus from every other known organism (sometimes the
cholera organism and the b. pyocyaneus produce a somewhat
similar appearance, but they can be readily distinguished by
their other characters). Potato is also a suitable medium for
starting cultures from the tissues; in this case minute trans-
parent colonies become visible on the third day, and afterwards
present the appearances just described.

Powers of Resistance. — The glanders bacillus is not killed at
once by drying, but usually loses its vitality after fourteen days
in the dry condition, though sometimes it lives longer. It is
not quickly destroyed by putrefaction, having been found to be
still active after remaining two or three weeks in putrefying
fluids. In cultures the bacilli retain their vitality for three or
four months, if, after growth has taken place, they are kept at
the temperature of the room ; on the other hand, they are often
found to be dead at the end of two weeks when kept constantly
a.t the body temperature. They have comparatively feeble

EXPERIMENTAL INOCULATION 311

resistance to heat and antiseptics. Loffler found that they were
killed in ten minutes in a fluid kept at 55° C., and in from two
to three minutes by a 5 per cent, solution of carbolic acid.
Boiling water and the ordinarily used antiseptics are very rapid
and efficient disinfectants.

We may summarise the characters of the glanders bacillus by
saying that in its morphological characters it resembles some-
what the tubercle bacillus, but is thicker, and differs widely
from it in its staining reactions. For its cultivation the higher
temperatures are necessary, and the growth on potato presents
most characteristic features.

Experimental Inoculation. — In horses, subcutaneous injection
of the glanders bacillus in pure culture reproduces all the
important features of the disease. This fact was established at
a comparatively early date by Loffler and Schutz, who, after one
doubtful experiment, successfully inoculated two horses in this
way, the cultures used having been grown for several generations,
outside the body. In a few days swellings formed at the sites
of inoculation, and later broke down into unhealthy-looking
ulcers. One of the animals died ; after a few weeks, the other,
showing symptoms of cachexia, was killed. In both animals, in
addition to ulcerations on the surface with involvement of the
lymphatics, there were found, post-mortem, nodules in the lungs,
softened deposits in the muscles, and also affection of the nasal
mucous membrane, — nodules, and irregular ulcerations. The
ass is even more susceptible than the horse, the disease in the
former running a more rapid course, but with similar lesions.
The ass can be readily infected by simple scarification and
inoculation with glanders secretion, etc. (Nocard).

Of small animals, field-mice and guinea-pigs are the most
susceptible; on the other hand, house-mice and white mice
enjoy an almost complete immunity. In field-mice, subcutaneous
inoculation is followed by a very rapid disease, usually leading
to death within eight days, the organisms becoming generalised
and producing numerous minute nodules, especially in the spleen,
lungs, and liver. In the guinea-pig the disease is less acute.
At the site of inoculation an inflammatory swelling forms, which
soon softens and breaks down, leading to the formation of an
irregular crateriform ulcer with indurated margins. The lym-
phatic vessels become infiltrated, and the corresponding lymphatic
glands become enlarged to the size of peas or small nuts, softened,
and semi-purulent. The animal sometimes dies in two or three
weeks, sometimes not for a longer period. Secondary nodules,
in varying numbers in different cases, may be present in the

312 GLANDERS

spleen, lungs, bones, nasal mucous membrane, testicles, ovaries,
etc. ; in some cases a few nodules are found in the spleen alone.
Intraperitoneal injection in the male guinea-pig is followed, as
pointed out by Straus, by a very rapid and semi-purulent
affection of the tunica vaginalis, shown during life by great
swelling and redness of the testicles, which changes may be
noticeable in two or three days, or earlier if material from man
has been used. This method of inoculation has been found of
service for purposes of cultivation and diagnosis. Rabbits are
less susceptible than guinea-pigs, and the effect of subcutaneous
inoculation is somewhat uncertain. Accidental inoculation of the
human subject with pure cultures of the bacillus has in more than
one instance been followed by the acute form of the disease and
a fatal result. •'

Mayer has found that when the glanders bacillus is injected along
with melted butter into the peritoneum of a guinea-pig, it shows
filamentous, branching, and club-shaped forms ; in other words, it
presents the characters of a streptothrix. Lubarsch, on the other hand,
in a comparative study of the results of inoculation with acid-fast and
other bacilli, found none of the above characters in the case of the
glanders bacillus (cf. Tubercle, p. 263).

Action on the Tissues. — From the above facts it will be seen
that in many respects glanders presents an analogy to tubercle
as regards the general characters of the lesions and the mode
of spread. When the tissue changes in the two diseases are
compared, certain differences are found. The glanders bacillus
causes a more rapid and more marked inflammatory reaction.
There is more leucocytic infiltration and less proliferative change
which might lead to the formation of epithelioid cells. Thus
the centre of an early glanders nodule shows a dense aggregation
of leucocytes, most of which are poly morpho-nucl ear, whilst in
the central parts many show fragmentation of nuclei with the
formation of a deeply staining granular detritus. And further,
the inflammatory change may be followed by suppurative
softening of the tissue, especially in certain situations, such as
the subcutaneous tissue and lymphatic glands. The nodules,
therefore, in glanders, as Baumgarten puts it, occupy an
intermediate position between miliary abscesses and tubercles.
The diffuse coagulative necrosis and caseation which are so
common in tubercle do not occur to the same degree in glanders,
and typical giant cells are not formed. The nodules in the lungs
show leucocytic infiltration and thickening of the alveolar walls,
whilst the vesicles are filled with catarrhal cells ; there may also
be fibrinous exudation, whilst at the periphery of the nodules con-

MODE OF SPREAD 313

nective-tissue growth is present in proportion to their age. The
tendency to spread by the lymphatics is always a well-marked
feature, and when the bacilli gain entrance to the blood stream,
they soon settle in the various tissues and organs. Accordingly,
even in acute cases it is usually quite impossible to detect the
bacilli in the circulating blood, though sometimes they have been
found. It is an interesting fact, shown by observations of the
disease both in the human subject and in the horse, as well as
by experiments on guinea-pigs, that the mucous membrane of the
nose may become infected by means of the blood stream — another
example of the tendency of organisms to settle in special sites.

Mode of Spread. — Glanders usually spreads from a diseased
animal by direct contagion with the discharge from the nose or
from the sores, etc. So far as infection of the human subject
goes, no other mode is known. There is no evidence that the
disease is produced in man by inhalation of the bacilli in the
dried condition. Some authorities consider that pulmonary
glanders may be produced in this way in the horse, whilst others
maintain that in all cases there is first a lesion of the nasal
mucous membrane or of the skin surface, and that the lung is
affected secondarily. Babes, however, found that the disease
could be readily produced in susceptible animals by exposing
them to an atmosphere in which cultures of the bacillus had
been pulverised. He also found that inunction of the skin
with vaseline containing the bacilli might produce the disease,
the bacilli in this case entering along the hair follicles.

Serum Reactions. — Shortly after the discovery of agglutination in
typhoid fever, McFadyean showed that the serum of glandered horses
possessed the power of agglutinating glanders bacilli. His later observa-
timis show that in the great majority of cases of glanders a 1 : 50
dilution of the serum produces marked agglutination in a few minutes,
whilst in the great majority of non -glandered animals no effect is
produced under these conditions. The test performed in the ordinary
\\.iy is, however, not absolutely reliable, as exceptions occasionally occur
in both directions, i.e. negative results by glandered animals and positive
results by non-glandered animals. He found that a more delicate and
reliable method is to grow the bacillus in bouillon containing a small
proportion of the serum to be tested. In this way he obtained a distinct
sediuienting reaction with a serum which did not agglutinate at all
distinctly in the ordinary method. Within recent times the sedimenta-
tion test by the ordinary method (p. 120) has been most generally used.
The general result seems to be that distinct sedimentation within thirty-
six hours with a serum dilution of 1 : 1000 may be taken as a positive
result, indicating the presence of glanders ; whilst reactions with dilutions
ln'twcen this and 1 : 500 are highly suspicious but not conclusive. The
deviation of complement test (p. 130) is also applicable in the case of
glanders, and this has given valuable results in the hands of various
observers ; a precipitin reaction mny also be obtained on the addition of

314 GLANDERS

mallein to the serum of a glandered animal. These reactions, which of
course depend on the presence of anti-substance in the blood in glanders,
form important auxiliaries to the method of diagnosis by means of
mallein.

Mallein and its Preparation.— Mallein is obtained from cultures of the
glanders bacillus grown for a suitable length of time, and, like tuber-
culin, is really a mixture comprising (1) substances in the bodies of the
bacilli, and (2) their soluble products, not destroyed by heat, along witli
substances derived from the medium of growth. It was at first obtained
from cultures on solid media by extracting with glycerin or water, but is
now usually prepared from cultures in glycerin bouillon. Such a culture,
after being allowed to grow for three or four weeks, is sterilised by heat
either in the autoclave at 115° C. or by steaming at 100° C. on successive
days. It is then filtered through a Chamberland filter. The filtrate
constitutes fluid mallein. Usually a little carbolic acid ('5 per cent.) is
added to prevent it from decomposing. Of such mallein 1 c.c. is usually
the dose for a horse (McFadyean). Foth has prepared a dry form of
mallein by throwing the filtrate of a broth culture, evaporated to one-
tenth of its bulk, into twenty-five or thirty times its volume of alcohol.
A white precipitate is formed, which is dried over calcium chloride and
then under an air-pump. A dose of this dry mallein is '05 to '07 grm.

The Use of Mallein as a Means of Diagnosis. — In using mallein for the
diagnosis of glanders, the temperature of the animal ought to be observed
for some hours beforehand, and, after subcutaneous injection of a suitable
dose, it is taken at definite intervals, — according to McFadyean at the
sixth, tenth, fourteenth, and eighteenth hours afterwards, and on the
next day. Here both the local reaction and the temperature are of
importance. In a glandered animal, at the site of inoculation there is a
somewhat painful local swelling, which reaches a diameter of five inches
at least, the maximum size not being attained until twenty-four hours
afterwards. The temperature rises 1'5° to 2° C., or more, the maximum
generally occurring in from eight to sixteen hours. If the temperature
never rises as much as 1*5°, the reaction is considered doubtful. In the
negative reaction given by an animal free from glanders, the rise of
temperature does not usually exceed 1°, the local swelling reaches the
diameter of*three inches at most, and has much diminished at the end
of twenty-four hours. In the case of dry mallein, local reaction is less
marked. Veterinary authorities arc practically unanimous as to the
great value of the mallein test as a means of diagnosis. It has recently
been shown that mallein instilled into the conjunctival sac, or inoculated
by scarification into the skin of glandered animals, gives corresponding
reactions to the ophthalmic and cutaneous tuberculin reactions in cases
of tuberculosis (pp. 285, 286) ; in the case of glanders the conjunctival
reaction would appear to be the more convenient and reliable.

Methods of Examination. — Microscopic examination in a
case of suspected glanders will at most reveal the presence of
bacilli corresponding in their characters to the glanders bacillus.
An absolute diagnosis cannot be made by this method. Cultures
may be obtained by making successive strokes on blood serum or
on glycerin agar (preferably the former), and incubating at 37° C.
The colonies of the glanders bacillus do not appear till two days
after. This method often fails unless a considerable number of

RHINOSCLEROM A 3 1 5

the glanders bacilli are present. Another method is to dilute the
-••(Tction or pus with sterile water, to varying degrees, and then
to smear the surface of potato with the mixture, the potatoes
being incubated at the above temperature. The colonies on
potatoes may not appear till the third day. The most certain
method, however, is by inoculation of a guinea-pig, either by
subcutaneous or intraperitoneal injection. By the latter method,
as above described, lesions are much more rapidly produced, and
are more characteristic. If, however, there have been other
organisms present, the animal may die of a septic peritonitis,
though even in such a case the glanders bacilli will be found to
be more numerous in the tunica vaginalis, and may be cultivated
from this situation. It is extremely doubtful whether the
application of mallein to diagnosis of the disease in the human
subject is justifiable. There is a certain risk that it may lead
to the lesions assuming a more acute character; moreover,
culture and inoculation tests are generally available. In the
case of horses, etc., a diagnosis will, however, be much more
easily and rapidly effected by means of mallein, or by one of
the scrum reactions described above. In some cases of acute
glanders in the human subject the bacillus has been obtained in
cultures from the blood during life.

RHINOSCLEROMA.

'I'll is disease is considered here as, from the anatomical
changes, it also belongs to the group of infective granulomata.
It is characterised by the occurrence of chronic nodular
thickenings in the skin or mucous membrane of the nose, or
in the mucous membrane of the pharynx, larynx, or upj)er part
of the trachea. The nodules are of considerable size, sometimes
as large as a pea ; in the earlier stages they are comparatively
smooth on the surface, but later they become shrunken and the
centre is often retracted. The disease is scarcely ever met with
in this country, but is of not very uncommon occurrence on the
Continent, especially in Austria and Poland. In the granulation
tissue of the nodules there are to be found numerous round and
rather large cells, which have peculiar characters and are often
known as the cells of Mikulicz. Their protoplasm contains a
collection of somewhat gelatinous material which may fill the
cell and push the nucleus to the side. Within these cells there
is present a characteristic bacillus, occurring in little clumps or
masses chiefly in the gelatinous material. A few bacilli also lie
free in the lymphatic spaces around. This organism was first

316 GLANDERS

observed by Frisch, and is now known as the bacillus of
rhinoscleroma. The bacilli have the form of short oval rods,
which, when lying separately, can be seen to possess a distinct
capsule, and which in all their microscopical characters correspond
closely with Friedlander's pneumobacillus. They are usually pre-
sent in the lesions in a state of purity. It was at first stated that
they could be stained by Gram's method, but more recent obser-
vations show that, like Friedlander's organism, they lose the stain.

From the affected tissues this bacillus can be easily cultivated
by the ordinary methods. In the characters of its growth in
the various culture media it presents a close similarity to that
of the pneumobacillus, as it also does in its fermentative action
in milk and sugar-containing fluids. The nail-like appearance
of the growth on gelatin is said to be less distinct, and the
growth on potatoes is more transparent and may show small
bubbles of gas ; but it is doubtful whether any distinct line of
difference can be drawn between the two organisms so far as
their microscopical and cultural characters are concerned.

The evidence that the organisms described are the cause of
this disease consists in their constant presence and their special
relation to the affected tissues, as already described. From
these facts alone it would appear probable that they are the
active agents in the production of the lesions. Experimental
inoculation has thrown little light on the subject, though one
observer has described the production of nodules on the con-
junctive of guinea-pigs. The relation of the rhinoscleroma
organism to that of Friedlander is, however, still a matter of
doubt, and the matter has been further complicated by the fact
that a bacillus possessing closely similar characters has been
found to be very frequently present in ozcena, and is often
known as the bacillus ozcenoe. The last-mentioned organism is
said to have more active fermentative powers. From what has
been stated it will be seen that a number of organisms, closely
allied in their morphological characters, have been found in the
nasal cavity in healthy or diseased conditions. There is no
doubt that rhinoscleroma is a specific disease with well-marked
characters, and it is quite possible that one member of this
group of organisms may be the causal agent, though indis-
tinguishable from others by culture tests. There is, however,
a tendency on the part of recent investigators, e.g. Perkins, to
consider the "bacillus of rhinoscleroma" to be identical with
the pneumobacillus, and its presence in the affected tissues to
represent merely a secondary invasion. The subject is one on
which more light is still required,

CHAPTER XIII.

ACTINOMYCOSIS AND ALLIED DISEASES.

ACTINOMYCOSIS is the most important example of a group of
diseases produced by streptothrix organisms. It is related, by
the characters of the pathological changes, to the diseases which
have been described. The disease affects man in common with
certain of the domestic animals, though it is more frequent in
the latter, especially in oxen, swine, and horses. The parasite
was first discovered in the ox by Bollinger, and described by
him in 1877, the name actinomyces or ray fungus being from its
appearance applied to it by the botanist Harz. In 1878 Israel
described the parasite in the human subject, and in the following
year Ponfick identified it as being the same as that found in the
ox. Since that time a large number of cases have been observed
in the human subject, the result of investigation being to show
that it affects man much more frequently than was formerly
supposed.

It is, however, to be noted that the term "actinomyces," as
originally used, does not represent one parasite but a number of
closely allied species, as cultures obtained from various sources
have presented considerable differences ; and, further, it is noted
that other distinct species of streptothrix have been cultivated
from isolated cases of disease in the human subject where the
lesions resembled more or less closely those of actinomycosis.
In one or two instances the organism has been found to be
" acid-fast," and there is no doubt that the actinomyces group
is closely related through intermediate forms with the tubercle
group (vide p. 264).

Naked-Eye Characters of the Parasite. — The actinomyces
ijrows in the tissues in the form of little round masses or colonies,
which, when fully develo^d, are easily visible to the naked eye,
the largest being about the size of a small pin's head, whilst all
si/.-s below this may be found. When suppuration is present,
they lie free in the pus ; when there is no suppuration, they are

317

318 ACTINOMYCOSIS AND ALLIED DISEASES

embedded in the granulation tissue, but are usually surrounded
by a zone of softer tissue. They may be transparent or jelly-
like, or they may be opaque and of various colours — white,
yellow, greenish, or almost black. The appearance depends
upon their age and also upon their structure, the younger colonies
being more or less transparent, the older ones being generally
opaque. Their colour is modified by the presence of pigment
and by degenerative change, which is usually accompanied by a
yellowish coloration. They are generally of soft, sometimes
tallow-like, consistence, though sometimes in the ox they are
gritty, owing to the presence of calcareous deposit. They may
be readily found in the pus by spreading it out in a thin layer
on a glass slide and holding it up to the light. They are some-
times described as being always of a distinctly yellow colour,
but this is only occasionally the case; in fact, in the human
subject they occur much more frequently as small specks of
semi-translucent appearance, and of greenish-grey tint.

Microscopical Characters. — The parasite, which is now
generally regarded as belonging to the streptothrix group of the
higher bacteria (p. 16), presents pleomorphous characters. In
the colonies, as they grow in the tissues, three morphological
elements may be described, namely, filaments, coccus-like bodies,
and clubs.

1. The filaments are comparatively thin, measuring about
•6 /A in diameter, but they are often of great length. They are
composed of a central protoplasm enclosed by a sheath. The
latter, which is most easily made out in the older filaments with
granular protoplasm, occasionally contains granules of dark
pigment. In the centre of the colony the filaments interlace
with one another, and form an irregular network which may be
loose or dense ; at the periphery they are often arranged in a
somewhat radiating manner, and run outwards in a wavy or even
spiral course. They also show true branching, a character
which at once distinguishes them from the ordinary bacteria.
Between the filaments there is a finely granular or homogeneous
ground substance. Most of the colonies at an early stage are
chiefly constituted by filaments loosely arranged ; but later, part
of the growth may become so dense that its structure cannot be
made out. This dense part, starting excentrically, may grow
round the colony to form a hollow sphere, from the outer
surface of which filaments radiate for a short distance (Fig. 91).
The filaments usually stain uniformly in the younger colonies,
but some, especially in the older colonies, may be segmented so
as to give the appearance of a chain of bacilli or of cocci, though

CHARACTERS OF THE ACTINOMYCES

319

the sheath enclosing them may generally be distinguished. Rod-
shaped and spherical forms may also be seen lying free.

2. Spores or Gonidia. — Like other species of streptothrix, the
actinomyces when growing on a culture medium shows on its
surface filaments growing upwards in the air, the protoplasm of
which becomes segmented into rounded spores or gonidia. In
natural conditions outside the body these gonidia become free,
and act as new centres by growing out into filaments. They

Flu. 91. — Actinomycosis of human liver, showing a colony of the
parasite composed of a felted mass of filaments surrounded by pus.
Paraffin section ; stained by Gram's method and with
safranin. x 500.

have somewhat higher powers of resistance than the filaments,
though less than the spores of most of the lower bacteria. An
exposure to 75° C. for half an hour is sufficient to kill most
streptothrices or their spores ; cultures containing spores can
resist a temperature from five to ten degrees higher than spore-
free cultures (Foulerton). It is probable that some of the
spherical bodies formed within filaments when growing in the
tissues have the same significance, i.e. are gonidia, whilst others
may be merely the result of degenerative change. Both the

320 ACTINOMYCOSIS AND ALLIED DISEASES

filaments and the spherical bodies are readily stained by Gram's
method.

3. Clubs, — These are elongated pear-shaped bodies which are
seen at the periphery of the colony, and are formed by a sort
of hyaline swelling of the sheath around the free extremity of
a filament (Figs. 92, 93). They are usually homogeneous and
structureless in appearance. In the human subject the clubs are

FIG. 92. — Actmomyces in human kidney, showing clnbs radially
arranged and surrounded by pus. The filaments had practically
disappeared.

Paraffin section ; stained with hsematoxylin and rubin. x 500.

often comparatively fragile structures, which are easily broken
down, and may sometimes be dissolved in water. Sometimes
they are well seen when examined in the fresh condition, but in
hardened specimens are no longer distinguishable. In specimens
stained by Gram's method they are usually not coloured by the
violet, but take readily a contrast stain, such as picric acid,
rubin, etc. ; sometimes a darkly-stained filament can be seen
running for a distance in the centre, and may have a knob-like
extremity. In many of the colonies in the human subject the
clubs are absent. In the ox, on the other hand, where there are

TISSUE LESIONS 321

much older colonies, the clubs constitute the most prominent
feature, whilst in most colonies the filaments are more or less
degenerated, and it may sometimes be imi>ossible to find any.
They often form a dense fringe around the colony, and when
stained by Gram's method retain the violet stain. They have,
in fact, undergone some further chemical change which produces
the altered staining reaction. Occasionally in very chronic

FIG. 93. — Colonies of actinomyces, showing general structural
arrangement and clubs at periphery. From pus in human subject.
Stained Gram and safranin. x 60.

lesions in the human subject the clubs stain with Gram's
method. Clubs showing intermediate staining reaction have
been described in the ox by McFadyean. The club-formation
probably represents a means of defence on the part of the
parasite against the phagocytes of the tissue ; the view, formerly
held, that the clubs are organs of fructification has now been
generally abandoned.

Tissue Lesions. — In the human subject the parasite produces
by its growth a chronic inflammatory change, usually ending
in a suppuration which slowly spreads. In some cases there
21

322 ACTINOMYCOSIS AND ALLIED DISEASES

is a comparatively large production of granulation tissue, with
only a little softening in the centre, so that the mass feels solid.
This condition is sometimes found in the subcutaneous tissue,
especially when the disease has not advanced far, and also in
dense fibrous tissue. In most cases, however, and especially
in internal organs, suppuration is the outstanding feature ;
this is associated with abundant growth of the parasite in the
filamentous form. In an organ such as the liver, multiple foci
of suppuration are seen at the spreading margin of the disease,
often presenting a honeycomb appearance, whilst the colonies
of the parasite may be seen in the pus with the naked eye. In
the older parts the abscesses have become confluent, and formed
large areas of suppuration. The pus is usually of greenish-
yellow colour, and of somewhat slimy character.

In cattle the tissue reaction is more of a formative type,
there being abundant growth of granulation tissue, which may
result in large tumour-like masses, usually of more or less
nodulated character, and often consisting of well-developed
fibrous tissue containing areas of younger formation, in which,
however, irregular abscess formation may be present. The cells
immediately around the colonies are usually irregularly rounded,
or may even be somewhat columnar in shape, whilst farther out
they become spindle-shaped and concentrically arranged. It is
not uncommon to find leucocytes or granulation tissue invading
the substance of the colonies, and portions of the parasite, etc.,
may be contained within leucocytes or within small giant-cells,
which are sometimes present. A similar invasion of old colonies
by leucocytes is sometimes seen in human actinomycosis.

Origin and Distribution of Lesions. — The lesions in the
human subject may occur in almost any part of the body, the
paths of entrance being very various. In many cases the
entrance takes place in the region of the mouth — probably
around a decayed tooth — by the crypts of the tonsil, or by
some abrasion. Swelling and suppuration may then follow in
the vicinity and may spread in various directions. The
periosteum of the jaw or the vertebrae may thus become affected,
caries or necrosis resulting, or the pus may spread deeply in
the tissues of the neck, and may even pass into the mediastinum.
Occasionally the parasite may enter the tissues from the
oesophagus, and in a considerable number of cases the primary
lesion is in some part of the intestine, generally of the large
intestine. The parasite penetrates the wall of the bowel, and
may be found deeply between the coats, surrounded by purulent
material. Thence it may spread to the peritoneum or to the

CULTIVATION OF ACTINOMYCES 323

extraperitoneal tissue, the retrocaecal connective tissue and that
around the rectum being not uncommonly seats of suppuration
produced in this way. A peculiar affection of the intestine has
been described, in which slightly raised plaques are found both
in the large and small intestines, these plaques being composed
almost exclusively of masses of the actinomyces along with
epithelial cells. This, however, is a rare condition. The path
of entrance may also be by the respiratory passages, the primary
lesion being pulmonary or peribronchial ; extensive suppuration
in the lungs may result. Infection may also occur by the skin
surface, and lastly, by the female genital tract, as in a case
recorded by Grainger Stewart and Muir, in which both ovaries
and both Fallopian tubes were affected.

When the parasite has invaded the tissues by any of these
channels, secondary or " metastatic " abscesses may occur in
internal organs. The liver is the organ most frequently affected,
though abscesses may occur in the lungs, brain (where a primary
meningitis may also occur), kidneys, etc. In such cases the
spread takes place by the blood stream, and it is possible that
leucocytes may be the carriers of the infection, as it is not
uncommon to find leucocytes in the neighbourhood of a colony
containing small portions of the filaments in their interior.

In the ox, on the other hand, the disease usually remains
quite local, or spreads by continuity. It may produce tumour-
like masses in the region of the jaw or neck, or it may specially
affect the palate or tongue, in the latter producing enlargement
and induration, with nodular thickening on the surface — the
condition known as "woody tongue."

Source of the Parasite. — There is a considerable amount of
evidence to show that outside the body the parasite grows on
grain, especially on barley. Both in the ox and in the pig the
parasite has been found growing around fragments of grain
embedded in the tissues. There are besides, in the case of the
human subject, a certain number of cases in which there was a
history of penetration of a mucous surface by a portion of grain,
and in a considerable proportion of cases the patient has been
< '\ posed to infection from this source. The position of the
lesions in cattle is also in favour of such a view.

Cultivation (for methods of isolation see later). — The descrip-
tions of the cultures obtained by various investigators differ in
essential particulars, and there is no doubt that the organisms
described are different. The following is the account of the
organism as cultivated by Bostrb'm : —

On agar or glycerin ayar at 37° C., growth is generally

324 ACTINOMYCOSIS AND ALLIED DISEASES

visible on the third or fourth day in the form of little trans-
parent drops which gradually enlarge and form rounded projec-
tions of a reddish-yellow tint and somewhat transparent
appearance, like drops of amber. The growths tend to remain
separate, and even when they become confluent, the nodular

character is maintained.
They have a tough con-
sistence, being with diffi-
culty broken up, and
adhere firmly to the sur-
face of the agar. Older
growths often show on
the surface a sort of cor-
rugated aspect, and may
sometimes present the
appearance of having been
dusted with a brownish-
yellow powder (Fig. 94).

In the cultures at an
early stage the growth is
composed of branching
filaments, which stain
uniformly (Fig. 95), but
later some of the super-
ficial filaments may show
segmentation into gonidia.
Slight bulbous thicken-
ings may be seen at the
end of some of the fila-
A B ments, but true clubs have

FIG. 94.— Cultures of the actinomyces on not been observed,
glycerin agar, of about three weeks' growth, Qn gelatin the same
showing the appearances which occur. The , i • r,,i

growth in A is at places somewhat corru- tendency to grow in little
gated on the surface. Natural size. spherical masses is seen,

and the medium becomes

very slowly liquefied. When this occurs the liquefied portion
has a brownish colour and somewhat syrupy consistence, and
the growths may be seen at the bottom, as little balls, from the
surface of which filaments radiate.

The organism obtained in culture by Wolff and Israel (vide
infra} is probably the same as the one which has been recently
described in detail by J. H. Wright, who obtained it in pure
condition from fifteen different cases of the disease. It differs
markedly from Bostrom's organism in being almost a strict

VARIETIES OF ACTINOMYCES

325

anaerobe and in ceasing to grow at a temperature a little below
that of the body. Under ordinary aerobic conditions either no
growth occurs or it is of a very slight character. On the surface
of agar under anaerobic conditions the organism produces dense
rounded colonies of greyish-white colour, which sometimes
assume a rosette form. A somewhat curious feature of growth
is described by Wright, namely, that in a shake culture in
glucose agar the colonies are most numerous and form a dense
zone about half an inch from the surface of the medium, that
is, at a level where there
is presumably a mere
trace of oxygen obtain-
able (Fig. 96). In
bouillon, growth takes
place at the bottom of
the medium in rounded
masses which afterwards
undergo disintegration.
Wright found when the
organism was grown in
the presence of serum or
other animal fluids, that
the formation of true
clubs occurred at the
extremity of some of
the filaments (Fig. 97).
From the conditions
under which growth
occurs, he is inclined to
regard it as a true para-
site, and doubts whether it can have a saprophytic existence out-
side the body, e.g. on grain. He is also of opinion that all cases
of true actinomycosis, i.e. cases where colonies visible to the
naked eye are present, are probably produced by one species, and
that the aerobic organisms obtained by Bostrom and others are
probably accidental contaminations. It is quite evident that
further investigations are required in the light of the results
detailed. Certainly the parasite in many cases of actinomycosis
in the human subject does not grow on ordinary media under
aerobic conditions as Bostrom's organism does.

Varieties of Actinomyces and Allied Forms. — It is probable that in
the cases of the disease described in the human subject there is more than
one variety or species of parasite belonging to the same group. Gasperini
has described several varieties of actinomyces boris according to the colour

95. — Actinomyces, from a culture on
glycerin agar, snowing the branching of
the filaments. See also Plate III., Fig. 10.
Stained with fuchsin. x 1000.

326 ACTINOMYCOSIS AND ALLIED DISEASES

FIG. 96. 1 — Shake cultures of actinomyces in
glucose agar, showing the maximum
growth at some distance from the surface
of the medium.

regarded as a distinct
species. Another species
was cultivated by Ep-
pinger from a brain
abscess, and called by
him ' ' cladothrix aste-
roides," from the appear-
ance of its colonies on
culture media. A case
of general streptothrix
infection in the human
subject described by
Mac Donald was pro-
bably due to the same
organism as Eppinger's.
In the tissues it grows
in a somewhat diffuse
manner, and does not

of the growths, and a similar
condition may obtain in the
case of the human subject.
Furthermore, a considerable
number of strcptothrices
have been found in cases of
disease in the human sub-
ject, the associated lesions
varying in character from
tubercle-like nodules on the
one hand to suppurative
processes on the other. The
organisms cultivated from
such sources differ accord-
ing to their microscopic
characters • (for example,
some form "clubs" whilst
others do not) according to
their conditions of growth,
staining reactions, etc. Of
these only a few examples
may here be mentioned, but
it may be noted that the
importance of the strepto-
th rices as causes of disease
is constantly being ex-
tended. Wolff and Israel
cultivated from two cases
of " actinomycosis " in man
a streptothrix which differs
in so many important points
from the actinomyces of
Bostrom that it is now

FIG. 97. — Section of a colony of actinomyces
from a culture in blood serum, showing the
formation of clubs at the periphery, x 1500.

i For Figs. 96 and 97 we are indebted to Dr. J. Homer Wright of
Boston, U.S.A.

METHODS OF EXAMINATION AND DIAGNOSIS 327

form clubs ; in rabbits and guinea-pigs it produces tubercle-like lesions.
Flexner observed a streptothrix in the lungs associated with lesions some-
what like a rapid phthisis, and applied the name "pseudo-tuberculosis
hominis streptothricea " ; an apparently similar condition has been
described by Buchholz. Berestnew cultivated two species of streptothrix
from suppurative lesions, one of which is acid -fast and grows only in
anaerobic conditions. Birt and Leishman have described another acid-
fast streptothrix obtained from cirrhotic nodules in the lungs of a man.
This organism grows readily on ordinary media, forming a white powdery
growth which afterwards assumes a pinkish colour ; it is pathogenic for
guinea-pigs, in which it causes caseous lesions. There is, further, the
streptothrix rnadurae described below.

In diseases of the lower animals several other forms have been found.
For example, a streptothrix has been shown by Nocard to be the cause
of a disease of the ox, — " farcin du boeuf," — a disease in which also there
occur tumour-like masses of granulation tissue. Dean has cultivated from
a nodule in a horse another streptothrix, which produces tubercle-like
nodules in the rabbit with club-formation ; it has close resemblances to
the organism of Israel and Wolff. The so-called diphtheria of calves and
" bacillary necrosis " in the ox are probably both produced by another
streptotlirix or leptothrix, which grows diffusely in the tissues in the
form of fine felted filaments. Further investigation may show that some
of these or other species may occur in the human subject in conditions
which are not yet differentiated.

Experimental Inoculation. — Inoculation of smaller animals,
such as rabbits and guinea-pigs, has usually failed to give positive
results. This was the case, for example, in the important series
of experiments by Bostrom, and it may be assumed that these
animals are little susceptible to the actinomyces. The disease
has, however, been experimentally produced in the bovine species
both by cultures from the ox and from the human subject.
Inoculation with the organism of Israel and Wolff produces
nodular lesions both in rabbits and in guinea-pigs, while Wright
found that characteristic colonies and lesions resulted although
the parasite did not grow to any great extent. Several of the
other species of streptothrix have been found to possess active
pathogenic properties.

Methods of Examination and Diagnosis. — As actinomycosis
cannot be diagnosed with certainty apart from the discovery of
the parasite, a careful examination of the pus in obscure cases of
suppuration should always be undertaken. As already stated,
the colonies can be recognised with the naked eye, especially
when some of the pus is spread out on a piece of glass. If one
of these is washed in salt solution and examined unstained, the
clubs, if present, are at once seen on microscopic examination. Or
the colony may be stained with a simple reagent such as picro-
carmine, and mounted in glycerin or Farrant's solution. To
study the filaments, a colony should be broken down on a cover-

328 ACTIffOMYCOSIS AND ALLIED DISEASES

glass, dried, and stained with a simple solution of any of the
basic aniline dyes, such as gentian-violet, though better results
are obtained by carbol-thion in-blue, or by carbol-fuchsin diluted
with five parts of water. If the specimen be over-stained, it
may be decolorised by weak acetic acid. Cover-glass pre-
parations of this kind, and also of cultures, are readily stained
by these methods, but in the case of sections of the tissues,
Gram's method, or a modification of it, should be used to show
the filaments, etc., a watery solution of acid fuchsin being after-
wards used to stain the clubs. By this method, very striking
preparations may be obtained.

Cultures should be made both under aerobic and anaerobic
conditions. Tubes of agar or glycerin agar should be inoculated
and incubated at 37° C. ; deep tubes of melted glucose agar
should also be used, the inoculated material being diffused
through the medium, separate colonies may thus be obtained.
Owing to the slow growth of the actinomyces, however, the
obtaining of pure cultures is somewhat difficult, unless the pus is
free from contamination with other organisms.

MADURA DISEASE.

Madura disease or mycetoma resembles actinomycosis both as
regards the general characters of the lesions and the occurrence
of the parasite in the form of colonies or "granules." There is
no doubt, however, that the two conditions are distinct, and it
also appears established that the two varieties of Madura disease
(vide infra) are produced by different organisms. This disease
is comparatively common in India and in various other parts of
the tropics : it has also been met with in Algiers and in America.
Madura disease differs from actinomycosis not only in its geo-
graphical distribution but also in its clinical characters. Its
course, for example, is of an extremely chronic nature, and
though the local disease is incurable except by operation, the
parasite never produces secondary lesions in internal organs.
Vincent also found that iodide of potassium, which has a high
value as a therapeutic agent in many cases of actinomycosis, had
no effect in the case of Madura disease studied by him. It most
frequently affects the foot ; hence the disease is often spoken of
as "Madura foot." The hand is rarely affected. In the parts
affected there is a slow growth of granulation tissue which has
an irregularly nodular character, and in the centre of the nodules
there occurs purulent softening which is often followed by the
formation of fistulous openings and ulcers. There are great

MADURA DISEASE

329

enlargement and distortion of the part and frequently caries and
necrosis of the bones. Within the softened cavities and also in
the spaces between the fibrous tissue, small rounded bodies or
granules, bearing a certain resemblance to the actinomyces, are
present. These may have a yellowish or pinkish colour, com-
pared from their appearance to fish roe, or they may be black
like grains of gunpowder, and may by their conglomeration
form nodules of considerable size. Hence a pale variety and a
black variety of the disease have been distinguished ; in both
varieties the granules mentioned reach a rather larger size than
in actinomycosis. These
two conditions will be
considered separately.

Pale Variety. — When
the roe-like granules an-
examined microscopically,
tIn-\ are found, like the
actiiiomyces, to show in
their interior an abundant
mass of branching fila-
ments witli mycelial
arrangement. There may
also be present at the
periphery club-like struc-
tures, as in actinomyces ;
sometimes they are ab-
sent. These structures
often have an elongated
\ve<lge-shape, forming an
outer zone to the colony,

and in some cases the filaments can be found to be connected with
them. Vincent obtained cultures of the parasite from a case in
Algiers, and found it to be a distinct species : it is now known
as the streptothrix or discomyces Madurce. Morphologically it
closely resembles the actinomyces, but it presents certain differ-
ences in cultural characters. In gelatin it forms raised colonies
of a yellowish colour, with umbilication of the centre, and there
is no liquefaction of the medium. On agar the growth assumes a
reddish colour ; the organism flourishes well in various vegetable
infusions in which" the actinomyces does not grow. On all the
media growth only takes place in aerobic conditions. Experi-
mental inoculation of various animals has failed to reproduce the
There is therefore no doubt that the streptothrix
madimr and the actiuomyces are distinct species.

Fie. 98. — Streptothrix Madura', showing
branching filaments. From a culture on

agar.
Stained with carbol-tliiomn-MiU'.

xlOOO.

330 ACTINOMYCOSIS AND ALLIED DISEASES

Black Variety. —The observations of J. H. Wright, who
obtained pure cultures of a hyphomycete, show that this variety
is a distinct affection from the pale variety. The pigment may
be dissolved by soaking the granules for a few minutes in
hypochlorite of sodium solution, and the granules may then be
crushed out beneath a cover-glass and examined microscopically.
The granules are composed of a somewhat homogeneous ground-
substance impregnated with pigment, and in this there is a
mycelium of thick filaments or hyphse, many of the segments
of which are swollen ; at the periphery the hyphae form a zone
with radiate arrangement. In many of the older granules the
parasite is largely degenerated and presents an amorphous
appearance. Wright planted over sixty of the black granules in
various culture media, and obtained cultures of a hyphomycete
from about a third of these. The organism grows well on agar,
bouillon, potato, etc. ; on agar it forms a felted mass of greyish
colour, and in old cultures black granules appear amongst the
mycelium. Microscopically the parasite appears as a mycelium
of thick branching filaments with delicate transverse septa ; in
the older threads the segments become swollen, so that strings of
oval-shaped bodies result. No signs of spore-formation were
noted. Inoculation of animals with cultures gave negative
results, as did also direct inoculation with the black granules
from the tissues. Brumpt, in a recent work, distinguishes
several varieties of parasite concerned in Madura disease, and
finds that a pale variety may be produced by a hyphomycete
as well as by Vincent's streptothrix ; in fact, with the exception
of Vincent's organism, all the parasites are considered by him to
be closely allied to aspergillus.

CHAPTER XIV.

ANTHRAX.1

OTHER NAMES. — SPLENIC FEVER, MALIGNANT PUSTULE, WOOL-

SORTER'S DISEASE. GERMAN, MILZBRAND ; FRENCH, CHARBON.2

Introductory.— Anthrax is a disease occurring epidemically
among the herbivora, especially sheep and oxen, in which
animals it has the characters of a rapidly fatal form of
septicaemia with splenic enlargement, attended by an extensive
multiplication of characteristic bacilli throughout the blood.
The disease does not occur as a natural infection from man to
man, but may be communicated to him directly or indirectly
from animals, and it may then appear in one of three forms.
In the first there is infection through the skin, in which a local
lesion, the "malignant pustule," occurs. In the second form
infection takes place through the respiratory tract. Here very
aggravated symptoms centred in the thorax, with rapidly fatal
termination, follow. Thirdly, an infection may occasionally
take place through the intestinal tract, which is now the first
part to give rise to symptoms. In all these forms of the affec-
tion in the human subject, the bacilli are in their distribution
much more restricted to the local lesions than is the case in the
ox, their growth and spread being attended by inflammatory
oedema and often by haemorrhages.

Historical Summary. — Historical researches leave little doubt that
from the earliest times anthrax has occurred among cattle. For along
time its pathology was not understood, and it went by many names.
Pollender in 1849 pointed out that the blood of anthrax animals con-
tained numerous rod-shaped bodies which he conjectured had some

1 In even recent works on surgery the term "anthrax" maybe found
applied to any form of carbuncle. Before its true pathology was known, the
local variety of the disease which occurs in man, and which is now called
" malignant pustule," was known as " malignant carbuncle."

-This must he distinguished from " charbou symptomatiqiuy' \vhu-his
quite a different disease.

331

332 ANTHRAX

causal connection with the disease. In 1863 Davaine announced that
they were bacteria, and originated the name bacillus anthracis. He stated
that unless blood used in inoculation experiments on animals contained
them, death did not ensue. Though this conclusion was disputed, still
by the work of Davaine and others the causal relationship of the bacilli
to the disease had been nearly established when the work of Koch
appeared in 1876. This constituted that observer's first contribution to
bacteriology, and did much to clear up the whole subject. Koch con-
firmed Davaine's view that the bodies were bacteria. He observed in the
blood of anthrax animals the appearance of division, and from this
deduced that multiplication took place in the tissues. He observed them
under the microscope dividing outside the body, and noticed spore forma-
tion taking place. He also isolated the bacilli in pure culture outside
the body, and, by inoculating animals with them, produced the disease
artificially. In his earlier experiments he failed to produce death by
feeding susceptible animals both with bacilli and spores, and as the
intestinal tract was, in his view, the natural path of infection, he con-
sidered as incomplete the proof of this method of the spontaneous occur-
rence of anthrax in herds of animals. Koch's observations were, shortly
afterwards, confirmed in the main by Pasteur, though controversy arose
between them on certain minor points. Moreover, further research
showed that the disease could be produced in animals by feeding them
with spores, and thus the way in which the disease might spread
naturally was explained.

Bacillus Anthracis. — Anthrax as a disease in man is of
comparative rarity. Not only, however, is the bacillus
anthracis easy of growth and recognition, but in its growth it
illustrates many of the general morphological characters of the
whole group of bacilli, and is therefore of the greatest use to the
student. Further, its behaviour when inoculated in animals
illustrates many of the points raised in connection with the
general pathogenic effects of bacteria, immunity, etc. Hence
an enormous amount of work has been done in investigating it
in all its aspects.

If a drop of blood is taken immediately after death from an
auricular vein of a cow, for example, which has died from
anthrax, and examined microscopically, it will be found to con-
tain a great number of large non-motile bacilli. On making
a cover-glass preparation from the same source, and staining with
watery methylene-blue, the characters of the bacilli can be better
made out. They are about 1'2 //, thick or a little thicker, and
6 to 8 //, long, though both shorter and longer forms also occur.
The ends are sharply cut across, or may be slightly dimpled so
as to resemble somewhat the proximal end of a phalanx. Their
protoplasm is very finely granular, and very frequently appears
surrounded by a capsule whose external margin is not, however,
so well defined as is the case with, e.g., the pneumococcus. When
several bacilli lie end to end in a thread, the capsule seems

BACILLUS ANTHRACIS

333

common to the whole thread (Fig. 103). They stain well with all
the basic aniline dyes and are not decolorised by Gram's method.

Methylene Blue Reaction. — This was introduced independently by
McFadyean and by Heim with a view to the easy recognition of the
bacilli in blood or other bodily fluids, and depends on a disintegration of
the bacillary capsules which occurs when these are imperfectly fixed.
Imperfect fixation is attained by drying a blood film on a slide and hold-
ing it three times for a second in a flame, film upwards (too great heating
fixes the capsules and prevents the reaction from occurring). The pre-
paration is stained for a few seconds with an old solution of methylene
blue, 1 per cent, in water (i.e., with a methylene blue possessing poly-
chromatic qualities, see p. 113). It is washed in water and dried with
filter paper, — preferably a cover glass is not applied. In such a prepara-
tion, between and near the bacteria there is a varying amount of an
irregularly disposed amorphous or finely granular material of a violet or
reddish -purple tint. Frequently the colour reaction in the preparation
is so marked as to be recognisable naked-eye. McFadyean states that
this reaction does not occur with putrefactive or other bacteria which
might be present under circumstances where the recognition of the
anthrax bacilli is the question under consideration.

Plate Cultures. — From a source such as that indicated, it is
easy to isolate the bacilli by making gelatin or agar plates. If,
after twelve hours' in-
cubation at 37° C., the
latter be examined under
a low objective, colonies
will be observed. They
are to be recognised by
1 MM ut if ul wavy wreaths
like locks of hair, radiat-
ing from the centre and
apparently terminating
in u point which, how-
ever, on examination
with a higher power, is
ol.xTved to be a fila-
ment which turns upon
itself (Fig. 99). Graham
Smith (vide p. 4) attri-
butes the appearance to
the toughness of the
bacterial envelope, which

prevents the separation of individuals from one another after
division. Tin- whole colony is, in fact, probably one long thread.
Such colonies are very suitable for making impression prepara-
tions (vide p. 138) which preserve permanently the appearances

FIG. 99. — Surface colony of the anthrax
bacillus on an agar plate, showing the
characteristic appearance, x 30.

334

ANTHRAX

described. On examining such with a high power, the wreaths
are seen to be made up of bundles of long filaments lying
parallel with one another, each filament consisting of a chain

FIG. 100. — Anthrax bacilli, arranged in chains,
from a twenty-four hours' culture on agar
at 37° C.

Stained with fuchsiu. x 1000.

of bacilli lying end to end, and similar
to those observed in the blood (Fig.
100).

On gelatin plates, after from twenty-
four to thirty-six hours at 20° C., the
same appearances manifest themselves,
and later they are accompanied by
liquefaction of the gelatin. In gelatin
plates, however, instead of the char-
acteristically wreathed appearance at
the margin, the colonies sometimes
give off radiating spikelets irregularly
nodulated, which produce a star-like
form. These spikelets are composed
of spirally twisted threads.

From such plates the bacilli can be easily isolated, and the
appearances of pure cultures on various media studied.

In bouillon, after twenty-four hours' incubation at 37° C.,
there is usually the appearance of irregularly spiral threads sus-
pended in the liquid. These, on being examined, are seen

FIG. 101. Stab culture of
the anthrax bacillus in
peptone-gelatin ; seven
days' growth. It shows
the "spiking," and also,
at the surface, com-
mencing liqiiefaction.
Natural size.

BIOLOGY OF THE B. ANTHRACIS 335

to be made up of bundles of parallel chains of bacilli. Later,
irmxvth is more abundant, and forms a flocculent mass at the
bottom of the fluid.

In yelatin stab cultures, the characteristic appearance can be
best observed when a low proportion, say 7J per cent., of gelatin
is present, and when the tube is directly inoculated from
anthrax blood. In about two days there radiate out into the
medium from the needle track numberless very fine spikelets
which enable the cultures to be easily recognised. These spike-
lets are longest at the upper part of the needle track (Fig. 101).
Not much spread takes place on the surface of the gelatin, but
here liquefaction commences, and gradually spreads down the
stab and out into the medium, till the whole of the gelatin may
be liquefied. Gelatin slope cultures exhibit a thick felted
growth, the edges of which show the wreathed appearance seen
in plate cultures. Liquefaction here soon ploughs a trough in
the surface of the medium. Sometimes "spiking" does not
take place in gelatin stab cultures, only little round particles of
growth occurring down the needle tract, followed by liquefaction.
As has been shown by Rd. Muir, this property of spiking can be
restored by growing the bacillus for twenty-four hours on blood
agar at 37° C. Agar sloped cultures have the appearance of
similar cultures in gelatin, though, of course, no liquefaction
takes place.

Blood serum sloped cultures present the same appearances as
those on agar. The margin of the surface growth on any of the
solid media shows the characteristic wreathing seen in plate
colonies. The occurrence of capsulation of the bacilli in such
cultures has been described.

On potatoes there occurs a thick felted white mass of bacilli
showing no special characters. Such a growth, however, is use-
ful for -studying sporulation.

The anthrax bacillus will thus grow readily on any of the
ordinary media. It can usually be sufficiently recognised by its
microscopic appearance, by its growth on agar or gelatin plates,
and by its growth in gelatin stab cultures. The growth on
plates is specially characteristic, and is simulated by no other
pathogenic organism.

The Biology of the B. Anthracis. — Koch found that the
bacillus anthracis grows best at a temperature of 35° 0. Growth,
i.e. multiplication, does not take place below 12° C. nor above
45° C. In the spore-free condition the bacilli have comparatively
low powers of iv<i<t;mce. They do not stand long exposure to
60° C., and if kept at ordinary temperature in the dry condition

336

ANTHRAX

they are usually found to be dead after a few days. The action
of the gastric juice is rapidly fatal to them, and they are accord-
ingly destroyed in the stomachs of healthy animals. They are
also soon killed in the process of putrefaction. They can, how-
ever, be cooled below the freezing-point without dying. The
bacillus can grow without oxygen, but some of its vital functions
are best carried on in the presence of this gas. Thus in anthrax
cultures the liquefaction of gelatin always commences at the
surface and spreads downwards. Growth is more rapid in the
presence of oxygen, and spore formation does not occur in its

absence. The organism
may be classed as a facul-
tative anaerobe.

Sporulation. — Under
certain circumstances
spor ulation occurs in
anthrax bacilli. The
morphological appear-
ances are of the ordinary
kind. A little highly
refractile speck appears
in the protoplasm about
the centre of the bacillus ;
this gradually increases
in size until it forms an
oval body about the same
thickness as the bacillus
iv;no. jn fhp K^lla™ ^vri
^mf u ' DaClUaiy pro-

toplasm (Fig. 102). The
latter gradually loses its
staining capacities and
finally disappears. The

spore thus lies free as an oval highly refractile body which does
not stain by ordinary methods, but which can be easily stained
by the special methods described for such a purpose (p. 109).
When the spore is again about to assume the bacillary form the
capsule is apparently absorbed, and the protoplasm within grows
out, taking on the ordinary rod-shaped form.

According to most observers, sporulation never occurs within
the body of an animal suffering from anthrax. Koch attributes
this, probably rightly, to the absence of free oxygen. The latter
gas he found necessary to the occurrence of spores in cultures
outside the body. Many, however, are inclined to assign as the
cause of sporulation the absence of the optimum pabulum. Be*

FIG. 102.— Anthrax bacilli containing spores
(the darkly coloured bodies) ; from a three
days' culture on agar at 37° C. See also
Plate III., Fig. 2.

Stained with carbol-fuchsin and methylene-

ANTHRAX IN ANIMALS 337

sides these conditions there is another factor necessary to sporula-
tion, namely, a suitable temperature. The optimum temperature
tor spore production is 30° C. Koch found that spore formation
did not occur below 18° C. Above 42° C. not only does
sponilation cease, but Pasteur found that if bacilli were kept at
this temperature for eight days they did not regain the capacity
when again grown at a lower temperature. In order to make
them again capable of sporing, it is necessary to adopt special
measures, such as passage through the bodies of a series of
susceptible animals.

Anthrax spores have extremely high powers of resistance.
In a dry condition they will remain viable for a year or more.
Koch found they resisted boiling for five minutes ; and dry heat
at 140° C. must be applied for several hours to kill them with
certainty. Unlike the bacilli, they can resist the action of the
gastric juice for a long period of time. They are often used
as test objects by which the action of germicides is judged. For
this purpose an emulsion is made by scraping off a surface
culture and rubbing it up in a little sterile water. Into this
sterile silk threads are dipped, which, after being dried over
strong sulphuric acid in a desiccator, can be kept for long
periods of time in an unchanged condition. For use they are
placed in the germicidal solution for the desired time, then
washed with water to remove the last traces of the reagent and
laid on the surface of agar or placed in bouillon, in order that if
death of the bacilli has not occurred growth may be observed
(see Chap. VI.).

Anthrax in Animals. — Anthrax occurs from time to time
epidemically in sheep, cattle, and, more rarely, in horses and
deer. These epidemics are found in various parts of the world,
although they are naturally most far-reaching where legal pre-
cautions to prevent the spread of infection are non-existent.
All the countries of Europe are from time to time visited by the
disease, but in some it is much more common than in others.
In Britain the death-rate is small, but in France the annual
mortality among sheep was probably 10 per cent, of the total
number in the country, and among cattle 5 per cent. These
figures, however, have been largely modified by the system of
preventive treatment which will be presently described. In sheep
ant I cattle the disease is specially virulent. An animal may
suddenly drop down, with symptoms of collapse, quickening of
pulse ami ie>piiation, and dyspnoea, and death may occur in
a few minutes. In less acute cases the animal is apparently
out of sorts, and does not feed ; its pulse and respiration are
22

338 ANTHRAX

quickened ; rigors occur, succeeded by high temperature ; there
is a sanguineous discharge from the bowels, and bloody mucus
may be observed about the mouth and nose. There may be
convulsive movements, and progressive weakness, with cyanosis,
is followed by death in from twelve to forty-eight hours. In
the more prolonged cases widespread oedema and extensive
enlargement of lymphatic glands are marked features ; and in
the glands, especially about the neck, actual necrosis with
ulceration may occur, constituting the so-called anthrax car-
buncles. Such subacute conditions are especially found among
horses, which are by nature not so susceptible to the disease as
cattle and sheep.

On post-mortem examination of an ox dead of anthrax, the
most noticeable feature — one which has given the name " splenic
fever " to the disease — is the enlargement of the spleen, which
may be two or three times its natural size. It is of dark-red
colour, and on section the' pulp is very soft and friable, sometimes
almost diffluent. A cover-glass preparation may be made from
the spleen and stained with watery methylene-blue. On ex-
amination it will be found to contain enormous numbers of
bacilli mixed with red corpuscles and leucocytes, chiefly
lymphocytes and the large mononucleated variety (Fig. 103).
Pieces of the organ may be hardened in absolute alcohol, and
sections cut in paraffin. These are best stained by Gram's
method. Microscopic examination of such shows that the
structure of the pulp is considerably disintegrated, whilst the
bacilli swarm throughout the organ, lying irregularly amongst
the cellular elements. The liver is enlarged and congested, and
may be in a state of acute cloudy swelling. The bacilli are
present in the capillaries throughout the organ, but are not so
numerous as in the spleen. The kidney is in a similar condition,
and here the bacilli are chiefly found in the capillaries of the
glomeruli, which often appear as if injected with them. The
lungs are congested and may show catarrh, whilst bacilli are
present in large numbers throughout the capillaries, and may
also be found in the air cells, probably as the result of rupture
of the capillaries. The blood throughout the body is usually
fluid and of dark colour.

The lymphatic system generally is much affected. The
glands, especially the mediastinal, mesenteric, and cervical
glands, are enlarged and surrounded by oedematous tissue, the
lymphatic vessels are swollen, and both glands and vessels may
contain numberless bacilli. The heart-muscle may be in a state
of cloudy swelling, and the blood in its cavities contains bacilli,

ANTHRAX IN ANIMALS 339

though in smaller numbers than that in the capillaries. The
intestines are enormously congested, the epithelium more or less
desquamated, and the lumen filled with a bloody fluid. From
all the organs the bacilli can be easily isolated by stroke cultures
on agar.

It is important to note the existence of great differences in
susceptibility to anthrax in different species of animals. Thus

;^k» - '••-•

Fi'i. 103. — Scraping from spleen of guinea-pig dead of anthrax,
showing the bacilli mixed with leucocytes, etc. (Same appearance
as in the ox.)

" Corrosive film " stained with carl >ol-thion in -blue, x 1000.

the ox, sheep (except those of Algeria, which only succumb to
enormous doses of the bacilli), guinea-pig, and mouse are all
very susceptible, the rabbit slightly less so. We have no data to
determine whether the disease occurs among the last three in the
wild state. Less susceptible than this group are the horse, deer,
goat, in which the disease occurs from time to time in nature.
Anthrax also occurs epidemically in the pig, often from the
ingestion of the organs of other animals dead of the disease. It
is, however, doubtful if all cases of disease in the pig described

340 ANTHRAX

on clinical grounds as anthrax are really such. A careful
bacteriological examination is here always advisable, especially
of any oedematous infiltration about the throat, or in the
neighbouring lymphatic glands ; often, in pigs dying of
anthrax, bacilli may not occur in the blood. Any hsemorrhagic
infarction in the spleen of a suspected animal should be carefully
investigated. The human subject may be said to occupy a

FIG. 104. — Portion of kidney of a guinea-pig dead of anthrax,
showing the bacilli in the capillaries, especially of the glomerulus.
Paraffin section : stained by Gram's method and Bismarck-brown.
x300.

medium position between the highly susceptible and the rela-
tively immune animals. The white rat is highly immune to the
disease, while the brown rat is susceptible. Adult carnivora are
also very immune, and the birds and amphibia are in the same
position.

With these differences in susceptibility there are also great
variations in the pathological effects produced in the natural or
artificial disease. This is especially the case when we consider
the distribution of the bacilli in the bodies of the less susceptible

ANTHRAX IN THE HUMAN SUBJECT 341

animals. Instead of the widespread occurrence described above,
they may be confined to the point where they first gained access
to the body and the lymphatic system in relation to it, or may
be only very sparsely scattered in organs such as the spleen
(which is often not enlarged), the lungs, or kidneys. Neverthe-
less the cellular structure of the organs even in such a case may
show changes, a fact which is important when we consider the
essential pathology of the disease.

Experimental Inoculation. — Of the animals commonly used
in laboratory work, white mice and guinea-pigs are the most
susceptible to anthrax, and are generally used for test inocula-
tions. If a small quantity of anthrax bacilli be injected into the
subcutaneous tissue of a guinea-pig, a fatal result follows, usually
within two days. Post-mortem, around the site of inoculation the
tissues, owing to intense inflammatory oedema, are swollen and
gelatinous in appearance, small haemorrhages are often present,
and on microscopic examination numerous bacilli are seen.
The internal organs show congestion and cloudy swelling, with
sometimes small haemorrhages, and their capillaries contain
enormous numbers of bacilli, as has already been described in
the case of the ox (Fig. 104) ; the spleen also shows a corre-
sponding condition. Highly susceptible animals may be infected
by being made to inhale the bacilli or their spores, and also by
being fed with spores, a general infection rapidly occurring by
both methods.

Anthrax in the Human Subject. — As we have noted, man
occupies a middle position in the scale of susceptibility to
anthrax. It is always communicated to him from animals, and
usually is seen among those whose trade leads them to handle
the carcases or skins of animals which have died of the disease.
It occurs in two principal forms, the main difference between
which is due to the site of entrance of the organism into the
lx>dy. In one, the path of entrance is through cuts or abrasions
in the skin, or through the hair follicles. A local condition
called a "malignant pustule" develops, which may lead to a
general infection. This variety occurs chiefly among butchers
and those who work among hides (foreign ones especially). In
Britain the workers of the latter class chiefly liable are the hide-
porters and hide- workers in South-Eastern London. In the
other variety of the disease the site of infection is the trachea
and bronchi, aud here a fatal result almost always follows. The
cause is the inhalation of dust or threads from wool, hair, or
IT! sties, which have been taken from animals dead of the disease,
and which have been contaminated with blood or secretions con-

342 ANTHRAX

taining the bacilli, these having afterwards formed spores.
This variety is often referred to as " woolsorter's disease," from
its occurring in the centres of the woolstapling trade .(in
England, chiefly in Yorkshire), but it also is found in places
where there are hair and brush factories.

(1) Malignant Pustule. — This usually occurs on the exposed
surfaces — the face, hands, fore-arms, and back, the last being a
common site among hide-porters. One to three days after
inoculation a small red painful pimple appears, soon becoming a
vesicle, which may contain clear or blood-stained fluid, and is
rapidly surrounded by an area of intense congestion. Central
necrosis occurs and leads to the malignant pustule proper, which
in its typical form appears as a black eschar often surrounded
by an irregular ring of vesicles, these in turn being surrounded
by a congested area. From this pustule as a centre subcutaneous
oedema spreads, especially in the direction of the lymphatics;
the neighbouring glands are enlarged. There is fever, with
general malaise. On microscopic section of the typical pustule,
the central eschar is noticed to be composed of necrosed tissue
and degenerating blood cells ; the vesicles are formed by the
raising of the stratum corneum from the rete Malpighi. Beneath
them and in their neighbourhood the cells of the latter are
swollen and oedematous, the papillae being enlarged and flattened
out and infiltrated with inflammatory exudation, w^hich also
extends beneath the centre of the pustule. In the tissue next
the eschar necrosis is commencing. The subcutaneous tissue is
also cedematous, and often infiltrated with leucocytes. The
bacilli exist in the periphery of the eschar and in the neigh-
bouring lymphatics, and, to a certain extent, in the vesicles. It
is very important to note that widespread oedema of a limb,
enlargement of neighbouring glands, and fever may occur while
the bacilli are still confined to the immediate neighbourhood
of the pustule. Sometimes the pathological process goes no
further, the bacilli gradually die out, the eschar becomes a scab,
the inflammation subsides, and recovery takes place. In the
majority of cases, however, if the pustule be not excised, the
oedema spreads, invasion of the blood stream may occur, and
the patient dies with, in a modified degree, the pathological
changes detailed with regard to the acute disease in cattle. In
man the spleen is usually not much enlarged, and the organs
generally contain few bacilli. The actual cause of death is
therefore a toxic effect. The early excision of an anthrax
pustule, especially when it is situated in the extremities, is
followed, in a large proportion of cases, by recovery.

TOXINS OF THE BACILLUS ANTHRACIS 343

(2) Woolsorter's Disease. — The pathology of this affection
was worked out in this country especially by Greenfield. The
local lesion is usually situated in the lower part of the trachea or
in the large bronchi, and is in the form of swollen patches in
the mucous membrane, often with haemorrhage into them, — small
ulcers may also be seen. The tissues are cedematous, and the
cellular elements are separated, but there is usually little or no
necrosis. There is enormous enlargement and engorgement of
the mediastinal and bronchial glands, and haemorrhagic infiltra-
tion of the cellular tissue in the region. There are pleural and
pericardial effusions, and haemorrhagic spots occur beneath the
serous membranes. The lungs show great congestion, collapse
and oedema. There may be cutaneous oedema over the chest
and neck, with enlargement of glands, and the patient rapidly
dies with symptoms of pulmonary embarrassment, and with a
varying degree of pyrexia. It is to be noted that in such cases,
though numerous bacilli are present in the bronchial lesions, in
the lymphatic glands, and affected tissues in the thorax, com-
paratively few may be present in the various organs, such as the
kidney, spleen, etc., and sometimes it may be impossible to find
any.

(3) Infection occasionally takes place through the intestine,
probably by ingestion of spores as in the case of animals ; but
this condition is rare. In such cases there occur single or
multiple local haemorrhagic lesions in the intestinal mucous
membrane, the central parts of the haemorrhagic areas tending
to be necrotic and yellowish, and there may be a corresponding
affection of the mesenteric glands.

A considerable number of cases have been recorded in which
hajmorrhagic meningitis, associated with the presence of the
anthrax bacilli in large numbers, has occurred as a complica-
tion of various primary lesions.

The Toxins of the Bacillus Anthracis. — Various theories
were formerly held as to the mode in which the anthrax bacillus
produces its effects. One of the earliest was the mechanical,
according to which it was supposed that the serious results were
produced by extensive blocking of the capillaries in the various
organs by the bacilli. According to another, it was supposed
that the bacilli used up the oxygen of the blood, thus leading to
starvation of the tissues. In modern times there has been a
tendency to attribute the effects produced to toxic action.
Sidney Martin investigated this subject, and isolated from
cultures, ] 'i-oto-albuinose, deutero-albumose, traces of pep-
tone, and alkaloidal bodies. By these, pathogenic effects were

344 ANTHRAX

produced in animals, closely similar to those produced by the
bacilli themselves. Martin, to account for the symptoms of the
disease, considered that the fever was mostly due to the
albumoses, while the oedema and congestion were caused by
the alkaloid acting as a local irritant. Hankin and Wesbrook
arrived at the conclusion that the bacillus anthracis produces
a ferment which, diffusing out into the culture fluid, elaborates
albumoses from the proteids present in it. The bacilli also pro-
duce albumoses directly without the intervention of a ferment.
Marmier isolated from cultures in peptone solution a body which
gave no reactions of albuminoid matter, peptone, propeptone,
or alkaloid, and which he considered to be the toxin. It was
chiefly retained within the bacilli when these were growing in
the most favourable conditions, and was not destroyed by heat-
ing to 110° C. The toxin produced by the b. anthracis growing
in a fluid medium remains intimately associated with the
bacterial protoplasm, as such cultures when filtered are relatively
non-toxic.

It cannot be said that great light has been thrown on the
pathology of the disease by these researches. The effects of
infection by the b. anthracis are those shared by all other
organisms producing inflammation, the tendency to oedema-
production of an unwonted degree being the chief special
feature. That toxic effects do occur in anthrax is probable,
for frequently, while the bacilli are still locally confined, there
may occur pyrexia and oedema spreading widely beyond the
pustule, but we have no definite information as to how these
effects are produced. In this disease we are probably dealing
with another example of the action of intracellular toxins,
regarding which, as in other cases, little is known.

The Spread of the Disease in Nature. — We have seen that
the b. anthracis rarely, if ever, forms spores in the body, and if
the bacilli could be confined to the blood and tissues of carcases
of animals dying of the disease, it is certain that anthrax in an
epidemic form would rarely occur. For it has been shown by
many observers that in the course of the putrefaction of such
a carcase the anthrax bacilli rapidly die out, and that after ten
days or a fortnight very few remain. But it must be remembered
that while still alive an animal is shedding into the air by the
bloody excretions from the mouth, nose, and bowel, myriads of
bacilli which may rapidly spore, and thus arrive at a very re-
sistant stage. These lie on the surface of the ground and are
washed off by surface water. At certain seasons of the year the
temperature is, however, sufficiently high to permit of their

SPREAD OF THE DISEASE IN NATURE 345

germination, and also of their multiplication, as they can un-
doubtedly grow on the organic matter which occurs in nature.
They can again form spores. It is in the condition of spores
that they are dangerous to susceptible animals. In the bacillary
stage, if swallowed, they will be killed by the acid gastric con-
tents ; but as spores they can pass uninjured through the
stomach, and gaining an entrance into the intestine, infect its
wall, and ultimately reach, and multiply in the blood. It is
known that in the great majority of cases of the disease in sheep
ami oxen, infection takes place thus from the intestine. It was
thought by Pasteur that worms were active agents in the natural
spread of the disease by bringing to the surface anthrax spores.
Koch made direct experiments on this point, and could get no
evidence that such was the case. He thinks it much more
probable that the recrudescence of epidemics in fields where
anthrax carcases have been buried is due to persistence of spores
on the surface which has been infected by the cattle when alive.
In Britain it is common to attribute the occurrence of sporadic
outbreaks to infection by imported feeding stuffs. Scientific
proof of such a method of infection being common is still
wanting,

The Disposal of the Carcases of Animals dead of Anthrax. — It is ex-
tremely important that anthrax carcases should be disposed of in such a,
way as to prevent their becoming future sources of infection. If anthrax
be suspected as the cause of death, no post-mortem examination should be
made, but only a small quantity of blood removed from an auricular
vein for bacteriological investigation. If such a carcase be now buried
in a deep pit surrounded by quicklime, little danger of infection will be
run. The bacilli being conftned within the body will not spore, and will
die during the process of putrefaction. The danger of sporulation taking
place is, of course, much greater when an animal has died of an unknown
disease, which, on post-mortem examination, has proved to be anthrax,
but similar measures for burial must be here adopted. In some countries
anthrax carcases are burned, and this, if practicable, is of course the best
means of treating them. The chief source of danger to cattle subsequently,
however, proceeds from the infection of fields, yards, and byres with the
offal and the discharge from the mouths of anthrax animals. All material
that can be recognised as such should be burned along with the straw in
which the animals have lain. The stalls or buildings in which the
anthrax cases have been must be limewashed. Needless to say, the
greatest care must be taken in the case of men who handle the animal or
its carcase that they have no wounds on their persons, and that they
thoroughly disinfect themselves by washing their hands, etc., in 1 to
1000 solution of corrosive sublimate or lysol, and that all clothes soiled
with blood, etc., from anthrax animals be thoroughly boiled or steamed
for half an hour before being washed.

The Immunising of Animals against Anthrax. — Having

346 ANTHRAX

ascertained that there was ground for believing that in cattle
one attack of anthrax protected against a second, Pasteur (in
the years 1880-82) elaborated a method by which a mild form
of the disease could be given to animals, which rendered
harmless a subsequent inoculation with virulent bacilli. He
found that the continued growth of anthrax bacilli at 42° to
43° C. caused them to lose their capacity of producing spores,
and also gradually to lose their virulence, so that after twenty-
four days they could no longer kill either guinea-pigs, rabbits,
or sheep. Such cultures constituted his premier vaccin, and
protected against the subsequent inoculation with bacilli which
had been grown for twelve days at the same temperature, and
the attenuation of which had therefore not been carried so far.
The latter constituted the deuxieme vaccin. It was further
found that sheep thus twice vaccinated now resisted inoculation
with a culture which usually would be fatal. The method was
to inoculate a sheep on the inner side of the thigh by the
subcutaneous injection, from a hypodermic syringe, of about
five drops of the premier vaccin \ twelve days later to again
inoculate with the deuxieme vaccin ; fourteen days later an
ordinary virulent culture was injected without any ill result.
This method was applicable also to cattle and horses, about
double the dose of each vaccine being here necessary. Extended
experiments in France generally confirmed earlier results, and
the method was, before long, used to mitigate the disease, which
in many departements was endemic and a very great scourge.
Since that time the method has been regularly in use. It is
difficult to arrive at a certain conclusion as to its merits.
Undoubtedly a certain number of animals die of anthrax either
after the first or second vaccination, or during the year following
vaccination. At the end of a year the immunity is lost in
about 40 per cent, of the animals vaccinated ; and thus to be
permanently efficacious the process would have to be repeated
every year. Further, the immunity is much higher in degree
if, after the first and second vaccinations, an inoculation with
virulent anthrax is performed. Everything being taken into
account, however, there is no doubt that the mortality from
natural anthrax is much diminished by this system.

During the twelve years 1882-93, 3,296,815 sheep were vaccinated,
with a mortality, either after the first or second vaccination, or during
the subsequent twelve months, of 0*94 per cent., as contrasted with the
ordinary mortality in all the flocks of the districts of 10 per cent.
During the same time 438,824 cattle were vaccinated, \vith a mortality,
of 0*34 per cent., as contrasted with a probable mortality of 5 per cent, if
they had been unprotected.

IMMUNISATION AGAINST ANTHRAX 347

The immunisation of animals against anthrax has always
been found to be a difficult proceeding. The most usual
technique has been to commence with Pasteur's vaccines, and to
follow these by careful dosage with virulent cultures. Marchoux
in this way produced immunity, and found that the serum of
immune animals had a certain degree of protective and curative
action. The most successful attempts in this direction have
been those of Sclavo and of Sobernheim. The former observer,
after trying various animals, came to the conclusion that the
ass was the most suitable. He first employed a method similar
to that of Marchoux ; later, however, after noting the effects
of the serum of an animal so immunised, he commenced the
immunisation by injecting 5 to 15 c.c. of this serum along with
a slightly attentuated culture of the bacilli. A few days later
this was followed up with injections of virulent cultures which
could now be periodically introduced for many months, and a
high degree of immunity resulted. What was even more
important, the serum of such an animal had strongly protective
and curative properties. It has been extensively used in the
treatment of anthrax in man. In a case of malignant pustule
30 to 40 c.c. are injected in quantities of 10 c.c. into the
abdominal wall, and if necessary the injection is repeated on the
following day. In cases treated by Sclavo himself the serum is
alone employed, and its action is. not aided by the excision
of the pustule usually practised. The results obtained have been
very good, — Sclavo, out of 164 cases, had only ten deaths or
about a fourth of the ordinary mortality in Italy. Sobernheim
independently elaborated an almost identical method of com-
bining passive with active immunisation for the obtaining of a
powerful anti-serum, and he has used this for the protective
inoculation of cattle. The technique is to inject a mixed serum
obtained from the ox, the horse, and the sheep, into one side
of the neck or into one thigh and the culture (Pasteur's second
vaccine) into the other side ; the doses given are for cattle or
horses 5 c.c. of serum and 0'5 c.c. culture, and for sheep 4 c.c.
of serum and 0*25 c.c. culture. The method has been widely
used in Germany and in Brazil, and its originator claims as its
advantages simplification of application, in that one operation
instead of two is sufficient, less risk of death following the
immunisation procedure, and higher degree and more lasting
character of the immunity resulting. During the development
of active immunity it is likely in every case (see Immunity) that
there is a . period of increased susceptibility to the disease.
Such a period would be more likely to occur with the Pasteur

348 ANTHRAX

method than with the Sobernheim procedure, where the
presence in the animal's body of the protective serum might tide
it over the stage when the action of the vaccine was lowering
its resistance.

The effects of the b. anthracis have been much studied with
a view to the shedding of light on the processes obtaining in
resistance and the development of immunity. Many puzzling
facts have long been known ; for example, in the dog, which
shows great natural resistance, the serum has little if any
bactericidal action, while in the susceptible rabbit ther.e is
present a serum capable of killing the organism. Such observa-
tions have hitherto been without explanation. Again, the
properties of the serum of immune animals have been much
discussed. Sobernheim and others have been unable to detect
in it any trace of special bactericidal action. Sclavo found that
the serum when heated to 55° C. did not lose its protective
properties; as the serum might have been complemented (see
Immunity) by the serum of the animal into which it was injected,
he simultaneously introduced an anti-complementary serum and
found that the heated serum was still effectual. From this he
deduces that in the action of the serum substances of the nature
of immune body and complement are not concerned. Many
have thought that the serum had a stimulating effect on the
leucocytes, but Cler has brought forward ground for supposing
that its effect is a sensitising one on the bacteria, and that thus
the effects are to be traced to opsonic action. With regard to
the formation of the protective substances, it is stated that the
spleen and bone-marrow are richer in these than the blood fluids.
In this connection an interesting fact may be mentioned, namely,
that Roger and Gamier found evidence of the liver and spleen
having special capacities for killing anthrax bacilli ; an otherwise
fatal dose could be introduced into the portal vein or the splenic
artery' without causing death. It has been thought that the
capsule frequently developed by the anthrax bacillus is a defen-
sive mechanism against bactericidal and bacteriolytic capacities
in an infected animal. It is stated that capsulation renders the
bacillus less susceptible to phagocytosis. Opinion on the signific-
ance of capsule formation is at present divided.

Methods of Examination. — These include (a) microscopic
examination ; (b) the making of cultures ; and (c) test in-
oculations.

(a) Microscopic Examination. — In a case of suspected
malignant pustule, film preparations should be made from the
fluid in the vesicles or from a scraping of the incised or excised

METHODS OF EXAMINATION 349

pustule, and stained with a watery solution of methylene-blue
and also by Gram's method. By this method practically con-
clusive evidence may be obtained; but sometimes the result
is doubtful, as the bacilli may be very few in number.
McFadyean's methylene-blue method (p. 333) should also be
applied. In all cases confirmatory evidence should be obtained
by culture. Occasionally bacilli are so scanty that both film
preparations made from different parts and even cultures may
give negative results, and yet a few bacilli may be found when
a section of the pustule is examined. It should be noted that
the greatest care ought to be taken in manipulating a pustule
before excision, as the diffusion of the bacilli into the
surrounding tissues may be aided and the condition greatly
aggravated. The examination of the blood in cases of anthrax
in man usually gives negative results, with the exception of very
severe cases, when a few bacilli may be found in the blood
shortly before death, though even then they may be absent.

(h] Cultivation. — A small quantity of the material used for
microscopic examination should be taken on a platinum needle,
and successive strokes made on agar tubes, which are then
incubated at 37° C. At the end of twenty-four hours anthrax
colonies will appear, and can be readily recognised from their
wavy margins by means of a hand lens. They should also be
examined microscopically by means of film preparations.

(c) Test Inoculation. — A little of the suspected material
should be mixed with some sterile bouillon or water, and
injected subcutaneously into a guinea-pig or mouse. If anthrax
bacilli are present, the animal usually dies within two days, with
the changes in internal organs already described. The diagnosis
of an organism as the anthrax bacillus cannot be said to be
substantiated till its pathogenicity has been proved.

CHAPTEE XV.

TYPHOID FEVER— BACILLI ALLIED TO THE
TYPHOID BACILLUS.

Introductory. — The organism now known as the bacillus
typhosus was first described in 1880-1 by Eberth, who observed
its microscopic appearance in the intestinal ulcers and in the
spleen in cases of typhoid fever (German, Abdominaltyphus).
It was first isolated (from the spleen) in 1884 by Gaffky, and its
cultural characters were then investigated. In 1885 Escherich
observed a bacillus, now known as the bacillus coli communis,
which occurs in the normal intestine, and which both micro-
scopically and culturally closely resembles the typhoid bacillus.
Ordinarily the b. coli is no doubt a harmless saprophyte, but
under experimental conditions in animals and also naturally in
man it may manifest pathogenic properties. Investigation has
shown that these two bacilli belong to a widespread group
of organisms isolated from various disease conditions, chiefly
of the intestine, which all bear close resemblances to one
another, and whose differentiation is often a matter of consider-
able difficulty. Other members of this group are the para-
typhoid bacillus, the organism of bacillary dysentery, the b.
enteritidis of Gaertner, the psittacosis bacillus, and the bacillus
of hog cholera.

The general characters of the group are as follows : the
organisms, which are microscopically indistinguishable, are thin
non-sporing bacilli, which in cultures often show variation in
length ; they are mostly motile, but this quality varies in differ-
ent members ; they possess flagella springing from all round
the bacillus ; they stain with ordinary dyes, and are all Gram-
negative ; they are all facultative anaerobes, i.e. they grow best
in the presence of oxygen, but can tolerate its absence ; in
growth characters on ordinary media they closely resemble one
another, and, generally speaking, they do not liquefy gelatin;
they show wide differences in fermentative capacities, and are
chiefly to be distinguished by their immunity reactions.

350

BACILLUS COLI COMMUNIS

351

THE BACILLUS COLI COMMUNIS.

Although the discovery of the bacillus coli communis was
subsequent to that of the bacillus typhosus, it is convenient to
commence with a description of the former, as it presents more
positive characters than any other member of the group to which
it belongs.

Bacillus Coli Communis. — Morphological Cliaracters. — These
are best seen in cultures. The bacillus is ordinarily from 2 to
4 p. long and about *5 /A broad ; longer forms up to 8 or 10 /A
are not infrequent (Fig. 105). It is usually found to be motile,
but the motility varies
in different strains and
under different growth
conditions in the same
strain. Here it is best
to use bouillon cultures
incubated at 37° C. for
from six to twelve
hours. The organism
may stain somewhat
faintly with watery dyes,
but is readily demon-
strated with carbol-
fuchsin even in fairly
\\vak solution (1 of the
Ziehl-Neelsen stain in
20 of water) ; it is
Gram-negative. By ap-
propriate staining 1). coli
derived from cultures

can be shown to possess fiagella springing from all round the
organism, varying in number and occasionally rather short.

Culture Reactions on Ordinary Media. — The following are
the appearances of the b. coli in the ordinary culture media : —

In bouillon, it produces a uniform turbidity. When grown in
fluid gelatin, it is stated by Klein to tend to form flocculi floating
on the surface rather than a uniform turbidity. In stab cultures
on jteptoue yelatin an abundant film-like growth takes place on
the surface, and there is a whitish or brownish-white line along
the stall. No liquefaction of the gelatin occurs, but occasionally
a few bubbles of gas develop (Fig. 109, C). In sloped cigar
tni>i>* a somewhat dense, glistening, white or brownish- white
growth occurs along the stroke. When ayar jjlates are used

Fie. 105. — Bacillus coli communis. Film

preparation from a young culture on agar.
Stained with weak carbol-tuchsin. x 1000.

352 TYPHOID FEVER

for the separation of the organism, the surface colonies are
somewhat large, and it may be irregular in outline, but
the deep colonies are smaller and lenticular in shape, and
under a low power of the miser oscope appear rather dense to
transmitted light. A similar growth occurs on blood serum.
On potatoes, in forty-eight hours, there is a distinct film of
growth of a brownish tint, sometimes with a moist surface,
which rapidly spreads and becomes thicker. The appearance
on potato, however, varies much with the different strains and
also with the reaction of the potato.

Culture Reactions on Special Media. — A great variety of
media have been used for the appreciation of special characters
in the b. coli. These reactions depend upon the capacities of the
organism to originate chemical changes in a variety of substances.

A. Fermentative Reactions on Carbo-hydrates. — B. coli shows
great powers of splitting up carbo-hydrates with the formation
of acids, especially lactic acid, and gas, chiefly carbon dioxide
and hydrogen.

Gelatin Shake Cultures. — If a tube of gelatin be melted,
infected with b. coli, shaken up, allowed to solidify, and kept
at room temperature, distinct growth of the organism occurs,
and round each little colony, bubbles of gas form, which in time
coalesce and give the tube a readily recognised appearance.
This phenomenon is due to the organism fermenting the sugars
originally present in the meat, and is thus to be grouped with
other carbo-hydrate reactions.

Fermentation of Sugars. — As stated on page 80, litmus or
neutral-red peptone water, or dextrose-free bouillon in Durham's
tubes is used, the sugar to be employed being added in the
proportion of half to . one per cent. The fermentative capacities
of the b. coli are very wide. It produces acid and gas in
glucose, lactose, laevulose, galactose, maltose, raffinose, mannite,
dulcite, sorbite, and very frequently in cane sugar (saccharose).1
It produces a similar change in the glucosides, salicin, and
arbutin.

The reactions of b. coli in some media other than simple
sugar solutions likewise depend on sugar fermentation, and of
these are the following : —

Curdling of Milk. — If the b. coli be grown in milk, preferably
litmus milk, acid is produced from the lactose present which
further curdles the milk. If litmus milk be used, the acid
reaction should be permanent when growth is allowed to go on

1 A strain of b. coli fermenting cane sugar was formerly referred to as
b. coli comnmnior, but this differentiating term has been discarded.

CULTURE REACTIONS ON SPECIAL MEDIA 353

for some days. A similar reaction is observed if litmus whey is
used (p. 51).

Measuring of Gas Formation. — As has been said, the gases produced
by the b. coli in fermenting sugars are chiefly carbon dioxide and
hydrogen. Many observers attach considerable importance, first, to the
amount of gas formed from a given quantity of glucose in a given time,
and, second, to the ratios of the two gases to one another, in such a
fermentation. For the observation of this, MacConkey recommends the
following method : fermentation tubes (p. 81, Fig. 36, c), with the
closed limb graduated, containing 2 per cent, peptone (Witte) and 1 per
cent, glucose in tap water, are inoculated and incubated for forty-eight
hours at 37° C. The tube is allowed to cool and the total amount of gas
noted. The bulb is then filled with 2 per cent, sodium hydrate solution,
the opening closed with the thumb and thoroughly shaken. After the
gas has been collected in the closed arm the thumb is removed and the
ratio of the hydrogen left to the original gas volume is read off.

Voges and Proskauer's Reaction. — This is a reaction which is
not given by the classical type of b. coli, but as it occurs with
many members of the coli group it may be described here. It
also depends on carbo-hydrate fermentation. A glucose peptone
solution tube is inoculated and growth allowed to take place for
three days. A solution of caustic potash is added and the tube
allowed to stand for twenty-four hours at room temperature. A
red fluorescent colour is produced, causing the medium to
resemble a weak alcoholic solution of eosin.

B. Action on Neutral-Red. — When b. coli is grown on neutral-
red lactose bouillon, a rosy red colour, the effect of the lactic
acid upon the dye, is at first seen. Frequently this is succeeded
by the appearance of a green fluorescence due to a direct action
of the organism upon the dye. This is evidenced by the fact
that the neutralisation of the lactic acid by an alkali does not
lead to a reproduction of the original alkaline tint in the
indicator. The reaction, however, varies with composition of
the medium, the important factors being the percentage of
sugar and the reaction.

C. Production of Indol. — The b. coli produces indol in
peptone water. The methods have been given on page 82, and
for the detection of the reaction the use of Ehrlich's rosindol
test is preferable ; (if the nitroso-indol reaction be used, a small
quantity of a nitrite must be added). Two peptone tubes
should always l>e inoculated, and if the reaction is not obtainable
in one after two or three days' growth, the other should be
incubated for from six to seven days and then tested. Where a
faint reaction is obtained, it is well to corroborate the presence of
indol by dissolving the rosindol out with amyl-alcohol as described.

354 TYPHOID FEVER

D. Reduction of Nitrates. — The b. coli is frequently capable
of reducing nitrates to nitrites. For this test, Savage recom-
mends the use of a medium made by dissolving 10 grm. of
peptone in 1 litre of ammonia-free distilled water, and adding
2 grm. of nitrite-free potassium nitrate. The medium is filtered,
tubed, and sterilised for half an hour on three days. Tubes
are infected and incubated for forty-eight hours, the forma-
tion of nitrites being now tested for by Ilosvay's method.
The following solutions are required : (a) sulphanilic acid, -5
grm. dissolved in 150 c.c. dilute acetic acid (s.g. 1'04); (6)
1 grm. a-naphthylamine is dissolved in 22 c.c. of water, the
solution filtered, and 180 c.c. dilute acetic acid added. In
using the test, 2 c.c. of each of these solutions is added to 10 c.c.
of culture. If reduction of the nitrates has occurred, a rose
pink colour should develop almost immediately. It is to be
noted that the pink colour first produced sometimes disappears
as it is formed or on shaking ; in such a case further portions
of the two reagents in equal quantities should be added.

Agglutination Eeactions of the B. coli. — When the b. coli has
produced a pathological condition in an animal, the serum of
the infected animal frequently manifests specific agglutinative
characters, especially towards the strain of the organism isolated
from the lesions. Under certain circumstances, also, the serum
of an animal infected by some other member of the b. coli group
may also agglutinate strains of this organism. This subject will
be treated of when we consider the differentiation of the
members of the group one from another.

Isolation of the B. coli. — In the case of abscesses or coli
infection of the kidney or bladder, etc. (p. 356), the isolation of
the organism is usually easy, the use of agar plates being here
sufficient. When, however, the organism is present along with
other bacteria, as in the case of water, sewage, etc., special means
must be adopted, the object of which usually is to inhibit the
growth of all organisms except those belonging to the coli group.
Formerly media containing small quantities of carbolic acid
were used for this purpose, but now the inhibition is usually
effected by the use of certain aniline dyes, by picric acid, or
by bile salts. The media of Conradi-Drigalski, Conradi, Endo,
Fawcus, 'and of MacConkey (pp. 47-51) are examples. All
these media have their uses, and it is best to select that with
which the worker has had most experience. In this country
MacConkey 's bile- salt lactose agar is perhaps most widely used.
The methods of the application of these media and the appear-
ances of b. coli have already been described (p. 47-51).

THE RECOGNITION OF TYPICAL B. COLI 355

The Recognition of typical B. coli. — The work on b. coli,
especially in relation to its occurrence in water, has revealed
the existence of a very large number of varieties of the organism.
These differ from one another in the absence of one or more
of the characters which may be elucidated by the application
of the biological methods given. Considerable difference of
opinion exists as to what characters are to be looked upon as
type characters, i.e. characters shared by the greatest number of
varieties isolated. In this connection it is to be noted that as
the b. coli was originally isolated from the human intestine, and
as the detection of such intestinal bacteria outside the body
constitutes a most important practical question, the inquiry for
type characters is to a certain extent limited to an attempt to
arrive at the type most frequently present in the human intestine.

Two standards may be alluded to. First, that of an English
Committee which reported in 1904 on the standardisation of
methods for the bacterioscopic examination of water. According
to this, the b. coli is a small, motile, non-sporing bacillus, capable
of growing at 37° C., decolorised by Gram, never liquefying
gelatin, producing clot and permanent acidity of milk within
seven days at 37°, fermenting glucose and lactose, with, in both,
acid and gas formation, — subsidiary points being the forma-
tion of indol, the formation of a thick yellowish-brown growth
on potato, production of fluorescence in neutral-red, reduction of
nitrates, and fermentation of saccharose. A similar American
Committee looked upon the typical organism as a non-sporing
bacillus, motile, fermenting dextrose-broth, with the formation
of about 50 per cent, of gas, of which about one-third is carbon
dioxide, causing acid and clot in milk in forty-eight hours, not
liquefying gelatin, producing indol and reducing nitrates.
These two standards differ in the fact that the English Committee
lay less weight on indol formation and the reduction of nitrates.

Generally speaking, the application in any case of the scheme
of the English Committee is to be recommended, and organisms
conforming to the tests laid down may be accepted in the
majority of cases as probably having come from an intestinal
source. The further differentiation of organisms conforming to
this type will be dealt with later (p. 391). Meantime it may be
said that, in addition to the type characters, lactose-fermenters
from the human intestine usually ferment saccharose and dulcite
and have no effect on adonite, inulin and inosite, and it may be,
no influence on mannite.

Pathogenic Properties of the B. coli. — In man, the b.
coli has been found as the only organism present in various

356 TYPHOID FEVER

suppurative conditions (see Chapter .. VII.), especially in con-
nection with the intestine (e.g. appendicitis) and about the
urinary tract. In the latter, it is also responsible for catarrhal
conditions in the pelvis of the kidney and in the bladder, these
being more common in the female, and frequently presenting
chronic characters. As a practical point, it may be said that
the treatment of the latter by vaccines, especially when made
from the strain isolated from the lesion, has been attended with
marked success. The b. coli is also apparently the cause of
some cases of summer diarrhoea (cholera nostras), of infantile
diarrhoea, and of some food poisonings.

The Pathogenicity of the B. coli and its Relation to that of the
Typhoid Bacillus. — Intraperitoneal injection in guinea-pigs is often
fatal. Subcutaneous injection may result in local abscesses, and some-
times in death from cachexia. Sanarelli found that the b. coli isolated
from typhoid stools was much more virulent than when isolated from
the stools of healthy persons. He holds that the increase in virulence is
due to the effect of typhoid toxins. This increased virulence of the
b. coli in the typhoid intestine makes it possible that some of the patho-
logical changes in typhoid may be due, not to the typhoid bacillus, but
to the b. coli. Some of the general symptoms may be intensified by
the absorption of toxic products formed by it and by other organisms.
Differences of behaviour of the two bacilli in connection with their
pathological effects have been brought forward as confirmatory of the
fact of their being distinct species. Thus Sanarelli accustomed the
intestinal mucous membrane of guinea-pigs to toxins derived from an
old culture of the b. coli, by introducing day by day small quantities
of the latter into the stomach. When a relatively large dose could be
tolerated, it was found that the introduction in the same way of a small
quantity of typhoid toxin was still followed by fatal result. Pfeiffer also
found that while the serum of convalescents from typhoid paralysed the
typhoid bacilli, it had no more effect on similar numbers of b. coli than
the serum of healthy men.

THE BACILLUS TYPHOSUS.

Bacillus Typhosus. — Microscopic Appearances. — It is some-
times difficult to find the typhoid bacilli in the organs of a
typhoid patient. Numerous sections of different parts of a
spleen, for example, may be examined before a characteristic
group is found. The best tissues for examination are a Peyer's
patch where ulceration has not yet commenced or where it is
just commencing, the spleen, the liver, or a mesenteric gland.
The spleen and liver are better than the other tissues named, as
in the latter the presence of the b. coli is more frequent. From
scrapings of such solid organs dried films may be prepared and
stained for a few minutes in the cold by any of the strong

APPEARANCES OF CULTURES

357

>taining solutions, e.g. with carbol-thionin-blue, or with Ziehl-
Xerlsen carbol-fuchsin diluted with five parts of distilled wrater.
As a rule, decolorising is not necessary. For the proper
observation of the arrangement of the bacilli in the tissues,
paratlin sections should be stained in carbol-thionin-blue for a
few minutes, or in Loftier'* methylene-blue for one or two hours.
The bacilli take up the stain somewhat slowly, and as they are
also easily decolorised, the aniline-oil method of dehydration
may be used with ad-
vantage (vide p. 100).
In such preparations the
characteristic appearance
to IK- looked for is tilt-
occurrence of groups of
bacilli lying between the
cells of the tissue (Fig.
106). The individual
bacilli are 2 p to 4 //.
Ion;:, with somewhat
oval ends, and '5 \L in
thickness. Sometimes
filaments 8 JJL to IQ p
long may be observed,
though they are less
common than in cultures.
It] is evident that one
of the bacilli may fre-
quently in a section be
viewed endwise, in which
case the appearance will
be circular. This ap-
pearance accounts for some, at least, of the coccus-like forms
which have been described. The bacilli are decolorised by
Grant 'j method.

Isolation and Appearances of Cultures.— To grow the
organism artificially it is best to isolate it from the spleen (for
method, see p. 146) as it exists there in greater numbers than in
the other solid organs, and may be the sole organism present
even some time after death. Agar or gelatin plates or agar stroke
cultures may be employed. On the agar media the growths are
\Mble after twenty-four hours' incubation at 37° C. On agar
plates the superficial colonies are thin and film-like, circular or
slightly irregular at the margins, dull white by reflected light,
bluish-grey by transmitted light. Colonies in the substance of

•Fie. 106. — A large clump of typhoid bacilli
in a spleen. The individual bacilli are only
seen at the periphery of the mass. (In
this spleen enormous numbers of typhoid
bacilli were shown by cultures to be present
in a practically pure condition.)

Paraffin section ; stained with carbol-thionin
blue, x 500.

358

TYPHOID FEVER

the agar are small, and appear as minute round points. Under a
low objective, the surface colonies are found to be very transparent
(requiring a small diaphragm for their definition), finely granular
in appearance, and with a very coarsely crenated and well-
defined margin. The deep colonies are usually spherical, some-
times lenticular in shape, and are smooth or finely granular on
the surface, and more opaque than the superficial colonies. In
cover-glass preparations, the bacilli are found to present the same
microscopic appearances as in preparations from solid organs,
except that there may be a greater number of the longer forms
which may almost be called filaments (Fig. 107). The same is

true of films made from
young gelatin cultures.
Sometimes the diversity
in the length of the
bacilli is such as to throw
doubt on the purity of
the culture. As a general
rule, in a young (twenty -
four or forty-eight hours
old) culture, grown at a
uniform temperature, the
bacilli are plump, and
the protoplasm stains
uniformly. In old cul-
tures, or in cultures
which have been exposed
to changes of tempera-
ture, the protoplasm
stains only in parts ;
there may be an appear-
ance of irregular vacuolation either at the centre or at the ends
of the bacilli.

Motility. — In hanging-drop preparations the bacilli are found
to be actively motile. The smaller forms have a darting or
rolling motion, passing quickly across the field, whilst some show
rapid rotatory motion. The filamentous forms have an undu-
lating or serpentine motion, and move more slowly. Hanging-
drop preparations ought to be made from agar or broth cultures
not more than twenty-four hours old. In older cultures the
movements are less active.

Flagella. — On being stained by the appropriate methods
(vide p. 110), the bacilli are seen to possess many long wavy
fiagella which are attached all along the sides and to the ends

FIG. 107.— Typhoid bacilli, from a young
culture on agar, showing some filamentous
forms.

Stained with weak carbol-fuschin. x 1000.

APPEARANCES OF CULTUKKs

359

(!•'!::. 108). They are more numerous, longer, and more wavy
than those of the b. coli.

Characters of Culture. — Generally speaking, on artificial
media growths of the b. typhosus appear less dense than
those of the b. coli. Stab cultures in peptone gelatin give
a somewhat characteristic appearance. On the surface of the
medium growth spreads outwards from the puncture as a thin
leaf-like film or pellicle, with irregularly wavy margin (Fig.

FIG. 108. — Typhoid bacilli, from a young culture on agar, showing

Hagella. See also Plate III., Fig. 20.
Stained l>y Van Ennengem's method, x 1000.

109, A). It is semi-transparent and of bluish-white colour.
Ultimately this surface growth may reach the wall of the tube.
Not infrequently, however, the surface growth is not well
marked. Along the stab there is an opaque whitish line of
growth, of finely nodose appearance. There is no liquefaction
of the medium, and no formation of gas. In stroke cultures
there is a thin bluish-white film, but it does not spread to such
an extent as in the case of the surface growth of a stab culture
(Fig. 109, B). In gelatin plates also the superficial and deep

360

TYPHOID FEVER

colonies present corresponding differences. The former are
delicate semi-transparent films, with wavy margin, and are
much larger than the colonies in the substance, which appear as
small round points (Fig. 110). These appearances, which are
well seen on the third or fourth day, resemble those seen in agar
plates, as already described in the method of isolation ; but on
gelatin the surface colonies are rather more transparent than

FIG. 109.

A. Stab culture of the typhoid bacillus in gelatin, five days' growth.

B. Stroke culture of the typhoid bacillus on gelatin, six days growth.

C. Stab culture of the bacillus coli in gelatin, nine days' growth"; the gelatin

is split in its lower part owing to the formation of yas.

those on agar. Their characters, as seen under a low power
of the microscope, also correspond. If a gelatin tube be
inoculated and incubated at 37° C., a uniform turbidity is
produced (cf. b. coli, p. 351).

In stroke cultures on agar there is a bluish-grey film of
growth, with fairly regular margins, but without any character-
istic features. This film is loosely attached to the surface, and
can be easily scraped off.

BIOLOGICAL REACTIONS 361

K The growth on potatoes is important. For several days (at
incubation temperature) after inoculation there is apparently no
growth. If looked at obliquely, the surface appears wet, and if
it is scraped with the platinum loop, a glistening track is left ; a
cover-glass preparation shows numerous bacilli. Later, however,
a slight pellicle with a dull, somewhat velvety surface may
appear, and this may even assume a brown appearance. These
characteristic appearances are only seen when a fresh potato
with an acid reaction lias
been used.

In bouillon incubated
at 37° C. for twenty-four
hours there is simply
a uniform turbidity.
Cover-glass preparations
made from such some-
times show filamentous
forms of considerable
length without apparent
segmentation.

CoHf/ifions of (,'rntrtl,,

ttc. — The optimum tem-

I't-nit mv of the typhoid "^^BBBB^^"

bacillus is about 37° C.,

tlmntrh it flkn flourish^* FlG- HO.— Colonies of the typhoid bacillus

(one superficial and three deep) in a gelatin

well at the room tern- plate. Three days' growth at room tem-
perature. It will not perature. x!5.
grow below 9° C. or

above 42° C. Its powers of resistance correspond with those
of most non-sporing bacteria. It is killed by exposure for
half an hour at 60° C., or for two or three minutes at 100° C.
Typhoid bacilli kept in distilled or in ordinary tap water have
usually been found to be dead after three weeks (Frankland).

Biological Reactions. — Very important means of identi-
fying the typhoid bacillus are found in testing its capa-
cities for growth on certain special media. This facilitates
its being differentiated from the b. coli and the other
members of the coli-typhoid group. The following results will
be best appreciated if considered in relation to what is said
regarding these other organisms, as the reactions of the typhoid
bacilli in differentiating media are largely negative. (See
Table, p. 394.)

The testa with sugars are important. The typhoid bacillus
produces acid without gas in maltose, laevulose, glucose, and

362 TYPHOID FEVER

mannite, but originates no change in lactose, cane-sugar, or
dulcite. Further, no gas production is observed in gelatin
shake cultures, and there is no curdling of milk, although in
litmus milk slight acid production occurs ; in a time varying
from a few days to a month the acid change may be succeeded
by alkali production. Under ordinary circumstances, the
typhoid bacillus is incapable of producing indol in peptone-salt
solution, and does not alter neutral-red in lactose bouillon.

A great many special tests were formerly in use for differ-
entiating the b. typhosus from the b. coli. The use of these
is not now so necessary, but the following may be described : —

The Media of Capaldi and Proskauer.—The first of these ("No. 1 ") is
a medium free of albumin, in which b. coli grows well and freely produces
acid, while the typhoid bacillus hardly grows at all, and certainly will
produce no change in the reaction. Its composition is as follows :
asparagin '2 parts, mannite '2, sodium chloride '02, magnesium sulphate
'01, calcium chloride '02, potassium monophosphate '2, distilled water
to 100 parts. The second medium ("No. 2") contains albumin, and
in it the b. coli produces no acid, while the typhoid bacillus grows well
and produces an acid reaction. It consists of Witte's peptone 2 parts,
mannite '1, distilled water to 100 parts. After the constituents of each
medium are mixed and dissolved, it is steamed for one and a half hours
and then made neutral to litmus — the first medium, being usually
naturally acid, by sodium hydrate, the second, being usually alkaline, by
citric acid. The medium is then filtered, filled into tubes containing
5 c.c., and these are sterilised. After inoculation for twenty hours the
reaction of the medium is tested by adding litmus.

The identification of the typhoid bacillus is best facilitated
by means of agglutination reactions which will be treated of
later (pp. 371, 393).

Pathological Changes in Typhoid Fever. — Here wre confine
our attention solely to the bacteriological aspects of the disease.
The inflammation and ulceration in the Peyers patches and
solitary glands of the intestine are the central features. In the
early stage there is produced an acute inflammatory condition,
attended with extensive leucocytic emigration and sometimes
with small haemorrhages. At this period the typhoid bacilli are
most numerous in the patches, groups being easily found between
the cells. The subsequent necrosis is evidently in chief part
the result of the action of the toxic products of the bacilli,
which gradually disappear from- their former positions, though
they may still be found in the deeper tissues and at the spread-
ing margin of the necrosed area. They also occur in the lym-
phatic spaces of the muscular coat. It is to be remarked that
the number of the ulcers arising in the course of a case bears

PATHOLOGICAL CHANGES 363

no relation to its severity. Small ulcers may occur in the
Ivmphoid follicles of the large intestine.

The me&enteric glands corresponding to the affected part of
the intestine are usually enlarged, sometimes to a very great
extent, the whole mesentery being filled with glandular masses.
In such glands there may be acute inflammation, and occasion-
ally necrosis in patches occurs. Sometimes on section the
glands are of a pale-yellowish colour, the contents being diffluent
and consisting largely of leucocytes. Typhoid bacilli may be
isolated both from the glands and the lymphatics connected with
them, but the b. coli is in addition often present.

The spleen is enlarged, — on section usually of a fairly firm
consistence, of a reddish-pink colour, and in a state of conges-
tion. Of all the solid organs it usually contains the bacilli in
greatest numbers. They can be seen in sections, occurring in
clumps between the cells, there being no evidence of local
reaction round them (Fig. 106). Similar clumps may occur in
the liver in any situation, and without any local reaction. In
this organ, however, there are often small foci of leucocytic
infiltration, in which, so far as our experience goes, bacilli
cannot be demonstrated. The bacillus is found, often in large
numbers, in the gall-bladder, where it may persist for years
(vide infra). Clumps of bacilli may also occur in the kidney.

In addition to these local changes in the solid organs, there are also
\\ i(lc>j)ic;i(l cellular degenerations in the solid organs which suggest the
action of toxic products.

In the lunys there may be bronchitis, patches of congestion and of
acute broncho-pneumonia. In these, typhoid bacilli may sometimes be
observed, but evidence of a toxic action depressing the powers of resist-
ance of the lung tissue is found in the fact that the pneumococcus
frequently occurs in such complications of typhoid fever.

The nervous system shows little change, though meningitis associated
either with the typhoid bacillus, with the b. coli, or with the strepto-
coccus pyogenes has been observed.

In typhoid fever the bacilli can in 90 per cent, of cases be isolated from
the blood during the course of the illness. The local lesions are thus associ-
ated with a general septica.jmic process. The bacilli have been found in
the rofteolar spots which occur in typhoid fever, but it cannot he yet
stated that such spots are always due to the presence of the bacilli. The
fact that the typhoid bacilli are usually confined to certain organs and
tissues shows that they probably have a selective action on certain tissues.

To sum up the pathology of typhoid fever, we have in it a
disease the centre of which lies in the lymphoid tissue in and
connected with the intestine. In this situation we must have
an irritant, against which the inflammatory reaction is set up,

364 TYPHOID FEVER

and which in the intestine is sufficiently powerful to cause
necrosis. The affections of the other organs of the body suggest
the circulation in the blood of poisonous substances capable of
depressing cellular vitality, and producing histological changes.
The occurrence of bacilli in the blood and organs links typhoid
fever with septicsemic processes.

Suppuration occurring in connection with Typhoid Fever.—
With regard to the relation of the typhoid bacillus to such
conditions, statements as to its isolation from pus, etc., can be
accepted only when all the points available for the diagnosis of
the organism have been attended to. On this understanding the
following summary may be given : — In a certain proportion of
the cases examined the typhoid bacillus has been the only organ-
ism found. This has been the case in subcutaneous abscesses,
in suppurative periostitis, suppuration in the parotid, abscesses
in the kidneys, etc., and probably also in one or two cases of
ulcerative endocarditis. But in the majority of cases other
organisms, especially the b. coli and the pyogenic micrococci,
have been obtained, the typhoid bacillus having been searched for
in vain. It has, moreover, been experimentally shown, notably
by Dmochowski and Janowski, that suppuration can be experi-
mentally produced by injection in animals, especially in rabbits,
of pure cultures of the typhoid bacillus, the occurrence of sup-
puration being favoured by conditions of depressed vitality, etc.
These observers also found that when typhoid bacilli were
injected along with pyogenic staphylococci, the former died out
in the pus more quickly than the latter. Accordingly, in clinical
cases where the typhoid bacillus is present alone, it is improbable
that other organisms were present at an earlier date.

