Dental Library

Gift of
Joseph D. Hodgen, D.D.S.

BACTERIOLOGY

JOSEPft p. HODGEN 0.0. S,

BY THE SAME AUTHOR,

PATHOLOGY: GENERAL AND SPECIAL
For Students of Medicine.

Third Edition. 32 Plates, 15 Text-figures. 10s. 6d. net.

SERUM AND VACCINE THERAPY,

BACTERIAL THERAPEUTICS AND

PROPHYLAXIS, BACTERIAL

DIAGNOSTIC AGENTS.

Second Edition.
With 32 Figures. Crown 8vo. 7s. Gd. net.

Edited by PROF. B. T. HEWLETT, M.D., F.E.C.P.
THE CELL AS THE UNIT OF LIFE

and other Lectures delivered at the Royal Institution,
London, 1899-1902.

AN INTRODUCTION TO BIOLOGY

By the late ALLAN MACFADYEN, M.D., B.Sc.

Fullerian Professor of Physiology, Royal Institution, London.

With 16 Illustrations. 8vo. 75. 6d. net.

A MANUAL

OF

BACTERIOLOGY

CLINICAL AND APPLIED

BY
R. TAN N E R |H EWL ETT

M.D., F.R.C.P.,^¥lH.(Lond.)

Professor of Bacteriology in the University of London ;

Director of the Bacteriological Department, King's

College, London ; Director of Pathology,

Seamen's Hospital, Greenwich : Lecturer

on Bacteriology, London School

of Tropical Medicine

FIFTH EDITION

JOSEPH D. I, D. D. S.

240 STOCKTON STREET Q P^ ^

SAINT LOUIS
C. V. MOSBY COMPANY

1914

Printed in C,>e<it Britain

PREFACE TO THE
FIFTH EDITION

IN this edition while the plan of the book remains sub-
stantially the same, the text has been completely revised.
Many alterations and additions have been made in the
preliminary chapters dealing with the structure and
classification of micro-organisms and with the methods
of cultivation, isolation, and staining. The chapter on
the side- chain theory and immunity reactions has been
extended and new details respecting anaphylaxis inserted.
Considerable changes have been effected in the chapters
dealing with particular micro-organisms, e.g. the Strepto-
cocci, Anthrax, Tubercle, and Leprosy bacilli, typhoid
fever, and cholera. More space is devoted to the Moulds,
and the subjects of Sporotrichosis and Thrush have been
transferred to this section.

In the chapter dealing with the Protozoa the cultivation
of the Spirochaetes has been inserted, mention is made
of the Luetin reaction, and the account of the Wasser-
mann reaction and the description of the life history of
the rabbit Coccidium have been re-written. The pages
dealing with Hydrophobia, Vaccinia, and Malignant
Disease have been revised, and new sections on Pellagra
and Rat-bite Disease have been inserted. Further details

vi PREFACE TO THE FIFTH EDITION

are given on the purification of water, and the bacteriology
of milk and foods has been revised. In fact, hardly a
page remains quite the same as previously, and it is hoped
that the text has been made clearer and the contained
matter brought up to date.

My grateful thanks are due to my friends and colleagues,
DR. FRANK TAYLOR, upon whom the labour of the revision
of the proof-sheets has largely fallen, and MR. J. E.
BARNARD, who has furnished some new photo- micrographs.

R. T. H.

UNIVERSITY OF LONDON] KING'S] COLLEGE
May 1914

PREFACE

IN the following manual I have endeavoured to give some
account of those portions of Bacteriology which are of
especial interest in clinical medicine and hygiene. The
preparation of tissues, methods of culture, descriptions of
pathogenic organisms and their detection, the examina-
tion of water, etc., have therefore been given at some
length. As it would be impossible in the space at my
disposal to include everything relative to the subject, a
selection has had to be made, and such details as the
celloidin method, Loffler's stain for flagella, the strictly
animal parasitic diseases (with a few exceptions), etc.,
have, among others, been omitted.

At the end of the sections dealing with the pathogenic
organisms which attack man, some directions have been
given for the bacteriological clinical diagnosis and exami-
nation, but these are in no way exhaustive ; in fact, it
would not be possible in a short work to give a scheme
of examination which would cover every case. These
directions will also render the book of service in the
laboratory, while I venture to hope that the details given
in the Appendix on the use of the remedies and diagnostic
agents of bacterial origin may be of value to the
practitioner.

I have to thank ME. PEYTON BEALE, DR. LAMBERT LACK,

viii PREFACE

and ME. F. J. TANNEK, for suggestions and criticisms, and
the last-named gentleman for the aid he has freely given
me in the revision of the proof-sheets. I am also
indebted, indirectly, in many ways to my colleagues,
DR. MACFADYEN and MR. FOULERTON. My thanks are due
to MR. J. BARNARD and to MR. FRANK STRATTON respec-
tively for the photo-micrographs and original drawings,
while for the eight borrowed illustrations blocks have
been kindly lent by Messrs. BAIRD and TATLOCK, and
Messrs. SWIFT and SON.

May 1898.

CONTENTS

CHAPTER PAGE

INTRODUCTION 1

I. THE NATURE, STRUCTURE, AND FUNCTIONS
OF THE BACTERIA: THEIR CLASSIFICATION,
GENERAL BIOLOGY, AND CHEMISTRY— BAC-
TERIA AND DISEASE 8

II. METHODS OF CULTIVATING AND ISOLATING

ORGANISMS 44

III. THE PREPARATION OF TISSUES AND ORGANISMS

FOR STAINING AND MOUNTING— STAINING AND
STAINING METHODS 86

IV. METHODS OF INVESTIGATING MICROBIAL DIS-

EASES—THE INOCULATION AND DISSECTION
OF ANIMALS— HANGING-DROP CULTIVATION—
INTERLAMELLAR FILMS— THE MICROSCOPE 1 1 ;

V. INFECTION— VEGETABLE AND ANIMAL PARA-
SITES—THE INFECTIVE PROCESS— ANTI- BODIES
—ANTI-SERA AND ANTITOXINS— IMMUNITY—
OPSONINS 144

VI. SUPPURATION AND SEPTIC CONDITIONS 223

VII. ANTHRAX 251

VIII. DIPHTHERIA 265

Diphtheria in England — The Diphtheria Bacillus — The
Pseudo- Diphtheria Bacillus — Clinical Diagnosis — The
Xerosis Bacillus — Diphtheritic Affections of Birds and
Animals

IX. " ACID-FAST " BACILLI— TUBERCULOSIS— LEP-
ROSY—THE SMEGMA BACILLUS— GLANDERS 299

X. TYPHOID FEVER — PARA - TYPHOID FEVER —
BACILLUS ENTERITIDIS AND THE GART-
NER GROUP— SWINE FEVER— BACILLUS DYSEN-
TERIC—BACILLUS COLI 351

XI. BUBONIC PLAGUE— CHICKEN CHOLERA— MOUSE

SEPTICAEMIA 391

ix b

x CONTENTS

CHAPTER PAGE

XII. PNEUMONIA, INFLUENZA, AND WHOOPING- 406
COUGH

XIII. ANAEROBIC ORGANISMS — TETANUS — MALIG-

NANT (EDEMA — BACILLUS BOTULINUS —
BACILLUS WELCHII— BACILLUS CADAVERIS
SPOROGENES— BLACK QUARTER— CLOSTRI-
DIUM BUTYRICUM 419

XIV. ASIATIC CHOLERA— SPIRILLUM METCHNIKOVI

—SPIRILLUM OF FINKLER AND PRIOR— SPI-
RILLUM TYROGENUM— SPIRILLUM RUBRUM 433

XV. STREPTOTHRIX INFECTIONS— ACTINOMYCOSIS
— MYCETOMA— LEPTOTHRIX BUCCALIS— CLA-
DOTHRIX DICHOTOMA — MYCOSIS TONSIL-
LARIS 450

XVI. THE SACCHAROMYCETACE/E 461

The Yeasts — The Pathogenic Yeasts — Saccharomyces
and Torulae — Yeasts and Fermentation

XVII. THE HYPHOMYCETES — ASPERGILLOSIS — SPO-

ROTRICHOSIS— THRUSH— RINGWORM 469

XVIII. THE PROTOZOA 480

The General Structure of the Protozoa — Pathogenic
Amoebae — Trypanosomata — Leishman-Donovan Body
— Spirochaetae — Syphilis — Coccidia — Malaria

XIX. SCARLET FEVER— HYDROPHOBIA— INFANTILE
PARALYSIS— TYPHUS FEVER— YELLOW FEVER
—DENGUE— PHLEBOTOMUS FEVER— VACCINIA
AND VARIOLA— MALIGNANT DISEASE 533

XX. SOME DISEASES NOT PREVIOUSLY REFERRED
TO, WITH A DISCUSSION OF THEIR CAUSA-
TION—MICRO-ORGANISMS OF THE SKIN AND
MUCOUS MEMBRANES 556

XXI. THE BACTERIOLOGY OF WATER, AIR, AND
SOIL, AND THEIR BACTERIOLOGICAL EX-
AMINATION — SEWAGE — BACTERIOLOGY OF
MILK AND FOODS 572

Some of the Commoner Organisms found in the Air,
Water, and Soil

XXII. DISINFECTION 623

Heat — Steam Disinfection — Chemical Disinfectants —
Theory of Disinfection — Methods of determining Dis-
infectant Power

FRENCH WEIGHTS AND MEASURES AND THEIR

ENGLISH EQUIVALENTS— SOLUBILITIES 648

INDEX 649

LIST OF PLATES

PLATE

I. PHAGOCYTOSIS AND M. PYOGENE8 to face p. 228

II. STREPTOCOCCUS PYOGENES „ 232

III. THE MENINGOCOCCUS AND GONOCOCCUS „ 244

IV. ANTHRAX „ 252

V. ANTHRAX „ 254

VI. DIPHTHERIA „ 268

VII. THE HOFMANN BACILLUS AND VINCENT'S

ANGINA „ 288

VIII. THE TUBERCLE BACILLUS „ 302

IX. THE TUBERCLE BACILLUS „ 310

X. LEPROSY AND B. SMEGMAT1S „ 352

XI. B. MALLEI AND GLANDERS NODULE „ 344

XII. BACILLUS TYPHOSUS „ 354

XIII. B. TYPHOSUS AND B. COLI ., 382

XIV. PLAGUE „ 392
XV. B. PESTIS AND CHICKEN CHOLERA „ 394

XVI. DIPLOCOCCUS PNEUMONIA „ 408

XVII. B. TETANI AND B. WELCH II ,. 428
XVIII. SPIRILLUM CHOLERA AND CULTURES

OF SPIRILLA „ 434

XIX. ACTINOM YCOSIS BO VIS AND M YCETOMA 452

xii LIST OF PLATES

PLATE

XX. ACT I NO MYCOSIS HO MINIS to face p. 454

XXL TRYPANOSOMA GAMBIENSE AND
8PIROCHAETA RECURRENT1S (OBER-
MEIERI) „ 494

XXII. TREPONEMA PALLIDUM „ 496

XXflT. TREPONEMA PALLIDUM AND COCCI-

DIUM OV I FORME „ 498

XXIV. THE MALARIA PARASITE „ 520

XXV. TERTIAN "ROSETTE" AND HALTERI-

DIUM DANILEWSKYI „ 522

XXVI. PIROPLA8MA CAN IS AND HMMOCYSTI-

DIUM OF COBRA 530

ERRATUM

On p. 230, Table, Col. 1, for Micrococcus epidermis read Micrococcus
epidermidis

A MANUAL OF

BACTERIOLOGY

INTRODUCTION

BACTERIOLOGY is a branch of Biology which deals with
the study of Micro-organisms, particularly the minute
vegetable ones known as Bacteria. The scope of bacterio-
logy is difficult to define exactly, for the term is often
used in a comprehensive sense equivalent to micro-
pathology, or even micro-biology, and all investigations
connected with micro-organisms, animal and vegetable,
may be included under it. So extensive, however, has
the subject become that the animal micro-organisms are
now being studied as a separate branch, PROTOZOOLOGY.
Bacteriology deals with micro-organisms particularly in
their relation to processes — disease, fermentation, putre-
faction, and the like — while the study of their structure,
functions, and life-history belongs to the botanist and
zoologist. There is no space in this work to enter into the
history of the science, but the names of Leeuwenhoek
(1675), Miiller (1786), Schwann (1837), Cohn, Pasteur,
Lister, and Koch will ever hold an honourable place in its
annals.

The study of micro-organisms must always be of
importance in general biology, for their vital phenomena
are comparatively simple, and shed light on the more
complex processes occurring in the higher orders of living

1 I

2 INTRODUCTION

beings. Weismann based his theory of heredity on the
fundamental conception of the immortality of these
unicellular organisms. Excluding accidents, they are
immortal — they reproduce themselves by a process of
simple division, an individual dividing, and two daughter
forms taking the place of the original parent one, and
although the parent has disappeared yet there has been no
death, no dissolution ; its protoplasm or living material
is still existent in its progeny and is immortal, since this
process of reproduction apparently may go on indefinitely.
Moreover, the study of the mutability and possible trans-
formation of species of micro-organisms is likely to throw
light on the theory of evolution. Organisms such as
bacteria multiply so rapidly that fifty or sixty generations
may develop in thirty hours, a number which would take
years to attain if even the most rapid breeder among
mammals were the subject of experiment, and as they
occur in vast numbers the opportunity for variation is
extensive. These are some of the relations which micro-
organisms have with general biology.

In what may be termed the economy of nature micro-
organisms are all-important ; without them there would
be no putrefaction, no decay, and the dead remains of
animal and vegetable life would accumulate and encumber
the earth, which would become barren for the want of
the organic matter originally derived from it, but of which
there was no return. In fact the higher plants, and
indirectly, therefore, animals also, are dependent for their
existence upon the presence of bacteria in the soil, which
break up and render assimilable complex substances used
as manures.

The question of life, animal and vegetable, without
bacterial activity is an important and interesting one.
It would seem from the experiments of Duclaux1 that the
1 Comp. Rend., t. 100, p. 66,

INTRODUCTION 3

higher plants in ordinary circumstances are unable to
obtain nutriment unless the complex compounds, pro-
teins, urea, and other nitrogenous bodies, which form
the important constituents of many manures, are broken
down into simpler ones through the agency of bacteria.
He sowed seeds in sterile soil free from nitrates, nitrites,
and ammonia, which was plentifully watered with sterile
milk and solutions of sugar and starch. No changes
occurred in these substances, the seeds lost weight, and
the seedlings dwindled and died. As regards the higher
animals various views have been expressed. Pasteur
considered that their life also would probably be im-
possible without the presence of bacteria in the intestinal
tract. Nencki expressed the opinion that this idea of
Pasteur's was an erroneous one, and his experiments in
conjunction with Macfadyen and Sieber1 showed that any
considerable decomposition of the food by bacteria first
takes place in the large intestine, and that the digestive
juices alone, without the co-operation of bacteria, are able
to prepare the constituents of the food for absorption.
Nuttall and Thierfelder obtained unborn guinea-pigs by
Caesarian section with antiseptic precautions, and after-
wards kept them in a sterile environment and fed them
on sterilised food. Not only did the animals live, but
they were even in a more thriving condition than those
naturally brought up. The intestinal tract was found to
be sterile on the eighth day. Schottelius, however, found
that chickens reared on sterile food were retarded in
development, and experiments by Moro on turtle larvae
point to the same conclusion, viz. that intestinal bacteria
are necessary for normal nutrition. On the other hand,
Cohendy2 finds that chickens can be reared perfectly well
without the presence of bacteria. Levin found that the

1 Journ.. of Anat. and PhysioL, xxv, p. 390. .

2 Ann. de rinst. Pasteur, xxvi, 1912, p. 106.

4 INTRODUCTION

intestinal tract in many Arctic animals — the polar bear,
reindeer, seal, eider duck, etc. — is generally sterile, and in
these instances, therefore, bacteria are not required for
normal nutrition.

Commercially, micro-organisms are of the utmost
importance. Without them there would be no fermen-
tation, and the wine, beer, and indigo industries, the
ripening of cheese and tobacco, and many like processes
would be non-existent. From a financial aspect also
micro-organisms cannot be ignored, for many of the so-
called " diseases " of beer and wine, which often occasion
great loss, are due to the entrance of adventitious forms,
while the silk industry and sheep farming in France were
once threatened with extinction owing to the ravages of
pebrine and of anthrax respectively, but through the
genius of Pasteur were restored to their former prosperity.
There is no need to emphasise the importance of micro-
organisms frorn a medical and hygienic point of view,
but the fact may be recalled that sixty years ago the
mortality after operations was very high, and that 40 per
cent, of these deaths were caused by pyaemia, septicaemia,
and hospital gangrene, conditions which are due to the
entrance of micro-organisms, and which are now almost
preventable, thanks to the antiseptic system introduced
by Lord Lister.

The theory of spontaneous generation or abiogenesis is
intimately connected with the study of bacteria. The
putrefaction of animal and vegetable fluids even after
boiling, and the growth in them of minute living forms,
were held by many to be a sure proof of the development
of life from inanimate matter, of the spontaneous genera-
tion of the living from the non-living. A succession of
investigators, however, showed (1) that if the fluids be
boiled sufficiently long, and be then sealed up so as to
prevent the access of air, they do not undergo putre-

INTRODUCTION 5

faction ; (2) that the sealing up may be dispensed with,
provided the air be first filtered through cotton-wool before
being admitted to the flasks ; (3) that even the cotton-
wool is not needed if the air be passed slowly through a
long and tortuous channel, so as to deposit its solid
particles. Tyndall showed that putrescible fluids may be
exposed in open vessels in a closed chamber in which the
air has been undisturbed for some time and its solid
particles thereby deposited on the walls of the chamber,
which had been smeared with glycerin ; he also proved
that vegetable infusions and the like, which putrefy after
having been boiled for ten minutes, do not do so if the
boiling be repeated on two or three successive days, and
explained this by the supposition that while the fully
developed bacteria are destroyed by the first boiling,
their more resistant spores remain alive, but these on
being left for twenty-four hours germinate into the less
resistant bacterial forms, which are destroyed by the
second boiling, and by the repetition of the process com-
plete sterilisation may ultimately be obtained. It is
this process of " discontinuous sterilisation," as it is
termed, which is employed by the bacteriologist for the
preparation of sterile culture media.1

The occurrence of abiogenesis (or as he prefers to term
it, " archebiosis ") is still maintained by Bastian. He
claims that certain saline solutions which have been boiled
or even heated above the boiling-point in sealed tubes
after a time show the development of various living
organisms, including bacteria and yeasts.2

Dunbar,3 as the result of a series of experiments con-

1 The writer believes that this explanation is only partially true,
and would ascribe some of the sterilising effect of repeated heatings
simply to the injurious action of alternate heating and cooling.

2 See various papers in the Proc. Roy. Soc. Lond. ; The Evolution of
Life, Methuen, 1907 ; and Proc. Roy. Soc. Med. 1913.

3 See Journ. Roy. Inst. Pub. Health, vol. xv, No. 11, 1907, p. 679.

6 INTRODUCTION

ducted over a long period and with every care to prevent
contamination, has come to the conclusion that the
bacteria are not an independent group of organisms, but
that the bacteria, yeasts, and moulds are stages in the
life-history of green algae. The observations were carried
out both by culture methods and by microscopical examina-
tion. A culture of a single-celled alga belonging to the
Palmellacea was obtained, but by modifying the culture
medium in which a pure culture of the alga was growing,
by the addition of acid, of alkali, or of traces of copper
salts, other organisms, generally bacteria, occasionally
moulds and yeasts, and even spirochaetes, made their
appearance. Granting that there is no flaw in the experi-
mental methods, and every care seems to have been taken
to exclude contamination, etc., the results are susceptible
of another explanation, viz. that the secondary growths
were derived by transformation of the algal cells, in fact
by the phenomenon of heterogenesis which has been
claimed by Bastian to occur.

Undoubtedly bacteria exhibit variations and mutations,
not only in morphology (see p. 16) but also in function.
Thus pathogenic organisms may become non-pathogenic,
and Twort has succeeded in training B. typhosus to fer-
ment lactose, which ordinarily it does not. Some experi-
ments by Horrocks suggest that the B. typhosus may, by
symbiosis with B. coli, be converted into B. alcaligenes,
and Revis has found many variations occur with coliform
organisms as a result of cultivating in malachite green
media, etc.1

As a result of exposure of sporing anthrax to ultra-violet
rays, Mme. Henri 2 states that stable coccoid and Gram-
negative, thin filamentous forms are obtained.

1 Proc. Roy. Soc. Lond., B, vol. 85, p. 192, and vol. 86, p. 373 ; Cenlr.
/. Bakt., Abt. II, ]912 and "1913.

2 Comp. Rend. Acad. Sc., vol. 158, No. 14, 1914, p. 1032.

INTRODUCTION 7

Minchin in a presidential address to the Quekett Micro-
scopical Club points out that syngamy (sexual reproduction,
e.g. conjugation) is of the greatest importance in preserving
differentiation of species, and that without it a species
will tend to break up into races. It therefore follows
that there are no true species among organisms of the
bacterial grade, if it be true, as is usually held, that
syngamy does not occur amongst them, and the so-called
species of bacteria are to be regarded as mere races or
strains capable of modification in any direction.

While much progress has been made during the last
two or three decades, a vast amount still remains to
be done. We have only touched the fringe of the explana-
tion of the perplexing problems of susceptibility and
immunity, and of the important question of cure in, and
prevention of, infective diseases, while the chemistry of
the products of bacterial activity is still in its infancy.

The literature of Bacteriology has now become somewhat exten-
sive. In the following pages references to original papers have
been freely introduced, many of which contain a more or less full
bibliography on the subject referred to, so that further information
may be obtained if required. Kolle and Wassermann's Handbuch
der Pathogenen Mikroorganismen, ed. ii, is the most encyclopedic
work on pathological bacteriology yet published.

CHAPTER 1

THE NATURE, STRUCTURE, AND FUNCTIONS OF THE
BACTERIA: THEIR CLASSIFICATION, GENERAL BIO-
LOGY, AND CHEMISTRY— BACTERIA AND DISEASE

THE Bacteria or Schizomycetes (" fission fungi ") are
minute vegetable organisms for the most part unicellular
and devoid of chlorophyll, which multiply by simple trans-
verse division or fission ; this distinguishes them from the
yeasts, in which multiplication takes place by budding or
gemmation. A certain number of filamentous forms are
also included, serving to connect the unicellular ones with
the multicellular true fungi. The " fission plants " may
be placed in a sub-kingdom, the Schizophyta, which
may be divided into two classes : Class I, Schizophycese,
the blue-green algae, and Class II, Schizomycetes, the
bacteria.

The unicellular plants are sometimes termed the " Proto-
phyta." It must be understood that there are connecting
links between the different groups, and that there is no
sharp line of demarcation between them.

The relation of the bacteria to other lower plants is
shown in the following scheme (p. 9) :

The Bacteria have been supposed to have affinities with
the Fungi, with the Protozoa, or with the Cyanophycese.
There is little or no evidence to connect them with the first
two groups and not much with the last one, though the
resemblances here are greater. Though usually regarded
as simple forms, the Bacteria display considerable morpho-

8

STRUCTURE OF BACTERIA 9

logical and structural differentiation and physiological
complexity and are by no means primitive forms.

The size of the bacteria is variable, but they are all
microscopic, measuring 0-3/x to 30-40^ in diameter or
in length.1 Their shape likewise is very different in the
different species ; some are spherical, others ovoid, others

Relation of Bacteria to Lower Plants

Thallophyta (lower plants without fibro-vascular bundles, and with
no distinction between root and stem)

Forms with chlorophyll Forms without chlorophyll

(Algae, desmids, etc.) |

Multicellular. Spores in Unicellular. Spores frequently

differentiated cells or spore- absent. Spore-bearing cells not,

bearing organs. Generally or but slightly, differentiated.

a sexual method of reproduction Sexual reproduction usually absent

I I

The true Fungi (Eumycetes)

including moulds (Hypho- Reproduction Reproduction

mycetes) by fission by budding

I I

The Schizomycetes The Saccharomycetes
or Bacteria or Yeasts

rod-shaped or filamentous, while in some the rod or fila-
ment is twisted into a spiral. The end of the cell is
occasionally almost rectangular, but is usually more or
less rounded ; it is probably never pointed except in
the Spirochaeta?, if these be true Bacteria (see p. 18).
The bacterial cell consists of a cell -membrane enclosing
the transparent, more or less structureless living matter
or protoplasm, the cell-plasma or cytoplasm. Biitschli
has described the bacterial plasma as having a reticular
structure, but in the young cell this is probably either
an artifact or a " false image " due to faulty illumination ;
the most that can be seen is a fine granulation. The

1 p. = micron = 0-001 mm.

10 A MANUAL OF BACTERIOLOGY

protoplasm frequently contains granules composed of
fatty or protein matter, pigment, and in some species of
sulphur ; occasionally certain granules stain blue with
iodine. In some species " metachromatic " granules occur,
chiefly at the poles ; these stain red or pink with many
blue dyes, e.g. methylene blue, are composed of nucleic
acid combined with an organic base and are to be regarded
as non-living reserve material (Dobell).

In the past many have regarded the bacteria as enucleate
cells. This is probably incorrect, and Dobell finds that
all bacteria investigated possess a nucleus which may be
'in the form of discrete granules (chromidia), a filament of
variable configuration, one or more relatively large aggre-
gated masses of nuclear substance, or a system of irregularly
branched or bent short strands, rods, or networks, and
probably also in the vesicular form. The granules observed
by Rowland to take part in cell division (see below) and
staining with roseine are probably chromidia.

The cell-membrane is usually invisible, but if the cell
is treated with salt -solution (2-5 per cent.) plasmolysis
takes place, the protoplasm shrinking away from the
membrane, which then becomes visible. It can also be
stained in vivo with very dilute solutions of roseine. The
cell-membrane sometimes becomes thickened, swollen,
and gelatinous on its outer surface, forming a layer or
so-called " capsule " around the organism. The clear
spaces frequently seen around bacteria in dried and stained
preparations, especially in those from blood and lymph,
are generally artifacts and not true capsules. In Clado-
ihrix and some other forms the cell-membrane becomes
hardened, leading to the production of a firm sheath.
When bacteria assume the resting stage groups of them
adhere together in a jelly-like matrix, forming what is
known as a " zooglcea."

The chemical composition of bacteria varies much, not

BROWNIAN MOVEMENT 11

only in different species, but even in the same species
when grown on different nutrient media. All bacteria
contain proteins, lipoid substances, and salts. Bacterial
protein, according to Nencki, differs from ordinary protein
matter in not being precipitated by alcohol and in not
containing sulphur ; it was termed by him " myko-
protein." This does not appear to be the case with the
proteins obtained by grinding bacterial cells, which seem
to agree with other proteins in heat-coagulation, etc.

The proteins are mainly globulins and nucleo-proteins.
The cell wall is relatively insoluble, and generally consists
chiefly of a material like ckitin, and not of cellulose ; in
this respect bacteria resemble animal rather than vegetable
cells. Carbohydrates are generally scanty. Spores differ
from the parent cells in containing a larger proportion of
solids and less water.

All species of bacteria, but especially the smaller ones,
when suspended in a fluid exhibit what is known as.
Brownian movement, consisting of an oscillation with
some amount of rotation about a fixed point, but there
is little actual movement of translation, unless due to
flotation. This Brownian movement is physical and not
vital in origin, and occurs with all fine particles suspended
in a fluid, and must be clearly distinguished from a true
vital motility.1 Some bacteria are always motionless,
others are more or less motile, but these, too, have a
resting stage. For motility to occur the cells must be
young, and the conditions favourable to growth and
development. Motility is due to delicate protoplasmic
threads termed " flagella " connected with the outer layer
of the cell protoplasm ; these vibrate to and fro and
propel the organism through the medium. A cell will,

1 Brownian movement is due to " the incessant movements of the
molecules of the -liquid which, striking incessantly the observed par-
ticles, drive them about irregularly through the fluid " (Perrin).

12 A MANUAL OF BACTERIOLOGY

however, move indifferently in either direction ; if a
motile organism be watched it will often be seen to proceed
rapidly in one direction, stop, and then return without
turning round. The flagella are not visible in the living
state, unless dark ground illumination be used, nor by the
ordinary methods of staining, unless previously treated
with a mordant, and are extremely liable to be broken
off. They vary considerably in number and in length ;
some organisms have but a single flagellum at one pole
(monotrichic), e.g. Bacillus pyocyaneus, others have two
or more flagella forming a brush or tuft (lophotrichic),
e.g. Spirillum rubrum, while others may be almost entirely
covered with them (peritrichic), e.g. B. typhosus ; in some
the flagella are short and straight, and in others long and
twisted. The motility of organisms does not necessarily
depend directly upon the number of flagella they possess,
an organism with a few flagella often being more active
than another possessing many, and some are apparently
non-motile, though well-marked flagella can be demon-
strated. Generally speaking, however, an organism with
several flagella will be more motile than a similar form
with a few.

Darwin says : " In looking at Nature it is most necessary
never to forget that every single organic being may be
said to be striving to the utmost to increase in numbers,"
and in no group perhaps of the animal and vegetable
kingdoms is this more marked than among the
Bacteria. Reproduction is probably always non-sexual,
and takes place in two ways — by simple division or fission
and by spore formation. Schaudinn described an apparent
conjugation in one species (B. flexilis) and Nadson states
that in a few species sister cells conjugate and from this
conjugation a spore arises. Dobell, however, considers
that all the evidence is definitely against the view that a
sexual process occurs at any stage in the life -history of

REPRODUCTION OF BACTERIA 13

Bacteria. Reproduction by transverse fission is common
to all bacteria ; the bacterial cell becomes constricted at
its middle and finally separates into two parts, and thus
two young cells take the place of the parent one ; repro-
duction by fission is therefore also an increase in numbers.
The fission is always transverse, never longitudinal,1 the
rule being in cell-division that the new membrane is
formed in the most economical manner. Longitudinal
division, on the other hand, is comparatively common
among the Protozoa. Previous to division the rod-forms
become elongated and the spherical ones ellipsoidal, and
there is an increase in the number of the roseine-staining
granules, partly by division of pre-existing ones and partly
by new formation. The constriction in the majority of
cases involves and passes through one of the granules.
In the monotrichous and lophotrichous bacteria it is
always the non-flagellated end of the dividing cell which
bears the flagella of the new cell. Under favourable
conditions reproduction may be very rapid, fission occurring
every twenty or thirty minutes (Klein), so that, the increase
being in a geometrical ratio, the number of individuals
which might arise from a single bacterium in three or four
days is almost inconceivable, and would en masse weigh
thousands of tons ; fortunately there are many checks to
such a rapid multiplication. Frequently, although the
protoplasm divides, the division of the cell-membrane is
incomplete, resulting in a loose union of the cells with the
formation of a pseudo-filament. These filaments often
become much curved and twisted, forming tangled masses,
owing to fission taking place in the cells in the middle of
the filament as well as at the ends, so that the filaments
have to become curved to make room for the new cells.

1 Longitudinal division has been described in a few species, but its
occurrence is so rare that a doubt must arise as to whether these forms
are true bacteria.

14 A MANUAL OF BACTERIOLOGY

Reproduction by spore formation is met with in some
species, and is generally described as being of two kinds.
In the first, " endogenous " spore formation, a bright
refractile round or ovoid body is formed within the bacterial
cell, the development of which can be watched under the
microscope. Rowland describes the process of spore
formation as follows : Refractile, oily-looking droplets,
which do not stain with roseine, appear and ultimately
coalesce, forming the spore. The cell-plasma at the same
time diminishes and retracts from the cell-membrane.
The roseine-staining granules increase in number and
aggregate into two spherical masses, which dispose them-
selves one at each end of the cell. The cell-membrane
collapses somewhat, and, when the spore is fully formed,
ruptures transversely, leaving two cup-shaped receptacles,
in which the granules and remains of the plasma are still
recognisable. Only one spore develops in each cell, and
the spores serve to perpetuate the race when it is threatened
with extinction from adverse circumstances. Each spore
consists of a little mass of protoplasm enclosed within a
very tough and resisting membrane, which tends to pre-
serve its vitality even under unfavourable conditions ;
for spores resist the action of desiccation and germicidal
agents to a much greater degree than the fully developed
organisms. Spores vary much in size and in the position
they occupy within the bacterial cell in the different
species ; their diameter is usually about the same as that
of the cell in which they are developed, but may be much
greater, and in position they may be central or terminal,
and sometimes the spore -bearing cells are swollen or
club-shaped ; these are termed " clostridia." Endospores
are still unknown in a large number of species. The
second variety of sporulation, " arthrospore " formation,
is of doubtful occurrence, but is stated to take place as
follows : Some of the elements formed by fission are

CLASSIFICATION OF BACTERIA 15

slightly larger, more refractile, and more resisting than
their fellows, and are stated to have the properties of
spores. Placed in favourable circumstances, the spore in
either case germinates, it becomes swollen and granular,
and loses its refractile appearance ; a slight protuberance
forms, this increases in size, and an organism similar
to the parent one is finally reproduced ; the empty
spore membrane at first frequently encloses one ex-
tremity, and is afterwards cast off. In certain instances
the spore germinates without casting its membrane,
the spore membrane becoming the cell-wall of the
young organism. The ellipsoidal spores of the B.
anthracis sprout from the end, those of B. subtilis from
the side (" polar " and " equatorial " germination
respectively).

On the Morphology, etc., of the Bacteria see Dobell, Quart,
Journ. Micr. Sci., vol. 56, 1911, p. 395 (Bibliog.) ; Penau, Comp.
Rend., clii, 1911, p. 53 ; Prazmowski, Bull. Interned, de VAcad. des
Sci. de Cracovie, No. 4, B, April 1913, p. 105 (Bibliog.).

Classification of the Bacteria

Many classifications of the Bacteria have been proposed,
but none can be said to be strictly scientific, or even
satisfactory from the standpoint of convenience. A some-
what heterogenous group of organisms has undoubtedly
been described under the term Bacteria, and many forms
exist intermediate between those unicellular organisms
with and those without chlorophyll, so that a hard and
fast line cannot be drawn. Moreover, bacterial cells are
so minute that only a few broad differences can be observed
in the morphology and reproductive processes of different
species, and therefore ordinary criteria are not available
for the classification of the Bacteria.

One of the most prominent of the older classifications

16 A MANUAL OF BACTERIOLOGY

was that of Cohn. He divided the Bacteria into four
principal groups :

I. The Sphserobacteria or spherical forms.
II. The Microbacteria or short rod-forms.

III. The Desmobacteria or long rod-forms.

IV. The Spirobacteria or spiral forms.

Zopf's classification (1885) has many points to commend
it, but is largely based on the occurrence of pleomorphism.
By pleomorphism is meant a variation in the form of an
organism during its life-cycle, a coccus, for example,
growing into a rod, or a straight rod becoming a spiral.
In a peach-coloured bacterium examined by Lankester,
coccoid, rod, filamentous, and spiral forms occurred, and
the doctrine of pleomorphism received considerable support
from his work, though it may be questioned whether he
was working with pure cultures. Be that as it may, a
certain amount of pleomorphism undoubtedly occurs in
some organisms. In the colon, typhoid, and plague
bacilli, for example, the rods may sometimes be so short
as to be almost cocci, while at others they are well-marked
rods and even filaments (see also p. 6).1 The following is
an outline of Zopf's classification, the Bacteria being
divided into four principal groups or families, which again
are subdivided into smaller groups or genera :

Family I. COCCACE^E. — Spherical forms only ; division

occurs in one or more directions.

Genus 1. MICROCOCCUS (Staphylococcus). — Division
in one direction only, but irregular, so that the
cocci after division form irregular clusters.

Genus 2. STREPTOCOCCUS. — Division in one plane,
but regular, so that the cocci form chains.

Genus 3. MERISMOPEDIA (Tetracoccus). — Division
in two directions at right angles to each other,
1 See Dobell, Journ. of Genetics, if, pp. 201, 325,

CLASSIFICATION OF BACTERIA 17

but in the same plane, so that lamellae or plates
are formed.

Genus 4. SAKCINA. — -Division in three directions at
right angles to one another and in two planes, eo
that cubical masses are formed.

Genus 5. Ascococcus. — Cocci which develop in a

gelatinous matrix.
Family II. BACTERIACE^E. — Rods, straight or curved, at

some period of the life-history, though coccoid and

other forms may occur.

Genus 1. BACTERIUM. — Straight rods ; endospore
formation does not occur.

Genus 2. BACILLUS. — Straight rods ; endospore
formation occurs.

Genus 3. LEUCONOSTOC. — Cocci and rods ; arthro-
spore formation occurs in the coccoid forms.

Genus 4. CLOSTRLDIUM. — The same as bacillus, but
the spore-bearing rods are enlarged and club-
shaped.

Genus 5. SPIRILLUM. — Spiral rods ; spore forma-
tion does not occur.

Genus 6. VIBRIO. — Spiral rods ; spore formation

occurs .
Family III. LEPTOTRICHE^J. — These are unbranching

thread forms.
Family IV. CLADOTRICHE^:. — These are thread forms

showing true but not dichotomous branching.

There are many features in this classification which
are of practical value. The distinction made between a
bacterium and a bacillus, for example, is convenient.
Formerly a short rod was termed a bacterium, and a long
rod a bacillus, but such a division is an arbitrary one, and
at one stage of its life-history an organism might be a bac-
terium and at another a bacillus. The term " bacterium "

2,

18 A MANUAL OF BACTERIOLOGY

is now but little used in this sense, and any straight rod
is termed a bacillus. The term " staphylococcus " is
one frequently met with ; it is practically synonymous
with micrococcus, and refers to cocci which are aggregated
into groups or clusters. Of the twisted rods, a simple
curved rod is now known as a vibrio, a definitely cork-
screw form of three or a few turns is a spirillum, a long and
flexible twisted filament is a spirochaeta. The systematic
position of the Spirochaetse has given rise to controversy.
The parasitic ones (e.g. that of relapsing fever) are com-
monly regarded as Protozoa, but Dobell1 dissents from
this view and considers them all to be much more closely
allied to the Bacteria, which he classifies as follows :

rTrichobacteria (£occ°id*a

{itaL^saa:

v SPIRO CHAETOIDEA

\Cristispira
Saprospira

Another classification is that proposed by Migula.2
The Bacteria are divided into two orders : the Eubacteria
— bacteria proper — the cells of which contain neither
sulphur granules nor a colouring matter, bacterio-purpurin ;
and the Thiobacteria, the cells of which contain sulphur
granules and may be coloured with bacterio-purpurin.
The Eubacteria are divided into five families : (1) Coccacese,
(2) Bacteriaceae, (3) Spirillacea3, (4) Chlamydo-bacteriaceae,
and (5) Beggiatoacea3. These, again, are subdivided into
many genera, based partly on the mode of division and
partly on the number and on the arrangement of the
flagella upon the organisms. The Coccacese — globular cells
— contain the genera Streptococcus, Micrococcus, Sarcina
(non-motile), and Planococcus and Planosarcina (motile) ;
the Bacteriaceae are defined as long or short cylindrical

1 Proc. Roy. Soc. Lond., B, vol. 85, 1912, p. 186.

2 System der Bakterien, 1897.

CONDITIONS OF LIFE , 19

rods, straight and never spiral ; division in one direction
only after elongation of the rods ; and this family has
three genera : (a) Bacterium — non- flagellated cells, often
with endospore formation ; (b) Bacillus — cells possessing
both lateral and polar flagella, often with endospore forma-
tion ; (c) Pseudomonas — cells with polar flagella only,
rarely endospore formation. The Spirillacese are curved
or spiral rods, and include (a) Spirosoma, non-motile
forms, (6) Microspira, motile forms with one polar flagellum,
(c) Spirillum, motile forms with two or more polar flagella.
Various pointed organisms have been described as
" fusiform* Bacteria," e.g. in Vincent's angina (see Chap.
VIII), but Dobell expresses the opinion that these more
probably belong to the Fungi.

The nomenclature of bacterial species is at present in a chaotic
condition. In botanical and zoological nomenclature every species
has a binomial name, the first being the generic, the second the
specific name. Many bacterial species have received trinomial or
multinomial names, which should be inadmissible. The specific
name first given to an organism must stand unless it has been used
for some other species.

Conditions of Life of Bacteria

Bacteria, being living organisms, must be supplied with
suitable nutritive substances in order that their life-
processes — nutrition, reproduction, and the like — may be
carried on in a normal manner. Being devoid of chloro-
phyll they are mainly dependent upon complex organic
compounds for the carbon, hydrogen, and nitrogen which
enter into their composition, these elements being derived
for the most part from proteins and carbohydrates. Some
bacteria, however, are able to obtain the requisite nitrogen
from such comparatively simple compounds as ammonia,
ammonium carbonate, or nitrates, and one group can
make direct use of the atmospheric nitrogen. Certain

20 A MANUAL OF BACTERIOLOGY

inorganic salts, sulphates, phosphates, and sodium chloride,
also seem to be necessary for normal development. These
nutrient substances must be presented to the bacteria in
association with water, for without water bacterial activity
ceases, though in the dry state many forms, and especially
their spores, may retain their vitality for a considerable
time ; absolute desiccation, however, is rapidly fatal to
many.

Temperature is also an important factor. Though the
growth of many species occurs through a wide range,
there is for almost all an optimum at which growth is best,
and of a range not exceeding 5° or 10°. Growth usually
ceases below 10° C., but cold does not destroy bacterial
life ; after exposure to the intense cold produced by the
evaporation of liquid oxygen (— 170° C.) for weeks, or of
liquid hydrogen (— 252° C.) for ten hours, bacteria and
their spores will grow and germinate, and their chromo-
genic and pathogenic properties seem to be unaltered.1
On the other hand, bacterial growth usually ceases when
the temperature exceeds 40° C. or thereabouts, and most
bacteria without spores are destroyed within half an hour
by a temperature of 65° C. The spores are far more
resistant ; some may even be boiled for a short time
without losing their vitality, but prolonged boiling is fatal
to both bacteria and their spores. There is, however,
a group of so-called thermophilic bacteria, which thrive
best at a temperature of 60° to 70° C. They occur in the
soil and in water, and are probably of considerable import-
ance in the natural fermentations accompanied by the
evolution of heat, such as are met with in manure heaps,
the heating of hay, and firing of moist cotton.2

Free oxygen is essential to the growth of some organisms ;

1 Macfadyen and Rowland, Proc. Roy. Soc. Lond., 1900.

2 Macfadyen and Blaxall, Journ. of Path, and BacL, November 1894.
and Trans. Jenner Inst. of Prev. Med., vol. ii, 1899, p. 162.

BACTERIAL DEVELOPMENT 21

these are termed strictly aerobic. Others will not develop
in its presence, strictly anaerobic ; others, again, while
preferably aerobic or anaerobic, will grow in the absence,
or in the presence, of oxygen, and are respectively termed
facultative anaerobic or facultative aerobic. Some
organisms are strictly parasitic on animals or plants ;
others live in water, soil, decaying matter, etc. — these are
termed saprophytes ; and many are able to exist either as
parasites or as saprophytes.

Bacterial development is much influenced by the presence
of foreign substances in the nutrient medium. A number
of metallic and other salts, chlorine, bromine, and iodine,
carbolic acid, salicylic acid, etc., have an injurious effect
upon bacterial life, inhibiting or stopping growth, or
killing the organisms outright ; these are of considerable
practical importance and are known as germicides, anti-
septics, and disinfectants. The products produced in the
nutrient medium by the bacteria themselves also sooner
or later inhibit or stop further growth ; a familiar instance
of this is seen in the alcoholic fermentation of sugar by
yeast, which ceases when the amount of alcohol reaches
12 to 14 per cent. The same reason probably accounts
for the fact that growths of bacteria in culture tubes
frequently do not spread all over the surface of the nutrient
medium, and why our cultures sometimes die out more
rapidly than might be expected.

Another point affecting bacterial life is the presence of
a mixture of organisms in the same nutrient medium. If
there be a very vigorous form, it may ultimately grow
and multiply to such an extent as to crowd out and finally
kill the other forms with which it is associated, and if the
nutrient medium equally favour two species, that one
which is in an excess at the beginning may outgrow the
other. The occurrence of what has been termed symbiosis
is of considerable interest in the life of micro-organisms,

22 A MANUAL OF BACTERIOLOGY

and too little attention has hitherto been paid to it. This
is the co-existence of two or more species which together
bring about certain changes. For example, in the well-
known ginger-beer plant, Marshall Ward1 isolated several
yeasts, bacteria, and moulds ; of these, one of the yeasts
and one of the bacteria together induce the particular
changes in a saccharine fluid to which ginger has been
added, which render the mixture like ginger-beer, and
these changes do not occur unless both species develop
together.

Another extraordinary feature exhibited by bacteria is
the selective action exerted on certain substances which
contain isomerides or right- and left-handed modifications
of a substance. The Bacillus eihaceticus attacks mannitol
but not dulcitol, two alcohols which are very similar in
taste and properties and possess the same simple chemical
formula.