Occurrence of Gallstones in those who have suffered from
Typhoid Fever. — As has been stated, foci of bacilli occur in the
liver in typhoid fever, and these bacilli are excreted with the
bile. In the gall-bladder they apparently not infrequently set
up a catarrhal process in the biliary ducts and gall-bladder
(cholecystitis typhosa\ and are then in a better position for multi-
plication, in consequence of the presence of albuminous catarrhal
secretions. There is evidence that the bacilli may persist in the
gall-bladder for many years, and probably the catarrhal inflam-
mation which they keep up is responsible for many of the cases
of gallstones wrhich occur — the albuminous matter produced
causing a deposit of the bile in a solid form. Typhoid bacilli
have actually been isolated from cases of gallstones operated on
years after an attack of typhoid fever, and the bacilli have even
been found within the calculi. They have also been demon-

PATHOGENIC EFFECTS OF B. TYPHOSUS 365

strated in chronic suppurations occurring in the gall-bladder.
It is to be noted that gallstones are more frequently found in
women than in men, the proportion being about four to one,
and probably a considerable proportion of the total number of
cases of gallstones are traceable to the previous occurrence of
typhoid fever.

Pathogenic Effects produced in Animals by the Typhoid
Bacillus. — There is no disease of animals which can be said
to be identical with typhoid, nor is there any evidence of the
occurrence of the typhoid bacillus under ordinary pathological
conditions in the bodies of animals. Attempts to communicate
the disease to animals by feeding them on typhoid dejecta have
been unsuccessful, and though pathogenic effects have been
produced by introducing pure cultures in food, the disease has
usually borne no resemblance to human typhoid. The most
successful experiments have been those of Remlinger, who, by
continuously feeding rabbits on vegetables soaked in water con-
taining typhoid bacilli, produced in certain cases symptoms
resembling those of typhoid fever (diarrhrea, remittent pyrexia,
etc.). An agglutinating action was observed in the serum, and
post mortem there was congestion of the Peyer's patches, and
typhoid bacilli were isolated from the spleen.

Feeding experiments are thus unsatisfactory, and the same
may be said of the results of subcutaneous or intraperitoneal
infection. Here, again, pathogenic effects can easily be produced
by the typhoid bacillus, but these effects are of the nature of a
short acute illness characterised by pyrexia, rapid loss of weight,
inability to take food, and frequently ending fatally in from
twenty-four to forty-eight hours. The type of disease is thus
very different from what occurs naturally in man. In such
injection experiments the results vary considerably, sometimes
scarcely any effect being produced by a large dose of a culture.
This is no doubt due to the fact that different strains of the
bacillus vary much in virulence. Ordinary laboratory cultures
are often almost non-pathogenic. They can, however, be made
virulent in various ways. Sanarelli used the method of injecting
sterilised cultures of the b. coli intraperitoneally at the same
time as the typhoid bacillus was introduced subcutaneously.
After this procedure had been repeated through a series of
animals, a typhoid culture of exalted virulence was obtained.
Sidney Martin has obtained virulent cultures by passing bacilli,
derived directly from the spleen of a person dead of typhoid
fever, through the peritoneal cavities of a series of guinea-pigs.

Sanarelli, studying the effects of the intraperitoneal injection

366 TYPHOID FEVER

of a few drops of a culture of highly exalted virulence, found
that the Peyer's patches and solitary glands of the intestine were
enormously infiltrated, sometimes almost purulent, and that they
contained typhoid bacilli, as also did the mesenteric lymphatics
and glands, and the spleen. These results are interesting, but
have not been confirmed.

The Toxic Products of the Typhoid Bacillus. — Here very
little light has been thrown on the pathology of the disease, but
the general results may be outlined. We may state that there
exist in the bodies of typhoid bacilli toxic substances, that in
artificial cultures these do not pass to any great degree out into
the surrounding medium, and that though they produce effects
on the intestine, there is evidence that such effects are not
characteristic and not peculiar to the toxins of the b. typhosus.
Sidney Martin found that the bodies of bacteria killed by chloro-
form vapour were very toxic, — more so than filtered cultures.
Diarrhoea was a constant symptom after injection, but no change
in the Peyerian patches was observed. He also found that
virulent cultures of the b. coli gave similar results when
similarly treated. Allan Macfadyen, by grinding up typhoid
bacilli frozen solid by liquid air, produced a fluid whose toxic
effect he attributed to the presence of the intracellular poisons.

The Immunisation of Animals against the Typhoid
Bacillus. — Earlier observers had been successful in accustoming
mice to the typhoid bacillus by the successive injections of
small* and gradually increasing doses of living cultures of the
bacillus. Later, Brieger, Kitasato, and Wassermann found that
the bacillus when modified by being grown in a bouillon made
from an extract of the thymus gland no longer killed mice and
guinea-pigs. These animals after injection were moreover
immune, and it was also found that the serum of a guinea-
pig thus immunised could, if transferred to another guinea-pig,
protect the latter from the subsequent injection of a dose of
typhoid bacilli to which it would naturally succumb. Chante-
messe and Widal, Sanarelli, and also Pfeiffer, succeeded in
immunising guinea-pigs against the subsequent intraperitoneal
injection of virulent living typhoid bacilli, by repeated and
gradually increasing intraperitoneal or subcutaneous doses of
dead typhoid cultures in bouillon. Experiments performed with
serum derived from typhoid patients and convalescents indicate
that similar effects occur in those who have successfully resisted
the natural disease. The serum of such patients has antibacterial
powers, but there is no evidence that it contains any antitoxic
bodies (see chapter on Immunity). Pfeiffer, for example, found,

RELATIONSHIP TO TYPHOID FEVER 367

on adding serum from typhoid convalescents to typhoid bacilli
killed by heat, and injecting the mixture into guinea-pigs, that
death took place as in control animals which had received these
toxic agents alone. Pfeiffer also found that by using the serum
of immunised goats, he could, to a certain extent, protect other
animals against the subsequent injection of virulent living
typhoid bacilli. On trying to use the agent in a curative way,
i.e. injecting it only after the bacilli had begun to produce their
effects, he got little or no result.

General View of the Kelationship of the B. typhosus to
Typhoid Fever. — 1. We see in typhoid fever a disease having
its centre in and about the intestine, and acting secondarily on
many other parts of the body. In the parts most affected there
is always a bacillus present, microscopically resembling other
bacilli, es[>ecially the b. coli, which is a normal inhabitant of
the animal intestine. The bacillus can be isolated from the
characteristic lesions of the disease and from other parts of the
body as described, and further, it is found by culture and serum
reactions to differ from other organisms. Here the important
point is that a bacillus giving all the reactions of the typhoid
bacillus has never been isolated except from cases of typhoid
fever, or under circumstances that make it possible for the
bacillus in question to have been derived from a case of typhoid
fever.

2. A difficulty in the way of accepting the etiological relation-
ship of the b. typhosus lies in the comparative failure of
attempts to cause the disease in animals. We have noted,
however, that in nature animals do not suffer from typhoid
fever.

3. The observations of Pfeiffer and others on the protective
power against typhoid bacilli shown, on testing in animals, to
belong to the serum of typhoid patients and convalescents, and
the peculiar action of such serum in immobilising and causing
clumping of the bacilli (vide infra), are also of great importance
as indicating an etiological relationship between the bacillus and
the disease. Additional important evidence is found in the fact
that vaccination by means of the dead bacilli (vide infra) has
a marked effect in preventing the disease from arising in a
population exposed to infection, and also in lowering the
mortality when the fever attacks those who have been inoculated.
These facts may thus be accepted as indirect but practically
conclusive evidence of the pathogenic relationships of the
typhoid bacillus to the disease.

According to our present results, we must thus hold that

368 TYPHOID FEVER

the b. typhosus constitutes a distinct species of bacterium, and
that it is the cause of typhoid fever. Evidence of an important
nature confirmatory of this view is, we think, found in the fact
that cases have occurred where bacteriologists have accidentally
infected themselves by the mouth with pure cultures of the
typhoid bacillus, and after the usual incubation period have
developed typhoid fever. Several cases of this kind have been
brought to our notice, and are not, we think, vitiated by the fact
that other similar instances have occurred without the subsequent
development of illness. These latter would be accounted for by
a low degree of susceptibility on the part of the individual or to
a want of pathogenicity in the cultures.

As there is thus strong evidence of the etiological relationship
of the typhoid bacillus to typhoid fever, the view of the
development of the disease usually taken has been that the
bacilli, being ingested, multiply in the intestinal tract, cause
inflammation and necrosis of the lymphoid tissue, and, gaining
an entrance to the general circulation, produce the septicsemic
phenomena which we have described.

Within recent years, considerable attention has been attracted to
another view of the course of infection put forward by Forster and
his co-workers in Strasburg. According to this, the process is primarily
a septicaemia, and the intestinal manifestations are looked on as
secondary. The bacilli are supposed to gain entrance to the circulation
possibly through the tonsils, sore throat being a not uncommon initial
symptom of typhoid fever. In the blood they multiply, and, passing
through the liver, gain access to the gall-bladder, set up a catarrhal
inflammation there on the products of which they flourish, and thence
pass out to infect the intestine. The intestinal lesions are either due
to an elective action of bacteria brought by the blood, or come from
infection by the bacilli which pass out from the gall-bladder, — the
former being apparently the alternative to which Forster leans. The
evidence on which this view is based consists, firstly, in the results
of animal experiments in which bacilli introduced intravenously have
been subsequently found chiefly or solely in the gall-bladder, — it may be,
persisting there for weeks. Further, it is stated that bacilli can be
isolated from the blood during the later parts of the incubation stage of
the disease, and before they can be demonstrated in the intestine, where
they are said not to appear until sometime during the first week of
active disease. And again it is stated that in the bodies of persons
dying from typhoid fever, while bacilli are always present in the gall-
badder and in the upper parts of the small intestine, they are frequently
absent from the lower part of the latter and from the colon. It cannot
be said that this view of the disease has been satisfactorily established.
Opinion differs as to the alleged late appearance of the bacilli in the
intestine, and the infectivity noticed during the incubation stage must be
explained. Further, there is strong reason for believing that multiplication
of the bacilli in the intestine can take place. The evidence of this rests
on the finding of bacilli, it may be in considerable numbers, in the faeces

TYPHOID CARRIERS 369

and even in the blood of healthy individuals who have merely been in
contact with typhoid cases or typhoid carriers, and who show no
symptoms of the disease.

There is evidence that certain individuals are relatively
insusceptible to typhoid fever. The cases of the occurrence
of typhoid bacilli in the healthy intestine support this view, and
it has been further shown that during an epidemic certain
persons may suffer from slight intestinal symptoms with typhoid
bacilli in the faeces without the disease going through its usual
course. The so-called " ambulatory " cases of typhoid fever
form a link between these mild infections and fully developed
typhoid fever.

The Epidemiology of Typhoid Fever. — Generally speaking,
the former prevalence of typhoid fever and the periodic outbreaks
which still occur even in well-regulated communities, have de-
pended on the capacity of the typhoid bacillus to live and it
may be to multiply outside the human body. The investigation
of the prevalence of the typhoid bacillus under such saprophytic
conditions is a matter of great difficulty, as, for its proper study,
the capacity of the organism to multiply when other intestinal
and putrefactive organisms are present constitutes the essential
problem. Enough is known, however, to show that the bacillus
can remain viable under such circumstances for some days, and
it may be wreeks. This fact explains the occurrence of the
epidemics due to water, and sometimes, it may be, to milk
supplies becoming contaminated with the excreta from typhoid
patients. Where surface wells are used, and where sewage,
instead of being properly disposed of, finds its way into ash-heaps
or cesspools, the way is opened up for communities becoming
infected with typhoid fever.

Typhoid Carriers. — In the great majority of cases of typhoid
fever, the bacilli disappear from the faeces within from two to
ten weeks of convalescence, but in a certain proportion of cases,
probably about 2 to 4 per cent., evidence is found of the
persistence of the bacilli for many months, and in certain cases
their existence has been ileiiioii-tratrd even thirty and, it may be,
fifty years after the attack of illness. Persons in whom this
phenomenon is present are a constant danger to those around
them, as the infectivity of the bacilli frequently remains, and
during recent years the importance of such " chronic " carriers
has been recognised as explaining many outbreaks of the disease.
The cases traceable to such an origin are of the type usually
< -lav-Til a- sporadic. They arise amongst persons associated with
carrier-, r-;|»eeially when the latter are concerned in the prepara-
24

370 TYPHOID FEVER

tion of food. From time to time, however, larger epidemics
have arisen from a carrier having contaminated a milk supply in
a dairy. The site of the multiplication of the bacteria in a great
many of these carriers is probably the gall-bladder (see p. 364).
As has been stated, the typhoid bacilli may persist there for
many years, often giving rise to gallstones. The fact that women
appear to be more liable to gallstones than men constitutes
a serious factor in relation to the problem of the typhoid carrier,
as women are more concerned in the preparation of food. An
additional danger lies in the fact that carriers usually appear to
be in perfect health or may only suffer from slight, and to them
unimportant, pains in the region of the gall-bladder, it being
well known that in only a proportion of patients suffering from
gallstones do severe symptoms arise. An additional factor in
the carrier problem lies in the fact stated above, that apparently
certain persons ingest the typhoid bacilli, and the latter may
multiply for some months in the intestinal tract without giving
rise to typhoid fever. Such persons have been referred to as
" paradoxical " carriers ; they represent those who either are
naturally insusceptible to typhoid fever or who have developed
immunity in consequence of a previous attack ; they may con-
stitute a danger to susceptible persons with whom they may
come in contact. The most serious danger to a community
arises, however, from the " chronic " carrier. In certain carriers,
the focus of multiplication of the typhoid bacillus may not be
the bowel but the kidney or bladder, the bacilli in such cases
passing out in the urine.

The tracking down of 'a typhoid carrier constitutes an impor-
tant and difficult problem. Firstly, the serum of all suspicious
persons ought to be subjected to the Widal test (vide infra).
Usually speaking the carrier gives a positive reaction, but
sometimes this is absent and sometimes is only obtained with a
low dilution of the serum. Further, it has been shown in
chronic carriers that the agglutinating capacity of the serum
varies from time to time and sometimes may be absent. The
proof of the presence of a carrier lies essentially in the isolation
of the typhoid bacillus from the faeces or the urine, and it is to
be noted that, especially in the former, the organism is not
constantly present, — in certain cases months of remission have
been recorded. This of course may be due to the difficulties of
the search, but whatever the explanation, it necessitates repeated
examinations. Much work has been directed to the question of
freeing the typhoid carrier from the organism, but although
various methods, such as intestinal antisepsis, vaccination,

SERUM DIAGNOSIS 371

excision of the gall-bladder, have been tried, success has hitherto
not been attained. From the public health standpoint, the
prevention of carriers from occurring in a population has been
considered, and it is a question whether in fever hospitals
means ought not to be taken for retaining convalescents from
typhoid until the bodily discharges are free from the typhoid
bacillus. This 1ms already been undertaken in the British army
in India.

The Serum Diagnosis of Typhoid Fever. — This method of
diagnosis is based on the fact that living and actively motile
typhoid bacilli, if placed in the diluted serum of a patient suffer-
ing from typhoid fever, within a very short time lose their
motility and become aggregated into clumps.

The methods by which the test can be applied have already
been described (p. 118).

(1) It will be there seen that the loss of motility and clumping
may be observed microscopically. If a preparation be made by
the method detailed (typhoid serum in a dilution of, say, 1 : 30
having been employed), and examined at once under the micro-
scope,'the bacilli will usually be found actively motile, darting
about in all directions. In a short time, however, these move-
ments gradually become slower, the bacilli begin to adhere to one
another, and ultimately become completely immobile and form
clumps by their aggregation. When this occurs the reaction
is said to be complete. If the clumps be watched still longer a
swelling up of the bacilli will be observed, with a granulation
of the protoplasm, so that their forms can with difficulty be
recognised. In a preparation similarly made with non-typhoid
serum the individual bacilli can be observed separate and
actively motile for many hours.

(2) A corresponding reaction visible to the naked eye is
obtained by the " sedimentation test," the method of applying
which has also been described (p. 120). The test in this form
has the disadvantage of taking longer time than the microscopic
method, but it is useful as a control ; in nature it is similar.

Such is what occurs in the case of a typical reaction. The
value of the method as a means of diagnosis largely depends
on attention to several details. The race of typhoid bacillus
employed is important. All races do not give uniformly the
same results, though it is not known on what this difference of
susceptibility depends. A race must therefore be selected
which gives the best result in the greatest number of undoubted
cases of typhoid fever, and which gives as little reaction a#
with normal sera or sera derived from other diseases.

372 TYPHOID FEVER

This latter point is important, as some races react very readily
to non-typhoid sera. Again, care must be taken as to the state of
the culture used. The suitability of a culture may be impaired
by varying the conditions of its growth. Continued growth of a
race at 37° C. makes it less suitable for use in the test, as the
bacilli tend naturally to adhere in clumps, which may be
mistaken for those produced by the reaction. Wyatt Johnson
recommended that the stock culture should be kept growing on
agar at room temperature and maintained by agar sub-cultures
made once a month. For use in applying the test, bouillon
sub-cultures are made and incubated for twenty-four hours at
37° C. The relation of the dilution of the serum to the
occurrence of clumping is most important. It has been found
that if the degree of dilution be too small a non-typhoid serum
may cause clumping. If possible, observations should always be
made with dilutions of 1 : 10, 1 : 30, 1 : 60, 1 : 100. To speak
generally, the more dilute the serum the longer time is necessary
for a complete reaction. Some typhoid sera have, however,
very powerful agglutinating properties, and may in a compara-
tively short time produce a reaction when diluted many hundreds
of times. With a too dilute serum not only may the reaction
be delayed, but it may be incomplete, — the clumps formed being
small and many bacilli being left free. These latter may either
have been rendered motionless or they may still be motile. No
diagnosis is conclusive which is founded on the occurrence of
such an incomplete clumping alone. Seeing that low dilutions
sometimes give a reaction with non-typhoid sera, it is important
to know what is the highest dilution at which complete
clumping indicates a positive reaction. The general consensus
of opinion, with which our own experience agrees, is that when
a serum in a dilution of 1 : 30 causes complete clumping in half
an hour, it may safely be said that it has been derived from a
case of typhoid fever. Suspicion should be entertained as to
the diagnosis if a lower dilution is required, or if a longer time
is required.

The reaction given by the serum in typhoid fever usually
begins to be observed about the seventh day of the disease,
though occasionally it has been found as early as the fifth day,
and sometimes it does not appear till the third week or later.
Usually it becomes gradually more marked as the disease
advances, and it is still given by the blood of convalescents from
typhoid, but cases occur in which it may permanently disappear
before convalescence sets in. How long it lasts after the end of
the disease has not yet been fully determined, but in many cases

SERUM DIAGNOSIS 373

it has been found after several months or longer. As a rule, up
to a certain point, the reaction is more marked where the fever
is of a pronounced character, whilst in the milder cases it is less
pronounced. In certain grave cases, however, the reaction has
been found to be feeble or almost absent. In some cases, which
from the clinical symptoms were almost certainly typhoid, the
reaction has apparently been found to be absent. Such cases
should always be investigated, from the point of view of their
possibly being paratyphoid fever.

It has been found that the reaction is not only obtained with
living bacilli, but in certain circumstances also with bacilli
that have been killed by heating at 60° C. for an hour, — if a
higher temperature be used, sensitiveness to agglutination is
impaired. Dreyer has introduced a simple technique which
enables an ordinary practitioner provided with dead cultures to
carry out the test for himself. The capacity is also still retained
if a germicide be employed. Here Widal recommends the
addition of one drop of formalin to 150 drops of culture. The
reaction, however, tends to be less complete.

Besides the blood serum, it has been found that the reaction
is given in cases of typhoid fever by pericardial and pleural
effusions, by the bile and by the milk, and also to a slight
degree by the urine. The blood of a foetus may have little
agglutinating effect, though that of its mother may have given
a well-marked reaction ; sometimes, however, the foetal blood
gives a well-marked reaction. It may here also be mentioned
that a serum will stand exposure for an hour at 58° C. without
having its agglutinating }>ower much diminished. Higher
temperatures, however, cause the proj>erty to be lost.

The Agglutination of Organisms other than the B. TyphoxuK
ly Typhoid Serum. — It was at first thought that the reaction in
typhoid fever would afford a reliable method of distinguishing
the typhoid bacillus from the b. coli. Though many races of
the latter give no reaction with a typhoid serum, there are others
which react positively. Usually, however, a lower dilution and
a longer time are required for a result to be obtained, and the
reaction is often incomplete. It has also been found that other
organisms belonging to the typhoid group (v. p. 382) react in a
similar way. The reaction as a method of distinguishing between
these forms is thus not absolutely reliable, but in certain cases
it is of great value in giving confirmation to other tests. The
important point here is the determination of the highest dilution
with which clumping is obtained (for methods, see p. 120).
There is a point in this connection regarding which further

374 TYPHOID FEVER

light is required. Many races of b. coli in use have been
isolated from typhoid cases, and we as yet do not know what
effect this circumstance may have on its subsequent sensitive-
ness to agglutination by typhoid serum. Again, Christophers
has pointed out that a large proportion of serum from normal
persons or from those suffering from diseases other than typhoid
will clump the b. coli in dilutions of from 1 : 20 to 1 : 200, and
no doubt many of the reactions shown by typhoid sera towards
b. coli are due to the pre-existence in the individuals of an
agglutinative property towards the latter bacillus.

With regard to the value of the serum reaction there is little
doubt. In nearly 95 per cent, of cases of typhoid it can be
obtained in such a form that no difficulty is experienced if the
precautions detailed above are observed. The causes of possible
error may be summarised as follows : The serum of the person
may naturally have the capacity of clumping typhoid bacilli ;
there may have been an attack of typhoid fever previously with
persistence of agglutinative capacity ; the case may be one of
disease caused by an allied bacillus ; the disease may have a
quite different cause, and yet the serum may react with typhoid
bacilli ; the disease may be typhoid fever and yet no reaction
may occur. The most important of these sources of error is that
with which diseases caused by allied organisms are concerned,
as it is probable that all the forms which these take in men
have not been recognised. The very wide application of the
reaction has elicited the fact that it is given in many cases of
slight, transient, and ill-defined febriculte, which occur especially
when typhoid fever is prevalent. Some of these may be aborted
typhoid, some may be paratyphoid. There is no doubt that, if
all the facts are taken into account, the cases where the reaction
gives undoubtedly correct information so far outnumber those in
which an error may be made that it must be looked on as a
most valuable aid to diagnosis. In conclusion, here we may say
that the fact of a typhoid serum clumping allied bacilli in no
way, so far as our present knowledge goes, justifies doubt being
cast on the specific relation of the typhoid bacillus to typhoid
fever.

In connection with the phenomenon that a serum either from
a normal person or a typhoid patient may clump several varieties
of bacteria, some points arise. The theoretical consideration of
agglutination is reserved for the chapter on Immunity, but here
it -may be said that agglutinating properties may be present
normally in a serum or they may be originated by an animal
baing infected with a particular bacterium, As the result of

VACCINATION AGAINST TYPHOID 375

injecting a bacterium, not only may agglutinins capable of acting
on that bacterium appear in the serum, but the serum may
become capable of agglutinating other, and especially kindred,
bacteria ; further, any normal agglutinins for the infecting
bacterium present in the serum may be increased in amount.
The agglutinin acting on the infecting organism has been called
the primary or homologous agglutinin, while the others have
been called the secondary or heterologous agglutinins. But
besides what we know to be a fact, that infection by a single
bacillary species can originate agglutinins acting both on itself
and on allied species, we must consider the possibility of
infections by more than one species occurring in an animal, e.g.
b. typhosus with b. coli or with b. paratyphosus (vide infra). In
such a case each organism may originate its primary agglutinin,
so that the presence of multiple agglutinins in a serum may
really be an indication of a mixed infection. Some attention
has been directed to the diagnosis and differentiation of these
conditions. Castellani introduced the absorption method for
their investigation (for method, see p. 121), and by this means
studied the primary and secondary agglutinins produced in
infections in rabbits ; he found that when an animal had been
infected with b. typhosus this organism would absorb from its
serum not only the primary typhoid agglutinins but also such
secondary agglutinins as those acting on the b. coli. If, how-
ever, an animal had undergone infection with, say, both the
b. typhosus and the b. coli, then the b. typhosus could not absorb
from its serum the b. coli (primary) agglutiuin. Castellani thus
put forward the view that by this means primary could be
differentiated from secondary agglutinins, and therefore pure
could be differentiated from mixed infections. There is little
doubt that this view possesses considerable validity, though it is
probably not of universal applicability. Safe deductions can
only be drawn when any serum is tested with several species of
fairly closely related organisms, such as those of the coli group.
Especially is it necessary that the highest dilutions in which
agglutination occurs should be compared. If such precautions
be adopted, the absorption method can be utilised for the differ-
entiation of the typhoid and paratyphoid organisms and their
infections, and for similar investigations.

Vaccination against Typhoid. — The principles of the im-
munisation of animals against typhoid bacilli have been applied
by Wright and Semple to man in the following way : — Typhoid
bacilli are obtained of such virulence that a quarter of a twenty-
four hours' old sloped agar culture when administered hypo-

376 TYPHOID FEVER

dermically will kill a guinea-pig of from 350 to 400 grammes.
Vaccination can be accomplished by such a culture emulsified in
bouillon, and killed by heating for five minutes at 60° C. For
use, from one-twentieth to one-fourth of the dead culture is
injected hypodermically, usually in the flank. The vaccine now
used, however, actually consists of a portion of a bouillon culture
similarly treated (see p. 133). The effects of the injection are
some tenderness locally and in the adjacent lymphatic glands,
and it may be local swelling, all of which come on in a few
hours, and may be accompanied by a general feeling of restless-
ness and a rise of temperature, but the illness is over in
twenty-four hours. During the next ten days the blood of the
individual begins to manifest, when tested, an agglutination
reaction, and further, Wright has found that usually after the
injection there is a marked increase in the capacity of the blood
serum to kill the typhoid bacillus in vitro. These observations,
there is little doubt, indicate that the vaccinated person possesses
a degree of immunity against the bacillus, a conclusion borne
out by the results obtained in the use of the vaccine as a
prophylactic against typhoid fever. Extensive observations
have been made in the British army in India, and in the South
African War the efficacy of the treatment was put to test.
Though in isolated cases not much difference was observed
among those treated as compared with those untreated, yet the
broad general result may be said to leave little doubt that on the
one hand protective inoculation diminishes the tendency for
the individual to contract typhoid fever, and, on the other, if the
disease be contracted, the likelihood of its having a fatal result
is diminished. Thus, in India, of 4502 soldiers inoculated, '98
per cent, contracted typhoid, while of 25,851 soldiers in the
same stations who were not inoculated, 2 '54 per cent, took the
disease. In Ladysmith during the siege there were 1705
soldiers inoculated, among whom 2 per cent, of cases occurred,
and 10,529 uninoculated, among whom 14 per cent, suffered
from typhoid. Wright has collected statistics dealing in all
with 49,600 individuals, of whom 8600 were inoculated, and
showed a case incidence of 2 '25 per cent., with a case mortality
of 12 per cent. ; in the remaining 41,000 uninoculated the case
incidence was 5 '75 per cent, and the case mortality 21 per cent.
The best results seemed to be obtained when ten days after
the first inoculation a second similar inoculation was practised.
Weight has found that in certain cases immediately after
inoculation there is a fall in the bactericidal power of the blood
(negative phase), and he is of opinion that this indicates a

METHODS OF EXAMINATION 377

temporary increased susceptibility to the disease. He therefore
recommends that when possible the vaccination should be carried
out some time previous to the exposure to infection. There can
be little doubt that in this method an important prophylactic
measure has been discovered.

Vaccine Treatment of Typhoid Fever. — As in the case of
other acute infections, vaccines have been recently used in the
treatment of typhoid fever during the acute stage (Leishman
and Smallman). The method is to inject hypodermically 100
million dead typhoid bacilli, i.e. a fifth of the first dose used for
the protective inoculation. If the temperature shows a tendency
to fall, this may be repeated about every four days. The results
obtained are hopeful, and justify the method being further
applied.

Antityphoid Serum. — Chanteim-sse has immunised animals with dead
cultures of the typhoid bacillus, and, having found that their sera had
protective and curative effects in other animals, lias used such sera
in human cases of typhoid with apparent good result. In the hands
of others, however, such a line of treatment has not been equally
successful.

Methods of Examination. — The methods of miscroscopic
examination, and of isolation of typhoid bacilli from the spleen
]*>st mortem, have already been described. They may be isolated
from the Peyer's patches, lymphatic glands, etc., by a similar
method.

During life, typhoid bacilli may be obtained in culture in the
following ways : —

(a) From the Blood. — The typhoid bacillus can often be
isolated from the blood, especially during the first week, by
ordinary methods (see p. 72). A special method has also been
used with success. In this 5 c.c. of blood are placed in 10 c.c.
of sterilised ox bile. The mixture is incubated for from twenty-
four hours to a week, and from time to time the presence of the
bacillus is tested for by sub-culturing on such media as those of
Conradi or MacConkey.

(l») From the, Spleen. — This is the most certain method of
obtaining the typhoid bacillus during the continuance of a case.
The skin over the spleen is purified, and, a sterile hypodermic
syringe being plunged into the organ, there is withdrawn from
the splenic pulp a droplet of fluid, from which plates are made.
In a large proportion of cases of typhoid the bacillus may be
thus obtained, failure only occurring when the needle does not
hap]K'n to touch a bacillus. Numerous observations have shown

378 TYPHOID FEVER

that, provided the needle be not too large, the procedure is quite
safe. Its use, however, is scarcely called for.

(c) From the Urine. — Typhoid bacilli are present in the
urine in at least 25 per cent, of cases, especially late in the
disease, probably chiefly when there are groups in the kidney
substance. For methods of examining suspected urine, see
p. 74.

(d) From the Stools. — During the first ten days of a case of
typhoid fever, the bacilli can be isolated from the stools by the
ordinary plate methods — preferably in McConkey's lactose bile-
salt neutral-red agar, or in the other media described on pp.
47-53. After that period, though the continued infectiveness of
the disease indicates that they are still present, their isolation is
difficult. We have seen that after ulceration is fairly estab-
lished by the sloughing of the necrosed tissue, the numbers
present in the patches are much diminished, and therefore there
are fewer cast off into the intestinal lumen, and that in addition
there is a correspondingly great increase of the b. coli, which
thus causes any typhoid bacilli in a plate to be quite outgrown.
From the fact that the ulcers in a case of typhoid may be very
few in number, it is evident that there may be at no time very
many typhoid bacilli in the intestine. The microscopic examina-
tion of the stools is of course useless as a means of diagnosing
the presence of the typhoid bacillus.

.Isolation from Water Supplies. — A great deal of work has been done
on this subject. It is evident that if it is difficult to isolate the bacilli
from the stools, it must a fortiori be much more difficult to do so when
the latter are enormously diluted by water. The b. typhosus has, how-
ever, been isolated from water during epidemics. The b. coli is, as
might be expected, the organism most commonly present in such
circumstances. In the case of both bacteria, the whole series of culture
reactions must be gone through before any particular organism isolated
is identified as the one or the other ; probably there are saprophytes
existing in nature which only differ from them in one or two reactions.
In examining waters, the ordinary plate methods are generally used, but
the McConkey or similar media may be employed with advantage.
Klein filters a large quantity through a Berkefeld filter, and, brushing
off the bacteria retained on the porcelain, makes cultures. A much
greater concentration of the bacteria is thus obtained. From time to time
various substances have been used with the object of inhibiting the growth
of the b. coli without interfering with that of the b. typhosus. Most of
these have not stood the text of experience. Lately caffeine has been
used for this end. For use in examining waters the following is the
method employed : To 900 c.c. of the suspected water there are added
10 grammes nutrose dissolved in 80 c.c. of sterile water, and 5 grammes
of caffeine dissolved in sterile distilled water, heated to 80° C. and cooled
to 55° C. before addition. After mixing the ingredients there is added
10 c.c. of '1 per cent, crystal violet. The flask is incubated at 37° C. for

FOOD-POISONING BACILLI 379

twelve hours, and then plates of Conradi-Drigalski medium are inoculated
from it. For investigation of faeces, a medium made up as above but with
ordinary st<-rilr water may be inoculated and a similar procedure followed.
On the whole there is little to be gained from this attempt to isolate the
typhoid bacillus from water in any particular case, and it is much more
useful for the bacteriologist to bend his energies towards the obtaining
of the indirect evidence of contamination of water by sewage, to the
nature of which attention has been called in Chapter V.

THE PARATYPHOID AND FOOD-POISONING BACILLI.

In the b. coli we have an organism having a definite habitat
in the animal intestine, and presenting certain cultural characters
by which it may be recognised. We may look on the bacillus
typhosus as an organism of the same class whose cultural
reactions as compared with b. coli present somewhat negative
characters, but which acquires definiteness from its association
with a well-known clinical condition. We have now to deal
with a group of organisms which occupy rather an intermediate
position between the two organisms referred to, and whose
cultural characters are such as to make their differentiation from
either fairly practicable. The members of this group have been
originally described in association with a variety of clinical
conditions, but, notwithstanding, they resemble each other so
closely that great difficulty arises, and the recognition of different
types which in literature receive different names can only be
effected by the application of the finest bacteriological tests.
Although in cultures the different types present slight differences,
these are not sufficient for the assignment of a name to an
organism of the class isolated from some fresh source, and, as a
matter of fact, in modern work relating to them, it is generally
impossible in identifying an organism to rely on merely noting
a correspondence with a described type. The method usually
adopted is to obtain from other workers cultures of what may
be called the historic strains isolated, and by comparing the
organism under investigation with these, to attempt to place it
in its proper position.

Organisms of the group are of great importance, not only
from their producing ordinary infective disease in man, but
because they are the agents at work in the great majority of
the not infrequently occurring cases of illness usually described
as " food poisoning." l Such poisoning is often referred to as
" ptomaine poisoning," from the idea originally prevailing that

1 A special type of food poisoning is associated with the Bfn-illi's l>«tulinm,
fj.v.

380 TYPHOID FEVER

the symptoms were caused by alkaloidal substances produced
during putrefactive processes occurring in meat. Certain cases
of illness arising within an hour or two of the taking of tainted
meat may be due to poisons of such a kind, but in the great
majority of single or multiple cases of illness traceable to food,
the symptoms do not appear so rapidly, and are associated with
the multiplication in the intestine of organisms of the type now
under consideration, and it may be also with an infection of the
blood. In such cases, the meat at fault may not, to taste or
smell, present any unusual features, but very often there can be
isolated from it an organism identical with organisms derived
from the sick individuals. Sometimes it has been proved that
the animals from which the meat was derived have been suffer-
ing from illnesses probably due to the organisms subsequently
found, but this has not always been the case, healthy meat being
here contaminated by contact with infective matter. The foods
giving rise to poisoning usually belong to the preserved food
class, or consist of sausages or similar products. There is every
reason to believe that the organisms in question may not be
killed in the ordinary processes of cooking, in which the internal
parts of the meat may not reach the temperature of blood
coagulation.

The organisms included in the paratyphoid and food-poisoning
group are as follows : The paratyphoid bacillus, varieties A and
B, originally isolated from pathological conditions in man ;
bacillus enteriditis Gaertner, isolated from meat-poisoning cases ;
bacillus ^Ertryck, also isolated from meat poisoning; bacillus
suipestifer (Salmon's bacillus of hog cholera) ; psittacosis bacillus,
occurring in a disease of parrots ; bacillus typhi murium, isolated
by Loffler from an epidemic of enteritis in mice ; and Danysz's
bacillus, isolated from an epidemic in field mice, and used by
him for originating epidemics in rats. The pathological effects
produced by these organisms include, on the one hand, general
septicsemic manifestations, and, on the other, gastro-enteritis.
The chief members of the group will be described below.

The Characters of the Paratyphoid and Food-Poisoning
Bacilli. — These bacilli are all miscroscopically indistinguish-
able from the bacillus typhosus. They are Gram-negative,
motile bacilli, the flagella being sometimes few in number, and
they do not form spores. On ordinary media, growths have the
general character of those of the b. coli and b. typhosus, some
members in certain reactions resembling the one, and in others
resembling the other. Opinion differs as to their capacity to
form indol, but usually the reaction to this test is negative.

FOOD-POISONING BACILLI 381

The methods for the isolation of the members of the group
vary with the nature of the infected material to' be examined.
In the case of abscesses caused by the paratyphoid bacillus,
the organism is usually accidentally discovered during the
application of ordinary methods. When deliberate search for a
member of the group is required, usually either the faeces or the
blood constitutes the material to be examined. In the former
case, advantage is taken of the fact that the food-poisoning
bacilli do not ferment lactose. Thus, if McConkey bile-salt
lactose-agar plates (p. 50) be used, the organisms sought for
will appear as colourless colonies which can be picked off for
systematic investigation. In the case of blood, ordinary methods
will prove sufficient.

Capacity for fermenting sugars has been largely applied in
work on this group. All the members produce practically the
same reactions. They originate acid and gas in glucose, lajvu-
lose, sorbite, mannite, dextrin, maltose, dulcite, galactose and
arabinose, like b. coli, but produce no change in lactose, cane-
sugar, salicin or inulin. Although differences in fermenting
capacity have been noted in different strains, the existence of
such cannot be relied upon for differentiating members of the
group from one another. The sugar reactions are only of use
in demarcating the lines between the food-poisoning group and
b. coli on the one hand, and b. typhosus on the other. The
differentiation of members of the group can only be effected by
applying the agglutination tests to the serum of animals suffer-
ing from natural or artificial infection. The chief point here is
that in such infections, the occurrence of group agglutinins in
the serum is much in evidence. Herein lies the necessity for
having at hand the historic strains of the organisms referred to
above. In dealing with an organism, it is first of all advisable
to take the serum of the ' infected individual, estimate the
highest dilution with which it clumps the strain isolated, and
compare the result obtained with the effect of the serum on the
historic strains. The unknown strain is most likely to be allied
to that strain which is agglutinated by a similar dilution of the
serum used. Frequently, in the investigation of an organism,
it is necessary to inject it into an animal and study the
agglutinating properties of its serum on the infecting strain
and upon allied organisms. Here considerable information may
be obtained by the use of the absorption method. If from such
a serum, for instance, an unknown organism has absorptive
qualities similar to that of a historic Gaertner, its being named
a Gaertner bacillus would be justified. It is customary in any

382 TYPHOID FEVER

case to note the action of a typhoid serum on an organism under
investigation, and also the action on the typhoid bacillus of an
antiserum to the unknown organism.

The Paratyphoid Bacillus. — This organism, which was when
first described often called the paracolou bacillus, was
primarily isolated from abscesses occurring in apparently non-
typhoid cases. Widal noted its resemblances to b. typhosus and
b. coli, from the latter of which it differed in not producing
indol and in not fermenting lactose. Gywnn first isolated it
from the blood of a case presenting typhoid symptoms, and since
then it has been recognised as being the probable cause of the
disease effects in about 3 per cent, of cases which clinically are
to be described as typhoid fever. The case mortality in para-
typhoid fever is low, being only from 1 to 2 per cent. The
organism has been isolated from the blood, the roseolar spots,
and from the stools. Several strains showing slight differences
in culture reactions have been obtained. Of these the two
chief are " paratyphoid A " and " paratyphoid B," the latter
being of commonest occurrence ; these appear to present slight
cultural differences. On gelatin, agar, and potato, A resembles
b. typhosus, B resembles b. coli ; in litmus milk A produces
slight permanent acidity, while after the third day, in the case
of B, acidity gives place to alkalinity ; on sugars the ferment-
ative activity of B is greater than that of A. Generally speak-
ing, the characters of both are those of the group to which they
belong. With regard to agglutinating reactions, the serum of
a paratyphoid patient will agglutinate the bacillus in high
dilutions. Observations on the behaviour of such sera towards
the b. typhosus have in different cases yielded some discordant
results, but usually a very much stronger concentration is
necessary to give clumping, and often a paratyphoid serum will
not clump the typhoid bacillus except in such concentrations as
might give similar effects when normal sera are under observation.
When any serum clumps both the paratyphoid and the typhoid
bacilli, the more closely the fmaximal clumping dilutions corre-
spond, the more likely is the case to be typhoid fever ; on the
other hand, if a high dilution will clump the paratyphoid
bacillus, while a low dilution is necessary for the typhoid
bacillus, then the case is likely to be paratyphoid fever. With
regard to the effects of other sera on the paratyphoid bacillus,
it may be said that usually a typhoid serum will require to be
used in greater concentration to clump this bacillus than is
necessary to obtain an effect with the typhoid bacillus itself.
Similar effects are observed when the sera of animals immunised

BACILLUS EXTERITIDIS 383

against Gaertner's bacillus or the bacillus of psittacosis are used.
In all serum tests the essential point is that deductions should
only be based on comparative observations of the highest
dilutions in which a clumping effect is produced with any series
of organisms compared.

While the paratyphoid bacillus originates a disease resembling
typhoid fever, it has also been found in the stools of typhoid
patients, and mixed infections may thus occur. Both organisms
have been observed together in the stools in typhoid carriers, and
pure paratyphoid carriers are also stated to occur. A meat
poisoning epidemic attributed to the paratyphoid bacillus has
been reported. Besides the septic cases already alluded to, the
organism has been isolated from cases of bone abscess, from
orchitis, and in Widal's case from a thyroid abscess, and in such
cases the history of a previous typhoid-like illness may not be
elicited. It has also been found in ordinary faeces. In animal
experiments it produces in rabbits and guinea-pigs a fatal illness
of a septictemic type with serous inflammations.

Bacillus Enteritidis (Gaertner). — In 1888, Gaertner, in
investigating a number of cases of gastro-enteritis resulting from
eating the flesh of a diseased cow, isolated, from the meat and
from the spleen of a man who died, a bacillus closely resem-
bling the typhoid bacillus. Since then, in a great number of
similar outbreaks, similar bacilli have been found both in the
stools and in the organs. The cultural characters are those of
the group, except that in some strains the presence of an effect on
lactose has been observed. Here again much information may
be obtained from the agglutinating properties of the serum.
It has also been found that the serum of persons suffering from
meat poisoning sometimes clumps the typhoid bacillus, though
a higher concentration is required than in the case of Gaertner's
bacillus. The Gaertner group of organisms is very pathogenic
for laboratory animals. Often, whatever the channel of infec-
tion, there is intense haemorrhagic enteritis, and very usually
there is a septicaemia with the occurrence of serous inflammations ;
the bacilli are recoverable from the solid organs and often from
the blood. In man, as the name of the bacillus indicates, the
symptoms are centred in the intestine, where there is usually
marked inflammation of the mucous membrane, sometimes
attended with haemorrhage into it ; evidence of a septicaemic
condition may also exist. Infection may take place by the
bacillus itself, and here the illness usually appears within
twenty-four hours of the food being partaken of, but symptoms
may appear almost at once, in which case they are no doubt due

384 TYPHOID FEVER

to the action of toxins ; here it is important to note that the
poisons formed by this group of organisms are relatively heat-
resisting, so that boiling for a time does not destroy the toxicity.

The Psittacosis Bacillus. —When parrots are imported from the
tropics in large numbers, many die of a septicaemio condition in which an
enteritis, it may be hsemorrhagic, is a marked feature. There is intense
congestion of all the organs and peritoneal ecchymoses. From the
spleen, bone marrow, and blood there has been isolated a bacillus having
the group characters, except that here also an effect on lactose has
been described. The parrot is most susceptible to its action, but it
also causes a fatal h?emorrhagic septicaemia in guinea-pigs, rabbits, mice,
pigeons, and fowls, the bacilli after death being chiefly in the solid
organs. From affected parrots the disease appears to be readily
communicable to man, chiefly, it is probable, from the feathers being
soiled by infective excrement. Several small epidemics have been
recognised and investigated in Paris. After about ten days' incubation,
headache, fever, and anorexia occur, followed by great restlessness,
delirium, vomiting, often diarrhoea, and albuminuria. Frequently
broncho-pneumonia supervenes, and a fatal result has followed in about
a third of the cases observed. The organism has been isolated from the
blood of the heart. The psittacosis bacillus is evidently one of the
typhoid group, a fact which is further borne out by the observation tlmt
it may be clumped by a typhoid serum. The clumping is, however, said
often to be incomplete, as the bacilli between the clumps may retain
their motility. It differs from the typhoid bacillus in its growth on
potatoes and in its pathogenicity.

Danysz's Bacillus and Rat Viruses. — Danysz isolated from an
epizootic in field mice an organism of this group, which he introduced
for the purpose of killing rats by originating in them through feeding a
similar epizootic, and several viruses of this kind are in commercial use
for this purpose. These have been investigated by Bainbridge, who,
however, finds that they owe any efficiency they possess to two organ-
isms, the bacillus Aertryck and the bacillus enteriditis of Caertner. The
efficacy of such agents varies, and the mortality in artificially originated
epizootics is from 20 to 50 per cent. Sometimes, apparently under
natural conditions, rats develop an immunity to those viruses, and it is
doubtful whether they are entirely innocuous to other animals which
may partake of the food containing them.

BACILLARY DYSENTERY.

Dysentery has for long been recognised as including a number
of different pathological conditions, and within more recent times
amoebic and non-amoebic forms have been distinguished. Of the
latter, bacteria have been believed to be the causal agents, and an
organism described by Shiga in 1898 has almost certainly been
established as the cause of a large proportion of cases. Shiga's
observations were made in Japan, and confirmatory results have
been obtained by Kruse in Germany, by Flexner and by Strong
and Harvie in the Philippine Islands, and more recently by Vedder

BACILLUS DYSENTERIC 385

and Duval in the United States. It is now further recognised
that the epidemics of dysentery which from time to time occur
in lunatic asylums are usually due to bacilli of this type, and in
America the organism has been demonstrated in summer
diarrhoea in children. The evidence for the relationship of the
organism to the disease consists chiefly in the apparently con-
stant presence of the organism in the dejecta in this form of
dysentery, and the agglutination of the organism by the serum
of patients suffering from the disease, but confirmatory evidence
has also come from animal experimentation. From different
epidemics a great many different strains of the dysentery bacillus
have been obtained, but these all possess common characters and
'are undoubtedly closely related to one another. The various
strains resolve themselves into two chief groups, whose differences
lie in their behaviour towards certain sugars, in their capacities
of forming indol, and in their agglutinating reactions. The
relation of amoebae to dysentery will be discussed in the
Appendix.

Bacillus Dysenteriae. — Morphological Characters. — This
bacillus morphologically closely resembles the typhoid bacillus,
but is on the whole somewhat plumper, and filamentous forms
are comparatively rare. Involution forms sometimes occur,
especially in glucose agar. Most observers have found no trace
of motility, whilst others say that it is slightly motile. Vedder
and Duval have, however, by a modification of Van Ermengen's
process, demonstrated in the case of one strain the presence of
numerous lateral flagella, which are of great fineness, but of
considerable length. No spore formation occurs ; the organism
is stained readily by the ordinary dyes, but is decolorised by
Gram's method.

Cultural Characters. — In these also considerable resemblance
is presented to the typhoid bacillus. In gelatin a whitish line of
growth occurs along the puncture, but the superficial film-like
growtli is usually absent, or at least poorly marked. In plate
cultures also the superficial growths are smaller and have less of
the film-like character, than those of the typhoid organism. On
ayar, growth occurs as a smooth film with regular margins, but
after two or three days, especially if the surface be moist, Vedder
and Duval describe an outgrowth of lateral offshoots on the
surface of the medium. On agar plates the colonies resemble
those of the typhoid organism, being of smaller size and less
'»)iaijue than those of the bacillus coli. In peptone bouillon a
uniform haziness is produced. As has l^een indicated, different
strains of the bacillus behave differently towards different suyarx,

386 TYPHOID FEVER

and the results of all observers do not agree, so that only general
statements can be made. Without going into the question of
the particular strains to be placed in the two groups, we may say
that, roughly, these may be classified into the Shiga-Kruse group
and the Flexner group. All produce acid in peptone-glucose and
in taurocholate peptone-glucose ; none produce change in lactose
or cane-sugar. The Shiga group do not produce acid in maltose or
mannite, while the Flexner group do, and, generally speaking, the
former do not produce indol, while the latter do. Forms inter-
mediate between the two groups occur. There is never any
evolution of gas observed in sugar media. In litmus milk there
is developed at first a slight degree of acidity, which is followed
by a phase of increased alkalinity ; no coagulation of the milk '
ever occurs. On potato the organism forms a transparent or
whitish layer, which, however, in the course of a few days assumes
a brownish-red or dirty grey colour, with some discoloration of
the potato at the margin of the growth.

Relation to the Disease. — The organism has been found in
large numbers in the dejecta, especially in the acute cases, where
it may be present in almost pure culture. In the thirty-six cases
examined, Shiga obtained it in thirty-four from the dejecta, and in
the two others post mortem from the intestinal mucous membrane.
The organism does not appear to spread deeply or to invade
the general circulation. In the more chronic cases it is difficult
to obtain, on account of the large number of the bacillus coli and
other bacteria present. Vedder and Duval found agar plates to
be the best method of culture, these being incubated at the
blood temperature. They also found that if the colonies which
appeared at twelve hours were marked with a pencil, there was
a greater probability of obtaining the bacillus of dysentery from
those which appeared later, most of those appearing early being
colonies of the bacillus coli. McConkey's agar medium with
lactose added may be used for isolation from stools. A little of
the fa3ces is rubbed up in broth and some of the mixture stroked
on the medium. The formation of acid by the b. coli colonies
enables them to be excluded, and therefore, as the b. dysenteriye
is not a lactose fermenter, the colourless colonies which develop
after twenty -four hours are picked out for further investigation.

As already stated, both acute arid chronic cases are marked
by the presence of this organism. In the former, where death
may occur in from one to six days, the chief changes, according
to Flexner, are a marked swelling and corrugation of the mucous
membrane, with haemorrhage and pseudo-membrane at places.
There is extensive coagulation-necrosis with fibrinous exuda-

BACILLUS DYSENTERIC 387

tion and abundance of polymorpho-nuclear leucocytes, and the
structure of the mucous membrane, as well as that of the
muscularis mucosae, is often lost in the exudation. There is
also great thickening of the sub-mucosa, with great infiltration of
leucocytes, these being chiefly of the character of " plasma cells."
In the more chronic forms the changes correspond, but are
more of a proliferative character. The mucous membrane is
granular, and superficial areas are devoid of epithelium, whilst
ulceration and pseudo-membrane are present in varying degree.
A feature of bacillary dysentery is the fact that abscess of the
liver does not occur as a complication.

Agglutination. — All the above-mentioned observers agree re-
garding the agglutination of this bacillus by the serum — that is,
in the cases of dysentery from which the organism can be cul-
tivated. The reaction may appear on the second day, and is
most marked after from six to seven days in the acute cases ; it
is usually given in a dilution of from one in twenty to one in
fifty within an hour, though sometimes much higher dilutions
give a positive result. In the more chronic cases the reaction
is less marked, and here the sedimentation method is to be
preferred. It is difficult to make any general statements with
regard to the effects of dysenteric sera on the different strains
of the bacilli, but it may be said that generally a serum
agglutinates the strain which produced it and the other strains
of the same group in higher dilutions than it does the strains of
the other group. Many observers have found that the serum
from a case associated with strains of the Shiga-Kruse group
has not agglutinated strains of the Flexner group, and
corresponding observations have been made in cases associated
with the Flexner group. Often the sera of animals immunised
with bacilli have been used for such tests, but apparently great
care must be exercised in basing diagnoses on such observations,
as the sera vary in different instances as regards their action on
strains allied to that used for injection. Agglutination of the
organism has not been obtained with serum from cases other
than those of dysentery, nor has a similar bacillus been cultivated
from such sources. The reaction is also absent in those cases
of dysentery which are manifestly of amoebic nature.

Pathogenic Properties. — The organism is pathogenic in guinea-
pigs and other laboratory animals, but, in these, characteristic
changes in the intestine are often awanting. Shiga, however,
obtained such effects by introducing the organism into the
stomach of young cats and dogs, and confirmatory results were
obtained by Flexner. Such attempts have been specially

388 TYPHOID FEVER

successful when the virulence of the organism has been
previously exalted by intraperitoneal passage. In two cases,
apparently well authenticated, a dysenteric condition has
followed in the human subject from ingestion of pure cultures
of the organism.

It is probable that in the action of the bacillus a toxin is
concerned. If the organism be grown for two or three weeks in
an alkaline bouillon, there appears, probably by autolysis of the
bacteria, a toxin in the culture medium separable by nitration in
the ordinary way. The optimum alkalinity is achieved by
adding '3 per cent, of soda to bouillon neutral to litmus, the
resulting precipitate not being removed ; free access of oxygen
is permitted during growth. Apparently, the Shiga-Kruse
strains yield the most toxic nitrates, and with the Flexner
strain, the results of most observers show that soluble toxins
cannot be obtained. The poison is very toxic to animals,
especially rabbits, and however introduced into the body it
causes after an incubation period haemorrhagic enteritis with a
diphtheritic-like exudate on the surface of the mucous membrane.
Toxins isolated from different strains differ as regards the
animals for which they are most toxic. The toxin is fairly
resistant to heat, standing temperatures up to 70° C. without
being injured.

It may be said that an aggressive reaction (vide p. 189) has
also been described in the case of the dysentery bacillus.

Immunisation Experiments. — Both large and small animals
have been immunised against the bacillus and also against its
toxic nitrates. In the former case the immunisation has been
commenced either with non-lethal doses of living cultures, or with
cultures killed by heat. The nature of the immunisation is
probably complex. When cultures have been used, a bactericidal
serum, in which immune bodies and complements (vide Immunity)
are concerned, is developed. When the toxin is used for
immunisation, a serum protecting against the toxin is produced.
According to some results, animals immunised with cultures are
immune against the toxin, and vice versa. All races of animals
do not lend themselves to immunisation.

Considerable work has been done in immunising large
animals (horses, goats) against the soluble toxins of the
dysentery bacillus with a view to obtaining therapeutic sera.
Doerr, using his toxin from the Shiga-Kruse strain, produced
in horses an antitoxic serum having protective and curative
properties in animals. This serum has been used in a number
of cases of bacillary dysentery in man with good results. Shiga

BACILLUS ENTERITIDIS SPOROGENES 389

produced a polyvalent serum by injecting horses with agar
cultures of different strains, and states that it has been used
in Japan with good results. Further observation is necessary
as to the therapeutic effects in cases associated with the Flexner
strain of an antitoxin produced by the Shiga strain.

It will be seen that the evidence furnished is practically
conclusive as to the causal relationship between this bacillus and
one form of dysentery, a form, moreover, which is both wide-
spread and embraces a large proportion of cases of the disease ;
and especially of importance is the fact that observations made
independently in different countries have yielded practically
identical results on this point.

Bacillus Dysenteriae (Ogata). — Ogata obtained this bacillus in an
extensive epidemic in Japan in which no amoebae were present. He
found in sections of the affected tissues enormous numbers of small
bacilli of about the same thickness as the tubercle bacillus, but very much
shorter. These bacilli were sometimes found in a practically pure
condition. They were actively motile, and could be stained by Gram's
method. He also obtained pure cultures from various cases and tested
their pathogenic effects. They grew well on gelatin, at the ordinary
temperature producing liquefaction, the growth somewhat resembling
that of the cholera spirillum. By injection into cats and guinea-pigs, as
well as by feeding them, this organism was found to have distinct
pathogenic effects ; these were chiefly confined to the large intestine,
haemorrhagic inflammation and ulceration being produced. It still
remains to be determined whether this organism has a causal relationship
to one variety of dysentery.

BACILLUS ENTEEITIDIS SPOROGENES.

This organism was first isolated by Klein from the evacuations in an
outbreak of diarrhoea following the ingestion of milk which contained the
microbe, and it was subsequently found by him in certain cases of
infantile diarrhoea and of summer diarrhoea, in certain instances in milk,
and as a constant inhabitant of sewage (see Chap. V.). In films made
from the stools in diarrhoea cases where it is present, it can be micro-
scopically recognised as a bacillus 1'6 /* to 4*8 /A in length and "8 /* in
breadth, staining by ordinary stains and retaining the dye in Gram's
method. It often contains a spore near one of the ends, or sometimes
nearer the centre. It is slightly motile, and in cultures can be shown
to possess a small number of terminal flagella. It grows well under
anaerobic conditions in ordinary media, especially on those containing
reducing agents. On agar the colonies are circular, grey, and translucent,
and under a low power are seen to have a granular appearance. On this
in- « Hum spore formation does not occur, but is easily obtained if the
organism is grown on solidified blood serum, which, further, is liquefied
(luring growth. On gelatin plates liquefaction commences after twenty-
four hours at 20° C. It produces acid and gas in bile-salt glucose media,
and in peptone-salt solution containing glucose or mannite. Spore
formation can be seen to take place in 2 per cent, dextrose gelatin, but

390 TYPHOID FEVEE

the degree seems to be in inverse ratio to the amount of gas formation.
Very typical is the growth in milk, and it is by this medium that
isolation can be best effected. A small quantity of the material
suspected to contain the bacillus is placed in 15 to 20 c.c. of sterile
milk, which is then heated for ten minutes at 80° C. to destroy all
vegetative bacteria ; the tube is cooled, placed under anaerobic con-
ditions, and incubated at 37° C. for from twenty-four to thirty-six hours.
If the bacillus be present there is abundant gas formation, and almost
complete separation of the curd from the whey takes place. The former
adheres to the sides of the tube in shreds, and large masses gather with
the cream on the top of the fluid, all being torn by the gas evolved. The
whey is only slightly turbid, and contains numerous bacilli. The growth
lias an odour of butyric acid. If a small quantity (say 1 c.c.) of
the whey be injected into a guinea-pig, the animal becomes ill in a few
hours and dies in twenty-four hours. At the point of inoculation, the
skin and subcutaneous tissues, and sometimes even the subjacent muscles,
are green and gangrenous and evil-smelling, there is considerable oedema,
and there may also be gas formation. The exudation is crowded with
bacilli, which, however, are not generally distributed in any numbers
throughout the body. These pathogenic properties of the bacillus
enteritidis sporogenes are important in its recognition, for its culture
reactions taken alone are very similar to those of the bacillus butyricus of
Botkin.

SUMMER DIARRHOEA.

As has been already stated, the bacillus of dysentery, the
b. coli, and the b. enteritidis sporogenes have been found
associated with epidemics of this disease. This indicates that
the condition may be originated by a variety of organisms, and
it is further probable that the clinical features in different
epidemics vary. This is to a certain extent illustrated by the
condition of the stools. In Britain these are usually green,
watery, slimy, and putrid, without blood or mucus, but in many
outbreaks in America blood and mucus are present. The
multiple origin of the disease has been illustrated by the work of
Morgan, who, in a careful investigation of the disease in Britain,
has been unable to find evidence of the dysentery bacillus being
present. He has, however, very frequently (in 63 per cent.
of the cases examined) found in the stools and intestine a
bacillus (" Morgan's No. 1 bacillus ") which is a motile Gram-
negative organism producing acid and slight gas formation in
glucose, laevulose, and galactose, and no change in mannite,
dulcite, maltose, dextrin, cane-sugar, lactose, inulin, amygdalin,
salicin, arabinose, raffinose, sorbite, or erythrite ; it further
causes indol formation, and in litmus milk slowly originates an
alkaline reaction. It produces diarrhoea • and death in young
rabbits, rats, and monkeys when these animals are fed on
cultures. It is thus possible that in this bacillus we have

REVIEW OF THE COLT-TYPHOID BACILLI 391

still another cause of the disease. Morgan has found that in
diarrhoea cases the lactose fermenters, so characteristic of normal
faeces, are relatively less frequent and tend to be replaced by
non-fermenters of lactose. His bacillus has been found in a
certain proportion of normal children, but this especially during
the epidemic season ; it has also been found in flies.

GENERAL REVIEW OF THE COLI-TYPHOID BACILLI.

A general view of the organisms belonging to the coli-typhoid
group which we have now considered indicates a close alliance
between the various members. All are microscopically indis-
tinguishable from one another, and react negatively to the
Gram stain. The chief sub-groups can be differentiated by
culture reactions, of which the action on sugars is most important.
Here important information is obtained by the study of the
glucose and lactose reactions. The typhoid sub-group produces
acid on glucose, but has no action on lactose. The dysentery
sub-group is similar, but is chiefly marked off from the typhoid
sub-group by its relative non-motility, by its tendency to form
alkali after a preliminary acid development on litmus milk, and
by the fact that it does not ferment sorbite. The food-poisoning
sub-group is differentiated from the typhoid sub-group by
forming acid and gas in glucose and from the coli-sub-group by
its producing no change on lactose. The positive features of the
coli sub-group are the formation of acid and gas in both glucose
and lactose.

From work done not only with bacteria isolated from patho-
logical conditions, but in connection with the bacteriology of
water, milk, and faeces, it has been found that an enormous
number of organisms exist, having the capacity of fermenting
glucose and lactose, but which, when further investigated, present
individual differences. Much has been done in attempting to
differentiate these so-called " lactose fermenters " from one
another. Here the work of McConkey may be taken as con-
stituting one of the best attempts at such further classification,
and it has the merit of simplifying a technique unduly compli-
cated l»y the use of fermentation tests in a great series of sugars,
on which the various sub-groups have all the same effect.
McConkey is of opinion that certain of the tests applied to the
lactose fermenters in reality give little information. These are,
first, the growth on litmus whey, observation of which only
corroborates what is observed with litmus milk ; second, observa-
tion of fluorescence on neutral-red lactose media (on account of

392 TYPHOID FEVER

the inconstancy of the occurrence of this change in lactose
fermenters, and from the fact that many other bacteria also
produce it) ; third, the reduction of nitrates, — this appears
to be a common property of nearly all the members of the
group ; fourth, observation of differences in the naked eye or
low power appearances on gelatin ; these are very inconstant, and
different colonies of the same organism may show different
appearances. On the other hand, important information may
be obtained by the observation of the Voges and Proskauer
reaction (p. 353). With regard to sugars, McConkey concludes
that in the differentiation of the lactose fermenters, the only
sugars necessary are lactose, saccharose, dulcite, adonite, inulin,
inosite, and mannite. Using these, a preliminary classification
can be made from the actions on cane-sugar and dulcite, and
four groups are constituted : I. Organisms not affecting either
cane-sugar or dulcite. II. Organisms having no action on cane-
sugar, but fermenting dulcite. III. Organisms fermenting
both cane-sugar and dulcite. IV. Organisms fermenting cane-
sugar but having no action on dulcite. Of the first, the
bacillus acidi lactici of Hiippe may be taken as a type ;
of the second, the bacillus coli communis of Escherich ;
of the third, bacillus Friedlander ; of the fourth, the bacillus
lactis aerogenes and the bacillus cloacae. Group IV. is further
sub-divided into sub-group 1, in which there is no lique-
faction of gelatin and an absence of the Voges and Proskauer
reaction ; 2, with no liquefaction of gelatin, presence of Voges
and Proskauer's reaction (bacillus lactis aerogenes) ; 3, with
liquefaction of gelatin, presence of Voges and Proskauer's
reaction (bacillus cloacae) ; 4, with liquefaction of gelatin and
production of a yellow pigment. Taking the properties named
as type characteristics, the great mass of lactose fermenters can
be further differentiated by the application of the other sugar
tests. It is well to refer any organism found as belonging to
one or other of the types, as in most cases no name has been
assigned. Examples are constantly met with in work on water
or faecal contents.

Although many of the named varieties were originally
described in connection with other bacterial processes, all these
bacteria are of frequent occurrence, especially in the human and
animal intestine. As in the case of the members of the food-
poisoning group, great difficulty has been experienced in
identifying the types from mere description, and considerable
complication has arisen from the fact that before the elaboration
of the modern differentiation technique, different observers

REVIEW OF THE COLI-TYPHOID BACILLI 393

identified organisms as belonging to a classical type, which have
now been found not to conform in properties with the historic
.strains ; here again, it is now customary during classification
work to have at hand such historic strains in order that com-
parative parallel observations may be made.

With regard to the type strains, a few words may be added.
The original bacillus coli communis of Escherich was isolated
from the intestine of newly-born infants in connection with the
first appearance of bacteria in the alimentary tract. About the
same time, an organism now known as the bacillus neapolitanus
was obtained by Emmerich in an outbreak of choleraic disease
in Naples, and this organism was looked upon as identical with
Escherich's bacillus, but it ferments saccharose, on which
Escherich's has no effect. The bacillus acidi lactici of Hiippe
was stated by this observer to be the chief cause of the souring
of milk. It is now known that a large number of organisms of
the same type, but differing slightly in cultural characters, are
concerned in this process, and, as a matter of fact, McConkey
found the presence of the classical strain to be relatively infre-
quent in milk. The bacillus lactis aerogenes was originally
described by Escherich, in connection with his work on the
bacteriology of the intestine in children, as an organism differing
from the ordinary milk-souring bacteria by its producing gas
from milk in the absence of air. Although it is a free gas-
producer, this property is not specific for it, and within recent
years it has attracted attention chiefly from its apparently being
closely allied to the bacillus pneumonias of Friedlander. Like
the latter, this organism is stated when injected into animals to
appear in a capsulated form. Another member of this group is
bacillus oxytocus perniciosus, which is said originally to have
been isolated from milk. This organism, along with the bacillus
vesiculosus and an organism denominated No. 71, were found
by McConkey to be of very common occurrence in human and
animal faeces.

In work of the kind with which we are dealing, two other
organisms are not infrequently observed which morphologically
belong to the coli-typhoid group, but neither of which is a
lactose fermenter. These are the bacillus faecalis alcaligenes,
and the bacillus coli anaerogenes. The reactions of these will be
found in the Table (p. 394). The latter bacillus somewhat
resembles the typhoid bacillus, but produces acid in lactose
and can be distinguished by agglutinating reactions.

When any question arises regarding the relationships of an
organism isolated under saphrophytic conditions and resembling

394

TYPHOID FEVER

s^an^itsoaij
put; saSoA

iiiiit +,+ +111

-.opuj

$$m$ m i < " ^ <

d & x x & o d d' ^ . d d d
4; ^! 3 9 9 3 ° 4 4 ^ ' 3 ''* 4 ^ .« ° '

d ..... o d .* . d d d d

-< •<" •< "" < <j <j <j

.„„„„,

- - i d i . i

'"—

f. • ^ d . .

•a,WoS

d.dd|||ddd d d

•„,

i , d d ( d ( , d

•et!o,na

d d d odd
i .iiiii, i

-< <3 •< •< •<* •<

•aimutjj^

d . d d ! . ! d d d ( . d d

•asojBqooBg

, .", d ( d , d , d d

•9SO^OUrI

d ddd . d d d d-<

• • •<...• >H

•asoon^o

d -idd . -dodo -dddd

•UW«P0

i i i , , , , i i + , i + , i i

•A,moM

+ + + +i 1 + 1 i++i i , + i

Bacterium.

IT 1? ^ § r

G^rjS ' 9* * • ' S '•

— P | 8l | r* . , g- g g .S 0

'S 7. i 8- ® ^ ^ s •? * 3 & I * V, tj

5 o^ffiS8-Mo" Soas^^

1 §/ 1, s 1 's 1 si 1 •^i = 3§!

S§S.-S^-e|?d^8.2c§-sg^
w^1so§S°--s^'5ci-20w^

1 S 1 i 1 1 * 1 1 1 • 1 1 1 ? 1

^M/WMCqp,^^^^^^^^^

i -\ A

REVIEW OF THE COLT-TYPHOID BACILLI 395

some definite pathogenic type, important information can often
be obtained by studying its agglutinating reactions. In such a
case the effect of sera produced by the pathogenic type upon
the unknown organism, and of sera produced by injection
into animals of the pathogenic type in question, ought to be
studied.

CHAPTER XVI.

DIPHTHEKIA.

THERE is no better example of the valuable contributions of
bacteriology to scientific medicine than that afforded in the case
of diphtheria. Not only has .research supplied, as in the case
of tubercle, a means of distinguishing true diphtheria from
conditions which resemble it, but the study of the toxins of the
bacillus has explained the manner by which the pathological
changes and characteristic symptoms of the disease are brought
about, and has led to the discovery of the most efficient means
of treatment, namely, the anti-diphtheritic serum. .

Historical. — The first account of the bacillus now known to be the
cause of diphtheria was given by Klebs in 1883, who described its
characters in the false membrane, *but made no cultivations. It was
first cultivated by Loffler from a number of cases of diphtheria, his
observations being published in 1884, and to him we owe the first
account of its characters in cultures and of some of its pathogenic effects
on animals. The organism is for these reasons known as the Klebs-
Loffler bacillus, or simply as Loffler's bacillus. By experimental in-
oculation with the cultures obtained, Loffler was able to produce false
membrane on damaged mucous surfaces, but he hesitated to conclude
definitely that this organism was the cause of the disease, for he did
not find it in all the cases of diphtheria examined, he was not able to
produce paralytic phenomena in animals by its injection, and, further,
he obtained the same organism from the throat of a healthy child. This
organism became the subject of much inquiry, but its relationship to
the disease may be said to have been definitely established by the
brilliant researches of Roux and Yersin, which showed that the most
important features of the disease could be produced by means of the
separated toxins of the organism. Their experiments were published in
1888-90. Further light has been thrown on the subject by the work of
Sidney Martin, who has found that there can be separated from the
organs in cases of diphtheria substances which act as nerve poisons, and
also produce other phenomena met with in diphtheria.

General Facts. — Without giving a description of the patho-
logical changes in diphtheria, it will be well to mention the out-
standing features which ought to be considered in connection

396

BACILLUS DIPHTHERIA 397

with its bacteriology. In addition to the formation of false
membrane, which may prove fatal by mechanical effects, the
chief clinical phenomena are the symptoms of general poisoning,
great muscular weakness, tendency to syncope, and albuminuria ;
also the striking paralyses which occur later in the disease,
and which may affect the muscles of the pharynx, larynx, and
eye, or less frequently the lower limbs (being sometimes of
paraplegic type), all these being grouped together under the
term " post-diphtheritic paralyses." It may be stated here that
all these conditions have been experimentally reproduced by the
action of the bacillus of diphtheria, or by its toxins. Other
bacteria are, however, concerned in producing various secondary
inflammatory complications in the region of the throat, such as
ulceration, gangrenous change, and suppuration, which may be
accompanied by symptoms of general septic poisoning. The
detection of the bacillus of Loffler in the false membrane or
secretions of the mouth is to be regarded as supplying the only
certain means of diagnosis of diphtheria.

Bacillus Diphtherise. — Microscopical Characters. — If a film
preparation be made from a piece of diphtheria membrane (in
the manner described below) and stained with methylene-blue,
the bacilli are found to have the following characters : — They are
slender rods, straight or slightly curved, and usually about 3 /n
in length, their thickness being a little greater than that of
the tubercle bacillus. The size, however, varies somewhat in
different cases, and for this reason varieties have been dis-
tinguished as small and large, and even of intermediate size.
It is sufficient to mention here that in some cases most are
about 3 /A in length, whilst in others they may measure fully
5 ft. Corresponding differences in size are found in cultures.
They stain deeply with the blue, sometimes being uniformly
coloured, but often showing, in their substance, little granules
more darkly stained, so that a dotted or beaded appearance is
presented. Sometimes the ends are swollen and more darkly
stained than the rest; often, however, they are rather tapered
off (Fig. 111). In some cases the terminal swelling is very
marked, so as to amount to clubbing, and with some specimens
of methylene-blue these swellings and granules stain of a violet
tint. Distinct clubbing, however, is less frequent than in
cultures. There is a want of uniformity in the appearance of
the bacilli when compared side by side. They usually lie
irregularly scattered or in clusters, the individual bacilli being
disposed in all directions. Some may be contained within
leucocytes. They do not form chains, but occasionally forms

398 DIPHTHERIA

longer than those mentioned may be found, and these specially
occur in the spaces between the fibrin as seen in sections.

Distribution of the Bacillus. — The diphtheria bacillus may
be found in the membrane wherever it is formed, and may also
occur in the secretions of the pharynx and larynx in the disease.
It may be mentioned that distinctions formerly drawn between
true diphtheria and non-diphtheritic conditions from the appear-

»•.

FIG. 111.— Film preparation from diphtheria membrane, showing
numerous diphtheria bacilli. One or two degenerated forms are seen
near the centre of the field. (Cultures made from the same piece of
membrane showed the organism to be present in practically pure
condition.)

Stained with methylene-blue. x 1000.

ance and site of the membrane, have no scientific value, the only
true criterion being the presence of the diphtheria bacillus. The
occurrence of a membranous formation produced by streptococci
has already been mentioned (p. 212).

In diphtheria the membrane has a somewhat different
structure, according as it is formed on the surface covered with
stratified squamous epithelium, as in the pharynx, or on a surface
covered by ciliated epithelium, as in the trachea. In the former

DISTRIBUTION OF THE BACILLUS - 399

-it nation necrosis of the epithelium occurs either uniformly or
in patches, and along with this there is marked inflammatory
reaction in the connective tissue beneath, attended by abundant
iibrinous exudation. The necrosed epithelium becomes raised
up by the fibrin, and its interstices are also filled by it. The
fibriuous exudation also occurs around the vessels in the tissue
beneath, and in this way the membrane is firmly adherent. In

Kic. ll^.--Sfction through a diphtheritic membrane in trachea,
showing diphtheria bacilli (stained darkly) in clumps, and also
>c 'uttered amongst the fibrin. Some streptococci are also shown.
towards the surface on the left side.

Stained by Gram's method and Bismarck-brown. x 1000.

the trachea, on the other hand, the epithelial cells rapidly
become shed, and the membrane is found to consist almost
exclusively of fibrin with leucocytes, the former arranged in a
reticulated or somewhat laminated manner, and varying in
density in different parts. The membrane lies uj)on the base-
ment membrane, and is less firmly adherent than in the case of
the pharynx.

The position of the diphtheria bacilli varies somewhat in
different cases, but they are most frequently found lying in oval

400 DIPHTHERIA

or irregular clumps in the spaces between the fibrin, towards the
superficial, that is, usually, the oldest part of the false membrane
(Fig. 112). There they may be in a practically pure condition,
though streptococci and occasionally some other organisms may
be present along with them. They may occur also in deeper
parts, but are rarely found in the fibrin around the blood vessels.
On the surface of the membrane they may be also seen lying in
large numbers, but are there accompanied by numerous other
organisms. Occasionally a few bacilli have been detected in the
lymphatic . glands. As Loftier first described, they may be
found after death in pneumonic patches in the lung, these
being due to a secondary extension by the air passages. They
have also been occasionally found in the spleen, liver, and
other organs after death. This occurrence is probably to
be explained by an entrance into the blood stream shortly
before death, similar to what occurs in the case of other
organisms, e.g. the bacillus coli communis. The diphtheria
bacillus may also infect other mucous membranes. It is found
in true diphtheria of the conjunctiva, and may also occur
in similar affections of the vulva and vagina ; some of these
cases have been treated successfully with diphtheria antitoxin.
The pseudo-diphtheria bacillus, however, may also occur in these
situations.

Association with other Organisms. — The diphtheria organism is
sometimes present alone in the membrane, but more frequently
is associated with some of the pyogenic organisms, the strepto-
coccus pyogenes being the commonest. The staphylococci, and
occasionally the pneumococcus or the bacillus coli, may be
present in some cases. Streptococci are often found lying side
by side with the diphtheria bacilli in the membrane, and also
penetrating more deeply into the tissues. In some cases of
tracheal diphtheria we have found streptococci alone at a lower
level in the trachea than the diphtheria bacilli, where the
membrane was thinner and softer, the appearance in these cases
being as if the streptococci acted as exciters of inflammation and
prepared the way for the bacilli. It is still a matter of dispute
as to whether the association of the diphtheria bacillus with the
pyogenic organisms is a favourable sign or the contrary, though
on experimental grounds the latter is the more probable. We
know, however, that some of the complications of diphtheria
may be due to the action of pyogenic organisms. The extensive
swelling of the tissues of the neck, sometimes attended by
suppuration in the glands, and also various hsemorrhagic con-
ditions, have been found to be associated with their presence ;

CULTIVATION OF THE BACILLUS

401

in fact, in some cases the diphtheritic lesion enables them to get
a foothold in the tissues, where
they exert their usual action and
may lead to extensive suppurative
change, to septic poisoning or to
septicaemia. In cases where a
gangrenous process is superadded,
a great variety of organisms may
be present, some of them being
anaerobic. Against such complica-
tions produced by other organisms
anti-diphtheritic serum produces no
favourable effect.

Cultivation. — The diphtheria
bacillus grows best in cultures at
the temperature of the body ;
growth still takes place at 22° C.,
but ceases about 20° C. The best
media are the following : Loffler's
original medium (p. 41), solidified
blood serum, alkaline blood serum
(Lorrain Smith), blood agar, and the ordinary agar media. If
inoculations be made on the surface of blood serum with a piece

of diphtheria membrane,

a b

FIG. 113. — Cultures of the
diphtheria bacillus on au
agar plate ; twenty-six hours'
growth. (Natural size.)
(a) Two successive strokes ; (ft)
isolated colonies from the same
plate.

colonies of the bacillus
may appear in twelve

hours, and are well formed
within twenty-four hours,
often before any other
growths are visible. The

I colonies are small circular

discs of opaque whitish
R ^^^ colour, their centre being
thicker and of darker
greyish appearance when
viewed by transmitted
light than the periphery.
Their margins are at first
regular, but later they
become wavy or even
crenated. On the second
or third day they may

reach 3 mm. in size, but when numerous they remain smaller.

On the agar media the colonies have much the same appearance
26

FIG. 114. — Diphtheria colonies, two days
old, on agar.

x8.

402

DIPHTHERIA

(Fig. 113) but grow less quickly, and sometimes they may be

comparatively minute, so
as rather to resemble
those of the streptococcus
pyogenes. In stroke cul-
tures the growth forms
a continuous layer of the
same dull whitish colour,
the margins of which
often show single colon-
ies partly or completely
separated. On gelatin
at 22° C. a puncture
culture shows a line of
dots along the needle
track, whilst at the sur-
face a small disc forms,

FIG. 115.— Diphtheria bacilli from a twenty- ra*her thicker in the
four hours' culture on agar. middle. In none of the

Stained with methylene-blue. x 1000. media does any liquefac-
tion occur. In bouillon

the organism produces a turbidity which soon settles to the

bottom and forms a

powdery layer on the

wall of the vessel. If

the growth is started on

the surface and the flask

is kept at rest, a distinct

scum forms, and this is

especially suitable for the

development of toxin.

Ordinary bouillon be- ^ ^

comes acid during the *-. / £tfc

first two or three days,

and several days later

again acquires an alka-
line reaction. If, how-
ever, the bouillon is

dextrose-free (p. 80) the FIG. 116. — Diphtheria bacilli of larger size

than in previous figure, showing also
irregular staining of protoplasm. From
a three days' agar culture.
Stained with weak carbol-fuchsin. x 1000.

'"im ~

does not

acid reaction
occur. The organism
not only ferments glu-
cose, but also galactose,
Isevulose, maltose, and usually also glycerine and lactose in

POWERS OF RESISTANCE OF BACILLUS 403

older cultures; mannite and saccharose are not fermented
(Graham-Smith).

In culture media the bacilli show the same characters as in
the membrane, but the beading is a more marked feature, except
in the very youngest cultures, and sometimes the stained proto-
plasm has a sort of septate appearance (Figs. 115, 116). They
are at first fairly uniform in size and shape, but later involution
forms may appear, especially on the less favourable media, such
as agar. Many are swollen at their ends into club-shaped masses
which stain deeply, and the protoplasm becomes broken up into
globules with unstained parts between (Fig. 117). Some
become thicker through-
out, and segmented so *•] *'
as to appear like large **
cocci, and others show •/
globules at their ends, >^
the rest of the rod

appearing as a faintly .- '-r „

stained line. Occasion- ; / ^

ally branched forms are •' ' - *^V

met with. The bacilli ,1

are non-motile, and do -^ «^ /

not form spores. ~^

Staining. — They take
up the basic aniline
dyes, e.g. methylene-
blue in watery solution, v-

with great readiness, FIG. 117. — Involution forms of the diphtheria
arul «taii dppnlv thp bacillus; from an agar culture of seven

PV» ' days' growth. See also Plate III., Fig. 13.

granules often giving the Stained with carbol-thionin-blue. x 1000.
metachromatic reaction

as described. They also retain the colour in Gram's method,
though they are more easily decolorised than the pyogenic cocci.
By Neisser's stain (p. 115) the granules are stained almost black,
the rest of the bacillary substance yellowish-brown, or by the
Dew method, pink (Plate III., Fig. 12).

Powers of Resistance, etc. — In cultures the bacilli possess
long duration of life ; at room temperature they may survive
for two months or longer. In the moist condition, whether in
cultures or in membrane, they have a low power of resistance,
being killed at 60° C. in a few minutes. On the other hand, in
the dry condition they have great powers of endurance. In
membrane which is perfectly dry, for example, they can resist a
temperature of 98° C. for an hour. Dried diphtheria membrane,

404 DIPHTHERIA

kept in the absence of light and at the room temperature, has
been proved to contain diphtheria bacilli still living and virulent
at the end of several months. The presence of light, moisture,
or a higher temperature, causes them to die out more rapidly.
Corresponding results have been obtained with bacilli obtained
from cultures and kept on dried threads. These facts, especially
with regard to drying, are of great importance, as they show that
the contagium of diphtheria may be preserved for a long time
in the dried membrane.

Effects of Inoculation. — In considering the effects produced
in animals by experimental inoculations of pure cultures, we
have to keep in view the local changes which occur in diphtheria,
and also the symptoms of general poisoning.

As Loffler stated in his original paper, inoculation of the
healthy mucous membranes of various animals with pure cultures
causes no lesion, but the formation of false membrane may
result wrhen the surface is injured by scarification or otherwise.
A similar result may be obtained when the trachea is inoculated
after tracheotomy has been performed. In this case the
surrounding tissues may become the seat of a blood-stained
cedema, and the lymphatic glands become enlarged, the general
picture resembling pretty closely that of laryngeal diphtheria.
The membrane produced by such experiments is usually less
firm than in human diphtheria, and the bacilli in the membrane
are less numerous. Rabbits inoculated after tracheotomy often
die, and Roux and Yersin were the first to observe that in some
cases paralysis may appear before death.

Subcutaneous injection in guinea-pigs of diphtheria bacilli in
a suitable dose produces death within thirty-six hours. At the
site of inoculation there is usually a small patch of greyish
membrane, whilst in the tissues around there is extensive
inflammatory oedema, often associated with haemorrhages, and
there is also some swelling of the corresponding lymphatic glands.
The internal organs show general congestion, the suprarenal
capsules being especially reddened and often haemorrhagic. The
renal epithelium may show cloudy swelling, and there is often
effusion into the pleural cavities. After injection the bacilli in-
crease in number for a few hours, but multiplication soon ceases,
and at the time of death they may be less numerous than when
injected. The bacilli remain practically local, cultures made
from the blood and internal organs usually giving negative results,
though sometimes a few colonies may be obtained. If a non-
fatal dose of a culture be injected, a local necrosis of the skin
and subcutaneous tissue may follow at the site of inoculation.

THE TOXINS OF DIPHTHERIA 405

In rabbits, after subcutaneous inoculation, results of the same
nature follow, but these animals are less susceptible than guinea-
pigs, and the dose requires to be proportionately larger. Roux
and Yersin found that after intravenous injection the bacilli
rapidly disappeared from the blood, and when 1 c.c. of a broth
culture had been injected no trace of the organisms could be
detected by culture after twenty-four hours ; nevertheless the
animals died with symptoms of general toxaemia, nephritis also
being often present (cf. Cholera, p. 455). The dog and sheep
are also susceptible to inoculation with virulent bacilli, but the
mouse and rat enjoy a high degree of immunity.

Klein found that cats also were susceptible to inoculation. The
animals usually die after a few days, and post-mortem there is well-marked
nephritis. He also found that after subcutaneous injection in cows, a
vesicular eruption appeared on the teats of the udder, the fluid in which
contained diphtheria bacilli. At the time of death the diphtheria bacilli
were still alive and virulent at the site of injection. The most striking
result of these experiments is that the diphtheria bacilli passed into the
circulation and were present in the eruption on the udder. He considers
that this may throw light on certain epidemics of diphtheria in which
the contagion was apparently carried by the milk. Other observers,
e.g. Abbott, have, however, failed to obtain similar results. Dean and
Todd,in investigating an outbreak of diphtheria traceable to a milk supply,
found a vesicular eruption on the teats of the udder in which diphtheria
bacilli were present. They, however, came to the conclusion that these
bacilli were not the cause of the eruption, but were the result of a
secondary contamination, probably from the saliva of the milkers. The
existence of a true diphtheria infection in cows must still be considered
doubtful. A case of true diphtheria in the horse has been described by
Cobbett.

The Toxins of Diphtheria. — As in the above experiments
the symptoms of poisoning and ultimately a fatal result occur
when the bacilli are diminishing in number, or even after they
have practically disappeared, Roux and Yersin inferred that the
chief effects were produced by toxins, and this supposition they
proved to be correct. They showed that broth cultures of three
or four weeks' growth freed from bacilli by filtration were highly
toxic. The filtrate when injected into guinea-pigs and other
animals produces practically the same effects as the living bacilli ;
locally there is fibrinous exudation but a considerable amount
of inflammatory (jedema, and, if the animal survive long enough,
necrosis in varying degree of the superficial tissues may follow.
The toxicity may be so great that '005 c.c. or even less may be
fatal to a guinea-pig in five days.

After injection either of the toxin or of the living bacilli,
\\lieii the animals survive long enough, paralytic phenomena

406 DIPHTHERIA

occasionally occur. The hind-limbs are usually affected first, the
paralysis afterwards extending to other parts, though sometimes
the fore-limbs and neck first show the condition. Sometimes
symptoms of paralysis do not appear till two or three weeks
after inoculation. After paralysis has appeared, a fatal result
usually follows in the smaller animals, but in dogs recovery may
take place. There is evidence that these paralytic phenomena are
produced by toxone, as they specially occur when there is injected
along with the toxin sufficient antitoxin to neutralise the more
rapidly acting toxin proper. This toxone is supposed by Ehrlich
to have a different toxic action, i.e. a different toxophorous
group (p. 198), from that of the ordinary toxin ; it produces
the late nervous phenomena, while its local action on the tissues
is very slight. It also has a weaker affinity for antitoxin, and
thus much of it may be left unneutralised. It is to be noted in
this connection that paralytic symptoms are of not uncommon
• occurrence in the human subject after treatment with antitoxin,
the explanation of which occurrence is probably the same as
that just given. One point of much interest is the high degree
of resistance to the toxin possessed by mice and rats. Roux and
Yersin, for example, found that 2 c.c. of toxin, which was
sufficient to kill a rabbit in sixty hours, had no effect on a
mouse, whilst of this toxin even — c.c. produced extensive
necrosis of the skin of the guinea-pig.

Preparation of the Toxin. — The obtaining of a very active
toxin in large quantities is an essential in the preparation of anti-
diphtheritic serum. Certain conditions favour the development
of a high degree of toxicity, namely, a free supply of oxygen, the
presence of a large proportion of peptone or albumin in the
medium, and the absence of substances which produce an acid
reaction. In the earlier work a current of sterile air was
made to pass over the surface of the medium, as it was found
that by this means the period of acid reaction was shortened and
the toxin formation favoured. This expedient is now considered
unnecessary if an alkaline medium free from glucose is used, as
in this no acid reaction is developed ; it is then sufficient to
grow the cultures in shallow flasks. The absence of glucose
may be attained by the method described above (p. 80), or by
using for the preparation of the meat extract flesh which is just
commencing to putrefy (Spronck). L. Martin uses a medium
composed of equal parts of freshly prepared peptone (by digest-
ing pigs' stomachs with HC1 at 35° C.), and glucose-free veal
bouillon. By this medium he has obtained a toxin of which
c.c. is the fatal dose to a guinea-pig of 500 grms. Park

NATURE OF THE TOXIN 407

and Williams and also Dean find that the amount of glucose
present in ordinary beef is not sufficient to interfere with toxin
formation, provided that a considerable amount of peptone, 2 per
cent., be added, and the medium be made sufficiently alkaline ;
after making it neutral to litmus they add to each litre of broth
7 c.c. of normal caustic soda solution. There is in all cases a
period at which the toxicity reaches a maximum ; Roux and
Yersin found this period to be two to three weeks, but later
observers find that in favourable conditions the greatest toxicity
is reached about the tenth to twelfth day, sometimes even
earlier. It may be added that the power of toxin formation
varies much in different races of the diphtheria bacillus, and
that many may require to be tested ere one suitable is
obtained.

Properties and Nature of the Toxin. — The toxic substance in
filtered cultures is a relatively unstable body. When kept in
sealed tubes in the absence of light, it may preserve its powers
little altered for several months, but, on the other hand, it
gradually loses them when exposed to the action of light and
air. As will be shown later (p. 527), the toxin probably does
not become destroyed, but its toxophorous group suffers a sort of
deterioration, so that a toxoid is formed which has still the
power of combining with antitoxins. Heating at 58° C. for
two hours destroys the toxic properties in great part, but not
altogether. When, however, the toxin is evaporated to dryness,
it has much greater resistance to heat. One striking fact,
discovered by Roux and Yersin, is that after an organic acid,
such as tartaric acid, is added to the toxin the toxic property
disappears, but it can be in great part restored by again
making the fluid alkaline.

Guinochet found that toxin was formed by the bacilli when
grown in urine with no proteid bodies present. After growth
had taken place he could not detect proteid bodies in the fluid,
but, on account of the very minute amount of toxin present,
their absence could not be excluded. Uschinsky also found that
toxic bodies were produced by diphtheria bacilli when grown in
a proteid-free medium.1 It follows from this that if the toxin
is a proteid, it may be formed by synthesis within the bodies of
the bacilli. Brieger and Boer have separated from diphtheria
cultures a toxic body which gives no proteid reaction (vide p.

i Uschinsky's medium has the following composition : water, 1000 parts ;
glycerin, 30-40 ; sodium chloride, 5-7 ; calcium chloride, 'I ; magnesium
sulphate, '2- '4 ; di-potassium phosphate, *2-'25 ; ammonium lactate, 6-7 ;
sodium asparaginate, 3-4.

408 DIPHTHERIA

193). Whether or not diphtheria toxin is of proteid nature
must, however, be considered to be a question not yet settled.

Toxic bodies have also been obtained from the tissues of those
who have died from diphtheria. Roux and Yersin, by using
a filtered watery extract from the spleen from very virulent cases
of diphtheria, produced in animals death after wasting and
paralysis, and also obtained similar results by employing the
urine. The subject of toxic bodies in the tissues has, however,
been specially worked out by Sidney Martin. He has separated
from the tissues, and especially from the spleen, of patients who
have died from diphtheria, by precipitation with alcohol, chemical
substances of two kinds, namely, albumoses (proto- and deutero-,
but especially the latter), and an organic acid. The albumoses,
when injected into rabbits, especially in repeated doses, produce
fever, diarrhoea, paresis, and loss of weight, with ultimately a
fatal result. He further found that this paresis is due to
well-marked changes in the nerves. Substances obtained from
diphtheria membrane have an action like that of the bodies
obtained from the spleen, but in higher degree. Martin con-
siders that this is due to the presence in the membrane of an
enzyme which has a proteolytic action within the body, resulting
in the formation of poisonous albumoses.

Immunity. — This is described in the general chapter on
Immunity. It is sufficient to state here that a high degree of
immunity, against both the bacilli and their toxins, can be
produced in various animals by gradually increasing doses either
of the bacilli or of their filtered toxins (vide Chapter XXI.).

Variations in the Virulence of the Diphtheria Bacillus. — In
cultures on serum the diphtheria bacilli retain their virulence
fairly well, but they lose it much more quickly on less suitable
media, such as glycerin agar. Roux and Yersin found that,
when the bacilli were grown at an abnormally high temperature,
namely, 39*5° C., and in a current of air, the virulence diminished
so much that they became practically innocuous. When the
virulence was much diminished, these observers found that it
could be restored if the bacilli were inoculated into animals
along with streptococci, inoculation of the bacilli alone not
being successful for this purpose. If, however, the virulence
had fallen very low, even the presence of the streptococci was
insufficient to restore it. The virulence is tested by the amount
of living bacilli necessary to produce a fatal result on injection,
and is to be distinguished from the power of producing toxin in
a fluid medium; as pointed out by Dean, the two properties
often do not correspond. It has been abundantly established

BACILLI ALLIED TO DIPHTHERIA BACILLUS 409

that, after the cure of the disease, the bacilli may persist in the
mouth for weeks, though they often quickly disappear. Roux
and Yersin found, by making cultures at various stages after
the termination of the disease, that these bacilli in the mouth
gradually become attenuated.

I j, Martin, moreover, has shown that some races of diphtheria
bacillus are so attenuated that 1 c.c. of a twenty-four hours'
growth in bouillon does not cause death in a guinea-pig, yet their
true nature is shown not only by their miscroscopical characters,
etc., but also by the fact that on more prolonged growth they
form small quantities of toxin, which is neutralisable by diphtheria
antitoxin. The persistence of these non-virulent bacilli in the
throats of those who have suffered from the disease, and their
occasional presence in quite healthy individuals, may manifestly
be of importance in relation to the continuance of the infection
and the reappearance of epidemics of the disease.

BACILLI ALLIED TO THE DIPHTHERIA BACILLUS.

Bacteriological examinations carried on within recent times
have shown that the diphtheria bacillus is merely a member of
a group of organisms with closely allied characters which are
of common occurrence and have a wide distribution. The terms
"pseudo-diphtheria bacilli " and "diphtheroid bacilli " have been
applied in a loose way to organisms which resemble the
diphtheria bacillus microscopically, especially as regards the
beaded appearance. Such bacilli have been obtained from the
mouth, nose, skin, genital organs, and even from the blood in
certain diseases. They are to be met with sometimes in condi-
tions of health, and they have been obtained from many diverse
morbid conditions — from skin diseases, from coryza, from
leprosy, and even from general paralysis of the insane. As
has been found with other groups, the differentiation is a matter
of considerable difficulty. Some are practically identical with
the diphtheria bacillus both morphologically and culturally, and
a few even give the characteristic reaction with Neisser's stain ;
others, again, differ in essential particulars. The fermentative
action on sugars1 has also been called into requisition as a
means of distinguishing them, but the results obtained cannot
be said to be of a definite character, and further work is
necessary. It may be stated, however, that most observers
have found the diphtheria bacillus of all the members of the
group to 'be the most active acid-producer, though here the
1 Vide a paper by Graham-Smith, Journal oj Hygiene, vi. 286.

410 DIPHTHERIA

difference seems to be one of degree rather than of kind. The
absence of the power of fermenting certain sugars, notably
glucose, may, however, be accepted in any particular case as
sufficient to exclude the organism from being the diphtheria
bacillus. From these facts, and from what has been stated
with regard to attenuated diphtheria bacilli, it will be seen that
an absolute decision as to the nature of a suspected organism
may in some cases be a practical impossibility. It may be that
some of the " diphtheroid " organisms cultivated have really
been non-virulent diphtheria bacilli. The bearing of this on
the practical means of diagnosis will be discussed below.

Ford Robertson and his co-workers have obtained from
numerous cases of general paralysis of the insane cultures of a
diphtheroid organism, which he considers is the chief agent in
producing the condition of chronic intoxication underlying the
disease. The organism has been obtained from various situations,
including the central nervous system, but it seems to flourish
specially in the respiratory and alimentary tracts. It closely
resembles the diphtheria bacillus ; the morphological and cultural
characters are indeed practically identical, but the diphtheroid
bacillus is non-pathogenic to the guinea-pig. Robertson and
Shennan found that when administered to rats by the alimentary
tract it produced certain nervous symptoms which were associ-
ated with changes in the brain of the same order as those in
general paralysis. Further research on this subject is still
necessary.

The term " pseudo-diphtheria bacillus " is often restricted by
present writers to an organism frequently met with in the
throat. This organism, which is also known as Hofmann's
bacillus, merits a separate description.

Hofmann's Bacillus — Pseudo-Diphtheria Bacillus. — This
organism, described by Hofmann in 1888, is probably the same
as one observed by Loffler in the previous year, and regarded
by him as being a distinct species from the diphtheria bacillus.
The organism is a shorter bacillus than the diphtheria bacillus,
with usually a single unstained septum running across it, though
sometimes there may be more than one (Fig. 118). The typical
beaded appearance is rarely seen, and the characteristic reaction
with Neisser's stain is not given, though in old cultures a few
granules which stain deeply may sometimes be found. It grows
readily on the same media as the diphtheria bacillus, but the
colonies are whiter and more opaque. It does not form acid
from glucose or other sugars, and is non-pathogenic to the
guinea-pig. Involution forms may sometimes be produced by

PSEUDO-DIPHTHERIA BACILLUS 411

it. It is usually a relatively easy matter to distinguish this
organism from the diphtheria bacillus.

Hofmann's bacillus is of comparatively common occurrence in
the throat in normal as well as diseased conditions, including
diphtheria ; it seems to be specially frequent in poorly nourished
children of the lower classes. Cobbett found it 157 times in an
examination of 692 persons examined, of whom 650 were not
suffering from diphtheria. Boycott's statistics show that the
time of its maximum seasonal prevalence precedes that of the
diphtheria bacillus. To what extent, if any, it is responsible
for pathological changes in the throat, must be considered a
question which is not
yet settled. Hewlett
and Knight have found
evidence that a true
diphtheria bacillus may
assume the characters of
Hofmann's bacillus, but
attempts to effect the
transformation have met
with negative results in
the hands of other
observers. The general
opinion is that the two
organisms are distinct
species with compara-
tively easily distinguished

characters. Flf; ns.—Pseudo-diphtheria bacillus (Hof-

mann's). Young agar culture. See also
Xerosis Bacillus. — This piate III., Fig. 14.

term has been given to an Stained with thionin-blue. x 1000.

organism first observed by
Kuschbert and Neisser in

xerosis of the conjunctiva, and which has been since found in many other
affections of the conjunctiva and also in normal conditions. Morpho-
logically it is practically similar to the diphtheria bacillus, and even in
cultures presents very minor differences ; it, however, grows more slowly
on serum, and its colonies have a tougher consistence and a more irregular
margin. It is non-virulent to animals, and does not produce an acid
reaction in glucose bouillon, or does so to only a slight extent ; in this
way it can be distinguished from the diphtheria bacillus. It is still
doubtful whether it is pathogenic to the human subject. Its morpho-
logical characters are shown in Fig. 119.

Action of the Diphtheria Bacillus — Summary. — From a
study of the morbid changes in diphtheria and of the results
produced experimentally by the bacillus and its toxins, the

412 DIPHTHERIA

following summary may be given of its action in the body.
Locally, the bacillus produces inflammatory change with
fibrinous exudation, but at the same time cellular necrosis is
also an outstanding feature. Though false membranes have not
been produced by the toxins, a necrotic action may result when
these are injected subcutaneously. The toxins also act upon
the blood vessels, and hence oedema and tendency to hemorrhage
are produced ; this action on the vessels is also exemplified by
the general congestion of organs. The hyaline change in the
walls of arterioles and capillaries so often met with in diphtheria
is another example of the action of the toxin. The toxins have
also a pernicious action on highly developed cells and on nerve

fibres. Thus in the kidney
g. ^ cloudy swelling occurs,

\ "•**^*iW» which may be followed

by actual necrosis of the
*tjj£ secreting cells, and along

with these changes albu-
minuria is present. The
action is also well seen in
the case of the muscle
fibres of the heart, which

„

V>*% m&y undergo a sort of

hyaline change, followed
by granular disintegra-
tion or by an actual

- " ; fatty degeneration. These

FIG. 119.— Xerosis bacillus from a young changes are of great
agar culture. xlOOO. importance in relation to

heart failure in the disease.

Changes of a somewhat similar nature have been recently
observed in the nerve cells of the central nervous system, those
lying near the capillaries, it is said, being affected first. There
is also the striking change in the peripheral nerves, which is
shown first by the disintegration of the medullary sheaths as
already described. It is, however, still a matter of dispute to
what extent these nerve lesions are of primary nature or
secondary to changes in the nerve cells.

Methods of Diagnosis. — The bacteriological diagnosis of
diphtheria depends on the discovery of the bacillus. As the
bacillus occurs in largest numbers in the membrane, a portion of
this should be obtained whenever it is possible, and transferred
to a sterile test-tube. (The tube can be readily sterilised by
boiling some water in it.) If, however, membrane cannot be

METHODS OF DIAGNOSIS 413

obtained, a scraping of the surface with a platinum loop may be
sufficient. Where the membrane is confined to the trachea the
bacilli are often present in the secretions of the pharynx, and
may be obtained from that situation by swabbing it with cotton-
wool (non-antiseptic), the swab being put into a sterile tube or
bottle for transport. A convenient method is to twist a piece of
cotton-wool round the roughened end of a piece of very stout
iron wire, 6 inches long, and pass the other end of the latter
through a cotton plug inserted in the mouth of a test-tube
(compare Fig. 46, the wire taking the place of the pipette), and
sterilise. In use the wire and plug are extracted in one piece,
and after swabbing are replaced in the tube for transit. A
scraping may be made off the swab for microscopic examination,
and the swab may be smeared over the surface of a serum tube
to obtain a culture. This method of taking and treating swabs
is that usually employed in routine public health work. The
results obtained ordinarily suffice for the diagnosis of cases
suspected to be diphtheritic in nature.

The means for identifying the bacillus are (a) By Micro-
scopical Examination. — For microscopical examination it is
sufficient to tease out a piece of the membrane with forceps and
rub it on a cover-glass ; if it be somewhat dry, a small drop of
normal saline should be added. The films are then dried in the
usual way, and stained with any ordinary basic stain, though
methylene-blue is on the whole to be preferred, used either as
a saturated watery solution or in the form of Loffler's solution.
After staining for two or three minutes, the films are washed
in water, dried, and mounted. As a rule no decolorising is
necessary, as the blue does not overstain. Neisser's stain (p. 1 1 5)
may also be used with advantage, although it is to be noted
that sometimes in a secretion the diphtheria bacillus does not
react typically to this stain. Any secretion from the pharynx
or other part is to be treated in the same way. The value of
microscopical examination alone depends much upon the experi-
ence of the observer. In some cases the bacilli are present in
characteristic form in such numbers as to leave no doubt in
the matter. In other cases a few only may be found, mixed
with large numbers of other organisms, and sometimes their
characters are not sufficiently distinct to render a definite opinion
possible. The bacillus may be frequently obtained by means
of cultures, when the result of microscopical examination is
inconclusive. As already said, however, microscopical examina-
tion alone is more reliable after the observer has had experience
in examining cases of diphtheria and making cultures from them.

414 DIPHTHERIA

(6) By making Cultures. — For this purpose a piece of the
membrane should be separated by forceps from the pharynx or
other part when that is possible. It should be then washed
well in a tube containing sterile water, most of the surface im-
purities being removed in this way. A fragment is then fixed in
a platinum loop by means of sterile forceps, and a series of
stroke cultures is made on the surface of any of the media
mentioned (p. 401), the same portion of the membrane being
always brought into contact with the surface. The tubes are
then incubated at 37° C., and, in the case of the serum
media and blood-agar, the circular colonies of the diphtheria
bacillus are well formed within twenty -four hours. A small
portion of a colony is then removed by means of a platinum
needle, stained, and examined in the usual way, Neisser's stain
being also applied. When the material has been taken from
the throat, an organism with all the morphological and cultural
characters of the diphtheria bacillus may for all practical
purposes be accepted as the diphtheria bacillus.

In cases where a suspicion arises that the organism found
is a ' pseudo - diphtheria bacillus, bouillon containing a trace
of glucose should be inoculated and incubated at 37° C.
The reaction should be tested after one and after two days'
growth. If it remains alkaline, the diphtheria bacillus may be
excluded. If an acid reaction results, then all the microscopical
and cultural characters must be carefully observed, and the
virulence of the bacillus may be ascertained by inoculating a
guinea-pig, say with 1 c.c. of a broth culture of two days' growth.
(See also pp. 404, 410.) A fatal result with characteristic
appearances may be taken as positive evidence ; but if the animal
survive there is still theoretically the possibility that the
organism is an attentuated diphtheria bacillus (p. 408).

CHAPTER XVII.

TETANUS1: CONDITIONS CAUSED BY OTHER
ANAEROBIC BACILLI.

Introductory. — Tetanus (German, Wundstarrkrampf) is a
disease which in natural conditions affects chiefly man and the
horse. Clinically it is characterised by the gradual onset of
general stiffness and spasms of the voluntary muscles, com-
mencing in those of the jaw and the back of the neck, and
extending to all the muscles of the body. These spasms
are of a tonic nature, and, as the disease advances, succeed
each other with only a slight intermission of time. There
are often, towards the end of a case, fever and rise of
respiration and pulse-rate. The disease is usually associated
with a wound received from four to fourteen days previously,
and which has been denied by earth or dung. The disease is,
in the majority of cases, fatal.

Historical. — The general association of the development of tetanus with
the presence of wounds-, though these might be very small, suggested that
some infection took place through the latter, but for long nothing was
known as to the nature of this infection. Carle and Rattone in 1884
announced that they had produced the disease in a number of animals by
inoculation with material from a wound in tetanus. They thus demon-
strated the transmissibility of the disease. Nicolaier (1885) infected mice
and rabbits with garden earth, and found that many of them developed
tetanus. Suppuration occurred in the neighbourhood of the point of
inoculation, and in this pus, besides other organisms, there was always
present, when tetanus had occurred, a bacillus having certain constant
microscopic characters. Inoculation of fresh animals with such pus
reproduced the disease. Nicolaier's attempts at its isolation by the
ordinary gelatin plate-culture method were, however, unsuccessful. He
succeeded in getting it to grow in liquid blood serum, but always in

1 This disease is not to be confused with the " tetany " of infants, which in
its essential pathology probably differs from tetanus (vide Frankl-Hochwart,
" Die Tetanic der Erwachsenen," Vienna, 1907). This remark of course does
not exclude the possibility of the occurrence of true tetanus in very young
subjects.