By a series of most brilliant researches Emil Fischer
succeeded in determining the constitution of the various
sugars, and, what is more, has produced them artificially
in the laboratory. The natural sugars are all compounds
with dissymmetric molecules, powerfully affecting the
beam of polarised light, but when prepared artificially
they are without action on polarised light, because the
artificial product consists of equal numbers of left-handed
and right-handed molecules, and the molecules of the one
neutralise the molecules of the other, thus giving rise to
a mixture which does not affect the polarised beam.

By the action of micro-organisms, however, on such an
inactive mixture the one set of molecules is sought out by
the microbes and decomposed, leaving the other set of
molecules untouched, and the latter now exhibit their
specific action on polarised light, an active sugar being
thus obtained.

1 Phil. Trans. Roy. Soc. Lond., vol. clxxxiii, 1892, p. 125.

EFFECT OF PHYSICAL AGENTS 23

Fructose was one of the principal artificial sugars pre-
pared by Fischer ; it is inactive, but consists of an equal
number of molecules of oppositely active sugars termed
" Isevulose." One set of these Isevulose molecules turns
the plane of polarisation to the right, another set to the
left — right- and left-handed Isevulose. The left-handed
Isevulose occurs in nature, while the right-handed Isevulose,
so far as is known, does not.

Now, on putting brewer's yeast into a solution of
fructose, the inactive artificial product, the yeast organisms
attack the left-handed Isevulose molecules and convert
them into alcohol and C02, while the right-handed Isevulose
is left untouched.

Pressure, unless very great, has little effect on bacteria.
Roger investigated the effects of high pressure on certain
organisms in bouillon cultures. Pressures of 200 to
250 kilos, per square centimetre had no effect ; by raising
the pressure to 3000 kilos, per square centimetre one-
third of streptococci were killed, and of anthrax without
spores a good many ; while sporing anthrax, Micrococcus
pyogenes, var. aureus, and the colon bacillus were un-
affected.1

Our countrymen Downes and Blunt first called attention
to the injurious effect of light upon bacteria. If plate
cultures be prepared and exposed to sunlight, a portion
of the plate being protected from its action, as by sticking
on a letter cut out of black paper, and the preparation
afterwards incubated, it will be found that the colonies
develop at the protected portion only, those parts which
have been exposed to sunlight remaining sterile. Although
this action of sunlight may occasionally be due to chemical
changes in the medium, resulting in the production of

1 Bacteria being so minute, the actual pressure on a bacterial cell,
even with these high pressures, is small. If, for example, a bacterium
measures 1 p. by 5 p., a pressure of 1000 kgrm. per square centimetre
\vould be but 0-05 grin, (f- grain) on the cell.

24 A MANUAL OF BACTERIOLOGY

ozone or other germicidal bodies, the experiments of
Marshall Ward and others have conclusively shown that
germicidal action may be caused by the direct action of
the light, the violet and ultra-violet rays being those
concerned, and the red end of the spectrum has no effect.
Ultra-violet rays may produce mutations of the anthrax
bacillus (see p. 6). The Rontgen rays seem to have little
or no influence upon bacteria, but the results that have
been obtained are somewhat contradictory.

The radium emanations with prolonged exposure and
near contact are germicidal to non-sporing organisms.1

Electricity, per se, has also usually little effect. When
the current is passed directly through the cultures electro-
lysis takes place, and the products formed may destroy
the bacteria ; currents of high potential, however, may
inhibit growth.

Living motile bacilli are very sensible to induced currents
of electricity, immediately orientating themselves in the
direction of the current, while dead or paralysed bacilli are
unaffected.

Bacterial Products

The chemical changes produced by micro-organisms are
chiefly analytic or destructive — the formation of simpler
from more complex bodies. This analytic faculty is
present to a marked degree in the process known as
putrefaction. Putrefaction is a term applied to the decom-
position of organic, especially protein, matter after the
death of the animal or plant. It is usually accompanied
by the evolution of foul-smelling gases and by solution
of the solid material. A large number of organisms are
concerned in this process, particularly a group to which
Hauser gave the name of Proteus. The first changes
which occur are the formation of proteoses and peptone,
1 See Green, Proc. Roy. Soc. Lond., vol. Ixxiii, 1904, p. 375.

INDOLE 25

then leucin, tyrosin, and glycocol, and basic compounds
to which the name of ptomine has been given ; next
indole, skatole, and phenol, and volatile fatty acids ; and
lastly, mercaptans, sulphuretted hydrogen, marsh gas,
ammonia, carbonic acid, and hydrogen.

In view of its practical importance in bacteriological
analysis and the identification of species, indole may here
be referred to at some length.

Indole. — Indole (C8H7N) is a product of the putrefactive
decomposition of proteins containing a tryptophane
nucleus and is formed during the growth of many organisms,
and, since one species may produce it and another allied
one may not, the determination of its presence or absence
in the culture may be of value in the identification of
organisms. The detection of indole is based on the
reaction with nitrous acid, with which it gives a fine
purplish-red coloration. In order to test for it, the
organism is grown in a fluid medium for twenty-four to
forty-eight hours or longer, 1 c.c. of a 0-1 per cent, solution
of sodium nitrite is added to every 10 c.c. of the culture,
and a few drops of pure concentrated sulphuric acid or
of hydrochloric acid are allowed to trickle slowly down the
side of the test-tube, which is inclined with its mouth
away from the operator. As the acid runs down, it is
mixed with the fluid ; a colour varying from pale pink to
pale purple indicates the presence of indole. A control
tube, uninoculated, should also be similarly tested to make
sure that the reaction is due to the products of the growth
of the organism. The culture fluid usually employed is
peptone water, preferably 2 per cent., but some samples
of " peptone " occasionally fail to give the indole reaction
when organisms are grown in media prepared from them ;
the right kind of peptone must, therefore, be used. As
the dilute solution of sodium nitrite is unstable, a stock
5 per cent, solution may be kept ; 2 c.c. of this solution

26 A MANUAL OF BACTERIOLOGY

are diluted to 100 c.c. with distilled water at the time of
making the test, and 1 c.c. of this dilution is added to
every 10 c.c. of the culture. The addition of the acid
liberates free nitrous acid, which reacts with any indole
present, and yields a pink colour. Sometimes when the
reaction is apparently absent or feeble, it may be obtained
or intensified by placing the tube in the blood-heat
incubator for half an hour. The sulphuric acid should be
pure and free from oxides of nitrogen, hence hydrochloric
acid is often preferable.

A more delicate method of testing is to run a little
hydrochloric acid down the side of the tube, so that a layer
forms at the bottom, the nitrite having been previously
added to the culture if required. A pink ring at the
juncture of the hydrochloric acid and culture indicates
the presence of indole. The pink pigment, the product
of the reaction, may be extracted by shaking with a little
amylic alcohol.

Other delicate and more certain reagents for the detection of
indole are para-dimethylamidobenzaldehyde (14 grm., dissolved in
absolute alcohol 380 c.c., hydrochloric acid 80 c.c.). To about
10 c.c. of culture 5 c.c. of this solution are added, and then 5 c.c.
of a saturated aqueous solution of potassium persulphate ; indole
gives a pink or red colour. Another test is /3-naphthaquinone-
sodium-mono-sulphonate (2 per cent, aqueous solution), which gives,
when the mixture is rendered alkaline with caustic potash, a blue
or blue -green colour or precipitate. The coloured compound may
be extracted with chloroform, in which it yields a red solution.

Peptone water is by no means a good culture medium,
and broth may therefore be employed, but it should be
free from dextrose. Peptone water with the addition of
a little rabbit's serum is perhaps the best culture medium
for the production of indole.

The presence of dextrose, saccharose, glycerin, or lactose
in quantity exceeding about 0-25 per cent, prevents the
formation of indole in broth by bacteria. Broth prepared

INDOLE REACTION 27

in the ordinary way usually contains a little dextrose
derived from the glycogen in the meat, and this probably
explains why the indole reaction is generally much more
marked in a peptone water than in a broth culture, although
the latter is a better nutrient soil. In order to prepare a
soil free from dextrose, the acid beef broth used in the
preparation of nutrient broth should be inoculated with
the colon bacillus and incubated for twenty-four hours,
and the peptone beef broth prepared from it. The
dextrose is consumed and no indole is formed.1

Some bacteria not only form indole but also produce
nitrites in the culture medium by the reduction of the
nitrates present in the peptone, etc., used in making the
nutrient medium, in which case the addition of pure
sulphuric or hydrochloric acid alone suffices to bring
out the pink indole reaction. This forms, therefore, an
additional means of distinguishing organisms, and is
employed especially for the recognition of the cholera
spirillum, which, if grown in peptone water, gives the
indole reaction (or, as it has been termed, " the cholera
red reaction ") on the addition of acid alone. The reaction
can be obtained as early as twelve hours after inoculation,
and becomes very marked in twenty-four to forty-eight
hours.

If indole is formed only in small quantities, 100 c.c.
of the culture may be distilled ; the first 20 c.c. of the
distillate will contain the bulk of the indole.

This " indole -reaction " is not necessarily always due
to indole ; the writer has shown2 that the indole-like
reaction obtained with cultures of the diphtheria and
pseudo-diphtheria bacilli is owing to the presence of
skatole-carboxylic acid. This substance is distinguished
from indole by being non- volatile. To make sure of the

1 T. Smith, Journ. of Exper. Med., vol. ii, 1897, p. 543.

2 Trans. Path. Soc. Lond., vol. lii, pt. ii, 1901, p. 113.

28 A MANUAL OF BACTERIOLOGY

presence of indole, the culture should therefore be made
alkaline with caustic soda and distilled.

Skatole (methyl indole) seems also to be formed by some organisms.
It is volatile like indole, but if a solution containing it be boiled
with an acid solution of dimethylamidobenzaldehyde (5 per cent.
in 10 per cent, sulphuric acid) it yields a blue colour, which gives a
blue solution in chloroform.

Nitrification. — Another important series of changes is
that included under the term " nitrification." As men-
tioned before, protein, albuminoid, and other complex
nitrogenous matters and urea, all of which are valuable
manures for plant life, cease to be so unless bacteria are
present.

It is necessary, in fact, for the nitrogenous matter to
be converted into nitrates, in which form alone is it avail-
able for the nutrition of plants.

Although so important, extremely small quantities of
nitrates are present in the soil ; in fertile soils, for example,
under some conditions there may be as little as one part
of nitrogen in 1,000,000, and there is often less than ten
parts. The bodies yielding nitric acid in the soil are :
(1) free nitrogen ; (2) small quantities of nitrates in rain-
water ; (3) ammonium salts, applied intentionally or
carried to the soil by rain or derived from the decay of
organic matter ; (4) various nitrogenous organic sub-
stances arising from the decay of animal and vegetable
matters.

With regard to the production of nitric acid from
nitrogenous organic matters very little was formerly
known. In 1877 Schloesing and Miintz by an ingenious
experiment showed that nitrification (as the production
of nitric acid is termed) of nitrogenous organic matter is
brought about by living organisms in the soil. Sewage
was passed continuously through a tube containing a
mixture of ignited quartz sand and limestone. After

NITRIFICATION 29

three weeks nitrates began to appear in the effluent and
increased to such an extent that finally the filtered sewage
contained no ammonia. After this had continued for
some weeks chloroform vapour was passed at the same
time through the tube, with the result that in ten days
after the introduction of the chloroform all nitrates dis-
appeared from the effluent.

Subsequently the passage of chloroform vapour was
discontinued, but nitrification did not resume until the
washings from 10 grm. of garden soil were added. Eight
days after this addition nitrates again appeared in the
effluent (this was confirmed by Warington). Evidently
the chloroform vapour acted as an antiseptic and killed
the nitrifying organisms, while the addition of soil washings
re-inoculated the material.

Shortly after this Schloesing and Miintz found that
exposure of soil to 100° C. for an hour destroyed the
power of mducing nitrification. Soils thus treated were
exposed to a current of air, purified by ignition, without
nitrification taking place ; the addition of a little un-
heated mould was, however, sufficient to cause nitrifica-
tion to recommence. They also tried seeding the sterilised
soils with various Hyphomycetes, etc., without result.

In 1884 Warington concluded that the factor deter-
mining the formation sometimes of nitric acid and some-
times of nitrous acid was a difference in the character of
the organisms ; for it is possible to have two similar
solutions under identical conditions, and for nitrites to be
produced in the one, and nitrates in the other.

In 1886 Munro showed that the process of nitrification
could take place in solutions practically destitute of
organic matter.

Nitrification in the soil takes place in three stages :

I. Ammonisation. — When complex organic compounds
such as albuminoids are applied to the land they are

30 A MANUAL OF BACTERIOLOGY

broken up ; first they become liquefied, peptone-like
bodies being produced ; these are then further acted upon
and we get alkaloidal substances in small quantity, indole,
skatole, leucin, and tyrosin and amino-acids, valerianic
acid, volatile fatty acids, lactic acid, etc.

These changes are brought about by numbers of
organisms, among which the varieties of Proteus (formerly
known as Bacterium termo), B. mesentericus, B. mycoides,
B. fluorescens liquefaciens, and B. putrificus are the more
important.

The nitrogenous compounds are then further acted upon
and ammonium salts are formed. According to Emile
Marchal, ammonisation takes place essentially under the
influence of microbes living in the upper layers of the soil.
The Bacillus mycoides is one of the most energetic of these,
and seems to play a double role, being ammonising in the
presence both of nitrogenous organic substances and of
nitrates. Urea is ammonised especially by the Micro-
coccus urece.

II. Nitrosation. — The ammoniacal salts are next con-
verted into nitrites. The nitrous organisms can probably
attack nitrogenous organic substances such as asparagine
and milk, but only feebly, milk being much more rapidly
nitrified when the nitrous organisms are mixed with other
species. The organisms bringing about this change are
short, stumpy, motile bacilli with single polar flagella
which are grouped under the generic name of Pseudo-
monas.

III. Nitratation. — These nitrites are then converted
into nitrates. The " nitric " organisms are minute non-
motile bacilli known as Nitrobacter.

Stages II and III are brought about by different species,
the nitric organisms having no effect whatever on ammonia,
but acting only after this has been oxidised into nitrous
acid by the nitrous forms.

NITRIFICATION 31

The discovery of Dr. Munro that organisms will grow in purely
inorganic solutions has been made use of for the isolation of the
different species. Solutions such as the following have been used :

For the Nitrous Organisms. For the Nitric Organisms.

Ammonium chloride, 0-5 grm. Potassium nitrite, 0-3 grin.

Potassium phosphate, 0-1 grm. Potassium phosphate, 0-1 grm.

Magnesium sulphate, 0-02 grm. Magnesium sulphate, 0-05 grm.

Calcium chloride, 0-01 grm. Calcium carbonate, 5 grm.

Calcium carbonate, 5 grm. Distilled water, 1000 c.c.
Distilled water, 1000 c.c.

These are seeded with traces of earth, and by carrying on the
cultivation for many generations a large number of organisms are
eliminated. This method does not lead to a pure cultivation, for
several forms besides the nitrifying organisms persistently maintain
themselves in these mineral solutions.

So recourse was had to gelatin plate cultivations. Although
several organisms were isolated in this manner, none of them
possessed the slightest nitrifying power.

Frankland, and later Warington (1890), succeeded in isolating
nitrous organisms by the dilution method. Nitrifying solutions
were diluted, and traces inoculated into ammoniacal solutions ; in
some of these nitrification occurred, although no growth could be
obtained on gelatin, and they were found to contain the nitrous
organism only. A little later Winogradsky isolated nitrous
organisms, first by modified gelatin plates, and afterwards by the
silica jelly method.

This is carried out as follows : Sodium carbonate is fused in the
blowpipe, and fine white sand is added so long as effervescence is
produced. The mass is allowed to cool, and is then dissolved in
water. The solution is poured into an excess of very dilute hydro-
chloric acid (silicic acid and sodium chloride being formed). The
solution is dialysed and sterilised. For use, some of this is placed
in a sterile dish and is mixed with the following solution and
inoculated :

Ammonium sulphate .... 0-4 grm.

Magnesium sulphate . . . 0-5 ,,

Di-potassium hydrogen phosphate . . 0-1 ,,

Calcium chloride ..... trace

Sodium carbonate ..... 0-6-0-9 grin.

Water 100 c.c.

This mixture sets to a jelly in five to fifteen minutes.
Winogradsky has also made use of agar for plates, but this

32 A MANUAL OF BACTERIOLOGY

medium is not so suitable as the silica jelly. A 2 per cent, aqueous
agar is prepared and poured into Petri dishes ; the film is then sown
with Proteus, and allowed to grow for seven to ten days. It is
then thoroughly washed, collected, melted, and mixed with the salts
mentioned above. The object of growing the Proteus upon it as a
preliminary is to eliminate the organic matter admixed with the agar.

Nitrification in the soil is thus brought about by two
groups of organisms. The first oxidises ammonia into
nitrous acid, and is isolated by successive cultivation in
solutions of ammonium carbonate. The second group
oxidises nitrous acid into nitric acid, and may be separated
by successive cultivations in a solution of potassium
nitrite containing a little sodium bicarbonate. In the
soil the nitric and nitrous organisms are equally active.

Besides the derivation of nitrogen from nitrogenous
compounds, the free atmospheric nitrogen is also " fixed "
through the agency of certain micro-organisms and
rendered available for plant life.

Thus, the Leguminosae are able to obtain their nitrogen
directly from the nitrogen of the air. If the roots of a
pea, bean, or vetch be examined, numerous little nodules
will be found upon them ; these contain minute irregular
and Y-shaped bodies, which have been termed " bac-
teroids," and seem to be of the nature of involution
forms. On inoculation into suitable culture media1 the
bacteroids give rise to a growth of a motile bacillus known
as Pseudomonas radicicola ; this " fixes " the atmospheric
nitrogen. The organisms penetrate the young roots
through, the root-hairs, multiply and form a filamentous
zooglcea, which grows into the tissue of the root and
penetrates the cells. Large amounts of nitrogen are taken
up by the bacteroids, and are converted into nitrogenous

1 Such as wood-ashes maltose agar. Boil 8 grm. of wood-ashes
with 500 c.c. of water for one minute ; filter. To 400 c.c. of this extract
add 4 grm. maltose and 4 grm. agar. Boil until dissolved ; filter, tube,
and sterilise.

AZOTOBACTER 33

compounds which can be assimilated by the plant. Legu-
minous plants grown from sterile seeds in a sterile soil
dwindle and die, but if inoculated with the organisms
derived from another plant of the same species growth
becomes vigorous ; if inoculated with those derived from
another species growth still takes place, but not nearly to
the same extent. The Leguminosse thus store up one of
the most important elements of plant food, and hence
their value in the rotation of crops. There is apparently
no increase of nitrogen compounds in the soil, the excess
found being due to the root residues remaining. A sub-
stance termed " nitragin," consisting of a culture of these
root organisms, has been prepared as a fertiliser. Nobbe's
" nitragin " did not prove a success, apparently because
the organisms soon lose their vitality. A better prepara-
tion, " nitro-bacterine," was devised by Moore of the
United States Department of Agriculture. Besides the
leguminous organisms, other bacteria are present in the
surface layers of the soil which fix atmospheric nitrogen.
The principal of these are ovoid organisms known as
Azotobacter. This group can be cultivated in a mannite
medium, e.g. di-potassium phosphate 0-2 grm., mannite
20 grm., water 1 litre. This may be used for isolation
by converting into an agar medium by the addition of
2 per cent. agar. Prof. Bottomley has succeeded in
obtaining a powder preparation of Azotobacter, which
retains its vitality for months, and the preparation properly
applied to poor soils produces astonishing results.

It has been found that partial sterilisation of the soil,
e.g. by heat, increases its fertility, whereas it might have
been supposed that such a procedure would decrease the
fertility by destruction of nitrogen-fixers. Russell and
Hutchinson suggest that in ordinary soil amoebae and other
protozoa devour and keep down the bacteria ; by the
sterilisation the protozoa are destroyed and the more

3

34 A MANUAL OF BACTERIOLOGY

resistant bacteria are then free to develop. Greig-Smith,1
however, denies that phagocytic protozoa possess any
power of limiting the number of bacteria in the soil, and
ascribes the effect of soil sterilisation to an action on the
bacterio-toxins and nutrients of the soil.

Besides nitrifying bacteria many de-nitrifying organisms
occur in the soil. They may (1) reduce nitrates to nitrites ;
(2) remove oxygen from nitrates and nitrites and form
ammonia ; (3) form nitrous and nitric oxides or nitrogen
from nitrates and nitrites.

Fermentation. — Another important group of changes
produced by micro-organisms is that comprised under
the comprehensive title of " fermentation," of which it is
difficult to give an accurate definition, for the distinction
between it and other chemical changes due to the activity
of micro-organisms is conventional rather than scientific.
The original conception of the term involved the occur-
rence of frothing of the fermenting liquid, owing to the
escape of gaseous products. Fermentation is brought
about by the action of ferments, two classes of which are
recognised, viz. the living or organised ferments, which,
in other words, are micro-organisms ; and the unorganised
or chemical ferments, bodies such as pepsin, which in
minute amount produce changes in a considerable quantity
of the substance acted upon, without themselves under-
going alteration.

It is better to reserve the term " fermentation " for
the changes brought about by the organised ferments or
living organisms, and to call the unorganised ferments
enzymes, and the changes which they produce zymolysis.
As fermentations are investigated more critically, the
tendency is to find that they are brought about by enzymes,
extra -cellular or intra-cellular, so that in course of time
this distinction may no longer hold good.

1 Proc. Linn. Soc. N.S.W., xxxvii, 1913, p. 655.

FERMENTATION 35

The following are the chief varieties of fermentation :
The alcoholic fermentation. — This is mainly brought
about by the decomposition of sugars of the hexose group
(C6H1206), principally dextrose and Isevulose, by yeasts
into alcohol and carbonic acid, but some of the bacteria
and moulds also produce appreciable quantities of alcohol.
Other carbohydrates by the action of enzymes secreted by
the organisms may be converted into hexoses, which are
then fermented. The general reaction is as follows :

C6H1206 = 2C2H60 + 2C02.

As a matter of fact small amounts of by-products
appear in addition to the alcohol and carbonic acid, viz.
glycerin, succinic acid, and higher alcohols. Until 1897
no enzyme had been obtained which would carry out this
change ; it only occurred when the living yeast-cells were
present, but in that year Buchner, by grinding up the
living yeast-cells, obtained a juice which decomposed
dextrose with the formation of alcohol and carbonic acid.
This " zymase " Buchner claimed to be the alcoholic
enzyme of yeast.