415

416 TETANUS

mixture with other organisms. Infection of animals with such a culture
produced the disease. These results were confirmed by Rosenbach, who,
though failing to obtain a pure culture, cultivated the other organisms
present, and inoculated them, but with negative results. He further
pointed out, as characteristic of the bacillus, its development of terminal
spores. In 1889, Kitasato succeeded in isolating from the local suppura-
tion of mice inoculated from a human case, several bacilli, only one of
which, when injected in pure culture into animals, caused the disease,
and which was now named the b. tetani. This organism is the same as
that observed by Nicolaier and Rosenbach. Kitasato found that the
cause of earlier culture failures was the fact that it could only grow in the
absence of oxygen. The pathology of the disease was further elucidated
by Faber, who, having isolated bacterium- free poisons from cultures,
reproduced the symptoms of the disease.

Bacillus Tetani. — If in a case of tetanus naturally arising
in man, there be a definite wound with pus formation or necrotic
change, the bacillus tetani may be recognised in film preparations
from the pus, if the characteristic spore formation has occurred
(Fig. 120). If, however, the tetanus bacilli have not formed
spores, they appear as somewhat slender rods, without present-
ing any characteristic features. There is usually present in such
pus a great variety of other organisms — cocci and bacilli. The
characters of the bacillus are, therefore, best studied in cultures.
It is then seen to be a slender organism, usually about 4 /JL to
5 JJL in length and *4 JJL in thickness, with somewhat rounded
ends. Besides occurring as short rods it also develops
filamentous forms, the latter being more common in fluid media.
It stains readily by any of the usual stains and also by Gram's
method. A feature in it is the uniformity with which the
protoplasm stains. It is very slightly motile, and its motility
can be best studied in an anaerobic hanging-drop preparation.
When stained by the special methods already described,
it is found to possess numerous delicate flagella attached
both at the sides and at the ends (Fig. 121). These flagella,
though they may be of considerable length, are usually
curled up close to the body of the bacillus. The formation of
flagella can be best studied in preparations made from surface
anaerobic cultures (p. 68). As is the case with many other
anaerobic flagellated bacteria, the flagella, on becoming detached,
often become massed together in the form of spirals of striking
appearance (Fig. 122). At incubation temperature b. tetani
readily forms spores, and then presents a very characteristic
appearance. The spores are round, and in diameter may be
three or four times the thickness of the bacilli. They are
developed at one end of a bacillus, which thus assumes what is
usually described as the drumstick form (Figs. 120, 123). In

BACILLUS TETANI

a specimen stained with a watery solution of gentian-violet or
methylene-blue, the spores are uncoloured except at the periphery,
so that the appearance of a small ring is produced ; if a powerful
stain such as carbol-fuchsin be applied for some time, the spores
become deeply coloured like the bacilli. Further, especially if

Fio. 120. — Film preparation of discharge from wound in a case
of tetanus, showing several tetanus bacilli of "drumstick" form.
(The thicker bacillus present is not a tetanus bacillus, but a
putrefactive anaerobe which was obtained in pure culture from the
wound. )

Stained with gentian -violet, x 1000.

the preparation be heated, many spores may become free from the
bacilli in which they were formed.

Isolation. — The isolation of the tetanus bacillus is somewhat
difficult. By inoculation experiments in animals, its natural
habitat has been proved to be garden soil, and especially the
contents of dung-heaps, where it probably leads a saprophytic
existence, though its function as a saprophyte is unknown. From
such sources and from the pus of wounds in tetanus, occurring
naturally or experimentally produced, it has been isolated^by

418

TETANUS

means of the methods appropriate for anaerobic bacteria. The
best methods for dealing with such pus are as follows : —

(1) The principle is to take advantage of the resistance of the
spores of the bacillus to heat. A sloped tube of inspissated
serum or a deep tube of glucose agar is inoculated with the pus
and incubated anaerobically at 37° C. for forty-eight hours, at
the end of which time numerous spore-bearing bacilli can often
be observed microscopically. The culture is then kept at 80° C.

FIG. 121.— Tetanus bacilli, showing flagella.
Stained by Rd. Mnir's method, x 1000.

for from three-quarters to one hour, with the view of killing all
organisms except those which have spored. A loopful is then
added to glucose gelatin, and roll-tube cultures are made in the
usual way and kept in an atmosphere of hydrogen at 22° C. ;
after five days the plates are ready for examination. Kitasato
compares the colonies in gelatin plates to those of the b.
subtilis. They consist of a thick centre with shoots radiating
out on all sides. They liquefy the gelatin more slowly than the
b. subtilis. This method of isolation is not always successful,
partly because along with the tetanus bacilli, both in its natural

ISOLATION OF THE BACILLUS

419

habitats outside the body and in the pus of wounds, other spore-
forming obligatory and

facultative anaerobes

occur, which grow faster

than the tetanus bacillus,

and thus overgrow it.

(2) If in any discharge /

the spore-bearing tetanus , '

bacilli be seen on micro- *

scopic examination, then

a method of isolation

based on the same prin-
ciple as the last may

be adopted. Inocula- ?"*"* *

tions with the suspected

material are made in

half a dozen deep tubes

of glucose bouillon, pre- FIG. 122.— Spiral composed of numerous

viously raised to a tem- twisted tiagella of the tetanus bacillus.

perature of 100° C. After Staiue(l b* R(L Muir's raethod' x 100°-

inoculation they are

again placed in boiling water and kept for varying times, say

for half a minute, for

one, three, four, five, and

six minutes respectively.

^ They are then plunged

-^— ' in cold water till cool,

and thereafter placed in

j.1, ^ ' «.-1* — j_^__

Fi<;. l'J:J. — Trtiinus liacilli ; some of which
possess sjxnvs. From a culture in glucose
agar, incubated for three days at 37° C.
See also Plate IV., Fig. 20.

Stained with carbol-fuchsin. x 1000.

Bulloch, may be employed. The

the incubator at 37° C.,
in the hope that in one
or other of the tubes all
the organisms present
will have been killed,
except the tenanus spores
which can develop in
pure culture. A series
of deep glucose agar
tubes may also be in-
oculated from the series
of bouillon tubes.

(3) Some method of
anaerobically making
plates, such as that of
isolation of the tetanus

420

TETANUS

bacillus is in many cases a difficult matter, and several methods
should always be tried.

Characters of Cultures. — Pure cultures having been obtained,
sub-cultures can be made in deep upright glucose gelatin or
agar tubes. On f/lucose gelatin in such a tube there commences,
an inch or so below the surface, a growth
consisting of fine straight threads, rather
longer in the lower than in the upper parts
of the tube, radiating out from the needle
track (Fig. 124). Slow liquefaction of the
gelatin takes place, with slight gas formation.
In agar the growth is somewhat similar,
consisting of small nodules along the needle
track, with irregular short offshoots passing
out into the medium (Fig. 128, A). There is
slight formation of gas, but, of course, no
liquefaction. On anaerobic agar plates
colonies have under a low power a feathery
outline (Fig. 125). Growth also occurs in
blood serum and also in glucose bouillon
under anaerobic conditions. The latter is the
medium usually employed for obtaining the
soluble products of the organism. There is
in it at first a slight turbidity, and later a
thin layer of a powdery deposit on the walls
of the vessel. All the cultures give out a
peculiar burnt odour of rather unpleasant
character.

Conditions of Growth, etc. — The b. tetani
grows best at 37° C. The minimum growth
temperature is about 14° C., and below 22° C.
growth takes place very slowly. Growth
takes place in the absence of oxygen, the
organism being an anaerobe. Sporulation
may commence at the end of twenty-four
hours in cultures grown at 37° C., — much
later at lower temperatures. Like other
spores, those of tetanus are extremely resistant. They can
usually withstand boiling for five minutes, and can be kept
in a dry condition for many months without being killed or
losing their virulence. They have also high powers of resistance
to antiseptics.

Pathogenic Effects. — The proof that the b. tetani is the cause
of tetanus is complete. It can be isolated in pure culture, and

FIG. 124.— Stab cul-
ture of the tetanus
bacillus in glucose
gelatin, showing
the lateral shoots
(Kitasato).
Natural size.

PATHOGENIC EFFECTS 421

when re-injected in pure culture it reproduces the disease. It
may be impossible to isolate it from some cases of the disease,
but the cause of this very probably is the small numbers in
which it sometimes occurs.

(a) The Disease as arising naturally. — The disease occurs
naturally, chiefly in horses and in man. Other animals may,
however, be affected. There is usually some wound, often of
a ragged character, which has either been made by an object
soiled with earth or dung, or which has become contaminated
with these substances. There is often a purulent or foetid dis-
charge, though this may be absent. In tetanus following

Kic. 125. — Colonies of the tetanus bacillus on anaerobic
agar plates, seven days old. x 50.

clean operation wounds, catgut ligatures may be the source of
infection. Microscopic examination of sections may show at the
edges of the infected wound necrosed tissue in which the tetanus
bacilli may be very numerous. If a scraping from the wound
be examined microscopically, bacilli resembling the tetanus bacillus
may be recognised. If these have spored, there can be practically
no doubt as to their identity, as the drumstick appearance which
the terminal spore gives to the bacillus is not common among
other bacilli. Care must be taken, however, to distinguish it from
other thicker bacilli with oval spores placed at a short distance
from their extremities, such forms being common in earth,
etc., and also met with in contaminated wounds. It is
important to note that the wound through which infection has

m TETANUS

taken place may be very small, in fact, may consist of a mere
abrasion. In some cases, especially in the tropics, it may be
merely the bite of an insect. The absence of a definite channel
of infection has given rise to the term "idiopathic" tetanus.
There is, however, practically no doubt that all such cases are
true cases of tetanus, and that in all of them the cause is
the b. tetani. The latter has also been found in the bronchial
mucous membrane in some cases of the so-called rheumatic
tetanus, the cause of which is usually said to be cold ; infection
of the intestinal mucosa may also occur.

The pathological changes found post mortem are not striking.
There may be haemorrhages in the muscles which have been the
subject of the spasms. These are probably due to mechanical
causes. Naturally it is in the nervous system that we look for
the most important lesions. Here there is ordinarily a general
redness of the grey matter, and the most striking feature is the
occurrence of irregular patches of slight congestion which are not
limited particularly to grey or white matter, or to any tract of
the latter. These patches are usually best marked in the grey
matter of the medulla and pons. Microscopically there is little
of a definite nature to be found. There is congestion, and there
may be minute haemorrhages in the areas noted by the naked
eye. The ganglion cells may show appearances which have
been regarded as degenerative in nature, and similar changes
have been described in the white matter. The only marked
feature is thus a vascular disturbance in the central nervous
system, with a possible tendency to degeneration in its specialised
cells. Both of these conditions are probably, due to the action
of the toxins of the bacillus. In the case of the cellular degenera-
tions the cells have been observed to return to the normal under
the curative influence of the antitoxins (vide infra). In the
other organs of the body there are no constant changes.

We have said that the general distribution of pathogenic
bacteria throughout the body is probably a relative phenomenon,
and that bacteria usually found locally may occur generally, and
vice versa. With regard to the tetanus bacillus, it is, however,
probably the case that very rarely, if ever, are the organisms
found anywhere except in the local lesion.

(b) The Artificially-produced Disease. — The disease can be
communicated to animals by any of the usual methods of inocula-
tion, but does not arise in animals fed with bacilli, whether
these contain spores or not. Kitasato found that pure cultures,
injected subcutaneously or intravenously, caused death in mice,
rats, guinea-pigs, and rabbits. In mice, symptoms appear in a

TOXINS OF THE TETANUS BACILLUS 423

day, and death occurs in two or three days, after inoculation
with a loopful of a bouillon culture. The other animals
mentioned require larger doses, and death does not occur so
rapidly. Usually in animals injected subcutaneously the spasms
begin in the limb nearest the point of inoculation. In the
case of intravenous inoculation the spasms begin in the
extensor muscles of the trunk, as in the natural disease in man.
After death there is found slight hypenemia without pus forma-
tion, at the seat of inoculation. The bacilli diminish in number,
and may be absent at the time of death. The organs generally
slio\\ little change.

Kitasato stated that in his earlier experiments the quantity of
culture medium injected along with the bacilli already contained
enough of the poisonous bodies formed by the bacilli to cause
death. The symptoms came on sooner than by the improved
method mentioned below, and were, therefore, due to the toxins
already present. In his subsequent work, therefore, he employed
splinters of wood soaked in cultures in which spores were
present, and subsequently subjected for one hour to a tempera-
ture of 80" C. The latter treatment not only killed all the
Yegetative forms of the organism, but, as we shall see, was
sufficient to destroy the activity of the toxins. When such
splinters are introduced subcutaneously, death results by the
development of the spores which they carry. In this way he
completed the proof that the bacilli by themselves can form
toxins in the body and produce the disease. Further, if a
small quantity of garden earth be placed under the skin of a
mouse, death from tetanus takes place in a great many cases.
[Sometimes, however, in such circumstances death occurs with-
out tetanic symptoms, and is not due to the tetanus bacillus but
to the bacillus of malignant cedema, which also is of common
occurrence in the soil (vide infrd)J\ By such experiments,
supplemented by the culture experiments mentioned, the natural
habitats of the b. tetani, as given above, have become known.

The Toxins of the Tetanus Bacillus. — The tetanus bacillus
being thus accepted as the cause of the disease, we have to
consider how it produces its pathogenic effects.

Almost contemporaneously with the work on diphtheria was the
attempt made with regard to tetanus to explain the general symptoms
by supposing that the bacillus could excrete soluble poisons. The
earlier results, in which certain bases, tetanin and tetanotoxin, were
said to have been isolated, have only a historic interest, as they were
obtained by faulty methods. In 1890, Brieger and Fraenkel announced
that they had isolated a toxalbumin from tetanus cultures, and this body
\\;is independently discovered by Faber in the same year. Brieger and

424 TETANUS

Fraenkel's body consisted practically of an alcoholic precipitate from
filtered cultures in bouillon, and was undoubtedly toxic. Within recent
years such attempts to isolate tetanus toxins in a pure condition have
practically been abandoned, and attention has been turned to the
investigation of the physiological effects either of the crude toxin
present in filtered bouillon cultures, or of the precipitate produced from
the same by ammonium sulphate (cf. p. 195).

The toxic properties of bacterium-free filtrates of pure
cultures of the b. tetani were investigated in 1891 by Kitasato.
This observer found that when the filtrate, in certain doses, was
injected subcutaneously or intravenously into mice, tetanic spasms
developed, first in muscles contiguous to the site of inoculation,
and later all over the body. Death resulted. He found that
guinea-pigs were more susceptible than mice, and rabbits less so.
In order that a strongly toxic bouillon be produced, it must
originally have been either neutral or slightly alkaline. Kitasato
further found that the toxin was easily injured by heat. Exposure
for a few minutes at 65° C. destroyed it. It was also destroyed
by twenty minutes' exposure at 60° C., and by one and a half
hours' at 55° C. Drying had no effect. It was, however,
destroyed by various chemicals such as pyrogallol and also by
sunlight.

In anaerobic bouillon cultures the maximum toxicity is de-
veloped in from ten to fifteen days. Behring pointed out that
after the filtration of cultures containing toxin, the latter may
very rapidly lose its power, and in a few days may only possess
yj^th of its original toxicity. This he attributed to such factors
as temperature and light, and especially to the action of oxygen.
Toxins should thus have a layer of toluol floated on the surface
and be kept in a cool, dark place. The effect of harmful agents
on the crude toxin is apparently to cause a degeneration of the
true toxin so as to form what it is convenient at present to call
toxoids similar to those produced in the case of diphtheria toxin,
and it is also true here that the toxoids while losing their
toxicity may still retain their power of producing immunity
against the potent toxin. Further, altogether apart from the
occurrence side by side in the crude toxin of strong and weak
poisons, it has been shown that such crude toxin contains toxic
substances of probably quite a different nature. Ehrlich has
shown that besides the predominant spasm-producing toxin
(called by him tetanospasmin), there exists in crude toxin a
poison capable of producing the solution of certain red blood
corpuscles. This hsemolytic agent he calls tetanolysin. It does
not occur in all samples of crude tetanus toxin, nor is it found

TOXINS OF THE TETANUS BACILLUS 425

when a bouillon culture of the bacillus is filtered through
I>orcelain. To obtain it the fresh culture must be treated by
ammonium sulphate, as described in the method of obtaining
concentrated toxins (p. 195). This substance also has the
power of originating an antitoxin, so that certain antitetanic sera
can protect red blood corpuscles against its action. Madsen,
studying the interactions of this anti-tetanolysin with the
tetanolysin, has shown that phenomena can be demonstrated
similar to those noted by Ehrlich as occurring with diphtheria
toxin, and which the latter interpreted as indicating the presence
of degenerated toxins (toxoids) in the crude poison. With
tetanus as with diphtheria toxin the action of an acid is to
cause an apparent disappearance of toxicity, but if before a
certain time has elapsed the acid be neutralised by alkali, then a
degree of the toxicity returns.

As with other members of the group, nothing is known of the
nature of tetanus toxin. Uschinsky has found that the tetanus
bacillus can produce its toxin when growing in a fluid containing
no proteid matter. The toxin may thus be formed independently
of the breaking up of the proteins on which the bacillus may be
living, though the latter no doubt has a digestive action on such
a protein as gelatin. There is, however, evidence that peptic
digestion and toxin formation are due to different vital processes
on the part of the tetanus bacillus.

Whatever the nature of the toxin is, it is undoubtedly one
of the most powerful poisons known. Even with a probably
impure toxalbumin Brieger found that the fatal dose for a
mouse was '0005 of a milligramme. If the susceptibility of
man be the same as that of a mouse, the fatal .dose for an average
adult would have been '23 of a milligramme, or about -^V^ths
of a grain. Animals differ very much in their susceptibilities
to the action of tetanus toxin. According to v. Lingelsheim, if
the minimal lethal dose per gramme weight for a horse be taken
as unity, that for the guinea-pig would be 6 times the amount,
the mouse 12, the goat 24, the dog about 500, the rabbit 1800,
the cat 6000, the goose 12,000, the pigeon 48,000, and the
hen 360,000.

A striking feature of the action of tetanus toxin is the
occurrence of a definite incubation period between the introduc-
tion of the toxin into an animal's body and the appearance of
symptoms. The incubation period varies according to the species
of animal employed, and the path of infection. In the guinea-
pi i: it is from thirteen to eighteen hours, in the horse five days,
and the incubation is shorter when the poison is introduced into

426 TETANUS

a vein than when injected subcutaneously. In man the period
between the receiving of an injury and the appearance of tetanic
symptoms is from two to fourteen days.

With regard to the action of the toxin, it has been shown to
have no effect on the sensory or motor endings of the nerves.
It acts solely as an exciter of the reflex excitability of the motor
cells in the spinal cord. The motor cells in the pons and
medulla are also affected, and to a much greater degree than
those in the cerebral cortex. When injected subcutaneoitsly
the toxin is absorbed into the nerves, and thence finds its way
to that part of the spinal cord from which these nerves spring.
This explains the fact that in some animals the tetanic spasms
appear first in the muscles of the part in which the inoculation
has taken place. This is not the case with man, in whom usually
the first symptoms appear in the neck. After subcutaneous
injection of toxin, part finds its way into the blood stream, and
if infected animals be killed during the incubation period there
is often evidence of toxin in the blood and solid organs. In the
guinea-pig there is little doubt that tetanus toxin has an affinity
solely for the nervous system. In other animals, e.(j. the
rabbit, an affinity may exist in other organs, and the fixation of
the poison in such situations may give rise to no recognisable
symptoms. In such an animal as the alligator, it is possible
that while some of its organs have an affinity for tetanus toxin
its nervous system has none. These facts are of great scientific
interest, and a possible explanation of them will be discussed in
the chapter on Immunity. If tetanus toxin be introduced into
the stomach or intestine, it is not absorbed, but to a large extent
passes through the intestine unchanged. Evidence that any
destruction takes place is wanting.

Within recent years some important light has been shed on
the mode of action of tetanus toxin. Marie and Morax studied
the path of absorption when the toxin was injected into the
muscles of the hind limb. The sciatic nerve in a rabbit was cut
near the spinal cord and toxin introduced into the muscles of the
same side ; after some hours the nerve was excised and introduced
into a mouse — the animal died of tetanus. But if the nerve were
cut near the muscles and the same procedure adopted, the mouse
did not contract the disease, though no doubt the cut nerve had
been surrounded by lymph containing toxin. If the same
experiment were performed and an excess of toxin injected into
the other limb, still only the nerve which was left in connection
with the muscle showed evidence of the presence of toxin. From
this it was deduced that the toxin was absorbed by the end-

TOXINS OF THE TETANUS BACILLUS 427

plates in the muscle and not from the lymphatics surrounding
the nerve. It was further shown that a nerve in the process of
degeneration following section did not absorb toxin after the
manner of a normal nerve. By a similar method it was shown
that the absorption by the nerve was fairly rapid, as one hour
after injection the toxin was present in it, and from other
experiments the view was put forth that the toxin was centripetal
in its flow and did not pass centrifugally in a nerve to which it
artificially gained access. Further observations have been made
on this subject by Meyer and Ransom. These observers found
evidence that toxin is only absorbed by the motor filaments of a
nerve, for while tetanus could be produced by injection into a
mixed nerve like the sciatic, the introduction of a lethal dose into
such a sensory nerve as the infra-orbital was not followed by
disease symptoms. If a small dose of toxin be injected into the
sciatic nerve, it reaches the corresponding motor cells of the cord,
and a local tetanus of the muscles supplied by the nerve results.
With a larger dose the poison passes across the commissure to
the corresponding cells of the other side, and if still further
excess is present it passes up the cord to higher centres. The
affection of such higher centres can be prevented by section of
the cord. Meyer and Hansom hold that when toxin is injected
subcutaneously or intravenously, it only acts by being absorbed
by the end-plates in muscles and thence passes to the cord, and
they consider that the incubation period is to be explained by the
time taken for this extended passage to occur. In this connection
they point out that it is in the larger animals" where the nerve
path is longest, that the incubation period is also long. Like
Marie and Morax, they believe that absorption of toxin by
its bathing the lateral aspects of uninjured nervous structures
does not occur. In support of this they bring forward the
observation that when intravenous injection is practised, the
occurrence of tetanus in a part of the body can be precipitated
by the injection of a drop of normal saline into the correspond-
ing part of the cord, — sufficient injury being thus caused to allow
the toxin in the surrounding lymph to obtain access to the
nervous elements. With regard to the action of tetanus toxin,
Meyer and Ransom believe that there is a double effect on the
nerve cells — first, an exaggeration of the normal tonus, which
accounts for the continuous stiffness of the muscles, and
secondly, an increase in reflex irritability, which is a pro-
minent factor in the recurring spasms. While no absorption
of toxin takes place by sensory filaments, they have found
evidence of atfection of the sensory apparatus in the

428 TETANUS

occurrence of what they call tetanus [dolorosus. This is a
great hypersesthesia and a paroxysmal hyperalgesia which can be
caused by injecting toxin into the spinal cord or into a sensory
root on the spinal side of the posterior root ganglion. These
symptoms are unaccompanied by motor spasms, but the animal
may die from exhaustion. The same observers have also made
interesting observations on the action of antitoxin. They found
that the injection of this substance into the course of a mixed
nerve could prevent toxin from passing up to the cord, but that if
antitoxin were injected even in great excess intravenously, and a
short time thereafter toxin were introduced into a nerve, the
death of the animal was not prevented. This they attribute
to the fact that antitoxin can only neutralise the toxin which
is still circulating in the blood. This is a very far-reaching
conclusion, as it throws doubt on what has been held to be a
possibility, namely, that toxin can be actually detached from
cells in which it is already anchored. But a still more
significant observation was made, for in one case of an animal
actively immunised against tetanus, and which contained in its
serum a considerable quantity of antitoxin, the injection of toxin
into the sciatic nerve was followed by tetanus. This would
appear to militate against Ehrlich's position that antitoxin is
manufactured in the cells which are sensitive to the toxin (see
Immunity).

Reference may here be made to the effects of injecting tetanus
toxin into the brain itself, as investigated by Roux and Borrel.
It was found th'at the ordinary type of the disease was not
produced, but what these observers called "cerebral tetanus."
This consisted of general unrest, symptoms of a psychic character
(apparent hallucinations, fear, etc.), and epileptiform convul-
sions. Death occurred in from twelve to twenty hours without
any true tetanic spasms. In this manifestation of tetanus the
incubation period was much shorter than with subcutaneous
injection, and the fatal dose was one twenty-fifth of the minimal
subcutaneous dose. Further, the injection of antitoxin forty-eight
to ninety-six hours previously did not prevent an animal from
succumbing to the intracerebral inoculation. In the light of
what has been already said, these results would seem to indicate
a special effect of the toxin when brought into direct contact
with the protoplasm of the brain cells.

We have seen that unless suitable precautions are adopted in
experiments with tetanus cultures in animals, death results not
from the multiplication of the bacilli, but from an intoxication
with toxin previously existent in the fluid in which the

IMMUNITY AGAINST TETANUS 429

bacilli have been growing. According to Vaillard, if spores
rendered toxin-free, by being kept for a sufficient time at
80° C., are injected into an animal, death does not take
place. It was found, however, that such spores can be
rendered pathogenic by injecting along with them such chemicals
as lactic acid, by injuring the seat of inoculation so as to cause
effusion of blood, by fracturing an adjacent bone, by introducing
a mechanical irritant such as soil or a splinter of wood (as in
Kitasato's experiments), or by the simultaneous injection of
other bacteria such as the staphylococcus j)t/ogenes aureus. These
facts, especially the last, throw great light on the disease as it
occurs naturally, for tetanus results especially from wounds
which have been accidentally subjected to conditions such as
those enumerated. Kitasato now holds that in the natural
infection in man, along with tetanus spores, the presence of
foreign material or of other bacteria is necessary. Spores alone
or tetanus bacilli without spores die in the tissues, and tetanus
does not result.

Immunity against Tetanus. — Antitetanic Serum. — The arti-
ficial immunisation of animals against tetanus has received much
attention. The most complete study of the question is found
in the work of Hehring and Kitasato in Germany, and of Tizzoni
and Cattani in Italy. The former observers found that such an
immunity could be conferred by the injection of very small and
progressively increasing doses of the tetanus toxin. The degree
of immunity attained, however, was not high. Subsequent
work has shown that the less rich a crude toxin is in modifica-
tions of the true toxin, the less useful it is for immunisation
procedures. In fact it is doubtful if small animals can be
immunised at all by fresh filtrates. In some cases it has been
found that the injection of non-lethal doses instead of commen-
cing an immunity actually increases the susceptibility of the
animal. This observation has recently acquired fresh interest
from its falling into line with the work on the development
of supersensitiveness to proteids, which is a very common
phenomenon (see " anaphylaxis " under Immunity). More
successful in producing immunity are the methods of accompany-
ing the early injections of crude toxin with the subcutaneous
introduction of small doses of iodine terchloride, or of using
toxin which has been acted on with iodine terchloride or with
iodine itself. Tizzoni and Cattani also used the method of
administering progressively increasing doses of living cultures
attenuated in various ways, e.y. by heat. By any of these methods
susceptible animals can be made to acquire great immunity, not

430 TETANUS

only against many times the fatal dose of tetanus toxin, but
also against injections of the living bacilli. The degree of
immunisation thus acquired remains in existence for a very long
time. Not only so, but when a high degree of immunity has
been produced by prolonged treatment, it is found that the
serum of immune animals possesses the capacity, when injected
into animals susceptible to the disease, of protecting them
against a subsequent infection with a fatal dose of tetanus
bacilli or toxin. Further, if injected subsequently to such
infection, the serum can in certain cases prevent a fatal result,
even when symptoms have begun to appear. The degree of
success attained depends, however, on the shortness of the time
which has elapsed between the infection with the bacilli or toxin
and the injection of the serum. In animals where symptoms
have fully manifested themselves only a small proportion of
cases can be saved. As with other antitoxins, there is no
evidence that the antitetanic serum has any detrimental effect
on the bacilli. It only neutralises the effects of the toxin.
The standardisation of the antitetanic serum is of the highest
importance. Behring recommends that for protecting animals
a serum should be obtained of which one gramme will protect
1,000,000 grammes weight of mice against the minimum fatal
dose of the bacillus or toxin. A mouse weighing twenty
grammes would thus require "00002 gramme of the serum to
protect it against the minimum lethal dose. In the injection
of such a serum subsequent to infection, if symptoms have
begun to appear, 1000 times this dose would be necessary; a
few hours later 10,000 times, and so on.

As the result of his experiments, Behring aimed at obtaining
a curative effect in the natural disease occurring in man. For
this purpose, as for his later laboratory experiments, he obtained
serum by the immunisation of such large animals as the horse,
the sheep, and the goat, by the injection of toxin accompanied
at first with the injection of iodine terchloride. It was found
that the greater the degree of the natural susceptibility of an
animal to tetanus, the easier was it to obtain a serum of a high
antitetanic potency. The horse was, therefore, the most
suitable animal. If now we take for granted that the relative
susceptibilities of man and the mouse towards tetanus are nearly
equal, a man weighing 100 kilogrm. would require '1 grm. of
the serum mentioned above to protect him from inoculation
with the minimum lethal dose of bacilli or toxin. If symptoms
had begun to appear, 100 c.c. at once would be necessary, and
as the injection of such a quantity might be inconvenient,

IMMUNITY AGAINST TETANUS 431

Behring recommended that for man a more powerful serum
should be obtained, namely, a serum of which one gramme would
protect 100,000,000 grammes weight of mice.1 The potency is
maintained for several months if precautions are taken to
avoid putrefaction, exposure to bright light, etc. To this end
•5 per cent, carbolic acid is usually added, and the serum is
kept in the dark. In a case of tetanus in man, 100 c.c. of
such a serum should be injected within twenty-four hours in
five doses, each at a different part of the body, and this
followed up by further injections if no improvement takes
place. Intravenous injection of the antitoxin has also been
practised, and, in cases which we have seen treated in this way,
has seemed to give better results than those obtained by the
subcutaneous method. The serum is warmed to the body
temperature and slowly introduced into a vein in the arm, the
pulse and respiration being carefully watched during the
proceeding. Ten to twenty c.c. can be injected every few hours,
and in all 100 c.c. should be given in as short a time as possible.
Henderson Smith has shown that when antitoxins to toxins of
the tetanus group are injected intravenously a high concentration
in the body fluid is maintained for some time, and the op-
portunity for neutralisation of toxin is thus great. He suggests
that both intravenous and subcutaneous injections should be
simultaneously practised. The former gives quickly the con-
centration which is desirable, and when the antitoxin injected
intravenously is beginning to be eliminated, that introduced
hypodermically comes into the circulation and the concentration
is maintained. The antitoxin has also been introduced intra-
cerebrally, very slow injection into the brain substance being
practised, but no better results have been obtained than by the
subcutaneous method.

Many cases of human tetanus have been thus treated, but
the improvement in the death-rate has not been nearly so
marked as that which has occurred in diphtheria under similar
circumstances. As in the case of diphtheria, however, the
results would probably be better if more attention were paid
to the dosage of the serum. The great difficulty is that
usually we have not the opportunity of recognising the
presence of the tetanus bacilli till they have begun to manifest
their gravest effects. In diphtheria we have a well-marked
clinical feature, — sore throat, — which draws attention to the pro-

1 The antitetauic serum sent out by the Pasteur Institute in Paris has a
strength of 1 : 1,000,000,000. Of this it is recommended that 50 to 100 c.c.
should be injected subcutaneously in one or two doses.

432 TETANUS

bable presence of the bacilli, and the curative agent can thus
be early applied. In tetanus the wound in which the bacilli
exist may be, as we have seen, of the most trifling character,
and even when a well-marked wound exists, the search for the
bacilli may be a matter of difficulty. Still it might be well,
when practicable, that every ragged, unhealthy-looking wound,
especially when contaminated with soil, should, as a matter of
routine, be examined bacteriologically. In such cases, un-
doubtedly, from time to time the tetanus bacillus would be early
detected, and treatment could be undertaken with more hope of
success than at present. However, in the existing state of
matters, whenever the first symptoms of tetanus appear, large
doses, such as those above indicated, of a serum whose strength
is known, should be at once administered. In giving a prognosis
as to the probable result, the two clinical observations on which
chief reliance ought to be placed are the presence or absence of
interference with respiration, and the rapidity with which the
groups of muscles usually affected are attacked. If dyspnoea
or irregularity in respiration or rise of temperature comes on
soon, and if group after group of muscles is quickly involved,
then the outlook is extremely grave. In addition to these
points, the duration of the incubation period is of high im-
portance in forming a prognosis. The shorter the time between
the infliction of a wound and the appearance of symptoms the
graver is the outlook.

The theory as to the nature of antitoxic action will be
discussed later in the chapter on Immunity.

Methods of Examination in a case of Tetanus. — The
routine bacteriological procedure in a case presenting the
clinical features of tetanus ought to be as follows : —

(a) Microscopic. — Though tetanus is not a disease in which
the discovery of the bacilli is easy, still microscopic examination
should be undertaken in every case. From every wound or
abrasion from which sufficient discharge can be obtained, film
preparations ought to be made and stained with any of the
ordinary combinations, e.y. carbol-fuchsin diluted with five parts
of water. Drumstick-shaped spore-bearing bacilli are to be
looked for. The presence of such, having characters corre-
sponding to those of the tetanus bacilli, though not absolutely
conclusive proof of identification, is yet sufficient for all
practical purposes. If only bacilli without spores resembling
the tetanus bacilli are seen, then the identification can only be
provisional.

The microscopic examination of wounds contaminated by soil,

MALIGNANT OEDEMA 433

etc., may, as we have said, in some cases lead to the anticipation
that tetanus will probably result.

(b) Cultivation. — The methods to be employed in isolating
the tetanus bacilli have already been described (p. 417). It
may be added, however, that if the characteristic forms are
not seen on microscopic examination of the material from the
wound, they may often be found by inoculating a deep tube of
one of the glucose media with such material, and incubating for
forty-eight hours at 37° C. At the end of this period, spore-
bearing tetanus bacilli may be detected microscopically, though
of course mixed with other organisms.

(c) Inoculation. — Mice and guinea-pigs are the most suitable
animals. Inoculation with the material from a wound should
be made subcutaneously. A loopful of the discharge introduced
at the root of the tail in a mouse will soon give rise to the
characteristic symptoms, if tetanus bacilli are present. With
suspicious organisms isolated by culture it is well to use the
splinter method (p. 423), as some strains of the b. tetani tend to
produce little toxin in artificial media, and may be injected
without causing tetanic symptoms.

MALIGNANT (EDEMA (Septicemie de Pasteur).

The organism now usually known as the bacillus of malignant
cedema is the same as that first discovered by Pasteur and
named vibrion septique. He described its characters, distin-
guishing it from the anthrax bacillus, which it somewhat
resembles morphologically, and also the lesions produced by it.
He found that it grew only in anaerobic conditions, but was
able to cultivate it merely in an impure state. It was more
fully studied by Koch, who called it the bacillus of malignant
oedema, and pointed out that the disease produced by it is not
really of the nature of a septicsjemia, as immediately after death
the blood is practically free from the bacilli.

" Malignant oedema " in the human subject is usually
described as a spreading inflammatory oedema attended with
emphysema, and ultimately followed by gangrene of the skin
and subjacent parts. In many cases of this nature the bacillus
of malignant oedema is present, associated with other organisms
which aid its spread, whilst in others it may be absent. One of
us has, however, observed a case in which the bacillus was
present in pure condition. Here there occurred intense oedema
with swelling and induration of the tissues, and the formation
of vesicles on the skin. Those changes were attended with a
28

434

MALIGNANT (EDEMA

reddish discoloration afterwards becoming livid. Emphysema
was not recognisable until the limb was incised, when it was
detected, though in small degree. Further, the tissues had a
peculiar heavy, but not putrid, odour. The bacillus, which was
obtained in pure culture, was present in enormous numbers in
the affected tissues, attended by cellular necrosis, serous
exudation, and at places much leucocytic emigration. The

FIG. 126. — Film preparation from the affected tissues in a case of
malignant oedema in the human subject, showing the spore-bearing
bacilli.

Gentian-violet, x 1000.

picture, in short, corresponded with that seen on inoculating a
guinea-pig with a pure culture. The term " malignant oedema "
should be limited in its application to cases in which the
bacillus in question is present. In most of these there is a
mixed infection ; in some the bacillus may be present alone.

This organism- has a very widespread distribution in nature,
being present in garden soil, dung, and various putrefying
animal fluids ; and it is by contamination of lacerated wounds
by such substances that the disease is usually set up in the

MICROSCOPICAL CHARACTER OF CULTURES 435

human subject. Malignant oedema can be readily produced by
inoculating susceptible animals, such as guinea-pigs, with garden
soil. The bacillus is also often present in the intestine of man
and animals, and has been described as occurring in some
gangrenous conditions originating in connection with the
intestine in the human subject.

Microscopical Characters. — The bacillus of malignant oedema
is a comparatively large organism, being slightly less than 1 /A
in thickness, that is, thinner than the anthrax bacillus. It
occurs in the form of single rods 3 to 10 /A in length, but both
in the tissues and in
cultures in fluids it fre-
quently grows out into
long filaments, which may
be uniform throughout or
segmented at irregular
intervals. In cultures on
solid media it chiefly
occurs in the form of
shorter rods with some-
what rounded ends. The
rods are motile, possessing
several laterally placed
flagella, but in a given
si>ecimen, as a rule, only
a few bacilli show active
movement. Under suit-
able conditions they form
spores, which are usually
near the centre of the rods
and have an oval shape,

their thickness somewhat exceeding that of the bacillus (Figs.
126, 127). The bacillus can be readily stained by any of the
basic aniline stains, but loses the colour in Gram's method, in
this way differing from the anthrax bacillus.

Characters of Cultures. — This organism grows readily at
ordinary temperature, but only under anaeroltic conditions. In
a puncture culture in a deep tube of glucose gelatin, the growth
appears as a whitish line giving off minute short processes, the
growth, of course, not reaching the surface of the medium.
Soon lii|iiet'a«-tion occurs and a long fluid funnel is formed, with
turbid contents and flocculent masses of growth at the bottom.
At the same time bubbles of gas are given off, which may split
up tin- gelatin. The colonies in gelatin plates under anaerobic

FIG. 127. — Bacillus of malignant oedema,
showing spores. From a culture in
glucose agar, incubated for tliree days
at 37° C.

Stained with weak rarbol-fuchsin. x 1000.

436

MALIGNANT (EDEMA

conditions appear first as small whitish points which under the
microscope show a radiating appearance at the periphery,
resembling the colonies of the bacillus subtilis. Soon, however,
liquefaction occurs around the colonies, and spheres with turbid
contents result ; gas is developed around the colonies.

In deep tubes of glucose agar at 37° C. growth is extremely

FIG. 128. — Stab cultures in agar, five days' growth at 37° C.
Natural size.

A. Tetanus bacillus. B. Bacillus of malignant oedema. C. Bacillus
of quarter-evil (Rauschbrand).

rapid. Along the line of puncture, growth appears as a some-
what broad white line with short lateral projections here and
there (Fig. 128, B). Here also gas may be formed, but this is
most marked in a shake culture, in which the medium becomes
cracked in various directions, and may be pushed upwards so
high as to displace the cotton-wool plug. The cultures possess
a peculiar heavy, though not putrid, odour.

EXPERIMENTAL INOCULATION 437

Spore formation occurs above 20° C., and is usually well seen
within forty-eight hours at 37° C. The spores have the usual
high powers of resistance, and may be kept for months in the
dried condition without being killed.

Experimental Inoculation. — A considerable number of animals
— the guinea-pig, rabbit, sheep, and goat, for example — are
susceptible to inoculation with this organism. The ox is said
to be quite immune to experimental inoculation, though it can,
under certain conditions, contract the disease by natural channels.
The guinea-pig is the animal most convenient for experimental
inoculation. When the disease is set up in the guinea-pig by
subcutaneous inoculation with garden soil, death usually occurs
in about twenty-four to forty-eight hours. There is an intense
inflammatory oedema around the site of inoculation, which
extends over the wall of the abdomen and thorax. The skin
and subcutaneous tissue are infiltrated with a reddish-brown
fluid and softened ; they contain bubbles of gas and are at places
gangrenous. The superficial muscles are also involved. These
parts have a very putrid odour. The internal organs are con-
gested, the spleen soft but not much enlarged. In such condi-
tions the bacillus of malignant oedema, both in short and long
forms, will be found in the affected tissues along with various
other organisms. Spores may be present, especially when the
examination is made some time after the death of the animal.
If the animal is examined immediately after death, a few of the
bacilli may be present in the peritoneum and pleurae, usually in
the form of long motile filaments, but they are almost invariably
absent from the blood. A short time after death, however, they
spread directly into the blood and various organs, and may then
be found in considerable numbers.

Subcutaneous inoculation with pure cultures of the bacillus of
malignant oedema produces chiefly a spreading bloody oedema,
the muscles being softened and partly necrosed ; but there is
little formation of gas, and the putrid odour is almost absent.

When the bacilli are injected into mice, however, they enter
and multiply in the blood stream, and they are found in con-
siderable numbers in the various organs, so that a condition not
unlike that of anthrax is found. The spleen also is much
swollen.

The virulence of the bacillus of malignant oedema varies con-
siderably in different cases, and it always becomes diminished in
cultures grown for some time. A smaller dose produces a
fatal result when injected along with various other organisms
(bacillus prodigiosus, etc.).

438 BACILLUS BOTtlLINUS

Immunity. — Malignant oedema was one of the first diseases
against which immunity was produced by injections of toxins.
The filtered cultures of the bacillus in sufficient doses produce
death with the same symptoms as those caused by the living
organisms, but a relatively large quantity is necessary. Chamber-
land and Koux (1887) found that if guinea-pigs were injected
with several non-fatal doses of cultures sterilised by heat or freed
from the bacilli by filtration, immunity against the living
organism could be developed in a comparatively short time.
They found that the filtered serum of animals dead of the
disease is more highly toxic, and also gives immunity when
injected in small doses. These experiments were confirmed
by Sanfelice.

Methods of Diagnosis. — In any case of supposed malignant
oedema, the fluid from the affected tissues ought first to be
examined microscopically, to ascertain the characters of the
organisms present. Though it is not possible to identify ab-
solutely the bacillus of malignant oedema without cultivating it,
the presence of spore-bearing bacilli with the characters described
above is highly suspicious (Fig. 126). In such a case the fluid
containing the bacilli should be first exposed to a temperature
of 80° C. for half an hour, and then a deep glucose agar tube
should be inoculated. In this way the spore-free organisms are
killed off. Pure cultures may be thus obtained, or this procedure
may require to be followed by the roll-tube method under
anaerobic conditions. An inoculation experiment, if available,
may also be made on a guinea-pig.

BACILLUS BOTULINUS.

The term " meat-poisoning " embraces a number of conditions
produced by different agents, and the relation of the bacillus of
Gaertner to one class of case has already been discussed. Another
group was shown by van Ermengem in 1896 to be caused by an
anaerobic bacillus to which he gave the name bacillus botulinus.
He cultivated the organism from a sample of ham, the inges-
tion of which in the raw condition had produced a number of
cases of poisoning, some of them followed by fatal result. The
symptoms in these cases closely corresponded with those occur-
ring in the so-called " sausage poisoning " met with from time to
time in Germany and other countries where sausages and ham
are eaten in an imperfectly cooked condition. Such cases form
a fairly well-defined group, the symptoms in which are chiefly
referable to an action on the medulla, and, as will be detailed

MICROSCOPICAL AND CULTURAL CHARACTERS 439

below, similar symptoms have been experimentally produced
by means of the bacillus mentioned or its toxins. The chief
symptoms of this variety of botulismus, as detailed by van
Krmengem, are disordered secretion in the mouth and nose, more
or less marked ophthalmoplegia, externa and interna (dilated
pupil, ptosis, etc.), dysphagia, and sometimes aphagia with
aphonia, marked constipation and retention of urine, and in
fatal cases interference with the cardiac and respiratory centres.
Along with these there is practically no fever and no interference
with the intellectual faculties. The symptoms commence at
earliest twelve to twenty-four hours after ingestion of the poison.
From the ham in question, which was not decomposed in the
ordinary sense, van Ermengem obtained numerous colonies of
this bacillus, the leading characters of which are given below.
It may be added that Romer obtained practically the same
results as van Ermengem in a similar condition, and that the
bacillus botulinus has been cultivated by Kempner from the
intestine of the pig.

Microscopical and Cultural Characters. — The organism is a
bacillus of considerable size, measuring 4 to 9 /x in length and
•9 to 1 '2 /x in thickness ; it has somewhat rounded ends and
sometimes is seen in a spindle form. It is often arranged in
pairs, sometimes in short threads. . Under certain conditions
it forms spores which are oval in shape, usually terminal in
position, and a little thicker than the bacilli. It is a motile
organism and has 4 to 8 lateral flagella of wavy form. It
stains readily with the ordinary dyes, and also retains the
colour in Gram's method, though care must be employed in
decolorising.

The organisms can be readily cultivated on the ordinary
media, but only under strictly anaerobic conditions. In glucose
gelatin a whitish line of growth forms with lateral offshoots,
but liquefaction with abundant gas formation soon occurs. In
gelatin plates the colonies after four to six days are somewhat
characteristic ; they appear to the naked eye as small semi-
transparent spheres, and these on examination under a low
power of the microscope have a yellowish-brown colour and are
seen to be composed of granules which show a streaming move-
ment, especially at the periphery. Cultures in glucose agar
resemble those of certain other anaerobes; there is abundant
development of gas, and the medium is split up in various
directions. The cultures have a rancid, though not foul, odour,
due chiefly to the development of butyric acid. The optimum
temperature is below that of the body, namely, between 20° and

440 BACILLUS BOTULINUS

30° C. ; at the body temperature growth is slower and less
abundant and spore formation does not occur.

Pathogenic Effects. — Like the tetanus bacillus, the bacillus
botulinus has little power of nourishing in the tissues, whereas it
produces a very powerful toxin. Van Ermengem. found that the
characteristic symptoms could be produced in certain animals
by administering watery extracts of the infected ham or cultures
either by the alimentary canal or by subcutaneous injection.
Here also there is a period of incubation of not less than six to
twelve hours before the symptoms appear, and when the dose is
small a somewhat chronic condition may result, in which local
paralyses form a striking feature. The characteristic effects
can also be produced by means of the filtered toxin by either of
the methods mentioned, though in the case of administration by
the alimentary canal the dose requires to be larger. Here also,
as in the case of the tetanus poison, the potency of the toxin is
remarkable, the fatal dose for a guinea-pig of 250 grm. weight
being in some instances '0005 c.c. of the filtered toxin. In cases
of poisoning in the human subject, the effects would accordingly
appear to be produced by absorption of the toxin from the
alimentary canal ; it is only after or immediately before death
that a few bacilli may enter the tissues. Van Ermengem
obtained a few colonies from the spleen of a patient who had
died from ham-poisoning. The properties of the botulinus toxin
have been investigated, and have been found to correspond
closely, as regards relative instability, conditions of precipitation,
combination with sensitive cells (i.e. of brain and cord), etc.,
with the toxins of diphtheria and tetanus. An antitoxin has
also been prepared by Kempner by the usual methods, and has
been shown not only to have a neutralising property, but to
have considerable therapeutical value when administered some
hours after the toxin. The subject has been recently studied by
Leuchs, and he has found that the combination toxin-antitoxin
can be split up by the action of acids and the two components
recovered, just as Morgenroth showed to occur in the case of
diphtheria (p. 529). The direct combining affinity of the toxin
for the central nervous system has been demonstrated by Kempner
and Schepilewsky by the same methods as Wassermann and
Takaki employed in the case of the tetanus toxin. The condition
of the nerve cells in experimental poisoning with the botulinus
toxin has been investigated independently by Marinesco and by
Kempner and Pollack, and these observers agree as to the
occurrence of marked degenerative changes, especially in the
motor cells in tfre spinal cord and medulla, Marinesco also

QUARTER-EVIL 441

observed hypertrophy and proliferation of the neuroglia cells
around them.

These observations, therefore, show that in one variety of
meat-poisoning the symptoms are produced by the absorption of
the toxins of the bacillus botulinus from the alimentary canal,
and, as van Ermengem points out, it is of special importance to
note that the meat may be extensively contaminated with this
bacillus, and may contain relatively large quantities of its toxins
without the ordinary signs of decomposition being present.
The production of an extracellular toxin by this organism, wdth
extremely potent action on the nervous system, is a fact of great
scientific interest, and has a bearing on the etiology of other
obscure nervous affections.

QUARTER-EVIL (GERMAN, RAUSCHBRAND ; FRENCH, CHARBON
SYMPTOMATIQUE).

The characters of the bacillus need be only briefly described, as, so far
as is known, it never infects the human subject. The natural disease,
which occurs especially in certain localities, affects chiefly sheep, cattle,
and goats. Infection takes place by some wound of the surface, and
there spreads in the region around, inflammatory swelling attended by
bloody u'denia and emphysema of the tissues. The part becomes greatly
swollen, and of a dark, almost black, colour. Hence the name "black-
leg" by which the disease is sometimes known. The bacillus which
produces this condition is present in large numbers in the affected tissues,
associated with other organisms, and also occurs in small numbers in the
blood of internal organs. For the isolation of the bacillus, Grassberger
and Schattenfroh recommend the use of anaerobic sugar agar plates con-
taining pieces of sterile ox flesh.

The bacillus morphologically closely resembles that of malignant
fiidema. Like the latter, also, it is a strict anaerobe, and its conditions
of growth as regards temperature are also similar. It is, however, some-
what thicker, and does not usually form such long filaments. Moreover,
the spores, which are of oval shape and broader than the bacillus, are
almost invariably situated close to one extremity, though not actually
terminal (Fig. 129). The characters of the cultures, also, resemble
those of the bacillus of malignant cedema, but in a stab culture in
glucose agar there are more numerous and longer lateral offshoots, the
growth being also more luxuriant (Fig. 128, c). This bacillus is actively
motile, and possesses numerous lateral flagella. When cultures derived
from disease conditions are continuously subcultured on sugar media, they
tend to lose their capacities of motility and spore formation. The
organism seems to occupy a position somewhat intermediate between the
b. saccharobutyricus (v. Klecki), which is a free sugar fermenter, and the
b. putrificus (Bienstock), which*has great powers of splitting up albumins.

The disease can be readily produced in various animals, e.g. guinea-
pigs, by inoculation with the affected tissues of diseased animals, and
also by means of pure cultures, though an intramuscular injection of a
considerable amount of the latter is sometimes necessary. The condition
product '1 in this way closely resembles that in malignant oedema, though

442

QUARTER-EVIL

there is said to be more formation of gas in the tissues. Rabbits are
more resistant to this disease, whilst they are comparatively suscep-
tible to malignant osdema. As in the case of tetanus, inoculation with
living spores which have been deprived of adherent toxin by heat does
not produce the disease. A toxin can be separated by nitration from
cultures of bouillon containing 5 per cent, glucose and a thick emulsion of
sterile calcium carbonate. It is fairly resistant to heat, withstanding
two hours at 70-75° C. without being destroyed, and it is also very rapid
in its action, being capable in appropriate dose of killing a horse in five
minutes. It is to be noted as an important fact, that while fresHy
isolated cultures possess a high degree of virulence they may have little

capacity for in vitro toxin
production. Grassberger and
Schattenfroh state that there
may be an antagonism be-
tween maximum virulence
and maximum toxin produc-
tion. One of the properties
of the toxin is said to be a
capacity for killing leuco-
cytes.

The disease is one against
which immunity can be pro-
duced in various ways, and
methods of preventive inocu-
lation have been adopted in
the case of animals liable to
suffer from it. This subject
was specially worked out
by Arloing, Cornevin, and

Thomas, and later by others.
FIG. 129.— Bacillus ot quarter-evil, showing

s^^r^ u<r -•

Stained with weak carbol-fuchsiu. xlOOO. the. intravenous and intra-

peritoneal routes) with a non-
fatal dose of the virus (i.e.

the cedematous fluid found in the tissues of affected animals and which
contains the bacilli), or by injection with larger quantities of the virus
attenuated by heat, drying, etc. It can be produced also by cultures
attenuated by heat and by the products of the bacilli obtained by
nitration of cultures. An antitoxin has been produced against the toxins
of the bacillus, and a method of protection in which the action of this
antitoxin is combined with that of the virus has been used (cf. Anthrax,
p. 348). The antitoxin is said to increase the chemiotactic properties of
the leucocytes.

BACILLUS AEROGENES CAPSULATUS.

This bacillus, though sometimes aiding in the production of patho-
•logical changes, is chiefly of interest on account of the extensive gaseous
development to which it gives rise in the tissues post mortem. It was
described by Welch and Nuttall in 1892 ; it is now recognised as being
identical with an organism found in gaseous phlegmon by E. Fraenkel,
and called by him the bacillus phlegmones emphysematosce. The organism
is a comparatively large one, measuring 3 to 6 /x in length and having a

BACILLUS AEROGENES CAPSULATUS

443

thickness about the same as that of the anthrax bacillus ; its ends are
square or slightly rounded (Fig. 130). It often occurs in pairs, sometimes
in chains ; occasionally filamentous forms are met with. It usually shows
a well-marked capsule, hence the name ; it is non-motile and does not
form spores. It stains readily with the basic aniline dyes and retains
the stain in Gram's method. It grows readily on the ordinary media,
but only under anaerobic conditions ; the optimum temperature is that
of the body, growth at the room temperature being comparatively slow.
In a puncture culture in agar there is an abundant whitish line of growth,
wiih somewhat indented margin ; the individual colonies are white and
of rounded or oval form. There is practically no liquefaction of gelatin,
though this medium becomes somewhat softened around the growth.
In all cases there is a tendency to abundant evolution of gas in the cul-
tures, and this is especially
marked when fermentable
sugars are present.

The organism appears to
be the most frequent cause
of rapid gaseous develop-
ment in the blood and
organs post-mortem, this
depending upon an invasion
of the blood immediately
before death. In sucn
cases, even within twenty -
four hours under ordinary
conditions, large bubbles of
gas may be present in the
veins, and the organs may
be beset with gas-containing
spheres of various sizes ; the
liver is usually the organ
most affected, and its ap-
pearance has been compared
n tliftf nf rrnv&rp r-lfpp
The iuvasion^bTthis ol^
ism is met with from time
to time in puerperal cases,

an 1 also in connection with ulcerative or gangrenous conditions of the
intestine ; the bacillus is also found not infrequently in the peritoneum
in cases of perforation. Although the striking changes in the organs
are due to a post-mortem development of the bacillus, there is no
doubt that its entrance into the blood stream often hastens death,
and may in some instances be the cause of it. As already stated, the
organism is also met with in some cases of spreading oedema with
emphysema as a leading feature.

When tested experimen tally, the bacillus by itself is found to have
little pathogenic action. Injection of pure cultures in rabbits and
guinea-pigs may be followed by little result, but sometimes in the latter
animals "gaseous phlegmon" is produced, without suppuration unless
other organisms are present. If a small quantity of culture be injected
intravenously, c.y. hi a rabbit, and then the animal be killed, bubbles
of gas are rapidly produced in the blood and organs, the picture
corresponding with that in the human cases,

Fl£: 130.-Barillus aerogenes eapsulatus ;
nlm preparation from bone-marrow in a

444 FUSIFORM ANAEROBIC BACILLI

FUSIFORM ANAEROBIC BACILLI PATHOGENIC TO MAN.

Babes in 1884 described organisms of this type in a
diphtheria-like affection of the fauces, and since that time the
presence of similar organisms has been noted in necrotic inflam-
mations, ulcerative stomatitis, noma, and like affections. The
association of fusiform bacilli with a form of angina has been
specially recognised since the work of Vincent (1898-99); and
this condition often goes now under the name of "Vincent's
angina." He recognised two forms of the affection — (a) a
diphtheroid type characterised by the formation of a firm
yellowish-white false membrane, very like that of diphtheria,
associated with only superficial ulceration ; and (b) an ulcerative
type where the membrane is soft, greyish, and foul-smelling,
attended with ulceration and surrounding oedema. In the
former type fusiform bacilli are present alone; in the latter,
which is distinctly the commoner, there are also spirochsetes.
The fusiform bacilli are thin rods measuring on the average
10 to 14 /x in length, and less than 1 /x in thickness ; they are
straight or slightly curved and are tapered at their extremities.
The central portion often stains less deeply than the extremities,
and not infrequently shows unstained points (Plate I., Fig. 4).
The organisms are non-motile. They stain fairly deeply with
Loffler's methylene-blue or with weak carbol-fuchsin. They
lose the stain in Gram's method. The spirochsetes are long
delicate organisms showing several irregular curves, and are
motile ; in appearance they resemble the spirochsete refringens
and similar organisms found in gangrenous conditions. They
stain less deeply than the bacilli. Sometimes they are numerous,
sometimes scanty ; they seem to be similar to spirochsetse found
in the mouth in a variety of other conditions. In a section
through the false membrane, when stained with methylene or
thionin blue, there is usually to be seen a darkly stained band,
a short distance below the surface, which is due to the presence
of large masses of the fusiform bacilli closely packed together ;
neither they nor the spirochsetes appear to pass deeply into the
tissues. Vincent's results have been confirmed by others, and
there is no doubt that fusiform bacilli, of which there are
probably several species, are associated with various spreading
necrotic conditions. Cultures of fusiform bacilli have been
obtained by E Hermann and by Weaver and Tunnicliffe. They
grow only under anaerobic conditions, and the best media are
those consisting of a mixture of serum or blood and agar (1 : 3).
The organisms form small rounded colonies of whitish or

FUSIFORM ANAEROBIC BACILLI 445

yellowish colour, somewhat like those of a streptococcus, but
rather felted in appearance on the surface. Injections of pure
cultures in animals sometimes produce suppuration but never
necrosis (Ellermann). Tunnicliffe finds that the spirochaetes
are only stages in the development of fusiform bacilli, as cultures
which at an early stage show only fusiform bacilli afterwards
contain spirochsetes, and intermediate forms can be found. These
results, however, have not yet been confirmed. It is also to be
noted that fusiform bacilli are sometimes present in the secretions
of the mouth in normal conditions, and may occur in increased
numbers in true diphtheria.

CHAPTER XVIII.

THE CHOLERA SPIRILLUM AND ALLIED
ORGANISMS.

Introductory. — It is no exaggeration of the facts to say that
previously to 1883 practically nothing of value was known
regarding the nature of the virus of cholera. In that year
Koch was sent to Egypt, where the disease had broken out, in
charge of a Commission for the purpose of investigating its
nature. In the course of his researches he discovered the
organism now generally known as the " comma bacillus " or
the "cholera spirillum." He also obtained pure cultures of
the organism from a large number of cases of cholera, and
described their characters. The results of his researches were
given at the first Cholera Conference at Berlin in 1884.

Since Koch's discovery, and especially during the epidemic in
Europe in 1892-93, spirilla have been cultivated from cases of
cholera in a great many different localities, and though this
extensive investigation has revealed the invariable presence in
true cholera of organisms resembling more or less closely Koch's
spirillum, certain difficulties have arisen. For it has been
found that the cultures obtained from different places have
shown considerable variations in their characters, and, further,
spirilla which closely resemble Koch's cholera spirillum have
been cultivated from sources other than cases of true cholera.
There has therefore been much controversy, on the one hand,
as to the signification of these variations, — whether they are
to be regarded as indicating distinct species or merely varieties
of the same species, — and, on the other hand, as to the means
of distinguishing the cholera spirillum from other species which
resemble it. These questions will be discussed belo\v.

In considering the bacteriology of cholera, it is to be borne in
mind that in this disease, in addition to the evidence of great
intestinal irritation, accompanied by profuse watery discharge,
and often by vomiting, there are also symptoms of general

446

MICROSCOPICAL CHARACTERS 447

systemic disturbance which cannot be accounted for merely by
the withdrawal of water and certain substances from the
system. Such symptoms include the profound general prostra-
tion, cramps in the muscles, extreme cardiac depression, the
cold and clammy condition of the surface, the subnormal
temperature, suppression of urine, etc. These, taken in their
entirety, are indications of a general poisoning in which the
circulatory and thermo - regulatory mechanisms are specially
involved. In some, though rare, cases known as cholera sicca,
general collapse occurs with remarkable suddenness, and is
rapidly followed by a
fatal result, whilst there
is little or no evacuation
from the bowel, though
post-mortem the intestine
is distended with fluid
contents. As the char-
acteristic organisms in
cholera are found only
in the intestine, the
general disturbances are
to be regarded as the
result of toxic substances
absorbed from the bowel.
It is also to be noted
that cholera is a disease

of which the onset and Fl(J. 131. -Cholera spirilla, from a culture on
course are much more agar of twenty-four hours' growth,

rapid than is the case ill Stained with weak carbol-fuchsin. xlOOO.
most infective diseases,

such as typhoid and diphtheria ; and also that recovery, when
it takes place, does so more quickly. The two factors to be
correlated to these facts are : (a) a rapid multiplication of
organisms, (6) the production of rapidly acting toxins.

The Cholera Spirillum. — Microscopical Characters. — The
cholera spirilla, as found in the intestines in cholera, are small
organisms measuring about 1 *5 to 2 /A in length, and rather less
than '5 in thickness. They are distinctly curved in one direction,
hence the api>earance of a comma (Fig. 131); most occur
singly, but some are attached in pairs and curved in opposite
directions, so that an S-shai>e results. Longer forms are rarely
seen in the intestine, but in cultures in fluids, as is especially
well seen in hanging-drop preparations, they may grow into
longer spiral filaments, showing a large number of turns. In

448

CHOLERA

film preparations made from the intestinal contents in typical

cases, these organisms are

< " * present in enormous

!.<?X numbers in almost pure

VL culture, most of the

^ , spirilla lying with their

-** C. long axes in the same

direction, so as to give
the appearance which
Koch compared to a num-
ber of fish in a stream.

They possess very active
motility, which is most
marked in the single
forms. When stained by
the suitable methods they
are seen to be flagellated.
Usually a single terminal
flagellum is present at
one end only (Fig. 132).
It is very delicate, and
length of the organism.

FIG. 132. — Cholera spirilla stained to show
the terminal flagella. See also Plate
IV., Fig 19. xlOOO.

the

measures four or five times
Cholera spirilla do not
form spores. In old cul-
tures the organisms may
present great variety in
size and shape. Some
are irregularly twisted
filaments, sometimes glo- , f
bose, sometimes clubbed , 4j
at their extremities, and *
also showing irregular
swellings along their
course ; others are short
and thick, and may
have the appearance of
large cocci, often stain-
ing faintly. All these
changes in appearance
are to be classed together
as involution forms.

Staining. — Cholera
spirilla stain readily with
the usual basic aniline stains, though Loffler's methylene-blue or

FIG. 133. — Cholera spirilla from an old agar
culture, showing irregularities in size and
shape, with numerous faintly - stained
coccoid bodies — involution forms.
Stained with fuchsin. x 1000.

CULTIVATION 449

weak carbol-fuchsin is specially suitable. They lose the stain
in Gram's method.

Distribution within the Body. — The chief fact in this con-
nection is that the spirilla are confined to the intestine, and are
not present in the blood or internal organs. In cases in which
there is the characteristic " rice-water " fluid in the intestines,
they occur in enormous numbers — almost in pure culture. The
lower half of the small intestine is the part most affected. Its
surface epithelium becomes shed in great part, and the flakes
floating in the fluid consist chiefly of masses of epithelial cells
and mucus, amongst which are numerous spirilla. The spirilla
also penetrate the follicles of Lieberkiilm, and may be seen
lying between the basement membrane and the epithelial lining,
which becomes loosened by their action. They are, however,
rarely found in the connective tissue beneath, and never pene-
trate deeply. Along with these changes there is congestion of
the mucosa, especially around the Peyer's patches and solitary
glands, which are somewhat swollen and prominent. In some
very acute cases the mucosa may show general acute congestion,
with a rosy pink colour but very little desquamation of
epithelium, the intestinal contents being a comparatively clear
fluid containing the spirilla in large numbers. In other cases of
a more chronic type, the intestine may show more extensive
necrosis of the mucosa and a considerable amount of haemorrhage
into its substance, along with formation of false membrane at
places. The intestinal contents in such cases are blood-stained
and foul-smelling, there being a great proportion of other
organisms present besides the cholera spirilla (Koch).

Cultivation. — (For methods, see p. 459).

The cholera spirillum grows readily on all the ordinary media,
and, with the exception of that on potato, growth takes place at
the ordinary room temperature. The most suitable temperature,
however, is that of the body, and growth usually stops about
16° C., though in some cases it has been obtained at a lower
temperature. Abundant growth occurs on media with suffi-
ciently alkaline reaction to inhibit the growth of many
intestinal bacteria, e.g. Dieudonne"'s medium, p. 44.

Peptone Gelatin. — On this medium the organism grows well
and produces liquefaction. In puncture cultivations at 22° C.
a whitish line appears along the needle track, at the upper part
of which liquefaction commences, and as evaporation quickly
occurs, a small bell-sliaprd <lcpression forms, which gives the
appearance of an air-bubble. On the fourth or fifth day we get
the following appearance: there is at the surface the bubble-
29

450 CHOLERA

shaped depression ; below this there is a funnel-shaped area of
liquefaction, the fluid being only slightly turbid, but showing at
its lower end thick masses of growth of a more or less spiral
shape (Fig. 134). The liquefied portion gradually tapers off
downwards towards the needle track. (This appearance is,
however, in some varieties not produced till much later, especi-
ally when the gelatin is very stiff, and, in
other varieties which liquefy very slowly,
may not be met with at all.) At a later
stage, liquefaction spreads and may reach
the side of the tube.

In gelatin plates the colonies are some-
what characteristic. They appear as minute
whitish points, visible in twenty-four to
forty-eight hours, the surface of which, under
a low power of the miscroscope, is irregularly
granular or furrowed (Fig. 135, A). Lique-
faction occurs, and the colony sinks into
the small cup formed, the plate then show-
ing small sharply-marked rings around the
colonies. Under the microscope the outer
margin of the cup is circular and sharply
marked. Within the cup the liquefied
portion forms a ring which has a more or
less granular appearance, whilst the mass of
growth in the centre is irregular and often
broken up at its margins (Fig. 135, B).

On the surface of the agar media a thin,
almost transparent, layer forms, which pre-
sents no special characters. On solidified
blood serum the growth has at first the same
appearance, but afterwards liquefaction of

tlle

spirillum in peptone superficial colonies under a low power are

gelatin — six days' circular discs of brownish-yellow colour, and

growth. Natural size. more transparent than those of most other

organisms. On potato at the ordinary

temperature, growth does not take place, but on incubation at

a temperature of from 30° to 37° C., a moist layer appears,

which assumes a dirty brown colour, somewhat like that of the

glanders bacillus ; the appearance, however, varies somewhat in

different varieties, and also on different sorts of potatoes.

In bouillon with alkaline reaction the organism grows very
readily, there occurring in twelve hours at 37° C. a general

CULTIVATION 451

turbidity, while the surface shows a thin pellicle composed of
spirilla in a very actively motile condition. Growth takes place
under the same conditions equally rapidly in peptone solution
(1 per cent, with *5 per cent, sodium chloride added).

In milk also the organism grows well, and produces no
coagulation nor any change in its apj>earance, at least for
several days.

On all the media the growth of the cholera spirillum is a
relatively rapid one, and especially is this the case in peptone
solution and in bouillon, a circumstance of importance in relation
tn its separation in cases of cholera (vide p. 459).

The cholera organism is one which grows much more rapidly
in the presence of oxygen than in anaerobic conditions ; in the
complete exclusion of oxygen very little growth occurs.

Kic. !#>.— Colonies of the cholera spirillum in a gelatin plate —
three days' growth. A shows the granular surface, liquefaction just
i ommencing ; in B liquefaction is well narked.

I l!"l H'-'i-t !<,n. — This is one of the most important tests
in the diagnosis of the cholera organism. It is always given by
a true cholera spirillum, and though the reaction is not peculiar
to it, the number of organisms which give the reaction under
the conditions mentioned are comparatively few. The test is
made by adding a few drops of pure sulphuric acid to a culture
in bouillon or in peptone solution (1 per cent.) which has been
incubated for twenty-four hours at 37° C. ; in the case of the
cholera spirillum a reddish-pink colour is produced. This is due
to the tact that both indol and a nitrite are formed by the
spirillum in the medium, and hence in applying the test for
indol tin- addition of a nitrite is not necessarily the red colour.
Here, as in testing for the production of indol by other bacteria.
it is found that not every specimen of peptone is suitable, and
it is advisable to select a peptone which gives the characteristic

452 CHOLERA

reaction with a known cholera organism, and to use it for
further tests. It is also essential that the sulphuric acid should
be pure, for if traces of nitrites are present the reaction may
be given by an organism which has not the power of forming
nitrites.

Htetnolytic Test. — This method, introduced by Kraus, is per-
formed by means of agar plates, a small quantity of sterile
defibrinated blood being added to the agar and thoroughly
diff used ; if any organism has haemolytic properties, a clear zone
or areola forms around each colony by the diffusion of haemo-
globin. In no instance has an undoubted cholera organism
been found to produce haemolysis, whereas many species of
spirilla closely resembling it possess marked hiemolytic action.
This test may accordingly be applied along with the others in
determining the identity of a supposed cholera organism.

Powers of Resistance. — In their resistance against heat,
cholera spirilla correspond with most spore-free organisms, and
are killed in an hour by a temperature of 55° C., and much
more rapidly at higher temperatures. They have comparatively
high powers of resistance against great cold, and have been
found alive after being exposed for several hours to the tempera-
ture of — 10° G. They are, however, killed by being kept in ice
for a few days. Against the ordinary antiseptics they have
comparatively low powers of resistance, and Pfuhl found that
the addition of lime, in the proportion of 1 per cent., to water
containing the cholera organisms was sufficient to kill them in
the course of an hour.

As regards the powers of resistance in ordinary conditions,
the following facts may be stated : In cholera stools kept at the
ordinary room temperature, the cholera organisms are rapidly
outgrown by putrefactive bacteria, but in exceptional cases they
have been found alive even after two or three months. In most
experiments, however, attempts to cultivate them even after a
much shorter time have failed. The general conclusion may be
drawn from the work of various observers, that the spirilla do
not multiply freely in ordinary sewage water, although they may
remain alive for a considerable period of time. On moist linen,
as Koch showed, they can nourish very rapidly. Though we
can state generally that the conditions favourable for the growth
of the cholera spirillum are a warm temperature, moisture, a
good supply of oxygen, and a considerable proportion of organic
material, we do not know the exact circumstances under which
it can nourish for an indefinite period of time as a saprophyte.
The fact that the area in which cholera is an endemic disease is

EXPERIMENTAL INOCULATION 453

so restricted, tends to show that the conditions for a prolonged
UTO \\tli nf tin- spirillum outside the body are not usually supplied.
Yrt. on tin- other hand, there is no doubt that in ordinary
conditions it can live a sufficient time outside the body and
multiply t<> a sufficient extent to explain all the facts known
with regard to the persistence and spread of cholera epidemics.
During an epidemic at St. Petersburg the cholera organism
was cultivated from the stools of a considerable number of
people suffering^ f rom slight intestinal disturbance, and even
(piite healthy individuals. The latter may be regarded as
"cholera-carriers," but the organisms were obtained from them
over a comparatively short period of time, and it is unknown to
what extent they maintain and spread the infection.

Cholera organisms are, as a rule, rapidly killed by being
thoroughly dried, and it is inferred from this that they cannot
be carried in the living condition for any great distance through
the air, a conclusion which is well supported by observations
on the spread of the disease. Cholera is practically always
transmitted by means of water or food contaminated by the
organism, and there is no doubt that contamination of the
water supply by choleraic discharges is the chief means by which
areas of population are rapidly infected. It has been shown
that if Hies are fed on material containing cholera organisms,
the organisms may be found alive within their bodies twenty-
four hours afterwards. And further, Haffkine found that
sterilised milk might become contaminated with cholera
organisms if kept in open jars to which flies had free access,
in a locality infected by cholera. It is quite possible that
infection may be carried by this agency in some cases.

Experimental Inoculation. — In considering the effects of
inoculation with the cholera organism, we are met with the
difficulty that none of the lower animals, so far as is known,
suffer from the disease under natural conditions. And, further,
attempts to produce the disease by feeding with cholera dejecta,
as well as with cultures, have been unsuccessful. As the
organisms are confined to the alimentary tract in the natural
disease, attempts to induce their multiplication within the
intestine of animals by artificially arranging favouring con-
ditions, have occupied a prominent place in the experimental
work. \Ve shall give a short account of such experiments : —

Xikati and Rietsch were the first to inject the organisms directly into
tin duodenum of dogs and rabbits, and they succeeded in producing, in
a considerable proportion of the animals, a choleraic condition of the
inti'.stine. These experiments were confirmed by other observers, in-

454 CHOLERA

eluding Koch. Thinking that probably the spirillum, when introduced
by the mouth, is destroyed by the action of the hydrochloric acid of
the gastric secretion, Koch first neutralised this acidity by administering
to guinea-pigs 5 c.c. of a 5 per cent, solution of carbonate of soda, and
some time afterwards introduced a pure culture into the stomach by
means of a tube. As this method failed to give positive results, he
tried the effect of artificially interfering with the intestinal peristalsis
by injecting tincture of opium into the peritoneum (1 c.c. per 200 grm.
Aveight), in addition to neutralising as before with the carbonate of
sodium solution. The result was remarkable, as thirty out of thirty-
five animals treated died with symptoms of general prostration and
collapse. Death occurs after a few hours. Post mortem the small
intestine is distended, its mucous membrane congested, and it contains
a colourless fluid with small flocculi and the cholera organisms in
practically pure cultures. Koch, however, found that when the spirilla
of Finkler and Prior, of Deneke, and of Miller (vide infra), were
employed by the same method, a certain, though much smaller,
proportion of the animals died from an intestinal infection. Though
the changes in these cases were not so characteristic, they were sufficient
to prevent the results obtained with the cholera organism from being
used as a demonstration of the specific relation of the latter to the
disease.

Within later years some additional facts of high interest have been
established with regard to choleraic infection of animals. For example,
Sabolotny found that in the marmot an intestinal infection readily takes
place by simple feeding with the organism, there resulting the usual
intestinal changes, sometimes with hjemorrhagic peritonitis — the
organisms, however, being present also in the blood. And of special
interest is the fact, discovered by Metchnikoff, that in the case of
young rabbits shortly after birth a large proportion die of choleraic
infection when the organisms are simply introduced along with the
milk, as may be done by infecting the teats of the mother. Further,
from these animals thus infected the disease may be transmitted to
others by a natural mode of infection. In this affection of young rabbits
many of the symptoms of cholera are present. Many of these experi-
ments were performed with the vibrio of Massowah, which is now
admitted not to be a true cholera organism, others with a cholera
vibrio obtained from the water of the Seine.

It will be seen from the above account that the evidence
obtained from experiments on intestinal infection of animals,
though by no means sufficient to establish the specific relation-
ship of the cholera organism, is on the whole favourable to this
view, especially when it is borne in mind that animals do not in
natural conditions suffer from the disease.

Experiments performed by direct inoculation also supply
interesting facts. Intraperitoneal injection in guinea-pigs is
followed by general symptoms of illness, the most prominent
being distension of the abdomen, subnormal temperature, and,
ultimately, profound collapse. There is peritoneal effusion,
which may be comparatively clear, or may be somewhat turbid

EXPERIMENTS ON THE HUMAN SUBJECT 455

and contain flakes of lymph, according to the stage at which
death takes place. If the dose is large, organisms are found
in considerable numbers in the blood and also in the small
intestine, but with smaller doses they are practically confined to
the peritoneum. Kolle found that when the minimum lethal
dose was used in guinea-pigs, the peritoneum might be free from
living organisms at the time of death, the fatal result having
taken place from an intoxication (cf. Diphtheria, p. 405). These
and other experiments show that though the organisms undergo
a certain amount of multiplication when introduced by the
channels mentioned, still the tendency to invade the tissues is
not a marked one. On the other hand, the symptoms of general
intoxication are always pronounced.

Experiments on the Human Subject. — Experiments have also
been performed -in the case of the human subject, both intention-
ally and accidentally In the course of Koch's earlier work, one
of the workers in his laboratory shortly after leaving was seized
with severe choleraic symptoms. The stools were found to
contain cholera spirilla in enormous numbers. Recovery, how-
ever, took place. In this case there was no other possible
source of infection than the cultures with which the man had
been working, as no cholera was present in Germany at the time.
AVithin recent years a considerable number of experiments have
been performed on the human subject, which certainly show that
in some cases more or less severe choleraic symptoms may follow
ingestion of pure cultures, whilst in others no effects 'may result.
The former was the case, for example, with Emmerich and
Pettenkofer, who made experiments on themselves, the former
especially becoming seriously ill. In the case of both, diarrhoea
was well marked, and numerous cholera spirilla were present in
the stools, though toxic symptoms were proportionately little
pronounced. Metchnikoff also, by experiments on himself and
others, obtained results which convinced him of the specific
relation of the cholera spirillum to the disease. Lastly, we may
mention the case of Dr. Orgel in Hamburg, who contracted the
disease in the course of experiments with the cholera and other
spirilla, and died in spite of treatment. It is believed that in
sucking up some peritoneal fluid containing cholera spirilla, a
little entered his mouth and thus infection was produced. This
took place in September 1894, at a time when there was no
cholera in Germany. On the other hand, in many cases the
experimental ingestion of cholera spirilla by the human subject
has given negative results. Still, as the result of observation of
what takes place in a cholera epidemic and of what has been

456 CHOLERA

established with, regard to cholera carriers, we may consider
that only a certain proportion of people are susceptible to
cholera, and the facts just mentioned are, in our opinion, of the
greatest importance in establishing the relation of the organism
to the disease.

Toxins. — The general statement may be made that filtered
cholera cultures as a rule have little toxic action ; that is, com-
paratively little extracellular toxin is produced by the organism.
It was, however, shown by R. Pfeiffer that the dead spirilla were
highly toxic, and that, in fact, they produced, on injection into
guinea-pigs, the same phenomena as living cultures, profound
collapse with subnormal temperature being a prominent feature.
Pfeiffer considers that the toxic substances are contained in the
bodies of the organisms — that is, they are endotoxins, — and
that they are only set free by the disintegration of the
latter. He showed also that when an animal is inoculated
intraperitoneally with the cholera organism, and then some
time later anti-cholera serum which produces bacteriolysis is
injected, rapid collapse with a fatal result may ensue, apparently
due to the liberation of the intracellular toxins. The dead
cultures administered by the mouth produce no effect unless
the intestinal epithelium is injured, in which case poisoning
may result. He considers that the desquamation of the
epithelium is an essential factor in the production of the
phenomena of the disease in the human subject. Pfeiffer found
that the toxic bodies were to a great extent destroyed at 60° C.,
but even after heating at 100° C. a small proportion of toxin
remained, which had the same physiological action. Later
A. Macfadyen found that the product obtained by grinding
up the spirilla frozen by means of liquid air had a very high
degree of toxicity when injected intravenously. Like Pfeiffer,
he found that the " endotoxin " was in great part destroyed at
60° C.

On the other hand, other observers (Petri, Ransom, Klein,
and others) have obtained toxic bodies from filtered cultures.
Metchnikoff, E. Roux, and Taurelli-Salimbeni have demon-
strated the formation of such diffusible toxic bodies in fluid
media. By means of cultures placed in collodion sacs in the
peritoneum of animals, they found that the living organisms
produce toxic bodies which diffuse through the wall of the
sac and cause toxic symptoms. By greatly increasing the
virulence of the organism, then growing it in bouillon and
filtering the cultures on the third and fourth day, they obtained
a fluid which was highly toxic to guinea-pigs (the fatal dose

IMMUNITY 457

usually being | c.c. per 100 gnu. weight). The symptoms
i-lust-ly resemble those obtained by Pfeiffer. They found that
the toxicity of the filtrate was not altered by boiling, — appar-
ently this toxic substance is different from Pfeiffer's endotoxin.
Within recent times numerous attempts have been made to
procure toxic fluids by disintegrating the cholera spirilla, e.g.
by methods of grinding, by solution by alkali, by autolysis, etc.,
and a certain measure of success has been reached. There has
also been much discussion as to whether the toxic bodies
obtained are merely liberated endotoxins or whether they
also represent true extracellular toxins. The true relations
of these bodies has not yet been determined, but it would seem
that in all probability the greater part of the toxic substance
is closely bound up with the bacterial protoplasm, and is only
set free on its disintegration.

Immunity. — As this subject is discussed later, only a few
facts will be here stated, chiefly for the purpose of making clear
what follows with regard to the means of distinguishing the
cholera spirillum from other organisms. The guinea-pig or any
other animal may be easily immunised against the cholera
organism by repeated injections (conveniently made into the
peritoneum) of non-fatal doses of dead spirilla; later the
living organisms may be used. In this way a high degree of
immunity against the organism is developed ; and further, the
blood serum of an animal thus immunised (anti-cholera serum)
has markedly protective power when injected, even in a small
quantity, into a guinea-pig along with five or ten times the fatal
dose of the living organism. Under these circumstances the
spirilla undergo a granular transformation and, ultimately,
solution; this phenomenon is generally known as Pfeiffer's
reaction, and was applied by him to distinguish the cholera
spirillum from organisms resembling it. The following are the
details : —

Pfeiffer's Reaction.— A. loopful (2 mgrm.) of a recent agar culture of the
organism to be tested is added to 1 c.c. of ordinary bouillon containing
'001 c.c. of anti-cholera serum. The mixture is then injected into the
peritoneal cavity of a young guinea-pig (about 200 gnu. in weigbt), and
the peritoneal fluid of this animal (conveniently obtained by means of
capillary glass tubes inserted into the peritoneum) is examined micro-
scopically alter a few minutes. If the spirilla injected have been cholera
spirilla, it will be found that they become motionless, swell up into
globules, and ultimately, break down and disappear — positive result. If
they are found active and motile, then the possibility of their being
true cholera spirilla may be excluded — negative result. In the former
case (positive result) there is, however, still the possibility that the
organism is devoid of pathogenic properties and has been destroyed by

458 CHOLERA

the normal peritoneal fluid. A control experiment should accordingly
be made with '001 c.c. of normal serum in place of the anti-cholera serum.
If no alteration of the organism occurs with its use, then the conclu-
sion is that a true reaction has been given. Corresponding bacteriolytic
effects may be obtained by in vitro methods, introduced since Pfeiffer's
original method (p. 534).

The serum of an animal immunised by the above method has
also marked agglutinative and other anti-bacterial properties
(p. 541) against the cholera spirillum, and these properties closely
correspond with Pfeiffer's reaction as regards specificity. Such
a serum has, however, little protective effect against the toxic
action of the dead spirilla, and Pfeiffer maintains that little or
no antitoxin to the endotoxin can be produced. On the other
hand, Macfadyen, by injecting the endotoxin derived from the
spirilla by grinding, obtained a serum which had antitoxic as well
as agglutinative and bacteriolytic properties (vide Immunity).
Metchnikoff and others have also obtained antitoxic sera which
act on the extracellular toxins obtained by filtration.

The serum of cholera convalescents has been found to possess
protective and increased bactericidal action. These properties of
the serum may be present eight or ten days after the attack of
the disease, but are most marked four weeks after ; they then
gradually diminish. Specific agglutinative properties have, how-
ever, been detected in the serum of cholera patients at a much
earlier date, in some cases even on the first day of the disease,
though usually a day or two later. The dilutions used have
been usually 1 : 15 to 1 : 120, and these had no appreciable
effect on organisms other than the cholera spirillum (Achard and
Bensande). In some cases, however, the agglutinative property
may not appear after recovery. Variations in the opsonic index,
analogous to those in other diseases, have recently been observed
in cholera, a marked fall on the acute onset of the disease being
a noteworthy feature.

Within recent times there have been introduced for therapeutic
purposes several so-called anti-sera which are supposed to be anti-
toxic as well as anti-bacterial, and of these the two most ex-
tensively used are those of Kraus and SchurupofF. Reports
regarding the effects of these sera during the Russian epidemic
are of somewhat conflicting character, but in any case it cannot
be said that they have a markedly beneficial action. They have
further been critically examined by others, who deny to them any
marked antitoxic action when tested experimentally. While,
therefore, it may be admitted that antitoxins to some of the
cholera toxins may be obtained, yet Pfeiffer's position, that

METHODS OF DIAGNOSIS 459

cholera anti-sera have little effect on at least most of the endo-
toxins, cannot 1x3 said to be shaken. It should be noted, how-
ever, that he disclaims having made the general statement, often
ascribed to him, that no antitoxins are formed to endotoxins.

Anti-Cholera Inoculation. — Haffkine's method for inoculation
against cholera exemplifies the above principles. It depends
upon (a) attenuation of the virus — that is, the cholera organism,
and (b) exaltation of the virus. The virulence of the organism
is diminished by passing a current of sterile air over the surface
of the cultures, or by various other methods. The virulence is
exalted by the method of passage — that is, by growing the
organism in the peritoneum in a series of guinea-pigs. By the
latter method the virulence after a time is increased twenty-fold
— that is, the fatal dose has been reduced to a twentieth of the
original. Cultures treated in this way constitute the virus exalte.
Subcutaneous injection of the virus exalte produces a local
necrosis, and may be followed by the death of the animal, but if
the animal be treated first with the attenuated virus, the sub-
sequent injection of the virus exalte produces only a local oedema.
After inoculation first by attenuated and afterwards by exalted
virus, the guinea-pig has acquired a high degree of immunity, and
Haffkine believed that this immunity was effective in the case
of every method of inoculation — that is, by the mouth as well as
by injection into the tissues. After trying his method on the
human subject and finding it free from risk, he extended it
in practice on a large scale in India in 1894. In the human
subject two or sometimes three inoculations were formerly made
with attenuated virus before the virus exalte" was used ; now,
however, a single injection of the latter is usually practised.
The results of preventive inoculation in India and also during
the recent epidemic in Russia have been such as to establish its
efficiency, both the case incidence and the mortality being
reduced.

Methods of Diagnosis. — In the first place, the stools ought
to be examinod microscopically. Dried film preparations should
be made and stained by any ordinary stains, though carbol-fuchsiu
diluted four times with water is specially to be recommended.
Hanging-drop preparations, with or without the addition of a
weak watery solution of gentian-violet or other stain, should also
be made, by which method the motility of the organism can be
readily seen. By microscopic examination the presence of spirilla
will be ascertained, and an idea as to their number obtained.
In some cases the cholera spirilla are so numerous in the stools
that a picture is presented which is obtained in no other con-

460 CHOLERA

dition, and a microscopic examination may be sufficient for
practical purposes. According to Koch, a diagnosis was made
in 50 per cent, of the cases during the Hamburg epidemic by
microscopic examination alone. In the case of the first appear-
ance of a cholera- like disease, however, all the other tests
should be applied before a definite diagnosis of cholera is made.
Dunbar has recently introduced a method for rapid diagnosis
which depends on the properties of an anti-cholera serum. Two
hanging-drop preparations are made, each consisting of a small
portion of mucus from the suspected stool broken up in peptone
solution. To one a drop of a 50-fold dilution of normal serum
is added, to the other a drop of a 500-fold dilution of an active
cholera serum. If the spirilla present are cholera organisms,
they retain their motility in the first preparation, while they lose
it and then become agglutinated in the second. By this method
a diagnosis may sometimes be given in a few minutes.

If the organisms are very numerous, gelatin or agar plates
may be made at once and pure cultures obtained.

If the spirilla occur in comparatively small numbers, the best
method is to inoculate peptone solution (1 per cent.) and incubate
for from eight to twelve hours. At the end of that time the spirilla
will be found on microscopic examination in enormous numbers
at the surface, and thereafter plate cultures can readily be made.
If the spirilla are very few in number, or if a suspected water is
to be examined for cholera organisms, the peptone solution
which has been inoculated should be examined at short intervals
till the spirilla are found microscopically. A second flask of
peptone solution should then be inoculated, and possibly again
a third from the second, and then plates may be made. In such
circumstances Dieudonne's medium (p. 44) has been found of
much service.

When a spirillum has been obtained in pure condition by
these methods, the appearance of the colonies in plates should
be specially noted, the test for the cholera-red reaction should
be applied, and in many cases it is advisable to test the effects
of intraperitoneal injection of a portion of a recent agar culture
in a guinea-pig, the amount sufficient to cause death being also
ascertained. The agglutinating or sedimenting properties of the
serum of the patient should be tested against a known cholera
organism, and against the spirillum cultivated from the case.
The action of an anti-cholera serum, i.e. the serum of an animal
immunised against the cholera spirillum, should be tested in a
similar manner.

Up till recent times there had been cultivated, from sources

GENERAL SUMMARY 461

other than cholera cases, no organism which gave all the cultural
and biological tests (agglutination and Pfeiffer's reaction) of the
cholera spirillum. In 1905, however, Gotschlich obtained six
different strains of a spirillum which conformed in all these
respects. The organisms were obtained at El Tor from the
intestines of pilgrims uho had died with dysenteric symptoms,
and tliciv \\viv no cases of cholera in the vicinity. The organisms
in question, however, differ In mi the cholera organism in having
marked ha-niolvtic action, and also in producing a rapidly acting
extracellular toxin. Kraus and others have found, on comparing
anti-sera to the cholera and El Tor spirilla, that while the anti-
bacterial properties are similar there is a difference in antitoxic
action. The El Tor antitoxin neutralises the cholera toxin, but
a cholera antitoxin has no effect on the El Tor toxin ; the El Tor
spirillum is thus peculiar as regards its toxic products. There
is accordingly difference of opinion as to whether these organisms
are to be regarded as a distinct species or as true cholera spirilla
which had been carried by the patients, though no symptoms
resulted from their presence. This instance exemplifies well
the great difficulty which may surround the identification of a
particular organism obtained from non-cholera cases, and one
can hardly doubt that if cholera-like symptoms had been present
in the El Tor cases, the spirilla would have been accepted as
varieties of the cholera organism. None of the facts ascertained,
however, really affect the question as to the causal relationship
of Koch's spirillum to cholera, although they indicate the
difficulties which may attend the bacteriological diagnosis in
isolated cases of disease.

General Summary. — We may briefly summarise as follows
the facts in favour of Koch's spirillum being the cause of cholera :
First, there is the constant presence of spirilla in true cases of
cholera, which on the whole conform closely with Koch's
description, though variations undoubtedly occur. Moreover,
the facts known with regard to their conditions of growth, etc.,
are in conformity with the origin and spread of cholera epidemics.
Secondly, the experiments on animals with Koch's spirillum or
its toxins give ;is definite results as one can reasonably look for
in view of the fact that animals do not suffer naturally from the
disease. Thirdly, the experiments on the human subject and
the results of accidental infection by means of pure cultures are
also strongly in favour of this view. Fourthly, the agglutinative
and protective properties of the serum of cholera patients and
convalescents constitute another point in its favour. Fifthly,
bacteriological methods, which proceed on the assumption that

462 CHOLERA

Koch's spirillum is the cause of the disease, have been of the
greatest value in the diagnosis of the disease. And lastly, the
results of Haffkine's method of preventive inoculation in the
human subject, which are on the whole favourable, also supply
additional evidence. If all these facts are taken together, we
consider the conclusion must be arrived at that the growth of
Koch's spirillum in the intestine is the immediate cause of the
disease. This does not exclude the probability of an important
part being played by conditions of weather and locality, though
such are very imperfectly understood. Pettenkofer, for example,
recognised two main factors in the causation of epidemics, wrhich
he designated x and y, and considered that these twro must be
present together in order that cholera may spread. The or is the
direct cause of the disease — an organism which he admitted
to be Koch's spirillum ; the y includes climatic and local con-
ditions, e.g. state of ground-water, etc.

Other Spirilla resembling the Cholera Organism. — These
have been chiefly obtained either from water contaminated by
sewage or from the intestinal discharge in cases with choleraic
symptoms. Some of them differ so widely in their cultural and
other characters (some, for example, are phosphorescent) that no
one would hesitate to classify them as distinct species. Others,
however, closely resemble the cholera organism.

The vibrio berolinensis, cultivated by Neisser from Berlin sewage
water, differs from the cholera organism only in the appearance of its
colonies in gelatin plates, its weak pathogenic action, and its giving a
negative result with Pfeiffer's test. It, however, gives the cholera-red
reaction. The vibrio Danubicus, cultivated by Heider from canal water,
also differs in the appearance of its colonies in plates, and also reacts
negatively to Pfeiffer's test ; in most respects it closely resembles the
cholera organism. Another spirillum (v. Ivanoff) was cultivated by
IvanofFfrom the stools of a typhoid patient after these had been diluted
with water. The organism differed somewhat in the appearance of its
colonies and in its great tendency to grow out in the form of long
threads, but Pfeiffer found that it reacted to his test in the same way as
the cholera organism, and he considered that it was really a variety of
the cholera organism. No spirilla could be found microscopically in the
stools in this case, and Pfeiffer is of the opinion that the organism
gained entrance accidentally. These examples will show how differences
of opinion, even amongst experts, might arise as to whether a certain
spirillum were really the cholera organism or a distinct species resem-
bling it.

A few examples may also be given of organisms cultivated
from cases in which cholera- like symptoms were present.

The vibrio of Massowah was cultivated by Pasquale from a case during

SPIRILLA RESEMBLING CHOLERA ORGANISM 463

a small epidemic of cholera. The organism so closely resembles Koch's
spirillum that it was accepted by several authorities as the true cholera
organism, and, as already stated, Metchnikolf produced by it cholera
symptoms in the human subject, and also the cholera-like disease in
young rabbits. It possesses four flagella, has a high degree of virulence,
producing septicnenna both in guinea-pigs and pigeons, and its colonies
in plates diller somewhat from the cholera organism. Moreover, it
reacts negatively to Pfeifl'er's test. Another organism, the v. (Undha,
was cultivated by Pasquale from a well, and was at Hrst accepted by
Pfeiffer as the cholera organism, but afterwards rejected, chiefly becaus'e
it failed to give the specific immunity reaction. It also differs somewhat
from the cholera organism in its pathogenic effects, and it fails to give
the cholera-red reaction, or gives it very faintly.

Pestana and Bettencourt also cultivated a species of spirillum from a
number of cases during an epidemic in Lisbon — an epidemic in which
there were symptoms of gastro-enteritis, although only in a few instances
did the disease resemble cholera. They also cultivated the same organism
from the drinking water. It differs from the cholera organism in the
appearance of its colonies and of puncture cultures in gelatin. It has
very feeble pathogenic effects, and gives a very faint, or no, cholera-red
reaction. To Pfeitfer's test it also reacts negatively. Another spirillum
(v. Romanus) was obtained by Celli and Santori from twelve out of forty-
four cases where there were the symptoms of mild cholera. This organism
does not give the cholera-red reaction, nor is it pathogenic for animals.
They look upon it as a "transitory variety" of the cholera organism,
though sufficient evidence for this view is not adduced.

We have mentioned these examples in order to show some of
the difficulties which exist in connection with this subject. It is
important to note that, on the one hand, spirilla which have
been judged to be of different species from the cholera organism,
have been cultivated from cases in which cholera-like symptoms
were present ; and, on the other hand, in cases of apparently true
cholera considerable variations in the characters of the cholera
organisms have been found. Such variations have especially
been recorded by Colonel Cunningham in India. It is there-
fore quite an open question whether some of the organisms
in the former class may not be cholera spirilla which have under-
gone variations as a result of the conditions of their growth.
The great bulk of evidence, however, goes to show that Asiatic
cholera always spreads as an epidemic from places in India where
the disease is endemic, and that its direct cause is Koch's spirillum
with the characters described above. It is sufficient to bear in
mind that choleraic symptoms may be produced by other causes,
and that in some of such cases spirilla which have a re-
semblance to Koch's orpin ism may be present in the intestinal
dis.-liarges, though rarely in large nimiU'rs.

A number of other spirilla have been cultivated, which are of
interest on account of their points of resemblance to the cholera

464 CHOLEKA

organism, though probably they produce no pathological con-
ditions in the human subject.

Metchnikoff's Spirillum (vibrio Metchnikovi). — This organism was
obtained by Gamaleia from an epidemic disease of fowls in Odessa, and
is of special interest on account of its close resemblance to the cholera
organism.

Morphologically the organism is practically identical with Koch's
spirillum (Fig. 136). It is actively motile, and has the same staining
reactions. Its growth in peptone-gelatin also closely resembles that of the
cholera organism, though it produces liquefaction more rapidly (Fig. 137,
A). In gelatin plates the young colonies are, however, smoother and more

circular. After liquefaction

* /~r»- • occurs, some of the colonies

y jfi "*•£„»' X are almost identical in

earance with those

/^'W

ap-

pearance with those of the
< " cholera vibrio, whilst others

show more uniformly turbid
contents. In puncture cul-
tures the growth takes place
more rapidly, but in appear-
ance closely resembles that
of the cholera organism a
few days older. Its growth
in peptone solution, too, is
closely similar, and it also
gives the cholora-red re-

action-

This organism can, how-
i (~, ** ' ever, be readily distinguished

from the cholera organism

FIG. 136.-Metclmikoff'.s spirillum, both in ^y the effects of inoculation

curved and straight forms ; from an agar on animals especially on

culture of twenty-four hours' growth. pigeons and guinea - pigs.

Stained with weak carbol-fuchsin. x 1000. Subcutaneous inoculation of

small quantities of pure cul-
ture in pigeons is followed

by septicaemia, which produces a fatal result usually within twenty-four
hours. Inoculation with the same quantity of cholera culture produces
practically no result ; even with large quantities death is rarely produced.
The vibrio Metchnikovi produces somewhat similar effects in the guinea-
pig to those in the pigeon, subcutaneous inoculation being followed by
extensive hsemorrhagic osdema and a rapidly fatal septicaemia. Young
fowls can be infected by feeding with virulent cultures. We have
evidence from the work of Gamaleia that the toxins of this organism
have somewhat the same action as those of the cholera organism.

The organism is therefore one which very closely resembles the cholera
organism, the results on inoculating the pigeon offering the most ready
means of distinction. It gives a negative reaction to Pfeiffer's test — that
is, the properties of an anti-cholera serum are not exerted against it. It
may also be mentioned that an organism which is apparently the same
as the vibrio Metchnikovi was cultivated by Pfuhl from water, and named
v. Nordhafen.

Finkler and Prior's Spirillum. — These observers, shortly after Koch's

FIXKLER AND PRIOR'S SPIRILLUM

465

discovery of the cholera organism, separated a spirillum, in a case of
cholera nostras, from the stools after they had been allowed to decompose
for several days. There is, however, no evidence that the spirillum has
any causal relationship to this or any other disease in the human subject.
Morphologically it closely resembles Koch's spirillum, and cannot be
distinguished from it by its microscopical characters, although, on the
whole, it tends to be rather thicker in the centre and more pointed at the
ends (Fig. 138). In cultures, how-
ever, it presents marked differences.
In puncture cultures on gelatin it
grows much more quickly, and lique-
faction is generally visible within
twenty- four hours. The liquefaction
spreads rapidly, and usually in forty-
eight hours it has produced a funnel-
shaped tube with turbid contents,
denser below (Fig. 137, B). In plate
cultures the growth of the colonies is
proportionately rapid. Before they
nave produced liquefaction around
them, they appear, unlike those of
the cholera organism, as minute
spheres with smooth margins. When
liquefaction occurs, they appear MS
little spheres with turbid contents,
which rapidly increase in size ; ulti-
mately general liquefaction occurs.
On potatoes this organism grows well
at the ordinary temperature, and in
two or three days has formed a slimy
layer of greyish -yellow colour, which
rapidly spreads over the potato. On
all the media the growth has a
distinctly foetid odour. A growth in
peptone solution fails to give the
cholera-red reaction at the end of
twenty -four hours, though later a
faint reaction may appear.

An organism cultivated by Miller
("Miller's Spirillum") from the
cavity of a decayed tooth in a human
subject is almost certainly the same
organism as Finkler and Prior's
spirillum.

Deneke's Spirillum— This organ-
ism was obtained from old cheese, and

is also known as the spirillum tyrogenum. It closely resembles Koch's
spirillum in microscopic appearances, though it is rather thinner and
smaller. Its growth in gelatin is also somewhat similar, but liquefaction
proceeds more rapidly, and the bell-shaped depression on the surface is
larger and shallower, whilst the growth has a more distinctly yellowish
tint. The colonies in plates also show points of resemblance, though the
youngest colonies are rather smoother and more regular on the surface,
and liquefaction occurs more rapidly than in the case of the cholera

3°

FIG. 137. — Puncture cultures in
peptone-gelatin.

A. Metchnikoff's spirillum. Five

days' growth.

B. Finkler and Prior's spirillum.

Four days' growth.
Natural size.

466 CHOLERA

organism. The colonies have, on naked-eye examination, a distinctly
yellowish colour. The organism does not give the cholera-red reaction,
and on potato it forms a thin yellowish layer when incubated above

x t, &

> . V 4

^ \'

FIG. 138. — Fiukler and Prior's spirillum ; from an agar culture

of twenty-four hours' growth.
Stained with carbol-fuchsin. x 1000.

30° C. When tested by intraperitoneal injection and by other methods,
it is found to possess very feeble, or almost no pathogenic properties.
Deneke's spirillum is usually regarded as a comparatively harmless
saprophyte.

CHAPTER XIX.

INFLUENZA, WHOOPING-COUGH, PLAGUE,
MALTA FEVER.

INFLUENZA.

THE first accounts of the organism now known as the influenza
bacillus were published simultaneously by Pfeiffer, Kitasato, and
Canon, in January 1892. The two first-mentioned observers
found, it in the bronchial _.

sputum, and obtained pure

cultures, and Canon ob-
served it in the blood in a

i
,'

l
* .v

•

» T

few cases of the disease.
It is, however, to Pfeiffer's
work that we owe most of
our knowledge regarding
its characters and action.
His results have been
amply confirmed by those
of others in various epi-
demics of the disease, and
this organism has been
generally accepted as the
cause of the disease, al-
though absolute proof is
.still wanting.

Microscopical Char-
acters.— The influenza bacilli as seen in the sputum are very
minute rods not exceeding 1 '5 /t in length and *3 p. in thickness.
They are straight, with rounded ends, and sometimes stain more
deeply at the extremities (Fig. 139). The bacilli occur singly
or form clumps by their aggregation, but do not grow into
chains. They show no capsule. They take up the basic aniline
-tains somewhat feebly, and are best stained by a weak solution

4«7

FIG. 139. — Influenza bacilli from a culture

on blood agar.
Stained with carbol-t'uchsin. x 1000.

468 INFLUENZA

(1 : 10) of carbol-fuchsin applied for five to ten minutes. They
lose the stain in Gram's method. They are non-motile, and do
not form spores.

In many cases of the disease, especially in the early stages of
the more acute, influenza bacilli are present in large numbers
and may be easily found. On the other hand, it is often
difficult or impossible to find them, even when the symptoms
are severe ; this may be due to the restriction of the organisms
to some part not readily accessible, or it may be that they
actually die out in great part while the effects of their toxins
persist. It has also been observed in recent epidemics, in which
the disease has been less widespread and on the whole less
severe, that the period during which the bacilli have been readily
demonstrable in the secretions has been on the average shorter
than in the previous epidemics.

Cultivation. — The best medium for the growth of the
influenza bacillus is blood-smeared agar (see p. 43), which
was introduced by Pfeiffer for this purpose. He obtained
growths of the bacilli on agar which had been smeared with
influenza sputum, but he failed to get any m6-cultures on the
agar media or on serum. The growth in the first cultures he
considered to be probably due to the presence of certain organic
substances in the sputum, and accordingly he tried the expedient
of smearing the agar with drops of blood before making the in-
oculations. In this way he completely succeeded in attaining
his object. The blood of the lower animals is suitable, as well
as human blood ; and the favouring influences of the blood
would appear to be due to the haemoglobin, as a solution of this
substance is equally effective. The colonies of the influenza
bacilli on blood agar, incubated at 37° C., appear within twenty-
four hours, in the form of minute circular dots almost trans-
parent, like drops of dew. When numerous, the colonies are
scarcely visible to the naked eye, but when sparsely arranged
they may reach the size of a small pin's head. This size is
generally reached on the second day. In cultures the bacilli
may show considerable variations in size and in shape ; they die
out somewhat quickly, and in order to keep them alive sub-
cultures should be made every four or five days. By this
method the cultures may be maintained for an indefinite period.
Even in sub-cultures growth on the ordinary agar media is slight
and somewhat uncertain ; there is, however, evidence that growth
is more marked when other organisms are present, that is, is
favoured by symbiosis. Neisser, for example, was able to
cultivate the influenza bacillus on plain agar through several

DISTRIBUTION OF BACILLI 469

generations by growing the xerosis bacillus along with it ; dead
cultures of the latter had not the same favouring effect, A very
small amount of growth takes place in bouillon, but it is more
marked when a little fresh blood is added. The growth forms
a thin whitish deposit at the bottom of the flask. The limits
of growth are from 25° to 42° C., the optimum temperature
being that of the body. The influenza bacillus is a strictly
aerobic organism.

The powers of resistance of this organism are of a low order.
Pfeiffer found that dried cultures kept at the ordinary tempera-
ture were usually dead in twenty hours, and that if sputum
were kept in a dry condition for two days, all the influenza
bacilli were dead, or rather, cultures could be no longer obtained.
Their duration of life in ordinary water is also short, the bacilli
usually being dead within two days. From these experiments
Pfeiffer concludes that outside the body in ordinary conditions
they cannot multiply, and can remain alive only for a short
time. The mode of infection in the disease he accordingly
considers to be chiefly by means of fine particles of disseminated
sputum, etc.

Distribution in the Body. — The bacilli are found chiefly in
the respiratory passages in influenza. They may be present in
large numbers in the nasal secretion, generally mixed with a
considerable number of other organisms, but it is in the small
masses of greenish-yellow sputum from the bronchi that they
are present in largest numbers, in many cases almost in a state of
purity. They occur in clumps which may contain as many as
100 bacilli, and in the early stages of the disease are chiefly
lying free. As the disease advances, they may be found in
considerable numbers within the leucocytes, and towards the
end of the disease a large proportion have this position. It is
a matter of considerable importance, however, that they may
persist for weeks after symptoms of the disease have disappeared,
and may still be detected in the sputum. Especially is this the
case when there is any chronic pulmonary disease. They also
occur in large numbers in the capillary bronchitis and catarrhal
pneumonia of influenza, as Pfeiffer showed by means of sections
of the affected parts. In these sections he found the bacilli
lying amongst the leucocytes which filled the minute bronchi,
and also penetrating between the epithelial cells and into the
superficial parts of the mucous membrane. Other organisms
also, especially Fraenkel's pneumococcus, may be concerned in
the pneumonic conditions following influenza. In some cases
influenza occurs in tubercular subjects, or is followed by tubercular

470 INFLUENZA

affection, in which cases both influenza and tubercle bacilli may
be found in the sputum. In such a condition the prognosis is
very grave. Regarding the presence of influenza bacilli in the
other pulmonary complications following influenza, much in-
formation is still required. Occasionally in the foci of sup-
purative softening in the lung the influenza bacilli have been
found in a practically pure condition. In cases of empyema
the organisms present would appear to be chiefly streptococci
and pneumococci ; whilst in the gangrenous conditions, which
sometimes occur, a great variety of organisms has been found.

Pfeiffer's observations on a large series of cases convinced him
that the organism was very rarely present in the blood — that in
fact its occurrence there must be looked upon as exceptional.
The conclusions of other observers have, on the whole, confirmed
this statement, and it is probable that 'the chief symptoms in the
disease are due to toxins absorbed from the respiratory tract
(vide infra). In a recent work, however, Ghedini was able to
cultivate the organism from the blood and spleen during life in
over 50 per cent, of the cases examined : he found that its
occurrence in these situations was specially frequent during
marked fever. The bacillus may be present in some of the
lesions complicating influenza. Pfeiffer found it in inflammation
of the middle ear, but in a case of meningitis following influenza
Fraenkel's diplococcus was present. In a few cases of meningitis,
however, the influenza bacillus has been found, sometimes alone,
sometimes along with pyogenic cocci (Pfulil and Walter, Cornil
and Durante) ; Pfulil considers that in these the path of
infection is usually a direct one through the roof of the nasal
cavity. This observer also found post mortem, in a rapidly fatal
case with profound general symptoms, influenza bacilli in various
organs, both within and outside of the vessels. In a few cases
also the bacilli have been found in the brain and its membranes
with, little tissue change in the parts around.

Extensive observations on the bacteriology of the respiratory
system show that influenza-like bacilli may be present in a great
variety of conditions ; we have, in fact, once more to do with a
group of organisms with closely allied characters, of which
Pfeifter's influenza bacillus was the first recognised example.
These " pseudo-influenza " bacilli have been obtained from the
fauces, bronchi, and lungs in inflammatory conditions, and also in
various specific fevers. To this group belongs the bacillus which
has been cultivated from cases of whooping-cough by Spengler,
Jochmann, Davis, and others, and which is present in consider-
able numbers in a large proportion of cases of this disease

EXPERIMENTAL INOCULATION 471

(p. 472). Miiller's "trachoma bacillus" (p. 219) is a member of
the same group. All these organisms are very restricted in their
growth, and require the addition of blood or haemoglobin to the
ordinary culture media ; hence they are sometimes spoken of as
hamiophilic bacteria. Some of the examples are a little larger
than the influenza bacillus, and tend to form short filaments,
but others are quite indistinguishable. All of them also seem to
have very feeble pathogenic properties towards the lower animals.
At present it can scarcely be claimed as possible to identify
Pfeiffer's bacillus by its microscopic and cultural characters.

Experimental Inoculation. --There is no satisfactory evidence
that any of the lower animals suffer from influenza in natural
conditions, and accordingly we cannot look for very definite
results from experimental inoculation. Pfeiffer, by injecting
living cultures of the organism into the lungs of monkeys, in
three cases produced a condition of fever of a remittent type.
There was, however, little evidence that the bacilli had under-
gone multiplication, the symptoms being apparently produced
by their toxins. In the case of rabbits, intravenous injection of
living cultures produces dyspnoea, muscular weakness, and
slight rise of temperature, but the bacilli rapidly disappear in
the body, and exactly similar symptoms are produced by
injection of cultures killed by the vapour of chloroform.
Pfeiffer, therefore, came to the conclusion that the influenza
bacilli contain toxic substances which can produce in animals
some of the symptoms of the disease, but that animals are not
liable to infection, the bacilli not having power of multiplying
to any extent in their tissues.

Cantani succeeded in producing infection to some extent in rabbits, by
injecting the bacilli directly into the anterior portion of the brain. In
these experiments the organisms spread to the ventricles, and then
through the spinal cord by means of the central canal, afterwards in-
fecting the substance of the cord. An acute encephalitis was thus pro-
duced, and sometimes a purulent condition in the lateral ventricles.
The bacilli were, however, never found in the blood or in other organs.
Similar symptoms were also produced by injection of dead cultures,
though in this case the dose required to be five or six times larger.
Cantani therefore concludes that the brain substance is the most suitable
nidus for their growth, but agrees with Pfeirl'er in believing that the
chief symptoms are produced by toxins resident in the bodies ol the bacilli.
He made control experiments by injecting other organisms, and also by
injecting inert substances into the cerebral tissue.

The evidence, accordingly, that the influenza bacillus is the
cause of the disease rests chiefly on the well-established fact that
it is always present in the secretions of the respiratory tract in

472 WHOOPING-COUGH

true cases of influenza, and often in very large numbers. The
observed relationships of the organism to lesions in the lungs
and elsewhere leave no room for doubt that it is possessed of
pathogenic properties, but we cannot yet maintain that its causal
relationship to epidemic influenza is absolutely established.

Methods of Examination. — (a) Microscopic. — A portion of the
greenish-yellow purulent material which often occurs in little
round masses in the sputum should be selected, and film prepara-
tions should be made in the usual way. Films are best stained
by Ziehl-Neelsen carbol-fuchsin diluted with ten parts of water,
the films being stained for ten minutes at least. In sections of
the tissues, such as the lungs, the bacilli are best brought out,
according to Pfeiffer, by staining with the same solution as above
for half an hour. The sections are then placed in alcohol
containing a few drops of acetic acid, in which they are
dehydrated and slightly decolorised at the same time. They
should be allowed to remain till they have a moderately light
colour, the time varying according to their appearance. They
are then washed in pure alcohol, cleared in xylol, and afterwards
mounted in balsam.

(b) Cultures. — A suitable portion of the greenish-yellow
material having been selected from the sputum, it should be
washed well in several changes of sterilised water. A portion
should then be taken on a platinum needle, and successive
strokes made on the surface of blood-agar tubes. The tubes
should then be incubated at 37° C., when the transparent
colonies of the influenza bacillus will appear, usually within
twenty -four hours. These should give a negative result on
inoculation on ordinary agar media.

WHOOPING-COUGH.

Up to the year 1906, the chief result of bacteriological
observations, of which those of Spengler, Krause and Jochmann,
and Davis may specially be mentioned, had been to demonstrate
the very frequent presence of minute influenza-like and haemo-
philic bacilli in the sputum and also in the lesions in this disease.
In the year mentioned, however, Bordet and Gengou published
an account of another minute organism, and brought forward
certain facts which gave strong support to its etiological
relationship. A short ^description of this bacillus may
accordingly be given.

Characters of the Bacillus (Bordet-Gengou). — The organism,
as seen, for example, in the sputum, occurs in the form of

CHARACTERS OF THE BACILLUS

473

minute oval rods scarcely larger than the influenza bacillus.
They stain rather faintly with ordinary stains, and their margin
and extremities are often more deeply coloured than the centre,
which may appear as an uncoloured spot; they are Gram-
negative and do not form spores. In cultures they present the
same characters and are less pleomorphous than the influenza
bacillus (Fig. 140). They are specially numerous at the beginning
of the disease, and they may be found in large numbers in almost
pure culture in the opaque whitish sputum expectorated from the
bronchi ; as the disease advances they become scanty, and may
disappear when the
symptoms of the disease
are still prominent.
Bordet and Gengou suc-
ceeded in obtaining pure
cultures on the blood-
agar medium described
on p. 44, and this was
found to be the most
suitable of all the media
tried. In the first cul-
tures growth is very
scanty and may be in-
visible, but later it
becomes much more
abundant, and sub-cul-
tures may also be

,., ,. FIG. 140.1— Film preparation from a twenty-

readily made on ordm- four hours> cuiture of the bacillus of whoop-
ary serum-agar media. ing-cough. (Bordet-Gengou).
As compared with that Stained with carbol-fuchsin. xlOOO.

of the influenza bacillus,

growth is thicker and less transparent and the margins are
more sharply marked off; the presence of haemoglobin, though
favouring the growth, is not so essential as in the case of
the latter organism. The organism is a strict aerobe, and in
the case of cultures in fluid media, e.y. serum bouillon, the
tubes ought to be placed in a sloped position, in order to
expose a greater surface to the air. Bordet and Gengou
completely confirmed the observations mentioned above as to the
very frequent, almost constant, presence of influenza-like bacilli.
They obtained growths of these organisms, and on comparing
them with their own bacillus found that distinct cultural

1 We are indebted to Dr. Bordet for the culture from which this preparation

was made.

'. »*• »«* *~
•*»*£. -,;,.:-

474 WHOOPING-COUGH

differences could be made out. The most important distinctions
were, however, obtained on studying the serum reactions of
convalescents from the disease. They found that in many cases,
though not invariably, such sera agglutinated their bacillus, but
none of the influenza-like organisms. The most important
result, however, was that in every case examined the serum of
convalescents gave the deviation of complement reaction very
markedly with the whooping-cough bacillus, but with none of the
others. This means, of course, that a true anti-substance to the
bacillus (immune-body or substance sensibilisatrice) was present
in the serum, and points to a true infection with the organism
(p.131).

Pathogenic Effects. — The general results obtained by
Bordet and Gengou were that the ordinarily used animals were
not susceptible to true infection with the bacillus, but that it
contained a powerfully acting endotoxin, which produced both
local and general effects. The injection of a small quantity of
the bacillus into the eye of a rabbit produced a local necrosis,
with little inflammatory change, and the introduction of dead,
as well as living, cultures into the peritoneal cavity of a guinea-
pig caused death from toxic action, there being great effusion
into the cavity and numerous haemorrhages in its lining.

They advanced the view that the bacillus is present in large
numbers at the beginning of the disease, and inflicts some local
damage on the bronchial tubes which may persist after the dis-
appearance of the bacillus and keep up the irritation.

Similar results were obtained with an endotoxin prepared accord-
ing to Besredka's method. It was not found possible to obtain an
anti-toxin to this toxin. Very important results have, however,
been since obtained by Klimenko, who succeeded in infecting
monkeys and young dogs by intratracheal injection of pure
cultures of the bacillus. After a period of incubation, there
occurred an illness in which symptoms of pulmonary irritation
and irregular pyrexia were outstanding features. Usually, in
the case of the dogs, a fatal result followed after two or three
weeks, and post mortem there were found symptoms of catarrh
of the respiratory tract and sometimes patches of broncho-
pneumonia, from which the bacillus could be recovered in pure
culture. The serum of the infected animals gave the deviation
of complement reaction. A specially interesting fact is that a
number of healthy young dogs contracted the disease by contact
with the inoculated. Fraenkel also obtained positive results,
closely similar to those of Kliinenko, on inoculation with pure
cultures of the bacillus. •

METHODS OF EXAMINATION 475

The results of Bordet and Gengou have received general con-
firmation, although it is to be noted that Fraenkel and also
Wollstein failed to obtain the deviation of complement reaction
with the serum of convalescents. Bordet and Gengou have
inquired into this discrepancy in the case of the former, and find
that it depends on the nature of the culture medium used. At
present it is not justifiable to make a definite pronouncement on
the subject. We can only say that Bordet and Gengou have
made out a strong case for the etiological relationship .of their
bacillus, and that their observations have been confirmed by
those of others.

Methods of Examination. — A portion of sputum expectorated
during a paroxysm of coughing should be obtained at as early as
possible a stage of the disease ; film preparations should be made
in the usual way and stained by carbol-thionin or carbol-methylene
blue. If the characteristic bacilli largely preponderate, tubes of
the Bordet-Gengou medium may then be inoculated and in-
cubated. If there are numerous colonies of other organisms in
the tubes, a portion of the intervening agar should be scraped
with a needle and fresh tubes inoculated. As already said,
growth is at first very scanty but becomes more luxuriant in
sub-cultures. On pure cultures being obtained, the deviation of
complement test is to be applied by the method described (p. 1 30).

PLAGUE.

The bacillus of Oriental plague or bubonic pest was discovered
independently by Kitasato and by Yersin during the epidemic
at Hong Kong in 1894. They cultivated the organism from a
large number of cases of plague, and reproduced the disease in
susceptible animals by inoculation of pure cultures. It is to
be noted that during an epidemic of plague, sometimes even
preceding it, a high mortality has been observed amongst certain
animals, especially rats and mice, and that from the bodies of
these animals found dead in the plague-stricken district, the same
bacillus was obtained by Kitasato and also by Yersin.

Bacillus of Plague — Microscopical Characters. — As seen in
the affected glands or buboes in this disease, the bacilli are
small oval rods, somewhat shorter than the typhoid bacillus,
and about the same tliirkiu-ss (Fi«j. Ill), though considerable
variations in size occur. They have rounded ends, and in
stained preparations a portion in the middle of the bacillus is
often left uncoloured, giving the so-called " i>olar staining." In
films from" the tissues they are found scattered amongst the cells,

476

PLAGUE

for the most part lying singly, though pairs are also seen. On
the other hand, in cultures in fluids, e.g. bouillon, they grow
chiefly in chains, sometimes of considerable length, the form
known as a streptobacillus resulting (Fig. 143). In young agar
cultures the bacilli show greater variation in size, and polar
staining is less marked than in the tissues : sometimes forms
of considerable length are present. After a time involution

FIG. 141. —Film preparation from a plague bubo showing enormous
numbers of bacilli, most of which show well-marked bipolar staining.
Stained with weak gentian-violet, x 1000.

forms appear, especially when the surface of the agar is dry;
but the formation of these is much more rapid and more marked
when 2 to 5 per cent, of sodium chloride is added to the medium,
constituting the so-called " salt agar " (Hankin and Leumann).
On this medium, especially with the higher percentage, the in-
volution forms assume a great size and a striking variety of
shapes, large globular, oval, or pyriform bodies resulting (Fig.
144) ; with about 2 per cent, sodium chloride, after twenty-
four hours' incubation, the most striking feature is -a general

CULTIVATION OF BACILLUS

477

/ ^

V>*^^i J\ » S v / -*r> t-

fouml w\y!>? ', V-~* v'-« x '»§& '

flagelk .> KVy^V'VVji ',
ith ''^c*^ ,' .. «

E <#'%>W^

"': vVCs^jy

of ^ , - ^ -_ v. .rx '

enlargement of all the bacilli. Sometimes in the tissues they
are seen to be surrounded _____

by an unstained capsule,
though this appearance is
by no means common.
They do not form spores. ^T,
Gordon, who has found >\^
that they possess
which, however, stain with
difficulty, states that they
are motile. Most ob-
servers, however, and with
these we agree, have
failed to find evidence ui r - t .

true inotility. They stain * ' ** • H^'

readily with the basic * J^s~~ '

aniline stains, but are

, FIG. 142.- Bacillus of plague from a young

decolorised by Grams culture on agar.

i lift hod. Stained with weak carbol-fuchsiii. x 1000.

Cultivation. — From the

atl'rcted glands, etc., the bacillus can readily be cultivated on

the ordinary media. It grows best at the temperature of the

body, though growth
occurs as low as 18° C.
On agar and on blood
serum the colonies are
whitish circular discs of
somewhat transparent
appearance and smooth,
shining surface. When
examined with a lens,
their borders appear
slightly wavy. In stroke
cultures on agar there
forms a continuous line
of growth with the
same appearance, show-
ing partly separated

FIG. 143 -Bacillus of plague in chains show- colonies at its margins,
ing polar staining. From a young culture ,I7,

in bouillon. When agar cultures are

Stained with thionin-blue. x 1000. kept at the room tempera-
ture, some of the colonies

may show a more luxuriant growth with more opaque appearance

than the rest of the growth, the appearance in fact being often

478 PLAGUE

such as to suggest the presence of impurities in the cultures. In
stab cultures in peptone gelatin, growth takes place along the
needle track as a white line, composed of small spherical colonies.
On the surface of the gelatin a thin, semi-transparent layer may
be formed, which is usually restricted to the region of puncture,
though sometimes it may spread to the wall of the tube ; some-
times, however, there is practically no surface growth. There is
no liquefaction of the medium. In gelatin plates the superficial
colonies develop first and form slightly raised semi-transparent
discs with somewhat crenated margins ; the deeper colonies are

smaller and of spherical
shape, with smooth out-
line. In bouillon the

^ 'v growth usually forms a

%••"" slightly granular or
• v powdery deposit at the

* ^°°* anc^ sides of the

• flask, somewhat resem-
bling that of a strepto-
coccus. If oil or melted

-i, *m+ butter is added to the

**t % ^" « V-*V"4«?I ' \S bouillon so that drops
-* ***«•* ^?£**:V*' float on the surface, then

a striking mode of growth
may result, to which the
term "stalactite" has been

FIG. 144.— Culture of the bacillus of plague applied. This consists in
on 4 per cent, salt agar, showing involution the growth starting from
forms of great variety of size and shape. ,, 11TU3PP allrfoPP Of flip
See also Plate IV., Fig. 17. tne Uinclern &™ce 01

Stained with carbol-thionin-blue. xlOOO. fat globules and extend-
ing downwards in the

form of pendulous, string-like masses. These masses are
exceedingly delicate, and readily break off on the slightest
shaking of the flask; accordingly during their formation the
culture must be kept absolutely at rest. This manner of
growth constitutes an important but not absolutely specific
character of the organism ; unfortunately it is not supplied by
all races of the organism, and varies from time to time with
the same race. The organism flourishes best in an abundant
supply of oxygen ; in strictly anaerobic conditions almost no
growth takes place.

The organism in its powers of resistance corresponds with
other spore-free bacilli, and is readily killed by heat, an exposure
for an hour at 58° C. being fatal. On the other hand, it has

ANATOMICAL CHANGES 479

remarkable powers of resistance against cold ; it has been exposed
to a temperature several degrees below freezing-point without
I iri ng killed. Experiments on the effects of drying have given
somewhat diverse results, but as a rule the organism has been
found to be dead after being dried for from six to eight days,
though sometimes it has survived the process for a longer period ;
exposure to direct sunlight for three or four hours kills it. When
cultivated outside the body the organism often loses its virulence,
but some races remain virulent in cultures for a long period of
time.

Anatomical Changes and Distribution of Bacilli. — The
disease occurs in several forms, the bubonic and the pulmonary
being the best recognised ; to these may be added the septiccemic.
The most striking feature in the bubonic form is the affection
of the lymphatic glands, which undergo intense inflammatory
swelling, attended with haemorrhage, and generally ending in
a greater or less degree of necrotic softening if the patient lives
long enough. The connective tissue around the glands is
similarly affected. The bubo is thus usually formed by a
collection of enlarged glands fused by the inflammatory swelling.
True suppuration is rare. Usually one group of glands is
affected first, constituting the primary bubo — in the great
majority the inguinal or the axillary glands — and afterwards
other groups may become involved, though to a much less
extent. Along with these changes there is great swelling of
the spleen, and often intense cloudy swelling of the cells of the
kidneys, liver, and other organs. There may also occur secondary
areas of haemorrhage and necrosis, chiefly in the lungs, liver,
and spleen. The bacilli occur in enormous numbers in the
swollen glands, being often so numerous that a film preparation
made from a scraping almost resembles a pure culture (Fig.
141). In sections of the glands in the earlier stages the bacilli
are found to form dense masses in the lymph paths and sinuses
(Fig. 145), often forming an injection of them; they may also
be seen growing as a fine reticulum between the cells of the
lymphoid tissue. At a later period, when disorganisation of
the gland has occurred, they become irregularly mixed with the
cellular elements. Later still they gradually disappear, and
wlp.-ii necrosis is well advanced it may be impossible to find any
— a point of importance in connection with diagnosis. In the
spleen they may be very numerous or they may be scanty,
according to the amount of blood infection which has occurred ;
in the secondary lesions mentioned they are often abundant,
lu the pulmonary form the lesion is the well-recognised "plague

480

PLAGUE

pneumonia." This is of broncho-pneumonic type, though large
areas may be formed by confluence of the consolidated patches,
and the inflammatory process is attended usually by much
hemorrhage ; the bronchial glands show inflammatory swelling.
Clinically there is usually a fairly abundant frothy sputum often
tinted with blood, and in it the bacilli may be found in large
numbers. Sometimes, however, cough and expectoration may

FIG. 145. — Section of a human lymphatic gland in plague, showing
the injection of the lymph paths and sinuses with masses of plague
bacilli — seen as black areas.

Stained with carbol-thionin-blue. x 50.

be absent. The disease in this form is said to be invariably
fatal. In the septiccemic form proper there is no primary bubo
discoverable, though there is almost always slight general en-
largement of lymphatic glands; here also the disease is of
specially grave character. A bubonic case may, however,
terminate with septicaemia ; in fact all intermediate forms occur.
An intestinal form with extensive affection of the mesenteric
glands has been described, but it is exceedingly rare — so much

EXPERIMENTAL INOCULATION 481

so that many observers with extensive experience have doubted
its occurrence. In the various forms of the disease the bacilli
occur also in the blood, in which they may be found during life
by microscopic examination, chiefly, however, just before death
in very severe and rapidly fatal cases. The examination of the
blood by means of cultivation experiments is, however, a much
more reliable procedure. For this purpose about 1 c.c. of blood
may be withdrawn from a vein and distributed in flasks of
bouillon (p. 72). It may be said from the results of different
investigators that the bacillus may be obtained by culture in
fully 50 per cent, of the cases, though the number will necessarily
vary in different epidemics. The Advisory Committee, ap-
pointed by the Secretary of State for India in 1905, found that
in some septicuemic cases the bacilli may be present in the blood
in large numbers, two, or even three, days before death, though
this is exceptional.

The above types of the disease are usually classified together
under the heading pestis major, but there also occur mild forms
to which the term pestis minor is applied. In these latter there
may be a moderate degree of swelling of a group of glands,
attended with some pyrexia and general malaise, or there may
!><• little more than slight discomfort. Between such and the
graver types, cases of all degrees of severity are met with.

Experimental Inoculation. — Mice, guinea-pigs, rats, and
rabbits are susceptible to inoculation, the two former being on
the whole most suitable for experimental purposes. After sub-
cutaneous injection there occurs a local inflammatory oedema,
which is followed by inflammatory swelling of the corresponding
lymphatic glands, and thereafter by a general infection. The
lesions in the lymphatic glands correspond in their main
characters with those in the human subject, although usually
at the time of death they have not reached a stage so advanced.
By this method of inoculation mice usually die in 1 to 3 days,
guinea-pigs and rats in 2 to 5 days, and rabbits in 4 to 7 days.
IVst mortem the chief changes, in addition to the glandular
enlargement, are congestion of internal organs, sometimes with
hemorrhages, and enlargement of the spleen ; the bacilli are
numerous in the lymphatic glands and usually in the spleen
(Fig. 146), and also, though in somewhat less degree, throughout
the blood. Infection can also be produced by smearing the
material on' the conjunctiva or mucous membrane of the nose,
and this method of inoculation has been successfully applied in
cases where the plague bacilli are present along with other
virulent organisms, e.g. in sputum along with pneumococci.

482 PLAGUE

Rats and mice can also be infected by feeding either with pure
cultures or with pieces of organs from cases of the disease,
though in this case infection probably takes place through the
mucous membrane of the mouth and adjacent parts, and only to
a limited extent, if at all, by the alimentary canal. Monkeys
also are highly susceptible to infection, and it has been showrn
in the case of these animals that when inoculation is made on
the skin surface, for example, by means of a spine charged with
the bacillus, the glands in relation to the part may show the
characteristic lesion and a fatal result may follow without there

being any noticeable

- - «,f~*>f jf*» .. lesion at the primary

*! r* V ** **** sea*- This fact throws

important light on in-
fection by the skin in
the human subject. The
disease may also extens-
ively affect monkeys by
natural means during an
epidemic.

Paths and Mode of
Infection. — Plague
bacilli may enter the sys-

through small wounds,
cracks, abrasions, etc.,

FIG. 146.-Film preparation of spleen of rat and in snch cases there
after inoculation with the bacillus of plague, IS usually no reaction
showing numerous bacilli, most of which are at the site of entrance,
somewhat plump. m-, . , ,. .

Stained with carbol-thionin-blue. x 1000. ims last tact 1S m

accordance with what

has been stated above with regard to experiments on
monkeys. The path of infection is shown by the primary
buboes, which are usually in .the glands through which the
skin is drained, those in the groin being the commonest site.
Absolute proof of the possibility of infection by the skin is
supplied by several cases in which the disease has been acquired
at post mortem examinations, the lesions of the skin surface
being in the majority of these of trifling nature ; in only two
was there local reaction at the site of inoculation. In most of
these cases the period of incubation has been from two to three
days ; under natural conditions of infection the average period
is within five days. While infection may occur by accidental
inoculation through small wounds of the skin surface, it appears

PATHS AND MODE OF INFECTION 483

in the majority of cases to take place by means of the bites
of fleas. For some time it had been known that plague bacilli
might be found for some time afterwards in the stomach of fleas
allowed to feed on animals suffering from plague, and some
observers, for example Simond, had succeeded in transmitting
the disease to other animals by means of the infected insects.
Most observers, however, had obtained negative results, and it
was only by the work of the Advisory Committee referred to
above,1 that the importance of this means of infection was estab-
lished. By carefully planned experiments, the Committee showed
that the disease could be transmitted from a plague rat to a
healthy rat kept in adjacent cages when fleas were present ;
whereas this did not occur when means were taken to prevent
the access of fleas, though the facilities for aerial infection
were the same. The disease can also be produced by fleas
removed from plague rats and transferred directly to healthy
animals, success having been obtained in fully 50 per cent, of
experiments of this kind. When plague-infected guinea-pigs
are placed amongst healthy guinea-pigs, comparatively few of
the latter acquire the disease when fleas are absent or scanty ;
whereas all of them may die of plague when fleas are numerous.
This result demonstrates the comparatively small part played
by direct contact, even when of a close character. Important
results were also obtained with regard to the mode of infection
in houses where there had been cases of plague. It was found
possible to produce the disease in susceptible animals by means
of fleas taken from rats in plague houses. When animals were
placed in plague houses and efficiently protected from fleas they
remained healthy ; whereas they acquired the disease when the
cages were free to the access of fleas in the neighbourhood.

The following are some of the experiments which were conducted : — A
series of six huts were built which only differed in the structure of their
roofs. In two the roofs were made of ordinary native tiles in which rats
freely lodge ; in two others, flat tiles were used in which rats live, but in
which they have not such facilities for movement as in the first set, and
in the third pair the roof was formed of corrugated iron. Under the
roof in each case was placed a wire diaphragm which prevented rats or
their droppings having access to the hut, but which would not prevent
fleas falling down on to the floor of the hut. The huts were left a
sufficient time to become infected with rats, and then on the floor in
each case healthy guinea-pigs mixed with guinea-pigs artificially infected
with plague were allowed to run about together. In the first two sets
of huts to which fleas had access the healthy guinea-pigs contracted
plague, while in the third set they remained unaffected, though they
were freely liable to contamination by contact with the bodies and excreta

1 See Journal of Hygiene, vi. 421 ; vii. 323.

484 PLAGUE

of the diseased animals. In the third set of huts no infection took place
as long as fleas were excluded, but when accidentally these insects
obtained admission, then infection of the uninoculated animals com-
menced. Other experiments were also performed. In one case healthy
guinea-pigs were suspended in a cage two inches above a floor on which
infected and flea-infested animals were running about. Infection occurred
in the cage, but if the latter were suspended at a distance above the
floor higher than a flea could jump, then no infection took place. Again,
in a hut in which guinea-pigs had died of plague, and which contained
infected fleas, two cages were placed, each containing a monkey. One
cage was surrounded by a zone of sticky material broader than the jump
of a flea. The monkey in this cage remained unaffected, but the other
monkey contracted plague.

Other experiments showed that when plague bacilli were
placed on the floors of houses, they died off in a comparatively
short period of time. After forty-eight hours it was not found
possible to reproduce plague by inoculation with material from
floors which had been grossly contaminated with cultures of the
bacillus. Afterwards, however, animals placed in such a house
might become infected by means of fleas. In all these ex-
periments the common rat-flea of India — Pulex cheopis (Roths-
child)— was used, but it has been shown that this flea, when a
rat is not available, will bite a man. Recent observations show
that not only is plague transferable by means of fleas, but that
this is practically the only method obtaining in natural condi-
tions, with the exception that rats may become infected by eating
the carcases of other animals containing large numbers of
plague bacilli. It is improbable from the experiments made
that plague is transmitted by direct contact even when of a
close nature; in fact, it has been shown that plague-infected
guinea-pigs may suckle their young without the latter acquiring
the disease. The general results show that in the human
subject direct infection by dust and other material through
small lesions of the skin plays, probably, a comparatively small
part in the spread of the disease, fleas apparently being in
nearly all cases the carriers of infection.

The more recent work of the Committee has supplied in-
formation of the highest value with regard to the epidemiology
of the disease ; it has shown, in short, that plague in its epidemic
form is dependent on the epizootic among rats, and with regard
to this some further facts may be given. Plague in Bombay
occurs in two chief species of rats, the mus rattus, the black
house-rat, and mus decumanus, the grey rat of the sewers.
The former, owing to its presence in dwelling-houses, is chiefly
responsible for the transmission of the disease to man ; while the
latter, on account of the large number of fleas which infest

TOXINS, IMMUNITY, ETC. 485

it, is of special importance in maintaining the disease from
season to season. The year may be divided into two portions
— an epizootic season, from December to May inclusive, and a
non-epizootic, from June to November. During the latter
period there are few cases of plague in rats on account of fleas
being scanty; especially is this so in the case of mus rattus.
In fact, in certain villages where this species alone is present,
tin- disease may actually die out at the end of the epizootic
season, and accordingly when plague reappears in these places
this is due to a fresh importation — a fact of great practical
i 1 1 1 { M »rtance. A fresh epizootic first affects chiefly mus decumanus,
and a little later spreads to mus rattux, while a little later still
the disease attacks the human subject in the epidemic form;
in each case fleas form the vehicle of transmission, and an
interval of from ten to fourteen days intervenes between the
outbreak of the epizootic and that of the epidemic. The
proportion of cases of plague in mus decumanus is much higher
than in /////« rattus, for the reason mentioned. It has been
further shown that the bacilli flourish in the stomach of the
flea and are passed in a virulent condition in the faeces, that a
large proportion of fleas removed from plague-infected rats
contain plague bacilli, and that the fleas may remain infective
for a considerable nmnlxsr of days, sometimes for a fortnight.
The repeated contamination of flea-bites by means of the
excrement of fleas seems to be the most likely means of infection
of the human subject.

hi primary plague pneumonia, from a consideration of the
anatomical changes and the clinical facts, the disease may be
said to be produced by the direct passage of the bacilli into the
respiratory passages. Nevertheless there must be certain factors,
still imperfectly understood, which determine the incidence of
this form : as in some epidemics of the highest virulence, plague
pneumonia has been practically absent, though opportunities for
infection by inhalation must have been present. On the other
hand, a case of plague pneumonia is of great infectivity in
producing other cases of plague pneumonia. If we except
infection through the respiratory passages in such cases, it may
In said that direct infection from patient to patient is relatively
uncommon. This is in accordance with the fact that in bubonic
plague the bacilli are not discharged from the unbroken surface
of the body, and are only present in the secretions in severe
cases.

Toxins, Immunity, etc. — As is the case with most organisms
which extensively invade the tissues, the toxins in plague

486 PLAGUE

cultures are chiefly contained in the bodies of the bacteria.
Injection of dead cultures in animals produces distinctly toxic
effects; post mortem haemorrhage in the mucous membrane
of the stomach, areas of necrosis in the liver and at the site
of inoculation, may be present. The toxic substances are
comparatively resistant to heat, being unaffected by an exposure
to 65° C. for an hour. By the injection of dead cultures in
suitable doses, a certain degree of immunity against the living
virulent bacilli is obtained, and, as first shown by Yersin,
Calmette, and Borrel, the serum of such immunised animals
confers a degree of protection on small animals such as mice.
On these facts the principles of preventive inoculation and
serum treatment, presently to be described, depend. It may
also be mentioned that the filtrate of a plague culture possesses
a very slight toxic action, and the Indian Plague Commission
found that such a filtrate has practically no effect in the
direction of conferring immunity.

1. Preventive Inoculation — Ilaffkine's Method. — To prepare
the preventive fluid, cultures are made in flasks of bouillon with
drops of oil on the surface (in India Haffkine employed a
medium prepared by digesting goat's flesh with hydrochloric
acid at 140° C. and afterwards neutralising with caustic soda).
In such cultures stalactite growths (vide supra) form, and the
flasks are shaken every few days so as to break up the stalactites
and induce fresh crops. The flasks are kept at a temperature
of about 25° C., and growth is allowed to proceed for about
six weeks. At the end of this time sterilisation is effected by
exposing the contents of the flasks to 65° C. for an hour ;
thereafter carbolic acid is added in the proportion of '5 per cent.
The contents are well shaken to diffuse thoroughly the sediment
in the fluid, and are then distributed in small sterilised bottles
for use. The preventive fluid thus contains both the dead
bodies of the bacilli and any toxins which may be in solution.
It is administered by subcutaneous injection, the dose, which
varies according to the " strength," being on an average about
7 '5 c.c. Usually only one injection is made, sometimes two,
though the latter procedure does not appear to have any
advantage. The method has been systematically tested by
inoculating a certain proportion of the inhabitants of districts
exposed to infection, leaving others uninoculated, and then
observing the proportion of cases of disease and the mortality
amongst the two classes. The results of inoculation, as attested
by the first Indian Commission, have been distinctly satisfactory.
For although absolute protection is not afforded by inoculation,

SERUM DIAGNOSIS 487

both the proportion of cases of plague and the percentage
mortality amongst these cases have been considerably smaller
in the inoculated, as compared with the uninoculated. Protec-
tion is not established till some days after inoculation, and lasts
for a considerable number of weeks, possibly for several months
( U;iiuu-i •man). In the .Punjab during the season 1902-3 the
case incidence among the inoculated was 1'8 per cent., among
the uninoculated 7 '7 per cent., while the case mortality was 23*9
iiiid 60*1 per cent, respectively in the two classes, the statistics
U'iiig taken from villages where 10 per cent, of the population
and upwards had been inoculated.

2. Anti-plague Sera. — Of these, two have been used as therapeutic
agents, namely, that of Yersin and that of Lustig. Yersin's serum is
prepared by injections of increasing doses of plague bacilli into the
horse. In the early .stages of immunisation dead bacilli are injected
subcutaneously, thereafter into the veins, arid, finally, living bacilli are
injected intravenously. After a suitable time blood is drawn oft' and
the serum is preserved in the usual way. Of this serum 10 to 20 c.c.
are used, and injections are usually repeated on subsequent days.
Lustig's serum is prepared by injecting a horse with repeated and
increasing doses of a substance derived from the bodies of plague bacilli,
probably in great part nucleo-proteid. Masses of growth are obtained
from the surface of agar cultures, and are broken up and dissolved in a
1 per cent, solution of caustic potash. The solution is then made slightly
acid by hydrochloric acid, when a bulky precipitate forms ; this is
collected on a filter and dried. For use a weighed amount is dissolved
in a weak solution of carbonate of soda and then injected. The serum
is obtained from the animal in the usual way. Extensive observations
with both of these sera show that neither of them can be considered
a powerful remedy in cases of plague, though in certain instances
distinctly favourable results have been recorded. The Indian Com-
mis.sion, however, came to the conclusion "that, on the whole, a certain
amount of advantage accrued to the patients both in case of those
injected with Yersin's serum and of those injected with Lustig's serum."
It may also be mentioned that the Commission found, as the result of
t-xperiments, that Yersin's serum modified favourably the course of the
disease in animals, whereas Lustig's serum had no such effect.

3. Serum Diagnosis. — Specific agglutinins may appear in the blood of
patients suffering from plague, as also they do in the case of animals
immunised against the plague bacillus. It is to be noted, however, that
in clinical cases the reaction is not invariably present, the potency of
the serum is not of high order, and the carrying out of the test is
complicated by the natural tendency of the bacilli to cohere in clumps.
For the last reason the macroscopic (sedimentation) method is to be
preferred to the microscopic (p. 120). A suspension of plague bacilli is
made by breaking up a young agar culture in "75 per cent, sodium
chloride solution ; the larger ttocculi of growth are allowed to settle, and
the fine, supernatant emulsion is employed in the usual way. According
to the results of the German Plague Commission and the observations of
Cairns, made during the Glasgow epidemic, it may be said that the
reaction is best obtained with dilutions of the serum of from 1 : 10 to

488 MALTA FEVER

1 : 50. Cairns found that the date of its appearance is about a week
after the onset of illness, and that it usually increases till about the end
of the sixth week, thereafter fading off'. It is most marked in severe
cases characterised by an early and favourable crisis, less marked in
severe cases ultimately proving fatal, whilst in very mild cases it is
feeble or may be absent. The method, if carefully applied, may be of
service under certain conditions ; but it will be seen that its use as a
means of diagnosis is somewhat restricted.

Methods of Diagnosis. — Where a bubo is present a little of
the juice may be obtained by plunging a sterile hypodermic
needle into the swelling. The fluid is then to be examined
microscopically, and cultures on agar or blood serum should be
made by the successive stroke method. The cultural and
morphological characters are then to be investigated, the most
important being the involution forms on salt agar and the
stalactite growth in bouillon, though the latter may not always
be obtained with the plague bacillus : the pathogenic properties
should also be studied, the guinea-pig being on the whole most
suitable for subcutaneous inoculation. In many cases a diagnosis
may be made by microscopic examination alone, as in no known
condition other than plague do bacilli with the morphological
characters of the plague bacillus occur in large numbers in the
lymphatic glands. The organism may be obtained in culture
from the blood in a considerable proportion of cases by with-
drawing a few cubic centimetres and proceeding in the usual
manner. On the occurrence of the first suspected case, every
care to exclude possibility of doubt should be used before a
positive opinion is given.

In a case of suspected plague pneumonia, in addition to
microscopic examination of the sputum, the above cultural
methods along with animal inoculation with the sputum should
be carried out; subcutaneous injection in the guinea-pig and
smearing the nasal mucous membrane of the rat may be recom-
mended. Here a positive diagnosis should not be attempted by
microscopic examination alone, especially in a plague-free dis-
trict, as bacilli morphologically resembling the plague organism
may occur in the sputum in other conditions.

MALTA FEVER.

Synonyms — Mediterranean Fever: Rock Fever of Gibraltar:
Neapolitan Fever, etc.

This disease is of common occurrence along the shores of the
Mediterranean and in its islands. Since its bacteriology has

MICROCOCCUS MELTTENSIS 489

been worked out, it has been found to occur also in India,
China, South Africa, and in some parts of North and South
America, its distribution being much wider than was formerly
supposed. Although from its symptomatology and pathological
anatomy it had been recognised as a distinct affection, and was
known under various names, its precise etiology was unknown
till the publication of the researches of Colonel Bruce in 1887.
From the spleen of patients dead of the disease he cultivated a
characteristic organism, now known as the Jficrococcus melitensis,
and by means of inoculation experiments established its causal
relationship to the disease. Wright and Semple applied the
agglutination test to the diagnosis of the disease, while within
recent years the mode of spread of the disease has been fully
studied by a Commission, and it has been demonstrated that
goat's milk is the chief means of infection.

The duration of the disease is usually long — often two or
three months, though shorter and much longer periods are met
with. Its course is very variable, the fever being of the con-
tinued type \\ith irregular remissions. In addition to the usual
symptoms of pyrexia, there occur profuse perspiration, pains
and sometimes swellings in the joints, occasionally orchitis,
whilst constipation is usually a marked feature. The mortality
is low — about 2 i>er cent. (Bruce).

In fatal cases the most striking post mortem change is in the
spleen. This organ is enlarged, often weighing slightly over a
pound, and in a condition of acute congestion ; the pulp is soft
and may be diffluent, and the Malpighian bodies are swollen and
indistinct. In the other organs the chief change is cloudy
swelling ; in the kidneys there may be in addition glomerular
nephritis. The lymphoid tissue of the intestines shows none of
the changes characteristic of typhoid fever.

Micrococcus melitensis. — This is a small, rounded, or slightly
oval organism about '4 /x in diameter, which is specially abundant
in the spleen. It usually occurs singly or in pairs, but in
cultures short chains are also met with (Fig. 147). (Durham
has shown that in old cultures kept at the room temperature
bacillary forms appear, and we have noticed indications of such
in comparatively young cultures : the usual form is, however,
that of a coccus.) It stains fairly readily with the ordinary
basic aniline stains, but loses the stain in Gram's method. It is
Lr< m-rally said to be a non-motile organism. Gordon, however,
is of a contrary opinion, and has recently demonstrated that it
possesses from one to four nagella, which, however, are difficult
to stain. In the spleen of a patient dead of the disease it

490

MALTA FEVER

occurs irregularly scattered through the congested pulp ; it may
also be found in small numbers post mortem in the capillaries of
various organs. It may be cultivated from the blood during life
in a considerable proportion of cases ; for this purpose 5 to 10 c.c.
of blood should be withdrawn from a vein and distributed in
small flasks of bouillon. The micrococcus was found by the
members of the Commission in the urine of Malta fever patients
in. 10 per cent, of the cases examined ; it was sometimes scanty,
but sometimes present in large numbers. It has also occasionally
been obtained from the faeces.

Cultivation. — This can usually readily be effected by making
stroke cultures on agar tubes from the spleen pulp and incub-
ating at 37° C. The
colonies, which are usually
not visible before the
third or fourth day, ap-
pear as small round discs,
;' l% " j. slightly raised and of some-
what transparent appear-
| ance. The maximum size
— 2 to 3 mm. in diameter
'"I — is reached about the
ninth day; at this period
by reflected light they
appear pearly white, while
by transmitted light they
have a yellowish tint in the
centre, bluish white at the
periphery. A stroke culture
shows a layer of growth
of similar appearance with

somewhat serrated margins. Old cultures assume a buff tint.
The optimum temperature is 37° C., but growth still occurs down
to about 20° C. On gelatin at summer temperature growth is
extremely slow — after two or three weeks, in a puncture culture,
there is a delicate line of growth along the needle track and a
small flat expansion of growth on the surface. There is no
liquefaction of the medium. In bouillon there occurs a general
turbidity with flocculent deposit at the bottom ; on the surface
there is no formation of a pellicle. The reaction of the media
ought to be very faintly alkaline, as marked alkalinity interferes
with the growth; a reaction of + 10 (p. 34) has been found
very suitable. On potatoes no visible growth takes place even
at the body temperature, though the organism multiplies to a

FIG. 147. — Micrococcus meliteusis, from a

two days' culture on agar at 37° C.

Stained with fuchsin. x 1000.

MODE OF SPREAD OF THE DISEASE 491

certain extent. Outside the body the organism has considerable
[•owns of vitality, as it has been found to survive in a dry con-
dition in dust and clothing for a period of two months.

Relations to the Disease. — There is in the first place ample
evidence, from examination of the spleen, both post mortem and
during life, that this organism is always present in the disease.
The exj>eriments of Bruce and Hughes first showed that by
inoculation with even comparatively small doses of pure cultures
the disease could be produced in monkeys, sometimes with a
fatal result. And it has now been fully established that inocula-
tion with the minutest amount of culture, even by scarification,
leads to infection both in monkeys and in the human subject.

liabbits, guinea-pigs, and mice are insusceptible to inoculation
by the ordinary method. Durham, by using the intracerebral
method of inoculation, however, succeeded in raising the virul-
ence, so that the organism is capable of producing in guinea-
pigs on intraperitoneal injection illness with sometimes a fatal
result many weeks afterwards. An interesting point brought
out by these experiments is that, in the case of animals which
survive, the micrococcus may be cultivated from the urine several
months after inoculation.

Mode of Spread of the Disease. — The work of the recent
Commission has resulted in establishing facts of the highest
importance with regard to the spread of the disease. In the
course of investigations Zammitt found that the blood of many
of the goats agglutinated the micrococcus melitensis, and
Horrocks obtained cultures of the organism from the milk.
Further observations showed that agglutination was given in
the case of 50 per cent, of the goats in Malta, whilst the organism
was present in the milk in 10 per cent. Sometimes the organism
was present in enormous numbers, and in these cases the animal
usually appeared poorly nourished, whilst the milk had a some-
what serous character. In other cases, however, the organism
was found when the animals appeared healthy, and there was
no physical or chemical change in the milk. It was also
determined that the organism might be excreted for a period
of two to three months before any notable change occurred in
the milk. Agglutination is usually given by the milk of infected
animals, and this property was always present when the micro-
coccus was found in the milk. It was, moreover,, found that
monkeys and goats could be readily infected by feeding them
with milk containing the micrococcus, the disease being contracted
by fully 80 per cent, of the monkeys used. It was therefore
rendered practically certain that the human subject was infected

492 MALTA FEVER

by means of such milk, and the result of preventive measures
by which milk was excluded as an article of dietary amongst
the troops in Malta has fully borne out this view. After such
measures were instituted, the number of cases in the second
half of 1906 fell to 11 per thousand, as contrasted with 47
per thousand in the corresponding part of the preceding year ;
further successful results have followed. The various facts with
regard to the epidemiology of the disease have thus been cleared
up. For example, it is more prevalent in the summer months,
when more milk is consumed, and there is a larger proportion
of cases amongst those in good social position, the officers, for
example, suffering more in proportion than the privates. Another
interesting fact, pointed out by Horrocks, is that the disease has
practically disappeared from .Gibraltar since the practice of im-
porting goats from Malta has stopped.

The work of the Commission, so far as it has gone, has been
to exclude other modes of infection than the ingestion of infected
milk as being of practical importance ; if the disease is conveyed
by contact at all, this is only when the contact is of an intimate
character, and even then it is probably of rare occurrence. Al-
though numerous patients suffering from the disease come to
England, there is no known case of fresh infection arising under
natural conditions.

There is distinct evidence that the disease may be acquired by
inoculation through small lesions in the skin, and this method
is probably not infrequent amongst those who handle infected
milk. It has been shown that the organism may remain alive
in the bodies of mosquitoes for four or five days, and possibly
these insects may occasionally be the means of carrying the
disease ; there is no evidence, however, that this takes place
to any extent.

Agglutinative Action of Serum. — The blood serum of patients
suffering from Malta fever possesses the power of agglutinating
the micrococcus melitensis in a manner analogous to what has
been described in the case of typhoid fever. The reaction
appears comparatively early, often about the fifth day, and may
be present for a considerable time after recovery — sometimes
for more than a year. Distinct agglutination with a 1 : 30
dilution of the serum in half an hour may be taken as a positive
reaction, sufficient for diagnosis. The reaction is, however,
usually given by much higher dilutions, e.g. 1 : 500, and even
higher. It is to be noted that normal serum diluted 1 : 5 may
produce some agglutination. As regards relation to prognosis,
the observations of Birt and Lamb and of Bassett-Smith

METHODS OF DIAGNOSIS 493

have given results analogous to those obtained in typhoid
(p. 372).

The Commission has found that vaccination with dead cultures
of the micrococcus confers a certain degree of protection amongst
those exposed to the disease. As a rule two injections were
made, 200-300 million cocci being the dose of the first injection,
and about 400 million the dose of the second. The use of vaccines
has also been carried out in the treatment of the disease, but
the observations are not sufficiently numerous to allow a definite
Muit'iiM'iit to be made as to its value.

Methods of Diagnosis. — During life the readiest means of
diagnosis is supplied by the agglutinative test just described
(for technique, vide p. 118).

Cultures are most easily obtained from the spleen either
during life or post mortem. Inoculate a number of agar tubes
by successive strokes and incubate at 37° C. Film preparations
should also be made from the spleen pulp and stained with
carbol-thionin-blue or diluted carbol-fuchsin (1 : 10). Cultures
may sometimes be obtained from the blood by the usual
methods.

CHAPTER XX.

DISEASES DUE TO SPIROCHAETES— THE RELAPSING
FEVERS, SYPHILIS, AND FRAMBCESIA.

THE diseases produced by spirochaetes — spirilloses or spiro-
chaetoses — fall into two main groups, one represented by the
human spirillar fevers and the corresponding affections of various
animals, and the second having as its two chief members
syphilis and yaws, though to the organisms of these diseases
various spirochaetes found in ulcerative and gangrenous con-
ditions seem to be closely related. The members of the first
group are essentially blood infections, and the organisms are in
most, if not in all cases, transmitted by blood-sucking ecto-
parasites ; in the second group the organisms are primarily
tissue-parasites, blood infection when it occurs being a later
phenomenon, and infection would appear to occur by direct
contact. As regards general morphology, staining reactions,
conditions of growth and culture, the various spirochaetes
present certain common characters, and, as already stated, it
is still uncertain whether they are to be regarded as bacteria
or as protozoa, though the balance of opinion is now distinctly
in favour of the latter.

RELAPSING FEVER AND AFRICAN TICK FEVER.

At a comparatively early date, namely in 1873, when prac-
tically nothing was known with regard to the production of
disease by bacteria, a highly characteristic organism was dis-
covered by Obermeier in the blood of patients suffering from
relapsing fever. This organism is usually known as the
spirillum or spirochcete Obermeieri, or the spirillum of relapsing
fever. He described its microscopical characters, and found
that its presence in the blood had. a definite relation to the
time of the fever, as the organism rapidly disappeared about
the time of the crisis, and reappeared when a relapse occurred.

41)4

THE SPIROCH^ETE OBERMETERI 495

His observations were fully confirmed, and his views as to its
causal relationship to the disease have been established as
correct.

Within recent years relapsing fever has been carefully studied
in different parts of the world, and the relationships of the
organisms have been the subject of much investigation and
discussion. This question will be referred to again below.
Recently also it has been shown that the so-called " tick fever "
prevalent in Africa is due to a spirochsete of closely similar
character, and results of the highest importance have been
established with regard to the part played by ticks in the
transmission of the disease. As a matter of convenience, we
shall give the chief facts regarding these diseases separately.
It has also been shown that spirillar diseases or " spirilloses,"
as they are called, are widespread amongst vertebrates; they
have been described, for example, in geese by Sacharoff, in
fowls by Marchoux and Salimbeni, in oxen and sheep by
Theiler, and in bats by Nicolle and Comte, and it is interesting
to note that in the case of the spirilloses of oxen and fowls the
infection is transmissible by means of ticks.

Characters of the Spirochaete. — The organisms as seen in
the blood during the fever are delicate spiral filaments which
have a length of from two to six times the diameter of a red blood
corpuscle. They are, however, exceedingly thin, their thickness
being much less than that of the cholera spirillum. They show
several regular sharp curves or windings, of number varying
according to the length of the organisms, and their extremities are
finely pointed (Fig. 148). There are often to be seen in the
spirals, portions which are thinner and less deeply stained than
tlio rest, and which suggest the occurrence of transverse division.
They are actively motile, and may be seen moving quickly
across the microscopic field with a peculiar movement which
is partly twisting and partly undulatory, and disturbing the
blood corpuscles in their course.

They stain with watery solutions of the basic aniline dyes,
though somewhat faintly, and are best coloured by the
Romanowsky method or one of its modifications. When thus
stained they usually have a uniform appearance throughout, or
may be slightly granular at places, but they show no division
into short segments. They lose the stain in Gram's method.
There is no evidence that they form spores.

Novy found that the spirochaete of American relapsing fever
remained alive and virulent in defibrinated rats' blood for
forty days. He also succeeded, by Levaditi's method (p. 503),

496

RELAPSING FEVER

in obtaining cultures in collodion sacs containing rats' blood
which were placed in the peritoneum of rats. By this method
cultures were maintained for many generations ; the organisms
were still virulent though the resulting infection was rather less
intense than at first. The spirochsetes are readily killed at a
temperature of 60° C., but may be exposed to 0° C. without
being killed. Novy and Knapp have found that there is a single
flagellum at one end of this organism.

Relations to the Disease. — -In relapsing fever, after a period
of incubation there occurs a rapid rise of temperature which
lasts for about five to seven days. At the end of this time a

crisis occurs, the tempera-
ture falling quickly to
normal. In the course
of about other seven
days a sharp rise of
temperature again takes
place, but on this occa-
sion the fever lasts a
shorter time, again sud-
denly disappearing. A
second or even third
relapse may occur after
a similar interval. The
organisms begin to appear
in the blood shortly
before the onset of the
pyrexia, and during the
rise of temperature rapid-
ly increase in number.
They are very numerous

during the fever, a large number being often present in
every field of the microscope when the blood is examined at
this stage. They begin to disappear shortly before the crisis :
after the crisis they are entirely absent from the circulating
blood. A similar relation between the presence of the organ-
isms in the blood and the fever is found in the case of the
relapses. Munch in 1876 produced the disease in the human
subject by injecting blood containing the spirochsetes, and this
experiment has been several times repeated with the same result.
Additional proof that the organism is the cause of the disease
has been afforded by experiments on animals. Carter in 1879
was the first to show that the disease could be readily produced in
monkeys, and his experiments were confirmed by Koch. In such

FIG. 148. — Spiroohretes of relapsing fever in
human blood. Film preparation. (After
Koch.) See also Plate IV., Fig. 18.
x about 1000.

IMMUNITY 497

experiments the blood taken from patients and containing the
spirochaetes was injected subcutaneously. In the disease thus
produced there is an incubation period which usually lasts about
three days. At the end of that time the organisms rapidly appear
in the blood, and shortly afterwards the temperature quickly
rises. The period of pyrexia usually lasts for two or three days,
and is followed by a marked crisis. As a rule there is no relapse,
but occasionally one of short duration occurs.1 White mice and

1 •'!«.. 119.— Spirochsete Obernieieri in blood of infected mouse.
xlOOO.

rats are also susceptible to infection. In the former animals
the disease is characterised by several relapses, in the latter there
is, however, no relapse.

Immunity.— Metclmikoff found that during the fever the
spirocluL-tes were practically never taken up by the leucocytes in
the circulating blood, but that at the time of the crisis, on dis-
;ijipi'jiring from the blood, they accumulated in the spleen and
were ingested in large numbers by the microphages or poly-
Morris, Pappeuheimer, and Flournoy, in their experiments on monkeys
with the organism of American relapsing fever, found that several relapses
occurred.

498 RELAPSING FEVER

morphonuclear leucocytes. Within these they rapidly under-
went degeneration and disappeared. It is to be noted in this
connection that swelling of the spleen is a very marked feature
in relapsing fever. These observations were confirmed by
Soudakewitch, who also found that when the disease was pro-
duced in splenectomised monkeys (cercocebus fuliyinosus) the
spirochaetes did not disappear from the blood at the usual time,
but rather increased in number, and a fatal result followed on
the eighth and ninth days respectively. Recent observations,
however, indicate that, as in the case of so many other diseases,
the all-important factor in the destruction of the organisms is
the development of antagonistic substances in the blood. Lamb
found in the case of the monkey (macacus radiatus) that the
removal of the spleen of an animal rendered immune by an
attack of the disease did not render it susceptible to fresh
inoculation, and attributed the immunity to the presence of
bactericidal bodies in the serum. He found, for example, that
in vitro the serum of an immune animal brought the movements
of the spirilla to an end, clumped them, and caused their dis-
integration; and further, that when the spirochaetes and the
immune serum were injected in one case into a fresh monkey no
disease developed. In opposition to Soudakewitch, Lamb found
that with a monkey from which the spleen had been removed
death did not occur after it was inoculated with the spirochaetes.
Observations by Sawtschenko and Milkich, Novy and Knapp,
and others, also show that in the course of infection there are
developed anti-substances of the nature of immune-bodies, with
protective properties, and agglutinins. Novy and Knapp pro-
duced a "hyper-immunity " in rats by repeated injections of blood
containing the spirochaetes, and found that the serum of such
animals had a markedly curative effect, and could cut short the
disease in rats, mice, and monkeys. The course of events in the
human disease might be explained by supposing that immunity
of short duration is produced during the first period of pyrexia,
but that it does not last until all the organisms have been
destroyed, some still surviving in internal organs or in tissues
where they escape the bactericidal action of the serum. With
the disappearance of the immunity the organisms appear in the
blood, the relapse being, however, of shorter duration and less
severe than the first attack. This is repeated till the immunity
lasts long enough to allow all the organisms to be killed.

Varieties. — As already stated, relapsing fever has been studied
in different parts of the world, and, apart from the African tick
fever, European, Asiatic, and American types have been dis-

VARIETIES 499

tinguished. Differences have been made out with regard to
clinical features, pathogenic effects, and immunity reactions.
Of these the last mentioned are probably the most important.
It has been shown, for example, by the work of Novy, Strong,
and Mackie, that the American spirochaete is probably a distinct
species, as animals immunised against it are still susceptible to
infection by the European and Asiatic organisms, and vice versa.
The relationship between the two latter is certainly closer, and
no distinct immunity differences have been established. Re-
lapsing fever in Asia is evidently a much more severe disease
than in Europe ; Mackie gives the mortality in Bombay at the
comparatively high figure of 38 per cent. But differences in this
respect, as well as in pathogenic effects, may simply depend on
variations in virulence. At present no definite statement can be
made on this point.

The fact that tick fever and other spirilloses are con-
veyed by the bites of insects makes it extremely probable that
relapsing fever is transmitted in this way. At first the bed-bug
was believed to be the vehicle of transmission, and the experi-
ments of Karlinski and of Tictin, which showed that the spiro-
clia-tes might remain alive and virulent in the body of this
insect for some time after it had sucked the blood of a patient,
lent some support to this view. Attempts to transmit the
disease by means of the bites of bugs were, however, generally
unsuccessful ; Mackie produced the disease in only one out of
six monkeys used for this purpose, though large numbers of bugs,
which had bitten relapsing fever patients, were used. On in-
vestigating an epidemic of the disease, however, he obtained
a considerable amount of evidence on epidemiological grounds
that the disease was carried by the body louse. He also found
that the spirochaites in the blood which had been sucked under-
went great multiplication about three days afterwards, and
formed large tangled masses in the stomach contents. The
view that the louse is the agent of transmission of the human
disease is strongly supported by the experiments of Manteufel,
who was able to transmit infection from rat to rat in nearly
60 per cent, of the experiments made, whereas he obtained
only negative results by means of bugs. Further observations
are still necessary.

African Tick Fever.

The disease long known by this name as prevalent in Africa
has also been shown to be caused by a spirillum or spirocha •!<•.

500 AFRICAN TICK FEVER

Sp. Duttoni. Organisms of this nature had been seen in the blood
of patients in Uganda by Greig and Nabarro in 1903, and Milne
and Ross in the end of 1904 recorded a series of observations
which led them to the conclusion that tick fever was due to a
spirochyete. It is, however, chiefly owing to the work of Button
and Todd in the Congo Free State, on the one hand, and of Koch
in German East Africa, on the other, that our knowledge of this
disease has been thoroughly established.

FIG. 150. — Film of human blood containing spiroehiete of tick fever,
x lOOO.i

The following are the chief facts regarding this fever :
Clinically the fever closely resembles relapsing fever, but the
periods of fever are somewhat shorter, rarely lasting for more
than two or three days. It is seldom attended with a fatal result
unless in patients debilitated by other causes. The organisms
are considerably fewer in the blood than in the European re-
lapsing fever, and sometimes a careful search may be necessary
before they are found. Morphologically, they are said to be

1"We are indebted to Lieut. -Col. Sir William Leishman, R.A.M.C.,
for the preparations from which Figs. 149-151 were taken.

ATKLCAN TICK FEVER 501

practically identical, although Koch thought that the organisms
in tick fever tended on the whole to be slightly longer; the
average length may )>e said to be 15 to 35 JJL. Button and Todd
showed that it was possible to transmit the disease to certain
monkeys (cercopitheci) by means of ticks which had been allowed
to bite patients suffering from the disease, the symptoms in
these animals appearing about five days after inoculation. The
disease thus produced is characterised by several relapses, and

o • - • -

FK.. 151. — Spirillum of human tick fever (Spirillum Duttoni) in
blood of infected mouse, x 1000.

often leads to a fatal result. In one case they produced the
disease by means of young ticks hatched from the eggs of ticks
which had been allowed to suck the blood of fever patients, and
they came to the conclusion that the spirochaetes were not simply
carried mechanically by the ticks, but probably underwent some
cycle of development in the tissues of the latter. The species of
tick concerned is the onutkodonu moubata. These results were
confirmed and extended by Koch. He found that after the ticks
had been allowed to suck the blood containing the organisms,
could be found for a da or two \n the stomachs of ^he

502 AFRICAN TICK FEVER

insects. After this time they gradually disappeared from the
stomach, but were detected in large numbers in the ovaries of
the female ticks, where they sometimes formed felted masses.
He also traced the presence of the spirochsetes in the eggs laid by
the infected ticks, and in the young embryos hatched from them.
On the other hand, Leishman has failed to find any evidence of
spirochsetes in the tissues of ticks later than ten days after
ingestion of blood containing them, or in the ova laid by them,
or in the young ticks when hatched, though these were proved
by experiment to be infective. After ingestion of the blood by
the. ticks, he found that morphological changes occurred in the
spirochsetes, resulting in the formation of minute chromatin
granules which traverse the walls of the intestine and are taken
up by the cells of the Malpighian tubules ; they also penetrate
the ovaries and may be found in large numbers within the ova.
Similar granules are to.be seen in the Malpighian tubules of the
embryo ticks, where they are also found in the subsequent stages
of their life. He has abundantly proved that infection of
animals may be produced by inoculation with crushed material
containing the granules but no spirocluvtes. He accordingly
considers that the granules in question probably represent a
phase in the life history of the parasite, and that infection probably
occurs by inoculation of the skin with the chromatin granules
voided in the Malpighian secretion and not by unaltered
spirochsetes from the salivary glands. It is also interesting to
note that Balfour has found similar granules in ticks (argas
persicus) infected with tpirochaste gallinarwn,

Koch also made extensive observations on the ticks in Ger-
man East Africa, and found that of over six hundred examined
11 per cent, of these insects along the main caravan routes con-
tained spirilla, and in some localities almost half of the ticks
were infected. In places removed from the main lines of com-
merce he still found them, though in smaller number. It has
also been demonstrated that in some places the ticks are found
to be infected with the spirilla although the inhabitants do not
suffer from tick fever, a circumstance which is probably due to
an acquired immunity against the disease.

It is now generally considered that the sp. Duttoni is a
species distinct from, though closely allied to, the organisms of
the relapsing fevers described above. We have mentioned some
differences in the clinical characters of the diseases, and there
are also differences in the pathogenic effects of the organisms on
inoculation. The sp. Duttoni, for example, produces a much
more severe disease in monkeys, and is pathogenic to more

SYPHILIS 503

«4>ecies of the laboratory animals than the sp. Obermeieri. The
most important differences are however brought out by immunity
reactions. It was shown by Breinl that the immunity produced
by the sp. Obermeieri did not protect against the sp. Duttoni,
and that the converse also held good ; and it has since been
established that a similar difference obtains between the sp.
Duttoni and the organisms of the Asiatic and American varieties
of relapsing fever. Corresponding results are obtained on
testing the various serum reactions in vitro.

Levaditi has succeeded in obtaining cultures of the spirochsete
of tick fever by inoculating sacs filled with monkey's serum,
heated at 70° C., and placing the sacs in the peritoneal cavity of
a rat or rabbit ; when opened at the end of five to seven days,
the sacs were found to contain an abundant growth of spiro-
chaetes, some of which were of unusually great length. Growth
was maintained in similar sub-cultures, and the virulence was
well preserved.

SYPHILIS,

Up till quite recent times practically nothing of a definite
nature was known regarding the etiology of syphilis. Most
interest for a long time centred around the observations of
Lustgarten, who in 1884 described a characteristic bacillus,
both in the primary sore and in the lesions in internal organs.
This organism occurred in the form of slender rods, straight, or
slightly bent, 3 to 4 /u, in length, often forming little clusters
either within cells or lying free in the lymphatic spaces ; it took
up basic aniline dyes with some difficulty, but was much more
easily decolorised by acids than the tubercle bacillus. The
etiological relationship of the organism to the disease was,
however, not generally accepted, and in view of the recent work
on syphilis, the organism cannot be regarded as having any
pathological importance.

Spirochaete pallida. — An entirely new light has been thrown
on the etiology of the disease by the work of Schaudinn and
Hoffmann which appeared in 1905. Since their first publication
a great amount of work has been undertaken in order to test
their conclusions, and the results have been of a confirmatory
nature. These observers found in cases of syphilis an organism
to which they gave the name spirochcete pallida ; it now also
goes by the name spironema pallidum. As described by them,
it is a minute spiral-shaped organism, showing usually from six
to eight curves, though longer forms are met with ; the curves

504 SYPHILIS

are small, comparatively sharp, and regular (Figs. 152, 153). It
may be said to measure 4 to 1 4 /A in length, while it is extremely
thin, its thickness being only '25 /x. In a fresh specimen, say a
scraping from a chancre suspended in a little salt solution, the
organism shows active movements, which are of three kinds —
rotation about the long axis, gliding movements to and fro,
and movements of flexion of the whole body. The ends are
pointed and tapering. Its detection is comparatively difficult,
as the organism is feebly refractile, and more difficult to see than
most other organisms ; the movement of small particles in the
vicinity, however, is of asistance in finding it. The use of the

FIGS. 152 and 153. — Film preparations from juice of hard chancre
showing spirochaete pallida, — Giemsa's stain. xlOOO. (From pre-
parations by Dr. A. MacLennan.)

parabolic sub-stage condenser (p. 93) is of great service in
searching for the organism.

In ulcerated syphilitic lesions other organisms are, of course,
present, and not infrequently another spiral organism, to which
the name spirochaete refringens has been given. This organism
is usually somewhat longer, and is distinctly thicker than the
spirochsete pallida. As the name implies, it is more highly
refractile, and it is much more easily detected than the latter
organism; its curves also are more open and much, less regular,
and they vary in their appearance during the movements. In
stained films (see p. 115), the differences between the organisms
come out more distinctly, as can be gathered from the accom-
panying photograph (Fig. 156). The spirochaete pallida by the
Giemsa stain is coloured somewhat faintly, and of reddish tint,
whilst the regular spiral twistings are preserved ; the spirochsete
refringens shows flatter, wave-like bends, and, like other organ-
isms, is stained of a bluish tint. By using Loffler's stain for the

SPIROCH^ETE PALLTDA 505

flagella <>f bacteria, Schaudinn was able to demonstrate a single
delicate fiagelluui at each pole of the spirochajte pallida, while
no undulating membrane could be detected; on the other hand,
several other species, including the spirochyete refringens, showed
a distinct undulating membrane. Two fiagella at one pole of
the spirorlui'te pallida were also seen, an appearance which
Schaudinn thought might represent the commencement of
longitudinal fission.

The number of publications with regard to the distribution of
the spiroclut'tf pallida is already very large, and a summary of

FIG. ].">!. Film preparation fnun juice of hard chancre showing
spiroi -ha-tr pallida. (iiemsa's stain, x 2000. (From a preparation
liy Dr. Haswell Wilson.)

tin- results may be given. In the primary sore and in the related
lymphatic glands, the juice of which can be conveniently
obtained by means of a hypodermic syringe, the organism has
l>eeii found in a very large majority of cases. It has been also
obtained in the papular and roseolar eruptions, in condylomata
and mucous patches — in fact, one may say generally, in all the
primary and secondary lesions. It has been obtained from
the splri'ii during life, and on a few occasions, e.g. by Schaudinn,
a No from the blood during life in secondary syphilis. In the
congenital form of the disease the organism may be present in
large numbers (Plate II., Fig. 6), as was first shown by Buschke
and Fischer, and by Levaditi. In the pemphigoid bullae, in

506 SYPHILIS

the blood, in the internal organs, the liver, lungs, spleen, supra-
renals, and even in the heart its detection may be comparatively
easy, owing to the large numbers present (Fig. 155). It has
been generally supposed that tertiary syphilitic lesions are non-
infective, and the results of the earlier observations on the
spirochsete pallida were apparently in accordance with this
view, as they gave negative results. More prolonged search
has, however, shown that the organism may occur in tertiary
lesions also. It has been found to be present in the peripheral
parts of gummata, especially at an early stage of their forma-

FIG. 155. — Section of spleen from a case of congenital syphilis,,
showing several examples of spirochrete pallida ; Levaditi's method,
x 2000.

tion ; and the observations of Schmorl, Benda, J. H. Wright
and others show that it is often to be found in syphilitic
aortitis, sometimes occurring in considerable numbers in the
thickened patches. That the spirochsete may persist in the
body for a very long time after infection, has been abund-
antly shown by different observers ; in one case, for example,
its presence was demonstrated sixteen years after the primary
lesion. It can readily be demonstrated in sections of the organs
by the method described on p. 112. In such preparations large
numbers of spirochsetes, chiefly extra-vascular in position, can
be seen, and many may occur in the interior of the more highly
specialised cells, for example, liver-cells ; in many cases examina-

CULTIVATION OF THE SPIROCH^ETE PALLTDA 507

tion has been made within so short a^" period after the death of
the child as to practically exclude the possibility of contamina-
tion from without. It also abounds sometimes on mucous
surfaces, e.</. of the bladder and intestine in cases of congenital
syphilis. The enormous numbers of the organism which may
be present in a well preserved condition in macerated foetuses
render it probable that the organism may multiply in the dead
tissues under anaerobic conditions. Although various organisms
may be associated with it in the lesions of the skin or mucous
membranes, then- is agreement amongst observers that this
organism occurs alone in syphilitic lesions where the entrance
of bacteria, etc., from outside is excluded. The high per-
centage of cases in which it is found would, in view of the
(lilliculty in detecting it, point to
its invariable presence, and, as a ^|J

matter of fact, Schaiidinn in his

last .series of cases, numbering over jm •*

seventy, found it in all. Shortly
after the discovery of the organism, .
Metclmikott' was able to detect it in m •

the lesions produced in monkeys by
inoculation with material derived
from syphilitic sores, and his obser-
vations have since been confirmed.
Another question of considerable

importance is, as to whether this Fj(; K((. Bpiro6hlete ren.ingens
organism has been found in other In film preparation from a ewe
conditions. Observations show of balanilis. xlOOO.
that in various conditions, such as

ulcerated carcinomata, balanitis, etc., spirochaetes are of com-
paratively common occurrence. There is no doubt whatever
that the great majority of these are readily distinguishable by
their appearance from the spirochsete pallida, but others re-
semble it closely. Hoti'mann, however1, who has seen many of
these spirocha-tes from other sources, considers that even by
their microscopic api>earance they are capable of being distin-
-uMied, though with considerable difficulty. It must, of course,
be borne in mind that the finding of an organism in non-syphilitic
lesions with the same microscopical characters does not show
that it is the same organism as the spirochaete pallida.

Cultivation. — Although numerous attempts have been made,
it has not yet been found possible to obtain pure cultures of the
-pirochsBte pallida outside the body. Levaditi and Mclntosh
inoculated with syphilitic material sacs of collodion containing

508 SYPHILIS

human serum, heated at 60°C., and placed them in the peritoneal
cavity of a monkey (macacus cynomolgus). After an interval of
about three weeks, they found in the sacs an abundant growth
of spirochsetes morphologically identical with spirochsete pallicla,
along with various anaerobic bacteria. They were able to
continue such cultures in like conditions, but were unable to
obtain any pathogenic effects on inoculating animals with the
material from the sacs, and considered that the organisms had
became avirulent owing to their conditions of growth. Recently
Schereschewsky claims to have obtained impure cultures of the
organism in test tubes. He used for this purpose horse serum
inspissated at 58° C., and then allowed to undergo autolysis for
three days at 37° C. Although abundant growth of spirochaetes
was obtained, he was unable to infect animals by means of them,
or to obtain any serum reactions which would go to show
that the organism was really the spirochsete pallida.

Transmission of the Disease to Animals. — Although various
experiments had previously been from time to time made by
different observers, in some cases with reported successful result,
it is to the papers of Metchnikoff and Roux (1903-5) that we
owe most of our knowledge. These observers have carried on a
large series of observations, and have shown that the disease can
be transmitted to various species of monkey. Of those the
anthropoid apes are most susceptible, the chimpanzee being the
most suitable for experimental purposes. Their results have
been confirmed by Lassar, Neisser, Kraus, and others. The
number of experiments on these animals is now very great, and
the general result is that the disease has been transmitted by
material from all the kinds of syphilitic lesions in which spiro-
chaetes have been demonstrated, including even the blood in
secondary syphilis and tertiary lesions. Inoculation is usually
made by scarification on the eyebrows or genitals ; the sub-
cutaneous and other methods of inoculation give negative
results. The primary lesion is in the form of an indurated
papule or of papules, in every respect resembling the human
lesion. Along with this there is a marked enlargement and
induration of the corresponding lymphatic glands. The primary
lesion appears on an average about thirty days after inoculation,
and secondary symptoms develop in rather more than half of the
cases after a further period of rather longer duration. These
are of the nature of squamous papules on the skin, mucous
patches in the mouth, and sometimes palmar psoriasis. As a
rule, the secondary manifestations are of a somewhat mild
degree, and in no instance up to the present has any tertiary

TRANSMISSION OF THE DISEASE TO ANIMALS 509

lesion been observed. By re-inoculation from the lesions, the
disease may be transferred to other animals. The disease may
also be produced in baboons and macaques (macacyj sinicus is
one of the most susceptible), but these animals are less susceptible,
and secondary manifestations do not appear. The severity of the
affection amongst aj>es would in fact appear to be in proportion to
the nearness of the relationship of the animal to the human
subject.

AJ <ho\vn tirst by Hansell, and more recently by Bertarelli, the
eye of the rabbit is susceptible to inoculation from syphilitic
lesions. The material used is introduced in a finely divided
state either into the tissue of the cornea or into the anterior
chamber, and syphilitic keratitis or iritis, or both, may result,
there being a period of incubation of at least two weeks.
Levaditi and Yamanouchi have recently studied the stages in
detail, and find that the spirochaetes remain in the inoculated
material unchanged for a time ; then organisation occurs and
the spirocha3tes multiply, and later still there is a more rapid
multiplication and invasion by them of the tissues of the eye.
The period of incubation is thus not due to the organism passing
through some cycle of development, but simply to its requiring
certain conditions for multiplying which are not supplied for
s« mil- time.

The production of the disease, experimentally, has supplied
us with some further facts regarding the nature of the virus.
It has been shown repeatedly that the passage of fluid con-
taining the virus through a Berkefeld filter deprives it completely
of its infectivity. In other words, the virus does not belong to
the ultra-microscopic group of organisms. The virus is also
readily destroyed by heat, a temperature of 51° C. being
fatal. With regard to the production of immunity, very little
of a satisfactory nature has so far been established. It has been
found that the virus from a macaque monkey produces a less
severe disease in the chimpanzee than the virus from the human
subject, inasmuch as secondary lesions do not follow; the virus
would thus appear to have undergone a certain amount of
attenuation in the tissues of that monkey. The effects of inject-
ing emulsions of tertiary lesions or of serum from syphilitic
patients, at the time of inoculation with the virus, appear to be
nil ; so also the employment of the virus rendered inactive by
heating has apparently no influence in acting as a vaccine.
There is some evidence that the serum from a patient suffering
from the disease when mixed with the virus before inoculation
modifies the disease to a certain extent, but further evidence on

510 SYPHILIS

this point is necessary. As mentioned above, the spirochsete
pallida has been found in the lesions in monkeys, Metchnikoff
and Roux obtaining positive results in more than 75 per cent,
of the cases, and it is to be noted that here also the organism
has been found deep in the substance of the papules, un-
accompanied by any other organisms. It is also to be noted
that the blood of infected apes after a time gives the Wasser-
mann reaction.

Serum Diagnosis — Wassermann Reaction. — The method of
applying this test has already been given (p. 131); we have now
to consider the results of its application. There is general
agreement amongst workers at the subject that the test affords
by far the most reliable means of diagnosis of the disease ; and
on comparing the results obtained it will not be an overestimate
to say that a positive result may be obtained in at least 90 per
cent, of cases where there is evidence of active general infection.
The reaction generally appears first on the fifteenth to thirtieth
day after appearance of the sore, and then gradually becomes
more marked ; during the period of secondary manifestations it
is practically always present ; in the tertiary stage with active
manifestations a positive result is only a little less frequent. As
the disease becomes inactive or is cured the reaction may disappear,
but it is to be noted that disappearance of the reaction after being
present does not necessarily imply cure of the disease. It may
only have become latent, and on its becoming once more active
the reaction may re-appear. Energetic treatment with mercury
may also diminish or annul the reaction ; in fact, its presence
and intensity would appear to be definitely related to the activity
of the syphilitic lesions. A positive reaction is also present in
the large majority of cases of general paralysis and of tabes, and
may be given by the cerebro-spinal fluid as well as by the blood
serum in such cases. As regards other diseases, a positive
reaction has been recorded as occurring in leprosy (p. 304) and
sleeping-sickness and also in yaws, but apart from these diseases
it is practically never met with. At present little can be said in
explanation of the Wassermann reaction. It seems to depend
on the interaction of lipoidal substances in the extract with
proteins in the serum, which are apparently contained in the
globulin fraction ; but we know nothing as to why this peculiar
modification of the serum should be present in syphilis. It is
now generally considered that it does not depend on the presence
of an anti-substance (immune-body), which in association with
the antigen (the spirochaite) fixes complement.

I-'KAMIUESIA OR YAWS 511

FRAMBCESIA OR YAWS.

l-'rambuesia is a contagious disease of the tropics, occurring in
the west coast of Africa, Ceylon, the West Indies, and other
parts. It is characterised by a peculiar cutaneous eruption, and
it is markedly contagious. Its resemblance in many respects
t«» syphilis has been noted, and the relation of the two diseases
1ms been the subject of much controversy. It is accordingly a
matter of great interest that an organism of closely similar
characters to the spirochuete pallida has been found in the lesions
of f ramboesia. This organism was discovered by Oastellani, who
gave to it the name spirochatte pertenuis or pallidula. Morpho-
logically, it is practically identical with the spirochaete pallida ;
when ulceration has occurred other spirochaetes of less regular
form may be present as contaminations. In the skin lesions
it has been shown by Levaditi's method to be present in con-
siderable numbers, especially in the epidermis and also amongst
the leucocytic infiltration, which comprises more polymorpho-
nuclear leucocytes than is seen in the case of syphilis. Castellani
showed that the disease could be transferred to monkeys (semno-
jiltkccus and macacus being used for this purpose), and that the
organism could be demonstrated in the unbroken skin lesions.
The lesions are as a rule confined to the site of inoculation, but
the infection is general, as is shown by the presence of spirochaetes
in the lymphatic glands and the spleen. These results with
regard to the presence of spirochaete pallidula in the lesions and
the inoculation of apes have been confirmed by other workers,
and the etiological relationship of the organism to the disease
may now be regarded as practically established. The immunity
reactions in monkeys infected with syphilis and frambcesia, as
experimentally studied by Castellani and by Neisser, Baermann,
and Halberstadter, go to show that the two diseases are distinct.
On the other hand, Levaditi and Nattan-Larrier found that,
although monkeys infected with syphilis were refractory to
framboesia (Fr. pian\ monkeys infected with frambcesia were sus-
cvptiblr to syphilis: they therefore concluded that frambcesia
is a modified or mild form of syphilis. The exact relationship
of the two diseases cannot be yet accurately defined, but they
are undoubtedly closely related, and probably have a common
parentage.

CHAPTER XXL

IMMUNITY.

Introductory. — By immunity is meant non-susceptibility to a
given disease or to a given organism, either under natural
conditions or under conditions experimentally produced. The
term is also used in relation to the toxins of an organism.
Immunity may be possessed by an animal naturally, and is then
usually called natural immunity, or it may be acquired by an
animal either by its passing through an attack of the disease, or
by means of artificial inoculation. It is to be noted that man and
the lower animals may be exempt from certain diseases under
natural conditions, and yet the causal organisms of these diseases
may produce pathogenic effects when injected in sufficient
quantity. Immunity is, in fact, of very varying degrees, and
accordingly the use of the term has a correspondingly relative
significance. This is not only true of infection by bacteria, but
of toxins also :— when the resistance of an animal to these is of
high degree, the resistance may in certain cases be overcome by
a very large dose of the toxic agent. On the other hand, in
cases where the natural powers of resistance are very high,
these can be still further exalted by artificial means, that is,
the natural immunity may be artificially intensified.

Acquired Immunity in the Human Subject. — The following
facts are supplied by a study of the natural diseases which affect
the human subject. First, in the case of certain diseases, one
attack protects against another for many years, sometimes
practically for a lifetime, e.g. smallpox, typhoid, scarlet fever,
etc. Secondly, in the case of other diseases, e.g. erysipelas,
diphtheria, influenza, and pneumonia, a patient may suffer from
several attacks. In the case of the diseases of the second group,
however, experimental research has shown that in many of
them a certain degree of immunity does follow; and, though
we cannot definitely state it as a universal law, it must be
considered highly probable that the passing through an attack

612

ARTIFICIAL IMMUNITY 513

of an acute disease produced by an organism, confers immunity
for a longer or shorter period. The immunity is not, however,
to be regarded as the result of the disease per se, but of the
bacterial products introduced into the system; as will be shown
ln-l«i\v, by suitable gradation of the doses of such products, or
by the use of weakened toxins, a high degree of immunity may
be attained without the occurrence of any symptoms whatever.

The facts known regarding vaccination and smallpox exemplify
another principle. We may take it as practically proved that
vaccinia is variola or smallpox in the cow, and that when
vaccination is performed, the patient is inoculated with a
modified variola (vide Smallpox, in Appendix). Vaccination
produces certain pathogenic effects which are of trifling degree
as compared with those of smallpox, and we find that the degree
of protection is less complete and lasts a shorter time than that
produced by the natural disease. Again, inoculation with lymph
from a smallpox pustule produces a form of smallpox less
severe than the natural disease but a much more severe con-
dition than that produced by vaccination, and it is found that
the decree of protection .or immunity resulting occupies an
intermediate position.

I iniinniity and Recovery from Disease. — Recovery from an
acute infective disease shows that in natural conditions the virus
may be exhausted after a time, the period of time varying in
different diseases. How this is accomplished we do not yet
fully know, but it has been found in the case of diphtheria,
typhoid, cholera, pneumonia, etc., that in the course of the
disease certain substances (called by German writers Antikorper)
appear in the blood, which are antagonistic either to the toxin
or to the vital activity of the organism. In such cases a process
of immunisation would appear to be going on during the pro-
of the disease, and when this immunisation has reached a
certain height, the disease naturally comes to an end. It cannot,
however, be said as yet that such antagonistic substances are
developed in all cases ; though the results already obtained make
this probable.

ARTIFICIAL IMMUNITY.

Varieties. — According to the means by which it is produced,
immunity may be said to be of two kinds, to which the terms
'I'-fitte and passive are generally applied, or we may speak of
immunity directly, or indirectly, produced. We shall first give
an account of the established facts, and afterwards discuss some

514 IMMUNITY

of the theories which have been brought forward in explanation
of these facts.

Active immunity is obtained by (a) injections of the organisms
either in an attenuated condition or in sub-lethal doses, or (b)
by sub-lethal doses of their products, i.e. of their "toxins," the
word being used in the widest sense. By repeated injections
at suitable intervals the dose of organisms or of the products
can be gradually increased ; or, what practically amounts to the
same, an organism of greater virulence or a toxin of greater
strength may be used. A degree of resistance or immunity
can thus be developed, and this in course of time may reach
a very high level. Such methods constitute the means of
preventive inoculation or vaccination. Immunity of this kind
is comparatively slowly produced and lasts a considerable time,
the duration varying in different cases. The principles of
vaccination have within recent years been extended by Wright
to the treatment of disease.

Passive immunity depends upon the fact that if an animal
be immunised to a very high degree by the previous method, its
serum may have distinctly antagonistic or neutralising effects
when injected into another animal along with the organisms, or
with their products, as the case may be. Such a serum,
generally known as an anti-serum, may exert its effects if intro-
duced into an animal at the same time as infection occurs or
even a short time afterwards ; it can, therefore, be employed
as a curative agent. The serum is also preventive, i.e. protects
an animal from subsequent infection, but the immunity thus
conferred lasts a comparatively short time. These facts form
the basis of serum therapeutics. When such a serum has the
power of neutralising a toxin it is called antitoxic ; when, with
little or no antitoxic power, it protects against the living
bacterium in a virulent condition, it is called antimicrobic or
antibacterial (vide infra).

In the accompanying table a sketch of the chief methods by
which an immunity may be artificially produced is given. It
has been arranged merely for purposes of convenience and to
aid subsequent description ; the principles underlying all the
methods are the same.

ARTIFICIAL IMMUNITY.

A. Active Immunity — i.e. produced in an animal by an in-
jection, or by a series of injections, of non-lethal doses of
an organism or its toxins.

ACTIVE IMMUNITY 515

1. By injection of the Hiring organisms.

(a) Attenuated in various ways. Examples : —

(1) By growing in the presence of oxygen, or in a

current of air.

(2) By passing through the tissues of one species

of animal (becomes attenuated for another
species).

(3) By growing at abnormal temperatures, etc.

(4) By growing in the presence of weak antiseptics, or

by injecting the latter along with the organism,
etc.

(b) In a virulent condition, in non-lethal doses.

2. By injection of the dead organisms.

3. By injection of jiltered bacterial cultures, i.e. toxins ; or of

substances derived from such filtrates.
These methods may also be combined in various ways.

B. Passive Immunity, i.e. produced in one animal by injection
of the serum of another animal highly immunised by the
methods of A.

1. By antitoxic serum, i.e. the serum of an animal highly

immunised against a particular toxin.
'2. It a antibacterial serum, i.e. the serum of an animal highly

immunised against a particular bacterium in the living

and virulent condition.

A. Active Immunity.

1. By Living Cultures. — (a) Attenuated. — In the earlier
work on immunity in the case of anthrax, chicken cholera, swine
plague, etc., the investigators had to deal with organisms of
high virulence, which had accordingly to be reduced before the
organisms could be injected in the living state. It is now found
most convenient a£ a rule to start the process of active immunisa-
tion with the injection of dead cultures. The principle is the
same as that of vaccination, and both attenuated cultures and
also the dead cultures used for injection are often spoken of as
rtirrfnes. The virulence of an organism may be diminished in
various ways, of which the following examples may be given : —

(1) In the first place, practically every organism, when culti-
vated for some time outside the body, loses its virulence, and in
the case of some this is very marked indeed, e.g. the pneumo-
coccus. Pasteur found in the case of chicken cholera, that

516 IMMUNITY

when cultures were kept for a time in ordinary conditions, they
gradually lost their virulence, and that when sub-cultures were
made the diminished virulence persisted. Such attenuated
cultures could be used for protective inoculation. He considered
the loss of virulence to be due to the action of the oxygen of
the air, as he found that in tubes sealed in the absence of oxygen
the virulence was not lost. Haffkine attenuated cultures of the
cholera spirillum by growing them in a current of air (p. 459).

('2) The virulence of an organism for a particular animal may
be lessened by passing the organism through the body of another
animal. Duguid and Burdon Sanderson found that the virulence
of the anthrax bacillus for bovine animals was lessened by its
being passed through guinea-pigs, the disease produced in the
ox by inoculation from the guinea-pig being a non-fatal one.
This discovery was confirmed by Greenfield, who showed that
the bacilli cultivated from guinea-pigs preserved their property
in cultures, and could therefore be used for protective inoculation
of cattle. A similar principle was applied in the case of swine
plague by Pasteur, who found that if the organism producing
this disease was inoculated from rabbit to rabbit, its virulence
was increased for rabbits but was diminished for pigs. The
method of vaccination against small-pox depends upon the same
principle. There is also evidence to show that the virulence of
the tubercle bacillus becomes modified according to its host,
being often diminished for other animals.

(3) Many organisms become diminished in virulence when
grown at an abnormally high temperature. The method of
Pasteur, already described (p. 346), for producing immunity
in sheep against anthrax bacilli, depends upon this fact. A
virulent organism may also be attenuated by being exposed to
an elevated temperature which is insufficient to kill it, as was
found by Toussaint in the case of anthrax.

(4) Still another method may be mentioned, namely, the
attenuation of the virulence by growing the organism in the
presence of weak antiseptics. Chamberland and Roux, for
example, succeeded in attenuating the anthrax bacillus by
growing it in a medium containing carbolic acid in the propor-
tion of 1 : 600.

These examples will serve to show the principles underlying
attenuation of the virulence of an organism. There are, how-
ever, still other methods, most of which consist in growing the
organism in conditions somewhat unfavourable to its growth, e.g.
under compressed air, etc.

(6) Immunity by living Virulent Cultures in Non-lethal

BY LIVING CULTURES 517

Doses. — Immunity may also be produced by employing virulent
cultures in small, that is non-lethal, doses. In subsequent
inoculations the doses may be increased in amount. For
example, immunity may thus be obtained in rabbits against the
bacillus pyocyaneus. Such a method, however, lias had only a
limited application, as it has been found more convenient to
commence the process by dead or attenuated cultures, and then
to continue with virulent cultures.

Exaltation of the Virulence. — The converse process to attenua-
tion, i.e. the exaltation of the virulence, is obtained chiefly by
the method of cultivating the organism from animal to animal —
the method of passage discovered by Pasteur (first, we believe,
in the case of an organism obtained from the saliva in hydro-
phobia, though having no causal relationship to that disease).
This is most conveniently done by intraperitoueal injections, as
there is less risk of contamination. The organisms in the
peritoneal fluid may be used for the subsequent injection, or a
culture may l>e made between each inoculation. The virulence
of a great number of organisms can be increased in this way,
the animals most frequently used being rabbits and guinea-pigs.
This method can be applied to the organisms of typhoid, cholera,
pneumonia, to streptococci and staphylococci, and in fact to
those organisms generally which invade tissues.

The virulence of an organism, especially when in a relatively
attenuated condition, can also be raised by injecting along with
it a quantity of a culture of another organism either in the living
or dead condition. A few examples may be mentioned. An
attenuated diphtheria culture may have its virulence raised by
being injected into an animal along with the streptococcus
pyogenes ; an attenuated culture of the bacillus of malignant
oedema by being injected with the bacillus prodigiosus; an
attenuated streptococcus by being injected with the bacillus coli,
etc. A culture of the typhoid bacillus may be increased in
virulence, as already stated, by being injected along with a dead
culture of the bacillus coli. In such cases the accompanying
injection enables the attenuated organism to gain a foothold in
the tissues, and it may be stated as a general rule that the
virulence of an organism for a particular animal is raised by its
growing in the tissues of that animal.

Combination of Methods. — The above methods may be com-
bined in various ways. By repeated injections of cultures at
first in the dead condition, then living and attenuated and
afterwards more virulent, and by increasing the doses, a high
degree of immunity may be obtained.

518 IMMUNITY

2. Immunity by Dead Cultures of Bacteria. — In some cases
a high degree of immunity against infection by a given microbe
may be developed by repeated and gradually increasing doses
of the dead cultures, the cultures being killed sometimes by
heat, sometimes by exposure to the vapour of chloroform. In
this method the so-called endotoxins will be injected along with
the other substances in the bacterial protoplasm, but the result-
ing immunity is chiefly directed against the vital activity of the
organisms — is antibacterial rather than antitoxic (vide infra).
The cultures when dead produce, of course, less effect than when
living, and this method may be conveniently used in the initial
stages of active immunisation, — to be afterwards followed by
injections of the living cultures. The method is extensively
used for experimental purposes, and is that adopted in anti-plague
and anti-typhoid inoculations, and in the treatment of infections
by means of vaccines.

3. Immunity by the Separated Bacterial Products or
Toxins. — The organisms in a virulent condition are grown in
a fluid medium for a certain time, and the fluid is then filtered
through a Chamberland or other porcelain filter. The filtrate
contains the toxins, and it may be used unaltered, or may be
reduced in bulk by evaporation, or may be evaporated to dryness.
The process of immunisation by the toxin is started by small
non-lethal doses of the strong toxin, or by larger doses of toxin
the power of which has been weakened by various methods (vide
infra). Afterwards the doses are gradually increased. This
method was carried out with a great degree of success in the
case of diphtheria, tetanus, malignant oedema, etc. It appears
capable of general application in the case of organisms where it
is possible to get an active toxin from the filtered cultures. It
has also been applied in the case of snake venoms by Calmette
and by Fraser, and a high degree of immunity has been
produced.

The following may be mentioned as some of the most
important examples of the practical application of the principles
of active immunity, i.e. of protective inoculation : — (1) Inocula-
tion of sheep and oxen against anthrax (Pasteur) (p. 346) ; (2)
Jenneriau vaccination against smallpox (p. 565) ; (3) Anti-
cholera inoculation (Haffkine) (p. 459) ; (4) Anti-plague
inoculation (Haffkine) (p. 486) ; (5) Anti-typhoid inoculation
(Wright and Semple) (p. 375) ; (6) Pasteur's method of inocula-
tion against hydrophobia, which involves essentially the same
principles (p. 579).

Vaccines as a Method of Treatment. — Up till recently the

BY BACTERIAL PRODUCTS OR TOXINS 519

principles of active immunity had not been directly applied in
the treatment of an existing disease except in the case of tuber-
culosis. The work of Wright, however, shows that active
immunisation in such circumstances is not only possible, but is
also probably capable of wide application. From his study of
the part played by phagocytosis in the successful combat of
bacteria by the body, he was led to advocate the treatment
of bacterial infections by carrying on an active immunisation
against the causal agents by the injection of dead cultures of
the latter. The justification for such a procedure lies in his
contention that in many cases infections are to be looked on as
practically localised, e.g. the cases of an acne pustule, or a boil.
The reason for the local growth of bacteria in the part of the
body affected is that there is for unknown causes a deficiency
of the opsonic power (vide p. 291) of the body fluids, which is
essential for the phagocytosis of the invading bacteria. Still
more marked in such cases is the deficiency in the opsonic
qualities of the fluids in the actual site of infection. Any
procedure which will raise the opsonic power of the body fluids
M ;i whole, and therefore of the fluids in the focus of infection,
will aid the destruction of the bacteria by sensitising them to
phagocytic action. Such a procedure is found in the active
immunisation which results from the injection of a vaccine
consisting of a dead culture of the causal bacteria. The
application of a vaccine of this kind can be controlled by
observation of the opsonic index of the patient's serum during
the treatment. When a local infection is present the general
opsonic index is usually found to be below unity. If dead
bacteria be injected into the individual there may occur during
the following few days a further fall in the opsonic index, —
what Wright calls the occurrence of a negative phase. • In a case
where the treatment is successful, this negative phase is succeeded
by a rise in the opsonic index above its original level, — occurrence
of positive phase, — and with this reaction there is an improve-
ment in the local condition. Usually in such cases repeated
injections are required to effect a cure, and the important point,
according to Wright, is to avoid giving an injection when a
negative phase is in progress. If this point is not attended to
an aggravation of symptoms may occur (vide p. 291).
With regard to the details of the preparation of the vaccines see
p. 133. The number of bacteria employed for a vaccination
varies from 5,000,000 to 500,000,000 or more. Such vaccines
have been used extensively in the treatment of acne, boils,
sycosis, tuberculosis, infections of the genito-urinary tract by the

520 IMMUNITY

b. coli, infections of joints by the gonococcus, and in many cases
considerable success has followed the treatment.

Active Immunity by Feeding. — Ehrlich found that mice
could be gradually immunised against ricin and abrin by feeding
them with increasing quantities of these substances (vide p. 199).
In the course of some weeks' treatment in this way the resulting
immunity was of so high a degree that the animals could tolerate
on subcutaneous inoculation 400 times the dose originally fatal.
Fraser also found in the case of snake venom that rabbits could,
by feeding with the poison, be immunised against several times
the lethal dose of venom injected into the tissues.

By feeding animals with dead cultures of bacteria or with
their separated toxins, a degree of immunity may in some cases
be gradually developed. But this method is so much less certain
in results, and so much more tedious than the others, that it has
obtained no practical applications.

Active immunity of high degree developed by the methods
described may be regarded as specific, that is, is exerted only
towards the organism or toxin by means of which it has been
produced. A certain degree of immunity, or rather of increased
general resistance of parts of the body (for example the peri-
toneum), can, however, be produced by the injection of various
substances — bouillon, blood serum, solution of nuclein, etc.
(Issaeff). Also increased resistance to one organism can be thus
produced by injections of another organism. Immunity of this
kind, however, never reaches a high degree.

B. Passive Immunity.

Action of the Serum of Highly Immunised Animals. — 1.
The serum of an animal A, treated by repeated and gradually
increased doses of a toxin of a particular microbe, may protect
an animal B against a certain amount of the same toxin when
injected along with the latter, or a short time before it. As
might be expected, it has less effect when injected some time
afterwards, but even then within certain limits it has a degree
of curative or palliative power. Seeing that the serum of animal
A appears to neutralise the toxin, the term antitoxic has been
applied to it.

2. The serum of an animal A, highly immunised against a
bacterium by repeated and gradually increasing doses of the
organism, may protect an animal B against an infection by
the living organism when injected under conditions similar to the
above. This serum is therefore antimicrobic, or antibacterial,

PASSIVE IMMUNITY 521

i.e. preventive against invasion by a particular organism. (In
addition to the preventive or protective action in vivo, such a
serum may exert certain recognisable effects on the corresponding
organism in vitro. Thus (a) it may lead to the death or solution
of the organism — bactericidal or lysogenic action ; when no such
effect follows, the presence of an immune-body (p. 128) may be
shown by the deviation of complement method ; (b) it may pro-
duce an increased susceptibility to ingestion by phagocytes —
opsonic action ; (c) it may lead to the clumping of the organism
— agglutinative action, or to precipitation with an extract of a
culture of the corresponding bacterium.)

Anti-substances and their Specificity. — The fundamental fact
in passive immunity, namely, that immunity can be transferred to
another animal, shows that the serum in question differs from
the serum of a normal animal in containing antagonistic sub-
stances to the toxin or bacterium as the case may be, — these
being generally spoken of as anti-substances. The development
of these bodies, first observed in the case of the injection * of
toxins, is found to occur when a great many different substances
are introduced into the tissues of the living body. We can, in
fact, divide organic molecules into two classes — those which give
rise to the production of anti-substances, and are thus known as
antigens, and those which have not this property. Amongst the
former are various toxins, ferments, molecules of tissue cells,
bacteria, red corpuscles, etc. They are all probably of proteid
nature, though their true constitution is not known, and none of
them have been obtained in a pure condition. Amongst the
latter may be placed the various poisons of known constitution,
glncosides, alkaloids, etc. We may also state at present that the
anti-substance forms a chemical or physical union with the
particular antigen which has led to its development, and we
shall discuss the evidence for this later. Furthermore, the anti-
substance has apparently a specific combining group which fits, as
it were, a group in the corresponding antigen, the two groups
having been compared to a lock and key. It is, however, to be
noted that this specificity is a chemical one rather than a
biological one. An anti-serum, for example, developed by the
injection of bacterium A may also have some effect on bacterium
B, and thus appear not to be specific. We have, however,
evidence to show that the antigens in bacterium A are not all
identical, and that some of them may be present though in smaller
proportion in bacterium B ; thus the theory of combining sjjeci-
t'n-ity is not invalidated. The number of different anti-substances,
as judged by their combining properties, would appear to be almost

522 IMMUNITY

unlimited, a fact which throws new light on the complexity of
the structure of living matter. When anti-substances are studied
as regards their action on the substances with which they com-
bine, they may be conveniently arranged in three classes
corresponding to Ehrlich's three classes of receptors (vide p. 549).
In the first place, the anti-substance may simply combine with
the substance without, so far as we know, producing any change
in it, and to this group the anti-toxins and anti-ferments belong.
In the second place, the anti-substance, in addition to combining,
may produce some recognisable physical alteration. In other
words, it possesses an active or zymotoxic group as well as
a combining group. The agglutinins may be mentioned as
examples of this group. In the third place, the anti-substance
after combination may lead to the combination of another body
normally present in serum called complement or alexine, and
this latter, which has a constitution very similar to that of a toxin,
may lead to physical change, for example, death or solution of a
cell. Anti-substances of this class are known as immune-bodies
or amboceptors (Ehrlich) or as sensitising substances — substances
sensibilisatrices of French writers. Their essential feature is that
they lead to the combination or fixation of complement, which
may or not produce some recognisable change such as bacterio-
lysis, etc. If no such effect follows, however, the union of com-
plement may be demonstrated by the indirect or deviation
method (p. 130).

After this preliminary statement in explanation, we shall con-
sider the actual properties of the two classes of serum, and later
we shall resume the theoretical consideration.

Antitoxic Serum. — In a previous chapter (p. 188) a distinction
has been drawn between extra- and intra-cellular toxins, and
with regard to these the general statement may be made that
while antitoxins are, as a rule, comparatively easily obtained in
the case of the former, the matter is quite otherwise in the case
of the latter. In fact some writers have gone so far as to say
that antitoxins to endotoxins cannot be obtained. Such an
extreme view is in our opinion unjustifiable in the light of the
recent work on antitoxins to the typhoid, cholera, and dysentery
endotoxins (pp. 366, 456, 388). Nevertheless we have the im-
portant fact that in many cases by the injection of dead cultures
an active anti-bacterial serum can be obtained which has no
neutralising action on the endotoxins, and we must conclude
either that a large proportion of the endotoxin does not lead to the
production of antitoxin or does so only with great slowness, the
latter alternative being on general grounds rather improbable.

ANTITOXIC SERUM 523

The best examples of antitoxic sera are those of diphtheria and
tetanus, though similar principles and methods are involved in
the case of the anti-sera to ricin and abrin, and to snake poison.
\\V shall here speak of diphtheria and tetanus. The steps in the
process of preparation may be said to be the following : First,
the preparation of a powerful toxin ; second, the estimation of
the power of the toxin ; third, the development of antitoxin in
the blood of a suitable animal by gradually increasing doses of
the toxin ; fourth, the estimation from time to time of the
antitoxic power of the serum of the animal thus treated.

1. Preparation of the Toxin. — The mode of preparation and
the conditions affecting the development of diphtheria toxin
have already been described (p. 406). In the case of tetanus the
growth takes place in glucose bouillon under an atmosphere of
hydrogen (vide p. 66). In either case the culture is filtered
through a Chamberland filter when the maximum degree of
toxicity has been reached. The term " toxin " is usually applied
for convenience to the filtered (i.e. bacterium-free) culture.

2. Estimation of the Toxin. — The power of the toxin is
estimated by the subcutaneous injection of varying amounts in
a number of guinea-pigs, and the minimum dose which will
produce death is thus obtained. This, of course, varies in pro-
portion to the weight of the animal, and is expressed accordingly.
In the case of diphtheria, in Ehrlich's standard, the minimum
lethal dose — known as M.L.D. — is the smallest amount which
will certainly cause death in a guinea-pig of 250 grms. within
four days. The testing of a toxin directly is a tedious process,
and in actual practice, where many toxins have to be dealt with,
it is found more convenient to test them by finding how much
will be neutralised by a certain amount of a standard antitoxic
serum, namely, an " immunity unit " (p. 524).

3. Development of Antitoxin. — The earlier experiments on
tetanus and diphtheria were performed on small animals, such
as guinea-pigs, but afterwards the sheep and the goat were used,
and finally horses. In the case of the small animals it was
found advisable to use in the first stages of the process either a
weak toxin or a powerful toxin modified by certain methods.
Sn.-h lilt-thuds are the addition to the toxin of terchloride of
in. line (Behring and Kitasato), the addition of Gram's iodine
solution in the proportion of one to three (Roux and Vaillard),
ami the plan, adopted by Vaillard in the case of tetanus, of
usini: a >rri«'> of toxins weakened to varying degrees by being ex-
posed to different temperatures, namely, 60°, and 55°, and 50° C.
In the case of large animals immunisation is sometimes started

524 IMMUNITY

with small doses of unaltered toxin ; and the doses are gradually
increased. The toxin is at first injected into the subcutaneous
tissues, later into a vein. Ultimately 300 c.c., or more, of active
diphtheria toxin thus injected may be borne by a horse, such a
degree of resistance being developed after the treatment has been
carried out for two or three months. The antitoxin content of
the serum is estimated from time to time, the object being, of
course, to raise it to as high a figure as possible. It is found
that each injection produces a certain amount of fall in the anti-
toxin value, and this, in favourable cases, is followed by a rise to
a higher, level than before, the former event being due in part
to the combination of a portion of the antitoxin with the toxin
introduced. (Similar phenomena are observed in the develop-
ment of all other classes of anti-substances.) In all cases of
immunising the general health of the animal ought not to suffer.
If the process is pushed too rapidly the antitoxic power of the
serum may diminish instead of increasing, and a condition of
marasmus may set in and may even lead to the death of the
animal. After a sufficiently high degree of antitoxic power
has been reached, the animal is bled under aseptic pre-
cautions, and the serum is allowed to separate in the usual
manner. It is then ready for use, but some weak antiseptic,
such as '5 per cent carbolic acid, is usually added to prevent its
decomposing. Other antitoxic sera are prepared in a correspond-
ing manner. Some further facts about antitetanic serum are
given on p. 429. (In immunisation of small animals an indica-
tion of their general condition may be obtained by weighing
them from time to time.)