The lactic acid fermentation. — This is brought about
chiefly by bacteria. Hexoses are converted into lactic
acid, the reaction being

C6H1206 = 2(HC3H603),

kit it is probably not actually so simple as this, for
carbonic acid is given off at the same time. A familiar
example of this form of fermentation is the souring of
milk, in which the lactose is acted upon as follows :

C12H22On + H20 = 4C3H603.

The butyric acid fermentation. — Butyric acid is formed
from carbohydrates by the action of bacteria, mainly the
Bacillus butyricus and Clostridium butyricum, the latter an
anaerobic organism, some by-products being formed in

36 A MANUAL OF BACTERIOLOGY

addition. Milk which has been just boiled usually under-
goes the butyric rather than the lactic fermentation, the
spores of the butyric organisms surviving. Lactic acid is
first formed, and this is then converted into butyric acid :

2C3H603 - C4H802 + 2C02 + 2H

The acetic acid fermentation. — The conversion of alcohol
into acetic acid is also due to bacteria, familiar examples
of which are the souring of beer and wine.

Bacterial enzymes. — Many changes brought about by
bacteria and other micro-organisms are due to enzymes,
which may be not only intra-cellular but may escape from
the cells into the medium in which they are. The most
familiar example is the peptonising enzyme produced by
bacteria which liquefy gelatin and digest coagulated
protein, fibrin, etc. The enzymes differ : an organism
which liquefies gelatin does not necessarily digest blood-
serum. The proteolytic enzyme is tryptic in nature and
escapes from the cells into the surrounding medium, so
that some of the liquefied gelatin free from cells or in
which their action is inhibited by an antiseptic, liquefies
other gelatin if added to it. Amylolytic enzymes are
also produced, such as amylase (digesting starch), maltase,
lactase, inulase, and invertase. Lipases and rennet-like
enzymes also occur. " Fermentation " of urea takes
place by means of an enzyme secreted by the Micrococcus
ureoe, etc., with the formation of ammonium carbonate.
These enzymes do not seem to possess any poisonous
action.

Formation of pigment. — Numerous organisms, especially
those of air and water, during their growth produce various
coloured pigments. They are termed " chromogenic
bacteria," examples of which are the Sarcina lutea and
Micrococcus cereus, var. flavus, which form citron-yellow
pigments ; the Bacillus prodigiosus and Spirillum rubrumt

PIGMENT FORMATION 37

red pigments ; the Bacillus violaceus forms a rich violet
one ; and the Bacillus pyocyaneus, a blue. A large number
of chromogenic organisms require oxygen for the production
of the pigment, and potato is often the most favourable
culture medium. In some cases the medium may become
coloured, and the property of fluorescence be conferred
upon it, as is the case with the Bacillus fluorescens lique-
faciens. Usually the pigment is extra-cellular, occasionally,
as in B. violaceus, it is intra-cellular.

A group of organisms producing purplish pigments has
been described under the name of " purple bacteria."
It is doubtful if these organisms are true bacteria, and the
pigment may exercise a respiratory function analogous
to chlorophyll.

Phosphorescence, or light-production, is developed by
some bacteria, notably by many marine forms, and is
well seen in decomposing fish. Some spirilla are also
known occasionally to produce phosphorescence.

A necrotic action on the tissues is produced by many
pathogenic organisms. For example, the tubercle and
glanders bacilli cause necrosis and caseation of the sur-
rounding tissues.

Gas production. — This is common to many organisms.
The gas may consist of carbonic acid, hydrogen, or marsh
gas, and in some cases of foul-smelling sulphur compounds,
sulphuretted hydrogen, mercaptans, etc.

Sulphuretted hydrogen may be detected by the blackening of
lead acetate paper. Methyl mercaptan may be detected by
aspirating a current of air through the culture, through a calcium
chloride drying-tube, and then through a test-tube or small flask
containing isatin dissolved in concentrated sulphuric acid. The
red colour of the isatin solution is changed to olive- or grass-green
by the mercaptan.

Toxic bacterial products. — Almost without exception the
pathogenic action of bacteria is brought about by means

38 A MANUAL OF BACTERIOLOGY

of the chemical substances produced in one way or another
by their metabolic processes (see also p. 24). The toxic
bacterial products may be classified as follows :

(1) Decomposition products. — These are substances pro-
duced by the decomposition of the medium upon which
the bacteria are growing. Thus proteoses appear to be
formed by the anthrax bacillus and the pyogenic cocci.

The ptomines form another group of these substances.
These are a very important group of nitrogenous bodies,
analogous to the vegetable alkaloids and mostly solid and
crystalline in nature, which are formed by the action of
bacteria on protein and albuminoid matter. They often
occur naturally in decomposing and putrefying food,
meat, fish, etc., and as many of them are virulent poisons
they are of considerable practical import. Poisoning by
tainted food may be due to the absorption of such toxic
ptomines, and this form of food-poisoning is known as
ptomine poisoning. A number of toxic ptomines were
isolated by Brieger from cultivations of pathogenic
microbes, and great importance was once attached to
them. They are referred to in the descriptions of the
various pathogenic organisms.

Brieger 's work, however, needs revision, for his methods
were not such as to exclude alteration by the reagents
employed.

Stevenson obtained traces of a highly poisonous crys-
talline ptomine from some sardines that had caused death.
Vaughan isolated a body, tyrotoxicon, apparently identical
with diazobenzene, from poisonous cheese and milk. It
seems to be developed by the action of organisms belonging
to the B. coli or B. lactis aerogenes types. Mytilotoxin
(C6H35N02) is the specific poison of toxic mussels. Such
mussels have invariably been subjected to sewage pollu-
tion and the poison is probably produced by the action of
bacteria derived from sewage. Neurin and muscarin are

TOXINS 39

extremely poisonous and may occur in decomposing
flesh. Some of the ptomines produced by putrefaction
are very similar to certain vegetable alkaloids and are
thus of considerable medico-legal importance. The
ptomines are not specific like the true toxins, and toxic
ones may be produced by non-pathogenic bacteria.

(2) Toxins. — These are the soluble poisons elaborated
by the bacteria and excreted by the cells into the sur-
rounding medium. They are regarded by Martin and
others as being allied to the proteoses. Roux and Yersin
suggested that the diphtheria poison might be an enzyme,
while Brieger and Frankel regard it as albuminous. The
toxins are non-basic substances closely related to the
proteins and hence have been named tox-albumins, and
are considered to be the specific toxic poisons of the
pathogenic bacteria. It is difficult or impossible to prepare
them in a state of purity and their chemical constitution
is therefore unknown, and they are characterised by
extreme specificity. Such are the poisons of the diphtheria
and tetanus bacilli.

(3) Endotoxins. — -These are toxic substances elaborated
by the bacteria which do not to any extent escape from
the cells. They are as specific as the toxins and possess
analogous properties (see below).

(4) Bacterial proteins.- — These are toxic constituents of
the bacterial cells which do not diffuse from the cells, are
not specific, and which probably usually play little part
in the production of the disease symptoms.

LITERATURE

On Nitrification, see Warington, Journ. Chem. Soc., 1886, et seq. ;
Frankland, Cantor Lectures, 1892 ; Nature, 1890, et seq. ; Lohnis,
Handbuch der landwirtschaftlichen Bakteriologie (Borntraeger, Berlin,
1910, full bibliography). On Bacterial Products, see Cellular Toxins,
by Vaughan and Novy, 1902 (Bibliog.), Ueber Ptomaine, by Brieger,
1885 ; Macfadyen, The Cell as the Unit of Life (Churchill, 1908) ;

40 A MANUAL OF BACTERIOLOGY

Wells, C/nniinil J'al/toluijy, 11)07; Jlcwlell, Ail. " Toxinfl and
Antiloxins," Thorpe's Diet, of Chemistry, 1J)I.*{. Kor Cicncial
Bibliography, HOC Kollo and Wassci maim, ratkuycntn J\l ikroor-
yaniarncn.

Endotoxins

The majority of pathogenic micro -organisms do not excrete any
appreciable amount of toxin ; the toxin remains within the cells.
To such an intra-cellular toxin the. name ol' " endotoxin " has been
given. The toxins of the staphylococci and streptococci, UK-
typhoid-colon group, plague, cholera, etc., arc cndotoxins. Various
methods have been employed to prepare these endotoxins, such as
extraction of the cells by the action of weak alkalies and enzymes,
and by autolysis or self-digestion.

The late l)r. Allan Maefadyen conceived that if the infra-
oellular toxins (endotoxins) of such organisms us the typhoid
bacillus, cholera vibrio, etc., could bo obtained free 1'rom the
bacterial cells, it might be possible to prepare sera (anti-endoto.xic
sera) of much more therapeutic potency than the ordinary anti-
microbic sera.

The disintegration of the bacterial cells in the presence of intense
cold, to prevent chemical change in the bacterial juice obtained,
was the method devised by J^lacfadyen to atla.in (Ins end. With
the aid of his colleagues, Mr. Rowland and Mr. Barnard, and ol his
laboratory assistants, Messrs. Burgess and Thompson, apparatus
and methods were evolved to ollect this.

By growing on the surface of agar or other suitable medium in
plate bottles (Fig. 16), scraping oil the growth and suspending this
in salt solution, ccntrifugalising at high speed, and collecting the
bacterial cell-mass on the walls of the centrifuge vessels, suilicient
material is readily obtained to grind or triturate, and thus dis-
integrate the bacterial cells so as to liberate their contents. This
is accomplished by means of a special machine, the essential part ol
which consists of a metal cone revolving at a high speed in a metal
pot, the bottom of which is shaped so as to lit the cone. The pot,
with its contents, is immersed in a vessel of liquid air or other
freezing mixture, and the bacterial mass is ground.

After grinding, the ground material is made up with distilled
water or with 0*1 per cent, sodium hydrate, so as to form a 10 per
cent, solution (calculated on the original weight of the moist
bacterial paste); this is centrifugalised, and the lluid is liltend
through a sterile Berkefeld lilt rr.

ENDOTOXINS

42 A MANUAL OF BACTERIOLOGY

The filtrate thus obtained is the endotoxin, and is used to
immunise horses and other animals in the same manner as with
any other toxin ; it should be used as fresh as possible. The
amounts of a typhoid or cholera endotoxin employed for immunising
must at first be small, O2-0-5 c.c., as it produces considerable
disturbance on injection, and the amount is gradually increased.
After some weeks' treatment a dose of 20-30 c.c. may be injected.
When tests show that the serum has attained the necessary potency,
the horse is bled and the serum obtained and bottled.

The endotoxins also possess immunising properties to a high
degree, and may be used as prophylactic or as curative vaccines ;
they markedly raise the opsonic index.

Another machine has been devised by Barnard for disintegrating
bacterial and other cells. It is supplied by Messrs. Baker, of High
Holborn, and is depicted in Fig. 1, p. 41.

The containing vessel consists of a phosphor-bronze body, A, in
which five hardened steel balls, B, are placed. The shape of the
containing vessel is such that when these balls are at its periphery
they accurately fit the inner side of the vessel. The balls are evenly
distributed round the vessel by means of a cage, c, and during
the time they are running this cage ensures that they are equi-
distant and do not collide one with another. At the centre of the
metal vessel is a steel cone, D, which is of such a size that it keeps
the balls in their proper position in close contact with the periphery
of the containing vessel. The vessel is closed by a screw cap, E,
through which the steel cone passes, and in which it is free to
rotate. Over the whole of this a metal cylinder, r, is placed, and
is screwed down, completely sealing the upper opening in the metal
vessel. In the top of this metal cylinder a steel bearing, G, is
placed, which has freedom of movement in a horizontal direction,
but is kept down on the top of the steel cone by the action of a
spring. It therefore follows that when this metal cylinder is
screwed down the steel cone is pressed on to the balls, and the balls
are in their turn forced out to the periphery of the metal pot. The
whole appliance is mounted on a cone, H, and a centre, I, which
are carried by two uprights attached to the base plate, J ; one end
of the shaft is attached to the electric motor.

The grinding action is brought about by retarding the revolu-
tion of the central cone, D. This has been effected by mounting
on the spindle of the central steel cone, D, a semi-cylindrical mass of
iron or lead, K, the weight of which must be such that when the
whole apparatus is rotated it is sufficient to hold the central cone
still.

ENDOTOXINS 43

By retarding the cone in this way a drag is placed on the balls,
they slide to a certain extent over the inner surface of the pot and
exert a grinding action.

See Hewlett's Serum Therapy, 1910 ; Hewlett, Proc. Ray. Soc.,
B, 1909 and 1911 ; Proc. Roy. Soc. Med., vol. iii, 1909-10 (Patho-
logical Section), p. 165 ; Barnard and Hewlett, Proc. Roy. Soc., B.,
1911.

CHAPTER II

METHODS OF CULTIVATING AND ISOLATING
ORGANISMS

IT is necessary for the satisfactory study of micro-organisms
in their relation to the various processes of infection and
disease, of fermentation, putrefaction, and the like, to
separate and isolate the different species occurring in a
mixture, and, having done so, to cultivate, grow, or
propagate each species on suitable soils through successive
generations. A slight consideration will show that unless
we work with pure cultures — that is, cultures consisting
of a single species — we can never be sure that a particular
result is due to a given organism ; in a mixture several or
all of the forms present may conduce to the effect pro-
duced. With regard to the pathogenic organisms, or
disease germs, Koch laid down certain conditions which
have been termed " Koch's Postulates " (p. 147), which
must be complied with before the relation of an organism
to a disease process can be said to be completely demon-
strated, one of which is that " the organism must be
isolated and cultivated outside the animal body on suitable
media for successive generations."

In order to isolate organisms in a state of purity it is
absolutely necessary to employ vessels, instruments, and
culture media which are sterile, that is, free from any
living organisms, and to possess the means of manipu-
lating them in such a way that the entrance of organisms
from without is prevented and contamination avoided.

44

METHODS OF CULTIVATION

45

Various methods of destroying and of getting rid of
organisms are known, such as the use of chemical " germi-
cides," heat, and filtration through porous porcelain.
The addition of chemical germicides, such as carbolic acid
or corrosive sublimate, is out of the question ; for although
the vessels and media might be rendered sterile thereby,

FIG. 2. — Hot-air steriliser.

the growth of the organisms which are being investigated
would equally be prevented, so that the two last, viz. heat
and filtration, are those which are employed, the former
being used for vessels, instruments, and culture media,
solid and fluid, the latter for fluid culture media only.

Various apparatus are needed for sterilisation and the
preparation of culture media. These will now be described.

Hot-air steriliser (Fig. 2).— This is a rectangular box of
sheet iron or copper with double walls, having an air-

46 A MANUAL OF BACTERIOLOGY

space of nearly an inch between them, and furnished
with a door. The joints should be brazed, riveted, or
folded, not soldered. The outer skin at the bottom should
have a large hole cut in it in which a loose piece of sheet
iron or copper should be inserted to protect the inner
skin from oxidation and may be renewed as it " burns "

away. The top is perforated
with a couple of holes, through
one of which a chemical thermo-
meter, registering to 200° C., is
inserted in a cork, while through
the other some form of mercurial
regulator can be introduced if
required, but is not usually
needed. In the hot-air steriliser
all thin- glass vessels and cotton-
wool are sterilised by heating to
a temperature of about 150° C.
by means of a Bunsen or a small
ring burner under the steriliser,
which is supported on a suitable
iron stand. If the steriliser is
placed on a table or other
wooden support, a piece of sheet

FIG. 3.— Steam steriliser. . ,

iron, asbestos cardboard or

uralite should be laid over the wood to protect it from
the heat. An inexpensive substitute for the hot-air steri-
liser may readily be devised, any iron box or even a
biscuit- tin being used for the purpose.

Steam steriliser (Fig. 3). — This consists of a cylindrical
or rectangular vessel of tinplate, galvanised iron, or
copper, covered on the outside with a layer of felt or
asbestos, having a false perforated bottom supported a
few inches above the true bottom, and provided with a
movable lid. In the steam steriliser or " steamer " the

AUTOCLAVE

47

culture media, and thick glass vessels and other apparatus
which would crack or be damaged by the high temperature
of the hot-air steriliser, are sterilised by steam. The
lower chamber of the steamer, below the false bottom,
is partly filled with water, which is boiled by means of
a Bunsen or ring burner. Above the false bottom the
culture media or apparatus are
placed, and are sterilised by the
steam at 100° C. which fills this
space.

Here again an inexpensive sub-
stitute may be devised ; the
ordinary kitchen saucepan with
steamer will do well for many
purposes, while a " warren pot "
answers admirably.

Autoclave (Fig. 4). — This is
most useful for many purposes,
but it is expensive and not a
necessity, as the steam steriliser
will serve almost every purpose
for which the autoclave is em-
ployed with the expenditure of
a little more time and trouble.
It consists of a strong boiler of
brass or gun-metal with a removable lid, which is attached
to the boiler by means of screw-bolts. The lid is provided
with a safety valve, a gauge for indicating the pressure
and temperature, and a stopcock to relieve the pressure
if required. A small quantity of water is placed in the
bottom, and the media or apparatus to be sterilised having
been introduced, the lid is screwed down. It is heated
by means of one or more Bunsen burners, which are
turned down when the required temperature has been
reached. The temperature usually employed is about

FIG. 4. — Autoclave.

48 A MANUAL OF BACTERIOLOGY

115° to 125° C. When sterilising media care should be
taken that the vessels are not too full, and that the auto-
clave is allowed to cool down below 100° C. before opening
the stopcock, or some of the contents may be lost by
violent ebullition. While the temperature is rising, the
stopcock should always be left open until steam is being
freely generated so that the air may be expelled.

Air-pump. — An exhaust pump is useful for many pur-
poses, such as evaporating to dryness in vacuo, filtration
through porous porcelain filters, etc. Any form will serve,
but of the more elaborate ones the Fleuss pump (Fig. 5,
p. 50) made by the Pulsometer Engineering Company is
one of the best. In using it care must be taken that no
fluid or moisture gains access to the barrel ; to avoid
this the connecting pipe may be intercepted with a vessel
containing strong sulphuric acid (D, Fig. 5), over the
surface of which the exhausted air has to pass. A double-
necked Woulfe's bottle is suitable for this, the inlet and
outlet tubes extending nearly down to, but not dipping
below, the surface of the sulphuric acid.

For greasing the vessels, etc., to make air-tight joints,
beeswax dissolved in the Fleuss pump oil with the aid of
heat to a stiff paste is a good composition, or the resin
ointment of the Pharmacopoeia may be used. Another
good grease is made by melting together one part of black
rubber, one part of vaseline, and one- third part of paraffin
wax.

Centrifuge. — A small centrifuge holding two or four
10 c.c. tubes is a necessity in the laboratory. A form
driven by hand may be used, but one driven by water or
electricity is almost essential. If milk is examined, a
centrifuge driven by power and containing two or more
tubes having a capacity of not less than 70 c.c. each is
required. Many forms of centrifuges are supplied by
Messrs, Hearson,

FILTERS 49

Bell-jars with ground rims and one or two tubules are
useful for evaporation in vacuo. They should stand on a
square of thick ground glass. To make an air-tight joint
the surface of the rim of the bell- jar, which must be quite
clean, should be well greased and pressed thoroughly
home on the ground-glass plate. A thick ridge of grease
should then be plastered all round the angle formed by the
rim of the bell- jar and the glass plate. Thick rubber
pressure tubing must be used for connections, and all
joints should be well greased. For evaporating large
quantities of fluid the writer devised a copper stand with
shelves, the shelves supporting glass dishes containing
alternately strong sulphuric acid and the fluid to be
evaporated, the whole being placed under a suitable
bell- jar. A mercurial gauge is a useful addition to show
the amount of exhaust and the occurrence of leakage.
The ordinary glass filter pumps used in chemical work
and actuated by a stream of water are also useful for many
purposes.

Porous porcelain filters. — The forms which are generally
employed are the Pasteur- Chamberland, the Doulton, and
the Berkefeld. These consist of " candles " composed in
the first two of unglazed porous porcelain, in the last of a
specially prepared diatomaceous earth. The filtration
through the Pasteur- Chamberland is much slower than
through the Berkefeld. All give a germ-free filtrate, but
the last should be employed if the fluid is thick or contains
many particles ; a preliminary filtration through paper
is an advantage. A useful method of conducting filtration
is the following : The filter " candle " (B, Fig. 5, p. 50)
is connected by a short length of pressure tubing with a
piece of glass tubing passing through a rubber cork in the
neck of an ordinary filtering flask c. The " candle " is
placed in a jar A, such as a glass measure or urine- jar,
which is filled up with the solution to be filtered. The

4

50

A MANUAL OF BACTERIOLOGY

lateral branch of the filter flask is then connected with the
air-pump. On exhausting, the fluid passes through the
filter "candle" over into the filtering flask; in which it
is collected. Before use the " candle " should be well
scrubbed and some water or J per cent, carbolic run
through to clean it, and the whole may be sterilised in the

FIG. 5. — Fleuss exhaust pump, arranged for filtration.

steamer for an hour or two. After use the same process
should be repeated to cleanse it.

Flasks, beakers, and test-tubes. — A good supply of these
of various sizes is required : Erlenmeyer and ordinary
shapes, tall and short forms of beakers, etc. A few " yeast
flasks " are also useful (see Fig. 13, p. 75). Beakers and
flasks of " Jena " glass are to be preferred. Enamelled
iron ware, jugs, saucepans, mugs, etc., may replace glass
for many purposes.

The most useful size of test-tube is 5" x f " ; a fow
larger sizes and " boiling tubes " should also be kept.

Platinum needles (Fig. 6). — Two or three platinum
needles are required. They consist of about two inches

PIPETTES 51

of platinum wire in a handle of glass rod. One end of
a glass rod is softened in the Bunsen or blowpipe flame,
and about an eighth of an inch of the platinum wire is
embedded in it with a forceps, the wire having been first
heated to a red heat. The glass- wire joint is then well
annealed in the flame and allowed to cool slowly. Metal
handles may also be used. Two thicknesses of platinum
wire are desirable, viz. 04 mm. (27-28 B.W.G.) for most
purposes, but a thicker wire of about 0-7 mm. where

FIG. 6. — Platinum needles.

stiffness is required, and one or two 3 in. or more in length
are useful.

Forceps, needles, etc.— Several forceps are necessary, the
ordinary dissecting form in two or three sizes, one or two
pairs of fine pointed, two or three small brass ones, and
two or three pairs of the " Cornet " pattern. A few
ordinary sewing needles of various sizes mounted in
wooden handles serve all purposes.