4. Estimating the Antitoxic Power q/, or "standardising" the
Serum. — This is done by testing the effect of various quantities
of the serum of the immunised animal against a certain amount
of toxin. Various standards have been used, of which the two
chief are that of Ehrlich and that of Roux. Ehrlich has adopted
as the immunity unit the amount of antitoxic serum which will
neutralise 100 times the minimum lethal dose of toxin, serum
and toxin being mixed together, diluted up to 4 c.c. and injected
subcutaneously into a guinea-pig of 250 grms. weight, the
prevention of the death of the animal within four days being
taken as the indication of neutralisation. 1 c.c. of a serum, of
which '02 c.c. will protect against a hundred times the lethal
dose, will possess 50 immunity units, and 20 c.c. of this serum
1000 imm unity units. Sera have been prepared of which 1 c.c.
has the value of 800 units or even more. As a standard in
testing, Ehrlich employs quantities of serum of known antitoxic

USE OF ANTITOXIC SERA 525

power in a dry condition, preserved in a vacuum in a cool place,
and in the absence of light. A thoroughly dry condition is
ensured by having the glass bulb containing the dried serum
connected \\ith another bulb containing anhydrous phosphoric
acid. With such a standard test-serum any newly prepared
serum can readily be compared.

Roux has adopted a standard which represents the animal weight in
grammes protected by 1 c.c. of serum against the dose of virulent bacilli
lethal to a control guinea-pig in thirty hours, the serum being injected
twelve hours previously. Thus, if '01 c.c. of a serum will protect a
guinea-pig of 500 grms. against the lethal dose, 1 c.c. (1 grin.) will pro-
tect 50,000 grms. of guinea-pig, and the value of the serum will be 50,000.

Use of Antitoxic Sera. — In all cases the antitoxic serum ought
to be injected as early in the disease as possible, and in large
doses. In the case of diphtheria 1500. immunity units of anti-
toxic serum was the amount first recommended for the treatment
of a bad case, but the advisability of using larger doses has
gradually become more and more evident. Sidney Martin
recommends that as much as 4000 units should be administered
at once, and that if necessary this quantity should be repeated.
A strong serum prepared by Behring contains 3000 units in
5 to 6 c.c., but even stronger sera may be obtained. Even very
large doses of antitoxic serum are without any harmful effects
beyond the occasional production of urticarial and erythematous
rashes (p. 563). Where large quantities of serum require to be
administered, as is always the case with antitetanic serum, in-
jections must be made at different parts of the body ; preferably
not more than 20 c.c. should be injected at one place. In recent
times intravenous injection has been introduced, the advantage
being greater rapidity of action. The immunity conferred by
injection of antitoxic serum lasts a comparatively short time,
usually a few weeks at longest.

Sera of Animals immunised against Vegetable and Animal
Poisons. — It was found by Ehrlich in the case of the vegetable
toxins, ricin and abrin, and also by Calmette and Fraser in the
case of the snake poisons, that the serum of animals immunised
against these respective substances had a protective effect when
injected along with them into other animals. Ehrlich found,
for example, that the serum of a mouse which had been highly
immunised against ricin by feeding as described above, could
protect another mouse against forty times the fatal dose of that
substance. He considered that in the case of the two poisons,
antagonistic substances — "anti-ricin" and " anti-abrin "— were

526 IMMUNITY

developed in the blood of the highly immunised animals. A
corresponding antagonistic body, to which Eraser gave the name
" antivenin," appears in the blood of animals in the process of
immunisation against snake poison.

These investigations are specially instructive, as such vegetable
and animal poisons, both as regards their local action and the
general toxic phenomena produced by them, present, as we have
seen, an analogy to various toxins of bacteria.

Nature of Antitoxic Action. — This subject is only part of the
general question with regard to the relation of anti-substances
to their corresponding antigens, but it is with regard to anti-
toxic action that most of the work has been done. We have to
consider here two points, namely, (a) the relation of antitoxin
to toxin, and (b) the source of the antitoxin. With regard to the
former subject there is now no doubt that the antagonism
between toxin and antitoxin is not a physiological one but that
the two bodies unite in vitro to form a compound inert towards
the living tissues, there being in the toxin molecule an atom
group which has a specific affinity for the antitoxin molecule or
part of it. We shall consider the facts in favour of this view,
and in doing so we must also take into account the anti-sera of
the vegetable toxins, of snake poisons, etc.

When toxin and antitoxin are brought together in vitro, it
can be proved that their behaviour towards each other resembles
what is observed in chemical union. Thus it has been found
that a definite period of time elapses before the neutralisation
of the toxin is complete, that neutralisation takes place more
rapidly in strong solutions than in weak, and that it is hastened
by warmth and delayed by cold. C. J. Martin and Cherry, and
also Brodie, have shown, that in the case of diphtheria toxin and
in that of an Australian snake poison the toxin molecules will
pass through a colloid membrane (p. 193), whilst those of the
corresponding antitoxin will not. Now, if a mixture of equivalent
parts of toxin and antitoxin is freshly prepared and at once
filtered, a certain amount of toxin will pass through, but the
longer such a mixture is allowed to stand before filtration the
less toxin passes, till a time is reached when no toxin is found
in the filtrate. Further, if the portion of fluid which at this
stage has not passed through the filter be injected into an animal
no symptoms take place ; this shows that after a time neutral-
isation is complete. Again, in cases where the toxin has some
definite physical effect, demonstrable in vitro, e.g. lysis, aggluti-
nation, coagulation, or the prevention of coagulation, its action
can be annulled by the antitoxin ; in such circumstances

NATURE OF ANTITOXIC ACTION 527

manifestly no physiological action of antitoxin through the
medium of the cells of the body can come into play. These
facts are practically conclusive in favour of antitoxin action
depending upon a direct union of the two substances concerned,
and Morgenroth has recently shown that the combination toxin-
antitoxin can be broken up by the 'action of hydrochloric acid
and the two constituents recovered.

Although authorities are now agreed as to the direct com-
bination of toxin and antitoxin, there is still much uncer-
tainty as to the exact nature of this union. Regarding this
subject there may be said to be three chief views — (a)
that of Ehrlich, according to which there is a firm chemical
union of toxin and antitoxin, and the former is not homo-
geneous but has a complex structure ; (6) that of Arrhenius and
Madsen, who consider that the phenomena correspond to the
behaviour of two substances in weak chemical union ; and (c) that
of Bordet, who regards the. combination not to be of chemical,
but of physical nature, corresponding to a process of adsorption.
Controversy on this question may be said to date from the
important work of Ehrlich on the neutralisation of diphtheria
toxin. Using an immunity unit of antitoxin (the equivalent of
100 doses of toxin) he determined with any example of crude
toxin the largest amount of toxin which could be neutralised
completely, so that no symptoms resulted from an injection of
the mixture. This amount he called the limes null dose, ex-
pressed as L0. He then investigated the effects of adding larger
amounts of toxin to the immunity unit and observed the
quantity which was first sufficient to produce a fatal result,
that is, which contained one M.L.D. of free toxin; this amount
he called the limes todtlich, fatal limit, expressed as Lt. Now
if, as he supposed, the union of toxin and antitoxin resembled
that of a strong acid and base, Lt - L0 ought to be the equiva-
lent of a minimum lethal dose of the toxin alone. This, how-
ever, was never found to be the case, the difference being always
considerably more than one M.L.D. For example, in the case
of one toxin, M.L.D. = '0165 c.c., Lt=l'26 c.c., L0=-9 c.c. ;
difference = -36 c.c., i.e. 21 '9 M.L.D. This, in brief, is what
is known as the "Ehrlich phenomenon," and it has been
explained by him as the result of the presence of toxoids (vide
p. 198), i.e. toxin molecules in which the toxophorous group has
become degenerated. He distinguishes three possible varieties
of such bodies according to the affinity of the haptophorous
group, namely, prototoxoid with more powerful affinity than the
toxin molecule, epitoxoid with less powerful affinity, and syntoxoid

528 IMMUNITY

with equal affinity. The presence of epitoxoids would manifestly
explain the above phenomenon. The L0 dose would represent
toxin + epitoxoid molecules all united to antitoxin molecules and
the addition of another M.L.D. of toxin would not result in
there being a free fatal dose, but in the added toxin taking the
place of epitoxoid. Several lethal doses would need to be added
before the mixture was sufficient to produce a fatal result ; that is,
Lt — L0 would equal several M.L.D.s. Ehrlich observed another
fact strongly in favour of the existence of toxoids, namely, that
in the course of time the toxin might become much weakened,
so that in one case observed the M.L.D. came to be three times
the original fatal dose, and still the amount of antitoxin neces-
sary to neutralise it completely was the same as before. Ehrlich
also investigated the effects of partial neutralisation of the L0
amount of toxin, that is, he added to this amount different
fractions of an immunity unit and estimated the toxicity of the
mixture. He found by this method that the neutralisation of
the toxin did not take place gradually, but as if there were dis-
tinct bodies present with different combining affinities — the
graphic representation of the effects of the mixture not being a
curve but a step-stair line. Thus he distinguished proto-, deutero-,
and trito-toxins (with corresponding toxoids). It will thus be
seen that Ehrlich regards the combination toxin-antitoxin to be
a firm one, and that the neutralisation phenomena are to be
explained by the complicated constitution of the crude toxin.

The chief criticism of Ehrlich's views has come from the
important work of Madsen and Arrhenius. Their main con-
tention is that the toxin-antitoxin combination is not a firm one
but a reversible one, and is governed by the laws of physical
chemistry. For example, in the case of a mixture of ammonia
and boracic acid (i.e. of a weak base and a weak acid)
in solution, there is a constant relation between the amounts
of each of the substances in the free condition and the
amounts in combination, — the combination is reversible, so that
if some of the free ammonia were removed a certain amount of
the combined ammonia would become dissociated to take its
place ; further, if to the mixture, in a state of equilibrium,
more ammonia or more boracic acid were added, part would
remain free while part would combine. Accordingly, if toxin
and antitoxin behaved in a similar manner an explanation of the
Ehrlich phenomenon would be afforded. Madsen and Arrhenius
have worked out the question in the case of a great many toxins,
and find that the graphic representation of neutralisation is in
every case a curve which can be represented by a formula. It

MODE OF PRODUCTION OF ANTITOXINS 529

should be noted in connection with this controversy that there are
two questions which may be independent of each other, namely,
(1) does the "toxin" in any particular case represent a single
substance or several1? (2) What is the nature of the combination
of any one constituent substance and its anti-substance — is it
reversible or is it not ? It may be said that it is practically
impossible to explain, the facts with regard to diphtheria toxin
on the hypothesis of a single substance, even if this should have
its combining and toxic actions equally weakened ; " toxoids " in
Ehrlich's sense must in our opinion be supposed. Then there is
an important fact established by Danysz and by v. Dungern,
namely that the amount of toxin neutralisable by a given amount
of antitoxin is different according as the toxin is added in several
moieties or all at once — in the latter case the amount of toxin
neutralisable is greater. There seems no explanation of this
according to the view of Madsen and Arrhenius, as the same state
of equilibrium ought to be reached in the two cases, that is, the
amounts of toxin neutralised should be the same.

An important t'ai-tnr in the union of toxin and antitoxin is
the time necessary for the union to be complete. Morgenroth
lias shown that in the case of diphtheria toxin this is considerable,
— about twenty-four hours. Up to this time, mixtures of toxin
and antitoxin, \\hrn injected intravenously, show decreasing
degrees of toxicity according to the time they have kept. On
the other hand, when the subcutaneous method of injection is
used the time interval has no effect, and this he considers to be
due to a catalytic action of the tissues which accelerates the
union of the two substances. A striking phenomenon, which
apparently points to the reversibility of the combination, was
noted by Behring in the case of diphtheria toxin, and afterwards
studied l>y Madsen and by Otto and Sachs in the case of
botulismus toxin, namely, that when a mixture of toxin and
antitoxin was found to be neutral on injection, a fraction of the
mixture might produce toxic phenomena or even death. This
wa«* apparently due to dissociation of the toxin in the greater
dilution, and in favour of this being the case Otto and Sachs
found that when the mixture was allowed to stand for twenty-
ton!1 hours, so that combination was complete, the phenomenon
no longer occurred. It was shown by Morgenroth and by Muir
independently that the union of a hasinolytic immune-body with
tin- ••onvspondinu; red corpuscle was of reversible nature, and
tin- latter observer found that in this case the union was not
increased in iirnmess after twenty-four hours. There is little
doubt that there are varying degrees of firmness of union of an

530 IMMUNITY

antigen and its anti-substance and varying periods necessary for
the combination to become complete ; and it is quite evident that
if there should be several toxic bodies in a " toxin," and that if
the union of some of these with antitoxin should be reversible,
the problem becomes one of extreme complexity.

There has recently been a tendency on the part of some
authorities to consider that the union of toxin-antitoxin does not
correspond to what takes place in ordinary chemical union, but
is a physical interaction of bodies in a colloidal state, the action
being one of the so-called adsorption phenomena. The smaller
toxin molecule becomes entangled, as it were, in the larger
antitoxin one, very much as a dye becomes attached to the
structure of a thread. Bordet has long maintained a theory of
this nature, and gives reasons for believing that there is no
definite quantitative relationship in the combination of the
molecules of the two substances, different amounts of antitoxin
affecting in varying degree all the molecules of a given amount
of toxin. A statement on the general question is at present
impossible ; we can only say that direct combination of the two
bodies does occur; that sometimes, probably often, the "toxin"
contains different toxic bodies with varying affinity ; and that
in a few instances the combination has been proved to be revers-
ible, but to what extent this is generally true remains still to be
determined.

The next question to be considered is the source of antitoxin.
The following three possibilities present themselves : (a) antitoxin
may be formed from the toxin, i.e. may be a " modified toxin " ;
(b) antitoxin may be the result of an increased formation of
molecules normally present in the tissues ; (c) antitoxin may be
an entirely new product of the cells of the body. It can now be
stated that antitoxin is not a modified toxin. It has been shown,
for example, that the amount of antitoxin produced by an animal
may be many times greater than the equivalent of toxin injected ;
and further, that when an animal is bled the total amount of
antitoxin in the blood may some time afterwards be greater
than it was immediately after the bleeding, even although no
additional toxin is introduced. This latter circumstance
shows that antitoxin is formed by the cells of the body. If
antitoxin is a product of the cells of the body, we are almost com-
pelled, on theoretical grounds, to conclude that it is not a newly-
manufactured substance, but a normal constituent of the living
cells which is produced in increased quantity. We have, how-
ever, direct evidence of the presence of antitoxin under normal con-
ditions,— the presence of such being shown by its uniting with

MODE OF PRODUCTION OF ANTITOXINS 531

toxin and rendering it inert. Normal horse serum, to mention
an example, may have a varying amount of antitoxic action to
the diphtheria poison, ox-bile has a similar action to snake poison,
whilst in the case of other anti-substances — such as agglutinins,
bacteriolysins, haemolysins, etc. — whose production is governed
by the same laws, numerous examples might be given. It is,
however, rather to the protoplasm of living cells than to the serum
that we must look for the source of antitoxins. In the first place,
we have evidence that in the living body bacterial toxins enter
into combination with, or, as it is often expressed, are fixed by
the tissues — presumably by means of certain combining affinities.
This has been shown by the experiments of Donitz and of
Heymans with tetanus toxin. We have, in such cases, however,
no evidence as to where the toxin is fixed beyond that supplied
l>y the occurrence of symptoms. Another line of research which
has been followed is to bring emulsions of various organs into
contact with a given toxin and observe whether any of the
toxicity is removed. This was first carried out by Wassermann
and Takaki, who investigated the action of emulsions of the
central nervous system of the susceptible guinea-pig on tetanus
toxin. They found in this way that the nervous system con-
tained bodies which had a neutralising effect on the toxin.
For example, it was shown that 1 c.c. of emulsion of brain and
spinal cord was capable of protecting a mouse against ten times
the fatal dose of toxin. These observations have been confirmed,
though their significance has been variously interpreted : and in
view of recently ascertained facts with regard to processes of
physical adsorption, it is quite possible that this neutralisation
of toxin does not represent a specific union as in the case of
antitoxin action. We may note, however, that it is not a serious
objection that in certain animals other tissues than that of the
central nervous system can combine with tetanus toxin — this
might take place with or without resulting symptoms.

It will be seen from what has been stated with regard to the
relation of toxin and antitoxin, that the fixation of toxin by the
tissues leads up theoretically to the possible production of anti-
toxin. In other words, the substance which, when forming part
of the cells, fixes the toxin and thus serves as the means of
jmisDiiiiitr, may act as an antitoxin when free in the blood.
This will be discussed below in connection with Ehrlich's theory
of jussive immunity. We may conclude by saying that anti-
tai-in is 2)?obably represented by molecules normally present in
the cells or (more rarely) in tJie fluids of the body.

Of the chemical nature of antitoxins we know little. From

532 IMMUNITY

their experiments C. J. Martin and Cherry deduced that while
toxins are probably of the nature of albumoses, the antitoxins
probably have a molecule of greater size, and may be allied to
the globulins. Such a supposed difference in the sizes of the
molecules might explain the fact, observed by Fraser and also
by C. J. Martin, that antitoxin is much more slowly absorbed
when introduced subcutaneously than is the case with toxin.
Hiss and Atkinson also came to the conclusion that antitoxin
belongs to the globulins. They found that the precipitate with
magnesium sulphate from anti-diphtheria serum contained
practically all the antitoxin, and that any substance obtained
which had an antitoxic value gave all the reactions of a globulin ;
and this result has been confirmed by others. They also found
that the percentage amount of globulin precipitated from the
serum of the horse increased after it was treated in the usual
way for the production of antitoxin. Ledingham observed an
increase of globulin during the process of immunisation of a
horse which yielded a high-grade antitoxic serum, and he ascer-
tained that while this increase was more on the part of the
euglobulin than of the pseudoglobulin fraction, most of the
antitoxin was contained in the latter.

Antitoxin, when present in the serum, leaves the body by the
various secretions, and in these it has been found, though in
much less concentration than in the blood. It is present in the
milk, and a certain degree of immunity can be conferred on
animals by feeding them with such milk, as has been shown by
Ehrlich, Klemperer, and others. Klemperer also found traces of
antitoxin in the yolk of eggs of hens whose serum contained
antitoxin. Bulloch also found in the case of hseinolytic sera
(vide infra) that the anti-substance ("immune-body ") is trans-
mitted from the mother to the offspring.

Antibacterial Serum. — The stages in the preparation of
antibacterial sera correspond to those in the case of antito'xic
sera, but living, or, in the early stages, dead cultures are used
instead of toxin separated by filtration, and in order to obtain a
serum of high antibacterial power a very virulent culture in large
doses must be ultimately tolerated by the animal. For this
purpose a fairly virulent culture is obtained fresh from a case of
the particular disease, and its virulence may be further increased
by the method of passage. This method of obtaining a high'
degree of immunity against the microbe is specially applicable
in the case of those organisms which invade the tissues and
multiply to a great extent within the body, and of which the
toxic effects, though always existent, are proportionately small

PROPERTIES OF ANTIBACTERIAL SERUM 533

in relation to the number of organisms present. The method
has been applk-d in the case of the typhoid and cholera organ-
isms, the bacillus of bubonic plague, the bacillus coli communis,
the pneumococcus, streptococcus (Marmorek), and many others.
In fact, it seems capable of very general application.

The important result obtained by such experiments is, that if
an animal be highly immunised by the method mentioned, the
development of the immunity is accompanied by the appearance
in the blood of protective substances, which can be transferred to
another animal. The law enunciated by Behring regarding
immunity against toxins thus holds good in the case of the
living organisms, as was first shown by Pfeiffer. The latter
found, for example, that in the case of the cholera organisms, so
high a degree of immunity could be produced in the guinea-pig,
that '002 c.c. of its serum would protect another guinea-pig
against ten times the lethal dose of the organisms, when injected
along with them. Here again is presented the remarkable
potency of the antagonising substances in the serum, which in
this case lead to the destruction of the corresponding microbe.

The anti-stre2>tococcic serum of Marmorek may be briefly described, as
it has come into extensive practical use. This observer found that he
could intensify the virulence of a streptococcus by growing it alternately
in the peritoneal cavity of a guinea-pig and in a mixture of human blood
serum and bouillon (vide p. 41). The virulence became so enormously
increased by this method, that when only one or two organisms were
introduced into the tissues of a rabbit a rapidly fatal septicaemia was
produced. Streptococci of this high degree of virulence were used first
by subcutaneous, afterwards by intravenous injection, to develop a high
degree of resistance in the horse. Injections were continued over a con-
siderable period of time, and the protective power of the serum was
tested by mixing it with a certain dose of the virulent organisms, and
then injecting into a rabbit. The serum of a horse highly immunised in
this way constitutes the anti-streptococcic serum which has been exten-
sively used in many cases of streptococcic invasion in the human subject.
Marmorek, however, found that this serum had little antitoxic power,
that is, could only protect from a comparatively small dose of toxin
obtained by filtration of cultures.

Anti-typhoid, anti-cholera,1 anti-pneumococcic, anti-plague,
and other sera are all prepared in an analogous manner.

Properties of Antibacterial Serum. — We have here to
consider the three main actions mentioned above, namely,
(a) bactericidal and lysogenic action, (b) opsonic action, and
(c) agglutinative and the closely allied precipitating action. Of

1 A true antitoxic cholera serum has been prepared by Metchnikoff,
E. Iloux, and Taurelli-Saliuibeui.

534 IMMUNITY

these the two first are concerned with the protective property of
an anti-bacterial serum.

(a) Bactericidal and Lysogenic Action. — Pfeiffer found that
if certain organisims, e.g. the cholera spirillum, were injected
into the peritoneal cavity of a guinea-pig highly immunised
against these organisms, they lost their motility almost immedi-
ately, gradually became granular, swollen, and then disappeared
in the fluid — these changes constitute "Pfeiffer's phenomenon."
Further, he showed that the same phenomenon was witnessed if
a minute quantity of the anti-serum was added to a certain
quantity of the corresponding organisms, and the mixture injected
into the peritoneal cavity of a non-treated animal. Pfeiffer
found that the serum of convalescent cholera patients gave the
same reaction as that of immunised animals. He obtained the
same reaction also in the case of the typhoid bacillus and other
organisms. From his observations he concluded that the reaction
was specific, and could be used as a means of distinguishing
organisms which resemble one another. He accordingly con-
sidered that a specific substance was developed in the process of
immunisation, and that this was rendered actively bactericidal
by the aid of the living cells of the body. It was subsequently
shown, however, by Metchnikoff and by Bordet that lysogenesis
might occur outside the body by the addition of fresh peritoneal
fluid or normal serum to the heated immune-serum. PfeifFer
also found that an anti-serum heated to 70° C. for an hour pro-
duced the reaction when injected with the corresponding organisms
into the peritoneum of a fresh animal. The outcome of these
and subsequent researches is to show that when an animal is
immunised against a bacterium a substance appears in its serum
with combining affinity for that particular organism. This
substance, which is generally known as the immune-body,
amboceptor (Ehrlich), or substance sensibilisatrice (Bordet) is
comparatively stable, resisting usually a temperature of 70° C.
for an hour. It cannot produce the destructive effect alone, but
requires the addition of a substance normally present in the serum,
which is spoken of under various names — complement (Ehrlich),
alexine or cytase (French writers). The complement is relatively
unstable, being rapidly destroyed by a temperature of 60° C.,
and it is not increased in amount during the process of
immunisation. Though ferment-like in its instability, it differs
from a ferment in being fixed or used up in definite quantities.

Eecent observations show that complement is not a single substance,
but is really made up of two components. Ferrata, who was the first to
establish this fact, employed the following method : Fresh guinea-pig's

PROPERTIES OF ANTIBACTERIAL SERUM 535

scrum is dialysed against running water for twenty-four hours ; the
precipitate which has formed at the end of that time is separated by the
c.'iitrifugp, washed several times in distilled water, and then dissolved in
normal salt solution. The separated fluid is passed through thick filter
paper. The component in the solution of the precipitate unites directly
with sensitised corpuscles — and then that in the separated fluid enters
into combination ; hence they have been called by Brand "middle-piece"
and "end-piece" respectively. The separation by such a method is,
however, far from being a complete one. Sachs and Altmann have
introduced the following method : To *5 c.c. of fresh guinea-pig's serum
is added 4'1 c.c. of a 7575—3^ normal solution of hydrochloric acid in
distilled water. The sediment is centrifuged off after the mixture has
been allowed to stand at room temperature for an hour, and, after being
washed, is made up with a suitable amount of distilled water. The
separated fluid is neutralised and made isotonic with '4 c.c. of a sV-3V
normal soda solution containing 10 per cent, sodium chloride.

The phenomenon of lysogenesis is, however, only seen in the
case of certain organisms when an animal is highly immunised
against them ; the typhoid and cholera group are outstanding
examples. It is also to be noted that it sometimes is seen in the
case of a normal serum (vide Natural Immunity). In other
cases the bactericidal effect of a serum may occur without the
rapid dissolution characteristic of lysogenesis though other
structural changes may be produced. In still other instances,
e.g. the anti-sera to staphylococci, streptococci, plague bacilli,
etc., a bactericidal effect may be wanting ; nevertheless it may
be shown that an immune-body is developed in the process of
immunisation. This may be done by observing the increased
amount of complement which is fixed through the medium of
the anti-serum (immune-body), sensitised red corpuscles being
used as the- test for the presence of free complement. The
method is described on pp. 128-131.

The all-important action of the immune-body is thus to bring
an increased amount of complement into union with bacteria ;
whether death of the bacteria will result or not will depend
ultimately on their sensitiveness to the action of the particular
complement.

It is to be noted that in the case of a bactericidal serum there
is an optimum amount of immune-body which gives the greatest
bactericidal effect with a given amount of complement. If this
amount of immune -body be exceeded, the bactericidal action
becomes diminished and may be practically annulled. This
result, which is generally known as the Neisser-Wechsberg
phenomenon, has been the subject of much controversy, and
cannot yet be said to be satisfactorily explained. It would
accordingly be out of place to discuss here the different views

536 IMMUNITY

with regard to it. (Regarding some theoretical considerations
as to the therapeutic applications of antibacterial sera, vide
p. 534.)

The laws of lysogenesis are, however, not peculiar to the
case of solution of bacteria by the fluids of the body, but, as has
been shown within the last few years, hold also in the case of
other organised substances, red corpuscles, leucocytes, etc., when
these are introduced into the tissues of an animal as in a process
of immunisation. Of such sera the hsemolytic have been most
fully studied, and, owing to the delicacy of the reaction and the
ease with which it can be observed, have been the means of
throwing much light on the process of lysogenesis, and thus on
one part of the subject of immunity. A short account of their
properties may now be given.

Hcemolytic and other Sera. — It has been known for some time
that in some instances the blood serum of one animal has, in
a certain degree, the power of dissolving the red corpuscles of
another animal of different species ; in other instances, how-
ever, this property cannot be detected. Bordet showed that
if one animal were treated with repeated injections of the
corpuscles of another of different species, the serum of the former
acquired a marked haemolytic property towards the corpuscles of
the latter, the property being demonstrated when the serum is
added to the corpuscles. He also found that the hsemolytic
property disappeared when the hsemolytic serum was heated at
55° C., but, as in the case of a bacteriolytic serum, was regained
on the subsequent addition of some serum from a fresh (i.e. non-
treated) animal. These observations have been fully confirmed
and greatly extended. Ehrlich and Morgenroth analysed the
phenomena in question, and showed that the specially developed
and heat-resisting substance, " immune-body," entered into com-
bination with the red corpuscles at a comparatively low
temperature, namely, at 0° C. ; whereas complement does not
combine at this temperature. In this way a method is supplied
by which the immune-body can be removed from a hamiolytic
serum while the complement is left. They came to the con-
clusion that immune-body combined with the complement
though the combination was less firm and only occurred at a
higher temperature — best about 37° C. They therefore consider
that the immune-body acts as a sort of connecting-link between
the red corpuscle and the complement, hence the term " ambo-
ceptor " which Ehrlich afterwards applied. It may be stated,
however, that the direct union of complement and immune-body
has not been conclusively demonstrated. Muir and Browning,

H^EMOLYTIC AND OTHER SERA 537

for example, found that when a fresh serum is passed through a
Berkefeld filter, complement is largely retained in the pores of
the filter, whereas immune-body passes through practically
unchanged : and that if a mixture of complement and immune-
body be made and filtered at a temperature of 37° C., the
amount of immune-body which passes through is not diminished,
whereas it would be if it had united with the retained comple-
ment. Accordingly by this method there was obtained no
evidence of the direct union of immune-body and comple-
ment. Bordet holds that the immune-body acts merely
as a sensitising agent — hence the term substance sensibili-
satrice — and allows the ferment-like complement to act. It
is quite evident from his writings, however, that he does not
mean, as is often assumed, that the immune-body causes some
lesion in the corpuscle which allows the complement to act, but
simply that it produces in the molecules (receptors) of the red
corpuscles an avidity for complement. All that we can say
definitely at present is that the combination of receptor -f
immune-body takes up complement in firm union while neither
does so alone ; whether the immune-body acts as a link be-
tween the two or not must be left an open question. Even
after the corpuscles are laked with water the receptors are not
destroyed. Muir and Ferguson have shown that they can still
take up immune-body and, through its medium, complement,
just as the intact corpuscles do. Ehrlich and Morgenroth showed
that in some cases the red corpuscles can take up much more
immune-body than is necessary for their lysis, and Muir found
in one case studied, that each further dose of immune-body led
to the fixation of more complement, so that as many as ten
times the htemolytic dose of complement might thus be used up.
It is a matter of considerable importance that the union of
immune-body and red corpuscles can be shown to be a reversible
action. If, as was found by Morgenroth and Muir indepen-
dently, corpuscles treated with several doses of immune-body
and then repeatedly washed in salt solution be mixed with
untreated corpuscles and allowed to remain for an hour, then
sufficient immune-body will pass from the former to the latter,
so that all become lysed on the addition of sufficient complement.
The combination of complement, on the other hand, is usually
of very firm nature. It has been a disputed point whether there
are several distinct complements in a normal serum with
different relations to different immune-bodies, for which Ehrlich
and his co-workers have brought forward a certain amount of
evidence, or whether, as Bordet holds, there is a single comple-

538 IMMUNITY

ment which may, however, show slight variations in behaviour
towards different immune-bodies. There is at least no doubt
that all the complement molecules in a serum are not the same.
For example, Muir and Browning have shown that the treat-
ment of a normal serum with a small amount of emulsion of a
bacterium will remove the bactericidal action for another
bacterium, whereas the amount of complement as tested by
haemolysis is practically unchanged. They accordingly con-
sider that there is a moiety of complement, " bacteriophilic
complement," which is specially concerned in bactericidal action.
On the other hand, many of the arguments adduced by Ehrlich
and his co-workers in favour of a multiplicity of complements
are open to another interpretation ; the truth probably lies
between Ehrlich's and Bordet's views. Workers of the French
school also hold that complement does not exist in the free
condition in the blood, but is liberated from the leucocytes when
the blood is shed. This cannot be held as proved. On the
contrary, there are facts which are strongly in support of the
view that complement exists in the free condition in the circu-
lating blood. There is, however, evidence that the amount of
free complement increases after the blood is shed and some time
later gradually diminishes.

The hsemolytic action of a normal serum can be shown in many cases
to be of the same nature as that of an immune-serum, that is, comple-
ment and the homologue of an immune-body can be distinguished. For
example, guinea-pig's serum is hsemolytic to ox's corpuscles ; if a portion
of serum be heated at 55° C. the complement will be destroyed ; if
another portion be treated with ox's corpuscles at 0° C., the natural
immune-body will be removed and only complement will be left.
Neither portion is in itself hremolytic, but this property becomes manifest
again when the two portions are mixed. Hsemolytic sera are of great
service in the study of the question of specificity. Each is specific in the
sense already explained (p. 521), but the serum developed against the
corpuscles of an animal may have some action on those of an allied
species, that is, some receptors are common to the two species. This fact
can be readily shown by the usual absorption tests, for example, in the
case of an anti-ox serum tested on sheep's corpuscles. A close analogy
holds to what has been established in the case of agglutinins. It is
further of great interest to note that by the injection of red corpuscles
into an animal its serum not only becomes hremolytic, but in many cases
when heated at 55° C. possesses also agglutinating and opsonic properties
towards the red corpuscles used. And further, it would appear that in
some cases at least the immune-body, haemagglutinin, and heemopsonin
are distinct substances. These facts abundantly show how close an
analogy obtains between anti-bacterial and haemolytic sera, and how
important a bearing hsemolytic studies have on the questions of im-
munity in general.

In addition to hsemolytic sera, anti-sera have been obtained by the

OPSONIC ACTION 539

injection of leucocytes, spermatozoa, ciliated epithelium, liver cells,
nervous tissue, etc. The laws governing the production and properties
of these are identical, that is, each serum exhibits a specific property
towards the body used in its production — i.e. dissolves leucocytes, im-
mobilises spermatozoa, etc. The specificity is, however, not so marked
as in the case of sera produced against red blood corpuscles ; thus
a serum produced against tissue cells is often haemolytic ; this is
probably due to various cells of the body having the same receptors.
Here again, when the anti-serum produces no destructive effect on the
corresponding cells, the presence of an immune-body may be demon-
strated by the increased amount of complement which is taken up
through its medium. It may also be mentioned that each anti-serum
usually exhibits toxic properties towards the animal whose cells bave
been used in the injections, e.g. a hrcmolytic serum may produce a fatal
result, with si^ns of extensive blood destruction, haemoglobinuria, etc.,
h.4. it is hiernotoxic for the particular animal ; a serum prepared by in-
jection of liver cells has been found to produce on injection necrotic
changes in the liver in the species of animal whose liver cells were
used. These are mentioned as examples of a very large group of specific
activities.

With regard to the sites of origin of immune-bodies our
information is still very deficient. Pfeiffer and Marx brought
forward evidence in the case of typhoid, and Wassermann in
the case of cholera, that the immune-bodies are chiefly formed
in the spleen, lymphatic glands, and bone-marrow. According
to certain workers of the French school, the chief source of anti-
substances acting on cells such as red blood corpuscles is the
large mononucleated leucocytes, whilst those acting on bacteria
are chiefly derived from the polymorpho-nuclear leucocytes (vide
p. 182). Another view is that immune-bodies are chiefly formed
by the large mononucleated leucocytes, whilst complements are
products of the polymorphs. That these cells are concerned in the
production of antagonistic and protective substances is almost
certain, though another possible source of wide extent, namely,
the endothelium of the vascular system, has been largely over-
looked. 'As yet, definite statements cannot be made on thisjpoint.

(b) Opsonic Action. — The presence of a substance in an
immune-serum which makes the corresponding organism sensi-
tive to phagocytosis was first demonstrated by Denys and Leclef
in 1895, in the case of an anti-streptococcal serum. They also
showed that the serum produced this effect by acting on the
organism, not on the leucocytes. It is, however, chiefly to the
researches of Wright and his co-workers that this subject has
come into special prominence. Wright and Douglas -in their
first paper showed that the phagocytosis of staphylococci by
leucocytes depended on a body in the normal serum which
became fixed to the cocci and made them a prey to the

540 IMMUNITY

phagocytes. To this they gave the name of " opsonin " (vide
pp. 122, 519). There is no phagocytosis of cocci by leucocytes
washed in salt solution ; normal serum heated to 55° C. is also
without effect in inducing this phenomenon. They could not
demonstrate any effect of the opsonin on the leucocytes. On the
other hand, if bacteria be exposed to the fresh serum, and they
be freed from the excess of serum and then exposed to phago-
cytes also washed free from serum, they will be readily taken
up by the cells. It has been abundantly shown that the opsonic
action of the serum is increased by the process of immunisation
against an organism, and the opsonic index represents the
degree o'f immunity in one of its aspects as already explained
(p. 122). The matter has, however, become complicated by the
circumstance that in an immune-serum an opsonin may still be
present after the serum is heated at 55° C., as has been shown
by Dean and others ; and Muir and Martin have shown that
this thermostable immune-opsonin (bacteriotropin of Neufeld)
has all the specific characters of anti-substances in general. On
the other hand, they have found that the thermolabile opsonin
of a normal serum has quite different properties. For example,
when a normal serum is tested on a particular bacterium, the
opsonic effect on that bacterium may be removed by treating the
serum with other bacteria; in other words, the thermolabile
opsonin of normal serum does not possess the specific character
of the opsonin developed in the process of immunisation. They
have also found that various substances or combinations of sub-
stances which act as " complement absorbers " also remove the
opsonic property from a normal serum, while they have no effect
on an immune-opsonin.

That this thermolabile normal opsonin can act in a non-
specific way is shown by the fact that particles of car-
mine and other substances become opsonised by the action
of normal serum. It is, however, to be noted that in certain
cases there have been found in a normal serum traces of sub-
stances which can be activated by thermolabile opsonin after
the manner of immune-body and complement (as seen in the
haemolytic action of a normal serum (p. 538) ; to this extent the
opsonic effect of a normal serum may have some degree of
specificity. From this and other facts some observers have
attempted to explain the whole of opsonic action according to
the scheme of immune-body and complement as seen in h&mo-
lysis. This, however, is not justifiable, ' since normal thermo-
labile opsonin can, as we have seen, act by itself, as can also
the specific immune-opsonin after normal opsonin has been

AGGLUTINATION 541

destroyed by heating, and we know of no corresponding action in
the case of an immune-body. The subject is one of considerable
complexity, but it may be said that the most important varia-
tions in the opsonic content observed in infections depend on
the specific immune-opsonins, though the presence of immune-
body may play a part in raising the index by leading to the
union of more normal-complement-opsonin.

Further study will be necessary before the exact relationships
of these substances are fully understood, and other questions with
regard to them have jis yet scarcely been touched upon.
Increased phagocytic action had long been known by the work
.•f Metelmikotf to be associated with the development of active
immunity, and the theory of stimulation of leucocytes was
supported liv many. The work on opsonins has caused a swing
of the pendulum in the other direction, and points to the
development of anti-substances in the serum as the all-important
factor. It remains to be determined to what extent the opsonic
and directly bactericidal properties taken together will explain
the phenomena of natural and acquired immunity.

(c) A<i<tliitin<itloi>. — Charrin and Roger in 1889 observed
that when the bacillus pyocyaneus was grown in the serum of
an animal immunised against this organism, the growth formed
a deposit at the foot of the vessel ; whereas a growth in normal
serum produced a uniform turbidity. Griiber and Durham, in
investigating Pfeiffer's reaction, found that when a small quantity
of an anti-serum is added to an emulsion of the corresponding
bacterium, the organisms become agglutinated into clumps, this
phenomenon depending upon the presence of definite bodies in
the serum called ayylutinins.

It had been already found that the serum of convalescents
from typhoid fever could protect animals to a certain extent
against typhoid fever, and, in view of the facts experimentally
established, it appeared a natural proceeding to inquire whether
such serum possessed an agglutinative action and at what stage
of the disease it appeared. The result, obtained independ-
ently by Griinbaum and Widal, but first published by the latter,
was to show that the serum possessed this specific action shortly
after infection had taken place ; in other words, the develop-
ment of this variety of anti-substance can be demonstrated at
an early stage of the disease. Agglutination may be said to be
observed generally in bacterial infections, though the degree of
the phenomenon and the facility with which it can be noted vary
greatly in different cases. Details will be found in the chapters
dealing with the individual diseases, etc. Furthermore, the

542 IMMUNITY

phenomenon is not peculiar to bacteria ; it is seen, for example,
when an animal is injected with the red corpuscles of another
species, hcemagglutinins appearing in the serum, which have a
corresponding specificity.

The physical changes on which agglutination depends cannot
as yet be said to be fully understood. Griiber and Durham
considered that the agglutinin produced a change in the envelope
of the bacterium, causing it to swell up and become viscous, but
the facts since established show that this is not the true explana-
tion. For example, it has been shown by Nicolle and by Kruse
that if an old bacterial culture be filtered through porcelain, the
addition of some of the corresponding anti-serum produces a
sort of granular precipitate in it; and that when, as in the
agglutination of bacteria, minute inorganic particles are added
to the mixture, they become aggregated into clumps. The
phenomenon would thus appear to be the result of the inter-
action of the agglutinin and some substance in the bacterial cell
which is known as the agglutinable substance or as the agglu-
tinogen. Joos has found in the case of the typhoid bacillus that
there are two agglutinable substances, which differ in their
resistance to heat — a and ft agglutinogen, and that they give
rise to corresponding agglutinins. Further, as the result of a
comparative study of the agglutinins of a motile and a non-
motile variety of the hog cholera bacillus, Theobald Smith has
come to the conclusion that there is an agglutinin which is pro-
duced by and acts on the flagella and another which is similarly
related to the bacterial bodies. The former acts in very much
higher dilutions than the latter, and this is regarded as an ex-
planation of the fact that in the case of non-motile organisms
the agglutinating serum acts only in proportionately high con-
centration as compared with the case of most motile forms.
Another factor necessary for the phenomenon of agglutination
is a proper salt content. Bordet showed that if the clumps of
agglutinated bacteria are freed from salt by washing in distilled
water they become resolved, and that on the addition of some
sodium chloride they are formed again, and Joos has also brought
forward striking confirmatory evidence as to the necessity for
the presence of salts. It is thus probable that in the pheno-
menon of agglutination as ordinarily understood more than one
factor is concerned, and it is possible that in part it may depend
on some altered molecular relationship of the bacteria to the
surrounding fluid analogous to altered surface tension.

In the phenomenon of agglutination we have to distinguish
two factors, namely, the combination of agglutinin and agglu-

AGGLUTINATION 543

tinaKle sul »st;mce (agglutinogen) and the actual clumping of the
bacteria, and it is to be noted that whether or not the latter
event follows depends on the physical condition of each of the
two substances concerned. For example, in some cases when
the bacteria are heated at a temperature of 65° C., for some time,
they may lose the faculty of being agglutinated while they may
still retain the property of combining with or binding agglutinin.
Dreyer and Jex Blake have observed the remarkable fact
that in certain instances on being heated to a still higher
temperature they may once more become agglutinable. Another
point of practical importance is that bacteria when freshly grown
from the tissues are very often less agglutinable than they after-
wards become when sub-cultured for some time. As stated
above, the agglutinins are usually placed in the second order
of anti-substances, and are regarded as possessing a combining
group and an active or agglutinating group. The constitution
would thus be analogous to that of a toxin, and in conformity
with this view Eisenberg and Volk consider that the agglutinat-
ing group may be destroyed while the combining group
remains, the result being an agglutinoid. The evidence for
this lies in the fact that when an agglutinating serum is heated
to a certain temperature, not only does it lose its agglutinating
action, but when the bacteria are treated with such a serum
their agglutination by active serum is interfered with, a sort of
plugging up of the combining molecules having apparently
taken place. Again, with agglutinating sera partially inacti-
vated by heat or other means, what are known as "^ojie_phenQi_
mejia " occur ; that is, when agglutination occurs with a given
dilution of such a serum a lower dilution may fail to agglutinate,
and this they suppose to be due to the interference of the union
of agglutinin by agglutinoid in the greater concentration of
serum. On the other hand, there are facts which cannot be
brought into harmony with this view. For example, Dreyer and
Jex Blake have shown that the inhibition zone may be slight
when there has been much destruction of agglutinin, and on
the other hand may be well marked when no weakening of the
agglutinating power has resulted from the heating. The physical
changes underlying such phenomena are still very obscure, but
we may say at present that the existence of agglutinoids has
not yet been proved.

Like immune-bodies, agglutinins are not destroyed at 55° C.
(a temperature sufficient to annul bactericidal action), but
different agglutinins show variations in this respect, some being
affected by a temperature little above that named. The resist-

544 IMMUNITY

ance to heat also varies when the serum is diluted with salt
solution, and it has been shown that conditions which interfere
with the coagulation of the proteins increase their resistance.
Like antitoxins, agglutinins seem to be chiefly contained in the
globulin fraction. Discussion has taken place as to the relation
of agglutinins to immune-bodies and as to how far agglutination is
an indication of immunity. It may be said that in the case of
certain sera investigated it has been shown that the immune-
body and the agglutinin are separate substances, but it would
not be justifiable to say this is always the case. And while the
agglutinative power cannot in itself be taken as the measure of
the degree of immunity, agglutinins and immune-bodies are the
products of corresponding reactive processes, and their forma-
tion is governed by corresponding laws. Agglutinins become
fixed in definite proportion by the receptors of the bacteria • that
is, the agglutinin becomes used up in the process of agglutination,
and it has been shown that bacteria may take up many times
the amount necessary to their agglutination — a corresponding
fact to what has been established with regard to immune-bodies
of hsemolytic sera. The agglutinins are specific in the sense
which has been explained above (p. 521). It can be shown by
the method of absorption that in an agglutinating serum there
may be several agglutinins with different combining groups, some
of which may be taken up by organisms allied to that which
has given rise to the anti-serum.

Besides those stated above, other phenomena have been
observed in the interaction of anti-sera and the corresponding
bacteria. For example, it has been shown that when certain
bacteria — e.g. the typhoid bacillus, b. coli, and b. proteus — are
grown in bouillon containing a small proportion of the homo-
logous serum, their morphological characters may be altered,
growth taking place in the form of threads or chains which are
not observed in ordinary conditions. In other instances a serum
may inhibit some of the vital functions of the corresponding
bacterium.

Precipitins. — Shortly after the discovery of agglutinins, Kraus
showed in the case of the organisms of typhoid, cholera and
plague, that the anti-serum not only caused agglutination, but
when added to a filtrate of a culture of the corresponding
bacterium produced a cloudiness and afterwards a precipitate.
To the substance in the immune-serum which brought about
this effect he gave the name of precipitin. Subsequent study
has shown that this phenomenon is closely related to agglutina-
tion ; in fact several authorities consider that they represent the

SERUM PRECIPITIN- 545

same reaction under different conditions, that is, that the sub-
stances which when present in the bacterial bodies give rise to
agglutination, on the addition of the anti-serum, produce a
precipitate when free in a fluid. To test the reaction it is
accordingly necessary to have as far as possible the substance of
the bacteria in solution, and for this purpose there have been
introduced various methods, of which the two following may
be given : —

(«) It is well known that in an old bouillon culture the bacteria undergo
disintegration and their constituents go into solution. Accordingly, if such
a culture which has been kept in the incubator for several weeks be
filtered through a porcelain filter, the filtrate will contain the interacting
substance or precipitinogen.

(b) The growth from a recent agar culture is scraped off and suspended
in normal salt solution, the mixture is made feebly alkaline with soda
solution and boiled for a few minutes. The mixture is then neutralised,
when a precipitate forms, and is filtered through filter-paper ; the filtrate
contains the precipitinogen.

The test is carried out by placing in a number of small test-
tubes a given amount of the bacterial nitrate along with varying
quantities of the homologous anti-serum. (The latter may be
obtained in the usual way by the repeated injection of dead
cultures or of bacterial nitrate.) As the precipitate forms
slowly the tubes should be placed in the incubator for twenty-
four hours, *5 per cent, carbolic acid being added to prevent the
growth of bacteria. This precipitin reaction has now been
observed in a great many bacterial diseases when the patient's
serum is added to the corresponding bacterial nitrate, and has
even been applied by some observers as a means of diagnosis.
It is, however, less delicate and more restricted in its applica-
tion than the agglutination methods.

Serum Precipitins.— This subject does not strictly belong to bacteri-
ology, but the general phenomena are so closely allied to those just
described, that some reference may be made to it. When the serum of
an animal is injected in repeated doses into another animal of different
species, after the type of an immunisation, there appears in the serum of
the animal treated a substance called precipitin, which causes a cloudi-
ness or precipitate when added to the serum (precipitinogen) used. (In
the case of rabbits doses of 3 to 4 c.c. of the serum may be injected intra-
peritoneally at intervals of four to five days, a precipitin usually appearing
at the end of about three weeks.) The reaction, which is a very delicate
one, is conveniently observed by adding a given amount of the anti-
serum, say '05 c.c., to varying amounts of the homologous serum •!, '01,
etc., c.c., in a series of small test tubes, the volume being then made up
with salt solution to 1 c.c. In this wa*y a definite reaction may be
observed with '001 c.c. of the serum or even less. Here again zone

546 IMMUNITY

phenomena, as in the case of agglutination, are met with. If the anti-
serum be heated to a temperature of 75° C. for some time it acquires
inhibitory properties, so that when added to a mixture of serum and anti-
serum which would otherwise give a precipitate, this no longer occurs.
Some observers consider that this is due to the presence of " precipitoid "
in the heated anti-serum ; but the observations of Welsh and Chapman
show that this view is not in accordance with the facts, and indicate that
the inhibition is related to a specific solvent action which the heated anti-
serum lias on the precipitate. They have also shown that the main mass
of the precipitate is furnished by the anti-serum (precipitin). and not as
is usually supposed by the precipitin throwing down the protein of the
homologous serum ; this result is of high importance in connection with
the action of anti-substances in general. The precipitin reaction is specific
in the sense explained above. It is always most marked towards the
serum of the species used in the immunisation ; but while this is so,
there may also be a slight reaction towards animals of allied species. An
anti-human serum, for example, gives the maximum reaction with human
serum, but also a slight reaction with the serum of monkeys, especially of
anthropoid apes ; it, however, gives no reaction with the serum of other
animals. The precipitin test has thus come to be employed as a means of
differentiating human from other bloods. Another interesting phenomenon
is what is known as the " deviation of complement," which is produced by
the combination of the two substances in the serum and anti-serum respect-
ively. If mixtures be made according to the above method, and then a small
quantity of complement, say fresh guinea-pig serum, be added, it will be
found that the complement becomes absorbed, as may be shown by sub-
sequently adding a test amount of sensitised red blood corpuscles. This
deviation phenomenon is even a more delicate reaction than the precipitin
test, it being often possible to demonstrate by its use from a tenth to a
hundredth of the smallest amount of serum which will give a perceptible
precipitate ; it also is specific within the same limits.1

Therapeutic Effects of Anti-Sera. — As will have been gathered,
the chief human diseases treated by anti-sera are diphtheria,
tetanus, streptococcus infection, pneumonia, dysentery, plague,
and snake bite. Of the results of such treatment most is known
in the case of diphtheria. Here a very great diminution in the
mortality has resulted. The diphtheria antitoxin came into
general use about October 1894, and the statistics published by
Eehring towards the end of 1895 indicated results which have
since been confirmed. In the Berlin Hospitals the average
mortality for the years 1891-93 was 36 '1 per cent., in 1894 it
was 21 '1 per cent., and in January-July 1895, 14'9 per cent. The
objection that in some epidemics a very mild type of disease
prevails is met by the fact that similar diminutions of mortality
have occurred all over the world. Loddo collected the results
of 7000 cases in Europe, America, Australia, and Japan, in

1 For an account of precipitins, vide Nuttall, "Blood Immunity and
Relationships," Cambridge, 19*04 ; and of complement deviation, Muir and
Martin, Journ. of Hyg. (1906), vi. p. 265.

THERAPEUTIC EFFECTS OF ANTI-SERA 547

which the mortality was 20 per cent, as compared with a former
mortality in the same hospitals of 44 per cent. It has also been
observed that if during an epidemic the supply of serum fails,
thr mortality at once rises; and in two instances recorded it
was doubled. It must here be ivmt-inbered that from the
spread <>f bacteriological knowledge the diagnosis of diphtheria
is now much more accurate than formerly. Another effect of
the antitoxic treatment has been that when tracheotomy is
necessary the percentage of recoveries is now much higher, being
73 per cent, instead of 27 per cent, in a group of cases collected
by the American Pediatric Society. In the London fever
hospitals, since 1894 the recoveries after tracheotomy have been
56 -4 as compared . with 32*1 per cent, previous to the intro-
duction of antitoxin. One of the most striking results obtained
in the same hospitals is a reduction of the death-rate in post-
scarlatinal diphtheria from 50 per cent, to between 4 per cent,
and 5 per cent. As the disease here occurs while the patient is
under observation, the treatment is nearly always begun on the
first day. It is a matter of prime importance that the treat-
ment should be commenced whenever the disease is recognised.
1 >fh ring showed that in cases treated on the first and second
days of the disease the mortality was only 7 '3 per cent., and this
has been generally confirmed, whilst after the fifth day it was of
little service to apply the treatment. In order to obtain such
results, it cannot be too strongly insisted on that attention
should be given to the dosage. When bad results are obtained,
it may be strongly suspected that this precaution has not been
observed. In the treatment of acute tetanus by the antitoxin
the improvement in results has not been marked, but some
chronic cases have been benefited, and, as already stated (p. 431)
better results are obtained in acute cases if intravenous in-
jection be practised. In the case of Yersin's anti-plague serum,
though some benefit has appeared to follow its use, this has
been of quite a limited nature. The same may be said to be
true of the anti-streptococcic and anti-pneumonic sera, though
in the case of the first mentioned numerous cases of apparently
successful result have been recorded. With regard to anti-
venin, Lamb has shown that, if a cobra with full glands bites a
man, many times the minimal lethal dose are probably injected.
In cases of slight bite, however, benefit may accrue from the use
of the anti-serum.

As has been shown above, antibacterial sera require for their
bactericidal action a sufficiency of complement, and as this
diminishes in amount when a serum is kept, the unsatisfactory

548 IMMUNITY

results with this class of sera may be due to a deficiency of
complement. Or it may be as Ehrlich has suggested, that the
complement naturally existing in human serum does not suit
the immune-body in the anti-serum — that is, is not taken up
through the medium of the latter and brought into combination
with the bacterium. And there is still the further possibility
that even though the complement should be taken up, the
zymotoxic group of the latter is not sufficiently active towards
the bacterium to effect its death. In both cases it will appear
that an extracellular bactericidal action cannot be produced by
the particular immune-body in association with the complement
of the animal in question. There is no doubt that this question
of complements is one of high importance, and that both com-
bining affinity and toxic action of complements must be con-
sidered in each case.

Theories as to Acquired Immunity.

The advances made within recent years in our knowledge
regarding artificial immunity, and the methods by which it may
be produced, have demonstrated the insufficiency of various
theories which had been propounded. Only a short reference
need be made to these. The theory of exhaustion, with which
Pasteur's name is associated, supposed that in the body of the
living animal there are substances necessary for the existence of
a particular organism, which become used up during the sojourn
of that organism in the tissues ; this pabulum being exhausted,
the organisms die out. Such a supposition is, of course, quite
disproved by the facts of passive immunity. According to the
theory of retention, the bacteria within the body were considered
to produce substances which are inimical to their growth, so that
they die out, just as they do in a test-tube culture before the
medium is really exhausted. Such a theory only survives now
in the view that antitoxins are modified toxins, the evidence
against which has already been discussed (p. 530). There then
came the humoral theory and the. theory of phagocytosis, but
neither of these is tenable in its pure form, and the distinction
between them need not be maintained. For, on the one hand,
any substance with specific property in the serum must be the
product of cellular activity, and, on the other hand, the facts
with regard to passive immunity go far beyond the ingestive and
digestive properties of phagocytes, though these cells may be in
part the source of important bodies in the serum. At the pre-
sent time interest centres around two theories, namely, Ehrlich 's

EHRLICH'S SIDE-CHAIN THEORY 549

side-chain theory and Metchnikoff s phagocytic theory as further
developed. These will now be discussed, and it may be noted
that the ground covered by each is not coextensive. For the
former deals chiefly with the production of anti-substances and
its biological significance, the latter deals with the defensive
pro forties of cells, either directly by their phagocytic activity
or indirectly by substances produced by them after the manner
of digestive ferments. It will be seen, however, that each has
a normal process as its basis, namely, that of nutrition.

1. Ehrlich's Side-Chain Theory. — This may be said to be an
application of his views regarding the nourishment of proto-
plasm. A molecule of protoplasm (in the general sense) may be
regarded as composed of a central atom group or executive
centre (Leistungskern) with a large number of side-chains
(Seitenketten), i.e. atom groups with combining affinity for
food-stuffs. It is by means of these latter that the living
molecule is increased in the process of nutrition, and hence
the name receptors given by Ehrlich is on the whole preferable.
These receptors are of three chief kinds corresponding to the
classes of anti-substances described (p. 521); the first has a
single unsatisfied combining group, and merely fixes molecules
of relatively simple constitution — receptor of the first order ;
the second has a combining group for the food molecule, and
another active or zymotoxic group, which leads to some physical
change in it — receptor of the second order ; the third has two
combining groups, one for the food molecule and another which
fixes a ferment '(or complement) in the fluid medium around —
receptor of the third order or amboceptor. These latter receptors
come into action in the case of larger food molecules which
require to be broken up by ferment-action for the purposes
of the cell economy. In considering the application of this idea
to the facts of acquired immunity, it must be kept in view that
all the substances to which anti-substances have been obtained
are, like proteids, of unknown but undoubtedly of very complex
chemical constitution, and that in apparently every case the
;mti -substance enters into combination with its corresponding
substance antigen. The dual constitution of toxins and kindred
substances, as already described (p. 198), is also of importance in
this connection. Now, to take the case of toxins, when these
aiv introduced into the system they are fixed, like fund stiitt's,
1>\ tlirir h;i|>ti>]iliorous groups to the receptors of the cell
protoplasm, but are unsuitable for assimilation. If they are in
sutliciently large amount, the toxophorous part of the toxin
molecule produces that disturbance of the protoplasm which

550 IMMUNITY

is shown by symptoms of poisoning. If, however, they are
in smaller dose, as in the early stages of immunisation, fixation
to the protoplasm occurs in the same way; and as the com-
bination of receptors with toxin is supposed to be of firm
nature, the receptors are lost for the purposes of the cell, and
the combination K.-T. (receptor + toxin) is shed off into the
blood. The receptors thus lost become replaced by new ones,
and when additional toxin molecules are introduced, these new
receptors are used up in the same manner as before. As a result of
this repeated loss the regeneration of the receptors becomes an over-
regeneration, and the receptors formed in excess appear in the
free condition in the blood stream and then constitute antitoxin
molecules. There are thus three factors in the process, namely,
(1) fixation of toxin, (2) over-production of receptors, (3) set-
ting free of receptors produced in excess. Accordingly these re-
ceptors which, when forming part of the cell protoplasm, anchor
the toxin to the cell, and thus are essential to the occurrence of
toxic phenomena, in the free condition unite with the toxin, and
thus prevent the toxin from combining with the cells and exert-
ing a pathogenic action. The three orders of receptors, when
separated from the cells, thus give the three kinds of anti-
substances. Ehrlich does not state what cells are specially
concerned in the production of anti-substances, but from what
has been stated it is manifest that any cell which fixes a toxin
molecule, for example*, is potentially a source of antitoxin.
Cells, to whose disturbance, resulting from the fixation of toxin,
characteristic symptoms of poisoning are due, will thus be
sources of antitoxin, e.g. cells of the nervous system in the case
of tetanus, though the cells not so seriously affected by toxin
fixation may act in the same way. The experimental investiga-
tion of the source of antitoxins has, however, yielded little result,
and no definite statement can be made on the subject.

When we come to consider how far Ehrlich's theory is in
harmony with known facts, we find that there is much in its
favour. In the first place, it explains the difference between
active and passive immunity, e.g. difference in duration, etc. ; in
the former the cells have acquired the habit of discharging anti-
substances, in the latter the anti-substances are simply present
as the result of direct transference. It is also in harmony with
the action of antitoxins, etc., as detailed above, and especially
it affords an explanation of the multiplicity of anti-substances.
For, if we take the case of antitoxins, we see that this depends
upon the combining affinity of the toxin for certain of the cells
of the body, and this again is referred back to the complicated

EHRLICH'S SIDE-CHAIN THEORY 551

constitution of living protoplasm. Furthermore, the^ biological
principle involved is no new one, being simply that of over-
regeneration after loss. It would appear likely that the integrity
of the executive centres of the protoplasm molecules would be
essential to the satisfactory production of side-chains, and this
would appear in accordance with the fact that antitoxin
formation occurs most satisfactorily when there is no marked
disturbance of the health of the animal.

It is to be noted, however, that it does not explain active
immunity apart from the presence of anti-substances in the
serum. For example, an animal may be able to withstand a
much larger amount of toxin than could be neutralised by the
total amount of antitoxin in its serum. This might theoretically
be explained by supposing a special looseness of the cell re-
ceptors so that the toxin-receptor combination became readily cast
oft'. The question, however, arises whether there may not be
really an increased resistance of the cells to the toxophorous
affinities. An observation made by Meyer and Ransom (vide
p. 427) is also difficult of explanation, according to the view
that antitoxin is formed by the cells with which the toxin
combines and on which it acts. They found that in an animal
actively immunised against tetanus and with antitoxin beginning
to appear in its blood, the injection of a single M.L.D. of
tetanus toxin into a peripheral nerve brought about tetanus
with a fatal result. On the other hand, the injection of anti-
toxin into the sciatic nerve above the point of injection of toxin
prevented the latter from reaching the cells of the cord. One
can scarcely imagine an explanation of these facts if antitoxin
molecules were in process of being shed off by the cells of the
nervous system. Further, when the serum of an animal con-
tains a large amount of antitoxin, how does the additional
toxin injected reach the cells in order to influence them as
we know it does'? This also is difficult to understand, unless
the toxin has a greater affinity for the receptors in the cells
than for the free receptors (antitoxin) in the serum. A super-
sensitiveness of the nerve-cells of an animal to tetanus toxin,
sometimes observed even when there is a large amount of
antitoxin in the serum, has been often brought forward as an
objection. But this also may perhaps be explained by there
having occurred a partial damage of the cell protoplasm
by the toxic action in the process of immunisation — an
explanation which, of course, demands that in some way the
fiv-hly introduced toxin may reach the cells in spite of the anti-
toxin in the blood, or it may belong to the group of anaphy lactic

552 IMMUNITY

phenomena described below (p. 558). Further investigation alone
will settle these and various other disputed points, and may
remove many of the apparent objections. At present we may
say, however, that Ehrlich's theory is the only one which even
attempts to explain the cardinal facts of this aspect of immunity.
2. The Theory of Phagocytosis. — This theory, brought
forward by Metchnikoff to explain the facts of natural and
acquired immunity, has been of enormous influence in stimu-
lating research on the subject. Looking at the subject from the
standpoint of the comparative anatomist, he saw that it was
a very general property possessed by certain cells throughout
the animal kingdom, that they should take up foreign bodies
into their interior and in many cases digest and destroy them.
On extending his observations to what occurred in disease, he
came to the conclusion that the successful resistance of an
animal against bacteria depended on the activity of certain cells
called phagocytes. In the human subject he distinguished two
chief varieties, namely — (a) the microphages, which are the
" polymorpho-nuclear " finely granular leucocytes of the blood ;
and (b) the macrophages, which include the larger hyaline
leucocytes, endothelial cells, connective tissue corpuscles, and, in
short, any of the larger cells which have the power of ingesting
bacteria. Insusceptibility to a given disease is. indicated by a
rapid activity on the part of the phagocytes, different varieties
being concerned in different cases, — an activity which may
rapidly destroy the bacteria and prevent even local damage. If
the organisms are introduced into the tissues of a moderately
susceptible animal, there occurs an inflammatory reaction with
local leucocytosis, which results in the intracellular destruction
of the invading organisms. Phagocytosis is regarded by
Metchnikoff as the essence of inflammation. He also showed
that the bacteria may be in a living and active state when they
are ingested by leucocytes. On the other hand, he found that
in a susceptible animal phagocytosis did not occur or was only
imperfect. He also showed that when a naturally susceptible
animal was immunised, the process was accompanied by the
appearance of an active phagocytosis. The ingestion of bacteria
by phagocytes is undoubtedly a phenomenon of the greatest
importance in the defence of the organism. It is known that
amoebae and allied organisms have digestive properties which
are specially active towards bacteria, and from what can be
directly observed, as well as indirectly inferred, there can be no
doubt that such a faculty is also possessed by the phagocytes of
the body. Thus bacteria within these cells are in a position

THE THEORY OF PHAGOCYTOSIS 553

favourable to their destruction, and do in many instances
become destroyed. In fact, observations on phagocytosis in
vitro show that such destruction may in the case of some
organisms occur so rapidly that the actual number observable in
the leucocytes is no indication of the activity of the process.
In other instances, e.g. in gonorrhcea, the ingested organisms
would appear to survive a- considerable time without undergoing
change. Undoubtedly phagocytosis is of the highest importance
in active immunity, as by its means organisms which would not
undergo an extra-cellular death may be killed off. In the process
of immunisation of a susceptible animal we see a negative or
neutral chemiotaxis becoming replaced by positive chemiotaxis.
This has been explained by Metchnikoff as due to an education
or stimulation, of the phagocytes. The recent work on opsonins
shows, however, that this is not the case, as leucocytes from an
immunised animal are not more active in this direction than
those of a normal animal, the all-important factor being the
development of an opsonin in the immune animal. Thus this
phase of immunity comes to be merely a part of the subject of
anti-substances in general.

The digestive ferments of phagocytes or cytases are, according
to Metchnikoff, retained within the cells under normal conditions,
but are set free when these cells are injured, for example, when
the blood is shed. They then become free in the serum by
the breaking up of the cells — the process known as phagolysis —
and they then constitute the alexines, or complements of Ehrlich.
Of these, as has already been said, Metchnikoff thinks there
are probably two kinds — one called macrocytase, contained in
the macrophages, which is specially active toward the formed
elements of the animal body, protozoa, etc. ; and the other,
microcytase, contained within the polymorpho-nuclear leucocytes,
which has a special digestive action on bacteria. It is the
microcytase which gives blood serum its bactericidal properties.
It api>ears to us, however, that Metchnikoff has gone too far in
distinguishing the activities of the two classes of cells so much
as he has done.

When the properties of antibacterial sera, as above described,
are considered in relation to phagocytosis, Ifetehnikoff irives the
following explanation. He admits that the Lmnrane-body is
fixed by the bacteria (or red corpu><-lr-. aa the ease may be),
though he doe- n«.t state that a chemical combination takes
place ; hence In- calls it a fixative (Jisateur). The immune-bodied
are to be regarded as auxiliary ferments (ferments adjuvants)
which aid the action of the alexine. Unlike the latter, however,

554 IMMUNITY

they are formed in excess during immunisation and set free in
the serum. He compares their action to that of enterokinase,
a ferment which is produced in the intestine and which aids
the action of trypsin. Thus, when the bacteria have fixed the
immune-body, their digestion is facilitated either within the
phagocytes, or outside of them when the alexine has been set
free by phagolysis. He, however, maintains that extracellular
digestion or lysogenesis does not take place without the
occurrence of phagolysis. The source of immune-bodies is, in
all probability, also the leucocytes, as these substances are
specially abundant in organs rich in such cells — spleen, lymphatic
glands, etc. ; here again the mono-nuclear leucocytes are probably
the source of the immune-bodies concerned in haemolysis, the
polymorpho-nuclear leucocytes the source of those concerned
in bacteriolysis. Although the immune-bodies are usually
set free in the serum, this is not always the case ; sometimes
they are contained in the cells, and this probably occurs when
there is a high degree of active immunity against bacteria
without a serum having an antibacterial action, the powers of
intracellular digestion being in such cases increased. In this
way the facts of immunity can be explained so far as these
concern the destruction of bacteria.

MetchnikofFs work has less direct bearing on the production of
antitoxins. He admits the fixation of the toxin by the antitoxin
to form a neutral compound, and he apparently considers that
leucocytes may also be concerned in the production of antitoxins.
Apart, however, from antitoxin formation, he considers the
acquired resistance of the cells themselves of high importance
in toxin immunity.

When we consider Metchnikoff's theory as thus extended to
cover recently established facts, it must be admitted that it
affords a rational explanation of a considerable part of the
subject, though the elucidation of the chemiotactic phenomena
during immunisation as explained above detracts from the im-
portance which he attached to the leucocyte. It, however, does
not afford explanation of the multiplicity and specificity of
antitoxins as Ehrlich's does; on the other hand, it is more
concerned with the cells of the body as destroyers or digesters of
bacteria. As regards the subject of antibacterial sera, the results
of these two workers may be said to be in harmony in some of
the fundamental conceptions. And it is of interest to note
'that Metchnikoff, starting with the phenomena of intracellular
digestion, has arrived at the giving off of specific ferments by
phagocytes ; whilst Ehrlich, from his first investigations on the

NATURAL BACTERICIDAL POWERS 555

constitution of toxins, has arrived at an explanation of antitoxins
and immune-bodies also with a theory of cell-nutrition as its
basis. Within the last few years marked progress has thus
been made towards the establishment of the fundamental laws
of immunity.

NATURAL IMMUNITY.

\\Y have placed the consideration of this subject after that of
acquired immunity, as the latter supplies facts which indicate in
what direction an explanation of the former may be looked for.
There may be said to be two main facts with regard .to natural
immunity. The first is, that there is a large number^of bacteria
— the so-called non-pathogenic organisms — which are practically
incapable, unless perhaps in very large doses, of producing patho-
genic effects in any animal ; when these are introduced into the
body they rapidly die out. This fact, accordingly, shows that
the animal tissues generally have a remarkable power of destroy-
ing living bacteria. The second fact is, that there are other
bacteria which are very virulent to some species of animals,
whilst they are almost harmless to other species; the anthrax
bacillus may be taken as an example. Now it is manifest that
natural immunity against such an organism might be due to a
special power possessed by an animal of destroying the organisms
when introduced into its tissues. It might also possibly be due
to an insusceptibility to, or power of neutralising, the toxins of
the organism. For the study of the various diseases shows that
the toxins (in the widest sense) are the weapons by which morbid
changes are produced, and that toxin-formation is a property
common to all pathogenic bacteria. There is, moreover, no
such thing known as a bacterium multiplying in the living tissues
without producing local or general changes, though, theoretically,
there might be. As a matter of fact, however, natural immunity
is in most cases one against infection, i.e. consists in a power
possessed by the animal body of destroying the living bacteria
\\ In -n introduced into its tissues : such a power may exist though
the animal is still susceptible to the separated toxins. We shall
now look at these two factors separately.

1. Variations in Natural Bactericidal Powers. — The funda-
mental fact here is that a given bacterium may be rapidly
destroyed in one animal, whereas in another it may rapidly
multiply and produce morbid effects. The special powers of
destroying organisms in natural immunity have been ascribed
to (a) phagocytosis, and (b) the action of the serum.

(a) The chief factors with regard to phagocytosis have been

556 IMMUNITY

given above. The bacteria in a naturally immune animal, for
example, the anthrax bacillus in the tissues of the white rat, are
undoubtedly taken up in large numbers and destroyed by the
phagocytes, whereas in a susceptible animal this only occurs to
a small extent ; and Metchnikoff has shown that they are taken
up in a living condition, and are still virulent when tested in a
susceptible animal. Variations in phagocytic activity are found
to correspond more or less closely with the degree of immunity
present, but are probably in themselves -capable of explanation.
The fundamental observations of Wright and Douglas show that
in many cases at least, leucocytes • do not ingest organisms in
normal saline solution, and that this is not due to the medium
in which they are, is readily shown by subjecting the organisms
to the action of fresh serum and then washing them ; thereafter,
they are rapidly taken up by the leucocytes in salt solution.
In most cases this result is due to the labile opsonin of normal
serum which has combining affinities for a great many organisms,
as already stated. In other cases more specific substances may
be concerned. But the all-important fact is that whether
phagocytosis occurs or not, appears to depend upon certain bodies
in the serum. As yet we cannot say whether the phagocytosis
in a given serum, observed according to the opsonic technique,
always runs parallel with phagocytosis in the tissues of the
animal from which the serum has been taken. This is a subject
on which extended observations are necessary. But whether or
not phagocytosis in vivo corresponds with that in vitro it is
probably to be explained in the same Avay ; that is, it probably
depends upon the content of the serum. The composition of the
latter, no doubt, is the result of cellular activity, and in this
the leucocytes themselves are in all probability concerned, but
the movements and phagocytic activity of these cells seem to
be chiefly if not entirely controlled by their environments.
Ingestion is, however, only the first stage in the process ; intra-
cellular destruction is the second, and is of equal importance.
What may be called intracellular bactericidal action probably
varies in the case of leucocytes of different animals, but regarding
this our knowledge is deficient, and, further, bacteria may some-
times survive the cells which have ingested them.

(6) When it had been shown that normal serum possessed
bactericidal powers against different organisms, the question
naturally arose as to whether this bactericidal power varied in
different animals in proportion to the natural immunity enjoyed
by them. The earlier experiments of Behring appeared to give
grounds for the belief that this was the case. He found, for

NATURAL SUSCEPTIBILITY TO TOXINS 557

example, that the serum of the white rat, which has a remark-
able immunity to anthrax, had greater bactericidal powers than
that of other animals investigated. Further investigation, how-
ever, has shown that this is not an example of a general law,
and that the bactericidal action of the serum does not vary pari
jrttMn with the degree of immunity. In many cases, however,
non-pathogenic and also attenuated pathogenic bacteria can be
M VM to undergo rapid solution and disappear when placed in a
drop of normal serum. The bactericidal action of the serum
wus specially studied by Nuttall, and later by Buchner and
Hankin, who believe that the serum owes its power to certain
substances in it derived from the spleen, lymphatic glands,
thymus, and other tissues rich in leucocytes. To these
substances Buchner gave the name of alexines ; as already
explained, they correspond with Metchnikoffs cytases and
Ehrlich's complements described above. They can be pre-
cipitated by alcohol and by ammonium sulphate, and in this
respect and in their relative lability correspond with enzymes or
unorganised ferments. Variations in bactericidal power of the
serum as tested in vitro, however, do not explain the presence
or absence of natural immunity against a living bacterium. In
some cases, for example, it has been found to be considerable,
while the organisms nourish in the body, and the animal has no
immunity. In such a case Metchnikoff says that there occurs in
the living body no liberation of alexines by the phagocytes, and
hence no bactericidal action such as occurs when the blood is
shed. In the case of the hsemolytic action of a normal serum,
it has been shown in many instances that in addition to com-
plement a natural immune-body is also concerned (p. 534), and
this would appear to be the rule ; the process being analogous to
what is seen in the case of an artificially developed haemolytic
serum. In certain instances an analogous condition appears to
obtain in a normal bactericidal serum. For example the dog's
serum heated at 58° C. contains a natural immune-body to
anthrax which can be activated by the addition of normal
guinea-pig's serum so as to produce a bactericidal action, though
the latter is by itself without any such effect. At present, how-
ever, the possibility of bactericidal action by complement alone
cannot be excluded, as it appears to combine with many bacteria
without any intermediary. Further work is necessary to deter-
mine whether all the facts regarding natural immunity are ex-
plainable by the opsonic and bactericidal properties of the serum.
2. Variations in Natural Susceptibility to Toxins. — We must
here start with the fundamental fact, incapable of explanation,

558 IMMUNITY

that toxicity is a relative thing, or, in other words, that different
animals have different degrees of resistance or non-susceptibility
to toxic bodies. In every case a certain dose must be reached
before effects can be observed, and up to that point the animal
has resistance. This natural resistance is found to present very
remarkable degrees of variation in different animals. The great
resistance of the common fowl to the toxin of the tetanus bacillus
may be here mentioned (vide p. 425), and large amounts of this
poison can be injected into the scorpion without producing any
effects whatever ; the high resistance of the pigeon to morphia
is a striking example in the case of vegetable poisons. This
variation in resistance to toxins applies also to those which
produce local effects, as well as to those which cause symptoms
of general poisoning. Instances of this are furnished, for
example, by the vegetable poisons ricin and abrin, by the snake
poisons, and by bacterial toxins such as that of diphtheria. We
must take this natural resistance for granted, though it is
possible that ere long it will be explained.

According to Ehrlich's view of the constitution of toxins, it
might be due to the want of combining affinity between the
tissue cells and the haptophorous group of the toxin ; or, on the
other hand, supposing this affinity to exist, it might be due to
an innate non-susceptibility to the action of the toxophorous
group. Certain investigations have been made in order to
determine the combining affinity of the nervous system of the
fowl with tetanus toxin, as compared with that obtaining in a
susceptible animal, but the results have been somewhat contra-
dictory. Accordingly, a general statement on this point cannot
at present be made, though in all probability variations in the
susceptibility to the toxophorous group will be found to play a
very important part. It has been shown by Muir and Browning
by means of haemolytic tests that the toxic activity of complement,
after it has been fixed to the corpuscles, varies very much ; in some
instances an amount of complement, which would rapidly produce
complete lysis of one kind of corpuscle, may have practically no
effect on another, even though it enters into combination. These
results are of importance in demonstrating how the corresponding
molecules of different animals may vary in sensitiveness to toxic
action.

•Super sensitiveness or Anaphylaxis.

Under this heading are to be grouped a number of phenomena
which in their character and results afford a striking contrast to
the state of immunity. The common feature is that repeated in

BUPERSENSinVENESS OR ANAPHYLAXLS 559

jections of certain substances in sub-toxic or non-toxic doses, — a
suitable interval of time elapsing between the injections, — may be
followed by markedly toxic or even fatal symptoms, and a similar
result may follow repeated injectionsof substances which are practic-
ally non-toxic in a single dose. In such cases, then, a condition of
sii] >LTsensitiveness to the particular substance has been established.
The substances which have been found to have the property of
calling forth this condition are of various kinds, including
bacteria and their toxins, animal poisons, and a great many
foreign proteins, e.g. those of serum, milk, egg albumin, etc., and
it is to be noted that they belong to the group of substances
which can act as antigens. Probably no body of known chemical
constitution develops supersensitiveness ; and, just as tolerance,
say to drugs, is to be distinguished from immunity, so ac-
cumulative action is to be distinguished from supersensitiveness.
Of the latter condition the earliest example observed was
probably the special susceptibility of tubercular patients to the
action of tuberculin, to which reference has already been made
(p. 284). At a comparatively early date also it was found, in
the case of diphtheria and tetanus toxins, that in certain
instances the injection of a minute dose followed by another at
a suitable interval might be attended by serious results; and
that this was not an example of accumulative action, was shown
by the fact that the sum of the doses might amount to only
a fraction of a lethal dose. Richet investigated a similar
phenomenon in the case of a toxic substance obtained from the
tentacles of actiniae, to which from its action he gave the name
of "congestin." He found that a certain time-interval between
the injections was necessary ; that after the second injection the
symptoms occurred with remarkable suddenness, and that they
appeared to be practically independent of the size of the first
dose. He applied the term anaphylaxis to the supersensitive
condition, and this has passed into general use ; he found also
that the condition lasted several weeks. Arthus found that
after repeated injections of horse serum in rabbits a stage was
reached at which an additional subcutaneous injection produced
marked oedema and even necrosis, while an intravenous injection,
harmless to an untreated animal, brought about a fatal result.
The period of active research on the subject, however, may be
said to date from the discovery of what is now known as the
" phenomenon of Theobald Smith." This observer found that
guinea-pigs which had been treated with a neutral mixture of
diphtheria toxin and antitoxin might, after a certain interval of
time, succumb on being injected with a quantity of normal horse

560 IMMUNITY

serum. It was afterwards shown — especially by the researches of
Otto and of Rosenau and Anderson — that the sensitising agent
had really nothing to do with the toxin or antitoxin, but was
contained in the normal serum.

After this brief review we may consider some of the
phenomena of serum anaphylaxis, as it is now called. In its
study horse serum has been chiefly employed, but other sera are
also efficient, and guinea-pigs are the most suitable test animals,
though rabbits have also been used ; in the case of mice it is
difficult if not impossible to bring about serum anaphylaxis.
There is first of all the sensitising injection; a guinea-pig is
injected subcutaneously with a minute quantity, e.g. '001 c.c. of
horse serum, though even smaller amounts may be sufficient and
other methods of injection may also be employed. After a
certain number of days, usually twelve as a minimum, anaphylaxis
has been established, and the test for this is usually made by
injecting subcutaneously 5 c.c. of horse serum. In the ana-
phylactic animal severe symptoms occur ; restlessness and hyper-
algesia are followed by evidence of collapse, the temperature falls
markedly, the heart's action becomes weak and the respiration
embarrassed ; finally death may occur. The intravenous injec-
tion of a smaller amount of serum brings about the same result
more rapidly. It is to be noted that anaphylaxis has the
character of specificity, apparently within corresponding limits
to immunity (p. 521); that is, it is manifested only on the re-
injection of the same protein substance as that used in the first
instance. There is also a passive anaphylaxis, as is shown by
the fact that if a certain amount of the serum of an anaphylactic
guinea-pig be injected into a normal one, the latter becomes ana-
phylactic, so that the characteristic symptoms appear in it when
the test amount of horse serum is injected. In most instances
an interval of about twenty-four hours must, however, elapse
between the injections (Otto) ; if the two injections are made
at the same time there is usually no result. Another interesting
observation has been made, namely, that the young of anaphylactic
mothers may also be anaphylactic, and the condition may last
for some time after birth. It is also possible to produce a
condition of anti-anaphylaxis, that is, to vaccinate against the
supersensitive condition. If, for example, the sensitising dose of
horse serum is injected, and then before anaphylaxis is established
(i.e. sometime before the twelfth day) another injection of a
considerable quantity is made, anaphylaxis does not appear, and
the animal is non-susceptible to further injections of small closes
for a considerable period of time.

SUPERSENSITIVENESS OR ANAPHYLAXIS 561

With regard to the mechanism underlying the phenomena
described, practically all observers are agreed that there is a
profound affection of the nervous system ; but it is still an open
question as to \\lirtlnT the severe and practically simultaneous
affections of the other systems are merely secondary, or whether
they are independently produced by some change common to
all. A great fall in the blood-pressure is an important
plu'iiomenon, and is due chiefly to a general vaso-dilatation ;
and it has been pointed out by Auer and Lewis that in the
case of guinea-pigs there occurs a spasm of the muscle fibres in
the fine bronchi and alveolar passages, the chest-wall being fixed
in full inspiration at the time of death. Besredka has shown
that the fatal symptoms are more rapidly produced in an
anaphylactic animal and with a smaller dose of serum, when
the injection is made directly into the brain, than by any other
method. Further, seeing that a single dose of horse serum is not
toxic to the guinea-pig, and that an interval of several days
must elapse before anaphylaxis is established, the majority of
observers consider that at least two substances are concerned,
one of which is contained in normal horse serum, whilst the
other is developed in the guinea-pig in response to the presence
of the first or of some other substance after the manner of an
anti-substance. To this newly developed substance the name
of " anaphylactic reaction-body " is often given. The phenomena
thus depend upon the co-operation of the reaction-body with a
substance or substances in the horse serum, and a rapid union
of the two, probably within the nerve-cells, brings about the
anaphylactic shock. Passive anaphylaxis would thus be due to
the transference of the reaction-body to a fresh animal, and the
interval necessary before the second injection might depend
upon the time required for the reaction-body to accumulate in
.sufficient quantity within the nerve-cells.

lU'sivdka considers that the sensitising and the toxic factors
in the horse serum are not one and the same. He finds that
serum heated to a certain temperature may still have the power
of inducing the condition of anaphylaxis, but has lost the power
of bringing about the toxic phenomena when injected into an
anaphylactic animal. Gay and Adler similarly find that the
sensitising substance (anaphylactin) is contained in the
i-ii^lobulin fraction of the serum while the other is not.
llrsrcdka accordingly puts forward the view that in the horse
si-rum there are two substances or rather factors, namely,
sen8ifii/i*''it»!/'-/i. which is thermostable, and anti-sensibilisin,
which is thermolabile. When the serum is injected the former

562 IMMUNITY

gives rise to sensibilisin as an anti-substance ; and when, after
a suitable time, fresh serum is injected, the anti-sensibilisin com-
bines with the sensibilisin, and thus the anaphylactic shock
results. In view, however, of the specific nature of the
phenomena, it would appear that both sensibilisinogen and anti-
sensibilisin must have the same special combining group for
sensibilisin, and it is accordingly difficult to see why the latter
should not also act as an antigen. He has also found that
when an animal is anaesthetised with ether the anaphylactic
shock may be averted. Other workers at this subject
hold that there are only two substances concerned, and
some consider that the phenomena depends on a process of
precipitation. Friedberger, for example, considers that the real
toxic agent is formed by the action of complement on serum-
precipitate (antigen + precipitin). To this substance he gives the
name " anaphylatoxin," and in support of his view he has shown
that guinea-pig's complement, after it has been allowed to
act for some time on such a precipitate and then removed by
the centrifuge, has acquired toxic properties, and produces the
symptoms of anaphylaxis when injected into a normal guinea
pig. He also points out that during anaphylactic phenomena,
especially in the case of passive anaphylaxis, there is a great
fall of complement in the blood of the animal, and Scott has
brought forward facts which indicate that there is a close
relationship between this fall in complement and the occurrence
of the symptoms in anaphylaxis. On the other hand, Gay and
Southard do not believe in the theory of a reaction-body.
They consider that the condition depends on the presence of a
substance in the serum which they call anaphylactin, and which
persists in the blood of the guinea-pig for a long period of time.
This acts as a slight irritant to the nerve-cells, and produces in
them an increased affinity for certain molecules in the serum.
Accordingly, when the second injection is made, the rapid com-
bination of these molecules with the cells results in the disturb-
ances described. This view has, however, received little support,
and there are various facts against it, especially in relation to
the transference of anaphylaxis. Others, again, e.g. Citron,
consider that supersensitiveness is so closely allied to immunity
as to really represent the earliest stage in its development. At
present it is impossible to express an opinion with regard to the
real nature of the phenomena. Manifestly, however, if they
depend upon the existence of an anti-substance in the nerve:cells
and the cells of other organs, the injection of fresh serum, before
the anti-substance is fully formed, say on the ninth day after

THE SERUM DISEASE IN MAN 563

the first injection, will lead to its combination and thus to its
being used up, and thus the condition of anti-anaphylaxis will
be established.

It is still an open question as to what extent the phenomena
of anaphylaxis just described are of the same nature as the
supersensitiveness manifested by patients suffering from disease
to the products of the corresponding organism, e.y. to tuberculin,
mallein, etc. (pp. 284, 314) ; though in all probability they are at
least similar in essence. It was held for some time as a distinc-
tion that this s u pel-sensitiveness in infections to bacterial products
could not be transferred to another animal, but recent observa-
tions show that in certain circumstances this is possible in the
case of tuberculin. There is no doubt that the supersensitive
condition must play an important part in the clinical manifesta-
tions of many diseases. For example, the sensitiveness of
tubercular patients to tuberculin shows that the symptoms in
this disease are evidently produced by the absorption from the
tubercular foci of a smaller amount of toxin than would be
necessary to produce effects in a normal individual. And the
sensitiveness of the conjunctiva in typhoid fever to the products
of the bacillus suggests that in this disease also supersensitive-
ness plays an important part. It is also possible that the
repeated absorption of proteins, harmless in single doses, may
I fat I to toxic symptoms, and in a similar way may possibly be
explained the relative non-toxicity of the products of certain
bacteria when tested in the usual manner. But with regard to
all these questions, which are of the highest importance, much
further research is still necessary.

The Serum Disease in Man. — This condition, which is
intimately related to suj>ersensitiveness, includes the phenomena
which have been observed after the injection of anti-diphtheric
and other sera. The real factor is the introduction of foreign
sera into the human tissues. As in the case of anaphylaxis, as
above described, there is here also a period of incubation, of eight
to twelve days on the average ; after which, in a certain proportion
of cases (in about 20 per cent, after the injection of a fairly
large amount of horse serum, a group of characteristic symptoms
appear. There may be as prodromal symptoms, swelling and
tenderness at the site of injection, and in the corresponding
lymphatic glands, and thereafter general exanthemata appear.
These are usually of an urticarial type, but may be erythematous
or morbilliform. There is usually moderate pyrexia of a
remittent type, and sometimes cedema and slight albuminuria
are present ; occasionally there are pains in the joints ; there is

564 IMMUNITY

also often leucopenia due to a fall in the number of polymorpho-
nuclear leucocytes. These symptoms last for a few days and
then disappear. Such are the phenomena of the serum disease
after a single injection of the foreign serum. There are, however,
two other types of reaction described by v. Pirquet and Schick,
namely, the immediate and the accelerated reactions. The im-
mediate reaction is seen when a large dose of serum has been
administered and then after a certain interval of time another
dose of serum is injected. This interval is usually from twelve
days to eight weeks, though sometimes as long as six months.
The symptoms of the immediate reaction, which appear shortly
after the injection, or at least within twenty-four hours, are an
intense oedema locally, general exanthemata and pyrexia, though
the general phenomena are often little marked. The symptoms
pass off comparatively quickly, usually within twenty-four hours.
The accelerated reaction is also seen after a second injection,
and it may occur from six weeks up to many months after the
first injection. In the case of the accelerated reaction there is
an incubation period, but it is shorter than in the case of the
first injection, being usually five to seven days ; the symptoms
resemble those in the ordinary reaction as described above, but
are of rather more acute onset and last a shorter time. In the
interval from about the sixth week to the sixth month, there
may occur both the immediate reaction, and also a few days
later an accelerated reaction.

The nature of the serum disease is not yet fully understood,
but in all probability depends upon the development of a reaction-
body or anti-substance, and the combination of this with a sub-
stance in the serum, probably an antigen, leads to the symptoms.
We suppose that the substances in the serum gradually disappear
from the body after the injection ; from about the eighth day
onward anti-substances appear in the blood in large amount,
and if antigens are still present, the combination of the two
brings about the phenomena described. Manifestly, if the
antigens have disappeared before the anti-substances appear in
quantity, there will be no symptoms. At a later period anti-
substances will be present alone in the serum, and then the
injection of fresh antigens brings about an immediate reaction.
After the anti-substances have disappeared, the injection of
fresh serum causes no immediate reaction, but the mechanism
of reaction has been stimulated by the first injection ; anti-
substances thus appear more quickly after the second injection,
hence the reaction is accelerated as compared with the reaction
after the first injection.

APPENDIX A.

SMALLPOX AND VACCINATION.

SMALLPOX is a disease to which much study has been devoted,
owing, on the one hand, to the havoc which it formerly wrought
among the nations of Europe, — a havoc which at the present
day it is difficult to realise, — and, on the other hand, to the
controversies which have arisen in connection with the active
immunisation against it introduced by Jenner. Though there
is little doubt that a contagium vivum is concerned in its
occurrence, the etiological relationship of any particular organ-
ism to smallpox has still to be proved ; and with regard to
Jennerian vaccination, it is only the advance of bacteriological
knowledge which is now enabling us to understand the prin-
ciples which underlie the treatment, and which is furnishing
methods whereby the vexed questions concerned will probably
be satisfactorily settled. We cannot here do more than touch
on some of the results of investigation with regard to the
disease.

Jennerian Vaccination. — Up to Jenner's time the only
means adopted to mitigate the disease had been by inoculation
(by scarification) of virus taken from a smallpox pustule,
especially from a mild case. By this means it was shown that
in the great majority of cases a mild form of the disease was
originated. It had previously been known that one attack of
the disease protected against future infection, and that the mild
attack produced by inoculation also had this effect. This
inoculation method had long been practised in various parts of
the world, and had considerable popularity all over Europe
during the eighteenth century. Its disadvantage was that the
resulting disease, though mild, was still infectious, and thus
might be the starting-point of a virulent form among un-
protected persons. Jenner's discovery was published when
inoculation was still considerably practised. It was founded on
the popular belief that those who had contracted cowpox from

565

566 SMALLPOX AND VACCINATION

an affected animal were insusceptible to subsequent infection
from smallpox. In the horse there occurs a disease known as
horsepox, especially tending .to arise in wet, cold springs, which
consists in an inflammatory condition about the hocks, giving
rise to ulceration. Jenner believed that the matter from these
ulcers, when transferred by the hands of men who dressed the
sores to the teats of cows subsequently milked by them, gave
rise to cowpox in the latter. This disease was thus identical
with horsepox in epidemics of which it had its origin. Jenner
was, however, probably in error in confounding horsepox with
another disease of horses, namely, grease. Cowpox manifests
itself as a papular eruption on the teats ; the papules become
pustules ; their contents dry up to form scabs, or more or less
deep ulcers occur at their sites. From such a lesion the hands
of the milkers may become infected through abrasions, and a
similar local eruption occurs, with general symptoms in the
form of slight fever, malaise, and loss of appetite. It is this
illness which, according to Jenner, gives rise to immunity from
smallpox infection. He showed experimentally that persons
who had suffered from such attacks did not react to inoculation
with smallpox ; and further, that persons to whom he communi-
cated cowpox artificially were similarly immune. The results
of Jenner's observations and experiments were published in 1798
under the title, An Inquiry into the Causes and Effects of the
Variola Vaccince. Though from the first Jennerian vaccina-
tion had many opponents, it gradually gained the confidence of
the unprejudiced, and became extensively practised all over the
world, as it is at the present day.

The evidence in favour of vaccination is very strong. There
is no doubt that inoculation with lymph properly taken from a
case of cowpox, can be maintained with very little variation in
strength for a long time by passage from calf to calf, and such
calves are now the usual source of the lymph used for human
vaccination. When lymph derived from them is used for the
latter purpose, immunity against smallpox is conferred on the
vaccinated individual. It has been objected that some of
the lymph which has been used has been derived from calves
inoculated, not with cowpox, but with human smallpox. It is
possible that this may have occurred in some of the strains of
lymph in use shortly after the publication of Jenner's discovery,
but most of the strains at present in use have probably been
derived originally from cowpox. The most striking evidence in
favour of vaccination is derived from 'its effects among the staffs
of smallpox hospitals, for here, in numerous instances, it is only

RELATIONSHIP OF SMALLPOX TO COWPOX 567

the unvaccinated individuals who have contracted the disease.
While vaccination is undoubtedly efficacious in protecting against
smallpox, Jenner was wrong in supposing that a vaccination in
infancy afforded protection for more than a certain number of
years thereafter. It has been noted in smallpox epidemics
which 1 1 avo occurred since the introduction of vaccination, that
whereas young unprotected subjects readily contract the disease,
those vaccinated as infants escape more or less till after the
thirteenth to the fifteenth years. It has become, therefore, more
and more evident that revaccination is necessary if immunity is
to continue ; and where this is done in any population, smallpox
becomes a rare disease, as has happened in the German army,
where the mortality is practically nil. The whole question of
the efficacy of vaccination was investigated in this country in
1896 by a Royal Commission, whose general conclusions were
as follows : — Vaccination diminishes the liability to attack by
smallpox, and when the latter does occur, the disease is milder
and less fatal. Protection against attack is greatest during
nine or ten years after vaccination. It is still efficacious for a
further j>eriod of five years, and possibly never wholly ceases.
The power of vaccination to modify an attack outlasts its power
wholly to wrard it off. Revaccination restores protection, but
this operation must be from time to time repeated. Vaccination
is beneficial according to the thoroughness with which it is
performed

The Relationship of Smallpox (Variola) to Cowpox
(Vaccinia). — This is the question regarding which, since the
introduction of vaccination the greatest controversy has taken
place ; a subsidiary point has been the inter-relationships within
the group of animal diseases which includes cowpox, horsepox,
sheep-pox, and cattle-plague. With reference to smallpox and
cowpox the problem has been, Are they identical or not 1 There
is no doubt that cowpox can be communicated to man, in whom
it produces the eruption limited to the point of inoculation, and
the slight general symptoms which vaccination with calf lymph
has made familiar. Apparently against the view that cowpox
is a modified smallpox are the facts that it never reproduces in
man a general eruption, and that the local eruption is only
infectious when matter from it is introduced into an abrasion.
The loss of infectiveness by transmission through the body of a
relatively insusceptible animal is a condition of which we have
already seen many instances in other diseases, and the uniformity
of the type of the affection resulting from vaccination with calf
lymph finds a parallel in such a disease as hydrophobia, where,

568 SMALLPOX AND VACCINATION

after passage through a series of monkeys, a virus of attenuated
but constant virulence can be obtained. We have seen there
are good grounds for believing that the virus of calf lymph
confers immunity against human smallpox. In considering the
relationships of cowpox and smallpox, this is an important
though subsidiary point ; for at present it is questionable
whether there are any well-authenticated instances of one
disease having the capacity of conferring immunity against
another. The most difficult question in this connection is what
happens when inoculations of smallpox matter are made on
cattle. Chauveau denies that in such circumstances cowpox is
obtained. He, however, only experimented on adult cows. The
transformation has been accomplished by many observers,
including, in this country, Simpson, Klein, Hime, and Copeman.
The general result of these experiments has been that if a series
of calves is inoculated with variolous matter, in the first there
may not be much local reaction, though redness and swelling
appear at the point of inoculation, and some general symptoms
manifest 'themselves. On squeezing some of the lymph from
such lesion as occurs, and using it to continue the passages
through other calves, after a very few transfers a local reaction
indistinguishable from that caused by cowpox lymph generally
takes place, and the animals are now found to be immune
against the latter. Not only so, but on using for human
vaccination the lymph from such variolated calves, results
indistinguishable from those produced by vaccine lymph are
obtained, and the transitory illness which follows, unlike that
produced in man by inoculation with smallpox lymph, is no
longer infectious. In fact, many of the strains of lymph in use
in Germany at present have been derived thus from the variola-
tion of calves. The criticism of these experiments which has
been offered, namely, that since many of them were performed
in vaccine establishments, the calves were probably at the same
time infected with vaccinia, is not of great weight, as in all the
recent cases at least, very elaborate precautions have been
adopted against such a contingency. And at any rate it
would be rather extraordinary that this accident should happen
in every case. We can, therefore, say that at present there
is the very strongest ground for holding not only that vaccinia
confers immunity against variola, but that variola confers
immunity against vaccinia. The experimentum crucis- for
establishing the identity of the two diseases would of course
be the isolation of the same micro-organism from both, and the
obtaining of all the results just detailed by means of pure

BACTERIA IN SMALLPOX ' 569

cultures or the products of such. In the absence of this
evidence we are at present justified in considering that there is
strong reason for believing that vaccinia and variola are the same
disease, and that the differences between them result from the
relative susceptibilities of the two species of animals in which
they occur naturally.

With regard to the relation of cowpox to horsepox, it is
extremely probable that they are the same disease. Some
epidemics of the former have originated from the horse, but in
other cases such a source has not been traced. Cattle plague
fnmi the clinical standpoint, and also from that of pathological
anatomy, resembles very closely human smallpox. Though
each of the two diseases is extremely infectious to its appropriate
animal, there is no record of cattle-plague giving rise to small-
pox in man or vice versa. When matter from a cattle-plague
pustule is inoculated in man, a pustule resembling a vaccine
pustule occurs, and further, the individual is asserted to be now
immune to vaccination ; but vaccination of cattle with cowpox
lymph offers no protection against cattle-plague, though some
have looked on the latter as merely a malignant cowpox. Sheep-
pox also has many clinical and pathological analogies with
human smallpox, and facts as to its relation to cowpox vaccina-
tion similar to those observed in cattle-plague have been
reported. Smallpox, cowpox, cattle-plague, horsepox, and sheep-
pox, in short, constitute an interesting group of analogous
diseases, of the true relationships of which to one another we
are, however, still ignorant.

Micro-organisms associated with Smallpox. — Burdon Sander-
son and other observers early pointed out that in matter derived
from variolous and vaccine pustules (especially the later stages
of the latter), pyogenic organisms are always present, e.g.
mtaphylococcuz aureus and staphylococcus cereus Jlavus, and many
of the ordinary skin saprophytes also are often present, but no
organism has ever been isolated which on transference to animals
lias been shown to have any specific relationship to the disease.
Streptococci have also been described as agglutinable by the sera
of smallpox patients and of vaccinated persons ; such sera, it may
be said, had no effect on other strains of streptococci. Calmette
and Guerin have described very minute granules in the lymph
\\hirli could not be cultivated, but which persisted after all the
bacteria had been removed. (The method by which the latter
was accomplished was by exciting a leucocytosis in a rabbit's
l>eritoneum and then introducing the vaccinal lymph ; the
leucocytes phagocyted the bacteria so that the lymph no longer

570 SMALLPOX AND VACCINATION

gave cultures on ordinary media. It was, however, still potent
to produce vaccinia.)

Klein and also, independently, Copeman, have observed an organism
in lymph taken from a vaccine pustule in a calf on the fifth and sixth
days, in human vaccine lymph on the eighth day, and in lymph from
a smallpox pustule on the fourth day. To demonstrate the bacilli,
cover-glass films are dried and placed for five minutes in acetic acid (1
in 2), washed in distilled water, dried, and placed in alcoholic gentian-
violet for from twenty-four to forty-eight hours, after which they are
washed in water and mounted. Copeman and Kent also found the
bacilli in sections of vaccine pustules stained by Loffler's methylene-blue,
or by Gram's method. The organisms are *4 to '8 /UL in length, and
one-third to a half of this in thickness. They are generally thinner and
stain better at the ends than at the middle. They occur in groups of
from three to ten in both the lymph and the tissues. In the centre of
their protoplasm there is often a clear globule, which is looked on as a
spore. They have hitherto resisted the ordinary isolation methods, a
fact which is rather in favour of their non-saprophytic nature. By
inoculating fresh eggs with the crusts of smallpox pustules Copeman. has,
however, obtained a growth of a bacillus resembling that found by him
in the tissues. Though subcultures on ordinary media have been
obtained, the pathogenic effects of these have not been fully investigated,
and thus the identity of this bacillus with that seen in the tissues is
not proved.

Various observers have described structures in the epithelial
cells in the neighbourhood of the smallpox or vaccine pustules,
which they have interpreted as being protozoa. Thus Ruffer
and Plimmer describe as occurring in clear vacuoles in the cells
of the rete Malpighii at the edge of the pustule (in paraffin
sections of vaccine and smallpox pustules carefully hardened in
alcohol, and stained by the Ehrlich-Biondi mixture) small round
bodies of about four times the size of a staphylococcus pyogenes,
coloured red by the acid fuchsin, sometimes with a central part
stained by the methyl-green. These are described as multiplying
by simple division, and in the living condition exhibiting
amoeboid movement. Similar bodies have been described by
Reed in the blood of smallpox patients and of vaccinated
children and calves.

These are probably the bodies described by Guarnieri and to
which considerable attention has been paid. They are from
1 to 8 /A in diameter, are round, oval, or sickle-shaped, and stain
by ordinary nuclear dyes. They lie in the cells in spaces
often near the nucleus, and are readily demonstrable in vaccine
pustules and also in the experimental lesions which can be
produced in the rabbit's cornea, the larger bodies being denned
in the cells towards the centre of the lesion. These bodies

NATURE OF VACCINATION" 571

have been looked on by many as protozoa, and Guarnieri himself
stated that multiplication could be seen occurring in them in
fresh lymph, but Ewing and also Prowazek have brought forward
strong evidence for the appearances being due to nuclear changes,
though the latter observer considers them to be the effect of a
specific reaction of epithelial cells against the variolous virus.
Here it may be said Wasielewski has shown that they persist
through 46 transfers on the cornea of the rabbit and, further, no
similar appearances have been found in other skin lesions.
Prowazek examined material fixed in a hot mixture of two-
thirds saturated perchloride of mercury and one-third 98 per
cent, alcohol, washed in 40 per cent, iodine alcohol and stained
in Grenacher's ha3matoxylin, and found bodies in the epithelial
cells 1 to 4 /u, in size, sharply contoured and having ragged
edges as if made up of massed chromosomes. These were often
broader at one end than at the other, and appearances have
been seen which suggest longitudinal division. Prowazek has
also seen these " lymph-bodies," as he has called them, in the
lymph, and he inclines to the idea that they may be protozoa.
Bonhoff and also Carini have described spirochsetes as occurring
in variolous lesions, but this has not been confirmed. Volpino
states that in the epithelial cells in corneal infection in rabbits,
minute motile bodies can be discerned which do not occur in
other corneal inflammations. Future investigations must show
what significance is to be attached to these various observations.

The causal organism of smallpox is probably very small, as,
though there has been some difference in opinion on this point,
there is little doubt that it will pass through the coarser porcelain
filters.

The Nature of Vaccination. — From the facts known regarding
vaccination we are justified in supposing that the principle
underlying the efficacy of this prophylactic is the establishment
of an active immunity against the causal organism, which is
sufficiently lasting to protect the vaccinated individual for a
considerable time. Although the virus of smallpox is unknown,
several attempts have been made by indirect methods to
establish the existence of reactions similar to those occurring
in other immunisations. Thus, in cases of human smallpox
and in animals intravenously injected with the vaccine lymph, it
is stated that the serum when mixed with vaccine lymph acquires
the property of deviating complement, and evidence has also been
obtained by Prowazek that the serum of monkeys infected
subcutaneously contains substances of the nature of anti-bodies,
for, when it is mixed with the lymph, the mixture is not

572 SMALLPOX AND VACCINATION

capable of originating a vaccine pustule in children. Phenomena
of hypersensitiveness on revaccination have also been described.

Considerable attention has been devoted to the study of the
effects of corneal and cutaneous infection in the rabbit and
monkey. Here it has been found that the infection of one
cornea protects that eye against re-inoculation but not the other
eye. Further, it is stated that while cutaneous vaccination
causes the general skin surface after about ten days to become
insusceptible, the cornea may still in the monkey be sensitive
(this last fact is said not to be true for the rabbit). Again,
intraperitoneal infection with lymph is said not to be followed
by cutaneous immunity. Such facts have led some to suppose
that smallpox is essentially a disease of the cutaneous tissues.
In it we would have another example of local infection such as
is found in tubercular leprosy, lupus, and certain other skin
infections. Prowazek strongly holds that in cutaneous vaccinal
infection there is never a distribution of the virus throughout
the organs, but this result has been disputed by other workers.
He also states that when the virus is injected intraperitoneally it
is soon taken up by leucocytes and is not absorbed into the body
fluids.

APPENDIX B.

HYDROPHOBIA.

SYNONYMS. — RABIES : FRENCH, LA EAGE : GERMAN, LYSSA,
DIE HUNDWUTH, DIE TOLLWUTH.

Introductory. — Hydrophobia is an infectious disease which in
nature occurs epidemically chiefly among the carnivora, especi-
ally in the dog and the wolf. Infection is carried by the bite
<>t' a rabid animal or by a wound or abrasion being licked by
such. The disease can be transferred to other species, and
\vhrn once started can be spread from individual to individual
by the same paths of infection. Thus it occurs epidemically
from time to time in cattle, sheep, horses, and deer, and can be
communicated to man. It is questionable whether infection
can take place from man to man, as the saliva of a person
suffering from hydrophobia appears not to contain the virus.
It is to be noted that the virus is apparently extremely potent,
a- cases of infection taking place through an unabraded mucous
membrane by the licking of a rabid animal are on record, and the
experimental applications of the virus to such surfaces as the
mucous membrane of the nose or the conjunctiva is often followed
by infection.

In Western Europe the disease is most frequently observed
in the dog; but in Eastern Europe, especially in Russia,
epidemics among wolves constitute a serious danger both to
other animals and to man. All the manifestations of the disease
point to a serious affection of the nervous system : but inasmuch
as symptoms of excitement or of depression may predominate,
it is customary to describe clinically two varieties of rabies, — (1)
rabies proper, or furious rabies (la raye vraie, la rar/e furieuse :
ill', ratk a' I- \\'utk) ', and (2) dumb madness or paralytic rabies (la
f">/>' mue : die stille Wuth). The disease, however, is essentially
the same in both cases. In the dog the furious form is the
more common. After a period of incubation of from three to

573

574 HYDROPHOBIA

six weeks, the first symptom noticed is a change in the animal's
aspect ; it becomes restless, it snaps at anything which it touches,
and tears up and swallows unwonted objects ; it has a peculiar
high-toned bark. Spasms of the throat muscles come on,
especially in swallowing, and there is abundant secretion of
saliva ; its supposed special fear of water is, however, a myth,
— it fears to swallow at all. Gradually convulsions, paralysis,
and coma come on ; and death supervenes usually about five days
after the appearance of symptoms. In the paralytic form, the
early symptoms are the same, but paralysis appears sooner.
The lower jaw of the animal drops, from implication of the
elevator muscles, all the muscles of the body become more or less
weakened, and death ensues without any very marked irritative
symptoms.

In man the incubation period after infection varies from
fifteen days to seven or eight months, or even longer, but is
usually about forty days. When symptoms of rabies are about
to appear, certain prodromata, such as pains in the wound and
along the nerves of the limb in which the wound has been
received, may be observed. To this succeeds a stage of nervous
irritability, during which all the reflexes are augmented — the
victim starting at the slightest sound, for example. There are
spasms, especially of the muscles of deglutition and respiration,
and cortical excitement evidenced by delirium may occur. On
this follows a period in which all the reflexes are diminished,
weakness and paralysis are observed, convulsions occur, and
finally coma and death supervene. The duration of the acute
illness is usually from four to eight days, and death invariably
results. The existence of paralytic rabies in man has been
denied by some, but it undoubtedly occurs. This is usually
manifested by paralysis of the limb in which the infection has
been received, and of the neighbouring parts ; but while in such
cases this is often the first symptom observed, during the whole
of the illness the occurrence of widespread and progressive
paralysis is the outstanding feature. In man there also occur
cases where the cerebellum and also the sympathetic system seem
to be specially affected.

The Pathology of 'Hydrophobia. — In hydrophobia as in
tetanus, to which it bears more than a superficial resemblance,
the appearances discoverable by an ordinary examination of the
nervous system, to which all symptoms are naturally referred, are
comparatively unimportant. On naked-eye examination, conges-
tions, and, it may be, minute haemorrhages are the only features
noticeable. Microscopically, leucocytic exudation into the peri-

PATHOLOGY OF HYDROPHOBIA 575

vascular lymphatic spaces in the nerve centres has been observed,
and in the cells of the anterior cornua of the grey matter in the
spinal cord, and also in the nuclei of the cranial nerves, various
degenerations have been described. Round the nerve cells in
the grey matter of the cord and medulla Babes described
accumulations of newly-formed cells, and Van Gehuchten observed
a phagocytosis of the cells in the posterior root ganglia and also
in the sympathetic ganglia. Both of these conditions were at one
time thought to be specific of rabies, but this has been found not
to be the case. In the white matter, especially in the posterior
columns, swelling of the axis cylinders and breaking up of the
myeline sheaths have been noted, and similar changes occur also
in the spinal nerves, especially of the part of the body through
which infection has come. In the nervous system also some
have seen minute bodies which they have considered to be cocci,
but there is no evidence that they are really of this nature. The
changes in the other parts of the body are unimportant.

Experimental pathology confirms the view that the nervous
system is the centre of the disease by finding in it a special
concentration of what, from want of a more exact term, we must
call the hydrophobic virus. Pasteur's first contribution to the
subject was to show that the most certain method of infection
was by inserting the infective matter beneath the dura mater.
He found that in the case of any animal or man dead of the
disease, injection, by this method, of emulsions of any part
of the central nervous system, of the cerebro-spinal fluid, or
of the saliva, invariably gave rise to rabies, and also that the
natural period of incubation was shortened. Further, the
identity of the furious and paralytic forms was proved, as
sometimes the one, sometimes the other, was produced, what-
ever form had been present in the original case. Inoculation
into the anterior chamber of the eye is nearly as efficacious as
subdural infection. Infection with the blood or solid organs
of rabid animals does not reproduce the disease. There is
evidence, however, that the poison also exists in such glands as
the pancreas and mamma. Subcutaneous infection with part of
the nervous system of an animal dead of rabies may or may not
give rise to the disease.

In consequence of the introduction of such reliable inoculation
methods, furthrr information has Urn acquired regarding the
spread and distribution of the virus in the body. Gaining
entrance by the infected wound, it early manifests its affinity for
the nervous tissues. It reaches the central nervous system
chiefly by spreading up the peripheral nerves. This can be

576 HYDROPHOBIA

shown by inoculating an animal subcutaneously in one of its
limbs, with virulent material. If now the animal be killed
before symptoms have manifested themselves, rabies can be
produced by subdural inoculation from the nerves of the limb
which was infected. Further, rabies can often be produced from
such a case by subdural infection with the part of the spinal
cord into which these nerves pass, while the other parts of the
animal's nervous system do not give rise to the disease. This
explains how the initial symptoms of the disease (pains along
nerves, paralysis, etc.) so often appear in the infected part of the
body, and it probably also explains the fact that bites in such
richly nervous parts as the face and head are much more likely
to be followed by hydrophobia than bites in other parts of the
body. Again, injection into a peripheral nerve, such as the
sciatic, is almost as certain a method of infection as injection
into the subdural space, and gives rise to the same type of
symptoms as injection into the corresponding limb. Intravenous
injection of the virus, on the other hand, differs from the other
modes of infection in that it more frequently gives rise to
paralytic rabies. This fact Pasteur explained by supposing
that the whole of the nervous system in such a case becomes
simultaneously affected. In certain animals the virus seems to
have an elective affinity for the salivary glands, as well as for
the nervous system. Roux and Nocard found that the saliva of
the dog became virulent three days before the first appearance
of symptoms of the disease.

The Virus of Hydrophobia. — While a source of infection
undoubtedly occurs in all cases of hydrophobia, and can usually
be traced, all attempts to determine the actual morbific cause
have been unsatisfactory. In this connection various organisms
(yeasts, diphtheroid bacilli) have been described as being
associated with the disease, but none of these have been shown
to possess the capacity of producing immunity against the
ordinary hydrophobic virus.

In 1903 Negri described certain bodies as occurring in the
nervous system in animals dying of rabies to which much
attention has since been devoted, and regarding the significance
of which opinion is still divided. As Negri's observations have
been generally confirmed, and as it is probable that the
occurrence of the bodies is specific to the disease, and that their
recognition is of value for diagnosis, we shall describe the
methods for their demonstration.

The method of "Williams and Lowden is to take a piece of the brain
tissue, to squeeze it between a slide and cover-glass, and, sliding off the

PATHOLOGY OF HYDROPHOBIA 577

latter, to make a smear which is then fixed in methyl alcohol for five
minutes and stained by Giemsa's stain (p. 115) for half an hour to three
hours ; the preparation is then washed in tap water for 2-3 min. and
dried. For rapid work, after fixation, equal parts of distilled water and
stain are used instead of the more dilute mixture.

For sections the tissues are left in Zenker's fluid l for 3-4 hours, then
placed in tap water for five minutes, 80 per cent, alcohol with enough
i i ii line added to give it a port wine colour for 24 hours ; 95 per cent,
alcohol and iodine, 24 hours ; absolute alcohol, 4-6 hours ; cleared with
cedar oil and embedded in paraffin of melting point 52° C. ; sections should
1»" •"> to 6 /x thick. For staining, Mallory's methylene-blue eosin is
recommended ; the steps are as follows : xylol ; absolute alcohol ; 95
I >i r cent, alcohol and iodine, J hour ; 95 per cent, alcohol, ^ hour ;
a I isolate alcohol, £ hour ; eosin solution (5-10 per cent, aqueous solution),
20 minutes ; rinse in tap water ; Unna's polychrome methylene-blue
solution diluted 1-4 with distilled water, 15 minutes ; differentiation in
!»5 per cent, alcohol for 1-5 minutes (the preparation being kept in
motion and its progress watched with a low power) ; rapid and careful
dehydration and clearing.

Frothingham recommends a method of making "impression prepara-
tions " of the brain. The part (e.g. hippocampus) is laid on a piece of
wood whose porosity causes it to adhere ; a clean slide is then lowered
upon the tissue and slight pressure applied ; on raising the slide a
thin film of cells preserving their original arrangement is lifted off,
and this can be fixed and stained like a smear, van Gieson's method
being used by this author.

The Xegri bodies (Plate IV., Fig. 16)2 vary much in size,
m'Msuring from *5 to 25 //,. They are round, oval, or angular
in outline. They are found in the protoplasm of the nerve cells
and of their processes. When examined in unstained prepara-
tions, they are seen to have a sharply defined outline, and some
of the features of the internal structure presently to be described
can be noted. With regard to staining reactions, they are
! rankly eosinophil for certain combinations containing eosin,
e.y. alcoholic eosin-methylene-blue, Mann's eosin mixture, and, in
certain circumstances, Leishman's stain. For the finer differen-
tiation of the internal structure, Negri employed Giemsa's stain.
With this stain and under high magnification the groundwork
of the body is a pale blue ; in it there appear certain round or
oval, multiple or single formations, of varying size, stained pink,
sometimes occupying nearly the whole of the body, sometimes
being relatively small (grosse Innenformationen). In addition,

1 Zenker's fluid is of the following composition : potassium bichromate
•J-f> .m-., sodium sulphate 1 gr., perchloride of mercury 5 gr., glacial acetic
a>-i'l 5 c.c., water to 100 c.c. Dissolve the perchloride of mercury and the
l.ulirouiate of potassium in the water with the aid of heat and add the
a.-rtir acid.

- For the material from which this preparation was made we are indebted to
nipt. W. F. Harvey, I. M.S.

578 ' HYDROPHOBIA

both inside the larger formations and in the general protoplasm
of the body are smaller red or violet-red granules, occurring
singly or in clumps (kleine Innenformationen). With the eosin
dyes named above, and magnifications of 800 to 1000, the
smaller bodies appear a homogeneous reddish pink, and in
the larger bodies the outlines of the larger internal formations
can be recognised (see Plate). With Mallory's stain they present
similar appearances with a bluish stippling of the protoplasm.

The Negri bodies have been found in practically 98 per cent,
of cases of street rabies examined by many observers in different
parts of the world. Numerous control observations on other
toxic conditions of the nervous system, especially where these
are characterised by spasms, have been made, and although
occasionally, e.g. in tetanus, a somewhat similar appearance has
been seen, at present the consensus of opinion is in favour of an
experienced observer being able to recognise the Negri bodies as
a specific appearance in nerve cells. The bodies occur in all
parts of the nervous system, but are most common in the
Purkinje cells of the cerebellum, and especially in the cells of
the cornu Ammonis (hippocampus major). It is in the last
situation, therefore, that they are generally looked for. They
are apparently not so readily found, and may be altogether
absent, in animals dying from inoculation with the exalted fixed
virus.- Hitherto they have not been found in the salivary
glands or saliva of a rabid animal.

While there is a general tendency to recognise the Negri
bodies as being specific to rabies, great difference of opinion
exists as to their true nature and as to their possessing any
etiological significance in the disease. Negri himself looks upon
them as protozoa, and the organism has been named by Calkins
neuroryctes hydrophobia*. The chief arguments advanced in
favour of this position have been the constancy of the occurrence
of the bodies in the brains of animals suffering from the natural
disease, and their peculiar structure which, such authorities as
Golgi state, does not correspond to any cellular degeneration.
Against their protozoal nature has been urged their absence
from the virulent brains of animals dying from fixed virus, their
non-discovery in the infected saliva, and the fact that the virus
can pass through a coarse filter. These objections have been met
with the argument that the smaller internal formations may be
the infective agent in its essential form, and a modification of
this view is that the Negri body is a cellular reaction against
an invasion with these ultimate forms. The whole question
must be looked upon as sub judice.

PROPHYLACTIC TREATMENT OF HYDROPHOBIA 579

There is no doubt that between rabies and the bacterial
disease's \\v have studied there are at every point analogies, the
most striking U'ing the protective inoculation methods, the <li-
co\ cry of which constitutes the great work of Pasteur ; and every-
thing points to a micro-organism being the cause. The organism,
whatever it is, is, in its infective form, probably very small, as it
can pass through the coarser Berkefeld filters, and also occasion-
ally through the coarser Chamberland candles. Evidence that it
is the organism itself which passes through, is found in the fact
that when iin animal dies from infection with the filtrate, a
small portion of its central nervous system will originate the
disease in a fresh animal. Judging from our knowledge of
similar diseases, we would strongly suspect that it is actually
present in a living condition in the central nervous system, the
>ali\a. etc., which yield what we have called the hydrophobic
virus, for by no mere toxin could the disease be transmitted
through a series of animals, as we shall presently see can be
done. A toxin may, however, be concerned in the production
of the pathogenic effects. Reinlinger found that death with
paralytic symptoms followed the injection of filtered virus, but
that the nervous system of the dead animals sometimes did not
reproduce rabies. He explains this occurrence by supposing that
the filtrate contained a toxin but not the actual infective agent.
The resistance of the virus to external agents varies. Thus a
iMTvous system containing it is virulent till destroyed by putre-
faction ; it can resist the prolonged application of a temperature
of from - 10° to — 20° C., but, on the other hand, it is rendered
non-virulent by one hour's exposure at 50° C. Again, its
potency probably varies in nature according to the source.
Thus, while the death-rate among persons bitten by mad dogs is
about 16 per cent., the corresponding death-rate after the bites
ot wolves is 80 per cent. Here, however, it must be kept in
view that, as the wolf is naturally the more savage animal, the
number and extent of the bites, i.e. the number of channels of
entrance of the virus into the body and the total dose, are
urreater than in the case of persons bitten by dogs. As we shall
see, alterations in the potency of the virus can certainly be
effected by artificial means.

The Prophylactic Treatment of Hydrophobia. — Until the
publication of Pasteur's researches in 1885, the only means
adopted to prevent the development of hydrophobia in a person
bitten by a rabid animal had consisted in the cauterisation of
the wound. Such a procedure was undoubtedly not without
effect. It has been shown that cauterisation within five minutes

580 HYDROPHOBIA

of the infliction of a rabic wound prevents the disease from
developing, and that if done within half an hour it saves a
proportion of the cases. After this time, cauterisation only
lengthens the period of incubation ; but, as we shall see
presently, this is an extremely important effect.

The work of Pasteur has, however, revolutionised the whole
treatment of wounds inflicted by hydrophobic animals. Pasteur
started with the idea that, since the period of incubation in the
case of animals infected subdurally from the nervous systems of
mad dogs is constant in the dog, the virus has been from time
immemorial of constant strength. Such a virus, of what might
be called natural strength, is usually referred to in his works as
the virus of la rage des rues,1 in the writings of German authors
as the virus of die Strasswuth. Pasteur found on inoculating
a monkey subdurally with such a virus, and then inoculating
a second monkey from the first, and so on with a series of
monkeys, that it gradually lost its virulence, as evidenced by
lengthened periods of incubation on subdural inoculation of
dogs, until it wholly lost the power of producing rabies in dogs,
when introduced subcutaneously. When this point had been
attained, its virulence was not diminished by further passage
through the monkey. On the other hand, if the virus of la
rage des rues were similarly passed through a series of rabbits
or guinea-pigs, its virulence was increased till a constant strength
(the virus fixe) was attained, — constancy of strength being in-
dicated by the unvarying recurrence of paresis on the sixth day.
Pasteur had thus at command three varieties of virus — that of
natural strength, that which had been attenuated, and that
which had been exalted. He further found that, commencing
with the subcutaneous injection of a weak virus, and following
this up with the injection of the stronger varieties, he could
ultimately, in a very short time, immunise dogs against subdural
infection with a virus which, under ordinary conditions, would
certainly have caused a fatal result. He also elucidated the
fact that the exalted virus contained in the spinal cords of
rabbits such as those referred to, could be attenuated so as no
longer to produce rabies in dogs by subcutaneous injection.
This was done by drying the cords in air over caustic potash (to

1 While Pasteur's original statements regarding the constancy of the
virulence of the street-virus were probably accurate for the street dogs of
Paris, it has been found that if the general virulence of virus derived from
animals in nature be studied, considerable variation occurs. It is now
usual to apply the term street- virus to any virus derived from an animal
becoming rabid under natural conditions of infection.

PROPHYLACTIC TREATMENT OF HYDROPHOBIA 581

absorb the moisture), the diminution of virulence being propor-
tional to the length of time during which the cords were kept.
Accordingly, by taking a series of such spinal cords kept for
various periods of time, he was supplied with a series of vaccines
of different strengths. Pasteur at once applied himself to find
whether the comparatively long period of incubation in man could
not be taken advantage of to " vaccinate " him against the
disease before its gravest manifestation took place. The
following is the record of the first case thus treated. The technique
was to rub up in a little sterile bouillon a small piece of the
oord used, and inject it under the skin by means of a hypodermic
syringe. The first injection was made with a very attenuated
virus, i.e. a cord fourteen days old. In subsequent injections
the strength of the virus was gradually increased, as shown in
the table :—

July 7, 1885, J) A.M., cord of June 23, i.e. 14 days old.

6 P.M.

9 A.M.

6 P.M.

11 A.M., cord o

July 1

1 day old.

The patient never manifested the slightest symptom of hydro-
phobia. Other similarly favourable results followed ; and this
prophylactic treatment of the disease quickly gained the con-
fidence of the scientific world, which it still retains.

An important modification in the method which further experience led
Pasteur to make was in the treatment of serious cases, such as multiple
bites from wolves, extensive bites about the head, especially in children,
cases which come under treatment at a late period of the incuhation
stage, and cases where the wounds have not cicatrised. In such cases
the stages of the treatment are condensed. Thus on the first day, say
at 11 A.M. and 4 P.M. and 9 P.M., cords of 12, 10, and 8 days respectively
are used ; on the second day, cords of 6, 4, and 2 days ; on the third
•lay, cords of 1 day ; on the fourth day, cords of 8, 6, and 4 days ; on
the fifth, cords of 3 and 2 days ; on the sixth, cords of 1 day ; and so on
for ten days. In each case the average dose is about 2 c.c. of the emulsion.

The details of the prophylactic treatment with regard to
dosage and virulence of material used vary in different Pasteur
institutes. The most important modification which has within

582 HYDROPHOBIA

recent times taken place is the substitution by Hogyes of
increasing concentrations of a fairly fresh virulent rabbit's
cord for emulsions of cords subjected to decreasing periods of
drying. Equally good results apparently are obtained by this
method, and it is stated that in cases so treated certain
symptoms sometimes following the ordinary treatment, the
gravest of which may be the occurrence of temporary paralyses,
are not so frequently observed. This, according to Harvey and
McKendrick, who have studied the subject very fully, may be
due to the fact that a smaller amount of nerve tissue is injected
under the Hogyes system.

The success of the treatment has been very marked. The statistics of
the cases treated in Paris are published quarterly in the Amiales de
I'lnstitut Pasteur, and general summaries of the results of each year are
also prepared. As we have said, the ordinary mortality formerly was
16 per cent, of all persons bitten. During the ten years 1886-95,
17,337 cases were treated, with a mortality of '48 per cent. It has been
alleged that many people arc treated who have been bitten by dogs that
were not mad. This, however, is not more true of the cases treated by
Pasteur's method than it was of those on which the ordinary mortality
of 16 per cent, was based, and care is taken in making up the statistics
to distinguish the cases into three classes. Class A includes only persons
bitten by dogs proved to have had rabies, by inoculation in healthy
animals of parts of the central nervous system of the diseased animal.
Class B includes those bitten by dogs that a competent veterinary surgeon
has pronounced to be mad. Class C includes all other cases. During
1895, 122 cases belonging to Class A were treated, with no deaths ; 940
belonging to Class B, with two deaths ; and 449 belonging to Class C,
with no deaths. Besides the Institute in Paris, similar institutions exist
in other parts of France, in Italy, and especially in Russia, as well as
in other parts of the world ; and in these similar success has been
experienced. It may be now taken as established, that a very grave
responsibility rests on those concerned, if a person bitten by a mad
animal is not subjected to the Pasteur treatment. Sometimes during or
after treatment there appear slight paralytic symptoms with obstinate
constipation and it may be retention of urine, but these usually pass off
within a few weeks and leave behind no ill effects.

The principles underlying the prophylactic treatment of
rabies raise questions of the highest interest from the standpoint
of immunity. The prime fact is, as has been stated, the taking
advantage of the long period of incubation of the disease in man
to neutralise an infection which may be supposed to be gradually
gathering force. We have here again to deal with an example
of the reinforcement of the natural powers of resistance of the
body in order to enable it to cope with a local pathological
change, the locus in this case being the nervous system. We
are thus unable at present to give a rational explanation of the

MKTHODS 5S3

ellicacy of the treatment, but again attention may be directed to
the bearing which the development of hypersensitiveness may
have to the occurrence of the phenomena of infective disease,
and Harvey and McKendrick draw attention to the fact that
some of the concurrent symptoms associated with the treatment
closely resemble anapliylactic phenomena.

J /////•'//,/»• .s'r rum. — In the early part of the nineteenth century
an Italian physician, Valli, showed that immunity against rabies
could In- conferred by administering through the stomach pro-
gressively increasing doses of hydrophobic virus. Following up
this observation, Tizzoni and Centanni have attenuated rabic
virus by submitting it to peptic digestion, and have immunised
animals by injecting gradually increasing strengths of such virus.
This method is usually referred to as the Italian method of
immunisation. The latter workers showed from this that the
serum of animals thus immunised could give rise to passive
immunity in other animals ; and further, that if injected into
animals from seven to fourteen days after infection with the
virus, it prevented the latter from producing its fatal effects,
even when symptoms had begun to manifest themselves. They
fiirthur succeeded in producing in the sheep and the dog an
immunity equal to from 1-25,000 to 1-50,000 (vide p. 525), and
they recommended the use, in severe cases, of the serum of such
animals in addition to the treatment of the patient by the
Pasteur method. A like serum has been obtained from animals
treated by the ordinary Pasteur method.

Methods. — (1) Diagnosis. — Recent work with regard to the
specificity of the Negri bodies for rabies has led to a modifica-
tion in the procedure to be adopted. Formerly it was advisable
if possible to keep an animal suspected of rabies alive for the
observation of symptoms. While the clinical history of the
animal ought to be carefully obtained, greater information will
be obtained by examination of its hippocampus. The animal
should therefore l»e killed and the brain removed after reflecting
the scalp and cutting through the calvarium with a sharp chisel.
The brain is laid down, vertex uppermost, and the upper parts
of one hemisphere are removed in thin horizontal slices till the
anterior part of the lateral ventricle is reached. The roof of the
ventricle is then cut away with a probe-pointed bistoury, and
the hippocampus will be recognised as the laterally arched ridge
which forms the floor of the ventricle. This may be transversely
incised and parts removed for the making of smears and sections
(p. r,76). ^

In addition to microscopic examination, a small piece of the

584 HYDROPHOBIA

medulla or cord of the suspected animal must be taken, with all
aseptic precautions, rubbed up in a little sterile '75 per cent,
sodium chloride solution, and injected by means of a syringe
beneath the dura mater of a rabbit, the latter having been
trephined over the cerebrum by means of the small trephine
which is made for the purpose. In rabies in the rabbit
symptoms of paresis usually occur in from six to twenty-three
days and death in fifteen to twenty-five days. When the material
for inoculation has to be sent any distance, this is best effected
by packing the head of the animal in ice. The virulence of
organs is not lost, however, if they are simply placed in sterile
water or glycerin in well-stoppered bottles.

(2) Treatment. — Every wound inflicted by a rabid animal
ought to be cauterised with the actual cautery as soon as
possible. By such treatment the incubation period will at any
rate be lengthened, and therefore there will be better opportunity
for the Pasteur inoculation method being efficacious. The
person ought then to be sent to the nearest Pasteur Institute for
treatment. It is of great importance that in such a case the
nervous system of the animal should also be sent, in order that
the diagnosis may be certainly verified.

APPENDIX C.

MALARIAL FEVER.

IT has now been conclusively proved that the cause of
malarial fever is a protozoon of which there are several species.
They belong to the haemosporidia (a sub-class of the sporozoa)
which are blood parasites, infecting the red corpuscles of mam-
mals, reptiles, and birds. The parasite was formerly known as
the licp.inatozoon or plasmodium malarice, although the use of the
latter form is incorrect ; the term hcemamoeba is, however, now
generally employed. The parasite was first observed by Laveran
in 1880, and his discovery received confirmation from the
i in k'i»endent researches of Marchiafava and Celli, and later from the
researches of many others in various parts of the world. Golgi
supplied valuable additional information, especially in relation
t<> the .sporulation of the organism and the varieties in different
types of malarial fever. In this country valuable work on the
subject was done by Mauson, and to him specially belongs the
credit of regarding the exfiagellation of the organism as a
preparation for an extra-corporeal phase of existence. By
induction he arrived at the belief that the cycle of existence
outside the human body probably took place in the mosquito.
It was specially in order to discover, if possible, the parasite in
this insect, that Ross commenced his long series of observations,
which were ultimately crowned with success. After patient and
persistent search, he found rounded pigmented bodies in the wall
of the stomach of a dapple-winged mosquito (a species of
Anopheles) which had been fed on the blood of a malarial
patient. The pigment in these bodies was exactly similar to
that in the malarial parasite, and he excluded the possibility of
their representing anything else than a stage in the life cycle of
the organism. He confirmed this discovery and obtained cor-
responding results in the case of the proteosoma infection of
birds, where the parasite is closely related to that of malaria.
In birds affected with this organism, he was able to trace all

586 MALAEIAL FEVER

stages of its development, from the time it entered the stomach
along with the blood, till the time when it settled 'in a special
form in the salivary glands of the insect. Ross's results were
published in 1898. Exactly corresponding stages were after-
wards found in the case of the different species of the human
parasite, by Grassi, Bignami, and Bastianelli; and these with
other Italian observers also supplied important information
regarding the transmission of the disease by infected mosquitoes.
Abundant additional observations, with confirmatory results,
were supplied by Koch, Daniels, Christophers, Stephens, and
others. Wherever malaria has been studied the result has been
the same. Lastly, we may mention the striking experiment
carried out by Manson by means of mosquitoes fed on the blood
of patients in Italy suffering from mild tertian fever. The
insects, after being thus fed, were taken to London and allowed
to bite the human 'subject, Manson's son, Dr. P. Thurburn
Manson, offering himself for the purpose. The result was that
infection occurred; the parasites appeared in the blood, and
were associated with an attack of tertian fever. Ross's discovery
has not only been a means of elucidating the mode of infection,
but, as will be shown below, has also supplied the means of
successfully combating the disease.

From the zoological point of view the mosquito is regarded
as the definitive host of the parasite, the human subject as the
intermediate host. But in describing the life history, it will be
convenient to consider, first, the cycle in the human body, and,
secondly, that in the mosquito. Various terms have been
applied to the various stages, but we shall give those now
generally used.

The Cycle in the Human Subject. — With regard to this
cycle (Plate V., Fig. 21 a — /), it may be stated that the parasite
is conveyed by the bite of the mosquito in the form of a small
filamentous cell — sporozoite or exotospore, which penetrates a red
corpuscle and becomes a small amoeboid organism or amcebula.
There is then a regularly repeated asexual cycle of the parasite
in the blood, the length of which cycle determines the type of
the fever. During this cycle there is a growth of the amoebuhie
or trophozoites within the red corpuscles up to their complete
development ; schizogony (formerly called sporulation) then
occurs. The onset of the febrile attack corresponds with the
stage of schizogony and the setting free of the merozoites or
enhaemospores, i.e. with the production of a fresh brood of
parasites. These soon become attached to, and penetrate into,
the interior of the red corpuscles, becoming intra-corpuscular

FORMS OF THE MALARIAL PARASITE 587

annul) u he ; the cycle is thus completed. The parasites are most
numerous in the blood during the development of the pyivxia,
and, further, they are also much more abundant in the internal
organs than in .the peripheral blood; in the malignant type, for
example, the process of schizogony is practically confined to the
former,

In addition to these forms which are part of the ordinary
asexual cycle, there are derived from the anuebuhe other forms,
\\hich arc called i/n/m-fnri/fes, or sexual cells. These remain
unaltered during successive attacks of pyrexia, and undergo no
further change until the blood is removed from the human body.
Fn the simple tertian and quartan fevers (vide infra) the gameto-
cvtes resemble somewhat in appearance the fully developed
aiiKL'huUe before sporulation, whereas in the malignant type they
have a characteristic crescent-like or sausage-shaped form ; hence
they are often spoken of as "crescentic bodies" (Plate V., Fig.

--.'.<; .-/).

The various forms of the parasite seen in the human blood
may now be described more in detail.

I . '/'//<• Merozoites (Enhcemospores, Lankester) are the youngest
and smallest forms resulting from the segmentation of the
adult amoebula or schizont. They are of round or oval
shape and of small size, usually not exceeding 2 //, in
dia meter; the size, however, varies somewhat in the different
t \pi-s of fever. A nucleus and peripheral protoplasm can be
distinguished (Fig. 162). The former appears as a small
rounded body which usually remains unstained, but contains a
minute mass of chromatin which stains a deep red with the
I (omanowsky method, the peripheral protoplasm being coloured
fairly deeply with methylene-blue. The merozoites show little or
no anneltoid movement; at first free in the plasma, they soon
attack the red corpuscles, where they become the intra-corpuscular
an in -Iniho. If the blood, say in a mild tertian case, be examined
in the early stages of pyrexia, one often finds at the same time
schi/onts, free merozoites, and the young amcebuhe within the
the red corpuscles.

•_' . / / / / / •" -•-,, , -f , uscular A mcebula' or Trophozoites. — These include
the parasites which have attacked the red corpuscles ; they are at
first situated on the surface of the latter but afterwards penetrate
their substance. They usually occur singly in the red corpuscles,
but sometimes two or more may be present together. As seen
in fresh blood, the youn^-st "i smallest forms are minute colour-
It -ss specks, of about the same si/e as the spores; they exhibit
more or less active amoeboid movement, showing marked

588 MALARIAL FEVER

variations in shape. The amount and character of the amoeboid
movement varies somewhat in different types of fever. As they
increase in size, pigment appears in their interior as minute dark
brown or black specks, and gradually becomes more abundant
(Figs. 158, 159, Plate V., Fig. 21 c, d, e, Fig. 22 e). This
pigment is elaborated from the haemoglobin of the red cor-
puscles, the parasite growing at the expense of the latter.
The red corpuscles thus invaded may remain unaltered in
appearance (quartan fever), may become swollen and pale
(tertian fever), or somewhat shrivelled and of darker tint
(malignant fever). In stained specimens a nucleus may be
seen in the parasite as a pale spot containing chromatin which
may be arranged as a single concentrated mass or as several
separated granules, the chromatin being coloured a deep red by
the Romanowsky method. The protoplasm of the parasite,
which is coloured of varying depth of tint with methylene-blue,
shows great variation in configuration (Fig. 159). The young
parasites not infrequently present a "ring-form," a portion of
the red corpuscle being often enclosed by the parasite. These
ring-forms are met with in all the varieties of the parasite, but
they are especially common in the case of the malignant parasite,
where they are of smaller size and of more symmetrical form
than in the others (Fig. 163).

Within the red corpuscles the parasites gradually increase
in size till the full adult form is reached (Fig. 160). In this
stage the parasite loses its amoeboid movement more or less
completely, has a somewhat rounded form, and contains a
considerable amount of pigment. In the malignant form it only
occupies a fraction of the red corpuscle. The adult parasites
may then undergo schizogony, but not all of them do so ; some
become degenerated and ultimately break down.

3. Schizonts. — In the process of schizogony the nuclear
outline becomes lost, and the chromatin becomes divided into
a number of small granules which are scattered through the
protoplasm ; the latter then undergoes corresponding segmenta-
tion and the small merozoites or enhsemospores result. The
pigment during the process becomes aggregated in the centre
and is surrounded by a small quantity of residuary protoplasm.
(Schaudinn has found in the case of the tertian parasite that
schizogony begins by a sort of primitive mitosis, which is then
followed by simple multiple fission.) The merozoites are of
rounded or oval shape, as above described, and are set free
by the rupture of the envelope of the red corpuscles. The
pigment also becomes free and may be taken up by leucocytes.

FIG. 159.

FIG. 160.

FIG. 161. FIG. 162.

FIGS. 157-162.— Various phases of the benign tertian parasite.

FIG. 157. Several young ring-shaped amuebulse within the red corpuscles, one of

the latter enlarged and showing a dotted appearance. Fig. 158. A larger amoebula

Containing pigment granules. Fig. 159. Two large amoebulae, exemplifying the great

variation in form. Fig. 160. Large amoebula assuming the spherical form and showing

isolated fragments of chromatin— preparatory to schizogony. Fig. 161. Schizont,

which has produced eighteen merozoites, each of which contains a small collection

of chromatin. Fig. 162. A number of merozoites which have just been sit free m

the plasma, x 1000.

FIG. 16-3.

FIG. 164.

FIG. 167.

FIG. 168.

FIGS. 163-168. — Exemplifying phases of the malignant parasite.

FIG. 163. Two small ring-shaped amoebulae within the red corpuscles. Fig. 164. A
"crescent" or gamete showing the envelope of the red corpuscles ; also an amoebula.
Figs. 165-168 illustrate the changes in form undergone by the crescents outside the
body. In the interior of the spherical form in Fig. 167 evidence of the flagella can be
seen. Fig. 168. A male gametocyte which has undergone exflagellation, showing the
thread-like microgametes or spermatozoa attached at the periphery, x 1000. (The
figures in this plate are from preparations kindly lent by Sir Patrick Manson.)

FOIJMS OF THE MALARIAL PARASITE 591

The number and arrangement of the merozoites within the
srhi/ont vary in the different types. In the quartan there are
6-12, and the segmentation is in a radiate manner, giving rise
to the characteristic daisy head appearance; in the tertian they
number ].") •_'<) or more, and have a somewhat rosette-like
arrangement (Fig. 161); in the malignant there are usually
6-20 merozoites of small size and somewhat irregularly arranged.

Gametocytes. — As stated above, these are sexual cells which
are formed from certain of the amoebulce, and which undergo no
further development in the human subject. In the mild tertian
and quartan fevers they are rounded and resemble somewhat
the largest amoebulae. The female cells, macroyametocytes, are
of large size, measuring up to 1 6 /x in diameter ; they contain
coarse grains of pigment, and the protoplasm stains somewhat
deeply witli methylene-blue. The male cells, microffometocytes,
are smaller, and the protoplasm stains faintly; the nucleus,
umerally in the centre, is rich in chromatin. In the malignant
fevers the gametocytes have the special crescentic or sausage-
shaped form mentioned above. They measure 8 to 9 /x in length,
and occasionally a fine curved line is seen joining the extremities
on the concave aspect, which represents the envelope of the red
corpuscle (Fig. 164). They are colourless and transparent, and
are enclosed by a distinct membrane ; in the central part there
is a collection of pigment and granules of chromatin. The male
crescents can be distinguished from the female by their appear-
ance ; in the former the pigment is less dark and more scattered
through the cell, and there are several granules of chromatin ;
in the latter the pigment is dark and concentrated, often in a
small ring, and there are one or two masses of chromatin in the
centre of the crescent (Plate V., Fig. 22 /, g). According to the
Italian observers, the early forms of the crescents are somewhat
fusiform in shape and are produced in the bone-marrow. The
fully developed crescents do not appear in the blood till several
days after the onset of the fever, and they may be found a
considerable time after the disappearance of the pyrexial attacks.
They are also little, if at all, influenced by the administration
of quinine.

It is well known that after a patient has apparently recovered
from malarial fever a relapse may take place without fresh
infection occurring, sometimes several years afterward, and
Schaudiim has published interesting observations bearing on
this point. He has found, and his observations on this point
have been confirmed, that the macrogametocyte of tertian fever
may by a process of parthenogenesis give rise to merozoites,

592 MALARIAL FEVER

which in their turn infect the red corpuscles and start the cycle
again. As described and figured by him, the chromatin of the
macrogametocyte divides first into two portions, one of which is
smaller and stains more deeply than the other. This more
deeply staining portion then divides, and the protoplasm becomes
segmented as in ordinary schizogony, and a young brood of
parasites results. The more faintly-staining chromatin along
with part of the protoplasm breaks up and disappears.

The Cycle in the Mosquito. — As already explained, this
starts from the gametocytes. After the blood is shed, or after it
is swallowed by the mosquito, two important phenomena occur,
namely, (a) the full development of the sexual cells or gameto-
cytes, and (b) the impregnation of the female (Plate V., Fig.
21 tn-y). If the blood from a case of malignant infection be
examined in a moist chamber, preferably on a warm stage, under
the microscope, both male and female gametocytes may be seen
to become oval and afterwards rounded in shape (Figs. 165-167).
Thereafter, in the case of the male cell, a vibratile or dancing
movement of the pigment granules can be seen in the interior, and
soon several flagella-like structures shoot out from the periphery
(Fig. 168). They are of considerable length but of great fine-
ness, and often show7 a somewhat bulbous extremity. By the
Romanowsky method they have been found to contain a delicate
core of chromatin, which is covered by protoplasm. They
represent the male cells proper, that is, they are sperm-cells or
spermatozoa; they are also known as microgametes. They
become detached from the sphere and move away in the
surrounding fluid. In the female cell, which has also assumed
the rounded form, maturation takes place by the giving off of
part of the nuclear chromatin, this process corresponding to
the formation of a polar body. Impregnation occurs by the
entrance of a microgamete, the chromatin of the two cells after-
wards becoming fused. Impregnation was first observed by
McCallum in the case of halteridium, and, he found that the
female cell afterwards acquired the power of independent move-
ment or became a " travelling vermicule." He also observed
the impregnation of the malignant parasite. The fertilised
female cell is now generally spoken of as a zygote or ookinete.

It has been established that the phenomena just described
occur within the stomach of the mosquito, and that the fertilised
cell or zygote penetrates the stomach wall and settles between
the muscle fibres ; on the second day after the mosquito has
ingested the infected blood small rounded cells about 6 to 8 //.
in diameter, and containing clumps of pigment, may be found in

VARIETIES OF THE MALARIAL PARASITE 593

this position. (It was in fact the character of the pigment
which led Ross to believe that he had before him a stage in the
development of the malarial parasite.) A distinct membrane
called a sporocyst forms around the zygote, and on subsequent
days a great increase in size takes place, the cysts coming to
project from the surface of the stomach into the body cavity.
The zygote divides into a number of cells called blastophores or
gporoMatts, and these again divide and form a large number of
tili form cells which have a radiate arrangement; these were
called by Ross "germinal rods," but are now usually known as
.<)><> rozoites or exotospores (in contradistinction to the enhaemospores
of the human cycle). The full development (sporogony) within
the sporocyst occupies, in the case of proteosoma, about seven
days, in the case of the malarial parasites a little longer.
When fully developed the cyst measures about 60 JJL in dia-
meter, arid appears packed with sporozoites. It then bursts, and
the latter are set free in the body cavity. A large number settle
within the large veneno-salivary gland of the insect, and are thus
in a position to be injected along with its secretion into the
human subject. The sporozoites enter red corpuscles and become
ama-lnihe as above described. Daniels found that in the case
of the malignant parasite an interval of twelve days at least
intervened between the time of feeding the mosquito and the
appearance of the sporozoites in the gland.

It will thus be seen that in the human subject the parasite
passes through an indefinite number of regularly recurring
asexual cycles, with the giving off of collateral sexual cells, and
that in the mosquito there is one cycle which may be said to
start with the impregnation of the female gamete.

Varieties of the Malarial Parasite. — The view propounded
by Laveran was that there is only one species of malarial
parasite, which is polymorphous, and presents slight differences
in structural character in the different types of fever. It may,
however, now be accepted that there are at least three distinct
species which infect the human subject. Practically all are
agreed as to a division into two groups, one of which embraces
the parasites of the milder fevers — " winter-spring " fevers of
Italian writers — there being in this group two distinct species,
!'•>;• the quartan and tertian types respectively ; whilst the other
includes the parasites of the severer forms — " sestivo-autumnal "
fevers, malignant or pernicious fevers of the tropics, or irregu-
larly remittent fevers. There is still doubt as to whether there
are more than one species in this latter group. Formerly
Italian writers distinguished (1) a quotidian; (2) a non-pig-

594 MALARIAL FEVER

mented quotidian ; and (3) a malignant tertian parasite, though
the morphological differences described were slight. Further
observations have, however, thrown doubt on this distinction,
and the evidence rather goes to show that there is a single
species. Opinion also varies as to the cycle of this parasite ;
according to some observers it is twenty-four hours, according to
others forty-eight hours, though there is more evidence in
support of the latter view ; and the term " malignant tertian "
is frequently used. The fever is often of an irregular type and
multiple infection is probably common. Although the question
cannot be considered as finally settled, we shall speak of three
species of human parasites. The zoological position may be
shown by the following scheme, generally followed by English
writers, the terminology being chiefly that of Grass! and
Feletti :—

Family : H.EMAJVKEBID^ (Wasielewski).

Genus I. Hsemamceba. The mature gametes resemble in form the
schizonts before segmentation has occurred.

Species 1. ffcemamceba Danileicski or halter idiuin.
Parasite of pigeons, crows, etc.

Species 2. ffccmamceba rclieta or proteosoma.
Parasite of sparrows, larks, etc.

Species 3. Hccmamccba malaria;.

Parasite of quartan fever of man.

Species 4. Hcemamceba vivax.

Parasite of tertian fever of man.

Genus II. Hsemomenas. The gametocytes have a special crescentic
form.

Species : Hccmomenas prcccox.

Parasite of malignant or festive-autumnal fever of man.

In addition there are other species belonging to the same
family of blood parasites, which infect monkeys, bats, frogs,
lizards, etc., especially in malarial regions.

We shall now give the chief distinctive characters of the three
human parasites : —

1. Parasite of Quartan Fever. — The cycle of development in
man is seventy-two hours, and produces pyrexia every third day ;
double or triple infection may, however, occur. In fresh speci-
mens of blood the outline is more distinct than that of the
tertian parasite, and amoeboid movement is less marked. Only
the smaller forms show movement, and this is not of active
character. The infected red corpuscles do not become altered

THE THREE HUMAN PARASITES 595

in si/A- or appearance, and the pigment within the parasite is in
the form of coarse granules, of dark brown or almost black
colour. The fully developed schizont has a " daisy-head "
appearance, dividing by regular radial segmentation into from
>ix to twelve merozoites, which, on becoming free, are rounded
in form.

2. The Para*<f< <>,' Mild Tertian Fever. — The cycle of de-
velopment is completed in forty-eight hours, though a quotidian
type of fever may be produced by double infection. The
amcebuki' have a less refractile margin than in the quartan type,
and arc thus less easily distinguished in the fresh blood; the
anueboid movements are, however, much more active, while
longer and more slender processes are given off. The infected
corpuscles become swollen and pale, and may show deeply
stained points by the Romano wsky method — " Schiiffner's dots."
The pigment within the parasite is fine and of yellowish-brown
tint. The mature schizont is rather larger than in the
quartan, has a rosette appearance, and gives rise to from fifteen
to twenty merozoites, though sometimes even more occur; these
have a somewhat oval shape.

In both the quartan and tertian fevers all the stages of
development can be readily observed in the peripheral blood.
The gametocytes have a rounded form as described above. '

3. The Parasite of Malignant or JKstivo-autumnal Fever or
Ti-'ij-ii-nl Mnlnriii. — The cycle in the human subject probably
occupies forty- eight hours, though this cannot be definitely stated
in l>e always the case (vide mpra). The amcebulae in the red
corpuscles are of small size, and their amoeboid movements are
very active ; they often, however, pass into the quiescent ring
form (Fig. 163). The pigment granules, even in the larger
forms, are few in number and very fine ; the infected red
corpuscles have a tendency to shrivel and assume a deeper or
coppery tint, sometimes they are swollen and decolorised. The
fully <leveloj>ed schizont usually occupies less than half the red
corpuscle, and gives rise to from six to twenty merozoites, some-
what irregularly arranged and of minute size. Schizogony
takes place almost exclusively in the internal organs, spleen,
etc., so that, as a rule, no sporocytes can be found in the blood
taken in the usual way. The proportion of red corpuscles infected
by the amcebula? is also much larger in the internal organs. The
gametocytes have the crescentic form, as already described.

Cases of infection with the malignant parasite sometimes
assume a pernicious character, and then the number of organisms
in the interior of the body may be enormous. In certain fatal

596 MALARIAL FEVER

cases with coma the cerebral capillaries appear to be almost
filled with them, many parasites being in process of sporulation ;
and in so-called algid cases, characterised by great collapse, a
similar condition has been found in the capillaries of the
omentum and intestines. The process of blood destruction,
present in all malarial fevers, reaches its maximum in the
malignant class, and the brown or black pigment elaborated by
the parasites — in part after being taken up by leucocytes, chiefly
of the mononuclear class — becomes deposited in various organs,
spleen, liver, brain, etc., especially in the endothelium of
vessels and the perivascular lymphatics. In the severer forms
also brownish yellow pigment is apparently derived from liberated
haemoglobin, and accumulates in various parts, especially in the
liver cells ; most of this latter gives the reaction of haemosiderin.
General Considerations. — The development of the malarial
parasites in the mosquito and infection of the human subject
through the bites of this insect, have, by the work of Ross and
others, as detailed above, become established scientific facts.
These facts, moreover, point to certain definite methods of pre-
vention of infection, which have to a certain extent already been
practically tested. The extensive observations recently carried
out go to show that all the mosquitoes which act as hosts of the
parasite belong to the genus anopheles ; of these there are a
large number of species, and in at least eight or nine the
parasite has been found. Some of these anopheles occur in
England, especially in regions where malaria formerly prevailed.
The opportunity for infection from cases of malaria returning
from the tropics to this country thus exists, and such infection
has occurred. The breeding places of the insects are chiefly,
though not exclusively, in stagnant pools and other collections
of standing water, and accordingly the removal, where practicable,
by drainage of such collections in the vicinity of centres of popu-
lation, the covering in of wells, etc., and the killing of the larvae
by petroleum sprinkled on the water, have constituted the most
important measures in localised areas. This procedure has been
carried out in various places, for example, in Freetown and
Ismailia, with marked success. On the other hand, where there
are large populous areas, as in India, it has been found almost
impracticable to carry out these measures with any success,.
Another measure is the protection against mosquito bites by
netting, it being fortunately the habit of the anopheles to rarely
become active before sundown. The experiments of Sambon
and Low in the Campagna proved that individuals using these
means of protection may live in a highly malarial district with-

THE PATHOLOGY OF MALARIA 597

out becoming iufected. The administration of quinine to
persons living in highly malarial regions, in order to prevent as
well as to treat infection, has also been recommended and
carried out, and the general agreement appears to be that in
India the properly controlled administration of quinine must,
in the meantime at least, be the chief means of combating the
disease. In the tropics the natives in large proportion suffer
from malarial infection, and one would accordingly expect that
infection of the mosquitoes in the neighbourhood of native
settlements would be common. This has been found to be
;ictually the ca.se, and it has accordingly been suggested that the
dwellings of whites should as far as possible be at some distance
from the native centres of population.

So far as is known, none of the lower animals have been
found to take the place of man as intermediate host to the
parasites of malaria, but the possibility of such being the case
cannot be as yet definitely excluded. On the death of infected
mosquitoes the exotospores or sporozoites will become set free,
and therefore theoretically there is a possibility that they may
enter the human subject by inhalation or by some other means.
\Ve have no facts, however, to show that this really occurs, and
the evidence already obtained establishes the bites of mosquitoes
as the most important if not the only mode of infection.

It may also be mentioned as a scientific fact of some interest,
though not bearing on the natural modes of infection, that the
disease can also be communicated from one person to another by
injecting the blood containing the parasites. Several experi-
ments of this kind have been performed (usually about J to 1 c.c.
of blood has been used), and the result is more certain in
intravenous than in subcutaneous injection. In such cases there
is an incubation period, usually of from seven to fourteen days,
after which the fever occurs; the same type of fever is re-
produced as was present in the patient from whom the blood \\a-
taken.

The Pathology of Malaria. — While much work has been
done on the malarial parasite, relatively less attention has been
directed to the processes by whish it produces its pathogenic
effects. It may be said that the organisms are not always
equally prevalent in the circulating blood, and probably at
certain stages tend to be confined in the solid organs; thus they
may be scanty at the height of the paroxysm. Some of the
pathogenic effects are probably associated with particular stages
in the life cycle. Thus the pyrexia occurs when the stage of
M hi/ogony is actively in progress. No opinion can be stated,

598 MALARIAL FEVER

however, as to the cause of the fever, — whether it is due to a
toxic process or to general disturbance of metabolism. We can
better explain the anaemia which is so pronounced in cases where
the disease is of long standing, and which is due to the actual
destruction of red blood corpuscles. The parasite in its sojourn
in these cells absorbs their pigment and thus destroys their
function ; this is further evidenced by the activity displayed by
the red marrow in its attempts to make good the loss sustained
by the blood. One of the most interesting events in malaria,
and one that links it with bacterial infections, is the reaction of
the colourless cells of the blood. It has been shown that during
the apyrexial stages the total number of leucocytes may be
diminished, but that there is always an increase of the mono-
nuclear cells, these frequently numbering 20 per cent, of the
whole, and sometimes even outnumbering the polymorphs.
This is such an important feature that in cases where the
parasites themselves cannot be demonstrated in the blood, the
mononuclear reaction along with the presence of pigment in the
mononuclear cells (due to phagocytosis of pigmented parasites),
has been taken as evidence that the case is really one of malaria.
The mononuclear reaction is specially interesting from the fact
that in other protozoal diseases an activity of the same elements
has been observed.

The question of the possibility of immunity to malaria being
developed naturally arises, and this is especially interesting in
the light of the leucocytic reaction which we have seen must be
looked on as an element in immunity against bacterial infection.
With regard to Europeans developing immunity, it is difficult to
speak. In such a malaria-stricken region as the West Coast of
Africa, the death-rate in residents of more than four years'
standing is less than in the previous years, but this may be due
to the survival of the more resistant immigrants. But there can
be little doubt that malaria in the negro is a much less serious
condition than in the European. Koch from his observations in
New Guinea attributes this to the infection of the native children
leading to the development of immunity in the adult community.
He found, what has been independently noted by Stephens and
Christophers in West Africa, that the greater number of the
children harboured malarial parasites in their blood. The wide-
spread presence of parasites in children might appear to preclude
the immunity of the adult being due to survival of the most
resistant, but the infant mortality in these regions may be very
high, and such a survival may be the real explanation. On the
other hand, Koch states that while an immunity appears to exist

THE PATHOLOGY OF MALARIA 599

in native adults in malarial districts, this is only true of those
born in the locality, — natives coming from neighbouring non-
malarial districts into the malarial region being liable to contract
the disease. At present it must be held that the facts available
do not enal tie us to determine the relative parts played by the
development of artificial immunity on the one hand, and the
existence of a natural immunity on the other, in apparent
insusceptibility to malaria.

Our knowledge on the relationship of blackwater fever to
malaria is also in an unsatisfactory condition. Blackwater fever
is a condition often occurring, especially in Europeans, in tropical
countries. It is characterised by pyrexia, darkly-coloured urine,
— the colour being due to altered haemoglobin pigment, — delirium
and collapse, frequently ending in coma and death. By some
the condition has been looked on as a separate disease, by others
as the terminal stage of a severe malaria. With regard to the
former view no special parasite has yet been demonstrated.
Stephens sums up the evidence for the second view by saying
that malaria, apart from the occurrence of blackwater fever, is a
relatively non-fatal disease; that in the great majority of cases
there is direct or indirect evidence of the subject of the condition
having suffered from repeated attacks of malaria ; that while in
all cases there must be an agent at work causing ha3molysis,
there is evidence that in many cases there is the possibility of
that agent being quinine. In a recent important work, Christo-
phers and Bentley come to the conclusion that the essential
feature in blackwater fever is an extra-cellular destruction of
red corpuscles in the blood plasma, a lysaemia as they call it,
but that this is not directly due to parasitic, osmotic, or chemical
actions, but to a specific haemolysin arising in the body as the
result of the repeated blood destruction. They have shown, for
example, that the addition of fresh serum (complement) to the
red corpuscles of blackwater fever, as well as of malarial, patients
may produce lysis, this apparently being due to a substance
corresponding to immune-body united to the corpuscles in
qui'stion. The development of this haemolysin (autolysin) results
from the extensive and repeated destruction of red corpuscles by
the malarial parasite. Thus though the latter is not the
immediate cause of the lysaemia, which is the essential feature of
l>laekwater fever, it is the means of inducing the development
of the haemolysin. If this view of the process is found to be
i oiroct, it would of course explain the relationship of malaria to
the condition. They also consider that in the conditions men-
tioned, i.e. where there has been repeated destruction of an

GOO MALARIAL FEVER

individual's corpuscles by the malarial parasite, the occurrence of
lysaemia may be precipitated by an acute attack of malaria
especially when under certain circumstances this is associated
with the administration of quinine. On this view, however, it
still remains to be determined whether the lysis at the onset of
an attack of blackwater fever is due to a sudden liberation of
complement or to some other cause.

Methods of Examination. — The parasites may be studied by
examining the blood in the fresh condition, or by permanent
preparations. In the former case, a slide and cover-glass having
been thoroughly cleaned, a small drop of blood from the finger
or lobe of the ear is caught by the cover-glass, and allowed to
spread out between it and the slide. It ought to be of such a
size that only a thin layer is formed. A ring of vaseline is
placed round the edge of the cover-glass to prevent evaporation.
For satisfactory examination an immersion lens is to be preferred.
The amoeboid movements are visible at the ordinary room
temperature, though they are more active on a warm stage.
With an Abbe condenser a small aperture of the diaphragm
should be used.

Permanent preparations are best made by means of dried
films, which films are fixed by one of the methods already given
(p. 94), or by placing in absolute alcohol for five minutes
(Manson). The films thus prepared and fixed may be stained
for two or three minutes in a saturated watery solution of
methylene blue or in carbol-thionin-blue (p. 105) ; the solutions
must be carefully filtered (especially the latter), and the films
must be washed well after staining. They are then dried and
mounted in balsam. In the case of thionin-blue, sharper results
are obtained by dehydrating in alcohol aud clearing in xylol
before mounting. The best results are, however, obtained by
one of the Romanowsky methods as described on p. 113.

The fact that in many cases the parasites may be few in
number led Ross to devise a method for making their recognition
more easy by using blood films of unusual thickness. Here
about as much blood as is used in a haemoglobin determination
(20 c.mm.) is taken on a slide, and, being spread out only so
much as to occupy the area of an ordinary cover-glass, is allowed
to dry. There is then dropped on it by means of a glass rod a
little of the watery eosin used in making up the Romanowsky
dye (vide p. 113). This is allowed to act for about a quarter
of an hour, and then very gently washed off with distilled water,
The Romanowsky methylene-blue solution is then applied for a
few seconds and also carefully washed off, and the preparation

METHODS OF EXAMINATION 601

dried and mounted. The hemoglobin of the red corpuscles is
washed out by the eosin solution, and the smaller forms of the
malarial parasite stand out as round circles containing the char-
acteristic chromatin dots ; and in consequence of the greater
number present in a given area as compared with an ordinary
preparation their recognition is very easy. For the large forms
of the parasite Ross has found it useful to make such a film
and, hemolysing the red cells with distilled water, to examine
it unstained. The presence of pigment in the parasites enables
them to be readily seen.

APPENDIX D.

AMCEBIC DYSENTERY.

IN a previous chapter it has been pointed out that the term
" dysentery " has been applied to a number of conditions of
different etiology, and the relations of bacteria as causal agents
have been there discussed (vide p. 384). We shall here consider
that variety of tropical dysentery which is believed to be due to
an amceba, and hence often known as amoebic dysentery.

Amongst the early researches on the relation of organisms to
dysentery probably the most important are those of Losch, who
noted the presence and described the characters of amoeba in
the stools of a person suffering from the disease, and considered
that they were probably the causal agents. Further observations
on a more extended scale were made by Kartulis with con-
firmatory results, this observer finding the same organisms also
in liver abscesses associated with dysentery. Councilman and
Lafleur, working in Baltimore, showed that this variety of
dysentery can be distinguished from other forms, not only by
the presence of amoebae, but also by its pathological anatomy.
The intestinal lesions, to which reference is made below, are
of a grave character, mortality is relatively high, and recovery,
when it occurs, is protracted on account of the extensive tissue
changes. The subject was, however, complicated by the fact
that a similar organism — the amoeba coli — had been previously
found in the intestine in normal conditions and in other diseases
than dysentery (by Cunningham and Lewis and others), and
additional research confirmed these results. The matter is still
far from being satisfactorily cleared up. While we may say
that the pathogenic role of amoebae has been established, much
remains to be done in determining what species have pathogenic
properties and how these species may be identified. The
characters of the common amoeba of the colon and an amoaba
of dysentery were carefully worked out by Schaudinn, who
recognised them to be quite distinct species, and gave to them

602

ENTAMCEBA HISTOLYTICA f>03

tlie names of entamoeba coli and entanm-ba hi&tolytica re-
spectively. We shall give the chief points in his description,
but it must be kept in view that amcebie of dysentery studied
by others present differences in character. To these also refer-
ence will be made below.

l',i>f<iin<i-li<i /tistolytica, as seen in dysenteric stools, occurs in
the form of rounded, oval, or pear-shaped cells, measuring
12 to r>0 //. in diameter (Fig. 169, and Plate VL, Fig. 23).
Considerable variations in size are met with in different cases
of dysentery; in some acute cases few am<eb;e may exceed
•_M) fi in diameter. When at rest, a somewhat clear, highly

/ <v,

a b

Fir;. 169.— Amcebse of dysentery.

<i and f>, amoebae as seen in the fresh stools, showing blunt amoeboid
processes of ectoplasm. The endoplasm of a shows a nucleus, three
red corpuscles, and numerous vacuoles ; that of 1>, numerous red
corpuscles and a few vacuoles.

r, an amoeba as seen in a fixed film preparation, showing a small
rounded nucleus (Kmsc and 1'asquale). x600.

refract! le ectoplasm and a granular endoplasm can be dis-
tinguished, a feature which differentiates the organism from
the entamoeba coli. The nucleus is rounded or oval, and
is seen with difficulty; its position is usually excentric, and is
sometimes quite at the margin of the ectoplasm. In the fresh
condition, and especially when examined on a warm stage, the
organism shows very active amoeboid movements. The pseudo-
podia, which are quickly protruded and retracted, are blunt
and apiiear to be of tough consistence, a property which
Sehaudinn considers of importance, as enabling the organism
to penetrate the mucous membrane, etc. The amoebic move-
ments are often of an active kind, and locomotion may be

604 AMCEBIC DYSENTERY

fairly rapid; and red corpuscles, bacteria, cells, etc. may
often be seen in the interior, though the ingestion of red
corpuscles is by no means a constant feature. The organism
usually dies and undergoes disintegration in a comparatively
short time after being removed from the body; the stools
ought therefore to be examined in as fresh a state as possible.
Multiplication takes place by simple amitotic division and also
by budding. The entamceba coli is an organism of about the
same size. When at rest it shows no differentiation into ecto-
plasm and endoplasm, and the nucleus, usually situated in the
centre, shows a highly refractile membrane with chromatin
masses scattered in the interior. During amoeboid movement
some delicate processes of ectoplasm come into view.

Both organisms have now been shown to pass into a resting
stage with formation of cysts, the character and mode of forma-
tion of which are markedly different in the two cases. The cyst
formation of the entamceba histolytica, as described by
Schaudinn, is specially seen when the disease is in process of
cure and the stools are beginning to have a less fluid character.
In the earliest stage of the change the nuclear membrane
becomes broader and fades into the protoplasm, whilst the
chromatin becomes dispersed through the endoplasm in the
form of small chromidia. Buds then form on the surface,
and into these some of the chromatin passes. Around these
buds concentric striation can be seen, and then a hyaline
cyst wall is formed, which is highly refractile in character.
The cyst then becomes separated from the rest of the cell.
Several cysts which measure 2 to 7 ^ in diameter may be formed
from the same amoeba, and the remnant of the cell undergoes
disintegration. These cysts, as will be shown below, repre-
sent a resting-stage with high powers of resistance to external
agencies, and are concerned in producing infection of another
subject. The cellular changes in the encysting of the entamoeba
coli have also been worked out by Schaudinn. They are of
a somewhat complicated character, involving the formation
of reduction bodies and copulation of nuclei, but the ultimate
result is the formation of a fairly large cyst, which contains
eight small cells. The process of cyst formation accordingly in
the two organisms is of a widely different character.

The description of the encystment of amoebae from cases of
dysentery as given by some other observers differs considerably
from that of Schaudinn. In fact, in the majority of the in-
vestigations published no process of encystment of buds on the
surface of the amoeba has been observed ; on the contrary, the

CULTIVATION 605

whole cell becomes enclosed in a cyst, which is of considerable
size. The facts already ascertained point strongly to there being
more than one pathogenic species which have not yet been
satisfactorily distinguished.

The whole subject of the classification and means of distinguishing the
species of pathogenic and non-pathogenic amoebae is still in a very un-
satisfactory state, and much further work is necessary. We may, however,
refer to some of the facts recorded. Musgrave and Clegg, working in

(Manila, cultivated amoebae from drinking water and from various other
external sources as well as from cases of dysentery, and found that they
possessed similar characters. The cysts as shown in their photographs
are of fairly large size, and do not correspond to Schaudinn's description.
By means of amoebse, cultivated from sources apart from dysentery, they
were able to produce dysenteric symptoms and lesions in monkeys,
Lesage cultivated amoebic from cases of dysentery in Saigon and Toulon,
and found that the process of encystment as studied in agar plates agreed
with the account given by Schaudinn. Craig, as the result of studies on
amoebae in San Francisco, confirms the work of Schaudinn with regard to
E. coli and E. histolytica. Viereck found an amoeba in two cases in
Hamburg which resembled E. coli, from which, however, it differed in
its cysts containing only four cells. He gave to it the name E.
tetragena. Hartmann found the same organism in African dysentery, and
was able by means of it to produce dysentery in cats, though the disease
was milder than with E. histolytica. !Noc, working in Cochin-China,
cultivated nim.-l»:< from the intestines in dysentery, from liver abscesses
and from drinking water, and found that they all had the same
characters. The process of encystment was different from that described
by Schaudinn, the whole cell becoming enclosed by the cyst. In
addition to the ordinary method of fission, it formed, numerous small
cells or merozoites by a process of budding within the protoplasm ; these
afterwards becoming free.

Cultivation. — Various attempts have been made to cultivate
the amoeba of dysentery, and Kartulis considered that he obtained
growth in straw infusions. Within recent years cultures of
amoebae in association with various bacteria have been obtained
on agar media by various workers, e.g. Lesage, Musgrave and
<']'"_rLr. X.H-, ami "tlirr-. |-'.,r this |»iir]m-.i- a plain a^ar without
I>eptone is used, and its reaction is made distinctly alkaline to
phenolphthaleine. The presence of .bacteria seems to be
essential for the growth of the amoeba?, and it is found that some
species favour growth whilst others act prejudiciously ; amongst
the former may be mentioned the sp. choleras, b. subtilis, and
various HUM nters of the coli group, though organisms from a
great variety of sources have been found to be equally efficient.

In such cultures, which are most conveniently made in Petri
dishes, the stages of growth and encystment of the amoebic can
be readily studied; the organisms seem to flourish best at a

606 AMCEBIC DYSENTERY

temperature of about 25° C. Although cultures without bacterial
growth have not been obtained, means have been devised to
ensure that only one species of amoeba is present. For this
purpose Musgrave and Clegg select, by means of a low-power
objective, an amoeba well separated on the agar plate, place it in
the middle of the field, then swing into position a high-power
objective, and, having ascertained by means of it that the amoeba
is still there, lower the point of the lens on to the agar. By
this means the amoeba may have been picked up, and it may
then be transferred to a fresh plate. These observers consider
suitable bacterial symbiosis to be of great importance in in-
creasing the virulence of the amoebae, and probably to play an
important part in the pathology of the disease.

Distribution of the Amoebae. — As already stated, they are
usually found in large numbers in the contents of the large
intestine in tropical amoebic dysentery. They also, however,
penetrate into the tissues, where they appear to exert a well-
marked action. In this disease the lesions are chiefly in the large
intestine, especially in the rectum and at the flexures, though
they may also be present in the lower part of the ileum. At
first there are seen local swellings on the mucous surface, chiefly
due to a sort of inflammatory gelatinous oedema with little
leucocytic infiltration ; soon, however, the mucous membrane
becomes partially ulcerated, more or less extensive necrosis of
the subjacent tissues occurs, and gangrenous sloughs result. The
ulcers thus come to have irregular and overhanging margins, and
the excavation below is often of wider extent than the aperture
in the mucous membrane. The amoebae are found in the mucous
membrane when ulcers are being formed, but their most
characteristic site is beyond the ulcerated area, where they may
be seen penetrating deeply into the submucous and even into
the muscular coats. In these positions they may be unattended
by any other organisms, and the tissues around them show
cedematous swelling and more or less necrotic change without
much accompanying cellular reaction beyond a certain amount
of swelling and proliferation of the connective tissue cells. This
action of the amoeba on the tissues explains the character of
the ulcers as just described. These lesions are considered to be
characteristic of amoebic dysentery.

As a complication of this form of dysentery, liver abscesses
are of comparatively common occurrence. They are usually
single and of large size ; sometimes there are more than one, and
occasionally numerous small ones may be present. The contents
are usually a thick pinkish fluid of somewhat slimy consistence,

EXPERIMENTAL INOCULATION

607

and are largely constituted by necrosed and liquefied tissue with

admixture of blood in varying amount. Microscopic examination

shows chiefly necrosed and granular cells and debris resulting

from their disintegration, whereas ordinary pus corpuscles are

-rant y or may be practically absent. In such abscesses associated

with dysentery the amoebae are usually to be found, and not

infrequently are the only organisms present, no cultures of

bacteria being obtainable by the ordinary methods (Fig. 170).

They are most numerous at the spreading margin, and this

probably explains a fact

I "tin ted out by Manson,

that examination of the

contents first removed may

give a negative result,

while they may be detected

in the discharge a day or

two later. The action here

on the tissues is of an

analogous nature, namely,

a necrosis with softening

and partial liquefaction,

attended by little or no

suppurative change. The

anni'baj have also been

found in the sputum when

a liver abscess has ruin FIG. 170.— Section of wall of liver absce-ss,

showing an amoeba of spherical form
with vacuolated protoplasm. From a
case published by Major D. G. Marshall.
xlOOO.

into

thp

not very infrequently
happens. Kartulis records
two cases of brain abscess

occurring secondarily to dysentery in which numerous amoeba;
were present.

Experimental Inoculation. — The anatomical changes in
dysentery, as above described, give strong presumptive evidence
as to the causal relationship of the amoebae, and practically con-
clusive evidence is afforded by animal exjjeriments. Dysentery
occurs occasionally in animals, e.g. in monkeys, but it is. of
comparatively rare occurrence. The disease sometimes results
in the dog by experimental inoculation with dysenteric material.
Kartulis, for example, records two cases, in one of which liver
abscess was present. Cats are, however, found to be more
susceptible, especially young animals. Dysenteric changes have
been produced in this animal by Kartulis, Kruse and Pasquale,
and others. The method generally adopted is the introduction

608 AMCEBIC DYSENTERY

of a small quantity of mucus from a dysenteric case into the
rectum. The resulting disease is of an acute character, and
sometimes leads to a fatal result. The changes in the large
intestine resemble those found in the human disease, and
microscopic examination shows the amoebae penetrating the
wall of the bowel in the characteristic manner. Kruse and
Pasquale obtained corresponding results when the material from
a liver abscess, containing amoebae without any other organisms,
was injected. Quincke and Roos obtained no effects when
the amoebae were administered by the mouth, but they ob-
tained a fatal result in two out of four cases when the cyst-
like forms were given. They also found that the cysts, unlike
the amoebae, were still present even after the material had been
kept for two or three weeks. Extremely important confirmatory
evidence with regard to infection by the cysts has been supplied
by experiments of Schaudinn. Dysenteric material was obtained
from China, and portions of it which were found to contain
the cysts were thoroughly dried. Some of this material was
given with food to cats by the mouth, and typical dysentery
resulted, the amoebae being found in the stools. No results
follow when the material ingested merely contains the vegetative
form of the organism, as it is readily destroyed in the contents
of the stomach.

Musgrave and Clegg produced amoebic colitis in monkeys
by means of cysts from cultures, and such results were obtained
whatever was the source of the amoebae, — that is, with those
obtained from water, vegetables, etc., as well as with those
from dysenteric material. They also produced liver abscess
by direct injection into the liver, and in some instances only
amoebae were present in the abscesses.

Investigations with regard to entamoeba, coli seem to show
that it is a harmless organism and that it is frequently present
in the intestines of healthy individuals. Schaudinn found that
in East Prussia as many as 50 per cent, of the population were
infected with it. The administration of the amoebae, or of the
cysts by the methods mentioned above, produced no result in
animals. It has, however, been shown that when the eight-
celled cysts are swallowed by persons who are free from the
parasite the entamoeba coli appears in the large intestine in
a comparatively short period of time. It accordingly appears
that in the case of both organisms it is the cysts alone which
give rise to infection. Confirmatory results with regard to
the common occurrence of E. coli were obtained by Craig in San
Francisco.

METHODS OF EXAMINATION 609

From the above facts, all of which have received ample
confirmation, there can be no doubt that the amoebae described
are the cause of the form of dysentery with which they are
associated. As already stated, much information is still required
as to the different species of pathogenic amoebae and as to the
means of distinguishing them, if this is possible, from harm-
less forms. In this connection it is interesting to note that
Musgrave and Clegg obtained pathogenic effects with amoebae
resembling the E. coli. But the causal relationship of amoebae
to dysentery has been completely established by the anatomical
and experimental evidence. It is also of importance to note
that the serum of patients suffering from amoebic dysentery
gives no agglutinating reaction with Shiga's bacillus of dysentery
(vide p. 384).

It is important to note that cases of amoebic dysentery have
been recorded both in France and England in patients who have
never resided outside these countries.

Methods of Examination. — The faeces in a case of suspected
dysentery ought to be examined microscopically as soon as
possible after being passed, as the amoebae disappear rapidly,
especially when the reaction becomes acid. A drop is placed
on a slide without the addition of any reagent, a cover-glass is
placed over it but not pressed down, and the preparation is
examined in the ordinary way or on a hot stage, preferably by
the latter method, as the • movements of the amoebae become
more active, and it is difficult to recognise them when they are
at rest. Hanging-drop preparations may also be made by the
methods described. Dried films are not suitable, as in the
preparation of these the amoebae become broken down; but
wet films may be fixed with corrosive sublimate or other
fixative (vide p. 96). In sections of tissue the amoebae may be
stained by methylene-blue, by safranin, by haematoxylin and
eosin, and iron hamnatoxylin, etc. Benda's method of staining
with safranin and light-green is also a very suitable one.
Sections are stained for several hours in a saturated solution
of safranin in aniline oil water (p. 105), they are then washed
in water and decolorised in a J per cent, solution of light-green
in alcohol till most of the safranin is discharged, the nuclei,
however, remaining deeply stained. In this method the nuclei
of the amoebae are coloured red (like those of the tissue cells),
the protoplasm being of a purplish tint.

APPENDIX E.

TRYPANOSOMIASIS— LEISHMANIOSIS—PIRO-
PLASMOSIS.

THE PATHOGENIC TKYPANOSOMES.

THE trypanosomata are protozoal organisms belonging to the
sub-class Flagellata, and many members of the genus have come
to be recognised as living in the blood and tissues in various
animals and as causing important disease conditions. As long
ago as 1878 the Trypanosoma Lewisi was observed infesting the
blood of rats, and it has been found to be sometimes capable of
causing death. Other diseases in which similar organisms have
been found are Surra, which occurs in cattle, horses, and camels
in India, and which is associated with the Tr. Evansi ; Dourine,
a condition affecting horses in especially the Mediterranean
littoral (Tr. equiperdum or Eougeti) ; Mai de Caderas, a disease
of South American horses (Tr. equinum or Elmassiani) ; Tse-tse
Fly Disease or Nagana, affecting horses and herbivora in South
Africa (Tr. Brucei) ; trypanosomiasis of African cattle (Tr.
Theileri) ; and — most important from the human standpoint —
the trypanosomiasis and sleeping sickness of West and Central
Africa associated with the Tr. gambiense and Tr. ugandense,
which are now believed to be the same organism. These
diseases present many general resemblances to one another.
They tend to be characterised by wasting, cachexia, anaemia,
fever often of an intermittent type and irregular oedemas, and
frequently have a fatal result. In many cases the infective
agent is conveyed from a diseased to a healthy animal by the
agency of blood-sucking insects.

Morphology and Biology of the Trypanosomata. — If a drop
of blood containing trypanosomes be examined, the organism
will be seen to be a fusiform mass of protoplasm which at one
end passes into a pointed flagellum. In the living condition the
trypanosome is usually actively motile by an undulatory move-

610

THE PATHOGENIC TRYPANOSOMES 611

incut of its protoplasm and a lashing of the flagellum. The
size varies, but those mentioned above are about 30 /x long and
about 1*5 to 3 /n broad. Much smaller forms exist, however,
and one, 7V. /////''"*, which is 7 to 10 /u, broad and 72 to 123 p.
long, has been described by Bruce. From the fact that in
progression the flagellum is in front, the flagellated end is
denominated the anterior end of the organism. It is stated
that the method of examining the fresh blood by merely allowing
it to spread itself out in a fairly large drop beneath a cover-glass
is more likely to reveal the presence of trypanosomes, if these are
present in small numbers, than is the examination of stained
specimens ; but the minuter structure of the organisms can
best be studied in dried preparations stained by Romanowsky
dyes such as those of Leishman or Giemsa.

For staining trypanosomata (or the Leishman-Donovan bodies) in
sections so as to bring out the chromatin structures, Leishman recom-
mends the following method : — Sections of 5 /A thickness are made and
carefully fixed on slides. The paraffin is very thoroughly removed by
melting it before applying the first xylol, and then washing with alter-
nate baths of alcohol and xylol three or four times. The last alcohol is
thoroughly washed oft" by distilled water, and the excess of water is
removed with cigarette paper. A drop of fresh blood serum is then
placed on the preparation and allowed to soak in for five minutes. The
excess is removed by blotting, and the remainder is allowed to dry on
the section, which is now treated with a mixture of two parts of Leish-
iii, in's stain and three of distilled water, and placed in a Petri dish for
1 to 1£ hours. The preparation is very deeply stained, the nuclei being
almost black, and decolorisation and differentiation are effected by alter-
nately applying the acetic acid and caustic soda solutions (commencing
with the acid) used in the application of the stain to ordinary histological
sections (r. p. 114), the effects being carefully watched with a low
power. The essential part of the method is the application of the blood
serum, though what effect this has is not known ; Leishman suggests
that it restores the normal alkalinity of the tissue.

In preparations stained by the above methods the protoplasm
of trypanosomata stains blue, and in some species some parts
are more intensely coloured than others. Sometimes it contains
violet-coloured granules (chromatin granules), and occasionally
there appears in it slight longitudinal striation. Two bodies
are always present in the protoplasm. Usually near the middle
there is an oval granular body staining purple, — the tropho-
nucleus or macronucleus, — and towards the posterior end is a
minute intensely stained purple granule known as the kineto-
nucleus, blepharoplast, micronucleus, or centrosome (that this
body represents the centrosome is strongly held by Laveran
from the analogy of appearances in certain spermatozoa which

6 1 2 TR YP ANOSOMI ASIS

closely resemble trypanosomes in structure). This micronucleus
is often surrounded by an unstained halo, and in its neighbour-
hood, in certain species, a vacuole has been described as exist-
ing ; this has been considered by some to be analogous to the
contractile vacuole present in many protozoa, and its shape and
position have been made the basis of specific distinctions ;
Laveran, however, thinks it is an artefact. From the micro-
nucleus or from its neighbourhood there arises an important
structure in the trypanosome, — the undulatory membrane.
This is of varying breadth, has a sharp undulating free margin,
and surmounts the protoplasm of the organism like a cock's
comb ; it narrows towards the anterior end where it passes into
the nagellum. Motion is chiefly effected by the undulations of
this membrane and of the nagellum. The latter is continuous
with the protoplasm of the body of the organism ; it stains
uniformly like it, except the free edge which has the reddish
hue of the chromatin. In different species of trypanosomes
variations occur in shape, in length, in breadth, in the position
of the micronucleus (and therefore in the length of the undulat-
ing membrane), in the breath of the membrane, in the length of the
free part of the nagellum, in the shape of the posterior end, which
is sometimes blunt, sometimes sharp, and in the presence or
absence of free chromatin granules in the protoplasm.

Multiplication in the body fluids ordinarily occurs by longi-
tudinal, amitotic division (see Fig. 171). First of all the micro-
nucleus divides, sometimes transversely, sometimes longitudin-
ally, then the macronucleus and undulating membrane, and lastly
the protoplasm. In some species the root of the nagellum only
divides, so that in the young trypanosomes the nagellum is short
and subsequently increases in length (Tr. Lewisi) ; usually the
whole nagellum takes part in the general splitting of the
organism.

In the cases of several of the trypanosomata it has been
found possible to cultivate them outside the body, the first work
here having been done by Novy- and MacNeal, who succeeded
with the Tr. Lewisi, Tr. Evansi, and Tr. Brucei. They used a
special medium (see p. 45), on which it was found that multipli-
cation went on readily, the organisms dividing longitudinally as
in the tissues. Sometimes very small forms result, and often
these are found in rosettes which are formed by a number of
individuals arranging themselves in a circle with the flagella
directed towards the centre of the agglomeration. These results
have been confirmed by other observers, and by repeated sub-
cultures several of the trypanosomata named have been kept

THE PATHOGENIC TRYPANOSOMES 613

alive for more than a year, and when re-introduced into appro-
priate hosts have been found not to have lost their infective
properties.

The main fact in the biology of those trypanosomata with
which the pathologist is concerned, is that in the higher animals
infection takes place by the parasite being transferred from one
host to another by the agency of biting or blood-sucking insects,
or by other similar agencies such as leeches. It may be said
that the mere mechanical transference of the parasite by, say,
a blood-flocking insect, while it may sometimes occur, probably
plays a subsidiary part in infection. Several instances will be
given in which it is known that an insect does not become
actively infective until some days have elapsed after it has
sucked the blood of an infected animal. The analogy of the
malarial organisms suggests the occurrence of a sexual con-
jugation within the insect, but definite proof of this is still
wanting, and, as Minchin points out, while we must admit the
existence of a cyclic development, it by no means follows that
this includes a definitely sexual stage, although many are of
opinion that such a stage does take place.

The starting-point of the sexual theory lies in the slight differences
in form which have been -observed in the organisms in the body fluids
of the vertebrate hosts. Such differences have been described in Tr.
Leuisi and Tr. Brucei by Prowazek, and in Tr. ugandense by Minchin,
and have been made the basis of a classification into three types, which
are looked on as representing male, female, and indifferent individuals.
The male type is rather slender both in body and in nucleus, the free
part of the flagellum is longer than the body, and the protoplasm is
free from granules ; the female is broader, its nucleus is larger and
rounder, the undulating membrane narrower, the free part of the flagellum
is shorter than the body, and the protoplasm contains many chromatin
granules, which are looked upon as reserve food material. The indifferent
individuals present intermediate characters. All multiply by fission
as described, and, according to the supporters of the sexual theory, the
indifferent individuals can on occasion become differentiated into male
or female forms. The females are the most hardy, and next come the
indifferent individuals ; if all but the females die out, these can undergo
parthenogenesis, and representatives of all three types can be again
reproduced. The sexual cycle is represented as occurring in the in-
vertebrate host. In Tr. Lewisi. according to Prowazek, this is found
in the rat louse, hwmaiopinus .yrinulosus. "When this insect sucks the
blood of an infected rat, copulation occurs by the male trypanosome
entering tin- female near the micronucleus and the various parts of the
two individual! becoming fused. A n on -flagellated obkinete results,
which, passing through a spindle-shaped grcgarinc-like stage (crUhidium),
can develop into a trypauosome in the stomach of the louse. A resting-
stage in an immature trypanosotne-like form is described as occurring
in or on the intestinal" epithelium, and the parasite is supposed to
reach the body cavity, and ultimately the pharynx of the insect, and

6 1 4 TR YP ANOSOMI ASIS

thus to find the opportunity for passing into the body of a fresh host.
Minchin, however, has been unable to find evidence of a sexual stage in
this trypanosome.

A still further development of the views held as to the life-history of
the trypanosomes is found in the work of Schaudinn, who investigated
the trypanosoma noctuce found in the owl (athene noctua), and which is
carried from bird to bird by the common mosquito (culex pipiens). In
the blood of the owl is a halteridium hsemamceba showing pigmented
male and female forms, closely corresponding to those observed by
Macallum in the crow. These, according to Schaudinn, on reaching the
mosquito's stomach undergo ordinary changes — the microgametocyte
develops microgametes, one of which fertilises the macrogamete. An
oval motile ookinete results, and in the formation of its nucleus from
the male and female elements a reduction of chromosomes takes place,
while the superfluous nuclear structures along with the pigment are cast
out of the cell. In these ookinetes a differentiation into male, female,
and indifferent forms can be recognised, but the important new departure
is that each can go on to develop into a trypanosome. In the indifferent
ookinete a portion of the new nucleus breaks off by a sort of heteropolar
division, and the broken-off part becomes the micronucleus or blepharo-
plast of the trypanosome and gives origin to the undulatory membrane.
Longitudinal division of these forms in the mosquito's intestine may
occur, and here it may be said that Schaudinn differed from other observers
in holding that on division the membrane does not split, but that one of
the individuals gets its membrane by this being laid down along the root
of that already in existence. Further, a resting-stage of the trypanosome
may occur, in which it becomes attached to the intestinal epithelium,
and, losing more or less its flagellate form, may resemble a gregarine.
The female ookinete is plumper and contains more chromatin granules
in its protoplasm than the male and indifferent forms, and is in virtue
of the reserve material in its protoplasm much more resistant than the
male or indifferent forms ; when these die out, as they do when, for
instance, the insect is starved, it reproduces all three forms by a process
of parthenogenesis. In the female ookinete the smaller nucleus which
goes to form the blepharoplast of the indifferent trypanosome divides
into eight small nuclei, all of which perish, and the blepharoplast and
membrane are formed by a fresh division in the large nucleus remaining.
In the male ookinete, which differs from the female in the clearness of
its protoplasm, a similar heteropolar mitosis takes place, and again eight
small nuclei are produced. These evidently represent the essentially male
element, for they persist, and each, appropriating to itself a portion of
the cellular protoplasm, detaches itself so that eight small trypanosomes
are budded off from the ookinete. This male ookinete Schaudinn holds
to be homologous with the microgametocyte occurring in the blood of
the owl, and the small trypanosomes are similarly homologous with the
microgametes formed when the blood of the host reaches the stomach
of the mosquito. These small trypanosomes and the male trypano-
somes readily die, probably because, by a reduction process in their
genesis, the assimilative powers of the larger nucleus have been
diminished. In degenerating they often are found in the intestinal
tract of the mosquito arranged in rosettes, with sometimes the anterior
ends, sometimes the posterior ends, directed towards the centre of the
rosette. The next stage of development takes place when the parasites
by the mosquito's bite reach the blood of the vertebrate host, and it is

THE PATHOGENIC TRYPANOSOMES 615

in connection with this stage that Schaudinn's observations are very
tar- reach ing. The indifferent forms by this time, by repeated longi-
tudinal division, have become small. In the owl they attach them-
selves to red blood corpuscles, which they penetrate, and, assuming a
halteridium form, grow in size for twenty-four hours ; they then leave
the cells, elongate, again assume the Hagellate form, move freely in the
blood till the next night, when they again enter fresh cells. This cycle
is repeated for six nights, the organism gaining in size with each sojourn
in a corpuscle. When the full size is thus attained, repeated longi-
tudinal division occurs, and when the smallest forms are again reached
the intra-cellular cycle recommences. The largest female forms prob-
ably cannot pass through the mosquito's proboscis, but when small
forms reach the owl they enter the red corpuscles. They assimilate
food material, and appear not to migrate so frequently as the indifferent
forms. They lose their capacity of assuming the free trypanosoma
form, and ultimately their capacity of migration, so that they are
found lying surrounded by the remains of the last cell they entered.
The male forms when they reach the blood of the host rapidly die, and
the microgametocytes are always recruited from the indifferent forms,
or from parthenogenesis of the female forms.

It may be said that, according to Schaudinn, the trypanosomes gain
access to the tissues of the mosquito by perforating the intestinal wall,
and, passing through the body till they reach the wall of the pharynx,
they bant through this, and are in a position to be ejected when next
the insect bites.

Such are the views put forward by this observer on the cycle
of life-history of the Tr. noctuse. It will be recognised that
the essential point is the occurrence, both in the vertebrate and
invertebrate host, of halteridium stages alternating with those
in which the trypanosome form is assumed. It is evident that,
if this were substantiated, important effects would follow in our
views as to the morphology not only of the trypanosome group,
but also of that to which the malarial parasites belong.

Certain criticisms of these results have been made, especially by
Novy and McNeal, who sought confirmatory evidence by means of their
culture method. These observers are of opinion that the appearances
described by Schaudinn were due to his dealing with a mixed infection
of the owls by trypanosomes on the one hand and hfemamcebse of the
halteridium type on the other. They examined a very large number of
different species of birds, and established the fact that infection with
halteridium parasites on the one hand and with trypanosomes on the
other is extremely common, and further, that in the blood of the same
lunl both halteridium and trypanosome infection could be observed.
The bird trypanosomes could be readily cultivated, and it was observed
that no cultures were obtainable from birds in which halteridia were
alone found, and further, that when a trypanosome isolated from a case
where both forms of parasite had been seen was injected into a fresh
bird, only trypanosome forms were found to develop in its body. These
results are, of course, not quite conclusive, as the halteridium stage
might only follow on the sexual portion of the cycle.

6 1 6 TR YPANOSOMI ASIS

Reference may here be made to the views put forward by
Schaudinn regarding the relationship of certain spirilla to the
trypanosomes. In the athene noctua, besides the Tr. noctuse
already referred to, there is a protozoal parasite infesting the
leucocytes known as spirillum Ziemanni or leucocytozoon
Ziemanni, whose invertebrate host is also the culex pipiens.
Ziemann had described the male and female forms in the owl,
and microgametes had been observed forming from the micro-
gametocytes. Schaudinn observed the formation of an ookinete
in the mosquito ; in certain cases this ookinete elongates, and
the vermicule rolls itself up into a ball with great proliferation
of the nucleus. Each little nucleus attaches to itself a portion
of the protoplasm, and, becoming a miniature trypanosome,
swarms off and becomes free. These minute trypanosomes
elongate and develop into typical spirilla by rolling their ribbon-
shaped bodies spirally along their longitudinal axes, the
individuals possessing male, female, or indifferent characters,
just as in Tr. noctme. These spirilla multiply by longitudinal
division, and often after fission the two individuals remain
attached to each other by their posterior ends, and in this way
there is made possible what is often seen in spirilla, namely, a
capacity to move in either direction. The spirilla often divide
so frequently that ultimately the individuals become invisible by
means of the microscope, and can only be seen when lying in
clumps. In this stage Schaudinn thinks the organism would be
able to pass through a Chamberland filter, and this may be a
very important observation, as throwing light on the etiology of
certain diseases, such as yellow fever, in which no visible parasites
have been found.

Schaudinn's views on the trypanosomal characteristics of Sp. Ziemanni
raised important questions regarding the morphology of other similar
forms which have been long familar, such as Sp. Obermeieri, and
also of the Spirochrete pallida which Schaudinn himself discovered.
It is as yet too soon to express any opinion on the ultimate effect of
these views. According to Schaudinn, a trypanosomal spirillum consists
of a central thread which represents the posterior nucleus of the
trypanosome ; round this thread the undulating membrane is spirally
wound, and the principal nucleus is represented by minute chromatin
dots sometimes seen in the course of the spiral. Whether all spirilla
have this structure must be left for future investigation to determine.
Difficulties arise with regard to the significance of an undulatory
membrane as a spirillary characteristic, and also with regard to the
terminal flagellum which Schaudinn himself found in Spiroclisete pallida,
and which he had previously thought did not occur in protozoal spirilla.
It is possible that two groups of organisms have hitherto been classed
together under the name spirillum, and that one of these may still
have to be placed with the bacteria.

TKYI'ANOSOMA LKWTSl 617

We now pass to consider in detail some of the more important
trypanosomes.

Trypanosoma Lewis!.— In 1878 Lewis described in rats in India the
occurrence of the parasite which now goes by his name, and since that time
this trypanosome has been found to be very common in the blood of rats
all over the world, though the percentage of animals affected varies in
different localities. Though the organism has no importance from the
standpoint of human pathology, several significant points arise in con-
nection with it. The condition in the rat is of great interest, as, though
the infection runs a very definite course, it is very rarely fatal ; in fact,
many observers have been unable to produce death by infecting even
large series of animals. There is, however, little doubt that a fatal issue
does occur sometimes in young individuals, especially when these are
infected with strains of the organism imported from other localities.
The trypanosome, which is actively motile, is of the usual length but is
somewhat narrow, and its protoplasm does not contain any granules. It
multiplies by fission, of which Laveran describes two varieties. In one,
the organism splits longitudinally and gives rise to smaller individuals
than the parent. In the other, the trypanosome loses its ordinary shape
and becomes more oval : nuclear division, which is often multiple, then
takes place, and on subsequent division of the protoplasm a number of
small flagellate organisms result ; these last may attain the full form and
size before dividing again, or they may divide when still small. When a
rat is infected by injection into the peritoneum, active multiplication goes
on in the cavity for a few days and then comes to an end. Very soon
after infection the organisms begin to appear in the blood and there
rapid multiplication occurs, the extent of which is sometimes so great
that the trypanosomes may seem to equal the red blood corpuscles in
number. The animal usually shows no symptoms of illness. The
infection goes on for about two months, and then the organisms gradually
disappear from the blood. In the great majority of cases the rat is now
immune against fresh infection. If trypanosomes be introduced into its
peritoneum they are, according to Laveran, taken up by mononucleate
phagocytes and destroyed. The serum of a rat which has been infected
shows agglutinating capacities towards the trypanosomes, causing them to
agglomerate in rosettes in which the flagella are directed outwards, and
tlie serum of immune rats has a certain degree of protective action if
injected along with the organism into a susceptible animal. As has
already been noted, this trypanosome has been cultivated on artificial
media, on which it multiplies freely, large numbers of small forms being
often produced. The-c when injected into rats give rise to the usual
infection, but not so rapidly as when blood from an infected animal is
used. The organism multiplies at the body temperature, but a lower
temperature is ] .referable, and at 20° C. Novy and McNeal succeeded in
carrying a growth through many sub-cultures. The trypanosome is very
resistant to cooling, and has been exposed for fifteen minutes to the
temperature of liquid air (-191°C.) without being killed. With
regard to this infection Minchin and Thomson have shown that the rat
flea, eeratoftkyttwfaueiatut transmits the parasite by the cyclical method
(mechanical infection not having been proved). The flea becomes
infective about a week after biting, and remains infective for a long
period,— possibly for the rest of its life. Infection may also take place
through another species of flea and through a louse.

6 1 8 TRYPANOSOMI ASIS

Nagana or Tse-tse Fly Disease. —This is a disease affecting
under natural conditions chiefly horses, cattle, and dogs; it is
prevalent especially in certain regions of South Africa, though
it probably may occur elsewhere. In the horse the chief
symptoms are the following : — The animal is observed to be
out of condition, its coat stares, it has a watery discharge
from the eyes and nose, and the temperature is elevated ;
swellings appear on the under surface of the abdomen and in
the legs; it gradually becomes extremely emaciated and
anaemic, and dies after an illness of from two or three weeks to
two or three months. In other animals the symptoms are of
the same order, though the duration of the disease varies much ;
thus in the dog the illness does not last more than one or two
weeks, while in cattle it may continue for six months. It is
doubtful whether a domestic animal attacked by the disease
ever recovers. The popular idea regarding the etiology of the
disease was that it was contracted by animals passing through
certain rather restricted and sharply denned areas or belts
characterised by heat and damp, usually lying beside rivers,
and always infested by the tse-tse fly (glossina morsitans), to the
bite of which the disease was attributed ; in this connection it
is important to note that though man is frequently bitten by
the tse-tse fly he does not contract nagana. Modern know-
ledge on the subject dates from the discovery made by Bruce
in 1894 that the blood of animals suffering from nagana
swarmed with a trypanosome now known as the Tr. Brucei,
and in 1895 he was instructed by the Governor of Natal to
undertake the investigation which led him to work out the true
etiology of the disease. It may be said that this research
forms the starting-point of the important work done during
the last decade with regard to infections by trypanosomes.
In his earlier work Bruce found that the parasite was present
in the blood of every animal suffering from nagana and absent
from the blood of healthy animals in the affected districts;
further, that the fever which marks the onset of the disease was
accompanied by the appearance of the trypanosome in the
blood ; and finally, that the transference of the smallest
quantity of blood from an affected to a healthy animal origin-
ated the disease. He then proceeded to investigate the part
played by the tse-tse fly in the condition. He found that if
flies taken from the fly belt were transported to a place where
nagana did not occur, kept for a few days, and then allowed to
bite susceptible animals, the latter did not contract the disease —
this result showing that it was not, as had been supposed by

NAG ANA OR TSE-TSE FLY DISEASE 619

some, a poison natural to the insect which was the pathogenic
agent. But if such a fly was allowed to bite a dog suffering
from the disease and then to bite a healthy dog, the latter
contracted the malady and abundant trypanosomes were found
in its blood. Again, threads dipped in the blood of an infected
animal and allowed to dry caused the disease in healthy animals
up to, but rarely beyond, twenty-four hours after being dried ; if,
however, the blood were kept moist, then it retained its infective-
ness up to between four and seven days ; up to forty-six hours
living trypanosomes could be seen in the tube of the fly's proboscis.
This corresponds roughly with what was found regarding the
limits of the infectiveness of the fly, in that twenty-four hours
after it has been fed on an infected animal its bite is usually in-
nocuous.1 Further, Bruce showed that infection did not occur by
any food or water partaken of by an animal while going through a
fly belt, for he took horses through such a region without allowing
them to eat or drink, and found that they still contracted the
infection, if during their few hours' journey through the belt
they had been bitten by the tse-tse fly. Finally, he showed
that if flies were taken from an infected area to a healthy one
a few miles off and allowed at once to bite infected animals, the
latter contracted nagana.

By those experiments it was thus determined that nagana
could be transmitted by the blood of the infected animal, that
is, without the agency of the fly ; that the latter had no inherent
power to produce the disease ; that it could, however, by
successively biting infected and healthy animals transmit the
disease to the latter; and that specimens of the insect caught in
infected areas harboured the parasite and were thus infective.
The question remained as to how the flies might become infected
in nature. It had been observed that in districts where the
tse-tse fly lived the prevalence of the disease in imported animals
was related to the presence in the locality of wild herbivora.
ISniee now found that, if considerable amounts of the blood
of the latter were taken to another locality and injected into
dogs, these in a proportion of cases contracted nagana, and from
this he deduced that the wild animals harboured the parasites
in small numbers in their blood and thus kept up the
possibility of infection. A further fact was that other blood-
sucking flies besides the tse-tse appeared incapable of acting as
carriers of infection. Bruce's work as a whole pointed to the

1 This observation probably only applies to infection so far as this may
l»r merely mechanical. There is evidence that a cyclic development
mvurs iii glossina. and that thus after an interval its bite is again infective.

620

TRYPANOSOMIASIS

trypanosome as the cause of nagana, and this has since been finally
established by the origination of the disease by artificial cultures
of the organism.

The Tr. Brucei (Fig. 171), according to Laveran, measures in
the horse from 28 to 33 //.long and from 1'5 to 2 '5 /x broad;
in the rat and dog it is somewhat shorter. It is motile, but its
activity is less than that of Tr. Lewisi. When stained it presents
the usual appearances; its posterior end is usually blunt, and

FIG. 171. — Trypansoma Brucei from blood of infected rat. Note in
two of the organisms commencing division of micromicleus and undu-
lating membrane, x 1000.

the body often contains granules in the anterior portion of its
protoplasm. It divides longitudinally, and, accordng to Brad-
ford and Plimmer, a form of longitudinal conjugation occurs in
the blood. According to the same observers, it can be kept alive
for five to six days in blood outside the body. It is less resistant
to the action of cold than Tr. Lewisi, perishing in a few days
at 5 to 7° C., but, like the other organism, it can withstand short
.exposures to temperatures down to -191° C. ; it is quickly

TIIYPANOSOMA OF SLEEPING SICKNESS 621

killed at 44 to 45° C. Novy and McNeal succeeded in cultivat-
ing this trypanosome also, though here it was very difficult to
obtain a first growth from the blood on their blood-agar medium ;
once >t!irtrd, however, it was kept alive through many sub-
cultures, the optimum temperature of growth being 25° C., and
it was from these sub-cultures that the infection was obtained
which definitely proved the organism to be the cause of the
disease. In cultures, as with Tr. Lewisi, short forms occur, and
there is sometimes a rosette formation with the flagella directed
outwards ; agglutination phenomena are also observable in
defibrinated blood. Under unfavourable conditions involution
forms occur, the organism dividing frequently to form round
flagellated individuals.

Nearly all laboratory animals are susceptible to infection,
and the duration of the illness corresponds to what has been
observed in the natural infection of these animals. The rat has
been largely used for experiment and usually succumbs in about
ten days, there being very few symptoms up till a few hours
before death. A very important fact has been observed with
regard to this animal, namely, that individuals which have gone
through infection with Tr. Lewisi and which are immune are
still susceptible to the Tr. Brucei ; from this it has been deduced
that the two organisms are to be looked on as distinct species.

Trypanosoma of Sleeping Sickness. — Since the year 1800
the disease called sleeping sickness, sleeping dropsy, or negro
lethargy has been recognised as prevailing on the West Coast of
Africa from the Senegal to Lagos, and in the parts lying behind
the coast between these regions. It has also been found to be
rife from Cameroon to Angola and in the Congo valley, and to a
less extent up the Niger and its tributaries. In 1901 it began
to appear in the Uganda Protectorate, and it is in that region
that the investigations have been carried on which have led to a
knowledge of its cause ; here it has wrought very serious havoc
amongst the native population. It is characterised in the early
stages by a change in disposition leading to moroseness, apathy,
disinclination for work or exertion, and slowness of speech and
gait. There may be headache, indefinite pains about the body,
the evening temperature may be elevated several degrees, the
pulse tends to be soft and rapid, and in a very large number of
cases the superficial glands of the body are enlarged. In a
rapid case the lethargy becomes more pronounced ; fine tremors,
especially of the tongue and arms, develop ; progressive emaci-
ation occurs ; blood changes appear, consisting of a progressive
diminution of the red cells and of the haemoglobin, and of a

622 TRYPANOSOMIASIS

lymphocytosis in which the percentage of both the large and
small mononuclear cells is increased, so that the former may
constitute from 20 to 30 and the latter from 30 to 40 per cent,
of all the white cells present. As the disease progresses the
drowsiness increases till it deepens into a coma from which the
individual cannot be roused. Often during the disease there
occur irregular cedematous patches on the skin, and sometimes
erythematous eruptions, and effusions into the serous cavities.
Not every case runs a progressively advancing course. Some-
times along with enlargement of glands the chief early feature
is the occurrence from time to time of attacks of fever which
may be mistaken for malaria, and from these apparently com-
plete recovery may take place ; recurrence, however, follows as
a rule, and ultimately the typical terminal phenomena may
commence. Such cases may go on for years, and it is probable
that many patients die of pneumonia without exhibiting typical
manifestations of the malady from which they really suffer.
The disease is an extremely fatal condition, and probably no
case where the actual lethargy is developed ever recovers.

On considering the disease from the standpoint of pathological
anatomy there is little to be said. As Mott described, the most
striking feature is the presence of a chronic meningo-encephalitis
and meningo-myelitis. The pia-arachnoid is sometimes opaque
and slightly thickened and may be adherent to the brain, and
its vessels usually show some congestion. The sub-arachnoid
fluid is sometimes in excess and occasionally may even be puru-
lent. The membranes of the spinal cord show similar changes.
The chief other feature is the presence of enlarged lymphatic
glands in the body, but otherwise there is nothing special to
note. With regard to the microscopic changes, the chief feature,
according to Mott, is a proliferation and overgrowth of the
neuroglia cells, especially of those which are related to the sub-
arachnoid space and the perivascular lymph spaces, with
accumulation and probably proliferation of lymphocytes in the
meshwork. He further points out that the changes in the
lymph glands are of similar nature and resemble the infiltration
of the perivascular lymphatics of the central nervous system.
These changes are specially significant in view of the lympho-
cytosis present in the blood, which has already been noted, and
which so often occurs in protozoal infections. In the nervous
structures there is comparatively little change, there being
merely, according to Mott, some atrophy of the dendrons of the
nerve cells, a diminution of Nissl's granules, and an excentricity
of the nucleus.

TRYPANOSOMA OF SLEEPING SICKNESS 623

7'/-t/j>anosoma f/ambiense. — Before going further we must refer
to the observation of a trypanosome in the blood of persons not
i-vidi'iitly suffering from sleeping sickness. The first case of
this was recorded by Dutton in 1901, the patient being a
European then living at Bathurst on the Gambia. The progress
of the disease was here very slow, and was characterised by
general wasting and weakness, irregular rises of temperature,
local oedemas, congested areas of the skin, enlargement of spleen,

FIG. 172.— Trvpanosoma garnbiense from blood of guinea-pig, x 1000.
See also Plate VI., Fig. 25.

and increased frequency of pulse and respiration ; death occurred
a year after the case came under observation after an access of
fever, and a striking fact was the absence of any gross causal
K'sion. During the time the patient was under observation
trypanosomes were repeatedly demonstrated in the peripheral
blood, and they also developed in the bodies of monkeys and
white rats inoculated with the blood. Pursuing further in-
quiries, Dutton and Todd demonstrated similar parasites] tin
other Europeans and in several natives in the Gambia region,

624 TRYPANOSOMIASIS

whilst about .the same time Manson reported a case of the same
kind in the wife of a missionary on the Congo. It thus came
to be recognised that in man there occurred a disease having
characters somewhat resembling nagana and in which trypano-
somes could be demonstrated in the blood, and this was usually
referred to as human trypanosomiasis, or trypanosoma fever, —
the trypanosome being named the Tr. gambiense.

Relation of Trypanosomes to Sleeping Sickness. — Several
views as to the etiology of this disease had been advanced. A
Portuguese Commission in 1902 described a diplococcus, tending
to grow in chains, which they isolated from the cerebro-spinal
fluid taken from cases during life, and to which they were
inclined from the constancy of its occurrence to attribute a
causal role. The seriousness of the epidemic in Uganda had led the
Royal Society of London in 1902, at the instigation of the Foreign
Office, to despatch a Commission to investigate the condition on
the spot. Soon after commencing work, Dr. Castellani found
in some cases in the cerebro-spinal fluid, especially when this
was centrifugalised, living trypanosomes resembling the Tr.
gambiense ; he also found in 80 per cent, of the cases post mortem
a coccus resembling if not identical with that observed by the
Portuguese Commissioners. At first Castellani was inclined to
look on the presence of the protozoon as accidental, but Colonel
Bruce, on going out with Nabarro and Greig in 1903 to pursue
the work of the Commission, realised the significance of the
observation, urged Castellani to further inquiries, which he
himself continued after the departure of the latter, with the
result that in a series of examinations carried out in several
infected localities, the trypanosome was demonstrated in every
case of the disease. This work formed the starting-point for
inquiries, the results of which make it practically certain that
the parasite is the causal agent of the condition. The organisms
were not demonstrable in the cerebro-spinal fluid of patients
dying of other diseases in the sleeping sickness area. On the
other hand, it was found that if cerebro-spinal fluid withdrawn
from cases of the disease was injected into monkeys (especially
macacus rhesus), trypanosomes appeared in the blood, and in
many cases in three or four months the animals died of an ill-
ness indistinguishable from sleeping sickness, and with the para-
sites in the central nervous system. It was further found that in
the parts round the north end of Victoria Nyanza where sleeping
sickness was rife, the distribution of the disease exactly corre-
sponded with the distribution of a blood-sucking insect, the
glossina palpalis, a species closely allied to the c/lossina morsitans

TKYPANOSOMA OF SLEEPING SICKNESS 625

of nagana. It was found that, when one of these Hies was
fed on a sleeping sickness patient and then allowed to bite a
monkey, frequently trypanosomes appeared in the animal's
blood, and that when fresh flies caught in the sleeping sick-
ness area were placed on a monkey a similar occurrence took
place.