Glass pipettes and capillary tubes. — These are useful for
preserving or storing blood or pus, etc., for examination,
for sterile water in making film specimens, and for many
other purposes. A blowpipe worked by a foot bellows
is required for making pipettes, etc. A piece of glass
tubing is heated in the blowpipe flame until quite soft ;
it is then taken out of the flame and the two ends are pulled
steadily apart ; this forms a capillary tube of greater or
lesser length and smaller or larger diameter, and it can be
sealed off in convenient lengths. To make a pipette
proceed in the same way : seal off the capillary tube two
or three inches from the wide tube, then heat this close

52 A MANUAL OF BACTERIOLOGY

up to where it was heated before, and draw out again
and seal off two or three inches from the bulb. In this
way a capillary tube with a bulb at its middle is formed
(Fig. 7). " Vaccine tubes," pipettes made of glass tubing
drawn out at one end, and Wright's capsules (see Fig. 35,
a and d, p. 215) are also useful.

India-rubber caps. — A few indiarubber caps for capping
test-tube or flask cultures are required. They retard

FIG. 7. — Glass pipette.

evaporation and the desiccation lof the medium, and
prevent the entrance of moulds. For use they should be
soaked in 1-500 corrosive sublimate solution; they
should not be kept pn the solution, as vulcanised rubber
absorbs mercuric chloride (Glenny and Walpole). Tinfoil,
gutta-percha tissue (sealed down by warming), paraffin
wax, sealing wax, or plasticine may also be used to cover
the tops of tubes and flasks.

Preparation of Sterile Test-tubes, Flasks, etc.,
for the Reception or Manipulation of Cul-
ture Media

To sterilise cotton-wool. — Non-absorbent cotton-wool,
best or No. 2 quality, is used for plugging purposes. The
wool should be pulled apart so as to assist the penetration
of heat ; in the compressed condition the interior is
difficult to sterilise The separated wool is placed in the
hot-air steriliser and the temperature is slowly raised to
145° C. and maintained at this for at least an hour. Above
150° C. cotton-wool becomes brown and brittle. It is a
common practice now to use various coloured wools for
the different culture media, especially the carbohydrate

GLASSWARE 53

ones, so that they are readily distinguishable by the eye.
The coloured wools may be purchased, or the ordinary
white wool may be dyed with the " Dolly " dyes.

Glass vessels. — The vessels (usually test-tubes, flasks,
and dishes) are thoroughly washed and rinsed in water,
then rinsed with 25 per cent, hydrochloric acid, and
afterwards washed well with tap-water and drained. A
final rinse with distilled water or alcohol is an advantage,
as no deposit then occurs on drying. The cleansed vessels
should be dried before sterilising, either in the air or by
placing in the hot-air steriliser for half an hour. When
dry, the vessels are plugged with a firm plug of the sterilised
cotton-wool, and are placed in the hot-air steriliser, the
temperature of which is then raised to about 150° C.
They should remain at this temperature for not less than
half an hour, after which the steriliser and its contents
are allowed to cool slowly.

Petri dishes for plate cultures, graduated pipettes, etc.,
are cleaned as described for tubes and flasks. They may
be sterilised and kept in sheet-iron or copper boxes of
appropriate size and shape.

If tubes, flasks, pipettes, etc., are required in a hurry
they may be rapidly sterilised as follows : After washing
in water they are rmsed with 5 per cent, carbolic, then
with absolute alcohol, and finally with ether, and are then
well flamed over a Bunsen flame, holding in a suitable
forceps or holder. The ether evaporates and burns at
the mouth, and when dry, a pledget of cotton-wool is held
in the forceps and singed in the flame, and, while burning,
the tube or flask is plugged with it.

When thick glass vessels, such as measures, etc./ have
to be sterilised, it is not safe to do this in the hot-air
steriliser unless the heating and cooling are carried out
very slowly, as they are very liable to crack. It is prefer-
able, after cleaning and plugging with sterile wool, to

54 A MANUAL OF BACTERIOLOGY

steam in the steam steriliser or the autoclave, the heating
and cooling being conducted slowly.

Culture Media

The ordinary methods of preparing culture media are
here given, but " Standard " media, having definite
reactions, are now largely employed (for the method of
standardisation, see p. 64). Certain special media will be
described as required. In all cases the media are filled
into the cleansed and sterilised vessels, test-tubes, flasks,
etc. (p. 53). For ordinary laboratory cultures test-tubes
are generally used. Media which are solid at ordinary
temperatures, e.g. agar, gelatin, and serum, are prepared
either as deep, upright tubes (Fig. 8, A), for which 8-15 c.c.
of the medium are required for a tube, or as sloping tubes
(Fig. 8, c), for which 4-5 c.c. are required for a tube. Of
fluid media 7-15 c.c. are used for a tube. The prepared
media having been introduced into the test-tubes, etc.,
sterilisation is effected in the steam steriliser (p. 46) by
steaming for twenty to thirty minutes on two or three
successive days, or in the autoclave (p. 47) by heating to
115°-120° C. for half an hour on one occasion. Culture
media may also be kept in bulk in flasks ; these need
somewhat longer sterilisation than tubes. Tubes of some
of the culture media can also be purchased ready for use.
Certain media can be obtained in powder form (Chopping's)
from Messrs. Baird and Tatlock, and in tabloid form
(Thompson's) from Messrs. Burroughs and Wellcome. These
are convenient when small quantities are required for
occasional use.

Acid beef-broth. — The basis of the most important
culture media, viz. peptone beef-broth, gelatin, and agar-
agar, is an infusion of meat prepared usually from beef.
In order to prepare this infusion, which may be termed

CULTURE MEDIA

55

acid beef-broth, proceed as follows : Take 1 Ib. of beef
(" gravy beef ") free from fat, chop fine or mince, add one
litre of tap-water, and allow it to simmer in a saucepan
for one hour ; cool, remove any solidified fat from the
surface, and filter through filter-paper into a clean glass
flask. If not required for immediate use, plug the neck
of the flask with cotton-wool, and
steam in the steam steriliser (or boil)
for three-quarters of an hour on two
successive days. It may then be kept
until required.

Peptone beef -broth. — Take one litre
of the acid beef-broth, add to this
10 grm. of peptone (Witte's) and 5
grm. of common salt (i.e. 1 per cent,
peptone and 0-5 per cent, sodium
chloride), mix in a flask, and steam
in the steam steriliser until dissolved.
When dissolved, remove from the
steam steriliser and render slightly
alkaline with a 10 per cent, solution

of caustic soda (preferably) or of

,. -, 77V, FIG. 8. — Tubes of culture

sodium carbonate , glazed litmus-paper media A Upright agar
being used as an indicator. Having B. Potato, c. Sloping
done this, return to the steamer for agar>
one hour, then filter through two thicknesses of German
filter-paper. It should now be quite clear and bright and
may be kept in bulk, after sterilising, or be introduced
into test-tubes, etc., and sterilised. Beef-broth, if pre-
pared in this manner, may need no clarifying, but if it
should filter at all cloudy, cool to 50° C., add the white of
an egg beaten up with the shell, and steam for half an
hour, filter, and finally sterilise as before. Other pre-
parations of peptone, e.g. Peptone Chapoteaut, may be
used.

56 A MANUAL OF BACTERIOLOGY

Instead of infusion made from meat, meat extracts are now
commonly used. The following is the composition of " Lemco "
broth :

Lemco ....... 10-20 grm.

Peptone (Witte) 10-20 grm.

Sodium chloride ..... 5-10 grm.

Water (preferably distilled) ... 1 litre

The constituents are dissolved with the aid of heat, neutralised,
clarified and filtered. Lemco may also be used to make all the
other media for which acid beef-broth is employed.

Veal-broth. — For some purposes veal presents advan-
tages over beef, e.g. for growing the tubercle bacillus.
When obtained from the butcher's the veal is frequently
powdered with flour ; this should be brushed and washed
off as completely as possible, as it renders the broth
turbid and difficult to clarify. The veal-broth is made
in precisely the same way as peptone beef-broth. It is,
however, often slightly alkaline, so that less alkali is
required for neutralisation. For the cultivation of the
tubercle bacillus about 4 to 6 per cent, of glycerin should
be added.

Glycerin beef-broth is prepared in the same manner,
4 to 6 per cent, of the best glycerin being added to the
fluid after filtration.

Glucose broth. — For the cultivation of anaerobic organisms
the addition of 0-5 to 2 per cent, of grape sugar is an
advantage. It should be added after filtration.

Egg broth.— Besredka and Jupille1 describe the com-
position of this as follows :

White of egg (10 per cent, solution) . . .4 parts
Yolk of egg (10 per cent, solution) . . .1 part
Ordinary nutrient broth . . . . .5 parts

The egg-white is beaten up with ten times its volume
of distilled water, filtered through cotton- wool, heated

1 Ann. de VInst. Pasteur, xxvii, 1913, p. 1009.

CULTUEE MEDIA 57

to 100° C., and filtered through " papier Chardin." The
liquid is tubed and sterilised at 115° C. for twenty minutes.
The yolk is beaten up with ten times its volume of distilled
water and a sufficiency of normal caustic soda solution
is added to clarify it (about 1 c.c. per 100 c.c.). It is then
treated as the egg-white. The authors recommend the
use of Martin's broth.

Peptone water. — Add to distilled or tap water 1 to 2 per
cent, of Witte's peptone and \ per cent, of common salt,
dissolve by heat, make faintly alkaline, steam for one
hour and filter.

For the cholera vibrio it is an advantage to add 1 per
cent, instead of J per cent, of common salt (Dunham's
solution).

Beer-wort. — Procure beer-wort (preferably unhopped)
from the brewery. Allow it to stand in a cool place
for twelve hours, filter, and then steam for an hour and
filter again. Fill into sterile test-tubes and sterilise.

Nutrient gelatin. — This is prepared in precisely the same
manner as peptone beef-broth with the addition of
100 grm. of the best " gold label " gelatin (Coignet's) per
litre. After the addition of the egg, steam for an hour
and then filter through two thicknesses of filter-paper in
a hot- water funnel (this is best, but it may be done in the
steamer at a low temperature, e.g. 35° C.). Fill into
test-tubes and sterilise. After the third steaming the tubes
are allowed to solidify, either in the upright or oblique
position, according as they are required for stab or surface
cultivation.

In hot summer weather 15 or even 20 per cent, of gelatin (150 grm.
or 200 grm. to the litre) are necessary for the product to remain
solid, as nutrient gelatin melts at 24° C. or a little under. Pro-
longed boiling diminishes and ultimately destroys the gelatinising
power of gelatin, so the less it is heated the better. It must not
be autoclaved.

58 A MANUAL OF BACTERIOLOGY

Glucose gelatin. — Ordinary gelatin with the addition of
1 to 2 per cent, of grape sugar.

Beer-wort gelatin. — This is one of the best culture media
for yeasts and some of the fungi (e.g. ringworm). Procure
from the brewery some beer-wort, preferably unhopped,
and add to every litre 100 grm. of gelatin. Dissolve,
clarify, and filter, as in the case of ordinary gelatin. It
is not neutralised.

Nutrient agar-agar. — This is one of our most valuable
culture media, and has the advantage over nutrient
gelatin that it remains solid at blood-heat.

Agar is a carbohydrate substance of high melting-point
and considerable gelatinising power, obtained from
Eastern seaweeds. The powdered form is now generally
used. Add 15 grm. (i.e. 1J per cent.) of powdered agar
to 1 litre of acid beef-broth, together with 10 grm. of
peptone and 5 grm. of common salt in a large glass flask,
place in the water-bath until dissolved (half an hour to
one hour), and then render alkaline as for peptone beef-
broth ; allow it to cool to 50° 0., and add the white of
an egg. Return to the steamer for an hour and a half,
then filter through an agar filter-paper (" papier Chardin ")
in a hot- water funnel or in the steamer. By this treatment
a litre of agar should pass through the filter in two to three
hours. If it does not come through clear, add another
white of egg and repeat the process.

If an autoclave is available, a quicker and better method
is, after neutralising and adding the white of an egg, to
place in the autoclave with a small beaker inverted over
the mouth of the flask, and heat to 134° C. (two atmospheres
pressure) for half an hour. Turn the gas out, and allow
to cool without opening the stopcock. When cool, open,
and filter through the special agar filter-paper in a hot-
water funnel ; the agar will pass through in about ten
minutes or a quarter of an hour. Fill into test-tubes and

CULTURE MEDIA 59

sterilise. Solidify in the upright or oblique position as
required.

In the case of bar or stick agar, first steep the agar in
1 per cent, acetic acid for a quarter of an hour, then drain
and wash it so as to thoroughly remove the acid. The
further procedure is the same as detailed above. This
yields a very clear, pale product, and is perhaps preferable
when an autoclave is not available.

Glycerin agar. — Add 4 to 6 per cent, of glycerin to
the nutrient agar after filtration and proceed as before.

Glucose agar. — One or two per cent, of grape sugar is
added to the nutrient agar after filtration.

Litmus media. — The addition of neutral litmus to the
various culture media is a useful method of demons-
trating the production of acid or of alkali by organisms.
To prepare the litmus solution take the lump litmus,
powder finely, and boil with distilled water so that a
saturated solution is obtained. Filter, and preserve in
a flask stoppered with cotton-wool, after sterilising by
boiling for half an hour on two successive days. For
some purposes a special solution of litmus, the Kubel-
Tiemann solution, which can be procured ready for use, is
employed. It must not have any antiseptic added to
it (as is sometimes done to preserve it for use in the
chemical laboratory).

Sufficient of this litmus infusion is added to the nutrient
media, after filtration, to tinge them a distinct purplish
colour. After steaming the colour has usually disappeared,
but returns as the tubes cool.

Milk. — Use separated milk, but failing this, centri-
fugalise ordinary new milk, or place it in a tall cylinder
and allow it to stand overnight in a cool place, preferably
in an ice safe. Then pipette off the milk from the bottom,
rejecting the cream. Introduce the separated milk into
test-tubes to the depth of about an inch to an inch and

60

A MANUAL OF BACTEEIOLOGY

a half and steam for one hour on two successive days.

The milk is usually tinged with litmus before tubing,

forming litmus milk.

Potatoes. — Choose sound potatoes, and scrub them well

with water to remove dirt. Cut off the ends, and with a
C^*. cork-borer, slightly smaller than the test-
tubes which are used, bore through the
potato so that a cylindrical piece is re-
moved. Push this out of the borer, and
divide it into two portions by a very
oblique transverse cut, so that two wedge-
shaped pieces are obtained, and in this
manner prepare as many pieces as there are
tubes to be filled. Place them in a basin
under the tap, and allow the water to flow
over them for about two hours. This pre-
vents the darkening of the potato in the
subsequent steaming, as does also the use
of a silver borer. The test-tubes for the
potato- wedges are prepared as follows :
After plugging and sterilising in the ordi-
nary way /introduce' a small pledget of steri-
lised wool into each, push to the bottom,
and moisten with a little sterilised distilled

^^ water. Drop the potato-wedges into the
FIG. 9. — Roux's . , i T .,. , ,

tube for potato. tubes> Plug> and sterilise by steaming for
three-quarters of an hour on two succes-
sive days (Fig. 8, B). The object of the moist wool
is to prevent drying, and for the same purpose Roux's
tubes (Fig. 9) may be used, the lower bulb being filled with
water.

Blood-serum. — Clean some glass jars of about 1 to
3 litres capacity, plug with wool, and sterilise in the
steamer for an hour on three successive days. Bleed
a horse, with aseptic precautions, and catch the blood

CULTURE MEDIA 61

in these sterilised jars. Allow the jars to stand in a
cool place for twelve hours. Then pipette off the clear
serum with a sterile pipette, and fill the sterilised test-
tubes to the depth of 2-4 cm. The tubes are then
arranged in a sloping position on the shelves of the serum
inspissator, or failing this in a hot- water oven, the tem-
perature of which should be about 50° C. At this tempera-
ture they remain for thirty hours ; it is then raised to
65° C., at which temperature the serum coagulates in from
four to six hours and the tubes are now ready for use. It
is well, however, to place them in the blood-heat incu-
bator for a night, so that any contaminating bacteria may
form colonies, and the contaminated tubes may then be
rejected.

Lqffler's blood-serum is prepared by adding one part
of glucose broth to three parts of the serum before inspissa-
tion.

The serum inspissator is practically an incubator (see
p. 68) with slightly inclined (10-15°) shelves, on which the
tubes rest, and thus the serum is coagulated in a sloping
position.

Fluid serum, etc. — Fluid blood-serum, ascitic and
hydrocele fluids, etc., are sometimes useful, and may be
used alone or mixed with peptone beef -broth in various
proportions.

Ascitic or hydrocele fluid may be obtained by using
sterile trocars, etc., and carrying out the tapping with
aseptic precautions, collecting the fluid in sterilised flasks.
It is better to collect in several small flasks than in one
large one.

Fluid blood- serum may be obtained by collecting blood
with aseptic precautions in sterilised flasks. When the
blood has coagulated and the serum separated, the serum
is pipetted off with a sterile pipette into sterile flasks.

The flasks of serum, etc., should be kept in a warm

62 A MANUAL OF BACTERIOLOGY

place for two or three days to make sure that they are
sterile, those in which a growth appears being rejected.

Serum, ascitic fluid, etc., may also be obtained sterile
by filtering through a sterilised Berkefeld filter into sterile
flasks.

Serum, ascitic and hydrocele fluids, etc., may be pre-
served in bulk and used as required. The material is
collected as aseptically as possible, 5 per cent, of chloro-
form is added, and the whole is well mixed and kept in a
cool place in the dark in a well-stoppered bottle. Sub-
sequently, during the process of sterilisation, the chloroform
is volatilised.

Serum agar (Kanthack and Stevens). — Ascitic, pleuritic,
or hydrocele fluid is collected in clean (not necessarily
sterilised) flasks, and allowed to stand overnight in a cool
place to allow the sediment or blood to deposit. The
clear fluid is then poured off, and to each litre enough
of a 10 per cent, caustic potash solution is added to render
it very distinctly alkaline — usually about 2 c.c. to every
100 c.c. of the fluid. The alkaline fluid is heated in the
autoclave for two to four hours. To this fluid 1-5 to 2 per
cent of agar is added, and the mixture is heated until the
agar dissolves. It is then filtered, introduced into test-
tubes, sterilised, and solidified in the ordinary way. The
addition of 5 per cent, of glycerin and 1 per cent, of glucose
is an advantage.

Serum agar may also be prepared by adding sterile
serum or hydrocele or ascitic fluid, warmed to 45° C.,
to sterile nutrient agar (2 to 3 per cent, agar) melted and
cooled to 45° C. Equal parts of the serum and agar may
be mixed, or 1 part of serum to 2 parts of agar.

Blood agar. — This may be prepared by smearing the
surface of the agar in sloping agar-tubes with blood
obtained aseptically from the finger or from a rabbit.
Or blood obtained aseptically may be defibrinated by

CULTUKE MEDIA 63

shaking with glass beads or with a coil of fine wire, and
the defibrinated blood, warmed to 45° C.? is added to
sterile agar liquefied by boiling and cooled to 45° C.
Haemoglobin agar may be prepared by laking defibrinated
blood by the addition of sterile distilled water and adding
to the liquid agar as before. Blood agar cannot be
sterilised after preparation, and the blood therefore must
be sterile.

Alkali albumen (Lorrain-Smith). — To 100 c.c. of fresh
serum add 1 to 1*5 c.c. of a 10 per cent, caustic soda
solution ; mix and introduce into test-tubes in the ordinary
way. Place the test-tubes in the slanting position in the
autoclave at 115° C. for twenty minutes, or in the steamer
on three successive days.

Egg cutiures (Hueppe). — These are very useful for some
purposes. A hen's egg is taken and one end sterilised by
washing with carbonate of soda solution, rinsing in sterile
water, soaking in 1-500 corrosive sublimate solution, and
washing in alcohol and in ether. A small hole is then
chipped in the shell with a sterile needle and the inocula-
tion made through this. The hole is afterwards closed
with a little sterilised wool and collodion.

Uschinsky's Fluid. Parts. Pasteur's Fluid. Parts.

Sodium chloride . . 5-7 Cane sugar ... 10

Calcium chloride . . 0-1 Tartrate of ammonia . 1

Magnesium sulphate . 0-2-0-4 The ash of 1 grm. of

Di-potassium phosphate 2-2-5 yeast ... —

Ammonium lactate . 6-7 Water .... 100

Sodium asparaginate . 3-4

Glycerin . . . 30-40

Water .... 1000

Uschinsky's fluid is a solution of known composition without
protein which can be used for investigating the chemical product
of bacteria. Pathogenic organisms grow well in it and produce
their toxins.

Pasteur's fluid is a good culture medium for yeasts, etc.1

1 Several formulae for synthesised media will be found in the Journal
of Experimental Medicine, vol. iii, p. 666.

64 A MANUAL OF BACTERIOLOGY

Standard Nutrient Media

Slight variations in the composition of the nutrient
media have a marked influence upon the characters of
the growths of micro-organisms developing upon them.
In order to obtain more uniformity for descriptive pur-
poses, etc., a committee of the American Public Health
Association drew up a scheme for the preparation of
nutrient media of approximately constant composition
and reaction. Eyre1 has devoted considerable attention
to this subject, and the following descriptions are based
largely upon his papers.

(1) Preparation of acid beef -broth.- — 1000 c.c. of distilled
water are introduced into a large flask, 500 grm. of finely
minced fresh lean beef added, and the mixture is heated
in a water-bath at 40°-45° C. for twenty minutes with
frequent agitation. It is then boiled for ten minutes,
strained, and filtered through paper. To the filtrate
sufficient distilled water is added to make up to 1000 c.c.

(2) Standardisation. — This may be most simply described
in the case of acid broth. A 100 c.c. Erlenmeyer flask is
rinsed out with boiling distilled water, 25 c.c. of the acid
beef-broth are introduced into it, and 0-5 c.c. of phenol-
phthalein solution is added (0-5 per cent, phenolphthalein in
50 per cent, alcohol). This is kept boiling and decinormal
caustic soda solution2 is run in from a 25 c.c. burette,
divided into tenths, until a faint pink tinge appears in the
boiling fluid. From the amount of soda solution used the

1 Brit. Med. Journ., 1900, vol. ii, p. 921 ; 1901, vol. ii, p. 788.

2 By a " normal " solution is meant the equivalent weight in grammes
of a substance dissolved in (i.e. made up to) a litre of water ; a " deci-
normal " solution contains one tenth of, a deka-normal ten times, this
amount. A normal solution of caustic soda contains 40 grm. of pure
NaOH (NaOH = 40), of sulphuric acid 49 grm. of pure H2S04

STANDARD MEDIA 65

amount of normal or deka- normal soda solution required
to neutralise a given volume of the acid beef-broth (e.g. a
litre) can be calculated, and this amount is then added.
Although neutral to phenolphthalein, the medium is now
strongly alkaline to litmus — too alkaline for the optimum
growth of most organisms. The reason for this is that
the di-sodium hydrogen phosphate (Na2HP04) present in
the medium is alkaline to litmus but neutral to phenol-
phthalein. To reduce the alkalinity (to litmus) normal
hydrochloric acid is then added. The American Com-
mittee recommended an acidity of -f- 1-5 — that is, to
every 100 c.c. of the medium neutral to phenolphthalein
1-5 c.c. of the normal hydrochloric acid are added. Eyre
advises a reaction of -{- 1-0 (i.e. 1 c.c. of normal hydro-
chloric to every 100 c.c.), while Chester considers that the
acidity should not exceed + 0-5. Whatever the reaction
adopted, it should be stated. Similarly, if a medium is
used which is alkaline to phenolphthalein, this is expressed
by the minus sign ; e.g. a reaction of — 1-5 indicates that
to every 100 c.c. 1-5 c.c. of normal hydrochloric acid must
be added to render it neutral to phenolphthalein, or, what
is almost (but not quite) the same thing, that to the
neutral medium 1-5 c.c. of normal caustic soda solution
have been added to every 100 c.c. Various methods are
adopted to obtain the final reaction ; the American
Committee recommend first neutralising and then adding
sufficient acid (or alkali) ; Eyre, having calculated the
acidity, adds only sufficient alkali to reduce the reaction
to the required point. Eyre describes the reaction as that
represented by the number of c.c.s of normal alkali or
acid per litre, e.g. -f 10 on Eyre's scale is equivalent to
the American -f 1-0. In making nutrient broth, agar
and gelatin, the salt and peptone and agar or gelatin are
added and dissolved, and the titration and neutralisation
are carried out as described, on the fluid medium itself,

5

66 A MANUAL OF BACTERIOLOGY

and after neutralisation the whole is heated over a water-
bath for half an hour before filtration.

The Cultivation and Isolation of Micro-
organisms

It should be clearly understood that micro-organisms
cannot usually be identified by their microscopical char-
acters alone. We can state from a microscopical examina-
tion the form of an organism, that it is a bacillus or a
micrococcus, or a sarcina, its size, that it is motile or non-
motile, sporing or non-sporing, but we cannot as a rule
go beyond this. It is necessary in most cases to ascertain
the characters of the growths of organisms on the various
culture media before species can be identified, and this is
the principal reason for having a varied assortment of
nutrient soils. It is likewise necessary for the successful
cultivation of pathogenic organisms, i.e. those connected
with disease processes and developing in or upon the
bodies of man and of animals, to maintain the cultures
at a temperature approximating to that of the host. For
this purpose some form of incubator is required. This
consists of a box or chamber of copper or iron with
double walls (Fig. 10), the space between which is filled
with water, the outside being covered with wood or felt,
or some other non-conductor. The water between the
walls is heated by means of a small burner, the gas supply
for which passes through some form of regulator inserted
in the water, so that the temperature, indicated by a
thermometer inserted through a hole in the top, can be
kept constant. The regulator is usually a mercurial one,
such as Page's or Reichert's, the principle of its action
being that as the temperature rises the mercury expands
and at a certain point cuts off the greater part of the gas
supply, only sufficient gas then passing to keep the flame

INCUBATORS

' (57

of the burner alight. This point can be varied either by
a sliding tube, in Page's, or by a screw, in Reichert's, so
that the temperature may be set at any desired point. In
Hearson's incubator, which is one of the best forms, the

FIG. 1'). — Hearson's incubator.

regulator consists of a capsule containing a fluid of a
certain boiling-point, which when ebullition takes place
raises a lever and so partially cuts off the gas supply.
While the Hearson regulator is a very constant one, it
has the disadvantage that it can only be used for a range
of temperature of a few degrees unless the capsule be

68 , A MANUAL OF BACTERIOLOGY

changed. At least one incubator is required, and it is
convenient to have two or three. If there be only one
the regulator should be set for a temperature of 37° C. ;
if more, another should be kept at about 20° C. The
incubator at 37° C. is termed the warm or blood-heat,
and that at 20° C. the cool or room temperature one. A
warm room or cupboard will serve most of the purposes
of the cool incubator. A third incubator set for 42° C.
is useful for water examination, and a fourth at 25° C. for
fermentation work.

A substitute for the large and expensive incubator can
readily be devised. An ordinary chemical hot- water oven
may be employed, or simply a smaller tin set in a some-
what larger one, the interspace being filled with water ;
and, with a little scheming, regulators can be dispensed
with by making use of a small gas or lamp flame, varying
its size and distance from the bottom until the right
temperature has been attained. Gas is a great con-
venience, but if not available, regulating oil lamps can be
obtained to take its place. Electricity has also been
adapted for heating incubators.

Gelatin will remain solid only at temperatures below
24° C., and cannot therefore be placed in the blood-heat
incubator without becoming for practical purposes a
fluid medium. Agar, however — and this is one of its
most valuable properties — does not liquefy below a
temperature of 97°r99° C., though when once liquefied
it does not set again until the temperature has fallen
to about 45° C. Gelatin is therefore usually reserved
for use at low temperatures, while agar, blood-serum,
potato, and the fluid media can be used indifferently
either at low or at high temperatures. Agar is often
a better cultivating medium than gelatin, even at low
temperatures, probably because it is so much moister.
The growths in fluid media are usually of the nature of a

INOCULATION OF MEDIA 69

general turbidity and are not particularly characteristic,
but sometimes an organism produces a film on the surface
which another similar organism does not, or the medium
remains clear, the growth forming a flocculent deposit,
thus affording a distinction. Not only do the characters
of the growths of organisms on media differ more or less,
but in some instances chemical changes occur in the media
which afford valuable information in the differentiation
of species. Thus many organisms exert a peptonising
effect on gelatin, and render it fluid sooner or later, while
others have no such action. Milk is coagulated by some
organisms, the coagulation being brought about in one
of two ways, either by the production of acids and pre-
cipitation of the caseinogen, or by the action of a rennet-
like ferment with the formation of a clot of casein. Most
organisms which liquefy gelatin coagulate milk, but the
converse is not the case. Agar is carbohydrate, not
albuminoid, in nature, and only two or three organisms
are known which liquefy it. In fluid media, such as
broth and peptone water, chemical tests can be applied,
especially for indole, which is formed by some organisms
but not by others.

Method of inoculating tubes. — The following is the
procedure by which sub-cultures are prepared from an
original test-tube or other culture : Tubes of the culture
media selected are placed in a test-tube rack. Their
mouths are then singed by holding in the Bunsen flame
for a few seconds, and with a forceps, also sterilised by
heating in the flame, the wool plugs are loosened by a
rotatory motion, and then partially withdrawn. The
mouth of the original culture- tube is similarly singed and
its plug partially withdrawn. A platinum needle is
selected and carefully straightened. The original tube
is then taken in the left hand between the thumb and
index finger with the palm upwards, and is held obliquely,

70 A MANUAL OF BACTERIOLOGY

the mouth of the tube pointing to the right, a tube of
sterile medium being held side by side with the original
culture in an exactly similar manner. The wire of the
platinum needle is then heated to redness by holding
nearly vertically in the flame, and the lower part of the
handle is also carefully heated. Holding the sterilised
needle between the finger and thumb of the right hand,
the plug of the original culture is now withdrawn by
grasping between the ring and little fingers of the right
hand, and is held there while the platinum needle is care-
fully introduced into the tube without touching the mouth
or sides, and a trace of the growth is picked up with it,
preferably from the margin. To ensure that the needle
is cool, it may first be touched on the medium where there
is no growth. The needle is quickly withdrawn without
touching the sides of the tube and the plug at once re-
placed. The plug of the sterile tube is now withdrawn
in the same manner, and the inoculated needle introduced.
If a typical surface culture is desired, a single light streak
is made with the needle from the bottom to the top of the
medium without penetrating the surface ; if an abundant
growth be required for any purpose the whole surface of
the medium may be rubbed with the needle ; if a stab
culture, the needle is plunged steadily into the centre of
the medium and withdrawn ; if a fluid one, the growth
removed is rubbed up on the side of the tube at the margin
of the fluid, and the emulsion washed down by tilting the
tube. The inoculation having been completed, the plug
is quickly replaced, and the needle is again heated in the
flame to destroy the remains of the growth upon it. If
the original culture is in a deep stab, or a fluid medium,
a looped platinum needle may sometimes be used with
advantage. The inoculations completed, the mouths of
the tubes are singed and the wool plugs pushed in level
with the lip. Before replacing the plugs each may, if

ANAEROBIC CULTURES 71

desired, for greater safety, be taken with the forceps, held
in the flame for a second or two, and pushed while
burning into the tube, and this procedure must always
be adopted if the plug be dropped or brush against
anything. If the tubes have to be kept for any length
of time, especially in the bloodheat incubator, each
should be capped with a rubber cap, tinfoil, or gutta-
percha tissue which has been soaked in 1-500 corrosive
sublimate solution.

Anaerobic cultures.- — Many organisms refuse to grow in
the presence of free oxygen, and various expedients have
to be adopted to exclude or remove it. The simplest of
all is to make the cultivation in a deep stab in glucose
agar or gelatin Narrow test-tubes filled three parts full
with the medium are best, and immediately before the
inoculation they should be placed upright in a beaker of
water, boiled for five minutes, and then cooled and
solidified in cold water. The object of this is to soften
the medium so that it does not split, as a dry medium
will, when the needle is plunged into it ; moreover, the
needle track closes up more readily, and the dissolved
oxygen is expelled. The tubes being cool, the inoculation
is made with a long thin wire, either straight or with a
closed loop at the end. It is inoculated and plunged
steadily into the centre of the medium, nearly to the
bottom, rotated, and then withdrawn, and the wool plug
is replaced and singed. The tube is then carefully heated
at the upper border of the medium so as to melt this
slightly and seal the puncture, and a well- fitting rubber cap
is applied while the tube is hot. The heating expels a
portion of the air, and, with a well-fitting cap, creates a
negative pressure within the tube, so that the residual
oxygen is not so readily absorbed, or the tubes may be
placed in a Buchner apparatus (see below). The tubes are
placed in the incubator at a suitable temperature, and

72

A MANUAL OF BACTERIOLOGY

it will be found that the most strictly anaerobic organisms
can be cultivated in this way.

When, however, an organism is required to grow
anaerobically on the surface of the medium, or in a fluid
medium, some other method must be -adopted. The tubes
may be placed under the receiver of
an air-pump and exhausted as com-
pletely as possible. This is not very
convenient, for it is difficult without great
care to maintain a vacuum, and special
receivers must be used when the cultures
have to be kept in the incubator, while
with fluid media ebullition may cause
considerable difficulty.

For fluid cultures Hamilton's method
is the simplest of all. The fluid in the
tubes is covered with a layer of olive oil
1-2 cm. thick, and the tubes are then
sterilised. The layer of oil prevents the
access and entrance of oxygen. The only
disadvantage is that the inoculation, or
the withdrawal of culture, must usually
be performed with a sterile glass pipette ;
FIG. 11.— Buchner's ^ a wire needle be used the material is
tube arranged for very liable to be detached in the oil.
vat^n.^ Another method (Buchner's) is that

usually adopted, and consists in absorb-
ing the oxygen by means of alkali and pyrogallic
acid, and so cultivating in an atmosphere of nitrogen.
This can be carried out in two ways — either in a wide-
mouthed bottle with well-fitting glass stopper, sufficiently
large to contain the test-tubes, or in a Buchner's tube.
For the first the inoculated culture tubes are placed
in the bottle, into which a few cubic centimetres of a
strong aqueous solution of pyrogallic acid have previously

ANAEROBIC CULTIVATION V\

been poured. By means of a thistle funnel, an equal
volume of 20 per cent, caustic potash1 or soda solution
is then added. As quickly as possible the thistle funnel
is withdrawn without mixing the solutions, and the stopper,
well vaselined, inserted and twisted well home, and some
melted paraffin may be poured all round the joint and
melted in with a hot iron. The solutions in the bottle are
now well mixed, and the whole is placed in a suitable
incubator. The Buchner's tube (Fig. 11) is convenient
for single test-tube cultures. It consists of a strong glass
test-tube, large enough to take an ordinary test-tube, and
having a constriction about an inch and a half from the
bottom. The constriction supports the test-tube culture,
while the mixture of pyrogallic acid and .caustic potash
fills the portion below the constriction. A well-fitting
rubber cork closes the mouth of the tube, and the joint
may be paraffined for additional security. If a Buchner's
tube is not available, the cotton- wool plug of the culture
tube may be pushed into the tube for an inch, some solid
pyrogallol is placed on the wool plug, this is just moistened
with caustic potash solution and the tube is stoppered
with a rubber cork.

The displacement of the atmosphere by means of
hydrogen may be adopted, and is to be preferred for fluid
cultures. Hydrogen does not seem to inhibit the growth
of any anaerobic organisms, whereas carbon dioxide gas,
which might be still more conveniently used, has a very
decided inhibitory action on some species. The hydrogen
is best generated from zinc and sulphuric acid in a Kipp
apparatus, or the compressed gas in cylinders, or even
coal-gas, may be used. Care must be taken that all

1 Thirty-two grm. of pyrogallic acid and 64 grin, of caustic potash
dissolved in 100 c.c. of water will absorb 9200 c.c. of oxygen. At the
same time some carbon monoxide is evolved (122-5 c.c.). The evolu-
tion of CO is a minimum when the potash is in excess and only one-
lifth or the theoretical absorbable amount of 0 is absorbed.

74

A MANUAL OF BACTERIOLOGY

joints are tight, and they may be paraffined with advan-
tage. The gas should be passed through a strong solution
of caustic potash, and may be passed through some
alkaline pyrogallic acid if the most rigorous condition of
anaerobiosis is desired, but for ordinary purposes this is
not essential ; it should also pass
through two or three fairly firm plugs
of cotton- wool to remove organisms ;
these must be dry, for if moist the
passage of the gas may be stopped.

For tube cultures Frankel's method
may be adopted (Fig. 12). The broth
or gelatin is introduced into a large
strong test-tube which is plugged with
a rubber cork, through which two
pieces of glass tubing pass, one to the
bottom of the tube, the other just
through the cork. Outside the cork
these tubes are bent over at right
angles, and each is drawn slightly out
so as to contract its lumen at about
the middle. The long tube is con-
FIG. l2.-^Frankel's tube nected with the hydrogen supply, and
for anaerobic cultiva- a current of the gas is passed through
and escapes by the shorter tube. After
the gas has been passing for twenty minutes to half an hour,
and all oxygen has been expelled, the distal, i.e. shorter, tube
is sealed off at the contracted portion in the Bunsen or
blowpipe flame, and then the proximal or longer one in
the same manner. The rubber cork must, of course, fit
well, and the joints should be paraffined. If gelatin be
the medium, it should be kept fluid in a bath of warm
water while the hydrogen is passing.

For broth or other fluid cultures, which are essential
for obtaining toxic products, flasks are used which are

ANAEROBIC CULTIVATION

75

fitted up like the Frankel tube described above. The ends
of the tubes are plugged with cotton- wool, and the whole
— flask, cork, tubes and medium — is sterilised. The
medium is inoculated from a recent culture by momentarily
removing the cork. Hydrogen is then passed through
from a Kipp apparatus, the
long tube being connected with
the hydrogen supply. After
passing for about half an hour,
the tubes are sealed off and the
flask is incubated. For con-
venience of sealing the tubes
should be drawn out slightly.

As many organisms produce
gas during their growth, it may
be necessary to provide for its
escape, or the flasks may burst
owing to the pressure. This can
be done by adjusting a mercury
valve, and may be carried out
in a simple manner by a method
devised by the writer. " Yeast
flasks," which can be obtained
in various sizes, are made use
of, and are filled three parts full
with a 2 per cent, grape-sugar
bouillon. The neck is corked

With a perforated rubber cork FIG. 13.-Yeast flask arranged

for anaerobic cultivation.
(A, Fig. 13), through which a

glass tube, B, passes to the bottom of the flask, projecting
two inches above the rubber cork and here plugged with

JT OO

cotton-wool. The lateral tube of the yeast flask is also
plugged with cotton- wool, care being taken that the
plugs are loose enough to allow air to pass freely. The
whole is sterilised and inoculated. The glass tube, B,

76 A MANUAL OF BACTERIOLOGY

which passes through the rubber cork, is then connected
with a Kipp or other hydrogen- genera ting apparatus by
means of a rubber tube, and a current of hydrogen is
passed through the flask. The hydrogen bubbles through
the bouillon and escapes by the lateral tube. After the
gas has been passing for half an hour a small tube con-
taining mercury, c, is applied to the end of the lateral
branch, so that the open end just dips below the surface
of the mercury, and the tube, B, which passes through the
rubber cork, is sealed off in the blowpipe flame, care being
taken that all the air has been expelled from the flask
by a free current of hydrogen. The flask, with the capsule
of mercury applied to the end of the lateral branch, can
then be placed in the incubator. The mercury thus forms
a valve through which air cannot enter, while gases
formed by the growth of the organism have free exit.

For large flasks, the lateral tube may be just bent down
and a little capsule of mercury attached.

The addition of \ to 1 per cent, of sodium formate to
the culture media much simplifies anaerobic cultivation ;
the tetanus bacillus, for example, can be grown in formate
broth in a stoppered bottle without any elaborate pre-
caution for excluding the last traces of air. The sodium
formate should be added immediately before the last
sterilisation, not previously, or decomposition may occur.
Sodium sulphindigotate (0-3 per cent.) may be similarly
used.

With such a broth, Dean's bottle may be used for
anaerobic cultivation. This consists of a bottle around
the neck of which a gutter for mercury is formed. A
glass cap loosely fits over the mouth of the bottle, and its
edge dips into the mercury in the gutter, thus sealing the
bottle.

Plate cultivations. — The method of plate culture is one
of the most important in bacteriology. It is used for

PLATE CULTIVATIONS 77

three purposes : (1) for obtaining pure cultivations, i.e.
cultures containing a single species, from a mixture of
organisms ; (2) for the enumeration of organisms ; and
(3) for ascertaining the characters of the colonies of
organisms as an aid in the identification of species.

Before the introduction of plate cultivations pure
cultures of organisms could only be obtained by chance,
or by the dilution method, which was also by no means
certain. The dilution method consisted in estimating
approximately the number of organisms in a given
volume of fluid by means of an instrument on the same
principle as the hsematocytometer. The fluid is then
diluted by the addition of some sterile fluid so that a
given volume of the dilution contains a single organism
only, assuming the organisms to be evenly distributed
throughout the fluid. By transferring this volume to
tubes of sterile media pure cultivations can in some cases
be obtained, a single organism having been sown in a tube.

It is obvious, however, that this method is at best an
uncertain one, but the plate-culture method to a large
extent obviates this uncertainty. It depends upon the
following principles : Gelatin and agar media, when
melted, remain fluid down to 25° and 45° C. respectively,
temperatures which will not affect the vitality even of
delicate organisms. By inoculating the fluid gelatin or
agar, thoroughly mixing, and then pouring on to a level
sterilised surface, so that the medium solidifies in a thin
film (" plating "), the organisms, wherever they may be
situated, are fixed and are unable to wander, and, being
in a good nutrient soil, grow and multiply and ultimately
form visible growths or colonies. Many of these colonies
will have arisen from a single organism ; the growth,
therefore, is " pure," i.e. consists of a single species, and
pure cultures can be obtained by inoculating tubes of
sterile media from them.

78 A MANUAL OF BACTERIOLOGY

When suitable, sterile nutrient gelatin is usually
employed for the preparation of plate cultivations, as
it is more easily manipulated than agar. Three tubes
of sterile nutrient gelatin are melted at a low tempera-
ture in a beaker of water (gelatin melts at 24° C. ; the
temperature should not exceed about 45° C.). The tubes
may be termed respectively 1, 2, and 3. Tube No. 1 is
inoculated, by means of a platinum needle, with a trace
of the growth from which pure cultivations are desired. The

trace of growth is thoroughly
mixed up and distributed
throughout the melted
gelatin. If this mixture be
" plated," so many organisms
may be present in the film
that the colonies which de-

™l°P wiH "<* ^ separate,
but will form a confluent

growth. To obviate this difficulty a second and a third
dilution are prepared. The second dilution is made
by inoculating the tube of melted gelatin No. 2 with
one platinum loopful from tube No. 1, and thoroughly
mixing up ; and to be quite sure that the resulting
colonies will be isolated from one another, a third dilu-
tion is prepared in the same manner by inoculating the
tube of melted gelatin No. 3 with two to four platinum
loopfuls from tube No. 2. The organisms having been
distributed throughout the gelatin by rolling and gentle
shaking, the wool plug is in each case withdrawn from
the mouth of the tube, the mouth of the tube is sterilised
in the Bunsen burner to prevent contamination, then
cooled for a few seconds, and finally the melted gelatin is
poured on to a level sterile glass surface. Formerly plates
of glass were used (hence the name) ; but now shallow
glass dishes with lids, about three or four inches in

PLATE CULTIVATIONS

79

diameter, known as Petri dishes (Fig. 14), are almost
always employed. They are previously sterilised in the
hot-air steriliser in suitable iron or copper boxes holding
a dozen or so ; the melted gelatin having been poured in,
the dish is tilted to diffuse the gelatin over the bottom of
the dish, placed on a level surface for the gelatin to set,
and then stored in the cool incubator. The plates are
examined daily, with a hand lens if neces-
sary, or with a low power of the micro-
scope, the dish being turned bottom
upwards on the stage of the microscope
for this purpose. When the colonies have
developed, inoculations can be made from
them by means of a platinum needle on
to tubes of sterile media. The colonies,
having arisen from single organisms, are
pure, and the resulting sub-cultures are
therefore also pure (it sometimes happens
that the colonies are mixed owing to two
or more organisms being dose together).
Different species of organisms usually form
colonies having different appearances, so
that the colonies are an aid in diagnosis and enable the
various species to be picked out from a mixture. The colonies
in gelatin are as a rule much more distinctive than those in
agar. Whereas the plate cultivation prepared from tube
No. 1 is generally too crowded, plates 2 or 3, or both, can
be made use of, and it is apparent that, to make certain
of isolating all the organisms from a mixture, several
sets of plates should be prepared. Flat bottles (Fig. 15)
may likewise be used for plate culturing, and are also very
useful for growing organisms in bulk for the examination
of the constituents and actions of the bacterial cells.

Golding has devised flat wedge-shaped flasks (having
sides at an appropriate angle) for plate-culturing, and

Fio. 15.— "Plate1
bottle.