The trypauosome of sleeping sickness is 17 to 28 //, long and
1*4 to 2 JJL broad (Fig. 172); when about to divide it is both longer
and broader. According to Laveran, the free part of the flagellum
often equals a fourth of the whole length, but occasionally the
body protoplasm extends quite to the end of the organism.
The undulating membrane is narrow, and the posterior end may
be either sharp or blunt. The trypanosome contains the macro-
and micronucleus characteristic of the group, and the proto-
plasm often shows chromatin granules. Castellaui attached great
importance to a vacuole often seen in the neighbourhood of the
inicronucleus, but, as stated above, Laveran holds this to be an
artefact. The organism divides longitudinally in the usual manner,
HI id often two can be seen to approach each other and lie more or
less side by side, but whether this indicates conjugation or not is
not known. The organism does not usually long survive removal
from the body, but it has been found to be motile for nineteen
days when kept on rabbit-blood agar at 22° C. As we have
said, when Tr. ugandense is inoculated into monkeys they often
contract an illness which ultimately presents the features of
typical sleeping sickness. Inoculation of other species of animals
is not usually so successful, though in nearly every case, e.g. in
the guinea-pig, a proliferation of the parasite, as indicated by its
appearing in the blood, takes place ; but often either no disease
occurs or this runs a very chronic course. The relative insuscepti-
bility of animals, especially of the dog, to the Tr. ugandense is
taken as evidence that this organism is essentially different from
Tr. Brucei.

By means of microscopic examination the organisms may be
found in the cerebro-spinal fluid, the blood, or the juice of
glands. In the case of the first about 10 c.c. of the fluid is to
l)o ceutrifugalised for fifteen minutes and the deposit placed under
a cover-glass for examination ; it is better to make a little cell
on a slide by painting a ring of ordinary embedding paraffin,
t'» |ilace the droplet of fluid in its centre, and to support the
cover-glass on the paraffin ; in this way injury to the delicate
structure of the organism is avoided. In • fresh cerebro-spinal
fluid the trypanosomes can be seen to be actively motile ; the
number in which they occur varies very much, and the same is
40

626 TRYPANOSOMIASIS

true to a greater degree of the blood, in which they are, however,
usually very scanty. With regard to the examination of the
blood, Bruce and Nabarro state that it is difficult by ordinary
centrifugalisation to concentrate the organisms, as these are not
readily precipitated. They accordingly recommend that the
blood be mixed with citrate of sodium solution (equal parts of
blood and of a one per cent, citrate solution) and centrifugalised
for ten minutes, that the plasma be removed and centrifugalised
afresh for the same time, and that this be repeated three times,
the deposit from each centrifugalisation after the first being
carefully examined. Greig and Gray have insisted that the
examination of the glands in a suspected case forms the most
ready means of arriving at a diagnosis, and this opinion has
found strong support from the work of Button and Todd. The
method is to push a hypodermic needle into the gland, suck up
a little of the juice, and blow it out on to a slide. In all cases
where films of any kind are to be prepared the staining methods
of Leishman or Giemsa are to be recommended. Often in
cerebro- spinal fluid and gland juice the staining of the chroma-
tin is difficult, but good preparations are obtained by the pro-
cedure recommended by Leishman for studying the parasite in
sections (p. 114).

Greig and Gray have studied the trypanosome in the body of
the glossina. They found evidence of its multiplying in the
stomach of the insect, and it also was seen to undergo changes
not elsewhere observed. These consisted in alterations in the
position of the micronucleus, which often became anterior to the
macronucleus ; there also occurred rosettes, consisting of from
four to twenty individuals attached by their posterior extremities.
Oval forms were also observed. It was at first supposed that
monkeys could not be inoculated with the trypanosomes from
the stomach of the fly, but recently Bruce has succeeded in
originating an infection in this wray; probably, however, the
organism remains alive in only a small proportion of flies biting
an infective case. Minchin in this connection has described in the
gut of the fly different types of the parasite corresponding with the
male, female, and indifferent forms found in other trypanosomes.
This was confirmed by Koch and by Klein, who also found in
the intestine agglomerations of immature forms which they
ascribed to the results of sexual conjugation. The most im-
portant fact established by the last observer was, however, that
when Gl. palpalis was allowed to bite an animal suffering from
nagana it did not become infective for some days. This has
been confirmed for Gl. palpalis in the case of monkeys suffering

TRYPANOSOMA OF SLEEPING SICKNESS 627

from Tr. gambiense by Bruce and those associated with him in
1908-9. Here it was found that infectivity did not appear till
about thirty-two days after the fly had fed, and continued until
at least seventy-five days. In this connection certain facts
having a serious bearing on the continued infectivity of a local-
ity have emerged. It was found that a certain island on Lake
Victoria Nyanza, which had been cleared of infective natives two
years previously, still harboured infective flies. To account for
this it must be supposed either that the glossina has an extended
duration of life or that the trypanosome exists among the wild
animals. As it has been found that cattle may be infected with
the parasite, and may through the medium of the fly infect
monkeys, it is possible that wild herbivora, while not suffering
in any serious way themselves, are the means of maintaining
infectivity. There is no definite evidence that, as Koch supposed,
the crocodile harbours the trypanosome.

Early in the Uganda investigations the question arose as to
whether the trypanosoma of sleeping sickness was different from
Tr. gambiense. This was forced on the inquirers by the fact
that a very large proportion of the natives in the sleeping
sickness area were found to harbour trypanosomes in their
blood, although not apparently suffering from the disease.
Several cases were carefully examined in which trypanosomes
were constantly present in the blood, but in which the patients
from time to time suffered from fever, and during these pyrexial
periods trypanosomes were found in the cerebro-spinal fluid. It
was suggested that these cases were on the way to develop sleep-
ing sickness. A very important observation was that while in
sleeping sickness areas a large proportion of the native popula-
tion harboured trypanosomes, this was not the case where sleep-
ing sickness did not occur. Further, it was found that
trypanosomes from the cerebro-spinal fluid of sleeping sickness
cases and from the blood of persons harbouring trypanosomes,
but not suffering from disease symptoms, gave rise in monkeys
to the same group of chronic effects which resembled the last
.stages of the disease in man. These facts led the Commissioners
to incline to the idea that trypanosoma fever and sleeping sick-
ness are due to the same cause, and represent different stages
<»f the same disease. It has already been pointed out that a
fatal termination can occur in trypanosoma fever by an acute
febrile attai-k or from intercurreut disease, and thus the terminal
lethargic stage may only develop in a certain proportion of cases.
Continued observation of prolonged cases of trypanosoma fever,
-both in Uganda by Greig and Gray, and in this country by

628 TRYPANOSOMIASIS

Manson, has shown that sometimes the termination of a case is
by the onset of typical sleeping sickness. There is now practi-
cally no doubt that the two conditions are etiologically identical.
The best authorities are agreed that morphologically no difference
between the Tr. gambiense and the Tr. ugandense can be
recognised, and from considerations of priority the former term is
now alone employed.

The prevalence of trypanosomes in the blood of apparently
healthy natives has raised the question of the possibility of
tolerance existing and of immunity being established. It is
possible that both phenomena occur, that not every infection
results in multiplication of the parasite in the body of the
victim, and that in certain cases where multiplication does occur
a resistance is developed which enables the body to kill the
parasites. The occurrence of the mononuclear reaction is here
significant ; it has been suggested that, when this resistance is
weak, the organism gains entrance to the spinal canal, and that
then sleeping sickness results.

The whole of the recent work on the disease is of the highest
interest and importance. The strongest evidence may be said
to exist that the Tr. gambiense is the cause of sleeping
sickness, and action taken on this supposition has had a very
important effect in checking the ravages of the disease in Uganda,
where the natives have been deported from the fly areas, and the
brushwood in which the insects lodge has been cut down in the
neighbourhood of ferries.

Not much success has attended remedial efforts in those suffer-
ing from infection. Here attention has been chiefly concentrated
on the action of organic arsenical compounds, the application of
which in the shape of atoxyl was first recommended by Thomas.
A great range of such substances and also of aniline derivatives
has been investigated by Ehrlich and his co-workers, and under
certain conditions of artificial infection in animals a complete or
partial destruction of the parasites has followed administration
of these bodies, but their application to natural infections has
not as yet met with decided success. Sufficient, however, is
known to justify further investigations of a similar kind. It
has been observed that a tolerance of such reagents can be
developed by the trypanosomes, and this fact may complicate the
problem at issue.

Other Pathogenic Trypanosomes. — Apart from sleeping sick-
ness no other important disease of man has been found to be
associated with trypanosomal infection, but some observations
on a condition observed in Brazil may be alluded to.

TRYPANOSOMA CRUZI 629

Trypanosoma Cruzi. — Chagas, working in Brazil, observed this
trypanosonir in a monkey, the intermediate host being a hemipterous
insect of tin- genus £'"//'//•/< //»'*. As this insect also feeds on man, the
|ius>il.le relationship of the trypanosome to a human disease occurring
in that region was considered. This disease affects children, and is
characterised, by pronounced anaemia, the occurrence of oedema, and
enlargement of lymphatic glands, the spleen and liver ; it is associated
with a mental condition of infantilism, and ends in death with convulsions.
The parasite was only in one case found in the blood of infected individuals,
but when the blood was injected into guinea-pigs, or into callithrix
monkeys, a definite disease occurred, leading to death. In the peripheral
blood in such animals, besides free forms, an infection of the red blood
corpuscles witli a body resembling a malarial merozoite was seen, this
body apparently developing into a trypanosoma-like organism. A special
development of the parasite seemed to occur in the lungs, the result of
which was the formation of cells containing eight bodies resembling the
merozoite forms seen in the circulation, and analogous to what Schaudinn
described in the mosquito's stomach in connection with trypanosoma
noctua?. A cycle of development was also observed in the intestinal tube
of the conorhinns, and cultures were obtained on Novy and MacNeal's
medium.

1 t is beyond the scope of this work to deal at length with the
other diseases of animals caused by trypanosomes. The chief of
these have been mentioned in the opening paragraph, but it may
be said that many others have been described in various species
of mammals, bin Is, and fishes, and that these are spread either
by flies or by leeches. The most interesting of those mentioned
is Dourine, a condition resembling in many ways nagana. It,
ho\\rver, presents this peculiarity, that infection does not take
place by an intermediate host, but apparently directly through
coitus, as it occurs only in stallions and in mares covered by
theea

In several of the trypauosomal infections of animals it appears
as if, as in the case of Tr. Lewisi, the animal suffers little
inconvenience from the presence of the parasite in its blood,
and the view has even been put forward that with all pathogenic
tr\ -pa i incomes there exists a host which acts as a "reservoir" and
carries the organism without being affected by its presence more,
for example, than is the rat by Tr. Lewisi. Though no opinion
can be expressed on this j>oint, it is necessary to bear the fact
in mind that either natural or acquired immunity can exist
against such protozoa. Not only is this important from the
point of view of the investigation of the conditions under which
sueh tolerance fttiaea, but also from the bearing which the
existence of this tolerance may have on the spread in nature of
the parasites to a susrrptibli- -| eCMfl from immune animals which
still liarbiuir trypanost.mes in their blood. \Ve are, however,

630 LEISHMANIOSIS

as yet quite ignorant of many of the processes at work in the
body during a trypanosomal infection, and of the causes of the
symptoms and other morbid effects.

LEISHMANIOSIS.

Under this term there are grouped three human diseases, but
the exact zoological place of the parasites among the protozoa
cannot be said to be at present definitely settled. These
organisms are the Leishmania donovani, associated with the
human disease, kala-azar ; Leishmania infantum, derived from a
similar disease occurring in children; and Leishmania tropica,
which has been found in a skin ulceration of widespread
geographical distribution. Microscopically the organisms are
practically identical, but at present it is convenient to look upon
the three species as being distinct.

Leishmania Donovani. — Leishman noticed in several soldiers
invalided from India for remittent fever and cachexia that the
most careful examination of the blood failed to reveal the
presence of the malarial parasite. From the fact that such
patients had almost invariably been quartered during their
service at Dum-Dum, an unhealthy cantonment near Calcutta,
he suspected he had to deal with an undescribed disease. In
1900 he noticed in the spleen of such a case peculiar bodies
which, from comparison with certain appearances found in
degenerating forms of Tr. Brucei, he suggested might be
trypanosomes, and on publishing his observations in 1903 he
put forward the view that trypanosomiasis might prevail in
India and account for the aberrant cases of cachexial fever met
with there. Soon after Leishman's paper appeared, his observa-
tions were confirmed in India by Donovan, and the bodies
associated with the disease are usually called the " Leishman " or
the " Leishman-Donovan " bodies. They were found by Bentley,
and later by Rogers, in the disease known in Assam as kala-azar,
the pathology of which had long puzzled those who had worked
at it, from the fact that, while it resembled malaria in many
ways, no parasite could be demonstrated to occur in those
suffering from it. This disease has gone under various synonyms,
e.g. cachetic fever, Dum-Dum fever, non-malarial remittent
fever, but is now recognised as a single entity.

Kala-azar (or "black disease," — so called from the hue
assumed by chocolate-coloured patients suffering from it) has
been known since 1869 as a serious epidemic disease in Assam,
where it has spread from village to village up the Brahmaputra

LEISHMANIA DONOVANI 631

valley. The disease is now known to occur in various sub-
tropical centres between the forty-ninth parallels — cases where
the Leishman bodies have been found having been met with
in many parts of India, China, the Malay Archipelago, North
Africa, the Soudan, and Arabia. The disease is characterised
by fever of a very irregular type, by progressive cachexia, and by
anaemia associated with enlargement of the spleen and liver,
and often with ulcers of the skin and dropsical swellings.
Rogers has pointed out that there occurs a leucopenia which
differs from that of malaria in that it is almost always more
marked, — the leucocytes usually numbering less than 2000, — and
further, in that the white cells are always reduced in greater ratio
than the red corpuscles, which condition, again, does not occur in
malaria. The disease is chronic, often going on for several years,
and, at any rate in the great majority of cases, has a fatal
issue. Post mortem, there is little to note beyond the enlarge-
ment of the liver and spleen, but in the intestine, especially
in the colon, there are often large or small ulcers, and there is
evidence of proliferation in the bone marrow, the red marrow
encroaching on the yellow.

In a film made from the spleen and stained by Irishman's
stain, the characteristic bodies can be readily demonstrated
(Fig. 173). They are round, oval, or, as Christophers has
pointed out, cockle-shell shaped, and usually 2*5 to 3*5 /x in
diameter, though smaller forms occur. The protoplasm stains
pink, or sometimes slightly bluish, and contains two bodies
taking on the bright red colour of nuclear matter when stained
by the Romanowsky combination. The larger stains less
intensely than the smaller, is round, oval, heart-shaped, or
bilobed, and lies rather towards the periphery of the body — in
the region of the " hinge " in the cockle-shaped individuals.
The other chromatin body is usually rod-shaped, and is set
perpendicularly or at a tangent to the larger mass, with which
only exceptionally it appears to be connected. Usually the
protoplasm contains one or two vacuoles. Though in spleen
smears many free bodies are seen, the study of sections shows
that ordinarily their position is intra-cellular, — the cells con-
taining them being of a large mononuclear type (Fig. 174). The
view held is that on their entering the circulation they are
taken iij> by the mononuclear leucocytes and by such cells as
the endothelial lining of the splenic sinuses or by those lining
capillaries or lymphatics, that in these cells multiplication takes
place — it may be to such an extent as to rupture the cell, — and
that if thus the bodies become free they are taken up by other

632

LEISHMANIOSIS

cells and the process is repeated. The clusters of bodies some-
times seen in smears are probably held together by the remains
of ruptured phagocytes. In capillaries the endothelial cells after
phagocyting the bodies probably become detached from the
capillary wall, as they are often observed free in the lumen of
the vessel, — this being well seen in the hepatic capillaries.
In the body generally the parasites are found in greatest
abundance in the spleen, liver, and bone marrow, and also in

FIG. 173. — Leishman-Donovan bodies from spleen smear, x 1000.

mesenteric glands, especially in those draining one of the
intestinal ulcers ; less frequently they occur in the skin ulcers,
and in other parts of the body. Donovan described them as
occurring in the peripheral blood, especially within the leucocytes,
and this has been confirmed by other observers, though sometimes
prolonged search is necessary.

In the body the parasite multiplies by simple fission, both
nuclei dividing amitotically, and two new individuals being
formed ; but sometimes a multiple division takes place, each
nucleus dividing several times within the protoplasm and a
corresponding number of new parasites resulting.

LEISHMANIA DONOVANI 633

In view of Leishman's original opinion an extremely important
discovery was made by Rogers and later confirmed by Leishman
himself, to the effect that in cultures a flagellate organism
developed from the Leishman-Donovan body. Cultivation was
effected by taking spleen juice containing the parasite, placing it in
10 per cent, sodium citrate solution and keeping it at 17 to 24° C.
Under such conditions there occurs an enlargement of the
organism, but especially of the larger nucleus. This is followed
by the appearance of a pink-staining vacuole in the neighbour-
hood of the smaller nucleus. Along with these changes, in from
twenty-four to forty-eight hours the parasite becomes elongated
and the smaller nucleus
and its vacuole move
to one end; from the
vacuole there then ap-
pears to develop a red-
staining tiagellurn, which
when fully formed seems
to take its origin from
the neighbourhood of
the small nucleus. The
body of the parasite is
now from 20 to 22 /x
long and 3 to 4 /u. broad,
with the flagellum about
22 fji long. The whole
development occupies
about ninety-six hours. Fl(; 174._Leishmai1-Don0van bodies within
The formation of an endothelial cell in spleen. See also Plate
undulating membrane VI., Fig. 24. xlOOO.
\\a- not observed, and,

although the flagellated organism moved flagellum first, like a
trypanosome, it is evident that here the relationship of the
micronucleus is different, as this structure lies anterior to the
macronucleus. In his cultures, which kept alive for four weeks,
Leishman made a still further important observation. In certain
of the flagellate forms he saw chromatin granules develop in the
protoplasm often in couples, a larger and a smaller. There then
occurred a very unequal longitudinal division of the protoplasm,
and a hair-like undulating individual containing one of the
pairs of chromatin granules would l>e split off. At first these
would be non-flagellate, but later a red-staining flagellum would
appear at one end. The analogies between these observations
and tho-r of Schaudinn (>'. p. GIG) on the relations of spirochsetes

634 LEISHMANIOSIS

to trypanosomes will be at once apparent ; the further develop-
ment of these spirillary forms in Leishman's organism could not,
however, be traced.

The facts just detailed have been the basis for discussion of the
classification of the organism, which now usually goes by the
name Leishmania donovani, originally given to it by Ross.
According to one view, it is to be looked on as a trypanosome,
and although, as we have noted, its flagellated form differs from
the typical trypanosoma form, it bears considerable resemblance
to the members of this group, and, as Leishman has pointed out,
his cultures may not represent the full development of the
organism in the trypanosoma direction. Others have looked on
it as a piroplasma, but Minchin's suggestion has been accepted
that in the present incomplete state of knowledge it is well to
place it and its congeners in a provisional genus, Leishmania, of
the flagellata.

The question arises, given that the Leishmania donovani is
the cause of kala-azar, how is infection spread ^ On this we
have as yet no certain information. The fact that in some
centres of the disease natives who are supplied with good water
are less liable than those who rely on the ordinary polluted
native cisterns, has led to the opinion that water may be the
carrier of infection. On the other hand, the possible relation-
ship of the organism to the trypanosomata naturally suggests
the idea of an insect as an intermediary, and Rogers adduced
some evidence that the bed-bug is the extra-human host. This
view was elaborated by Patton, who brought forward facts to
show that multiple cases might occur in a house while
neighbouring houses were free from the disease. This observer
also fed the common insect parasites of man in Madras on
patients whose peripheral blood contained the Leishmania, and
found that the parasite could be observed only in the pediculus
capitis, and in the bug, cimex macrocephalus. In the midgut of
the latter, forms similar to those seen in the earlier stages of
cultures could be found. Patton compares the organism to an
allied protozoon occurring in the intestine of the common fly.
The rarity of the Leishmania in the peripheral blood has been
advanced as an argument against infection taking place by means
of a blood-sucking insect, but it has been pointed out that
invisible spirillary forms may be instruments of infection. It
may be said here that all attempts to communicate the disease
to animals have been hitherto unsuccessful.

With regard to kala-azar as a whole, we may say that we are
dealing with a distinct disease fairly widespread in various sub-

LEISHMANIA INFANTUM G3.r>

tropical regions. All attempts to include it among the malarial
cachexias, which clinically it so much resembles, have failed.
In this atypical cachexial fever there is always present a parasite
of very special characters belonging or closely allied to a group
which contains many varieties capable of giving rise to similar
diseases. Beyond this we cannot go, but at present we must
admit that there is strong presumptive evidence of the parasite
described being the cause of the disease.

Methods of Examination. — The Leishmania douovani can be
readily seen in films or sections of the organs in which we have
mentioned its occurrence. These should be stained by the
liomanowsky stains. Fluid taken from the enlarged spleen with
a perfectly dry needle during life may be examined, but it is
probable that in this disease puncture of the spleen may not be
a very safe operation, as death from haemorrhage from this
organ is a not uncommon natural terminal event. During life .
the main points on which a pathological diagnosis may be based
a iv the demonstration of the parasite in the circulating blood
which should always be attempted, the absence of the malarial
parasites from the blood, and the features of the leucopenia
which have been alluded to.

Leishmania Infantum. — Nicolle, working in Tunis, observed
a disease clinically identical with kala-azar, but presenting the
peculiarity of only affecting children of about two years of age.
Mr found in the spleen in such cases an organism, microscopically
indistinguishable from the Leishmania donovani. It was
cultivated on a modified Novy and MacNeal's medium, the
cultures presenting characters similar to those observed by
Rogers and by Leishman in the other Leishmania. It was
found that dogs could be infected with the parasite, and, taking
into account the fact that this animal is not susceptible to infec-
tion with the Leishmania donovani, and the further fact that the
disease is apparently confined to infants, Nicolle considered the
organism to be a separate species and gave it the name,
Leishmania infantum. In his view, the infection of the dog
possesses a further significance in that this animal may be the
reservoir from which, by means at present unknown, children
become infected. In support of this, he observed the fact that
a certain proportion of dogs destroyed in Tunis contained the
parasite in the spleen. This disease occurs in other parts of the
Mediterranean littoral; in 1905 Pianese described it in children
in Italy, and the parasites have been found in cases in that
country and also in Sicily and Malta.

Leishmania Tropica.— In various tropical and sub-tropical

636 LEISHMANIOSIS

regions (India and the East, Northern Africa, Southern Russia,
South America) there is widely prevalent a variety of very
intractable chronic ulceration which goes by various names in
different parts of the world — Delhi sore, tropical ulcer, Aleppo
boil, etc. Various views were formerly held as to the pathology
of the condition, but the work of J. H. Wright makes it
practically certain that a protozoal parasite is concerned in its
etiology. In the discharge from the ulcer and in sections of a
portion of tissue excised from a case coming from Armenia,
Wright observed great numbers of round or oval, sharply denned
bodies, 2 to 4 /x in diameter. When stained by a Romanowsky
combination there was found to be a peripheral portion coloured
a pale blue and a central portion tending to be unstained ; there
were also two chromatin bodies, one larger, occupying a fourth
or a third of the whole and situated in the periphery, another
smaller, round or rod-shaped, and of a deeper colour than the
larger mass. It was found that the bodies were usually intra-
cellular in position in the lesion, as many as twenty being in one
cell, and that the type of cell containing them was, as in kala-
azar, that derivable from endothelial tissues.

Wright's observations have been . fully confirmed by workers
in various parts of the world, and it is now recognised that in
these tropical ulcers we have a third example of the activity of
a Leishmania. This is corroborated by the work of Row, who
has obtained cultures in citrated blood, corresponding to those
of the other two species. Nicolle and Manceaux have also
cultivated the organism on Novy and MacNeal's medium, and
have reproduced the condition in man, the monkey, and the dog,
both by virus obtained from the natural infection and from
cultures. The lesions were identical with those naturally occur-
ring, but the incubation period was often many months. It
may be said that Thompson and Balfour have described in the
Soudan a condition in which subcutaneous nodules without
ulceration occurred in man and these contained Leishmania
bodies.

At present the tendency is to look upon the three Leishmanise
as representing different species, but further investigation is here
necessary. It has been pointed out that in kala-azar, skin
ulcerations occur which might link this condition with tropical
ulcer, but it is to be noted that, while in the latter enormous
numbers of the parasite are found, in the ulcers of kala-azar, on
the other hand, parasites are difficult to find. Again, Nicolle
has found that dogs infected with Leishmania tropica appeared
to be not so susceptible to subsequent infection with Leishmania

PIROPLASMOSIS 637

infantum. These facts, however, might be consistent with the
existence of three species.

Histoplasma Capsulatum. — Under this name, Darling has described a
parasite observed by him in Panama, in certain cases characterised
during life by continued irregular fever, spleno-megaly, emaciation, and
ana-mia, and post mortem showing minute granulomata in the lungs,
irregular necrosis and cirrhosis of the liver, — the spleen, naked-eye,
resembling that of spleno-mvelogenous leukemia. In smears from the
lung nodules, the liver, and spleen, stained by Irishman's method,
there were observed enormous numbers of small bodies sometimes crowd-
ing endothelial cells, often free. These bodies were round or oval and
from 1 to 4 /x in diameter. Each contained an irregularly placed
chromatin mass, the shape of which was globular, oval or kidney-shaped,
the remainder of the parasite consisting of bine-staining basophilic
substance. The parasite is surrounded by a non-staining refractile
capsule, one-sixth of the diameter of the parasite in width and sometimes
containing a single minute chromatoid dot, and similar granules are
sometimes seen in the non-chromatoid part of the body of the parasite.
Darling considers this organism to be different from the Leishmania
donovani in the form and arrangement of its chromatin and in not
possessing a blepharoplast.

PlKOI'LASMOSIS.

Up to the present no human disease has been proved to be associated
with the presence of piroplasmata. The observations of Donovan,
which seemed to indicate that the parasite of kala-azar might be found
within the red blood corpuscles, and which led Laveran to denominate
the Leishmania donovani the piroplasma donovani, have, as already
indicated, not been confirmed ; the same is true of the association of
piroplasms with the occurrence of the Rocky Mountain spotted fever
sometimes prevalent in Montana. But several important diseases of the
lower animals are almost certainly caused by protozoan parasites of this
group, and a short account of the organisms may be given.

The piroplasmata are pear-shaped unicellular organisms about 1 to 1 '5 /*
long and varying in breadth. The peripheral part is denser than the
central, whicl'i often appears as if vacuolated, and at the broad end there
is ;i well-staining chromatin mass. Sometimes irregular and ring-, rod-,
or oval-shaped individuals occur. The organisms are found within the
red Mood corpuscles of the infected animal and also free in the blood. In
the former situation there is sometimes only one within a cell, but the
numbers vary under different circumstances and in different species.
Multiplication takes place by fission, and the new individuals, remaining
fui longer or shorter times in apposition, account for some of the appear-
•noea seen in cells. Especially in the forms free in the blood pseudopodial
prolongations of the protoplasm, usually from the. pointed end, arc
developed, and it may be by means of such pseudopodia that entrance to
the red cells is obtained. Infection fs usually carried from infected
unimals by means'of ticks. In one case Koch has described the develop-
ment in the organism, in the stomach of the tick, of spiked protoplasmic
processes sprouting out from the broad end of the piro plasm, and the
occurrence of conjugation of two such individuals by their narrow ends
to form a xygote. Further observations, however, here are necessary,

638 PIROPLASMOSIS

and nothing is known of the further history of the parasite within the
insect except that the eggs in the ovary may become infected, so that
insects developed from these can carry infection to animals. Frequently
when an animal has passed through an attack of a piroplasmosis it is
immune to the disease, and with regard to this immunity in certain cases
very interesting facts have been observed. For instance, the condition
may not be associated with the disappearance of the parasite from the
blood of the immune animal, and the latter may thus be a source of
danger to other non-immune animals with which ticks harboured by it
may come in contact.

The following are the chief piroplasmata causing disease in animals : —
(i) Piroplasma bigeminum. This was first described by Theobald Smith
and is the cause of Texas or red- water fever, a febrile condition associated
with hffimoglobinuria, which occurs in the Southern States of America,
the Argentine, South and Central Africa, Algeria, various parts of
Northern Europe, and in Australia. The organism gets its name of
bigeminum from the fact that it is often present in the red cells in pairs,
which may be attached to one another by a fine thread of protoplasm ;
this probably results from the complete separation of two individuals
being delayed after division has occurred. Infection is here spread by
the tick boophilus bovis, and some of the characteristics of the disease
epidemiologically are explained by the fact that this insect goes through
all its moultings on the same individual host. (2) Piroplasma parvum.
This organism was discovered by Theiler in the blood of cattle suffering
from African East Coast fever, a disease closely resembling Texas fever,
which prevails endemically in a narrow strip along a long extent of the
east coast, and which occurs epidemically inland. As its designation
implies, the organism is small, and it is also attenuated. Its insect host
is the tick rhipicephalus appendiculatus, and it may be noted that this
tick drops off the animal on which it may be feeding when it is about to
go through one of its several moultings. It can thus carry an infection
much more quickly and widely through a herd than can the carrier of
ordinary red-water fever. It may be said that in England there occurs a
red-water fever also associated with the presence of a piroplasm in the
blood, but the relationship of this organism to the other varieties has not
yet been fully worked out. (3) Piroplasma equi. This organism gives
rise to biliary fever in horses, another South African disease, and it is
carried by the tick rhipicephalus evertsii. In this disease Theiler made
the interesting observation that when the blood of a donkey which had
recovered from the disease was injected into a horse, the latter suffered a
slight illness only, although the organisms were present in the blood
injected. Such a fact is of importance, as attenuation of virulence in
pathogenic protozoa seems, so far as our present knowledge goes, a not
very common event. (4) Piroplasma canis. This causes a piroplas-
mosis occurring in dogs.

With regard to the pathology of infection by piroplasmata we know
nothing. The diseases are often extremely fatal, carrying off nearly
every individual attacked, but we do not know the nature of the changes
originated.

APPENDIX F.

YELLOW FEVER.

YKLLOW fever is an infectious disease which is endemic in the
\\ Yst Indies, in Brazil, in Sierra Leone and the adjacent parts
of West Africa, though it is probable that it was from the
first-named region that the others were originally infected. .
I Y<>m time to time serious outbreaks occur, during which
neighbouring countries also suffer, and the disease may be
carried to other parts of the world. In this way epidemics
have occurred in the United States and elsewhere, infection
usually being carried by cases occurring among the crews
of ships. In the parts where it is endemic, though usually
a few cases may occur from time to time, there is some
evidence that occasionally the disease may remain in abeyance
for many years and then originate de novo. There is, there-
t'mv, reason to suspect that the infective agent can exist for
considerable periods outside the human body. It is possible,
however, that continuity may be maintained by the occurrence
of a mild type of the disease, which may be grouped with the
" bilious fevers " prevalent in yellow fever regions. This would
explain the degree of immunity which is shown during a serious
epidemic by the older immigrants.

Great variations are observed in the clinical types under
which the disease presents itself. Usually after from two to
six days' incubation a sudden onset in the form of a rigor
occurs. The temperature rises to 104-105° F. The person is
livid, with outstanding bloodshot eyes. There are present great
prostration, pain in the back, and vomiting, at first of mucus,
later of bile. The urine is diminished and contains albumin.
About the fifth day an apparent improvement takes place, and
this may lead on to recovery. Frequently, however, the remission,
which may last from a few hours to two days, is followed
by an aggravation of all the symptoms. The temperature rises,
jaundice is observed, haemorrhages occur from all the mucous

63»

640 YELLOW FEVER

surfaces, causing, in the case of the stomach, the " black vomit "
— one of the clinical signs of the disease in its worst form.
Anuria, coma, and cardiac collapse usher in a fatal issue. The
mortality varies in different epidemics from about 35 to 99
per cent, of those attacked. Both white and black races are
susceptible, but those who have resided long in a country are
less susceptible than newr immigrants. An attack of the disease
usually confers complete immunity against subsequent infection.

Post-mortem the stomach is found in a state of acute gastritis,
and contains much altered blood derived from haemorrhages
which have occurred in the mucous and submucous coats. The
intestine may be normal, but is often congested and may be
ulcerated ; the mesenteric glands are enlarged. The liver is in
a state of fatty degeneration of greater or less degree, but often
resembling the condition found in phosphorus poisoning. The
kidneys are in a state of intense glomerulo-nephritis, with fatty
degeneration of the epithelium. There is congestion of the
meninges, especially in the lumbar region, and haemorrhages
may occur. The other organs do not show much change,
though small haemorrhages under the skin and into all the
tissues of the body are not infrequent. In the blood a feature
is the excess of urea present, amounting, it may be, to nearly
4 per cent.

Etiology of Yellow Fever. — Although a large amount of
bacteriological work has been done on yellow fever, this has
merely a historical interest, as it is now known that the
causal agent is not one of the ordinary bacteria, but belongs to
the group of ultra-microscopic organisms.1 A mosquito acts as
the intermediate host, and the facts detailed below point to the
organism passing through some cycle of development in the
body of the insect. The analogy of malaria makes it extremely
probable that the organism is a protozoon, but this has not yet
been completely proved. As bacteriological work led up to the
establishment of our knowledge regarding the nature of the
disease, some reference must be made to it.

A very full research into the bacteriology of yellow fever was-
that of Sternberg, and one of the organisms isolated, which he
called the bacillus x, appeared possibly to have some relationship
to the disease. Sanarelli in 1897 isolated an organism Avhich he
called bacillus icteroides, and which he considered to be the
cause of yellow fever ; it was probably identical with the bacillus

1 In several diseases the existence of such causal factors is probable.
Examples in animals are foot and mouth disease, South African horse sickness,
and the contagious pleuro-pueumonia of cattle.

ETIOLOGY OF YELLOW FEVER 641

x of Sternberg, but subsequent observations made by others
gave conflicting results. The bacillus icteroides, as described by
Sanarelli, belongs to the paratyphoid group, possessing lateral
fiagella, growing on gelatin without liquefaction, and fermenting
glucose but not lactose. Reed and Carroll found that it was
practically identical with the bacillus of swine cholera. It
must now be considered merely as an organism which may
occur in the organs and tissues in yellow fever as a secondary
infection, but without any etiological significance.

The facts of importance which have been established
regarding the etiology of the disease are due to the labours of
the United States Army Commission, which began its wrork in
1900. The members of the Commission first directed their
inquiries towards determining whether the bacillus icteroides
was present in the blood during life, and a series of cases was
investigated bacteriologically, with entirely negative results in
each instance. They then resolved to test the hypothesis of
Dr. Carlos Finlay of Havana, promulgated several years pre-
viously, that the disease wras carried by mosquitoes. Selecting
mosquitoes which they reared from eggs, they allowed them to
bite yellow fever patients and then to bite healthy men. Of
several experiments of this nature two were successful in the
first instance, the first individual to be infected in this way
being Dr. James Carroll, a member of the Commission, who
passed through a severe attack of typical yellow fever. Experi-
ments were then performed on a larger scale, with completely
confirmatory results, as to the conveyance of the disease by
mosquitoes. Of twelve non-immunes living under circumstances
which excluded natural means of infection, ten contracted
yellow fever after hiving been bitten by mosquitoes which had
previously bitten yellow fever patients ; happily all of these
recovered. Two of the men who were thus infected had been
previously exposed to contact with fomites from yellow fever
patients without results. These results were confirmed by
(luiu'-ras, whose investigations were carried out along similar
lines ; of seventeen individuals bitten by infected mosquitoes,
eight took yellow fever, and three of these died.

The species of mosquito used by the American Commission
was the Stefjomyia fasciata, and up to the present time no other
s]iecies has been found capable of carrying the infection. It has
also been determined that a certain }>eriod must elapse after the
insect has bitten a yellow fever patient before it becomes infec-
tive to another subject. In summer weather this period is about
twelve days ; at a lower temperature somewhat longer. This

642 YELLOW FEVER

probably means that, as in the case of malaria, the parasite must
pass through certain- stages of development before it reaches the
salivary gland and is thus in a position to be transferred to a fresh
subject. Infected mosquitoes, however, retain the power of
infection for a considerable time afterwards, probably as long as
sixty days. It has also been shown that mosquitoes may become
infective after biting a patient on the first, second, or third day
of the disease, but at a later period the results are usually
negative, apparently because the virus is no longer present in
the blood.

Interesting results were also obtained with regard to the
communication of the disease directly from patient to patient,
the conclusion arrived at, after careful experiments, being that
the disease cannot be transferred in this way, even when the
contact is of a close character. In a specially constructed house
seven men were exposed to the most intimate contact with the
fomites of yellow fever patients for a period of twenty days each,
the soiled garments worn by the patients being in some cases
actually slept in by these men ; the result was that not one of
those thus exposed contracted the disease. The conclusions on
this point have been subsequently confirmed by other workers.

The American Commission also found it possible to transmit
yellow fever to a healthy man by injecting small quantities of
blood or of serum taken from a yellow fever patient at any
period up till the third day of the disease. The period of
incubation in this case is somewhat shorter than when the disease
is conveyed by the bite of mosquitoes, the average duration in
the former case being about three days, and in the latter about
four days, though these times may be considerably exceeded.
It is also interesting to know that in these experimental injec-
tions the blood or serum used was found to be free from bacteria.
Up till the present time, we know of only these two methods of
infection, namely, indirectly by the bite of a mosquito infected
with the yellow fever germ; or directly by the injection of some
of the blood from a yellow fever patient. In these respects
there is a striking similarity to what has been established in the
case of malarial fever.

Experiments with regard to the nature of the yellow fever
organism were carried out by Reed and Carroll, and interesting
results were obtained. They found that the organism of the
disease was very easily killed by heat, as blood from a yellow
fever patient lost its infective power on being heated to 55° C.
for ten minutes. On the other hand, blood or serum was found
to be still infective after having been passed through a Berkefeld

ETIOLOGY OF YELLOW FEVER 643

tilt.-]. This has been confirmed by the French Commission,
with tin- additional result that the virus passes through a
( 'liaiiil>er]and F filter, but not through a Chamberland B. These
facts would show that the parasite is of extremely minute size,
and apparently belongs to the group of ultra-microscopic
organisms. l.Tp till the present time all attempts to find by
microscopic examination the yellow fever parasite, either in the
blood of patients suffering from the disease or in the tissues of
infective mosquitoes, have been attended with negative results.
It has U vii recently stated that it is possible to produce yellow
fever in the chimpanzee by the injection of blood from a patient.
Though nothing has been determined regarding the actual
nature of the virus, yet the results already obtained have
supplied the basis for preventive measures against the disease,
these being directed towards the destruction of mosquitoes and
the protection of those suffering from yellow fever, and also the
healthy, against the bites of these insects. Already a striking
dfgivc of success has been obtained in Havana. Such measures
came into force in February 1901, and in ninety days the town
was free of yellow fever, and for fifty-four days later no new
cased occurred; and although subsequently the disease was
reintroduced into the town, no difficulty was experienced in
stamping it out by the same measures. In recent years the
results have also been highly gratifying, and the disease may
be said to be practically eradicated from Havana. In other
places also successful results have been obtained, and epidemics
would ap]n'«ir to be now under control if the proper measures
are taken.

APPENDIX G.

EPIDEMIC POLIOMYELITIS.

DURING the past twenty years, the occurrence at one time and
place of groups of cases of acute anterior poliomyelitis has favoured
the idea that the condition might be of an infective nature.
This view was supported by the work of Landsteiner and Popper,
who, in 1909 in Vienna, succeeded in producing the disease in a
monkey by the intraperitoneal injection of an emulsion of the spinal
cord of a child who had succumbed on the fourth day of illness.
The occurrence in New York in the summer of 1907 of an
epidemic in which probably over 2000 cases occurred, 762
of which were carefully investigated by a Commission,
concentrated attention in America upon the condition, and a
recrudescence in the summer of 1909 furnished Flexner with
material for investigation. In earlier experiments it had been
observed that the cerebro-spinal fluid was non-infective, and
that, while it was possible to infect monkeys with the disease,
the transference of the condition from monkey to monkey wTas
not successful. Flexner found that if for intraperitoneal
injection intracerebral inoculation was substituted, the brain and
cord of the infected animals was infective for other monkeys,
the incubation period being from four to thirty-three days. In
this way the disease was kept up by subdural and intravenous
injection, and also by injection into the sheath of nerves such as
the sciatic, — the intraperitoneal and subcutaneous methods being
found also to give results. The disease produced resembled in
every way the disease naturally occurring in children, and
frequently resulted fatally, as in the natural illness. When
the virus was injected into the sheath of a nerve, the paralytic
symptoms first appeared in relation to that part of the cord
from which the nerve emerges. Infective material preserved in
glycerin retained its virulence, — a fact which rather militates
against there being a bacterial virus, — and the virus could also
withstand prolonged freezing and drying for a considerable period.

644

EPIDEMIC POLIOMYELITIS 645

In animals other than monkeys no result followed infection. A
raivful microscopic examination of the nervous system in natural
and experimental cases did not disclose the presence of bacteria
or protozoa, and it was found that the virus could pass through
a llrrkefeld lilter without losing its infectivity, and that the cords
of animals thus infected were still infective for further animals, —
men- toxic action being thus excluded. These results have been
continued by Levaditi working with the chimpanzee, and then- is
thus little doubt that here again we are dealing with an ultra-
micruscnj.ic vims. Flexner found that when the virus was
mixed with bouillon a turbidity developed, but no formed
organic body could be detected. He further found that monkeys
which had passed through the illness following inoculation \\ere
insusceptible to re-inoculation, and both he and Levaditi noted
that the serum of such insusceptible, monkeys had the capacity
of neutralising the virus, a similar result being obtained with
the serum of human cases which had recovered.

With regard to the distribution of the virus in an infected
animal, the chief concentration is found in the nervous system,
but it was also found in lymphatic glands ; as has been stated,
the cerebro-spinal fluid is apparently inert. No facts are known
bearing on the channel of natural infection or on the path by
which the poison reaches the cord, — whether by the general
lymph stream or by the sheaths of nerves, but the virus can
be eliminated by the nose, 'and infection can also be effected
through the scarified nasal mucosa. In its pathology the
condition bears many resemblances to rabies.

These recent observations are of the greatest importance in
relation to the etiology of the condition and also possibly in
relation to treatment.

APPENDIX H.

PHLEBOTOMUS FEVER.

IN Dalmatia, Herzegovina, and neighbouring parts of the
Adriatic littoral there occurs a disease known as "pappataci,"
characterised by fever lasting for about three clays, followed
by somewhat prolonged prostration, but very rarely having a
fatal issue. Doerr, after failing to isolate any organism from
the blood, found that the subcutaneous injection of from '5 to
1 c.c. of the serum from a case into a healthy individual was
followed about eight days later by an attack of the disease.
A similar effect was produced with the serum after it had been
passed through a Berkefeld filter, — all the inoculation experiments
being performed at a distance from the original location of the
disease. The view is therefore put forward that here we have
to deal with another example of an ultra-microscopic virus. The
disease has been only observed in the summer season, and Doerr
considered there was justification for the popular view that
it was associated with the bite of the dipterous fly, phlebotomm
pappatasii. This was borne out by the fact that on feeding
such flies on a sick person, transporting them to a locality free
from the disease and allowing them to bite healthy individuals,
the affection was reproduced. An apparently identical disease
occurs in Malta and has been investigated by Birt under the
name of "Phlebotomus Fever." This observer fully confirmed
Doerr's results, the condition again being reproduced by infected
flies which, however, were found not to manifest infectivity
earlier than seven days after biting.

These results are of importance of themselves as throwing
light on the etiology of a troublesome disease of the Mediter-
ranean littoral, but they are also interesting as having a possible
bearing on the pathology of a group of similar affections
occurring in various parts of the world, — chiefly in coastal areas,
— and going under a variety of names. Examples are dengue,
the three-day fever of various regions, Canary fever, Shanghai

C46

PHLEBOTOMUS FEVER 647

fever, Chitral fever, and the seven-day fever or simple continued
fever of India. Of these, that presenting the most definite
clinical picture is dengue, — a condition for long well known and
having an extensive distribution, and it may be said that
Ashburn and Craig in the Philippines found the blood in dengue
as in pappataci to be infective even after filtration. Whether
all tlie.se disease conditions are identical further research must
decide; at present Birt believes that at any rate pappataci
and dengue are distinct, and certainly Doerr does not in his
description allude to the terminal skin eruption which Manson
believes to be of very constant occurrence in the latter. The
rarity of a fatal result in these diseases makes their investiga-
tion by inoculation of the human subject relatively safe.

APPENDIX J.

TYPHUS FEVER.

UP till recently all attempts to elucidate the etiology of this dis-
ease by ordinary bacteriological methods have given equivocal
results. Certain experiments, however, performed by Nicolle
in 1909, during an outbreak in Tunis, are of importance. This
observer found that the inoculation of the monkeys, macacus
cynomolgus and macacus sinicus, with typhus blood gave a
negative result. On injecting 1 c.c. of such blood into a
chimpanzee, however, an illness presenting the features of the
disease in man, including the eruption, resulted three days later,
and death occurred. It was then found that the blood of this
animal was capable of originating a similar disease after an
incubation period of ten to fourteen days in the lower apes
referred to, and this was kept up through several passages.
The virulence of the material, however, gradually died out.
The dog and the white rat were insusceptible. It was found
that macacus sinicus could be infected by means of the body
louse, the incubation period, however, being lengthened to
forty days. These experiments probably throw important light
on the etiology of the condition, and on the means by which
the disease is spread in man.

648

BIBLIOGRAPHY.

L TKXT- BOOKS. — In English the student may consult the follow-
ing : " Micro-organisms and Disease," E. Klein, 3rd ed., London, 1896.
"Bacteriology and Infective Diseases," Edgar M. Crookshank, London,
1898. "A Manual of Bacteriology," George M. Sternberg, New York,
1st ed. 1893, 2nd ed. 1896 (this book contains a full bibliography).
"Text-Book upon the Pathogenic Bacteria," Joseph M'Farland, London,
f.th ed. 1906. "Practical Bacteriology," A. A. Kanthack and J. H.
Diysdalc, London, 189;"). " Bacteria and their Products," 0. S. Wood-
head, London, 1891. " Bacteriological Technique," Eyre, London, 1902.
Tin- articles on bacteriological subjects in CliHbnl Allbutt's " System of
Medicine," London, 1906-10, are of the highest excellence, and have full
bibliographies appended. For the hygienic aspects of bacteriology, see
" System of Hygiene," Stevenson and Murphy, London, 1892-94.

For non-pathogenic bacteria occurring in connection with pathological
\\<>rk consult Heim, op. cit. infra. For fungi, see I)e Bary, "Comparative
Morphology and Biology of the Fungi, Mycetoxoa and Bacteria/' transl.
by Garnsey and Balfour, Oxford, 1887 ; Sachs, "Text-Book of Botany,"
ii., transl. by Garnsey and Balfour, Oxford, 1887.

In German: "Die Mikroorganismen," by Dr. C. Fliigge, 3rd ed.,
Leipzig, 1896. " Lehrbuch der pathologischen Mykologie," by Baum-
-•••u ten, Braunschweig, 1890. " Handbuch der pathogenen Mikro-
ui-gaiiismen," Knlle and \Vassermann, Fischer, Jena, 1904. "Handbuch
<1. i Tfrluiik and Methodikderlmmunitiitsforschung," Krausand Levaditi,
Jena, 1908 and 1909.

In French : Roger, " Les maladies infectieuses," Paris, 1902.

I'KKioiui ALS.— For references to current work see (1) Uentralbl. f.
lla.ktcriol. u. Parasitenk., Jena. This publication commenced in 1887.
Two volumes are issued yearly. In 1895 it was divided into two parts.
Abtheilung I. deals with Medizinisch-hygienischc Bakteriologie und
ihi'-rixrhc I'anisitenkuwle. The volumes of this part are numbered cpn-
st( utively with those of the former series, the first issued thus being
vol. xvii. Commencing in 1902 with volume xxxi., each volume of
Abtheilung I. was further divided into two parts, one consisting of
Originate, the other of Jlefcmte. Abtheilung II. deals with A //</•;-
,,i ''iin- /<iii>lifirlsch(tftli<-)i-t«-liiinlogisclie Bakterioloyie, O&rwtgs-^kynoloffif
nii'l rjlanzenpatli »/<»//'<'. The first volume is entitled Zweite Abtheilung,
1 5-1. L It contains original articles, Referate, etc. (2) Bull. dcVInst.
Pasteur, Paris, Masson. Besides bacteriological abstracts this journal
contains many valuable reviews and analyses relating to protozoology.

649

650 BIBLIOGRAPHY

(3) "Ergebnisse der allgemeinen Pathologic," Lubarsch and Ostertag,
Wiesbaden, Bergrnann. This from time to time contains valuable
critical reviews.

The most complete account of the work of the year is found in the
Jahresb. it. d. Fortschr. . . . d. path. Mikrooryanismen, conducted by
Baumgarten, and published in Braunschweig. This publication com-
menced in 1887. Its disadvantage is that the volume for any year does
not usually appear till two years later.

Bacteriology is also dealt with in the Index Medicus. For valuable
lists of papers by particular authors see Royal Society Catalogue of
Scientific Papers and Internat. Cat. Sc. Lit. (Section, Bacteriology).

The chief bacteriological periodicals are the Journ. Path, and Bacteriol, ,
Cambridge, edited by G. Sims Woodhead ; the Ztschr. f. Hyg. n. Infec-
fionskrankh., Leipzig, edited by Koch and Flligge ; the Ztxchr. f.
Immunitatsforschuny, Frankfort, edited by Ehrlich ; and the Ann. de
Vlnst. Pasteur, Paris, edited by Duclaux ; Journ. Krper. Med., New
York, edited by Flexner ; Journ. Hyij., Cambridge, edited by Nuttall ;
Journ. Med. Research, Boston, edited by Ernst ; Journ. Infect. Disease*,
Chicago, edited by Hektoen ; Journ. of Royal Army Medical Corps, edited
by Bruce.

Valuable papers also from time to time appear in the Lancet, Brit.
Med. Journ., Deutsche med. Wchnschr., Bcrl. klin. Wchnsclir., tiemaine
med., Arch. f. Hyg., Arch. f. cxpcr. Path. u. PharmaJcol. Besides these
periodicals the student may have to consult the Reports of the Med. Off.
of the Local Government Board, which contain the reports of the medical
officers, also the Proc. Roy. Soc. London, the Compt. rend. Acad. d. .ST.
Paris, the Compt. rend. Soc. de biol., Paris, and the Arb. a. d. k. Gsndht-
samte. (the first two volumes of the last were denominated Mitthciluhgen}.

For general reviews on Portozoal and Tropical Diseases generally, see
Manson, "Tropical Diseases," London, 1908; and Mense. " Handbuch
der Tropenkrankheiten, Leipzig, 1906.

CHAPTER I. — GENERAL MORPHOLOGY AND BIOLOGY.

Consult here especially Fliigge, "Die Mikroorganismen." De Bary,
"Bacteria," translated by Garnsey and Bay ley Balfour, Oxford, 1887.
Zopf, "Zur Morphologic der Spalitpflanzen, " Leipzig, 1882; " Beitr. z.
Physiologic und Morphologie niederer Organismen," 5th ed., Leipzig,
1895. Graham-Smith, " Parasitology," iii. 17. Cohn, Beitr. z. Biol.
d. Pflanz., Bresl. (1876), ii. V. Nageli, "Die niederen Pilze," Munich,
1877; " Untersuchungen liber niedere Pilze/' Munich, 1882. Migula,
"System der Bakterien," Jena, 1897. Duclaux, "Traite de micro-
biologie," Paris, 1898-99. For general morphological relations, see
Marshall Ward, art. " Schizomycetes," Ency. Brit., 9th ed. xxi. 398;
xxvi. 51. Engler and Prantl, "Die natiirlichen Pflanzenfamilien,"
Lieferung, 129. " Schizophyta " (by W. Migula). STRUCTURE OF BAC-
TERIAL CELL. — Biitschli, " Uber den Bau der Bakterien," Leipzig, 1890 ;
" Weitere Ausfiihrungen liber den Bau der Cyanophyceen und Bakterien,"
Leipzig, 1896. Fischer, op. cit. in text. Buchner, Longard and Riedlin,
Centralbl. f. Bakteriol. u. Parasitenk. ii. 1. Ernst, Ztschr. f. Hyg.
v. 428. Babes, ibid. v. 173. Neisser, ibid. iv. 165. MOTILITY. —
Klein, Biitschli, Fischer, Cohn, loc. cit. Loftier, Centralbl. f. Bakteriol.
u. Parasitenk. vi. 209 ; vii. 625. PIGMENTS.— Zopf, loc. cit. ; Galeotti,
ref. in Centralbl. f. Bakteriol. u. Parasitenk. xiv. 696. Babes, Ztschr.

BIBLIOGRAPHY 651

f. ////'/. xx. 3. SPORULATION. — Prazmowski, Biol. Cent mill. viii. 301.
A. Koch. Boto*. Ztii. (1888), Nos. 18-22. Buclmer, Sitzunysb. d. math.-
phys. Cl. d. k. buyer. Akad. d. Wissmsch. zu Miinchen, 7th Feb. 1880.
II. Kifh. Mi'ffli. '«. d. /,-. (,'x)nth/mnnf>: i. 65. Dobell, Quart. Journ.
Ml,-,-, .sv. (190i'\ liii. CHKMICAL STRUCTURE OF BACTERIA. — Nencki,
/*'••/-. 'I. d.-utsi'lt. chfin. Uwllxch. (1884), xvii. 2605. Cramer, Arch. /.
////.-/. xvi. 1".4. Buehner, Bcrl. klin. //'.-A//..*-///-. (1890), 673, 1084;
rid,- Klii^i:*', "p. fit. CLASSIFICATION OF BACTERIA. — For general review,

Marshall Ward. Ami. of Botany, vi. 103; Migula, loc. clt. xti/n-n.
!•'<••>!> OF HA- rr.Kiv. --Xiigeli, Cohn, op. cit. Pasteur, "Etudes sur la
bii'-iv," 1S7«>. Hiieppr, Miftli. a. d. k. Gsndhtsamtc. ii. 309. RELATIONS
TO OXYCJKN.— Pasteur, t'ompt. rend. Acad. d. sc. Hi. 344, 1142. Kitasato
and \\Vyl. Zlscli /-../'. ////;/. viii. 41, 404 ; ix. i»7. TEMPERATURE.— Vide
Hiiirjr,..' ,,y;. c/V. For thermopliilic bacteria, Rabinowitsch, Ztschr. f.
HIKJ. \\. l.")4. Mai'fad\-en and Blaxall, Journ. Path, and Bacterial, iii.
-7. ACTION OF BACTKEIAL FERMENTS. — Salkowski, Ztschr. /. Biol.,
N. I-'., \ ii. 92. Pasteur and Joubert, Conijit. rend. Acad. d. sc. Ixxxiii. 5.
Shnidaii Lea, Jnnrn. /'////x/W. vi. 136. Beijerinck, Ccntralbl. f. Bak-
t' ,-lnf. a. I'aritsitenk. Abth. II. i. 221. E. Fi.seber, /ter. ^. deutsch. chem.

Vx.-//. \\viii. 1430. Liborius, Xtx.hr. f. Hyy. i. 115. See also
Pasteur. •' Royal Society Catalogue of Scientific Papers." Green, "The
Soluble Ferinriits and Fermentation," Cambridge, 1899. VARIABILITY.
— Cohn, Xii-^'li, Kliigge, op. cit. Winogradski, " Beitr. •/.. Morph. u.
Pliysiol. d. Bakt./' Leipzig, 1888. Ray Lankester, Quart. Joum. M'n-r.
,Vc., N.S. (1873), xiii. 408 ; (1876), xvi. 27, 278. NITRIFYING OROAMSMS.
—Winogradski, Ann. de VInst. Pasteur, iv. 213, 257, 760; v. 92, 577.
Ma/r. ill,!. \i. 11 ; xii. 1, 263.

CHAPTER II.— METHODS OF CULTIVATION OF BACTEIMA.

For CKNKHAI. PIMM IIM.KS. — Pasteur, Compt. rend. Acad. d. sc. 1.
303 ; Ii. 348, 675 ; Ann. de chem. Ixiii. 5. Tyndall, "Floating Matter
of the Air in Relation to Putrefaction and Infection," London, 1881.
H. C. Bastian, "The Beginnings of Life," London, 1872. METHODS OF
Si I.I;II.ISATION.— R. Koch, Gatt'ky, and Li -tiler, Mitth. a. d. k. Gsndhts-
'•utte. i. 322. Koch and WolfFhiigel, ibid. i. 301. CULTURE MEDIA.

S-i- tr\t -Looks, csj.oc-ially Kanthack and Drysdale, Eyre. Pasteur,
" Btndea sur la bi.'-re." Paris, 1876. R. Koch, Mitth. a. d. k. Gsndhts-
<nn/,'. i. 1. Roux et Nocard, Ann. de Vlnst. Patteur, i. 1. Roux, ibid.
ii. 28. Marmorek, ibid. ix. 593. Kitasato and Weyl, op.^ cit. .supra. P.
and Mrs. Percy Frankland, "Micro-organisms in Water," London, 1894.
Fuller, Rep. "Amer. Pub. Health Ass. xx. 381. Theobald Smith,
Centralbl. f. Baktcriol. u. Parasitenk. vii. 502 ; xiv. 864. Durham,
Hi-it. Med. Journ. (1898), i. 1387. "Report of American Committee
on Bacteriological Methods," Concord, 1898. MacConkey, Tkompson-
}~,if,-s ,i,id Johnston Lab. Rep. vol. iii. pt. iii. 151 ; vol. iv. pt. i. 151 ;
Journ. Hyy. v. 333. Griinbaum and Hume, Brit. Mcd. Journ.
June 14, 1902. Drigalski and Conradi, Ztschr. f. Hyg. xxxix. 283.
Kudo, <', lttmm. f. Bakteriol. u. Parasitenk. (Orig.), xxxv. 109. Con-
radi. ;/,/,/., Beilage zu. Abth. I. Bd. xlii. (1908) (Referate), p. *47.
Fawcus, Journ. R.A.M.C. xii. 147. Sabouraud, " Les Teignes, " Paris,
1910. INDOL REACTIONS.— Bohme, Centralbl. /. Bakteriol. u. Para-
sitfiik. Abth. I. (Orig.) xl. 129. Steensma, ibid. xii. 295. Marshall,
Journ. Hyy. vii. 581. MacConkey, ibid. ix. 86.

652 BIBLIOGRAPHY

CHAPTER III.— MICROSCOPIC METHODS.

Consult text-books, especially Klein, Kanthack and Drysdale, Hueppe,
Gimther, Heim. Also Bolles Lee, "The Microtomist's Vademecum,"
London, 1905 (this is the most complete treatise on the subject).
Rawitz, op. cit. in text. Koch, Mitth. a. d. k. Gsndhtsamte. i. 1.
Ehrlich, Ztschr. f. Tclin. Med. i. 553; ii. 710. Gram, Fortschr. d.
Med. (1884), ii. No. 6. Nicholle, Ann. de I'lnst. Pasteur, ix. 666.
Kiihne, " Praktische Anleitung zum mikroscopischen Nachweis der
Bakterien im tierischen Gewebe," Leipzig, 1888. Van Ermengen, ref.
Centralbl. f. Bacterial, u. ParasitenTc. xv. 969. Richard Muir, Journ.
Path, and Bacterial, v. 374. Mann, "Physiological Histology,"
Oxford, 1902. For Romanowsky methods, see Jenner, Lancet (1899),
i. 370. Leishman, Brit. Med. Journ. (1901), i. 635 ; (1902), ii. 757 ;
Journ. R.A.M.C. (1904), ii. 669. Geimsa, Deutsche Med. Wchnschr.
(1905), 1026 ; Ann. de I'lnst. Pasteur, xix. 346. MacNeal, Journ.
Inf. Diseases, iii. 412. Wright, ,T. H., Journ. Med. Research, vii. 138.
Wilson, Journ. Exp. Med. ix. 645.

CHAPTER IV.— METHODS OF EXAMINING THE PROPERTIES OF SERUM
— PREPARATION OF VACCINES — GENERAL BACTERIOLOGICAL
DIAGNOSIS — INOCULATION OF ANIMALS.

AGGLUTINATION. — Delepine, Brit. Med. Journ. (1897), ii. 529, 967.
Widal and Sicard, Ann. de I'lnst. Pasteur, xi. 353. Wright, Brit. Med.
Joum. (1897), i. 139 ; (1898), i. 355. Park and Collins, Journ. Med.
Research (1904), xii. 491. Bainbridge, Journ. Path, and Baeteriol. xiii.
443. Win slow and Rogers, "Biological Studies by the Pupils of
William Thompson Sedgwick," Boston, 1906.

GENERAL METHODS. — Wright, A. E., "Studies on Immunity," London,
1909. Muir, Robert, "Studies on Immunity," London, 1909. Ehrlich,
" Gesammelte Arbeiten zur Immunitiitsforschung," Berlin, 1904. These
works contain methods applied in the investigation of the subjects dealt
with in this chapter. The following are additional references relating to
special points : —

OPSOXIC METHODS. — Klien, H., John* Hopkins Hosp. Bull. (1907),
xviii. 245. Simon, Journ. Exp. Med. (1907), 487.

WASSERMANN REACTION. — Gengou, Ann. de FInst. Pasteur (1902),
xvi. 734. Moreschi, Berl. Jclin. Wchnschr. (1905), 1181 ; (1906), 100.
Wassermann and Brack, Deutsche med. Wchnschr. (1906), 100. Wasser-
mann, Neisser, and Bruck, ibid. (1906), 745. McKenzie, Journ. Path, and
Baeteriol. (1909), xiii. 311. Neisser, Milnchen. med. Wchnschr. (1909),
No. 21, 1076.

PREPARATION OF VACCINES. — Harrison, Journ. R.A.M.C. (1905),
iv. 313. Leishman, Harrison, G rattan, and Archibald, Hid. (1908), x.
583 ; (1908), xi. 327.

CHAPTER V. — BACTERIA OF AIR, SOIL, WATER — 'ANTISEPTICS.

AIR, SOIL, AND WATER.— -Petri, Ztschr. f. Hyg. iii. 1 ; vi. 233.
Fliigge, ibid. xxv. 179. Sticher, ibid. xxx. 163. Weyl, "Handbuch
der Hygiene," Jena, 1896, et seq. Houston, Rep. Med. Off. Local Gov.
Bd. xxvii. (1897-98) 251 ; xxviii. (1898-99) 413, 439, 467 ; xxix. (1899-
1900) 458, 489. Sidney Martin, ibid. xxvi. (1896-97) 231 ; xxvii.

BIBLIOGRAPHY 653

(1897-98) 308; xxviii. (1898-99) 382. Horrocks, "Bacteriological
Examination of Water," London, 1901. Percy and G. C. Frankland,
" Micro-organ isma in Water," London, 1894. Dibdin, " Purification of
Sewage and Water," London, 1897 ; Ann. Rep. Bd. Health, Mass.,
Boston, 1890, et seq. Savage, "The Bacteriol. Exam, of Water Supplies,"
London, 1906. Lewis, Rideal, and Walker, Journ. Roy. San. Inst.
(1903), xxiv. 424. Prescott and Winslow, "Elements of Water
Bacteriology," New York, 1908. Houston, "Annual Reports of Metro-
politan Water Board," 1907, el seq. ; "Reports on Research Work, Metro-
politan Water Board," 1907, et seq. MacConkey, Journ. Hyg. (1908),
vol. viii. 322 ; (1909), vol. ix. 86. Mair, ibid. (1908), vol. viii. 609.
Lorrain Smith, "Third Rep. Roy. Comm. on .Sewage Disposal " (1903), ii.
As i ISI.ITK s. — R. Koch, Mitth. a. d. k. Gsndhtsamte. i. 234. Behring,
/T/.v///-. f. Jlijij. ix. 395. Ritchie, Trans. Path. Soc. London, 1. 256.
Rideal, "Disinfection and Disinfectants," London, 1898. -Chick and
Martin, Journ. Hi/if. (1908), vol. viii. 654, 698. Chick, ibid. vol.
viii. 93.

CHAPTER VI. — RELATIONS OF BAOTKRIA TO DISEASE, ETC.

As the observations on which this chapter is based are scattered
through the rest of the book, the references to them will be found under

the ditl'crcnt disra---.

< IIAPTER VII.— INFLAMMATORY AND SUPPURATIVE CONDITIONS.

Ogston, 11 fit. Med. Journ. (1881), i. 369. Rosenbach, "Mikro-
organismen bei den Wundinfectionskrankheiten des Menschen," Wies-
baden, 1884. Passet, Fortschr. d. Med. (1885), Nos. 2 and 3. W.
Watson Cheyne, "Suppuration and Septic Diseases," Edinburgh, 1889.
Grawitz, Virchow's Archiv, cxvi. 116 ; Deutsche med. Wchnschr. (1889),
No. 23. Steinhaus, Ztschr. f. Hyg. v. 518 (micrococcus tetragenus) ;
"Die Aetiologie der acuten Eiterung," Leipzig, 1889. Christmas-
Dirckinck-Holmfeld, " Recherches expe>imentales sur la suppuration,"
Paris, 1888. Muir, Journ. Path, and Bacteriol. vii. 161 ; Trans. Path.
Soc. London, 1902. Garre, Fortschr. d. Med. (1885), No. 6. Marmorek,
Ann. dc, VInst. Pasteur, ix. 593. Petruschky, Ztschr. f. Hyg. xvii. 59;
xviii. 413; xxiii. 142 (with Koch, xxiii. 477). Lubbert, " Biologische
Spaltpilzuutersuchung," Wiirzburg, 1886. Krause, Fortschr. d. Med.
(1884), Nos. 7 and 8. Ribbert, ibid. (1886), No. 1. Widal and
Bcsancon, Ann. de VInst. Pasteur, ix. 104. V. Lingelsheim, Ztschr.
I. Hyg. x. 331 ; xii. 308. Behring, Centralbl.f. Bakteriol. u. Parasitenk.
xii. 192. Thoinot et Masselin, Rev. dc med. (1894), 449. Orth and
Wyss.,k..\vits,-h, Ccntralbl. f. d. med. Wisscnsch. (1885), 577. Netter,
Arch. 'l< jtlujxhJ. norm, rt path. (1886), 106. Weichselbaum, Wien. med.
Wchnschr. (1885), No. 41 ; (1888), Nos. 28-32 ; Central bl. f. Bakteriol. u.
Parasitenk, ii. 209 ; Beitr. r. path. 'Anat. u. c. allg. Path. iv. 127.
Becker, Deutsche med, Wchnschr. (1883), No. 46. Lannelongue et
Acliard, Ana. dc VInst. Pasteur, v. 209. Fehleisen, "Die Aetiologie
des Erysipel-v' Berlin, 1883. Welch, Am. Med. Journ. Sc. (1891), 439.
Lemoine, Ann. de I Inst. Pasteur, ix. 877. Kurth, Arb. a. d. k.
Gsndhtsamte. vii. 389. KnoiT, Ztschr. f. Hyg. xiii. 427. Bnlloch,
Lancet (1896), i. 982, 1216. Bordet, Ann. de VInst. Pasteur, xi. 177.
Booker (streptococcus enteritis), Johns Hopkins Hoxp. Rep. vi. 159.

654 BIBLIOGRAPHY

Hirsch, Centralbl. f. Bakteriol. u. Parasitenk. xxii. 369. Libman, ibid.
xxii. 376. Wright and Douglas, Proc. Roy. Soc. Loud. Ixxiv. 147.
Wright, Clinical Journal (1906), May 16.

STREPTOCOCCI. — Hiss, Journ. Exper. Med. vi. 317. Schottmiiller,
Munchen. med. Wchnschr. (1903), 849. Gordon, Reports Med. Officer
Local Gov. Board (1905), 388 ; Lancet (1905), ii. 1400. Andrewes and
Border, Lancet (1906), ii. Ruediger, Journ. Infect. Diseases, iii. 755.
Besredka, Butt, de I'Inst. Pasteur, iii. Nieter, Ztschr. f. Hyg. (1907), Ivi.
307. Mandelbaum, ibid. (1907), Iviii. 26. Levy, Arch. f. path. Anal.
(1907), ccxxxvii. 327. Centralbl. f. Bakteriol. u. Parasitenk. Abtheil I.
xliii. 793, et seq. (hsemolytic properties).

CONJUNCTIVITIS. — Morax, Ann. d I'Inst. Pasteur (1896), x. 337. Eyre,
Journ. Path, and Bacteriol. vi. 1. M tiller, Wien. med. Wchnschr.
1897 ; Inglis Pollock, Trans. Ophthalm. Soc. 1905 ; Axenfeld, in Lubarsch
and Ostertag, "Ergebnisse der allgem. Pathol. u. Path. Anat.," 1901 ;
" Die Bakteriologie in der Angenheilkunde," 1907 (full references).

ACUTE RHEUMATISM. — Tribouletand Cayon, Bull. Soc. med. d. h6p. de
Paris (1898), 93. Westphal, Wassermann, and Malkoff, Berl. klin.
Wchnschr. (1899), 638. Poyntoii and Paine, Lancet (1900), ii. 861, 932
(full references). Trans. Path. Soc. Lond. (1902), liii. 221 ; Lancet,
December 1905. Beaton and Walker, Brit. Med. Journ. (1903), i. 237.
Shaw, Journ. Path, and Bacteriol. (1903), ix. 158. Beattie, ibid. ix. 272,
xiv. 432 ; Journ. Med. Research, xiv. 399 ; Journ. Exper. Med. ix. 186.
Cole, Journ. Infect. Diseases, i. 714. Beattie, Journ. Path, and Bacteriol.
xiv. 432.

CHAPTER VIII. — INFLAMMATORY AND SUPPURATIVE CONDITIONS,
CONTINUED : ACUTE PNEUMONIAS, EPIDEMIC CEUEBRO - SPINAL
MENINGITIS.

Friedliinder, Fortschr. d. Med. i. No. 22 ; ii. 287 ; Virclww's Archiv,
Ixxxvii. 319. Fraenkel, A., Ztschr. f. klin. Med. (1886), 401. Salvioli
and Zaslein, Centralbl. f. d. med. Wissensch. (1883), 721. Ziehl, ibid.
(1883), 433 ; (1884), 97. Klein, ibid. (1884), 529. Jiirgensen, Berl. klin.

Wchnschr. (1844), 270. Seibert, ibid. (1884), 272, 292. Senger, Arch,
f. exper. Path. u. Pharmakol. (1886), 389. Weichselbaum, Wien. med.

Wchnschr. xxxvi. 1301, 1339, 1367; Monatschr. f. Ohrenh. (1888), Nos.
8 and 9 ; Centralbl. f. Bakteriol. u. Parasitenk. v. 33. Gamaleia, Ann.
de I'Inst. Pasteur, ii. 440. Guarnieri, Atti d. r. Accad. med. di Roma
(1888), ser. ii. iv. Kruse and Pansini, Ztschr. f. Hyg. xi. 279. E.
Fraenkel and Reiche, Ztschr. f. klin. Med. xxv. 230. Sanarelli, Centralbl.
f. Bakteriol. u. Parasitenk. x. 817. Launelongue, Gaz. d. hop. (1891),
379. Netter, Bull, et mem. Soc. med. d. h6p. de Paris (1889) ; Compt.
rend. Acad. d. sc. (1890) ; Compt. rend. Soc. de biol. Ixxxvii. 34.
G. and F. Klemperer, Berl. klin. Wchnschr. (1891), 869, 893. Foa and
Bordoni-Uffreduzzi, Deutsche med. Wchnschr. (1886), No. 33. Emmerich,
Milnchen. med. Wchnschr. (1891), No. 32. Issaeff, Ann. de I'Inst.
Pasteur, vii. 260. Grimbert, Ann. de I'Inst. Pasteur, xi. 840. Wash-
bourn, Brit. Med. Journ. (1897), i. 510 ; (1897), ii. 1849. Eyre and
Washbourn, Journ. Path, and Bacteriol. iv. 394 ; v. 13. See also
Brit. Med. Journ. (1901), ii. 760. Neufeld and Rimpau, Ztschr. f. Hyg.
Ii. 283. Rosenow, Journ.: Am. Med. Ass. (1908), Ii. No. 19. Neufeld and
Handel, Ztschr. f. Immunitdtsforschung (1909), iii. 159. Tschistowitch and
Jurewitch, Compt. rend. Soc, de biol. (1908), Ixiv. pp. 1044, 1095.

BIBLIOGRAPHY 655

Commission to Invi-stigate Acute Resp. Dis. (Hiss and others), see
Jour*. A'./-//. M'd. (1905), vii. pp. 403-632.

MK\IM;ITIS.— Weichselbauiu, Fortschr. d. Med. (1887), v. 573, 620.
JHCIOT. Xtxi-hr.f. Ifii'.i. xix. 351 ; xliv. 225. Councilman, Mallory, and
Wright. " Epidemic Cerebro-spinal Meningitis," Rep. Bd. Health, Mass.,
.11, 1898 (full references). Gwyn, Johns Hopkins Hosp. Bull. (1899),
109. v. Lingt'lslu'ini, Klin. Jahrb. xv. 373. Kolle and Wassermann,
ibul. p. 507. Kutscher, Deutsche med. Wchnschr. (1906), 1071. Betteu-
court and Franca, Ztschr. f. Hyg. xlvi. 463. Durham, Journ. Infect. Dis.
XniH>1. No. 2, p. 10. Goodwin and von Sholly, ibid. p. 21. Arkwriglit,
Jnurn. nf J/fft/. vii. 145 ; ix. 104. Flexner, Journ. Exper. Med. ix. 105.
Klt-xner and jobling, ibid. x. 141, 690. Vansteenberghe et Grysez, Ann.
</< ////.>/. /v/.sV, ///•. xx. 69. Gordon, Report to Local Govt. Board on the
Mi( rococcus of Cerebro-spinal Meningitis," London, H.M. Stationery Office,
1907. Albrecht and Ghou, Wien. Tclin. Wchnschr. (1901), xiv. 984 ;
/,' /-. Newr. and I'^ichlat. (1907), v. 593, 686. Stuart M 'Donald, Journ.
/'»th. //,/// Bncteriol. (1908), xii. 442. Shennan and W. T. Ritchie, ibid.
xii. 4f>(5. M'Kenxie and Martin, ibid. xii. 539. J. Ritchie, ibid.
(1910), xiv. 615. Elser and Huntoon, Journ. Med. Research (1909), xx.
371. Discussion in Brit. Med. Journ. (1908), ii. 1334.

CHAPTER IX. — GONORRHCEA, SOFT Son i..

<;<>N«II;KH<KA. — Neisser, Centralbl. f. d. med. Wissensch. (1879), 497 ;
lt,-nt*'lt,< »ned. Wchnachr. (1882), 279 ; (1894), 335. Bumm, "Der Mikro-
or^anismus der gonorrhoischen Schleimhauterkrankungen," Wiesbaden,
1885, 2nd ed. 1887 ; Miinchcn. med. Wchnschr. (1886), No. 27 ; (1891),
Nos. 50 and 51 ; Centralbl. f. Gyndk. (1891), No. 22 ; Wien. med. Presec
(1891), No. 24. Bockhart, Monatsh. f. prakt. Dcrmat. (1886), v. No. 4 ;
(1887), vi. No. 19. Steinschneider, Berl. klin. Wchnschr. (1890), No.
•Jl ; H893), No. 29 ; Verhandl. d. deutsche dermat. Gescllsch. I. Congress,
Wirn (1889), 159. Wertlu-im, U'icn. klin. Wchnschr. (1890), 25 ;
It. -nfv-fte med. ll'<-hnschr. (1891), No. 50 ; Arch. f. Gyniik. xii. Heft 1 ;
CetitraW. f. (.'••fl,«k. (1891), No. 24; (1892), No. 20; Wien. klin.
FTdbudtr. (1894), 441. Gerhardt, Charity-Ann. (1889), 241. Leyden,
Zf.«rlu: f. klin. ^f«/. xxi. 607; Deutsche, med. Wchnschr. (1893), 909.
Bordoni-Uflreduzzi, ibid. (1894), 484. Councilman, Am. Journ. Med.
A'--. <-vi. 277. Finger, Ghou, and Schlagenhaufer, Arch. f. Dermat. ».
,S'///'//. xxviii. 3, 276. Lang, ibid. (1892), 1007; Wien. med. Wchnschr.
M>!'i;, No. 7; "Der Venerische Katarrh, dessen Pathologic und
Therapie," Wit-sluulcn, 1893. Klein, Mmmtschr.f. Gcburtsh. u. Gynaek.
(1895), 33. Michaelis, Ztschr. f. klin. Med. xxix. 556. Heimaii, New
York Med. Rec. (1895), 769 ; (1896), Dec. 19. Foulerton, Trans. Brit.
Inst. Prer, ,i . M,,l. \. 40. De Christmas, Ann. de Vlnst. Pasteur, xi.
609. Nicolaysen, Centralbl. f. Bakteriol. u. Parasitenk. xxii. 305.
K.-ndu, /-''/•/.'/.•/;//. Jl'r/inschr. (1898). 431. Wassermann, Ztschr. f.
////v. xx vii. 2!»s : .!/;/,/«•//»•,/. ,,inl. )lrchnschr. (1901), No. 8. Leiihartz,
/;.'/•/. /•////. Wchiwhr. (1897), 1138. Thayer and Lazear, Journ. Exper.
.]/<->/. iv. 81. Kiinig. Berl. klin. Wchnschr. (1900), No. 47.. De Christmas,
Ann. '/. /'Inst. Pasteur (1900), xiv. 331. Raskai, Deutsche med. WchMclir.
(1901), No. 1. Jundell, Centralbl. f. Bakteriol. u. Parasitenk. xxix.
•J-J4. ("lo ml. ini, ibiil. xxiv. 955. Hrcssel, Miinchen. med. Wchnschr.
(1903\ No. 13. Moller, Arch. f. Dermat: u. Syph. (1904), Ixxi. 269.
Wyiin. Lnn'-.t (1905), i. 352. Prochaska, Arch. f. klin. Med. Ixxxiii.

656 BIBLIOGRAPHY

Heft 1-2. Strong, Journ. Am. Med. Ass. May 1904. Gurd, Journ. Med.
Research (1908), xviii. 291. Th. Vanned, Centralbl. f. Bakteriol. u.
Parasitenk. Abth. I. (Orig.) (1907), xliv. 10, 110. Hamilton, Journ. Infect.
Diseases (1908), v. 133. Brons, Klin. Monatsbl. f. Augenheilk. (1907),
xlv. 1. Torrey, Journ. Med. Research (1908), xix. 471. Martin, Journ.
Path, and Bacterial. (1910), xv. 76.

SOFT SORE. — Ducrey, Monatsh. f. praTct. Dermat. ix. 221. Krefting,
Arch. f. Dermat. u. Syph. (1892), 263. Jullien, Journ. d. mal. cutan. et
syph. (1892), 330. Unna, Monatsh. f. prakt. Dermat. (1892), 475 ; (1895),
61. Quinquand, Semaine mid. (1892), 278. Petersen, Centralbl. f.
Bakteriol. u. Parasitenk. xiii. 743 ; Arch. f. Dermat. n. Syph. (1894),
419. Audrey, Monatsh. f. prakt. Dermat. (1895), 267. Colombini,
Centralbl. f. Bakteriol. u. Parasitenk. xxv. 254. Nicolle, Presse medicale
(1900), 304. Bezan9on, Griifon, and Le Sourd, Ann. de dermat. et de
syphilolog. (1901), tome ii. 1. Lenglet, ibid. (1901), tome ii. 209. Simon,
Compt. rend. Soc. biol. (1902), 547. Tomasczewski, Ztschr. f. Hyg.
(1903), Bd. 43, p. 327. Davis, Journ. of Med. Research (1903), ix. 401.

CHAPTER X.— TUBERCULOSIS.

Klencke, " Untersuchungen und Erfuhrungen im Gebiet der Anatomic,"
etc., Leipzig, 1843. Villemin, "De la virulence et de la specificite de
la tuberculose," Paris, 1868. _ Cohnheim and Fraenkel, " Experimentelle
Untersuchungen iiber der Ubertragbarkeit der Tuberculose auf Thiere."
Cohnheim, "Die Tuberculose vom Standpunkt der Infectionslehre,"
1879. Various Authors, "Discussion sur la tuberculose," Bull. Acad.
de med. (1867), xxxii., xxxiii. Armanni, " Novimento med.-chir.,"
Naples, 1872. Baumgarten, " Lehrb. d. path. Myk.," 1890. Straus,
"La tuberculose et son bacille," Paris, 1895. Koch, Berl. klin.
Wchnschr. (1882), 221; Mitth. a. d. k. Gsndhtsamte. 1884 ; Deutsche
med. Wchnschr. (1890), No. 46a. ; (1891), Nos. 3 and 43 ; (1897), No. 14.
Bulloch, Lancet (1901), ii. 243. Nocard, "The Animal Tuberculoses,"
(transl.), London, 1895. Cornet, Ztschr. f. Hyg. v. 191. Nocard and
Roux, Ann. de VInst. Pasteur, \. 19. Pawlowsky, ibid. ii. 303.
Sander, Arch. f. Hyg. xvi. 238. Coppen Jones, Centralbl. f. Bakteriol.
u. Parasitenk. xvii. 1. Prudden and Hodenpyl, New York Med. Rec.
(1891), 636. Vissman, Virchow's Archiv, cxxix. 163. Straus and
Gamaleia, Arch. de. med. exper. et d'anat. path. iii. No. 4. Courmont,
Semaine med. (1893), 53 ; Revue de med. (1891), No. 10. Hericourt and
Richet, Bull. med. (1892), 741, 966. Williams. Lancet (1883), i. 312.
Pawlowsky, Ann. de I'Inst. Pasteur, vi. 116. Maffucei, " Sail' azione
tossica dei prodotti del bacillo della tuberculosi " ; Centralbl. f. allg.
Path. u. path. Anat. i. 404. Kruse, Beitr. z. path. Anat. u. z. allg.
Path. xii. 221. Bollinger, Miinchen. med. Wchnschr. (1889), No. 37 ;
Vcrhandl. d. Gesellsch. deutsche. Naturf. u. Aertze (1890), ii. 187-
Hofmann, Wien. med. Wchnschr. (1894), No. 38. Straus and Wiirtz.
Cong. p. r etude de la tuberculose, Paris, July 1888. Gilbert and Roger,
Mem. Soc. de biol. (1891). Diem, Monatsh. f. prakt. Thierh. iii. 481.
Weyl, Deutsche med. Wchnschr. (1891), 256. Buchner, Centralbl. f.
Bakteriol. u. Parasitenk. xi. 488. Courmont and Dor, Province mtd.
(1890), No. 50. Tizzoni and Centanni, Centralbl. f. Bakteriol. u.
Parasitenk. xi. 82. Ribbert, Deutsche med. Wchnschr. (1892), 353.
Virchow, ibid. (1891), 131. Hunter, Brit. Med. Journ. (1891), July
25. Kiihne, Ztschr. f. Biol. xxix. 1 ; xxx. 221. Krehl, Arch. f. exper.

BIBLIOGRAPHY 657

Path. u. Pharmakol. xxxv. 222. Krehl and Matthes, ibid, xxxvi. 437.
Bang, "La lutte centre la tuberculose en Danemark," Geneva, 1895.
Maniirliano, " Le serum antituberculeux et son antitoxin," Paris, 1896;
Bcrl. klin. Wchnschr. (1896), 409, 437, 773. Nocard, Ann. de I'Inst.
Pasteur, xii. 561. Stockman, Brit. Med. Journ. (1898), ii. 681. Mara-
gliano, ref. Brit. Mcd. Journ., Epitome (1896), i. 63. Baumgarten and
Walz, Centralbl. f. Bakteriol. u. Parasitcnk. xxiii. 587. T. Smith,
Journ. Exper. Mcd. iii. 451. Koch, Brit. Med. Journ. (1901), ii. 189 ;
1'rans. Internal. Congr. of Tuberc., London, 1901. Delepine, Brit.
.]/•••/. Journ. (1901), ii. 1224. Bataillon, Dubard, and Terre (fish tuber-
culosis), Compt. r> /"/. .w. de biol. 1897, 446. Dubard, Rev. de la tubercul.
(1898), 13, 129. Ravenel. Univ. Pennsylvania Med. Bulletin, May 1902.
Koch, Deutsche med. Wchnschr. (1902), No. 48. Koch, Schutz, Neufeld,
and Miessner, Ztschr. f. Hyg. 51, 300. De Jong, Centralbl. f. Baklcrlnl.
u. Parasitcnk. xxxviii. (Orig.), 146. Ravenel, Univ. of Pennsylvania
.}fed. Bulletin, 1902. Kossel, Weber, and Heuss, Tuberkulosearbeiten
ausd. kaiserl. Gsndhtsmte., Berlin, 1904-1905. Weber and Tante, ibid.
1905. Salmon and Smith, "Tuberculosis," U.S. Department of Agri-
culture, Washington, 1904. Wolbach and Ernst, Journ. Med. Research,
x. 313. " Interim Reports of the Royal Commission on Tuberculosis,"
London, 1904, 1907. Wright and Douglas, Proc.Roy. Soc. Lond.,]xxiv.
159. Wright, Clinical Journal, Nov. 9, 1904; ibid., May 15, 1906;
Med. Chir. Tran. (1905) Ixxxix. Wright and Reid, Proc. Hoy. Soc.
Loud., Ixxvii. 194, 211. v. Pirquet, Berlin, klin. Wchnschr. (1907). Vide
also article on " Kutane u. konjunktivale Tuberkulinreaktion," in Kraus
.ind Levaditi's Techniku. Methodik der Immunitdtsforschung, Bd. I. 1035.
Wolf!'- Eisner, Berlin, klin. Wchnschr. (1907). Calmette, Compt. rend.
Acad. d. Sciences (1907), 1324. Calmette, Breton, Painblon et Petit,
J'ri'sse med. (1907), xv. 443. Petit, " Le diagnostic de la tuberculose par
I'ophthalmo-reaction" (full references), Paris, 1908. Much, Beitr. Klin. d.
Tuberc. (1907), viii. 85, 99, 357- Wirths, Miinchen. med. Wchnschr.
(1908), Iv. 1687. Trauholz, New York Med. Rcc. (1908), 60. Herman, Ann.
de I'Inst. Pasteur, xxii. 92. Frugoni, Centralbl. f. Bakteriol. u. Parasitenk.
(Orig.) (1910), liii. 553. Twort, Proc. Roy. Soc. Lond., B. Ixxxi., March
1909.

acid-fast bacilli.— Moeller, Deutsche med. Wchnschr. (1898), 376.
. Bakteriol. u. Parasitenk. xxv. 369; ibid. xxx. 513. Petri,
Arl>. a. d. k. Gsndhtsamlc. (1898), 1. Rabinowitch, Deutsche med.
H'cluwhr. (1897), No. 26; (1900), No. 16; Ztschr. f. Hyg. xxvi. 90.
Koi'ii, Arch. f. Hyg. xxxvi. 57; Centralbl. f. Bakteriol. u. Parasitenk.
\\vii. 481. Schulze, Ztschr. f. Hyg. xxxi. 153. M. Toblcr, ibid, xxxvi.
120. Lubarsch, ibid. xxxi. 'l87. Holscher, Centralbl. f. Bakteriol. u.
1',1,-nsif, 'nJ:. xxix. 425. Potet, " fitude sur les bacilles dites 'acido-
philes,'" Paris, 1902. Abbott and Gildersleeve, Univ. of Pennsylvania
M.il. /;////<•////., June 1902 Johne and Frothingham, Deutsche Ztschr. f.
T/i termed. (1895), 438. McFadyean, Journ. Compar. Path. xx. (1907), 48.
Tliilibert, "Les pseudo-bacilles acido-resistants," Paris, 1908.

CHAPTER XI.— LEPROSY.

Hansen, Norsk. Mag. f. L«gevidensk., 1874; Virtliows Archir, Ixxix.
:\-2 ; xc. 542 ; ciii. 388 ; Virchow's Feslschr. (1892), iii. See papers by
Ni-isser and Cornil and Suchard in " Microparasites in Disease" (New
Soc., 1886). Hansen and Looft, " Leprosy," Bristol, 1895.

658 BIBLIOGRAPHY

Doutrelepont and "VVolters, Arch. f. Dermat. u. Syph. (1892), 55.
Thoma, Sitzungsb. d. Dorpater Naturforsch., 1889. Unna, Dermat.
Stud. Hamburg (1887), iv. Bordoni-Uff'reduzzi, Ztschr. f. Hyg. iii.
178 ; Berl. klin. Wchnschr. (1885), No. 11. Avning and Nonne,
Virchow's Archiv, cxxxiv. 319. Gairdner, Brit. Med. Journ. (1887),
i. 1296. Hutchinson, Arch. Surg. (1889), i. v. Torok, "Summary of
Discussion on Leprosy at the first Internat. Congr. for Dermatol. and
Spyh." v. Moiiatsh. f. prakt. Dermat. ix. 238. Profeta, Gior. intcrnaz.
d. sc. med., 1889. See Journal of the Leprosy Investigation Committee,
1890-91. Philip pson, Virchow's Archiv, cxxxii. 529. Daniclssen,
Monatsh. f. prakt. Dermat. (1891), 85, 142. Wesener, CcntralbL f.
BakterioL u. Parasitenk. ii. 450 ; Miinchen. mcd. Wchnschr. (1887),
No. 18. Uhleuhuth and Westphal, Centralbl. f. BakterioL u. Parasitenk.
xxix. 233. Babes in " Erganzungsband " of Kolle and Wassermann's
Handbuch dcr Pathogenen Mikro-organismen. Dean, Journ. of Hyg. v.
99. Wherry, Journ. Infect. Diseases (1908), v. 507. Kitasatoj Ztschr. f.
Hyg. (1909), Ixiii. 507. March oux and Bourret, Ann. de f'lnst. Pasteur
(1909), xxiii. 513. Clegg, Philippine Jour. Sc., Series B. iv. (1909).
Slatineano and Danielopolu, Compt. rend. Soc. biol. (1908), Ixv. 347 ;
(1909), Ixvi. 332.

CHAPTER XII.— GLANDERS AND RHINOSCLEROMA.

Loffler and Schultz, Deutsche med. Wchnschr. (1882), No. 52. Loftier,
Mitth. a. d. k. Gsndhtsamte. i. 134. Weichselbaum, Wien. med.
Wchnschr. (1885), Nos. 21-24. Preusse, Berl. thierarztl. Wchnschr.
(1889), Nos. 3, 5, 11 ; ibid. (1894), Nos. 39, 51. Gamaleia, Ann. de
Vlnst. Pasteur, iv. 103. A. Babes, Arch, de med. exper. et d'anat. pat/t.
(1892), 450. Straus, Compt. rend. Acad. d. sc. cviii. 530. M'Fadyean
and Woodhead, Hep. National Vet. Assoc., 1888. Baumgarten, Centralbl.
f. BakterioL u. Parasitenk. iii. 379. Silviera, Semainc mtd. (1891),
No. 31. Bonome, Deutsche med. Wchnschr. (1894), 703, 725, 744.
Kalning, Arch. f. Veterindrwissensch. (St. Petersburg), i. Apr. May.
Foth, Centralbl. f. BakterioL u. Parasitenk. xvi. 508, 550. M'Fadyean,
Journ. Comp. Path, and Therap., 1892, 1893, 1894. Leclainche and
Montane, Ann. de Vlnst. Pasteur, vii. 481. Leo, Ztschr. f. Hyg. vii.
505. Marx, Centralbl. f. BakterioL u. Parasitenk. xxv. 275. Mayer,
ibid, xviii. 673. Bonome, Centralbl. f. BakterioL u. Parasitenk. (Reler.),
xxxviii. 97. Anderson, Chalmers, and Buchanan, Glasgow Med. Journ.,
Oct. 1905. Nicolle, Ann. de Vlnst. Pasteur, xx. 623, 698, 801 ; ibid. (1907),
xxi. 281. Schnurer, Centralbl. f. BakterioL u. Parasitenk. (Refer.),
(1909), xlii. : Supplem. 167 ; Ztschr. f. Infektionskrank. d. Hausthiere.
(1908), iv. 216. Collins (agglutination), Journ. Infect. Diseases (1908), v.
401. Miiller, Ztschr. f. Immnnitdtsf. (Orig.) (1909), iii. 401. Valenti,
ibid. (1909), 98.

RHINOSCLEROMA.— Frisch, Wien. med. Wchnschr. (1882), No. 32.
Cornil and Alveraz, Arch, de physiol. norm, et path. (1895), 3rd series,
vi. 11. Paltauf and Eiselsberg, Fortschr. d. med. (1886), Nos. 19, 20.
Wolkowitsch, Centralbl. f. d. med. Wisscnsch. (1886). Dittrich, Ztschr.
f. Heilk. viii. 251. Babes, Centralbl. f. BakterioL u. Parasitenk. ii.
617. Pawlowski, ibid. ix. 742; " Sur 1'etiologie et la pathologic du
rhinosclerome," Berlin, 1891. Paltauf, Wien. med. Wchnschr. (1891),
Nos. 52, 53; (1892), Nos. 1, 2. Wilde, Semaine med. (1896), 336.
Klemperer and Scheier, Ztschr. f. klin. Med. xlv. Heft 1-2. Lanzi,

BIBLIOGRAPHY r>59

Cfitfralbl.f. H.il-trri,,/. a. /Wr/N/7, ,//•. (KdV-r.), \x.\iv. «!27. SehaMowski,
ibid, \\xviii. 714. Perkins, Jo urn. Infect. Diaauct (1907), iv. 51.

CHAPTER XIII.— ACTINUMV. osis, KT« .

llollinger, < '< -uti-nllil. f. d. tncd. IV-isscHsck., 1877. J. Israel, F/n7/«//-'.s-
Ari'hir. Ixxiiv. 15; Ixxviii. 421. Ponfick, Brcxlnn. aertol. Z(*>-/,,:, :
"Die Aktinomykose des Menschen," 1882. 0. Israel, Vircltmr's Arc/iir,
xcvi. 17"'. Chiari, Pray. med. Wchnschr., 1884. Langhans, t'or.-Bl. /'.
sr/. //•/»/;. .lcr-Jc (1888), xviii. Liining and Hanau, ibid. (1889), xix.
Shattook, Trans. Path. Soc. London,, 1885. Aclaud, ibid. 1886. Delepine.
//>/</. ISK'.i. Harley. M.d.-Chir. Trans., London, 1886. Crookshank.,
ibiil. 1889; "Manual of Bacteriology," London, 1896. Ransome, J/"/.-
Chir. Trans., London, 1891. M'Fadyean, Journ. Comp. Path, mxl
Th- r>'i>., 1889. Bostroni. Beitr. t. path. Anat. •//. c. a////. Pa//t., 1890.
Woltt' and Israel, Virclioid's Archiv, cxxvi. 11. Illich. ''Beitrage xur
Klinik der Aktinomykose," Wien, 1892. Grainger Stewart and Muir,
K'f.i. J/ns/t. ];<-p., 1893. Leith, ibid. 1894. Gasperini, Centralbl. f.
Baktfi-lot. >'. I'n I'tixitc nk. xv. 684. Hummel, Beitr. z. klin. Chir. xiii.
No. 3. Pawlowsky and Maksutoff, Ann. dc I'Inst. Pasteur, vii. 544.
Nenkirch, Uebcr Mrahlcnjrilze, Strassburg, 1902. Doepke, M a m-l •••>!.
///"/. 11',-hnsrhr.i 1902. Sillurschmidt, Ztschr. f. Hyy. xxxvii. 345.
.1. Homer Wright, Publications of the Massachusetts General Ho.y/fn/,
liuston, May 1905. Neuhaiiser, Deutsche med. Wchnschr. (1907), 1457.
\Vi)o!ilridge, Jimrn. Compar. Path, and Therap. (1907), xx. Fritzsche,
A >•<•/, iv.f. Hyy. (1908), Ixv. 181.

Allied Mrcptothrices;—'8ocsird, Ann. de I'Inst. Pasteur (1888), ii. 293.
Kppinger, Beitr. z. path. Anat. u. c. allg. Path. ix. 287 ; in Lubarsch and
Ostortag, " Krgebnisse der allgem. Path." iii. 328. Buchholz, Ztschr. f.
////.'/. xxiv. 470. Berestnew, ibid. xxix. 94. Cozzolino, ibid, xxxiii. 36.
Flexner, Journ. E.'-^r. Med. iii. 435. Dean, Trans. Path. Soc. London
(1900), 26. Birt and Leishman, Journ. of Hyg. ii. 120. Mertens,
' '. ,/t, -it/hi, f. Bakteriol. u. Parasitenk. xxix. 694. Foulerton, Trans.
I'ntli. S»c. 'London (1902), 56. M'Donald, Trans. Med. -Chir. Soc. Edin.
xxiii. 131. N orris and Larkins, Journ. Exper. Med. v. 155. Butter-
field, Journ. Infect. Diseases, vi. 421. Litten and Levy, Deutsche med.
//W/,/.sr///\ (1906), 1772.

MADTKA DISKASK. — Carter "On Mycetoma or the Fungus Disease of
India," London. Ba.ssini, ref. in Ccntralbl.f. Bakteriol. u. Parasitenk.
iv. 652. Lewis and Cunningham, Eleventh Ann. Rep. San. Com. India.
K.'.bner, Fortschr. d. Med. (1886), No. 17. Kanthack, Journ. Path, and
Ji'H-trriol. \. 140. Boyee and Surveyor, Proc. Roy. Soe. London, 1893.
Vandyke Carter, Trans. Path. Soc. London, 1886. Vincent, Ann. </<•
/'/nut. Pasteur, viii. 129. J. H. Wright, Journ. Exper. Med. iii. 421.
Oppenheim, Arch. f. Dermat. u. Syph. Jxxi. 209. Bruin pt, " Les
.My '-tomes," Paris, 1906.

CHAPTER XIV.— ANTHRAX.

Bellinger in Ziemssen's " Cycloptedia of Medicine." Greenfield,
"Malignant Pustule/' in Qnaut'a "Dictionary of Medicine," London,
1894. Pollender, I'rtljschr. f. [jcrlchtl. Med. viii. Davaine, Compt.
r, ,i,l. Acad. d. sc. Ivii. 220, 351, 386 ; lix. 393. Koch, Cohn's Beitr. z.
l. d. Pftanz. (1876), ii. Heft 2 ; Mitth. a. d. k. Gsndhtsamtf.. i. 49.

660 BIBLIOGRAPHY

Pasteur, Compt. rend. Acad. d. sc. xci. 86, 455, 531, 697 ; xcii. 209.
Buchner, Virchow's Archiv, xci. Chamberland, Ann. de I'lnst. Pasteur,
viii. 161. Chauveau, Com.pt. rend. Acad. d. sc. xci. 33, 648, 880 ; xcvi.
553. Czaplewski, Beitr. z. path. Anat. u. z. allg. Path. vii. 47.
Gamaleia, Ann. de I'lnst. Pasteur, ii. 517. Marshall Ward, Proc. Roy.
Soc. London, Feb. 1893. Petruschky, Beitr. z. path. Anat. u. z. allg.
Path. iii. 357. Weyl, Ztschr. f. Hyg. xi. 381. Behring, ibid. vi. 117 ;
vii. 171. Osborne, Arch. f. Hyg. xi. 51. Roux, Ann. de I'lnst. Pasteur,
iv. 25. Hankin, Brit. Med. ' Journ. (1889), ii. 810; (1890), ii. 65.
Hankin and Wesbrook, Ann. de I'lnst. Pasteur, vi. 633. Sidney Martin,
Rep. Med. Off. Local Govt. Board (1890-91), 255. Marmier, Ann. de I'lnst.
Pasteur, ix. 533. Rd. Muir, Journ. Path, and Bacterial, v. 374. Sclavo,
Rivista d'Igiene e Sannita pubblica, vii. Nos. 18, 19 ; Sulla stato
presente delict Sieroiherapia anticarbonchiosa. Turin, Pozzo, 1903 (see
Legge, Lancet (1905), i. 689, 765, 841). Sobernheim in Kolle and Wasser-
niann's Handbuch, iv. 793. Cler, Centralbl. f. Bakteriol. u. Para-
sitenk. (Orig.) xl. 241. Bail, ibid, xxxiii. 343, 610. Saufelice, ibid.
xxxiii. 61. Roger and Gamier, Compt. rend. Soc. de liol. Iviii. 863.
Teacher, Lancet (1906), i. 1306. M'Fadyean, Journ. Comp. Path. (1903),
xiv. 35, 360. Cave, ibid. (1908), 320. Heim, Archiv. f. Hyg. xl. 55.
Sobernheim in Kraus and Levaditi's " Handbuch der Technik und
Methodik der Immunitiitsforschung," Jena, 1908, ii.

CHAPTER XV.— TYPHOID FEVER, ETC.

BACILLUS COLT. — Escherich, Centralbl. f. Bakteriol. u. Parasitenk.
(1887), i. 705 ; ibid. (1888), iii. 675, 801 ; Deutsche med. Wchnschr.(\888),
No. 24. Gordon, Journ. Path, and Bacteriol. iv. 438. MacConkey,
Journ. of Hyg. (1905), v. 333 ; (1906), vi. 385 ; (1909), ix. 86. Wilson,
ibid. (1908), viii. 543. Prescott and Winslow, " Elements of Water
Bacteriology," New York, 1908. Voges and Proskauer, Ztschr. f. Hyg.
(1898), xxviii. 20.

EARLY WORK ON B. TYPHOSUS. — Eberth, Virchow's Archiv, Ixxxi. 58 ;
Ixxxiii. 486. Koch, Mitth. a. d. k. Gsndhtsamte. i. 46. Galfky, ibid.
ii. 80. Klebs, Arch. f. exper. Path. u. Pharmakol. xii. 231 ; xiii. 381.
Escherich, Fortschr. d. Med. (1885), Nos. 16, 17. Emmerich, Arch. f.
Hyg. iii. 291. Rodetand Roux, Arch, de 'intd. exper. et d'anat. path. iv.
317. Weisser, Ztschr. f. Hyg. i. 315. Klein, "Micro-organisms and
Disease," London, 1896 ; Rep. Med. Of. Local ffovt. Board (1892-93), 345 ;
(1893-94), 457 ; (1894-95), 399, 407, 411. Babes. Ztschr. f. Hyg. ix.
323. Vincent, Compt. rend. Soc. de biol. se"r. ix. ii. 62. Birch -Hirseh-
feld, Arch. f. Hyg. vii. 341. Buchner, Centralbl. f. Bakteriol. u. Para-
sitenk. iv. 353. Pfuhl, ibid. iv. 769. Petruschky, ibid. vi. 660. Hunter,
Lancet (1901), i. 613. Kitasato, Ztschr. f. Hyg. vii. 515. Chantemesse
and Widal, Bull. med. (1891), No. 82 ; Ann. de TImt. Pasteur, vi. 755 ;
vii. 141. Pere, Ann. de I'lnst. Pasteur, vi. 512. Neisser, Ztschr. f. klin.
Med. xxiii. 93. Nicholle, Ann. de I'lnst. Pasteur, viii. 853. Quincke
and Stiihlen, Berl. klin. Wchnschr. (1894), 351. A. Fraenkel. Centralbl.
f. klin. Med. (1886), No. 10. E. Fraenkel and Simmonds, ibid. (1886), No.
39. Achalrne, Semaine mtd. (1890), No. 27. Grawitz, Charite-Ann.
xvii. 228. Beumer and Peiper, Centralbl. f. klin. Med. (1887), No.
4 ; Ztechr. f. Hyg. i. 489 ; ii. 110, 382. Sirotinin, ibid. i. 465. R.
Pfeiffer and Kolle, Ztschr. f. Hyg. xxi. 203. R. Pfeifter, Deutsche med.
Wchnschr. (1894), 898. Sanarelli, Ann. de I'lnst. Pasteur, vi. 721 ;

BIBLIOGRAPHY 661

viii. 193, 353. Brieger and Fraenkel, Bcrl. klin. Wchnschr. (1890),
241, 268. Siduey Martin, Brit. Med. Journ. (1898), i. 1569, 1644 ; ii.
11, 73. Bokenham, Trans. Path. Soc. London (1898), xlix. 373.
Macta.lv. 'ii. l>roc. Roy. Soc. London, B. Ixxvii. 548. Macfadyen and
Rowland, Centralbl. f. Bakteriol. u. Parasitcnk. (Orig. ) xxxiv. 618,
765. Lorrain Smith and Tennant, Brit. Med. Journ. (1899), i. 193.
< 'astdlani, Ztschr. f. Hyg. u. Infectionakrankh. xl. i.

Ki'iiu-'.MioiAXJY OF TYPHOID.— Forster and Kayser, Munchen. med.
Wchnschr. (1905), 4173. Forster, ibid. (1908), 1. Forster, Discussion at
Unterelsiissischer Artzverein, Deutsche med. Wchnschr. (1907), 85, 1767.
Klinger, Arb. a. d. k. gsndhtsamle. (1906), xxiv. 91. Conradi and
their authors, Klin. Jahrb. (1907), xvii. 115-433 ; ibid. (1909), xxi. 171-
-121. (Typhoid Carriers) Dean, Brit. Med. Journ. (1908), i. 562.
Lcdingluiiii, M. and J. C. G., ibid. i. 15. Sacquepee, Bull, de I' List.
J'ii*ti-nr, viii. 1, 49 (with literature). Browning and Gilmour, Glasgow
.}/,,/. Jnnrn.. (1910), Ixxiv. 81.