80 A MANUAL OF BACTERIOLOGY

these are very useful, as the culture medium may be kept
in them ready for use.

In addition to the isolation of species from mixtures
and for diagnosis, plate cultures are also used to enumerate
organisms. Assuming that every colony arises from a
single organism, which is approximately the case, the
number of colonies represents the number of organisms
originally introduced into the gelatin, and if a known
weight or volume of the material inoculated be used, the
number of organisms in it can be calculated. For example,
in the bacteriological examination of water a measured
volume of the water is added to melted gelatin by means
of a sterilised pipette, and by counting the resulting
colonies the number of organisms originally present in
1 c.c. of the water can be estimated.

Agar plate cultures may be prepared in a similar way.
The agar must, however, be brought to a temperature of
nearly boiling before it melts ; it is then allowed to cool
to nearly 45° C. and the tubes are inoculated in the same
manner as for a gelatin plate culture described above.
Unless the manipulations be carried out expeditiously the
agar will solidify, or the agar film in the Petri dish be
lumpy.

Agar plates should usually be inverted during incuba-
tion, or the growth may become confluent owing to the
condensation water carrying the organisms all over the
film.

The plate-culture method can be modified to suit
particular circumstances : for example, the melted gelatin
or agar, uninoculated, may be poured into the dishes and
allowed to solidify, and the film then inoculated by
streaking or painting with the material, or by pouring a
few drops of broth containing the organisms upon it.
This is practically the only way in which blood-serum can
be used, the sterile blood-serum being placed in the Petri

SINGLE-CELL CULTURES 81

dish, solidified in the inspissator in the same manner as
for blood-serum tubes, and the coagulated film inoculated.

For many purposes plates are unnecessary, the same
result being obtained by rubbing over the surface of two
or three tubes of sloping agar or gelatin successively the
once charged needle, straight or looped. In the second or
third tubes isolated colonies generally develop.

The plate-culture method often fails if the organism
to be isolated forms but a small minority of the total
organisms present in the mixture ; the only alternative
then is to multiply the number of plates, which, however,
may entail great labour in their examination.

Single-cell cultures. — With large cells, such as yeasts,
it is not difficult to obtain growths from single cells by
making miniature plate cultures on ruled cover-glasses
and ascertaining where single cells are located in the film
by examining the preparation with a J or £ in. objective
(see Chapter XVI). But with the minute bacterial cells
this method is inapplicable. By the use of Burri's Indian
ink method,1 however, single- cell cultures of bacteria can
be obtained. Fluid Indian ink is diluted with 6-10
volumes of distilled water and the mixture is sterilised
in the autoclave. Several loopfuls of this are deposited
in series on a sterile slide. The first drop is inoculated
with the culture which is being investigated, the second
drop is inoculated from the first, the third from the second,
and so on. A fine mapping-pen, sterilised in the flame,
is then dipped into the third, fourth, or fifth drops, and
the trace of Indian ink mixture so picked up is deposited
on a gelatin or agar plate. The droplet is covered with
a sterilised cover-glass and is examined with a J in. or
J in. objective, with a high eyepiece. An organism shows
up white on a black background. Many drops are de-

1 Das Tuschverfahren (G. Fischer, 1909). Giinther Wagner's ink
Hanover) is recommended and is supplied by Griibler.

6

82 A MANUAL OF BACTERIOLOGY

posited on the plate and examined, and those in which
only a single organism can be found are noted and the plate
is then incubated so that colonies may form, from which
sub-cultures may be prepared.

Esmarctis roll cultures. — Another modification of the
plate-culture method is known as Esmarch's roll culture.
For this purpose large test-tubes (" boiling tubes "), at
least an inch in diameter and six inches long, are sterilised
and plugged with cotton- wool. The sterile melted gelatin,
about 10 c.c., is poured in and inoculated, the wool plug
replaced, and the tube held in the horizontal position and
rotated under a stream of cold water, or in warm weather
on a block of ice, until the gelatin has set. In this way
the gelatin forms a thin film over the inside of the tube,
but a little practice is required to get it evenly distributed.
The colonies then develop in the film of gelatin, which is
quite analogous to a film in a Petri dish.

Anaerobic plate cultivations are sometimes required.
The plate culture after preparation as described above,
using a deep Petri dish, is inverted, and some alkaline
pyrogallol is placed in the lid ; this absorbs the oxygen
within the dish. The preparation must be kept under
observation for the next hour or so, and more alkaline
pyrogallol is added from time to time to compensate for
the rise of fluid within the dish until absorption of the
oxygen from the contained air is complete.

McLeod has devised a useful porcelain dish for con-
taining the alkaline pyrogallol over which the Petri dish
is inverted, the joint being made air-tight with plasticine.

In Botkin's method a bell-jar standing in a glass dish
is made use of. The Petri dishes are placed on a support
within the bell- jar, and mercury or oil is poured into the
glass dish. By means of a piece of bent glass tubing a
stream of hydrogen is passed into the bell- jar under its
rini so as to displace the air, which bubbles out through

FERMENTATION TUBES 83

the oil or mercury. When the air has been entirely
displaced the glass tube is removed, the bell- jar weighted,
and the whole placed in the incubator. Bulloch's apparatus
is somewhat similar to this. Wide- mouthed jars with
well -ground glass lids, which are luted down, are very
convenient, the oxygen being absorbed with alkaline
pyrogallol placed at the bottom, and the Petri dishes
stacked on a glass capsule or other support to raise them
above the fluid.

The Esmarch roll cultures can be adapted for anaerobic
cultures. The wool plug is replaced by a rubber cork
with two holes, through which inlet and outlet glass
tubes pass, as in Frankel's anaerobic tubes (Fig. 12). The
gelatin (or agar) having been melted and inoculated, the
medium is kept melted in a water- bath at appropriate
temperature, the hydrogen is passed through for a quarter
of an hour, the tubes are sealed oft', and the roll- culture is
prepared.

Golding's flask (p. 79) or a " plate " bottle (Fig. 15)
may be similarly used, or a Golding flask may be inverted
over a beaker of alkaline pyrogallol.

For the detection of fermentation and gas production,
stab cultures in glucose agar or shake cultures in gelatin
may be employed. For the latter a tube of gelatin1 is
melted at a low temperature, inoculated with the organism,
and allowed to solidify in the upright position ; the
organism is thereby distributed throughout the medium.
Fermentation with gas production is indicated by the
presence of gas bubbles, or even by the disruption of the
medium. Durham's fermentation tubes are very con-
venient for showing fermentation. These are test-tubes
containing suitable fluid media (10 c.c. each) into which
small glass tubes closed at the upper end are placed ; the

1 Lemco gelatin frequently gives no gas ; a meat-broth gelatin
should therefore be used for gelatin shake cultures.

84 A MANUAL OF BACTERIOLOGY

latter become filled during the sterilisation. The tubes
are inoculated and incubated, and if fermentation occurs
the little tube becomes filled with gas (Fig. 16). Einhorn's
saccharimeter may also be used (Fig. 17). The tube is
filled with the medium, sterilised, inoculated, and in-

FIG. 16. — Durham's
fermentation tube.

FIG. 17. — Einhorn's sacchari-
meter.

cubated. Any gas produced collects in the closed limb
of the tube. When the amount of gas ceases to increase,
a little strong caustic potash solution may be added ; this
absorbs the C02, the residue probably being hydrogen,
and thus the H : C02 ratio may be determined. The
most suitable media for fermentation are peptone broth,
the acid beef-broth for which has been treated with the

FERMENTATION TUBES 85

colon bacillus (see p. 27), 1-2 per cent, peptone water,
or a medium which has been largely used by Houston,
Gordon, and others, consisting of a 1 per cent, solution
of " Lemco " in distilled water with the addition of peptone
1 per cent., sodium bicarbonate O'l per cent. ; to either
medium is added 1-2 per cent, of glucose, lactose, sac-
charose, starch, inulin, mannitol, dulcitol, etc., and the
mixture is tinged with litmus.

The fermentation tube has been much used of late for
the examination of faeces in abnormal intestinal conditions.
For this purpose 1 grm. of faeces is thoroughly emulsified
in 10 c.c. of physiological salt solution and 1 c.c. of the
suspension is introduced into the fermentation tube, the
long arm of which is 95 mm. long. The media employed
are 1 per cent, dextrose, lactose, and saccharose broths
made with " Lemco " (as above) or with sugar- free meat
broth (see p. 27). With such tubes normal stools yield
the following amounts of gas : 1

Dextrose. Lactose. Saccharose.

26-75 29-9 19-5 mm.

1 See Herter and Kendall, Studies from the Rockefeller Institute
(Reprints), x, 1910.

CHAPTER III

THE PREPARATION OF TISSUES AND ORGANISMS FOR
STAINING AND MOUNTING— STAINING AND STAINING
METHODS

A SELECTED few of the numerous methods devised for the
preparation and staining of tissues, bacteria, etc., are here
given. Special methods occasionally employed will be
described when required.

Preparation of Tissues

In bacteriological work the demonstration of the bacteria
in the tissues is the primary object, and, therefore,
the elaborate methods which have been devised for fixing
the tissue elements are not usually required, unless it
be that the minuter changes in the latter are being studied.
The tissues should always be obtained as fresh as possible,
because within a few hours of death they are invaded by
numerous bacteria, derived from the air and from the
intestine, which may mask the original bacterial infection
and lead to serious mistakes if this source of error be not
carefully borne in mind. In all cases the tissue should be
cut into pieces of convenient size, not more than about
1 cm. in thickness, and organs if kept en masse should be
sliced. Having been thus prepared, the material may
be treated by one of the following methods :

(a) Place directly in 'alcohol1 for a week or a fortnight.

1 Methylated spirit may usually be employed for all purposes when
an alcohol of not more than 90 per cent, strength suffices. //

86

PREPARATION OF TISSUES 87

(b) Place in alcohol 1 part, water 2 parts, for twenty-
four to forty- eight hours, transfer to alcohol and water,
equal parts, and finally to absolute alcohol, for like periods.

(c) Place in rectified spirit (8(5 per cent, alcohol) con-
taining 1 per cent, of corrosive sublimate for twelve . to
forty- eight hours, and pass through increasing strengths
of alcohol as in (6).

(d) Place for six to twenty hours in a saturated aqueous
solution of corrosive sublimate. This is prepared by
saturating boiling distilled water with the corrosive
sublimate, cooling, and filtering. Keep in the dark.
When removed from the corrosive sublimate solution the
tissues must be washed in a stream of running water for
an hour, or, better, placed for a day in 70 per cent, alcohol
deeply coloured with iodine, to remove the excess of
corrosive sublimate and prevent precipitation. The tissues
are then passed through increasing strengths of alcohol,
as in (b).

(e) Formalin, a 40 per cent, aqueous solution of formic
aldehyde, is an excellent fixing agent. A solution of
1 part of formalin and 9 parts of water, or better, physio-
logical salt solution, may be used, the pieces of tissue
remaining in this for twelve to twenty- four hours. They
are then washed in running water for an hour or two and
passed through increasing strengths of alcohol, as in (b).

All tissues after fixing and hardening should be pre-
served in alcohol — 70-80 per cent.

The methods (c), (d), and (e) are to be recommended,

however, be free from mineral naphtha, which is present in all " shop "
methylated spirit. Methylated spirit free from mineral naphtha can
be obtained in quantities of five bulk gallons, " for scientific purposes
only," by special order from the Inland Revenue Authorities, Somerset
House, W.C. If it cannot be procured, absolute alcohol must be
employed. Duty-free absolute alcohol can also be obtained at a low
price under somewhat similar conditions. In the following pages,
when the unqualified term " alcohol " is used, the naphtha-free methy-
lated spirit may generally be employed.

88 A MANUAL OF BACTERIOLOGY

especially the two last, as the tissue elements are well
fixed thereby. In all cases the fixing fluid should be used
in considerable excess. Fixing fluids containing potassium
bichromate (as in Miiller's fluid) and chromic acid seem
to prevent the bacteria from staining with any certainty,
and should be avoided.

Section Cutting

In order satisfactorily to demonstrate bacteria in tissues,
and their relation to the tissue elements, it is usually
necessary to prepare sections. For this purpose either the
freezing or the paraffin method should be employed.

(a) Freezing method. — The tissue, in suitable pieces,
must first be soaked in water to remove the alcohol. A
convenient way of doing this is to place the material in
a wide- mouthed bottle, into the mouth of which an
ordinary glass funnel is introduced, and the bottle with
the funnel is placed under a stream of running water ;
the funnel, while allowing the water to flow out, retains
the pieces of tissue in the bottle. With running water
the alcohol will be completely removed in from one to
two hours ; in still water, which should be changed two
or three times, this result may not be attained for several
hours, during which time there is an ever-increasing risk
of bacterial contamination from without. It is essential
to remove all the alcohol, or the tissue will not freeze.

When the alcohol has been removed, which is known
by the tissue sinking in the water (lung is an exception
— it always floats unless solid from any cause), the pieces
are transferred to a strong mucilage of gum acacia :

Gum acacia ...... 5 grm.

Cane sugar . . . . . . 0-5 grm.

Water 100 c.c.

Add a piece of thymol or a little carbolic acid to prevent decom-
position. Hamilton saturates the solution with boric acid.

SECTION CUTTING 89

In this gum solution the pieces remain for twelve to
forty- eight hours, according to their size and the time
at the disposal of the investigator, and are then cut on
one of the numerous ether-freezing microtomes now to be
obtained, such as Swift's (Fig. 18) or Cathcart's. A

FIG. 18. — Swift's ether-freezing microtome.

microtome in which the freezing is effected by carbonic
acid is now frequently employed and acts well. Liquid
carbonic acid, contained in a cylinder, sprays by its own
pressure on to the under surface of the plate on which the
block of tissue rests ; the tissue quickly freezes and is
then cut. This form of microtome works satisfactorily in
the hottest weather. The material must not be frozen so
hard that the sections roll up and fall off the knife ; the
sugar in the above solution should prevent this. The
sections are transferred successively to two or three lots

90 A MANUAL OF BACTERIOLOGY

of distilled water, preferably slightly warmed, to remove
the gum, and can then be stained at once, or may be
preserved in equal parts of alcohol and water.

Bacteria seem to retain their staining properties better
in the tissue in bulk than in sections preserved in alcohol.
This objection does not hold with paraffin sections.

(b) Paraffin method. — Nothing can surpass the paraffin
method for the thinness and beauty of the sections obtain-
able by it, and for some friable tissues, such as actino-
mycosis, it is almost essential. The tissue, in suitable
pieces for cutting, is transferred from the diluted spirit
preservative solution to pure methylated spirit for two
or three hours, and then to absolute alcohol — which may
have to be changed once unless a fairly large volume is
employed — for from four to twenty- four hours. It is
then removed from the alcohol, lightly dried between the
folds of a dry cloth or piece of blotting-paper to remove
the superfluous alcohol, and placed in an excess of xylol,
in which it remains for from four to twenty-four hours
until cleared. This is recognised by the material assuming
a more or less semi-transparent condition, and the pro-
cess may be much accelerated by warming the xylol to
from 37° to 50° C. in the blood-heat incubator or paraffin
oven, the bottle containing the xylol being well stoppered.
When cleared it is ready to go into the bath of melted
paraffin. A paraffin of a fairly high melting-point is
perhaps the best, viz. 45° to 55° C., and is placed in glass
capsules in an oven which can be kept uniformly heated
to the required temperature. An ordinary chemical hot-
water oven answers the purpose quite well, and is heated
by a special form of small Bunsen burner with mica
chimney, the temperature being regulated by some form
of mercurial regulator, which is set a degree or two above
the melting-point of the paraffin employed. The tissue
is taken out of the xylol, blotted to remove the excess, and

PARAFFIN SECTIONS 91

placed in the melted paraffin for from six to twenty hours.
It is then embedded by pouring a little of the melted
paraffin into a watch-glass, or into a small box formed
of folded paper or lead-foil, or by bringing together
two L- shaped pieces of brass on a glass plate so that a
rectangular cavity is produced. The pieces of tissue are
then taken out with a small warmed forceps or needle,
adjusted to the position they are required to occupy,
and more melted paraffin is poured in, so as to cover them.
When a film of solid paraffin has formed, the whole is
immersed in cold water so as to cool it rapidly.

A new paraffin is frequently crystalline in structure,
and acts much better after it has been kept melted for
some weeks, or is much improved by heating nearly to
its boiling-point for five or six days (P. T. Beale). The
xylol for clearing may be used several times and the
paraffin repeatedly, the remains of old tissues being
removed. The time which the tissues require to remain
in the alcohol, xylol, and paraffin depends upon their
size ; very small pieces may be treated in a few hours,
large ones may require two or three days.

Other clearing agents, such as chloroform, turpentine,
and cedar oil, may be used instead of xylol. The paraffin
method is usually straightforward, but small pieces of
tissue must not be left too long either in absolute alcohol
or in the paraffin bath, for they are liable to become too
hard to cut. Thyroid tissue and skin are also rather
troublesome ; they become very hard unless the whole
process is carried out as rapidly as possible. If the pieces
of tissue be large, the capsule of melted paraffin containing
the tissue may be placed under the receiver of an air-
pump, which is then exhausted. This causes the paraffin
to penetrate better, and the process may be repeated two
or three times during the period of infiltration. A special
form of paraffin oven has been devised by Cheatle for

92 A MANUAL OF BACTERIOLOGY

infiltrating under diminished pressure, and is made by
Messrs. Hearson, of Regent Street, London.

In order to prepare sections from material embedded in
paraffin some form of microtome must be employed. An
ether-freezing microtome can be made use of with some
manipulation, the paraffin block being placed in a little
melted paraffin on the freezing plate so that it is cemented

FIG. 19. — Cambridge rocking microtome.

there, and sections are cut with the razor or plane iron,
as though it had been frozen (it is not to be frozen). It is
better, however, to use some special form of microtome,
the Cambridge " Rocker " (Fig. 19), or a modification of
it, or the " Minot," being perhaps the best. The block of
paraffin containing the tissue is trimmed with a knife to
remove the excess, and is cemented to the carrier of the
microtome with a little melted paraffin, or by melting
the paraffin on it with a hot iron (end of a file, etc.) or a
match. The union may be made more secure by melting
the paraffin around the base of the block with a hot iron.
Having fixed the paraffin block to the carrier, sections
may then be cut of any degree of thinness. In order to

MICROTOMES 93

do this it is essential for the knife or razor to have a keen
edge and one of the right nature, for a knife may be
perfectly sharp and yet the sections as they are cut may
roll up in such a manner that it is difficult to flatten them.
Though this may be due to a wrong consistence of the
paraffin, owing to cold weather or some other factor, in
the majority of instances it is the edge of the knife which
is at fault. Provided the knife be sharp, stropping on
the palm of the hand will usually remedy this difficulty.
The paraffin being of the right consistence, and the knife
in good order, the sections as they are cut should be flat
and should adhere together at adjacent margins so that
a ribbon of greater or shorter length is formed.

Satisfactory sections having been obtained, they are
transferred with a needle or camel's-hair brush to a tin
pan containing a little water, or spirit and water warmed
to about 40° C. The sections float and the paraffin softens
so that they spread out perfectly flat (the water must
not be hot enough to melt the paraffin). A clean slide is
then introduced underneath the section, raised so that
the section is lifted up on it, and by fixing the section with
a needle and tilting the slide the section is deposited in
the required position on the slide and allowed to dry.
If preferred, the section may be transferred to a slide
flooded with water, which is warmed over the Bunsen.
The slides can be manipulated in an hour or two if dried
at 37° C., but it is best to allow them to dry in the incubator
all night. It will be found after this treatment that thin
sections generally adhere sufficiently firmly to the slides
for all the ordinary methods of staining to be carried out
without detaching them ; thick sections, however, do not
adhere nearly so well.

To prevent the risk of detachment, it is generally better
to cement the sections to the slides by the following
method., Equal parts of egg-white and glycerin are mixed

94 A MANUAL OF BACTERIOLOGY

and filtered through muslin, and to every 100 c.c. of the
mixture 1 grm. of sodium salicylate is added. The slide
is smeared thinly with this, the section is transferred to it
and afterwards dried in the manner above described.

Supposing that the sections, in spite of all precautions,
curl up as they are cut, it is still often possible to obtain
a few that can be mounted. They may sometimes be
unrolled by cautious manipulation with a couple of
needles after having been softened by warming, or a
needle or knife- blade may be held close to the edge of
the microtome knife during cutting, so that curling is
prevented.

Tissues embedded in paraffin may be kept indefinitely in labelled
pill-boxes and cut all at once or from time to time as required, or
the ribbons of sections may be preserved in a box in a cool place
until wanted. The slides also, with the sections attached, can be
kept until it is convenient to stain, if preserved free from dust in
a slide box.

Cover-glass and Film Specimens

The satisfactory preparation of cover- glass and film
specimens is one of the most important in bacteriology,
for they are used for the examination of cultivations of
bacteria, and of blood or other fluids or secretions, organs,
etc., for the presence of micro-organisms.

Films and smears are now usually made on the slide,
but may be made on the cover- glass (" cover- glass speci-
mens "). In either case the glass must be clean and free
from grease. Cover- glasses must be thin, otherwise the
higher powers cannot be employed to examine the prepara-
tions, and those described as " No. 1 " should be purchased,
" f-in. squares " being a convenient size. These serve
both for cover- glass specimens and for covering sections ;
it is well also to have a few of the same thickness but
larger, viz. |-in. or 1-in. squares, for large sections.

FILM PREPARATIONS 95

Slides and cover-glasses may be cleaned by boiling them
in a porcelain dish with 10 per cent, carbonate of soda
solution for a few minutes, well washing, and then treating
with strong sulphuric acid, warmed carefully in a porcelain
dish, for a few minutes. The acid having been poured off,
they are well rinsed in several changes of water, and
should be kept in a stoppered glass pot or capsule in
absolute alcohol.

A clean slide (or cover- glass) is taken, dried with a
clean soft linen or silk rag or handkerchief, or with
Japanese paper, or it may be momentarily introduced into
the Bunsen flame and the spirit burnt off, and placed
flat on a convenient support on the work-table — a white
glazed tile is excellent — with the end or corner projecting
so that it can be conveniently picked up.1 A droplet
(i.e. small drop) of tap-water or of physiological salt
solution (not distilled water) is then placed on it, in the
middle, by means of a looped platinum needle, or with
a small glass pipette (Fig. 7). Theoretically, physiological
salt solution2 sterilised by boiling should be used, but
ordinary tap- water may generally be employed. A thin
film of organisms has now to be formed on the glass, and
the following is the method of procedure with a culture
on a solid medium such as agar or gelatin. The culture
tube and platinum needle are held and manipulated in
precisely the same manner as that described for the
inoculation of tubes (p. 69).