IMMUNITY PHENOMENA AND SERUM DIAGNOSIS. — Brieger, Kitasato,
and Wassermanu, Ztschr. f. Hyg. xii. 137. Widal, Semainc m<!d. (1896),
295, 303. Achard, ibid. 295, 303. Griinbaum, Lancet, Sept. 1896.,
IMrj.iue, Brit. Med. Journ. (1897), i. 529, 967 ; Lancet, Dec. 1896.'
Remlinger and Schneider, Ann. de Vlnst. Pasteur, xi. 55, 829. Widal
and Sicard, ibid. xi. 353. Peckham, Journ. Expcr. Med. ii. 549.
Richardson, ibid. iii. 329. Wright and Semple, Brit. Med. Journ. (1897),
i. 256. Wright and Lamb, Lancet (1899), ii. 1727. Wright, ibid.
(1900), i. 150 ; ii. 1556 ; ibid. (1901), i. 609, 858, 1272, 1532 ; ii. 715,
1107 ; ibid. (1902), ii. 651 ; Brit. Med. Journ. (1900), ii. 113 ; ibid.
(1901), i. 645, 771. Wright and Leishman, ibid. (1900), i. 622. See
also discussion at the 0/.in. Soc. London, Brit. Med. Journ. (1901), ii.
1342. Cha'nternesse and Widal, Ann. de I'List. Pasteur, vi. 755.
Christophers, llrit. M><>. Journ. (1898), i. 71. Remy, Ann. de. Vlnst.
Pitstein; xiv. 55fi, 705. Wyatt Johnson, Brit. Med. Journ. (1897), i. 231 ;
Lancet (1897), ii. 1746. Durham, Lancet (1898), i. 154 ; ibid. ii. 446.
(Vaccination treatment of Typhoid Fever) Smallman, Journ. R.A.M.C.
(1909), vii. 136. Leishman, Journ. Roy. Intst. Pub. Health (1910), viii.
385, 513.

PARATYPHOID AND FooD-PoisoNix<; BACILLI. — Boycott, Journ. Hyg.
vi. 33. Gaertner, refs. inde Baumgarten's Jahrcsbericht, iv. 249 ; vii.
297 ; xii. 508. van Ermengem, in Kolle and Wassermann's Hnndbuch,
vol. ii. Conradi, Deutsche Med. Wchnschr. (1904), 1165. Foruet, Arb.
a. i/. /.. gtndktmmie. (1907), xxv. 247. Levy and Gaehtgeus, ibid. xxv.
250. Goehtgeas, Ibid. (1909), xxx. 610. Rimpau, ibid. xxx. 330.
llainhridge, Journ. of Path, and Bacterial. (1909), xiii. 443. Saoqn6pe£,
Hull, de Vlnst. Pasteur (1907), v. 888. Sacquepee and Chevrel, ibid.
49, 97. (Psittacosis) Baumgartcn'.s Jahrcsbericht. xii. 496. (Bacillus
Enteritidis Sporogenes) Klein, Rep. Med. Off. Local Govt. Board, xxv.
171 ; xxvii. 210.

P. A< TI.IMAL DYSENTERY.— Shiga, Centralbl. f. Bakteriol. u. Parasitenk.
xxiii. 599 ; xxiv. 817, 870, 913. Kruse, Deutsche med. Wchnschr.
(1900), 637. Flexner, Bull. Johns Hopkins Hasp. (1900), xi. 39, 231 ;
Hfit. .]/,,/. Jnnrn. (1900), ii. 917. Strong and Musgrave, Journ. Amcr.
Med. Assoc. (1900), xxxv. 498. Vedder and Duval, Journ. Exper. Med.
(1902), vi. 181. Ogata, Centralbl. f. Bakteriol. u. Parasitenk. xi. 264.
See various authors in Studies from the Rockefeller Institute for Medical
ll< search (1904), vol. ii. Park, Collins, and Goodwin, Journ. Med.

662 BIBLIOGRAPHY

Research (1904), xi. 553. Hiss, ibid. (1905), xiii. 1. Torrey, Journ.
fixper. Med. (1905), vii. 365. Weaver, Tuniiiclifle, Heiiiemaiin, and
Michael, Journ. Infect. Diseases, ii. 70. Doerr, Das Dysenterietoxin,
Jena, 1907 ; Kraus and Levaditis' Handbuch (1908), ii. 164. Pane and
Lotte, Centralbl.f. fiaktcriol. n. Para site n.k. (Orig.j (1907), 811. Shiga,
Ztschr. f. Hytj. (1908), Ix. 75. Anmko, ibid. Ix. 85. Franchietti, ibid.
Ix. 127.'

SUMMER DIAUUIKEA. — Morgan, Brit. Med. Journ. (1906), i. 908 ;
(1907), ii. 16. Morgan and Ledingham, Proc. Roy. Soc. Med. (1909),
ii. (2) (Epidcmiological section), 133.

OH A PTEK XVI. — Di r HT 1 1 v. i: i A .

Klebs, Vcrhandl. d. Cong. f. iiim-rc Med. (1883), ii. Loftier, Mitth.
a. d. k. Gsndhtsamte. (1884), 421. Roux and Yersin, Ann. de I'lnst.
Pasteur, ii. 629 ; iii. 273 ; iv. 385. Brieger and Fraenkel, fieri, kiln.
Wchnschr. (1890), 241, 268. Spronck, Cenfralbl. f. ally. Path. u. path.
Anat. i. No. 25 ; iii. No. 1. Welch and Abbott, ^ Johns Hopkin* Hosp.
Bull. 1891. Reining and Wernicke, Ztschr. f. Hyy. xii. 10. Loftier,
Centralbl. f. fiaktcriol. u. Parasitcnk. ii. 105. v. Hofmann. Wicn.
med. Wchnschr. (1888), Nos. 3 and 4. Cobbett and Phillips, Journ.
Path, and Barteriol. iv. 193. Peters, ibid. iv. 181. Wright, 7,Wr,/,
Med. and >S'. Journ. (1894), 329, 357. Kanthack and Stephens, Journ.
Path, and Bacteriol. iv. 45. Klein, Brit. Med. Journ. (1894), ii. 1393 ;
(1895), i. 100; Hep. Med. O/. Local (fort. Board (1890-91 ), 219; (1891-
92), 125. Abbott, Journ. Path, and Bacterial, ii. 35. Guinochet, Com/it,
rend. $oc. de biol. (1892), 480. Roux and Martin, Ann. dr. I'lnst.
Pasteur, viii. 609. Cartwriglit Wood, Lancet (1896), i. 980, 1076 ;
ii. 1145. Sidney Martin, " Goulstonian Lectures," Brit. Med. Journ.
(1892), i. 641, 696, 755 ; Rep. Med. Off'. Loral ffoct. Board ( 1891-92), 147 ;
(1892-93), 427. Escherich, Wicn. 'med. IVchnschr. (1893), Nos. 47-50 ;
Wicn. klin. Wchnschr. (1893), Nos. 7-10; (1894), No. 22; fieri, klin.
Wchnschr. (1893), Nos. 21, 22, 23. Behring, "Die Geschichte der
Diphtherie," Leipzig, 1893; " Abhandlungen /. atiol. Therap. v. anst.
Krankh.," Leipzig, 1893; ">Bckampfung der Infectionskrankheiteji,"
Leipzig, 1894. Ehrlich and Wassermann, Ztschr. f. llyg. xviii. 23i».

Deutsche med. Wchnschr. (1894), 353. Funck, Ztschr. f. //)/</. xvii. 401.
Prochaska, ibid. xxiv. 373. Madaen, ibid. xxiv. 425. Neisser, ibid.
xxiv. 443 ; Hyy. Rundsch. xiii. 705. L. Martin, Ann. dc. Flnst.
Pasteur, xii. 26. Park and Williams, Journ. E;qtcr. Med. i. 164.
Salomonsen and Madsen, ibid. xii. 763. Wood head, Brit. Med. Journ.
(1898), ii. 893 ; Rep. Metrop. Asyl. Bd., London, 1901. Mi- tin, Ann. de
I' List. Pasteur, xii. 596. Madsen, ibid. xiii. 568, 801. Dean and Todd,
Journ. of Hyy. ii. 194. Cobbett, ibid. i. 485. Graham-Smith, ibid. iv.
258 ; vi. 286. Petrie, ibid. v. 134. Hist, Compt. rend. Moc. de biol.
(1903), No. 25. Neisser, fieri, klin. Wchnschr. (1904), No. 11. Knapp,
Journ. Med. Research (1904), 475. Morgenroth, Ztschr. f. Hyg. xlviii.
177. Bolton. Lancet (1905), i. 1117. Theobald Smith, Journ. Med.
Research (1905), xiii. 341. Boycott, Journ. of Hyg. v. 223. Ford
Robertson, Brit. Med. Journ. (1903), ii. 1065, and Rev. of Neurol. and
Psych, vols. i.-iii. Nuttall and Graham-Smith, "The Bacteriology of
Diphtheria" (with full literature, etc.), Cambridge, 1908.

BIBLIOGRAPHY 663

CHAPTER XVII.— TETANUS,

Nicolaicr, " Beitriige zur Aetiologie des Wundstarrkrampfes," Inaug.
Diss. Giittingen, 1885. Rosenbach, Arch. f. kiln. ('/«'>-. xxxiv. 306.
Carle and Rattoiie, Gior. <t. r. Accad. di med. di Torino, 1884. Kitasato,
Ztxch,: f. lhj<j. vii. 225; x. 267; xii. 256. Kitasato an. I Weyl, ibid.
viii. 41, 404. Vaillard, Ann. de I' lust. Pasteur, vi. 224, 676. Vaillard
and Rouget, ibid. vi. 385. Hehring, " Abhandlungcn. z. iitiol. Therap.
v. anst. Krankh.," Leipzig, 1893; Ztschr. f. Hy<j. xii. 1,45; " Blut-
smimtherawe," Leipzig, 1892; " Das Tetanusheilserum," Leipzig, 1892.
Brii-ger and Fraenkel, JJerl. klin. Wchnschr. (1890), 241, 268. Sidney
Martin, lleji. Med. Off. Local Go vt. Board (1893-94), 497 ; (1894-95), 505.
Tsdiinsky, Centralbl. f. Bakteriol. u. Parasitenk. xiv. 316. Tizzoni ;md
Cattani, Arch. f. cxj>cr. Path. n. Pharmakol. xxvii. 432; Centralbl. f.
Balctcriol. u. Parasitenk. ix. 189, 685 ; x. 33, 576 (Ref.) ; xi. 325 ; Ji,-i-/.
kiln. U'chiixrhr. (1894), 732. Madsen, Ztschr. f. Hyg. xxxii. 214. Ritchie,
./'>n ni. of Jfit'f. i. 125. Danysz, Ann. de Vlnst. Pasteur, xiii. 155.
Marie and Morax, Ann. dc Vlnst. Pasteur, Paris, xvi. 818. Meyer and
Ransom, Proc. Roy. Soc. London, Ixxii. 26 ; Arch. /. cxper. Path. n.
riinrmukol., Leipzig, xlix. 269. Roux and Borrel, Ann. dc Vlnst. Pasteur,
Paris, xii. 225. Henderson Smith, Journ. Hyy. vii. 205. Kitt, see ref. in
('••nh-albl. f. Bakteriol. u. Parasitenk., Referate, xxxii. 359. Eisler and
I'ribram in Krans and Levaditi's Hantlbuch, i. 103.

MALK;NANT(KIH..MA.— Pasteur, Bull. Acad. denied., 1881, 1887. Koch,
Mitt/i. ". 'I. /.-. ttxmUiisamtc. i. 54. Kitt, -A////r.s/>. '/. /,-. <'<'nlr.- 7'/n'< ,-« /-./"<-
s,-h>'/> In Mihich'',!, 1883-84. W. R. Hesse, Deutsche med. Wchnschr.
(1885), No. 14. Chauveau and Arloing, Arch. vtt. (1884), 366, 817.
Liborius, Zfsrit,: f. //////. i. 115. Roux and Chamberland, Ann,, tie Vlnst.
, i. ;")t;-J. Charriii and Roger, Couijrt. rent/. Soc. dc bio/. (1877), st'-r.

viii. vol. iv. ]>. 408. Kerry and S. Fraenkel, Ztschr. /. Ify</. xii. 204.
Sanfelice, -ibid. xiv. 339. Leelainche and Velle, Ann. dc Vlnst. Pasteur,
xiv. 202, 590.

BACILM-S PxnTLixrs.— v. Ermengeni, Ccntralbl. f. Bakteriol. ».
Pui-iixiteiik. xix. 443; Ztschr. f. Hyy. xxvi. 1. Kempner, ibid. xxvi. 481.
Kcnipner and Schepilewsky. ibid, xxvii. 213. Kempner and Pollack,
It.-,!/*-/,,- med. Wchiischr. (1897), No. 32. Brieger and Keinpner, ibid.
(1897), No. 33. Marinesco, Compt. rend. Soc. dc biol. (1896), No. 31.
Schneidemiihl, Centralbl. f. Bakteriol. u. Parasitenk. xxiv. 577, 619.
R.'inrr. ibid, xxvii. 857. Madsen in Kraus and LevinlitVsHaiidbtrch, i. 137 ;
ii. 134. Leuchs, Ztschr. f. Hijy. u. Infektionskrankh. (1910), Ixv. 55.

QuAUTKi:-K\ IL.— See Nocard and Leclainche, "Les maladies micro-
l>ieniics dcs aninianx," Paris, 1896. Arloing Cornevin, et Thomas,
" Le charbon syniptomatiqne du brenf," Paris, 1887. Nocard and Roux,
./////. ill- r/nsf. PasUur, i. 256. Roux, ibid. ii. 49. See also Joi'/-».
<'»,,ir. /'nt/i. u, ,,l Therap. iii. 253, 346 ; viii. 166, 233. Grassberger and
Schattentroh in Kraus and Levaditi's Handbuch, i. 161 ; ii. 186. Eism-
berg, Comp. rend. Soc. dc biol. No. 62, 491, 537, 613.

BACILLUS AEKOGRKES CAPSULATUS. — Welch and Nuttall, Bull. Johns
Hopkins Il»*i>. (1892), 81. Welch and Flexner, Journ. Expcr. Med. i.
5. E. Fraenkel, Centralbl. f. Balteriol. n. Parasitenk. xiii. 13. Durham,
Hulf. Johns Hopkins Hosp. (1897), 68. Norris, Am. Journ. Med.
cxvii. 172.

I'l siFOKM BACILLI.— Babes in Kolle and Wassj-rmann's Handbuch,
Erg;inz-Bd. i. 271. Vincent, Ann. dc l'J,>sf. p.isteur (1896), x. 492-;

664 BIBLIOGRAPHY

(1899), xiii. 609. Areillon and Zuber, Arch, de med. exper. (1898), x.
517. Bernheim, Centralbl. f. Bakteriol. u. Parasitenk. (1898), xxiii. 171.
Plant, Deutsche, med. Wchnschr. (1904), 920. Beitzke, Centralbl. f.
Bakteriol. u. Parasitenk. (Ref.) (1904), xxxv. 1. Ellermann, ibid. (Orig.)
(1904), xxxvii. 729 ; xxxviii. 383 ; Ztschr. f. Hyg. (1907), Ivi. 453.
Veszpremi, Centralbl. f. Bakteriol. u. Parasitenk. (Orig.) xxxviii. 136.
Weaver and Tunnicliffe, Journ. Infect. Diseases, (1905), ii. 446 ; (1906),
iii. 190. Blumer and MacFarlane, Amer. Journ. Med. Sc. clxl. 122.

CHAPTER XVIII.— CHOLERA.

Koch, Rep. of 1st Cholera Conference, 1884 (v. " Microparasites in
Disease, "New Sydenham 8oc., 1886). Nikuti and Rietsch, Compt. rend.
Acad. d. sc. xcix. 928, 1145. Bosk, Ann. de I'lnst. Pasteur, ix. 507. Petten-
kofer, Munchen. med. Wchnschr. (1892), No. 46 ; (1894), No. 10. Sawts-
clienko, Centralbl. f. Bakteriol. u. Parasitenk. xii. 893. Pfeiffer, Ztschr. f.
Hyg.xi.393. Kolle.^'oJ. xvi. 329. Issaeff and Kolle, ibid, xviii. 17. Wasser-
mann, ibid. xiv. 35. Soberuheini, ibid. xiv. 485. Metchnikoff, Ann. de
I'lnst. Pasteur, vii. 403, 562 ; viii. 257, 529. Fraenkel and Sobefnheim,
Hyg. Rundschau, iv. 97. Dunbar, Arb. a. d. k. Gmdhtsamte. ix. 379.
Pf'eiffer and Wasserman, Ztschr. f. Hyg. xiv. 46. Wesbrook, Ann. de
I'lnst. Pasteur, viii. 318. Scholl, Berl. klin. Wchnschr. (1890), No. 41.
Griiber and Wiener, Arch. f. Hyg. xv. 241. Cunningham, Sclent. M em.
Med. Off. India, 1890 and 1894. Hueppe, Deutsche med. Wchnschr.
(1889), No. 33. Klemperer, ibid. (1894), 435 ; Berl. klin. Wchnschr.
(1892), 969. Lazarus, ibid. (1892), 1071. Reincke, Deutsche med. Wchnschr.
(1894), 795. Koch, Ztschr. f. Hyg. xiv. 319. Voges, Centralbl. f.
Bakteriol. u. Parasitenk. xv. 453. Pastana and Bettencourt, Centralbl.
f. Bakteriol. u. Parasitenk. xvi. 401. Dieudonne, ibid. xiv. 323. Celli
and Santori, ibid. xv. 289. Neisser, ibid. xiv. 666. Sanarelli, Ann. de
I'lnst. Pasteur, vii. 693. Ivanoff, Ztschr. f. Hyg. xv. 485. Issaeff, ibid.
xvi. 286. Pfuhl, ibid. x. 510. Rurnpel, Deutsche med. Wchnschr. (1893),
160. Klein, Rep. Med. Off. Local Govt. Board, 1893: "Micro-organisms
and Disease," London, 1896. Haffkine, Brit. Med. Journ. (1895), ii.
1541; Indian Med. Gaz. (1895), No. 1 ; "Anti-cholera Inoculation,"
Rep. San. Com. India, Calcutta, 1895. Pfeiffer in Fliigge, "Die Micro-
organismen," 3rd ed. 1896. Gamaleia, Ann. de I'lnst. Pasteur, ii. 482,
552. Archard and Bensande, Semaine med. (1897), 151. Rumpf, "Die
Cholera Asiatica und Nostras," Jena, 1898. Kraus and Pribram, Centralbl.
f. Bakteriol. xli. (Orig.), 15, 155. Kraus and Prantschoff, ibid. 377, 480.
A. Macfadyen, ibid. xlii. (Orig.), 365. Gotschlich, Scientific Reps.
Sanitary, Maritime, and Quarantine Council of Egypt, Alexandria, 1905,
1906. For discussion, vide Supplements to Centralbl. f. Bakteriol. (Ref. ),
(1906), xxxviii. 84 ; and (1908), xlii. 1. Dunbar, Berlin, klin. Wchnschr.
(1902), No. 39. Kraus, in Kraus and Levaditi's " Handbuch der
Immunitatsforschung (with literature on toxins and anti-toxins). For
Russian epidemic, vide Centralbl. f. Bakteriol. u. Parasitenk. (Ref.)
(1909), xliv. 1 et seq. ; Dieudonne, Centralbl. f. Bakteriol. u. Parasitenk.
(Orig.)l. 107.

CHAPTER XIX. — INFLUENZA, ETC.

INFLUENZA. — Pfeiffer, Kitasato, and Canon, Deutsche med. Wchnschr.
xviii. 28, and Brit. Med. Journ. (1892), i. 128. Babes, Deutsche med.

BIBLIOGRAPHY 665

r. xviii. 113. Pfeiffer and Beck, ibid. (1892), 465. Pfuhl,
C<-,,tr«lbl. f. Bakteriol. n. Parasitenk. xi. 397. Klein, Rep. Med. Off.
/,<«•»// <;,,,-/. /,w.>-^(1893), 85. Pfeiffer, Ztschr. f. Hyg. xiii. 357. Huber,
yjwh,: f. ////</. xv. 454. Kruse, Deutsche med. Wchnschr. (1894), 513.
Pelicke, BerL kiln. Wchnschr. (1894), 524. Pfuhl and Walter, Devtsclt,-
ini'il. Wchnschr. (1896), 82, 105. Cantani, Ztschr. f. Hyg. xxiii. 265.
I'lulil, Ztschr. f. Hyg. xxvi. 112. Wasserniann, Deutsche med. Wchnwltr.
(1900), No. 28. Clemens, Miinchen. med. Wchmchr. (1900), No. -11.
\\\ n., ..op, Journ. Med. Ass., February 1903. Neisser, Deutsche med.
//V////.sr/i/\ (1903), No. 26. Auerbach, Ztschr. f. Hyg. (1904), xlviii. 259.
ini, Centralbl.f. Bakteriol. u. Parasitenk, (Orig.) (1907), xliii. 407.
r, Ibid. (Orig. ) (1909), 1. 503.

Nu-CouGH. — Jochmann, Arch. f. klin. Med. Ixxxiv. 470.
and Krause, Ztschr. f. Hyg. (1901), xxxvi. 193. Sperigler,
Deutsche med. Wchnschr. (1897), 830. Davis, Journ. Infect. Diseases, iii. 1.
Bordet and Gengou, Ann. de I'lnst. Pasteur, xx. 731 ; xxi. 720 ;
< ',„(,•'• H./.f. Bakteriol. u. Parasitenk. (1909) (Ref. ), xliii. 273. Bordet, Bull,
de 1'Acad. Jioy. de Medicine de Belgique (1908), 4th ser. tome xxii. 729.
Arnhuini, Berlin, klin. Wchnschr. (1908), 1453. Fraenkel, Miinchen. med.
Wchnschr. (1908), 1683. Klimenko, Centralbl.f. Bakteriol. u. Parasitenk..
(Orig.) xlviii! 64. Wollstein, Journ. Exper. Med. (1909), xi. 41.

PLAGUE. — Kitasato, Lancet (1894), ii. 428. Yersin, Ann. de I'lnst.
J'nsti'in; viii. 6»!2. Lowson, Lancet (1895), ii. 199. Yersin, Calmctte,
:ind l>orrel, Ann. de I'lnst. Pasteur, ix. 589. Aoyama, Centralbl. f.
Hnklr.riol. u. Parasitenk. xix. 481. Zettnow, Ztschr. f. Hjig. xxi. 164.
Yersin, Ann. de I'lnst. Pasteur, xi. 81. Gordon, Lancet (1899), i. 688.
Simond, Ann. de VInst. Pasteur, xii. 625. Haflkine, Brit. Med. Journ.
(1S97), i. 424. Wyssokowitz and Zabolotny, Ann. de I'lnst. Pasteur,
xi. 663. Ogata, Centralbl.f. Bakteriol. u. Parasitenk. xxi. 769. Childe,
llfit. Med. Journ. (1898), ii. 858. See also Brit. Med. Journ. and
Lancet, 1897-99. Ltistigand Galeotti, Deutsche med. Wchnschr. (1897),
No. 15. Markl, Centralbl.f. Bakteriol. u. Parasitenk. xxiv. 641, 728;
xxix. 810. Cairns, Lancet (1901), i. 1746. Montenegro, "Bubonic
Pl.-igue," London, 1900. Netter, " La peste et son bacillc," Paris, 1900.
Mitth. d«T Deutschen Pest-Kommission, Deutsche med. Wchnschr. (1897),
Nos. 17, 19, 31, 32. "Report of the India Plague Commission (1898-
99)," London, 1900-1901. Also numerous papers in the Lancet and Brit.
M><l. Jowr*.t 1897-1901. Regarding Glasgow epidemic, see ibid. (1900),
ii. "Reports on Plague Investigations in India," Journ. Hyg. (1906),
vi. 422; (1907), vii. 323; (1908), viii. 162. Lamb, "The Etiology
;in«l Epidemiology of Plague," Calcutta, 1908. Lfston, Report Bombay
J!»<-/. /,////. (1908), ii.

MALTV KKVKK -Bruce, Practitioner, xxxix. 160; xl. 241; Ann. de
/'/us/. r«*t',ir, vii. 291. Bruce, Hughes, and Westcott, Brit. Med.
Journ. (1887), ii. 58. Hughes, Ann. de. I'lnst. Pasteur, vii. 628 ; Lancet
(1892), ii. 1265. Wright and Semple, Brit. Med. Journ. (1897), i. 1214.
Wright and Smith, ibid. (1897), i. 911 ; Lancet (1897), i. 656. Welch,
ihiil. (1897), i. 1512. Gordon, ibid. (1899), i. 688. Durham, Journ.
/'iitli. a ad Bacteriol. v. 377. Bruce in Davidson's " Hygiene and
Diseases of Warm Climates," Edinburgh and London, 1893. Birt and
Lamb, Lancet (1899), ii. 701. Brunner, Wien. klin. Wchnschr. (1900),
No. 7. Bruce, Journ. Roy. Army Med. Corps (1904), ii. 487, 731 ;
(1907), viii. 225. Horrocks, Proc. Roy. Soc. London, Series B (1905),
Ixxvi. 510. "Reports of the Commission on Mediterranean Fever,"

666 BIBLIOGRAPHY

1904-1907 (reprinted in Journ. Roy. Army Med. Corps,}. Eyre in Kolle
and Wassermann's Handbuch d. Patho<j. Mikro-orr/anismen, Eryaiizunys-
band, 1906. Milroy, "Lectures on Militensis Septicaimia," Lancet (1908),
i. 1677, et seq. Sergent, Gillot, et Lemaire, Ann. de Vlnst. Pasteur, xxii.
209. Siere, ibid. xxii. 616.

CHAPTER XX.— DISEASES DUE TO SPIROCILKTES.

RELAPSING FEVERS. — Obermeier, Centralbl. f. d. med. Wissensch.
(1873), 145; and Berl. klin. Wcknschr. (1873), No. 35. Munch,
Centralbl. f. d. med. Wissensch., 1876. Kocli, Deutsche med. Wchnschr.
(1879), 327. Moczutkowsky, Deutsches Arch. f. klin. Med. xxiv. 192.
Vandyke Carter, Med.-Chir. Trans., London (1880), 78. Lubinoff,
Virchow's Archiv, xcviii. 160. Metchnikolf, ibid. cix. 176. Soudake-
witch, Ann. dc I Inst. Pasteur, v. 545. Lamb, Sclent. Mem. Med. Off.
India (1901), pt. xii. 77. Sawtschenko and Melkich, Ann. de VInxt.
Pasteur, xv. 497. Tictin, Centralbl. f. Bakteriol. xxi. 179. Karlinski,
Centralbl. f. Baktcriol. (1902) (Orig.) xxxi. 566. Gabritschewsky, Ztschr.
f. klin. Med. (1905), Bd. 56. Norris, Pappenheimer, Flournoy, Journ.
Infect. Diseases, iii. 266. Novyand Knapp, ibid. 291. Zettnow, Ztschr. f.
Hyg. (1906), Hi. 485; Deutsche med. Wchnschr., 1906. Maiiteufel, Arl.
a. d. k. Gsndhtsamte. xxix. 337. Shellack, ibid. xxx. 351. Novy,
Journ. Amer. Med. Assoc xlvii. 215. Mackie, Lancet (1907), ii. 832 ;
Brit. Med. Journ. (1907), ii. 1706 ; New York Med. Jonni.., Aug. 22,
1908. Strong, Philippine Journ. Med. Sc. iv. 187.

AFRICAN TICK FEVER. — Ross and Milne, Brit. Med. Journ. (1904),
ii. 1453. Dutton and Todd, Thompson- Yules Laboratory Rep. (1905), vi.
pt. ii. Koch, Deutsche mr.d. JTchnschr., 1905; Berl. klin. Wchnschr.,
1906. Hodges and Ross, Brit. Med. Journ. (1905), i. 713. Breuil and
Kinghorn, ibid. i. 668. Bivuil, Lancet (1906), i. 1806. Levaditi, Comjtt.
Acad. Sc. (1906), tome 142, 1099. Leishman, Journ. R.A.M.C. (1909),
xii. 123. Levaditi and Manom'lian, Ann. dc Vlnst. Pasteur (1907),
xxi. 205.

SYPHILIS. — Lustgarten, Wicn. med. Wchnschr. (1884), No. 47.
Sabouraud, Ann. de Vlnst. Pasteur, vi. 184. Golas/,, Journ. d. maJ.
cutan. et syph. (1894), 170. Markuse, Vrtljschr. f. Jkrmat. u. Sijph.
(1883), No. 3. van N lessen, Centralbl. f. Bakteriol. n. Parasitcnk.
xxiii. 49. Metchnikoff and Roux, Ann. de V Inst. Pasteur, xvii.-xix.
Lassar, Berl. klin. Wchnschr. (1903), 1189. Neisser, Deutsche med.
Wchnschr. (1904), 1369, 1431. Schaudiun and Hoffmann, Arb. a. d.
k. Gsndhtsamte. (1905), Bd. 22 ; Deutsche med. Wchnxclir. (1905),
No. 18; Berl. klin. Wchnschr. (1905), Nos. 22, 23. Schaudinn,
Deutsche med. Wchnschr. (1905), No. 22. Hoffmann, Berl. klin.
Wchnschr. (1905), No. 46. "Selected Essays on Syphilis and Smallpox,"
New Sydenham Society, 1906. Metchnikoff, La Semaine med. (1905),
234. Levaditi, ibid. (1905), 247. Siegel, Miinchen. med. Wchnschr.
(1905), 1321, 1384. Herxheimer, ibid. (1905), 1857. Shennan, Lancet
(1906), i. 6b3, 746. Maclennan, Brit. Med. Journ. (1906), i. 1090.
Levaditi, Ann. de I'Inst. Pasteur (1906), xx. 41. Levaditi and M'Intosh,
ibid. (1907), 784. Levaditi and Yamanouchi, ibid. (1908), 763. Hoff-
mann, "Die Atiologie der Syphilis," Berlin, 1906. Neisser, "Die
experimentelle Syphilisforschung," Berlin, 1906.

SERUM DIAGNOSIS. — "Wassermann, Neisser, and Bruck, Deutsche med.
Wchnschr; (1906), 745. Wassermaun and Plant, ibid. 1769. Wassermann,

BIBLIOGRAPHY 067

Wien. kit a-, ll'i-liiiwhr. (1907), 745. Marie and Levaditi, Ann. </<• /'///*/.
Patteur (1H07), xxi. 138; (frm/it. rend. Soc. dc biol. (1907), Ixii. 872.
POINTS iiii<l Meier, forl. Win. IVfhnncIn'. (1907), 1655, and (1908), 731.
Sachs and Altmaiin, /W. (1908), 494, 699. Sachs and Rondoni, Ztscl,,. /.
II,H,LI> nitii/sf. i. 132. M'Kenzie, Journ. 1'iifli. «n<l Jlactcriol. (1909), xiii.
311. IJrowniug, Cruickshank, and McKenzie, ibid. (11)10), xiv. 484. Plaut,
" Wnssennann's Serodiagnostik der Syphilis in ihrer Anwendung auf die
Ps\ -i -lii.-iti if." .Irna, 1909. Landsteiner, CcntrnUil. /'. AV//-/C/W. /'. Para-
ntenk. (Kef.) (190S), xli. 785. Stern, Ztschr. /.* Immunitiitsf. i. 422.
Nnguchi, Compt. rend. &»: d,- Hoi. (1909), Ixvi. No. 11. See also dis-
cussion in /,'/•/'/. M'li. Jin/rii. 1910, ii.

Fi; AM mi > A UK V.v\vs. — Castellani, /,'///. M<il. Jin/rn. (1905), ii. 282,
1280, l:i:'.U: Journ. 7/////. (1907), r»"»s. Ncisscr, liaemiann, and Halber-
st:i(itrr, Miim-h'-n. wed. ll'i-finrr/ir. (1906), 1337. Halberstadter, Jr/>.
f. if. /.-. (;*ni/h/*ttnt./c. (1907), xxvi. 48. Levaditi and Nattan-Larrier,
Ann. •!'• r l.ixt. /'.isb-iir (1908), xxii. 260. Shennan, Journ. 1'olh. «,nl
/;<><•/, ,•/<>/. (190S), xii. 426. Ashburn and Craijr, J'/ii///>/n'iif Jnuni. Mcd.
Sc. l!»i)7): ii. 111. ShiiiriiiT, Mii,i.'h<-n.. Died. Hrc/tnschr. (1907), 1364.

rilAPTKK XXI. IM.MI-MTV.

Fur early inoculation methods (e.ij. against anthrax, chicken cholera,
etc.). >ci " MicrojiaraMtcs in Disease," New Syd. Soc. 1886. Duguid
and Sanderson. ./'//'/•//. /to//. J<//-i'c. Site. (1880), 267. Orecniicld, U>!<I.
(1880), 273 ; I'm,-. /,'/,//. »/,-. London, June 1880. Toussaint, Comjif.
rend. Aead, <L te. xoi. !:•:•. liatl'kine, Brit. Mcd. ./»///•/*. (1891), ii.
127*. Klein, //////. (l,S!»:j), i. (332, tS3S», b~>\. Klenipcrer, Arch. f. ('.••/» r.
/'"'It. ii. rinn-iiKiknl. \\xi. :!.")(;. IJiichncr. Mi< iirln'n. n«<<l. ITc/msc/tr.
(1898), M1.'. isO. Khrlich, Dmi^l,,- med. Wchnachr. (1891), 976, 1218.
K. IT. iilri, Ztvhr. /. ////;/. xviii. 1 ; xx. 198. Pfeiffcr and Kolle, ibid.
xxi. 203. Uonk't, Ami. <!<• /'/n*f. Pasteur, ix. 462; xi. 106. Metchni-
kotf, Vin-hnir* Ai-chlr, xcvi. 177; xcvii. 502; cvii. 209; <-ix. 176;
Ann. <!>• /7,/s/. /'<>x/cur, iii. 289 ; iv. 65; iv. 193; iv. 493; v. 465 ; vi.
289 : \ ii. iirj ; vii. 562 ; viii. 257 ; viii. 529 ; ix. 433. Calmette, Ann.
'/>- I' lust. /'<iist.-iir, viii. 275 : xi. 95. Fraser, Proc. Hoy. ,S'w. Kdin. xx.
11^. Marmorek, Ami: dc I'Jnuf. J'astt ur, ix. 593. Metchnikoff, Koux,
ami Taun-lli-Salinibeni, ibid. x. 257. Charrin and Roger, Compt. rend.
Si,,-, ilr hint. (1887), 667. Griibcr and Durham, Miiiichcn. mfd. Wehntekf.
(1896), March. Dm ham, Jourit. Path, mid Jlactrritil. iv. 13. Cart-
wright Wood; Liini-i't. (1896), i. 980; ii. 1145. Sidney Martin, "Serum
Treatment of Diphtheria," Lancet (1896), ii. 1060. Kansome, "On
Immunity t«> DlMMe," London, 1896. Burdoii Sanderson, " Croonian
Lectures." Jlrit. M"'. J.mrn. (1891), ii. 983, 1033, 1083, 1135. Dis.-u>-
sion on Immunity, Path. Soc. London, Brit. Mcd. Jmtrn. (1892;, i. 373.
Fodoi. /1,-nfK-l,, med. ll'chnschr. (1887), No. 34. Hu.-ppi-. Jlcrl. klin.
//V////.sr///-. (1892), No. 17. Nicholle, Ann. dc VJnst. Pasteur, xii. 161.
Salonioiisen and Madsen, ib'nl. xi. 315 ; xii. 763. Roux and Horrell, ibid.
xii. 22.">. Salimlieni, Hid. xi. 277. Wassermann and Takaki, Bcrl.
/.-////. W<-l, ,i*<'t, i\ (1898), xxxv. 4. Blumenthal, Deutsche Died. Wchnschr.
xxiv. 185. Ransom, ibid. xxiv. 117. Meade Bolton, ,/-////•/*. Exper.
M"i. i. MS, T. R. Fraser, Brit. Med. Jouni. (1895), i. 1309; ii. 415.
416 ; (1896), i. 957 ; (1896), ii. 910 ; (1897), ii. 125, 595. Calmette,
Ann. de I'lnst. Pasteur, vi. 160, 604 ; viii. 275 ; ix. 225 ; x. 675 ; xi.
21 1 : xii. 343. C. J. Martin, Journ. Physiol. xx. 364; Proc. Roy. Soc.

668 BIBLIOGRAPHY

London, Ixiv. 88. C. J. Martin and Cherry, ibid. Ixiii. 420. Gautier,
" Les Toxines microbiennes et animales," Paris, 1896. Wassermann,
Berl. klin. Wchnschr. (1898), 1209. Pfeiffer and Marx, Ztschr. f. Hyg.
xxvii. 272. Bordet, Ann. de I'lnst. Pasteur, xii. 688. Ehrlich, Deutsche
med. Wchnschr. (1898), xxiv. 597. "Die Wertbemessung des Diph-
therieheilserums," Jena, 1897. Croonian Lecture, Proc. Roy. Soc.
London, Ixvi. 424; Deutsche med. Wchnschr. xxvii. (1901), 866, 888,
913. Nothnagel's *' Specie! le Pathologic und Therapie," Bd. viii.
Schlussbetrachtnngen. Ehrlich and Morgenroth, Berl. klin. Wchnschr.
(1899), xxxvi. 6, 481 ; (1900), xxxvii. 453, 681 ; (1901), xxxviii. 251,
569, 598. Weigert, in Lubarsch and Ostertag, "Ergebnisse der
Allgemeinen Pathologic" (1897); iv. Jahrg. (Wiesbaden, 1899). Morgen-
roth, Centralbl.f. Bakteriol. u. Parasitenk. xxvi. 349. Bulloch, Trans.
Jenner Inst. 2nd ser. p. 46. Dimitz, Deutsche med. Wchnschr. (1897),
xxiii. 428. Bordet, Ann. de VInst. Pasteur, xii. 688 ; xiii. 225, 273 ;
xiv. 257 ; xv. 303 ; xvii. 161 ; xviii. 593 ; Metchnikoff, ibid. xiii. 737 ;
xiv. 369 ; xv. 865. Gengou, ibid. xv. 232. Sawtschenko, ibid. xvi. 106.
Ritchie, Journ. of Hyg. ii. 215, 251, 452 (with full references).
"General Pathology of Infection," in Clifford Allbutt's "System of
Medicine," 2nd ed. 1906, vol. ii. pt. i. p. 1. Neisser and
Wechsberg, Manchen. med. Wchnschr. (1901), No. 18. von Dungern,
ibid. (1899), 1288 ; (1900), 677, 973. Joos, Ztschr. f. Hyg. xxxvi.
422; xl. 203; Centralbl.f. Bakteriol. (Orig.), xxxiii. 762. Eisen-
berg and Volk, Ztschr. f. Hyg. xl. 155. Dreyer and Jex-Blake, Journ.
PatJi. and Bacteriol. xi. 1. (Precipitins) Welsh and Chapman, Proc.
Hoy. Soc. London, B., Ixxix. (1907), 465 ; Journ. Path, and Bacteriol.
(1909), xiii. 206. Kraus, Wien. klin. Wchnschr. (1907), No. 32. See
also the articles on precipitins by Uhlenhuth and Weidanz, on Bacterial
Precipitins by v. Eisler, and on Agglutinins by Volk, in Kraus and
Levaditi's "Handbuch." Nuttall, "Blood Immunity and Blood Rela-
tionship," Cambridge, 1904.

OPSONINS. — Denys and Leclef, "La cellule," 1895, 177. Sawtschenko,
Ann. de V Inst. Pasteur (1902), 106. Wright and Douglas, Proc. Roy.
Soc. London, Ixxii. 357 ; Ixxiii. 128 ; Ixxiv. 147. Wright and Reid, ibid.
Ixxvii. 211. Bulloch and Atkin, ibid. Ixxiv. 379. Dean, ibid. Ixxvi.
506 ; Brit. Med. Journ. (1907), ii. 1409. Discussion in Centralbl. f.
Bakteriol. u. Parasitenk. Referate xliv. Supplement 14.* Bulloch and
Western, Proc. Roy. Soc. London, Ixxvii. 531. Neufeld and Rimpau,
Deutsche med. Wchnschr. (1904), 1458. Neufeld, Berl. klin. Wchnschr.
(1908), No. 21 : Med. Klinik. (1908), No. 19. Hektoen and Ruediger,
Journ. Infect. Diseases (1905), 128. Hektoen, ibid. (1908), 259 ; (1909),
78. Leishman, Trans. Path. Soc. Lond., 1905. Muir and Martin, Brit.
Med. Journ. (1906), ii. ; Proc. Roy. Soc. London, B., Ixxix. 187.
Fornet and Porter, Centralbl.f. Bakteriol. u. Parasitenk. (Orig.) (1908),
xlviii. 461.

The following works dealing with the subject of Immunity have been
published within recent years: — Metchnikoff, " Immunity in Infective
Diseases" (Engl. TransL), Cambridge, 1905 ; Ehrlich, "Studies in Im-
munity" (Engl. Transl.), 2nd ed., New York, 1909 ; Bordet, "Studies in
Immunity," New York, 1909; Kraus and Levaditi, "Handbuch der
Technik und Methodik der Immunitatsforschung," Jena, 1908 ; Wright,
" Studies on Immunisation," London, 1909. D'Este Emery, " Immunity
and Specific Therapy," London, 1909; Muir, "Studies on Immunity,"
London, 1909; Wolff-Eisner, " Klinische Immunitiitslehre und Serodiag-

BIBLIOGRAPHY 669

nostik," Jena, 1910. The most important papers dealing with current
work on the subject are published in the Zeitschrift fur Immunitdts-
forschung.

ANAPHYLAXIS. — Richet, Compt. rend. Soc. de biol., 1903-5, Ann. de
r/nxt.. I'«xt..',ir (1907), xxi. 497 ; (1908), xxii. 465. Arthus, Compt. rend.
Soc. de biol. (1903), Iv. 817. Arthus and Breton, ibid. Iv. 1478. Th.
Smith, Discussion on " Hypersensibility," in Journ. Amer. Mcd. Assoc.
(1906), xlvii. 1010. Rosenau and Anderson, Hya. Lab. Bull., Washington,
Nos. 29, 39, 45 ; Journ. Infect. Diseases (1907), vol. iv. 1. Otto, in v.
Lriithold-Gedenkschrift, Bd.i. art. " Anaphylaxie/'inKolle-Wassermann's
" Handbuch," Ergiinz.-Bd. ii. Hft. 2. Gay and Southard, various papers
iu Journ. M«l. Itixearch (1907), xvi. et seq. Doerr, art. " Anaphylaxie,"
in Kraus-Lcvaditi's " Haudbuch." Various papers by Besredka in Ann. de
r Inxt. I'ltxt.-m: 1907. i'f wq., and by Biedl and Kraus, Friedberger, Doerr

and Russ, in Ztschr. f. Jmmnnitdtsf. Bd. ii. et seq. Bail, ibid. (1909),
iv. 470. v. Pirquet and Schick, "Die Seiumkrankheit," Wien, 1907.
Currie, Journ. Hy<j. (1907), vii. 35, 61. Goodall, ibid. 607. Scott,
./mini. I'nih. mid Bacterial. (1909), xiv. 147 and (1910), xv. 31. Auer
and Li-wis, Journ. Amer. Med. Assoc. (1909), liii. 458.

APPENDIX A.— SMALLPOX.

Jenner, "An Inquiry into the Causes and Effects of the Variola
Vaccime," London, 1798. Creighton, art. "Vaccination" in Ency.
I'.rlt., 9th cd. Crookshank, "Bacteriology and Infective Diseases."
M'Vail, "Vaccination Vindicated." Chauveau, Viennois et Mairet,
'• Vaccine et variole, nouvelle e"tude experimentale sur la question de
ridentitc- de res deux affections," Paris, 1865. Klien, Jb-p. Med. Off.
Loot/ &ovt. Hoard (1892-93), 391 ; (1893-94), 493. Copeman, Brit. Med.
Jour*. (1894), ii. 631 ; Journ. Path, and Bacterial, ii. 407 ; art. in
Clifford Alllmtt's "System of Medicine," vol. ii. L. Pfeiffer, "Die
I'roto/oi-n als Krankheitserreger," Jena, 1891. Ruffer, Brit. Med. Jotmi.
(1894), June 30. Becli-re, Chambon, and Meiiard, Ann. de VInst. Pasteur,
\. 1 ; xii. 837. Copeman, "Vaccination," London, 1899. Calmette and
Hiu'riii. Ann. dr. V Inst. Pasteur, xv. 161. Guarnieri, Centralbl. f.
l'»ikt' r'ml. u. /'araffitenk. xvi. 299. Ewing, Journ. Med. Research, xiii.
'!'•'>'.). Pi'owa/ck, Arb. «. <l. kaiscrl. Ge.sumllteitsamte, xxii. 535 ; xxiii.
.".-jr.. \V;i<irlc\\ski, Z/si-Jir. f. Hyg. xxxviii. 212. Bonhoff, Berl. /•////.
//V////.sr/«/\ (I'.'O'.X ].. 11-12. Carini, CwtiralW. f. BaJct<>rioL u. Pmrasttcrik.
(Orig.) xxxix. 6S5.

APPENDIX B.— HVDUOPHOIHA.

Pastoar, r,t,n^. revd. Acad. d. sc. xdi. 1259; xcv. 1187; xcviii. lf,7,
1229 ; ci. 765 ; cii. 459, 835 ; ciii. 777. Schaffer, Ann. de. VInst. Pant fin:
iii. 644. Fit-mill^. Trims. 1th Internal. Cony. Hyg. and Demo;/, iii. 16.
Hdinan, Ann. de I' List. Pasteur, ii. 274 ; iii. 15. Babes and Lepp, ibid.
iii. :JS-I. Xocard and Roux, ibid. ii. 341. Roux, ibid. i. 87 ; ii. 479.
Bruschettini. <:,;i/,-albl. f. Bakteriol. it. Parasitenk. xx. 214 ; xxi. 203.
Mi-mum, ii'iit. \\. 209 ; xxi. 657. Frantzins, ibid, xxiii. 782 ; xxiv. 971.
K.Mi.liimiT. Ann. /A- n,<*i. Pasteur, xvii. 834; xviii. 150; xix. «;2.'..
Harv.-y :md M.-K.-ndrick, Sc. Mem. by Officers of Med. a<t>' San it. />•/•'*.
n'ort. huiia (New Series), No. 30 (1907), Calcutta. L-mm and M.-Ki-n-
dri.-k. ibid. (1909), No. 3(5. Hiigyes, Lyssa, in Nothnagers "SjM-c. Path.
u. Th. T," Vienna, 1897.

670 BIBLIOGFxAPHY

NEGRI BODIES. — Negri, Ztschr. f. Hyg. u. Infcctionskrankh. xliii.
507 ; xliv. 519 ; Ixiii. 421. Williams and Lowden, Journ. Inf. Diseases,
iii. 452. Bertarelli, Centralbl. f. Baktcriol. xxvii. 556. D'Amato and
Faggella, Ztschr. f. Hyg. (1910), Ivi. 351. Frosch, in Kolle and
Wassermann's "Handbuch der Patliogenen Mikro-organismen," Erganx-
ungsband, i. 626. Frothingham, Am. Journ. Pub. Hyg. (1908), xviii.

APPENDIX C. — MALARIAL FKVEK.

Laveran, Bull. Acad. de. med. (1880), ser. ii. vol. ix. 1346; " Du
paludisme et de son hematozoaire," Paris, 1891. Marchiafava and Celli,
Fortschr. d. Med., 1883 and 1885 ; also in Virchow's Festschrift. Golgi,
Arch. per. le. sc. med., 1886 and 1889 ; Fortschr. d. Med. (1889), No. 3 ;
Ztschr. f. Hyg. x. 136 ; Deutsche med. Wclmschr. (1892), 663, 685, 707,
729. Steinberg, New York Med. Rec. xxix. No. 18. James, ibid, xxxiii.
No. 10. Councilman, Fortschr. d. Med. (1888), Nos. 12, 13. Osier,
Trans. Path. 8oc. Philadelphia, xii. xiii. Grass! and Feletti, Riforma
med. (1890), ii. No. 50. Carialis, , Fortschr. d. Med. (1890), Nos. 8, 9.
Danilewsky, Ann. dc I'lnst. Pasteur, xi. 758. "Parasites of Malarial
Fevers," New. Syd. Soc., 1894 (Monographs by Marchiafava and
Bignami, and by Mannaberg, with Bibliography). Manson, Brit. Med.
Journ. (1894). i. 1252, 1307 ; Lancet (1895), ii. 302 ; Bril. Med. Journ.
(1898), ii. 849 ; Koch, JBerl. klin. Wchnschr. (1899), 69. Ross, Indian
Med. Gas. xxxiii. 14, 133, 401, 448. Nuttall, Centralbl. f. Bakteriol. u.
Parasitenk. xxv. 877, 903; xxvi. 140; xxvii. 193, 218, 260, 328 (with
full literature). Manson, Lancet (1900), i. 1417 ; (1900), ii. 151. Gray,
Brit. Med. Journ. (1902), i. 1121. Leishnian, ibid. (1901), i. 635 ; ii.
757. Daniel, ibid. (1901), i. 193. Celli, ibid. (1901), i. 1030. Nuttall
and Shipley, Journ. of Hyg. i. 45, 269, 451 (with literature). Ross,
Nature, Ixi. 522; "Mosquito Brigades and how to organise them,"
London, 1902. Celli, " Malaria," trans, by Eyre, London, 1900. Lan-
kester, Brit. Med. Journ. (1902), i. 652. Ewing, Journ. Exper. Med. v.
429 ; vi. 119. Schaudinn, Arbeit, a. d. kaiserl. Gesundheitsamtc, xix ;
Arrjutinsky Archiv mikroskop. Anat. lix. 315 ; Ixi. 331. Ruge in Kolle
and Wassermann's " Handbuch der Pathogenen Mikro-organism," Erganz-
ungsband, 1907 (full literature). Ross, Lancet (1903), i. 86. Minchin,
"The Sporozoa," London, 1903. Stephens, art. " Blackwater Fever,"
in Allbutt's "System of Medicine," vol. ii. pt. ii., London, 1907.
Laveran, "Traite du paludisme," 2nd ed., Paris, 1907. Stephens and
Christophers, " The Practical Study of Malarial and other Blood Para-
sites," 3rd ed., Liverpool, 1908. Christophers and Bentley (Blackwater
Fever), " Scientific Memoirs published by the Government of India,"
No. 35, Simla, 1908.

APPENDIX D. — AMCEBIC DYSENTERY.

Lbsch, Virchow's Archiv, Ixv. 196. Cunningham, Quart. Journ. Micr.
•S'c., N.S. xxi. 234. Kartulis, Virchow's Archiv, cv. 118 ; Centralbl. f.
Bakteriol. u. Parasitenk. ii. 745 ; ix. 365. Koch, Arb. a. d. k. Gsndht-
samte. iii. 65. Councilman and Lafleur, John? Hopkins Hosp. Rep. (1891).
ii. 395. Maggiora, Centralbl. f. Bakteriol. u. Parasitenk. xi. 173. Ogata,
ibid. xi. 264. Schuberg, ibid. xiii. 598, 701. Quincke and Roos, Bed.
klin. Wchnschr. (1893), 1089. Kruse and Pasquale, Ztschr. f. Hyg. xvi.
i. Cieohanowsld and Nowak, Centralbl. f. Bakteriol. u. Parasitrnk. xxiii.

BIBLIOGRAPHY 671

1 !."•. Howard and Hoover, Am. Joum. Med. Sc. (1897), cxiv. 150, 263.
Harris, 'Vir<'1unrs Archiv, clxvi. 67. Schaudinn, Arbeit, a. d. IcaiserL
tlx.iilhlMtntt''. (190:5), xix. r.-17. Lesage, Ann. de I'List. Paste,,,- (1905),
xix. 9. Kartulis in Kolle aiul Wasscrmaun's " Haudbuch der Pathogcnen
Mikro-organismcn," Krgiin/ungsband, 1906 ; Centralbl. /. Bakteriol.
<>rij{). (1904), xxxvii. 527. Musgrave and Clcgg, " Amccbas, their
Cultivation and Etiologic Signification," Bureau of Government Labora-
tories, Manila, 1904 ; Philip]). Journ. of Science (1906), i. Craig, Jnvrn.
////<-•/. 1 >;*•„*•* (1908), v. 324. Viereck, Bull. dc. Vlnst. Pnxteur (1907),
v.' 819. Hartmann, ibid. (1908) vi. 100. Werner, Arch. J Set/. ?/.
7V"/"'/'/'//.'/. xii. 11. Noc, Ann. del'lnst. Pasteur (1908), xxiii. 177.

ArTKNDIX E. — TllYPANOSOMIASlS, KTC.

C.KNKKAI,. Lavcian and Mesnil, " Trypanosomes et trypanosomiasis,"
Paris, Masson, 1904. Minehin, in Clifford Allbntt's "System of
Medicine," 2nd cd. vol. ii. pt. ii. p. 9, London, Macmillan, 1907.
Schnudinn. Arlxif. <>. <I. hiiscrl. Gcsundhcitsamte, xx. -387. Mensc,
" Handbueh der Tropenkrankheiten," Leipzig, 1906, Barth. Novy and
Mai-NVal. ,l»nr,,. Inf. Diseases, ii. '256. Leishman, Journ. Hyg. iv. 434.
Minchin and Thomson, Proc. Roy. Soc. London, B. (1909), Ixxxii. 273. '
(Trypanosoma Crnzi), Chagas, Ref. in Bull, de VInsl. Pasteur (1910),
viii. 373.

S i . i . i •: i- 1 N < ; Si< K N KS s. — Mott, Reports of the Sleeping Sickness Commission
"/ flic Royal Society, pt. vii. No. 15, London, Bale, Sons & Dannielsson,
r.'Oi;. Dutton and Todd, Brit. Med. Journ. (1903), i. 304. Duttou,
.ind Tod<l, Tli<>,,ii>x(>n-Yates Lab. Rep. v. pt. ii. i. ; v. pt. ii. 97. Button,
Todd, and Christy, ibid. vi. pt. i. p. 1. Manson and Daniels, ibid. (1903),
i. 1249. Idem, ibid. (1903), ii. 1461. Low and Mott, ibid. (1904), i.
1000. Bettenconrt, Kopke, Resende, and Mendes, ibid. (1903), i. 908.
Castellani, Jfcjiorta of the Sleejiimj Sickness Commission of the Royal
Surety, No. 1, i. 1, London, Harrison & Sons, 1903. Bruce and
Nabarro, ibid. (1903), No. 1, ii. 11. Bruce, Nabarro, and Greig, ibvl.
(1903), No. 4, viii. 3. Greig and Gray, ibid. (1905), No. 6, ii. 3.
L'-ishman, Journ. Hyg. iv. 434. Minchin, Gray, and Tulloch, Reports of
thr stc?itiiig Sickness Commission of the Royal Society, No. 8, xxi. 122,
London, H.M. Stationery Office, 1907. Manson, Brit. Med. Journ.
(1903), ii. 1249, 1461. See discussions at British Medical Association,
/;/•/'. M>'d. Jonrn. (1903), ii. 637 ; (1904), ii. 365. Thomas, Thompson-
y.ifc* Lnb. Urp. vi. pt. ii. 1. Kleine, Deutsche mcd. Wchnschr. (1909), pp.
I'.1.'. 924, 1257, 1956. Bnice, Hamerton, Bateman and Mackie (Sleeping
Sirkncss Commission of Royal Society, 1908-9), Proc. Roy. Soc. London, P..,
Ixxxi. 40f> ; ibid. Ixxxii. pp. 5t», 63, 256, 368, 480, 485, 498.

LI.I>HM \NIA DONOVANI.— Leishman, Brit. Med. Journ. (1903), i. 1252.
/</• in. in Clifford ADbntt'a " System of Medicine," 2nd ed. vol. ii. pt. ii. 226,
London, Macmillan, 1907. Idem, Mense, " Handbnch der Tropenkrank-
heiten," iii. 591, Leipzig, Barth., 1906. Leishman and Statham, Journ
<>f Roy. Army Med. Corp*, iv. 321. Donovan, Brit. Med. Journ. (1903),
ii. 79. Rogers, Quart. Journ. Micr. Soc. xlviii. 367. Idem, Brit. J/"/.
Jowrn. (1904), i. 1249 ; ii. 645. Idem, Proc. Roy. Soc. Ixxvii. 284.
IVntley, Brit. Mcd. Journ. (1904), ii. 653 ; ibid. (1905), i. 705. Chris-
tophera, Scientif. Mem. by Off. of the Mfd. and San. Dept. of the (,'ovt. of
Iii'lin, Nos. 8, 11, 15. Ross, Brit. Mcd. Journ. (1903), ii. 1401. See
discussion at Brit. Med. Ass< ...... ///•//. M«l. Journ. (1904), ii. 642. Patton,

672 BIBLIOGRAPHY

Sc, Mem. by Officers of Med. and San. Depts. Gov. India, Calcutta, 1907,
No. 27. (Histoplasmosis), Darling, Journ. Ex. Med. (1909), xi. 515.

LEISHMANIA INFANTUM. — Nicolle, Ann. del'Inst. Pasteur (1909), xxiii.
361, 441. See also references, Bull, de Vlnst. Pasteur, viii. 164, 680.
Pianese, G-azz. intern, di Medicin. viii. 8.

LEISHMANIA TROPICA. — Wright, J. H., Journ. Med. Research, x. 472.
Marzinowsky, Ztschr. f. Hyg. Iviii. 327. Row, Quart. Journ. Med. Sc.
liii. 747. Nicolle and Manceaux, Ann. de Vlnst. Pasteur, xxiv. 673.
Thomson and Balfour, Journ. Roy. Army Med. Corps (1910), xiv. 1.

PIROPLASMOSIS. — See Minchin, loc. cit. supra. Koch, Deutsche med.
Wchnschrft. (1905), No. 47 ; Ztschrft.f. Hyg. u. Infektionskrankh. liv. i.
Nuttall, Journ. Hyg. iv. 219. Nuttall and Graham-Smith, ibid. v.
237 ; vi. 586.

APPENDIX F. — YELLOW FEVER.

Sternberg, Rep'. Am. Pub. Health Ass. xv. 170. Sanarelli, Ann,, de
Vlnst. Pasteur, xi. 433, 673, 753 ; xii. 348. Davidson, art. in Clifford
Allbutt's "System of Medicine," vol. ii., London, 1897. Sternberur,
Centralbl. f. Bakteriol. u. Parasitenk. xxii. 145 ; xxiii. 769. Sanarelli,
ibid. xxii. 668. Reed and Carroll, Medical News, April 1899. Reed,
Journ. of Hyg. ii. 101 (with full references). Durham, Thompson - Yates
Laboratory Rep. (1902), iv. pt. ii. 485. Gorgas, Lancet, 1902, Sept. 9 ;
1903, March 28. Marchoux, Salimbeni, and Simond, Ann. de Vlnst.
Pasteur, xvii. 665 ; xx. 16, 104, 161. Bandi, Ztschr. f. Hyg. (1904), xivi.
81. Otto and Neumann, Ztschr. f. Hyg. (1905), Ii. Heft 3. Reed, Carroll,
Agramonte, Lazear, Proc. Amer. Health Ass., 1900; Journ. Amer. Med.
Ass., Feb. 1901. Carroll, Neiv York Med. Journ., Feb. ]904 ; Amer.
Medicine (1906), xi. 383. Thomas, Brit. Med. Journ. (1907), i. 138.

APPENDIX G.— EPIDEMIC POLIOMYELITIS.

Landsteiner and Popper, Ztschr. f. Immunitatsforsclmng (Orig.) (1902),
ii. 377. "Epidemic Poliomyelitis," Report on New York Epidemic of
1907, New York, 1910. Flexner and Lewis, Journ. Am. Med. A*s.
(1909), liii. 1639, 1913, 2095 (1910), liv. 45, 1140, 1780. Landsteiner
and Levaditi, Comp. rend. Soc. de biol. Ivii, 592, 787. Levaditi and
Landsteiner, ibid. Iviii. 3, 11, 417. Netter and Levaditi, ibid. Iviii.
617, 855. Levaditi, Presse med. (1910), 43.

APPENDIX H. — PHLEROTOMUS FEVEIJ.

"System of Medicine" (1907), ii. (2) 345. Ashburn and Craig,
Philippine Journ. Sc. Med. ii. 93 (Ref. in Bull. de. Vlnst. Pasteur
(1907), v. 773).

APPENDIX J. — TYPHUS FEVER.
Nicolle, Ann. de Vlnst. Pasteur (1910), xxiv. 243.

INDEX.

Al.rin, 196

iinm unity against, 520, 525
Abscesses (see also Suppuration) :

bacteria in, 202

in dysentery, 607
A 1 isolate alcohol, fixing by, 96
Absorption of complement test, 121
Acid-fast bacilli, 264, 278

stain for, 107
A i -ill formation, observation of,

51, >2
Ai-ijiiired immunity in man, 528

theories of, 5-18
Actinomyces, 16

characters of, 317, 318

cultivation of, 323

inoculation with, 327

varieties of, 325
Actinomvooau, 317

anaerobic streptothricos in, 326

diagnosis of, 327

lesions in, 321

origin of, 323

Active immunity, 514, 515
Aerobes, 18

culture of. ."7

separation of, 56
JEstivo-autumnal fevers. .V.i:;
African tick fever, 494
Agar media (.sw also Culture media),

separation by, 60
Agglutinable substance, 543
Agglutination by sera, .'. 1 1

in relapsing fever, 498

methocuL 118

of b. mallei, 313

of b. typhosus, etc., :',71

Agglutination of cholera vibrio,
458

of in. melitensis, 492

of plague bacillus, 487

of red blood corpuscles, 537, 542

theories regarding, 542
Agglutinins, measurement of group,
120

primary (homologous), 375

secondary (heterologous), 375
Agglutinogen, 543
Agglutinoids, 543
Aggressins, 189
Air, bacteria in, 147

examination of, for bacteria, 147
Albumose of anthrax, immunity

by, 343
Albumoses, 193

in diphtheria, 408
Alcohols, higher, fermentation of,

Aleppo boil, 636
Alexines, 534, 557
Amanita phalloides, toxin of, 197
Amboceptors, 536, 549
Amcebic dysentery, 602
Amoibulae of malaria, 587
Anaerobes, 18

cultures of, 65

fusiform, 414

separation of, 63

toxins of, 60

Anaerobic Buchner tubes, 66
Anaerobic Esmarch's tubes, 64
Anaerobic fermentation tubes, 65
Anaerobic plate cultures, liulloch'.s

apparatus for, 64
Ana-fcthetio leprosy, 299

674

INDEX

Anaphylactin, 561
Anaphylaxis, 192, 558

mechanism of, 561

phenomena in, 559

reaction-bodies in, 561, 562

in relation to rabies, 583

supersensitiveness to tetanus,
429

toxic phenomena, 192

tubercular sensitiveness, 287
Aniline oil, dehydrating by, 100

water, 105

Aniline stains, list of, 101
Animals, autopsies on, 145

inoculation of, 141
Anthrax, 331

anti-serum, 347

bacillus, 332
biology of, 335
cultivation of, 333
inoculation with, 341
toxins of, 343

diagnosis of, 348

in animals, 337

in man, 341

protective inoculation, 346

spread of, 344
Anti-abrin, 625
Anti-anthrax serum, 346
Anti-bacterial sera, 532

properties of, 533
Anti-cholera vaccination, 458
Anti-diphtheritic serum, 523
Antiformin, 295
Antigens, 521
Antikorper, 513
Anti-plague inoculation, 486
Anti-plague sera, 486
Antipneumococcic serum, 239
Antirabic serum, 583
Anti-ricin, 525
Anti-sensibilisin, 561
Antiseptics, 166

actions of, 168

standardisation of, 167

testing of, 166
Antisera, therapeutic action of,

546

Antistreptococcic serum, 547
Anti-substances, specificity of, 521
Antitetanic serum, 429

preparation of, 522 et seq.
Antitoxic action, nature of, 526
bodies in normal tissues, 530

Antitoxic sera, use of, 525
serum, 522

standardisation of, 524
Antitoxins, chemical nature of,

526

origin of, 530
Antitubercular sera, 294
Antityphoid serum, 377
Aortitis, syphilitic, 506
Appendicitis, 212
Arthrospores, question of occurrence

of, 7

Arthus on anaphylaxis, 559
Artificial immunity, varieties of,

513 et seq.

Attenuation of virulence, 514
Auer and Lewis on anaphylaxis,

561

Autoclave, 30
Autolysis of bacteria, 188
Autopsies on animals, 145
Avian tuberculosis, 276

Bacilli, acid-fast, 264, 278
stain for, 107
anaerobic fusiform, 444
arthrosporous, 7
characters of, 14
Bacillus acidi lactici, 21, 392
aerogenes capsulatus, 208, 442
^Ertryk, 380, 384
anthracis, 332
botulinus, 438
coli anaerogenes, 393
coli communis, lesions caused by,
211 et seq.

agglutination reactions, 354

characters of, 351

culture media for, 51, 57, 154

culture reactions, 351 et seq.

gas formation by, 353

isolation of, 354

morphological characters of,
351

pathogen icity of, 356

in soil, 154, 155

type characters of, 355

in water, 157, 161
of cholera, 447
cloacae, 392
of Danysz, 384
diphtheria, 397
dysenterise Shiga-Flexner, 385
of Emmerich, 393

INDEX

675

Bacillus, cntcritidis (Gaertner), 383
euteritidis sporogenes, 389

iu soil, 154, 155
of Eschcrich, 350
f.i-'.-alis alcaligenes, 393
'•!' glanders, 308
of hog cholera, 380
of Hiippe, 393
icteroides, 640
<>f influenza, 467
Koch-Weeks, 219
lactis aerogenes, 208, 393
lacunatus, 220
of leprosy, 299
of malignant it'dema, 433
Muller's, 219
mycoides in soil, 153
neapolitanus, 393
o/.i-na-, 316

oxytocus perniciosus, 393
paratyphosus, 382, 379
of plague, 475
pneumonia', 226
pseudo-diphtherieus, 410
of psittacosis, 384
putrificens, 441
I'V.ti-yaiHMi.s, 208

agglutination of, 541

occurrence of, 213
pyogi'iu's t'n-tidus, 202
of quarter-evil, 441
of rhinoscleroma, 315
saccliam !>utyricus, 441
"f sinegma, 380
• •Is,, ft sore, 257
Mibtilis, 62
suipestifer, 380
of syphilis, 50-'5
tetaiii, 416

of Timothy grass, 278
of tubercle,
typhi murium, 380
of typhoid, 35»;
of whooping-cough, \l'i
of xerosis, 411

Bacti-ria, action of dead, 180
aerobic («<•<• Ai-mln-.s), 18
anaerobic (scr Anaerobes), 18
In'ology of, 17
< .i|»uluted, 3
chemical action of, 22

composition of, 10
classification of, 11
cultivation of, 26

- Bacteria, death of, 166

effects of light on, 19

food supply of, 17

higher, 15

lower, 12

microscopic examination of, 91

morphological relations of, 2

motility of, 8

movements of, 20

multiplication of, 4

nitrifying, 24

parasitic, 22

pathogenic, action of, 175
effects of, 181

saprophytic, 22

separation of, 56

species of, 24

spore formation in (so U!M.
Spores), 5, 62

structure of, 3

sulphur-containing, 10

temperature of growth of, 19

toxins of, 187

variability among, 24

virulence of, 176, 517
Bacterial ferments, 23, 195

pigments, 10

protoplasm, structure of, 9

treatment of sewage, 163
Bactericidal methods, 126

powers of serum, 533

substances, 534
Bacteriological diagnosis, 138

examination of discharges, 136
Bainbridge on agglutination, 121
Beer wort agar, 52
Beggiatoa, 16
Behring on immunity, 429
Besredka on anaphylaxis, 561
Bile-salt media, 50
Bismarck-brown, 101
Blackleg, 441
Blackwater fever, 599
Blastophores (malaria), 593
Blood-agar (see also Culture media),

43
Blood, examination of, 72, 94

in malarial fever, 584

in relapsing fever, 494

serum, coagulated, as medium, X)
Blood-smeami agar, 13
Bone-marrow in leucocytosis, 182
Bordet's phenomenon, 534
| Bordet and Gengou's medium, 44

676

INDEX

Bordet and Gengou on whooping-
cough, 472 et seq.

Botulism, bacillus of, 438
toxin of, 529

Bouillon (see also Culture media), 33

Bovine tuberculosis, 274

Bread paste, 47

Brieger and Boer, 193

Brieger and Fraenkel, 187

Buboes, 258

Bubonic pest, 475

Buchner on alexines, 557

Buchner's anaerobic tubes, 66

Bulloch's apparatus for anaerobic
culture, 64

Biitschli on bacterial structure, 10

Butter bacilli, acid-fast, 279

Calmette, 486, 518, 525
ophthalmo-reaction of, 285

Canary fever, 646

Canon on influenza, 467

Cantani on influenza, 471

Capaldi and Proskauer, media of,
362

Capsules, staining of, 109

Carbol-fuchsin, 105
-methylene-blue, 104
-thionin-blue, 105

Carbolic acid as antiseptic, 173

Carroll's method of making anae-
robic cultures, 65

Carter on relapsing fever, 496

Castellani on frambcesia, 511

Cattle plague, 569

Cerebro-spinal fluid, examination
by lumbar puncture, 73

Chagas on trypanosomiasis, 629

Chamberland and Roux, attenua-
tion of b. anthracis, 516

Chamberland's filter, 75

Chemiotaxis, 20, 553

Chitral fever, 646

Chlorine as antiseptic, 170

Cholera, 446
anti-sera, 458
culture methods, 44, 449
immunity against, 457
inoculation of man with, 457
methods of diagnosis of, 459
preventive inoculation against,

459
-red reaction, 451

Cholera carriers, 453

Cholera spirillum, 447

distribution of, 449

inoculation with, 453

powers of resistance of, 452

relations to disease, 461

toxins of, 456
Cladothrices in soil, 153
Cladothrix, 16

asteroides, 326
Clubs in actinomyces, 320
Coccaceae, 140
Cocci, characters of. 12
Coli-typhoid bacteria, 391, 394
Collodion capsules, preparation of,

144

Colonies, counting of, 70
Comma bacillus, 446
Commission on tuberculosis, 274

on vaccination, 527
Complement, 534

bacteriophilic, 538

constitution of, 534, 538

deviation of, 545

method of estimating, 126, 130

in glanders, 313

in relation to precipitins, 546

in tuberculosis, 289
Congestin, 559
Conjunctivitis, 219
Conradi-Drigalski medium, 47
i Conradi's picric acid method, 49
i Copeman on smallpox, 570
Copper sulphate method, 109
Cornet's forceps, 94
Corrosive films of blood, etc., 95
| Corrosive sublimate, as antiseptic,
171

fixing by, 96

Councilman and Lafleur on dysen-
tery, 602
Counting of colonies, 70

dead bacteria in a culture, 133

living bacteria in a culture, 71
Cover-glasses, cleaning of, 94
Cowpox, relation to smallpox, 567
Crescentic bodies in malaria, 587
Cultivation of anaerobes, 62
Culture media, preparation of, 31
et seq.

agar, 37

alkaline blood serum, 42

blood agar, 43
serum, 40

bouillon, 33

INDK.X

677

( 'ulture media, preparation of :

1 >rra<l paste, 47

glucose agar, 38
broth, 36
gelatin, 37

•^Ivccrin agar, 38
broth, 36

litmus whey, 51

LiifHer's serum medium, 41

Marmorek's serum media, 42

meat extract, 32

milk, 46

peptone gelatin, 36
solution, 39

potatoes, 45

serum agar, 43
Cultures, destruction of, 89

filtration of, 74

from organs, 137, 145

hanging-drop, aerobic, 69

incubation of, 85

microscopic examination of, 91

permanent preservation of, 88

plate, 59

pure, 54

"shake," 81

Cutaneous tuberculin reaction, 285
< 'utting of sections, 97
( 'ystitis, 212, 255
Cytases, 553
( 'ytoly tic sera, 538

1 ) ni\ >/.'s bacillus, 384

I )arling on hisLoplusma capsulatum,

637

Dead cultures, counting of, 133
I)e Bary, definition of species, 24
Decolorising agents, 103
Deep cultures, 65
Dehydration of sections, 100
Delepine, agglutination method, 119
Delhi sore, 636
Deneke's spirillum, 465
Dengue fever, 646
Deviation of complement, 126, 130,

546

Dextrose-free bouillon, 80
Diagnosis, bacteriological, 135, 138
Dieudonne's medium, 44
Diphtheria, 396

diagnosis of, 412

immunity against, 523

origin and spread of, 398

paralysis in, 397, 405

Diphtheria, results of treatment,

546
Diphtheria bacillus, action of, 402

bacilli allied to, 409

characters of, 397

distribution of, 398

fermentation reactions of, 402

inoculation with, 404

isolation of, 412

powers of resistance of, 403

staining of, 115, 403

toxins of, 191, 405, 407, 409

variations in virulence of, 408
Diphtheroid bacillus, 410
Diplo-bacillus of conjunctivitis, 220
Diplococcus, 12

catarrhalis, 247

crassus, 247

endocarditidis encapsulate, 216

intracellularis meningitidis, 242

mucosus, 247

pharyngis, 247

pneumonias, 227
Disaccharides, 79
Disturbances of metabolism by

bacteria, 185

Doerr on phlebotomus fever, 646
Dorset's egg media, 267
Dreyer and Jex-Blake on aggluti-
nation, 543

Drigalski-Conradi medium, 47
Drying of sera, etc. , in vacuo, 84
Ducrey's bacillus, 257

cultivation of, 258
Dum-Dum fever, 630
Durham's fermentation tubes, 81
Dysentery, amoebic, 602

bacteria in, 384

characters of amoeba of, 602
cultivation of, 605
distribution of, 606
inoculation experiments, 607
Dysentery, methods of examination
in, 385

East coast fever in cattle, 638

Eberth's bacillus, 350

Eel serum, 198

Egg media for tuberculosis, 267

Ehrlich on ricin and abrin, 520, 525

on toxins, 198

rosindol reaction, 83

side-chain theory of antitoxin
formation, 649

678

INDEX

Eisenberg on anthrax, 190
Eisner's medium, 51
Embedding in paraffin, 97
Emmerich's bacillus, 393
Empycnia, 234, 470
Endows medium, 49
—-Endocarditis, bacteria in, 216

Endotoxins, 522. See Intracellular

toxins.

Enhaeinospores (malaria), 587
Entamoeba coli, 603
Entamceba histolytica, 603

cultivation of, 605
Enteritis, dysenteric, 386, 606
Epidemic cerebro-spinal meningitis,
242

poliomyelitis, 644
Eppinger's streptothrix, 326
Ermengem on botulism, 438

stain for flagella, 110
Erysipelas, 218
Escherich's bacillus, 350
Esmarcli's roll-tubes, 60

anaerobic, 63

Exaltation of virulence, 517
Examination of water, 156
Exhaust-pump, 75
Exotospores (malaria), 586
Extracellular toxins, 190, 522

False membrane, 212, 397

Farcy, 307

Fawcus' picric acid method, 49

Feeding, immunity by, 520

Fermentation by pneumo-bacillus,

232

by bacillus coli, 352, 392
by b. diphtherias, 402 -
methods of observing, 78
of sugars by bacteria, 79
test of bacterial action, 79
tubes, 81
anaerobic, 65

Ferments formed by bacteria, 23,

195
in diphtheria, 402, 408

Ferrata on complement, 534

Fever, 185

Film preparations, dry, 93
wet, 95
staining of, 102

Filter, porcelain, gelatined, 193

Filtration of cultures, 74

Finkler and Prior's spirillum, 464

Fish, tuberculosis in, 277

Fixateurs, 553

Fixation of complement, 130

of tissues, 96
Klagella, nature of, 8

staining of, 110
j Flagellated organisms in malaria,

592
Flexner on epidemic poliomyelitis,

644

Fliigge, 15

Food-poisoning bacilli, 379
Forceps for cover-glasses, 94
Ford Robertson on diphtheroid

bacilli, 410

Formalin as antiseptic, 171
Forster on typhoid fever, 368
Foth's dry inallein, 314
Fraenkel's pneumococcus, 226, 229,
230

stain for tubercle, 107

on whooping-cough, 474
Framboesia, spirochpetes in, 511
Frankland on water bacteria, 160
Fraser, T. R., 518, 525, 532
Friedberger on anaphylaxis, 562
Friedliinders pneumobacillus, 226,

232

Frisch on rhinoscleroma, 316
Frothingham on Negri bodies, 577
Fuchsin, carbol-, 105, 108
Fusiform anaerobic bacilli, 444

Gallstones in relation to typhoid

fever, 364

| Gamale'ia on pneumonia, 235
i Gametocytes (malaria), 587
Gangrenous emphysema, 433, 437

pneumonia, 470
Gas formation, observation of, 50

Gas-regulator, 86

Gay and Adler on anaphylaxis, 561
| Geissler's exhaust-pump, 75
Gelatin media, 36

separation by, 56
Gelatined porcelain filter, 193
General paralysis, diphtheroid

bacilli in," 410

Wasserman reaction in, 510
j Gentian-violet, 105
I Germicides, 166
i Geryk pump, 85
i Giemsa's stain, 115

INDEX

G79

Giemsa's stain for spirochrotes in

films, 115
Glanders, 306

diagnosis of, 314

in horses, 307

in man, 307

loii ins in, 312
Glanders bacillus, 308

agglutination of, 313

inoculation with, 311
Globulin, constituent in antitoxin,

532
Glossina morsitans, 618

palpalis, 624
( llucosc media, 3»> r.t .*••"/.
Clucosides, fermentation of, 79
Glycerin media, 36 et scq.

potato as culture medium, 46
(Jolgi on malaria, 585
Gonidia, 16
Gonococcus, characters of, 249

comparison with meningococcus,
252

culture methods, 42, 43

inoculation with, 253

toxin of, 253
Gonorrhoea, 249
Gonorrhoeal conjunctivitis, 255

endocarditis, 256

septicaemia, 256
Craham-Smith on identification of

bacilli, 4
Gram's method, 105

Kiihne's modification of, 106

Much's modification of, 265

Wcigrrt's modification of, 106
(irassberger and Schattenfroh on
quarter-evil, 441, 442

on symptomatic anthrax, 192
Grease, 566

Civrnticld on anthrax, 343,480,516
Group agglutinins, measurement of,

120
Griiber and Durham's phenomenon,

541

Guarnieri bodies in smallpox, 570
Gulland (methods), 95, 99

Hsemamceba Danilewski, 594

malaria;, 594

pr;«'cpx, 594

relicta, 594

vivax, 594
Hrematozoon malaria? , 585

Hremolytic sera, 536

Hremolytic tests, methods of, 128,

538

Ha triune on anti-cholera inocula-
tion, 459
Haffkine's inoculation method

against plague. 486
Haltcridium, 592, 594
Hanging-drop cultures, 69

examination of, 91
Hankin, 344

Hansen, leprosy bacilli, 299
Harrison's method for counting

bacteria, 134
Hesse's tube, 148
Hiss's serum water media, 47

method of capsule staining, 109
Histoplasma capsulatum, 637
Hermann's bacillus, 410
Hog cholera, 380

Hogyes on treatment of hydro-
phobia, 582
Horsepox, 569

Houston on bacteriology of soil, 152
Hiippe, 7, 15
Hiippe's bacillus, 393
Hydrogen, supply of, 62
Hydrophobia, 573

diagnosis of, 583

Negri bodies in, 576

prophylactic treatment of, 575,
579

virus of, 579
Hypodermic syringes, 143

Immune-bodies, 534

origin of, 536

Immunity (sec also Special Diseases),
512

acquired, theories of, 548

active, 515, 516

artificial, 513

by feeding, 520

by toxins, 518

methods, 515

natural, 555

passive, 514, 520

unit of, 524

Impression preparations, 138
Incubators, 85
Indol, formation of, 82
Infection, conditions modifying,
175

nature of, 179

G80

INDEX

Inflammatory conditions due to

bacteria, 183
Influenza, 467

lesions in, 471

sputum in, 469
Influenza bacillus, 467

cultivation of, 43, 468

inoculation with, 471

pseudo-bacilli, 470
Inoculation, methods of, 141

of animals, 141

of tubes, 55

protective, 518 etseq.

separation by, 61
Intestinal changes in cholera, 449

amoebic dysentery, 606

bacterial dysentery, 387

typhoid fever, 365
Intestinal infection in cholera (ex-
perimental), 453
Intracellular toxins, 188, 522
Involution forms in bacteria, 5
Iodine solution, Gram's, 106

terchloride, 523

as antiseptic, 170
lodoform as antiseptic, 173
Issaeff, 520
Ivanoff's vibrio, 462

Japanese dysentery, 389
Jenner on vaccination, 565
Jenner's stain, 113
Johne's bacillus, 279
Joints, gonococci in, 255

Kala-azar, 630, 635
Keratitis, syphilitic, 509
Kipp's apparatus, 63
Kitasato on bacillus of influenza,
467

of plague, 475

of tetanus, 416 etseq.
Klebs-Lbffler bacillus, 396
Klein, 378, 570
Klemperer on pneumonia, 239
Klimenko on whooping-cough, 474
Knapp and Novy on relapsing fever,

498
Koch on avian tuberculosis, 276

bacillus of malignant redema,
433

bovine tuberculosis, 274

cholera spirillum, 446

cultivation of b. anthracis, 333

Koch on tubercle bacillus, 260
Koch's blood serum, 40

glass plates. 59

leveller for plates, 59

new tuberculin, 288

tuberculin, 284

"tuberculin 0," and "R," 288
Koch- Weeks bacillus, 219
Korn's acid -fast bacillus, 279
Kraus on cholera, 461
Kruse and Pasquale on dysentery,

607
Kubel-Tiemann litmus solution,

48
Kiihne's methylene-blue, 105

modification of Gram's method,
106

Lactose fermenters, 391
Lamb on relapsing fever, 498
Laveran on malarial parasite, 585
Leishman-Donovan bodies, 630

cultivation of, 633
Leishman's opsonic technique, 121

serum method for staining try-
panosomes, 611

stain, 114

Leishman on tick fever, 502
Leishmania donovani, 630

infantum, 635

tropica, 635
Leishmaniosis, 630
Lenses, 91
Lepra cells, 299
Leprosy, 297

bacillus, 299

distribution of, 301
staining, 107, 300

diagnosis of, 304

etiology of, 302

Leprosy-like disease in rats, 30-°>
Leptothrix, 16

Lesions produced by bacteria, 181
Leucocytosis, 182, 552
Leucomaines, 187

Levaditi's collodion sac method,
503

method for staining spirocheetes,
112

on tick fever, 503

and Mclntosh on Sp. pallida,

507

Levy on streptococci, 207
Litmus media, 39

INDEX

081

Litmus solution, Kubel-Tiemann's,
48

\vlu-y, 51

Liver aliMv<s in dvH-ntcry, 60
Lockjaw, 415
l/>tll.Vs bacillus, 3!)G

methylene-blue, 104

scrum medium, 41

and Schut/c' glanders bacillus,

306

Lbsch, amoeba of, 602
Lumbar puncture, 73
Lustgarten's bacillus, 503
Lustig's anti-plague serum, 487
Lymph, vaccine, 568
Lymphangitis, 212
Ly.sa-mia in l)lackwater fever, 599
Lysogenic action of serum, 534

towards blood corpuscles, 536

MacConkey's bile-salt media, 50
medium, use of in dysentery,

386

on coli-typhoid group, 391
in examining water, 157
in paratyphoid fever, 381
MacDonald on meningitis, 244
MeFadyean on glanders, 313

methylene-blue reaction in anth-

rax> 333

Macrocytase, 553
Macrophages, 552
Madura disease, 328
Malaria, cycle in man, 586

in mosquito, 592
pathology of, 597
prevention of, 596
question of immunity against,

598
Malarial fever, examination of

blnod in, 600
malignant, 587, 595
mosquitoes in, 596
parasite, 585
inoculation of, 586
.staining of, Irishman's method,

114

Romanowsky methods, 113
varieties of, 593

Malignant oedema, bacillus of, 433
diagnosis of, 438
immunity against, 438
Malignant pustule, 342
Mallein, 314

Malta fever, 488

methods of diagnosis, 493
spread of disease, 4(.*i

Mann's method of fixing sections, 99
Manson, 584

Manteufel on relapsing fever, 49'J
Maragliano's anti-tubercular serum,

294
Marchiafava and Celli on malaria,

584
Marmorek on streptococci, 210

antistreptococcic serum, 533
Marmorek's serum media, 42

antitubercular serum, 294
Martin, C. J., on toxins, 193

on antitoxins, 532
Martin, Sidney, on alburn^, s.
etc., 194

on anthrax, 343

on diphtheria, 408
Massowah vibrio, 462
Measuring bacteria, 140
Meat extract, 32

Meat-poisoning by bacillus botu-
linus, 438

by Gaertner's bacillus, 383
Mediterranean fever, 488
Meningitis, bacteria in, 247

epidemic cerebro- spinal, 202, 242

in influenza, 470

pneumococci in, 234

posterior basal, 245
Meningococcus, 242

allied diplococci, 247

anti-sera, 246

comparison with gonococcus, 252

serum reaction, 245
Mercury perchloride as antiseptic,

171

Merozoites in malaria, 587
Metabolism, disturbances of, by

bacteria, 185

Metachromatic granules, 9
Metacoccaceae, 141
Metchnikoff on cholera in rabbits,
454

relapsing fever, 454

on syphilis, 508

MetchnikofTs phagocytosis theory,
552

spirillum, 464
Methylene-blue, 104, 105

reaction in anthrax, McFadvean,
333

682

INDEX

Methyl-violet, 101

Meyer and Ransom on tetanus

toxin, 427

Micrococci of suppuration, 202
Micrococcus, 12

of gonorrhoea, 249

melitensis, 489

pyogenes tennis, 202

tetragenus, 209

lesions caused by, 213

urese, 21

Microcytase, 553
Microphages, 552
Microscope, use of, 90
Microtomes, 97
Migula, 15

Mikulicx, cells of, 315
Milk as culture medium, 46
Minchin on trypanosomiasis,

613
Mceller's Timothy-grass bacillus,

278

Mb'ller's stain, for spores, 109
Monosaccharides, 79
Moore's medium for coli-typhoid

bacilli, 51

Morax, bacillus of, 220
Mordants, 103

Morgan's bacillus No. 1, 390
Mosquitoes, in malaria, 592, 596

role in yellow fever, 641
Moulds, media for growing, 52
Much's modification of Gram's

method, 265
Muencke's filter, 77
Miiller's bacillus, 219
Musgrave and Clegg on amoebic

dysentery, 605, 606, 608
Mycetoma, 328'
Myelocytes, neutrophile, 181

Nagana, 618

Nasgar medium, 43

Natural immunity, 555

Neelsen's stain for tubercle, 108

Negative phase in immunisation,
291, 519

Negri bodies in rabies, 576

Neisser and Wechsberg's bacteri-
cidal method, 127

Neisser's gonococcus, 249
stain for b. diphtheria, 115

Nencki, 11

Neuroryctes hydrophobice, 578

Neutral-red as indicator for media,

50
use of, 39

Avith b. coli, 353
Neutrophile leucocytes, 181

myelocytes, 181
Nicolaier, tetanus bacillus, 415
Nicolle on Leishmania infantum,

635

on Leishmania tropica, 636
on typhus fever, 648
Nicolle's modification of Gram's

method, 107

Nikati and Rietsch on cholera, 453
Nitrifying bacteria, 24
Nitroso-indol body, 82
Nordhafen vibrio, 464
Novy on relapsing fever, 495, 498
Novy and MacNeal, medium for cul-
ture of trypanosomes, 45, 612

Obermeier's spirillum, 494
(Edema, malignant, 433
Ogata's dysentery bacillus, 389
Ogston, 202
Oil, aniline, for dehydrating, etc.,

100

Oil immersion lens, 91
Ookinete, 592
Ophthalmic tuberculin reaction,

285
Opsonic action, nature of, 539

technique, 121
Opsonins, 122

absorption of, 540

in tuberculosis, 289

therrnolabile, 540

thermostable, 540
Organisms lower than bacteria, 2,

640

Oriental plague, 475
Osteomyelitis, 217
Otitis, 234, 470

Oxygen, nascent, as antiseptic, 170
Ozoena bacillus, 316

Pappataci fever, 646
Parabolic condenser, 504
Paracoccaceae, 141
Para-colon bacillus, 382
Paraffin embedding, 97
Paratyphoid bacillus, 379, 382
Park and Collins on agglutination,
-121

INDEX

683

Park ami Williams on diphtheria
tuxin, 407

. .">]»;

Passive immunity. ."•] 1, 520
Pasteur on exaltation of virulence
of liaeteria, 517

on hydrophobia, 575

nil vaccination against anthrax,
846

septio'inic «lr, 433
Pathogenicity of bacteria, 175
Peptone gelatin (sec Culture media),
36

solution, 39, 451

Periostitis, acute suppurative, 217
Peritonitis, 212, 255
iVrlsucht, 261
Pesti^ major, 481

minor, 481
Pctri's arid-last liacillus, 279

capsules, 57

sand-filter for examining air, 149
Petruschky's litmus whey, 52
Pettenkofer on cholera, 455, 462

rtcfier, 20

ITciH'er on anti-serum, 534

cholera, 457

influenxa, 466.

tvphoid, 369
Pfeiffer'a media, l:;
Pl'ciifer's phenomenon, 457, 533,

534

Phagocytes. ls-j
Phagocytosis theory of Metchnikoll',

552

Phenol broth, 154
Phenol-phthalein as indicator, 34
Phenomenon of Bordet, 534

Crul •.«•!• and Durham, 541, 542

I'fcill.-r, 457, 533, 534
Phlebotomus fever, 646
Picric acid media, 49
Pigment.-, bacterial, 10
Pipettes, 71, 116, 119, 124
Piroplasmata as causes of disease,

638

I'iroplasmosis, 637
Pitfield's flagella stain, 110
1' la^uc. bui-illus of, 475 tt seq.

Hafl'kinc's inoculation against,
486

immunity against, 485

infn-tioii in, 482

involution forms, 478

Plague, part played by rat fleas in
the spread of, 484

preventive inoculation again>t.
486

serum diagnosis, 187

stalactite growths of, 478

varieties of, 481
Plasmolysis, 9
Plate cultures, agar, 60

gelatin, 56
Platinum needles, 54
Pneumobacillus(Friedliinder's),227,

232 ct ,sv Y.

Pneumococcus (Fraenkel's), 227,
229 ft seq.

capsulation of, 230

culture methods, 43

fermentation reactions of, 231

immunity against, 239

in endocarditis, 216

lesions caused by. 233

relation to streptococci, 231

toxins of, 238
Pneumonia, bacteria in, 225

gangrenous, 470

in influenza, 469

methods of examination of, 241

septic, 225

varieties of, 224
Polar granules, 8
I Poliomyelitis, 644
Polysaccharides, 79
Positive phase in immunisation,

292, 519

Potassium permanganate as anti-
septic, 173

Potatoes as culture material, 45
Poynton and Payne on acute rheu-
matism, 220
Precipitinogen, 545
Precipitins, 544
Precipitoid, 546
Preparations, impression, 138
Protective inoculation, 578 ctscq.
Proteosoma, 594

Protozoa described in hydrophobia,
578

smallpox, 570
Protozoon malaria, 585
Prowazek on smallpox, 571

on the trypanosomata, 613
Pseudo-diphtheria bacillus, 410

-tuberculosis streptothrices, 327
Psittacosis bacillus, 384

684

INDEX

Ptomaine poisoning, 379
Ptomaines, 187
Puerperal septicaemia, 212
Pus, examination of, 94, 222
Pustule, malignant, 342
Pyaemia, 212 ct seq.

"nature of, 201

Pyogenic cocci, culture of. 133
Pyrogallate of potassium for anae-
robic cultures, 63
Pyrogallol saturated tubes, 68

Quartan fever, 594
Quarter-evil, bacillus of, 441
Quotidian fever, 593

Rabies, 573

Rabinowitch's acid-fast bacillus,

279

Rat viruses, 384
Rauschbrand bacillus, 441
Ray-fungus (actinomyces), 317
Reaction of media, standardising of,

Receptors, 549
Recovery from disease, 513
Red stains, 101
Red-water fever in cattle, 638
Reichert's gas-regulator, 86
Relapsing fever, agglutination of

spirillum, 498
i bactericidal serum in, 498

spirillum of, etc., 495

transmission of, 499

varieties of, 498

Reversibility of toxin-antitoxin re-
action, 528

Rheumatism, acute, 220
Rhinoscleroma, bacillus of, 315
Richet on anaphylaxis, 559
Ricin, 196

immunity against, 520, 525
Rivers, bacteria in, 160
Robin, 196
Rock fever, 488
Roll-tubes, Esmarch's, 60, 63
Romanowsky stains, 113
Rosenbach (bacteria in suppura-
tion), 202

Rosindol reaction (Ehrlich), 83
Ross on malaria, 585

thick film method for malarial

parasite, 600
Roux on antitoxic sera, 523, 525

Roux on syphilis, 508

and Yersin (diphtheria), 407 ct
scq.

Sabouraud's media, 52

Safranin, 101

Salt-agar as medium for b. pcstis

476

Sanarelli (typhoid fever), 356
Sanderson, Burdon, 516, 569
Saprophytes, 175
Sarcina, 14
Sausage poisoning, bacillus botu-

linus in, 438

Schaudinn on biology of trypano-
somes, 614

on amoebae of dysentery, 602,
604, 605

on morphology of spirilla, 616

on spirochaete pallida, 503

on spirillum Ziemanni, 616
Schizogony, 586
Schizomycetes, 3
Schizonts, 588
Schizophyceae, 3
Schizophyta, 3
Schiitfner's dots, 114
Sclavo's anti-anthrax serum, 347
Scorpion poison, 197
Section-cutting, 97
Sections, dehydration of, 100
Sedimentation methods, 118

test for typhoid, 371
Seitenketten, 549
Sensibilisinogen, 561
Septicaemia, nature of, 201

puerperal, 212

sputum, 225

Septicemie de Pasteur, 433
Septic pneumonia, 225
Sera, hgemolytie, 536
Serum agar, 43
Serum, agglutinative action of, 541

anaphylaxis, 558

antibacterial, 532

anti-cholera, 458

antidiphtheritic, 523

anti-plague, 487

antipneumococcic, 239

antirabic, 583

antistreptococcic, 533

antitetanic, 429

antitoxic, preparation of, 523

INDEX

685

Serum, antitubercular, 294

antityphoid, 377

bactericidal action of, 533

blood (see Culture media), 40

diagnosis, 541
methods, 118
of syphilis, 131, 510
of typhoid, 371

inspissator, 41

Ivsogenic action of, 534

towards blood corpuscles, 536
Scrum disease, 563
Serum media, 40
Serum-water media, 47
Seven-day fever, 646
Sewage, bacterial treatment of, 163

contamination of water by, 160
Shake cultures, 81
Shanghai fever, 646
Sheep-pox, 569
Shiga's bacillus, 384
Side-chain theory, Ehrlich's, 549
Sleeping sickness, 621
Slides for hanging-drops, 69
Sloped cultures, aerobic, 53

anaerobic, 68
Smallpox, 564

bacteria in, 569

Guarnieri bodies in, 570
Smegma bacillus, 280
Smith's, Lorrain, serum medium,

42
Smith, Theobald, phenomenon of,

559
Snake poisons, 197

activating of, by serum, 197

constituents of, 197

immunity against, 518
Sobernheim's anti-anthrax serum,

347
Soft sore, 257

bacillus of, 257

culture methods, 43, 258
Soil, examination of, for bacteria,

151
Soudakewitch on relapsing fever,

498

Spirilla, characters of (xee also
Vibrio), 14, 616

like cholera spirillum, 462
Sjiirillosis in animals, 4'.C>
Spirillum Metchnikovi, 464

of cholera, 447

Deneke, 465

Spirillum Duttoni, 502

Finkler and Prior, 464

Miller, 465

Obermeieri, 494, 503

relapsing fever, inoculation with,

etc., 496
Spirochsete, 15, 503, 616

gallinarum, 502

pallida, 507

staining of, 112, 115

pallidula, 511

pertenuis, 511

refringens, 504
Spirochsetes, diseases due to, 494

in syphilis, 503

in tick fever, 503

in yaws, 511

staining of, in films, 115

staining of, in sections, 112
Spironema pallidum, 503
Splenic fever, 331
Spore formation, arthrosporous, 7

endogenous, 5

in b. anthracis, 336
Spores, staining of, 109
Sporoblasts, 593
Sporocyst (malaria), 593
Sporogony (malaria), 593
Sporozoites, 586. See Schizonts
Sporulation of malarial parasite,

586
; Sputum, amoebae in, 607

influenza, 469, 472

in plague, 481

in pneumonia, 228

phthisical, 268, 281, 295

septicaemia, 225
Staining methods, 101 et seq.

of capsules, Hiss's method, 109
Welch's method, 109
Richard Muir's method, 110

of flagella, 110

of leprosy bacilli, 300

of spores, 109

of tubercle bacilli, 107

principles, 101
Stains, basic aniline, 101
Standard of immunity, ">2 1
Standardising reaction of media,

33
Staphylococci, lesions caused by

212
Staphylococcns, 12

cereus albus, 204

686

INDEX

Staphylococcus, cereus flavus, 204

pyogenes albus, 204

aureus, characters of, 202

inoculation with, 210
citreus, 202

Steam steriliser, Koch's, 28
Stegomyia f'asciata, 641
Sterilisation by heat, 27 et *cq.

at low temperatures, 30

by steam at high pressure, 29
Streptococci in diphtheria, 400

in false membrane, 212

hsemolytic action of, 207

lesions caused by, 212

varieties of, 206
Streptococcus, 12

anginosus, 207

brevis, 206

conglomerates, 206, 207

equinus, 207

erysipelatis, 218

fsecalis, 159, 207

longus, 206

mitior, 207

mucosus, 231

pneumonia, 226

pyogenes, characters of, 204
inoculation with, 222
in air, 151
in soil, 154

salivarius, 206

saprophyticus, 207
Streptothrices allied to actino-

myces, 326
Streptothrix, 16

actinomyces, 318

Streptothrix, anaerobic, in actino-
mycosis, 325

madune, 329
Subcultures, 54
Sugars, classification of, 79

fermentation of, 78

by b. coli group, 392
Sulphurous acid as antiseptic, 172
Summer diarrhoea, bacteria in, 390
Supersensitiveness, 558. See Ana-

phylaxis
Suppuration, bacteria of, 202

gonococci in, 254

methods of examination of, 222

nature of, 200

origin of, 213

pneumococci in, 234

typhoid bacillus in, 364

Symptoms caused by bacteria, 186

Syphilis, bacillus of, 503
lesions in, 505
serum diagnosis, 131, 510
spirochate pallida in, 503
transmission to animals, 508

Syringes for inoculation, 141, 142

Tabes mesenterica, 283

Tarozzi's method of anaerobic

cultures, 68
Taurocholate media, 50
Tertian fever, 594, 595
Test-tubes for cultures, 52
Tetanolysin, 424
Tetanospasmin, 424
Tetanus, 415

anti-serum of, 429, 523 et scq.
intravenous injection of, 431

cerebral, 428

dolorosus, 428

immunity against, 429

methods of examination in, 432

treatment of, 430, 547
Tetanus bacillus, 416

inoculation with, 422

isolation of, 417

spores of, 417

toxins of, 191, 423
Tetrads, 12
Texas fever, 638
Theory of exhaustion, 548

of phagocytosis, 552

of retention, 548

humoral, 548
Thermophilic bacteria, 19
Thermostable opsonins, 540
Thionin-blue, 101, 105
Thiothrix, 16
Three-day fever, 646
Tick fever, African, 494
Timothy-grass bacillus, 278
Tissues, action of bacteria on, 181

fixation of, 96
Tizzoni and Cattani on tetanus,

429

Toxalbumins, 187
Toxic action, theory of, 198
Toxicity, estimation of, 523
Toxin - antitoxin combination, re-
solution of, 527, 529
Toxins, concentrated, method of
obtaining, 194

constitution of, 549

INDEX

687

Toxins, early work on, 187

effects of, 181

immunisation by, 518

intra- and extra-cellular, 188

nature of, 193

non-proteid, 194

of anthrax, cholera, etc. (sec
Special Diseases)

production, 179

susceptibility to, 549

vegetable, 196
Toxoids, 198, 527
Trachoma, bacteria in, 219, 471
Trichophyta, media for growing,

Trophozoitea (malaria), 587
Tropical ulcer, 636
Trypanosoma cnizi, 629

gambiense, 623

L-uisi, 617, 620

noctuae, 614

of sleeping sickness, 621

ugandense, 610, 624

ngandense, relation to Tr. Gam-

biense, 628

Trvpanosomata associated with
various diseases, 610

biology of, 610, 613

culture of, 43, 45, 612

morphology of, 610

sexual cycle in, 614
Ti ypmosoniiasis, 610
Tse-tse fly disease, 618
Tubercle bacillus, 262

action of dead, 281

avian, 276

cultivation of, 265

distribution of, 270

immunity against, 290

inoculation with, 273

microscopic methods. 295

powers of resistance of, 268

in sputum, etc., 281, 295

toxins of, 284

<pecitii- reactions, _:> l

stains for, 108, 264

u'iant: cells, 268

met hod of examination of, 295
Tubercles, structure of, 268
Tubercular leprosy, 298
Tuberculin, 284, 288

" Bazillenemulsion," 288
Tuberculin, "0" and "R," 288

therapeutic application of, 290

Tuberculin reactions, 284 ct s>-q.
Tuberculosis, 260

in animals, 261

avian, 276

bovine. 274
its relation to human, 2/4

diagnosis by tuberculin, 287

in tish, 277

immune-bodies and precipitins
in, 288

immunity phenomena in, 284,
288

modes of infection, 282

precautions in diagnosis of, 281
Tubes, cultures in, 52
Typhoid bacillus, 356

biological reactions, 361

comparison with b. coli, 356

culture methods, 47, 49

distribution of, 368

epidemiology of, 369

examination for, 377

immunity against, 366

inoculation with, 365

isolation from water supplies,
378

occurrence of gallstones in,
364

serum diagnosis, 371

suppuration in, 364

toxins of, 366

vaccination against, 375, 377
Typhoid carriers, 369
Typhoid fever, 362

pathological changes in, 362
Typhus fever, 648

Ulcerative endocarditis, 216

experimental, 217

gonococci in, 256
Unit of immunity, 524
Urine, examination of, 74

staining of bacteria in, 94

tubercle bacilli in, 272, 295

typhoid bacilli in, 378
1 "-< -hinsky's medium for diphtheria
bacilli, 407

Vaccination against smallpox, ^\\

nature of, 571
against hydrophobia, 579
against typhoid, 375
for infection by pyogenic bacteria,
222

688

INDEX

Vaccines, preparation of, 133
Variola, 567 ct seq.
Vegetable poisons, 196
Venins, 197
Vibrio (see also Spirillum), 15

berolinensis, 462

of cholera, 447

Danubicus, 462

Deneke's, 465

Finkler and Prior's, 464

Gindha, 463

Ivanoff, 462

Massowah, 454, 462

Metchnikovi, 464

Nordhafen, 464

of Pestana and Bettencourt, 463

Romanus, 463
Vibrion septique, 433
Vincent's bacillus, 444
Virulence, attenuation of, 515

exaltation of, 517

of bacteria, 176
Voges and Proskauer's reaction,

353

Volpino on smallpox, 571
Von Pirquet's test, 285

Wassermann reaction in syphilis,
131, 510

in general paralysis, 510

in leprosy, 304
Water, bacteria in, 156

contamination of, by sewage,
169

examination of, 156

supplies, typhoid bacilli in, 378
Weichselbaum on pneumonia, 226
Weigert's method of dehydration,
100

modification of Gram's method,

106
Whooping-cough, bacteria in, 472

culture methods, 43, 44, 473

inoculation experiments, 474

methods of examination, 475

pathogenic effects, 474

serum reaction, 474
WTidal on serum diagnosis, 541

Widal's reaction, synonym for
agglutination of b. typhosus,
q.v., 118, 382
Williams and Lowden on Negri

bodies, 576
Winogradski, 24
Winslow and Rogers on coccacea?,

140

Winter-spring fevers, 595
Wolff and Israel's streptothrix, 326
Woodhead on tuberculosis, 283
Woody tongue, 323
Woolsorter's disease, 343
Wright's, A. E., bactericidal

method, 127
calibrated pipette, 117
diluting pipette, 71
method of counting dead bacteria,

133

opsonie technique, 122
vaccination against tuberculosis.

291
vaccination treatment of pyo-

genic infections, 222
Wright, J. H., on anaerobic

streptothrices, 324
on Leishmania tropica, 63
Romanowsky stain, 113

Xerosis bacillus, 411
Xylol, 100

Yaws, spirochsetes in, 511
Yellow fever, 639

bacteria in, 640

etiology of, 640

mosquitoes in relation to, 641
Yersin (see also Roux) on plague,

475, 486
Yersin 's anti-plague serum, 487

Ziehl-Neelsen stain, 108

Fraenkel's modification, 108
Ziemanni, spirillum, 616
Zone phenomena in agglutination,

543, 545
Zoogloea, 3
Zygote (malaria), 592

Printed by MORKISON & GIP.B LIMITED, Edinburgh

RETURN TO the circulation desk of any
University of California Library
or to the

NORTHERN REGIONAL LIBRARY FACILITY
Bldg. 400, Richmond Field Station
University of California
Richmond, CA 94804-4698

ALL BOOKS MAY BE RECALLED AFTER 7 DAYS
2-month loans may be renewed by calling

(415)642-6233
1-year loans may be recharged by bringing books

to NRLF
Renewals and recharges may be made 4 days

prior to due date

DUE AS STAMPED BELOW

65602

yJOLOGY

LIBRARY

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Books

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "Manual of bacteriology"

See other formats