A mere trace of the growth from a culture should be
taken, just sufficient to soil the tip of the straight platinum
needle, or the preparation will be too crowded, and this
is well rubbed up with the droplet of water on the glass,

1 The writer has devised a useful support for staining. It consists
of a square of plate glass, painted half white and half black at the
back, and having a narrow strip of thick glass cemented across it on
which the gl":js rests. It is made by Messrs. Baird and Tatlock.

2 0-75-0-95 per cent, of sodium chloride dissolved in distilled water.

96 A MANUAL OF BACTERIOLOGY

so as to form an emulsion, which is then spread over the
surface. As a general rule the material should be well
emulsified, but in some instances this is inadvisable, as a
particular formation or characteristic grouping may be
disturbed thereby, in which case, after a slight admixture
with the water, the emulsion is gently spread. The thin
moist film is allowed to dry, or may be dried by gentle
warming over the Bunsen flame, preferably holding the
preparation in the fingers and moving backwards and
forwards over the flame. The film, when dry, must next
be fixed, which is accomplished by passing the slide,
film side up, six times through the Bunsen flame (a cover-
glass is held in the forceps and passed three times through
the flame). Films may also be fixed in alcohol and ether
(p. 97). The object of this " fixing " is to thoroughly
dry the film and coagulate albuminous material, whereby
the film adheres better to the glass, and is not so likely
to be detached in the subsequent processes of staining
and washing, etc. Fixing may also tend to diminish the
staining capacity of the extraneous matter mixed with
the organisms. The preparations are now ready for
staining.

When the culture is in a fluid medium, such as broth,
the tube is manipulated in the same way, the deposit at
the bottom having been shaken up if necessary, and a
loopful or two of the fluid removed with a looped platinum
needle, transferred to the glass, spread, dried, and fixed
as before, but as the medium is fluid there is usually no
need to add any water.

If a specimen of blood, pus, or sputum is required, the
procedure is much the same. A little of the material is
taken up with a looped platinum needle and spread in a
thin film over the slide or cover- glass, which is then dried
and fixed, as described above. If necessary, a droplet
of tap water or physiological salt solution may be used

SMEAR PREPARATIONS 97

to dilute the material so as to obtain a thinner film. If a
specimen is to be made from an organ, a particle of the
pulp is picked up and an emulsion made as before, or a
small piece of the organ may be held in sterile forceps and
the cut surface gently smeared over the slide or cover-
glass, which is then similarly dried and fixed ; these are
termed " smear preparations."

To obtain the best results it is preferable before staining
to submit films of blood1 or pus or smear preparations to
the action of some chemical fixing agent, unless the film
is stained with Leishman's solution, which both fixes and
stains. The simplest method of doing this is to immerse
the films, after owr-drying, in a mixture of equal parts of
absolute alcohol and ether for ten to thirty minutes. In
hot countries a saturated aqueous solution of corrosive
sublimate (five to fifteen minutes) is perhaps as satisfactory
as anything. Another method, combining both fixing and
staining, is to immerse the films as soon as they are pre-
pared and without drying for a few minutes in the following
solution :

Absolute alcohol, saturated with eosin . . 25 c.c.

Pure ether ....... 25 c.c.

Alcoholic solution of corrosive sublimate (2 grm.

in 10 c.c.) ....... 5 drops

The specimens are then removed with a forceps and well
rinsed in water, stained for not more than a minute in a
saturated aqueous solution of methylene blue, washed
quickly, dehydrated in absolute alcohol, cleared in xylol,
and mounted in xylol balsam. This solution may be
used for fixing blood, pus, sputum, etc., if the eosin be
omitted, and the preparations may then be stained or
otherwise treated in any desired manner.2

1 For the method of preparing blood-films see the section on " Ma-
laria," Chapter XVIII.

2 Gulland, Brit. Med. Journ., 1897, vol. i, p. 65.

7

98 A MANUAL OF BACTERIOLOGY

Scott1 recommends the following as giving the most
perfect results with blood films, etc. :

(1) Hold the freshly prepared and still wet film in the
mouth of a wide-mouthed bottle half filled with the
ordinary formalin solution, film side downwards, for five
seconds.

(2) Drop, while still wet, film downwards, into absolute
alcohol. Leave for fifteen minutes, or, for convenience,
for any time up to forty- eight hours.

The preparations may then be stained with methylene
blue, hsematoxylin and eosin, or with the Leishman or
Giemsa stain. (See also under " Malaria," Chapter XVIII.)

Impression specimens. — These are employed to examine
and preserve permanently the colonies or growths of
organisms so that their characteristic formation may be
observed. With plate cultivations this is very simple.
A clean cover-glass is sterilised in the flame and, having
cooled, is cautiously lowered on to a selected surface
colony with a sterile needle, avoiding all lateral movement.
It is then gently pressed on to the colony and then care-
fully raised by means of a couple of needles ; the colony
should adhere to the glass, and may be dried and fixed.
The colonies in gelatin tube cultures may also be used if
the gelatin is removed from the tube. This can be done
by dipping the tube for a few seconds into hot water ;
the gelatin round the walls of the tube will be melted,
and the gelatin mass can then be tilted out of the tube
on to a glass dish or tile.

Stains and Staining Methods

Micro-organisms being so minute and transparent, it
is usual to stain or dye them, so that they can be more
readily examined. In some instances organisms may have

1 Journ. Path, and Bact., vol. vii, No. 1, p. 131.

STAINS AND STAINING METHODS 99

a peculiar staining reaction which may serve as an aid to
their identification. But when an organism is being
investigated, examination in the fresh and living condition
must never be omitted, for it is only thus that its motility
and life-history can be studied. Only general methods
are detailed here ; special ones will be given when they
are required.

(1) LofHer's alkaline methylene blue :

Saturated alcoholic solution of methylene blue . 30 c.c.
Solution of caustic potash, 0-01 per cent. . . 100 c.c.

A very useful staining solution. Cultures should be quite fresh,
or the organisms do not stain well. When the organisms are
mixed with extraneous material, as in smears, or there is much
debris, this is one of the best staining solutions to employ. Methyl-
ene blue preparations are, however, not very permanent, and in
hot countries rapidly fade. Thionine blue is then preferable. (See
also p. 100.)

Film specimens are stained for three to ten minutes, and
sections half to twenty-four hours.

(2) Carbol-methylene blue (Kiihne) :

Methylene blue . . . . . .1-5 grin.

Absolute alcohol . . . . . 10 c.c.

Five per cent, aqueous solution of carbolic acid . 100 c.c.

A more intense staining solution than the former, and very
useful for sections, which are stained for from half to six hours.

(3) Anilin gentian violet :

Saturated alcoholic solution of gentian violet . 30 c.c.
Anilin water ....... 100 c.c.

The anilin water is prepared by shaking 3 c.c. of anilin with
90 c.c. of distilled water, allowing the mixture to stand for a few
minutes, and filtering.

This solution is a useful general stain for films, which are stained
for two or three minutes, and is employed in Gram's method of
staining. It does not keep well.

Instead of anilin gentian violet, a carbol-gentian violet may be
used, and keeps much better than the foregoing (saturated alcoholic

100 A MANUAL OF BACTERIOLOGY

solution of gentian violet, 1 part ; 5 per cent, aqueous solution of
carbolic acid, 10 parts).

For anilin gentian violet two stock solutions may be employed,
and these seem to keep indefinitely, viz. :

No. 1
Gentian violet ...... 2 grm.

Anilin ........ 9 c.c.

Alcohol (95 per cent.) ..... 33 c.c.

No. 2
Gentian violet ...... 2 grm.

Distilled water 100 c.c.

For use, mix 1 c.c. of No. 1 with 9 c.c. of No. 2, and filter ; this
mixture will keep for about a fortnight.

(4) Carbol-fuchsin (Ziehl-Neelsen solution) :

Fuchsin ....... 1 part

Absolute alcohol . . . . . • . 10 parts

Five per cent, aqueous solution of carbolic acid . 100 parts

The fuchsin is dissolved in the absolute alcohol and then mixed
with the carbolic acid solution. It must always be filtered before
use.

An intense staining solution. For films it is best diluted with
five to ten parts of water ; stain for two to five minutes.

(5) Carbol-thionine blue (Nicolle) :

Saturated solution of thionine blue in alcohol

(90 per cent.) 10 c.c.

One per cent, aqueous solution of carbolic acid . 100 c.c.

Sections can be stained in from a few minutes to half an hour.
This solution may be used for a modified Gram's method (see
p. 106). Can be substituted for methylene blue for all purposes,
and is more permanent than the latter.

(6) Eosin (alcohol-soluble and water-soluble) :

A somewhat diffuse stain. Is used for counter-staining the
tissues in Gram's method, and for staining red blood-corpuscles
and acidophile granules in leucocytes.

A | to 1 per cent, aqueous or alcoholic solution may be used,
and the staining should not, as a rule, be prolonged for more than
about half a minute.

STAINING SOLUTIONS 101

(7) Bismarck brown (Vesuvin) :

A saturated aqueous solution should be prepared and diluted
somewhat for use. A good counter-stain for the tissues in Gram's
method. Stain for two to five minutes.

(8) Orange-rubin :

Prepare saturated aqueous solutions of orange G. and rubin S.
Mix equal volumes and dilute with water until of a light port-wine
colour. Stain tissues for five to fifteen minutes. A good contrast
stain for tuberculosis and actinomycosis.

(9) Picro-carmine :

This is best bought ready prepared. Sections are stained in the
solution for half to one hour, washed, then placed in a watch-glass
of spirit, to which three or four drops of hydrochloric acid have
been added, for two or three minutes, then well washed in water.
The section can now be counter-stained with Loffler's blue or by
Gram's method.

(10) HaBmatoxylin :

Ehrlich's formula is one of the best and simplest to use, and
can be obtained ready for use. It must be " ripe." It is a histo-
logical and not a bacterial stain. Sections are treated as follows :

(1) Distilled water, one to two minutes.

(2) Stain with the hsematoxylin solution for five to thirty minutes.
In some cases the solution is preferably diluted somewhat with
distilled water.

(3) Rinse in distilled water.

(4) Rinse in distilled water containing a trace of acetic acid.

(5) Treat with distilled water containing a trace of ammonia.
The sections remain in this until they assume a deep blue colour.
(Tap-water, five to ten minutes, may also be used.)

(6) They can be dehydrated, cleared and mounted, or counter-
stained with eosin, orange-rubin, or Van-Gieson, and then mounted.

Hsematoxylin makes a good contrast stain for the tubercle and
the leprosy bacillus and for Actinomyces.

Mayer's hsemalum (see section on the " Amoeba coli ") and
Delafield's hsematoxylin are also good hsematoxylin stains.

(11) Ehrlich-Biondi triple stain :

This is best bought ready for use. It is a good histological stain
for tissues and blood films, and actinomycosis stains well by it.
Stain for ten to sixty minutes, then treat with methylated spirit

102 A MANUAL OF BACTERIOLOGY

until the section becomes greenish. Pass through absolute alcohol,
clear, and mount.

(12) Leishman's stain :

Like the Jenner, Wright, and other similar ones, a modification
of the Romanowsky stain, a double compound of eosin and methyl -
ene blue. The solution will keep for some time, but is best freshly
prepared. Griibler's powder or Burroughs Wellcome 's soloid
may be used, and is dissolved in pure (Merck's or Kahlbaum's)
methyl alcohol. Failure frequently proceeds from the use of a
so-called pure methyl alcohol, which is not really so. (For method
of using, see " Malaria," Chapter XVIII.)

(13) Giemsa stain:

An eosin-azur mixture dissolved in pure glycerin and methyl
alcohol. Useful for blood-films, smears, etc., and has been much
used to demonstrate the spirochaetes in syphilitic material. (For
method of using, see " Syphilis " and "Malaria.")

Safranin and acid fuchsin are also used as counter-stains.
Malachite green, neutral red, and rosein may be used for intra-vitam
staining of protozoa, etc.

Eosin, orange-rubin, hsematoxylin, and picro-carmine keep well
in solution ; the remainder may or may not, and are best used
fairly fresh. All stains should be filtered before use, and may be
conveniently kept in bottles having a funnel fitted with a filter-
paper, so that they are always ready. Or smaller bottles may be
used, fitted with pipettes, and several arranged in a stand.

Methylene-blue, Leishman and Giemsa preparations are more
permanent if kept unmounted. After examination with the oil-
immersion, the oil may be removed from the film with xylol. Coles
mounts these preparations in parolein.

The best stains are Griibler's, which can be obtained from many
agents in this country. Messrs. Burroughs, Wellcome and Co. supply
most of the anilin dyes and some other reagents, iodine, etc., in
" soloids," which are very convenient and good.

Gram's method. — This is a most useful method, especially
for sections, specimens of blood, or films or impression
preparations, as the tissue or ground substance can be
counter- stained so that the organisms show up in marked
contrast. Ordinary films of cultures do not usually require
this method, unless debris or ground substance is present and

GRAM'S METHOD 103

the best result is desired. Unfortunately Gram's method
is not applicable for all organisms, as many do not retain
their colour by the process. This disadvantage, however,
is counter- balanced by the fact that it forms a valuable
means of distinguishing organisms, and is always one of
the points to be noted in bacteriological diagnosis. Most
of the moulds, yeast, streptothrix and sarcina forms, and
cocci stain by it, though there are exceptions ; the spirilla
and protozoa do not stain by it, but as regards the bacilli
no rule can be laid down (see p. 105). Films are stained
for five to ten minutes, and sections for ten minutes to
half an hour, in anilin- or carbol- gentian violet solution.
The superfluous stain is then drained or blotted off, not
washed away, the specimen is rinsed with Gram's iodine
solution and is treated with fresh iodine solution for from
one-half to two minutes.

GRAM'S IODINE SOLUTION

Iodine .... 1 part

Potassium iodide ..... 2 parts

Distilled water 300 parts

The purple colour of the gentian violet changes to a
dirty yellowish- brown, and sections become much like
a used tea-leaf. The specimens must not be passed on
to the next solution until they have assumed the brown
colour. Cover- glass specimens are best immersed in the
solution in a watch- glass, film side up.

The specimens are removed from the iodine solution,
drained, and then immersed in alcohol, preferably methyl-
ated spirit. In this the purple colour of the gentian
violet returns and is dissolved out, so that they ultimately
become colourless ; this is aided by moving them gently
about, and for sections two or more baths of alcohol may
be an advantage, a fresh one being substituted when the
first has become deeply coloured. Films decolorise much
more readily than sections, and they should be removed

104 A MANUAL OF BACTERIOLOGY

from the alcohol when no more colour dissolves out, or
the stain may be entirely removed ; usually twenty to
forty seconds in the alcohol suffices, thick preparations
taking longer than thin ones. After decolorising, films
are washed in water, dried, and mounted, or, after washing,
the ground substance may be counter-stained, if required,
with eosin for a few seconds, or Bismarck brown for two
or three minutes, washed again in water, dried, and
mounted. With films it is important to remember on
which side of the glass the film is, for it may be very
difficult to ascertain this after decolorisation. Sections
after decolorising are passed through absolute alcohol and
xylol before mounting, or, if required to be counter-
stained, are immersed in eosin for fifteen to thirty seconds,
or Bismarck brown for three to five minutes, and then
passed through alcohol, absolute alcohol, and xylol.

Sections frequently are somewhat difficult to decolorise
with alcohol alone, in which case it is well to treat them
with a slightly acid alcohol (3 per cent, of hydrochloric
acid) for a few seconds, and then return to the alcohol
(Giinther's method).

The iodine in Gram's method seems to act as a mordant,
precipitating the stain in a relatively insoluble form in
certain species of bacteria. The staining of organisms
by Gram is relative ; some forms do not stain at all, are
Gram-negative — i.e. the colour is removed by the alcohol
with the greatest facility ; others stain intensely, are
Gram-positive, but even these may become decolorised
by prolonged treatment with alcohol. In order to ascer-
tain whether an organism is or is not stained by Gram's
method, it is sometimes useful to mix with it in making
the preparation some undoubted Gram-staining organism
— e.g. if a bacillus, the Micrococcus pyogenes ; if a coccus,
B. anthracis or B. subtilis. The admixed organism then
serves as an index.

WEIGERT'S METHOD 105

The following organisms are Gram-positive : B. anthracis,
B. diphtherice, B. tetani, B. \Vefchii, B. botulinus, B. tuberculosis,
B. smegmatis, B. leprce, B. murisepticus, Actinomyces, B. subtilis,
B. mesentericus, B. megaterium, B. mycoides, the pyogenic cocci,
the streptococci, including the pneumococcus, most cocci, yeasts,
moulds, and streptothrices.

The following organisms are Gram-negative : B. typhosus,
B. enteritidis, B. dysenterice, B. coli, B. pestis, B. inftuenzce, B. mallei,
B. pseudo-tuberculosis, B. pyocyaneus, B. osdematis maligni, B.
Chauvcei (usually), B. prodigiosus, B. proteus, the septicaemic bacilli,
such as chicken cholera, the spirilla and vibrios, spirochaetes and
protozoa, M. gonorrhoeas, M. meningitidis, M. melitensis, and
M. catarrhalis.

Gram's method of staining depends upon the formation of an
iodine-pararosanilin-protein compound which is not readily dis-
sociable in the case of the Gram-positive organisms. Pararosanilin
dyes, such as gentian violet, methyl violet and victoria and
thionine blues, are alone suitable for the method.

In Claudius's modification of Gram's method,1 staining
is done in a 1 per cent, aqueous solution of methyl violet
(films for one minute, sections for two minutes). The
preparations are washed, treated with a half -saturated
aqueous solution of picric acid for one to two minutes,
washed again, and dried with filter- paper. Decolorisation
is then carried out in the case of films with chloroform,
in that of sections with clove oil. After decolorising, the
preparations are treated with xylol and mounted. By
this method the ordinary Gram-positive organisms are
stained ; also the bacilli of malignant cedema and of black
quarter. Counter-staining may be carried out with lithium
carmine.

W eigert's modification of Gram's method. — In this process
the sections, whether frozen or paraffin ones, should be
manipulated on the slide. They are stained with the
anilin gentian violet and treated with Weigert's iodine
solution (iodine 4-5 per cent., potassium iodide 6 per
cent.) as in the simple Gram's method. The iodine is then
1 Ann. de VInst. Pasteur, xi, 1897, p. 332.

106 A MANUAL OF BACTERIOLOGY

removed with filter-paper and the sections are flooded
with anilin oil two or three times. This removes the
colour and dehydrates. The anilin oil is removed by
flooding two or three times with xylol.

Thionine blue may be used for Gram's method, the
carbol solution being employed (No. 5, p. 100). Sections
are stained for two or three minutes, then treated with
an iodine solution somewhat stronger than Gram's (200
parts of water instead of 300 parts). The sections, after
remaining in this for one to two minutes, are decolorised in
alcohol containing 1 per cent, of acetone (methylated spirit
does very well), and subsequently treated as in Gram's
method.

The Staining of Film Specimens

To stain films, smear, and impression preparations,
the film is flooded after fixing with a drop or two of the
solution, or the preparation, if a cover- glass, may be
floated, film side down, on the solution contained in a
watch-glass ; if it should sink it makes little difference.
Various baths or pots can be obtained for staining slides.
Warming intensifies the staining properties of all staining
solutions, and may be necessary if deep staining is required
or if the temperature of the laboratory be low (see also
p. 110). When stained sufficiently, the preparation is
rinsed in a beaker or tumbler of water, or in a fine stream
of water, preferably distilled, to remove the superfluous
colour, after which it is dried and mounted in a drop
of solution of Canada balsam in xylol. The preparation
may be dried either by gentle warming over the Bunsen
flame after the film has been blotted with filter-paper, or
the film may be allowed to dry spontaneously in the air,
in which case it should always be set up on edge to drain.
The preparations must be completely dried before being
mounted in balsam.

FILM STAINING 107

To prevent the stain flowing all over a slide, two lines
may be drawn across the slide with a grease pencil, one
on either side of the area to be stained.

If there be much debris or other material which, when
stained, would interfere with a clear view of the organisms,
various expedients may be adopted. One is to stain for
a short time with a solution which does not give a very
dense colour, the best for this purpose being Loffler's
methylene blue, or Gram's method may be made use of
if the organism stains by it, and will give the best result
of any. Another plan is to treat the specimen with acetic
acid before staining ; it may be just dipped in glacial
acetic acid and immediately washed in distilled water, or
immersed in 20 per cent, acetic acid for five to ten minutes,
washed in distilled water, and then stained. A third is,
after staining and washing, to rinse the preparation in
dilute alcohol (alcohol 1 part, water 1 or 2 parts), and
immediately to wash again in water to stop the further
action of the alcohol. If the film be thick, two or three
rinses in the dilute alcohol may be necessary. This process
gives excellent results with the sarcinae, but the staining
agent should be anilin gentian violet or dilute carbol-
fuchsin and not LofHer's blue, unless it is allowed to act
for fifteen to twenty minutes. The treatment with acetic
acid before staining may be combined with decolorisation
with alcohol after.

Preparations can always be examined in water with the
J-in. objective, after washing and before permanently
mounting, in order to see whether they are satisfactory.
If the film is on a slide, a drop of water is put on and
covered with a cover-glass, if on a cover-glass, this is
mounted in a drop of water on a slide. If satisfactory,
the preparation can be dried, and mounted in balsam ; or
if not stained sufficiently, or if stained too deeply, it can be
stained again, or further decolorised, as the case may be.

108 A MANUAL OF BACTERIOLOGY

Treatment of Sections for Staining and
Mounting

(a) Frozen sections. — If preserved in spirit they should
be rinsed in distilled water or in fresh alcohol before
staining, according as the staining solution is an aqueous
or an alcoholic one. After staining they are well rinsed
in water or alcohol to remove the excess of stain, and are
then dehydrated and cleared before being mounted. For
dehydrating, if they have been washed in water, they
should be well rinsed in methylated spirit l to remove the
excess of water, and then transferred to absolute alcohol
for a few seconds to two minutes, the time varying with
the size and thickness of the section. In many cases — for
instance, when the anilin dyes have been used for staining
• — the sections must be passed as rapidly as possible, con-
sistent with thorough dehydration, through the absolute
alcohol to avoid removing too much of the colour. If
it is important to avoid any decolorisation, anilin oil may
be used for dehydration, as in Weigert's method (pp. 105
and 106). For clearing, xylol or cedar oil is the best agent,
for neither dissolves the anilin dyes ; they will only clear,
however, out of absolute alcohol : hence the preliminary
rinsing of water- washed sections with methylated spirit
to prevent dilution of the subsequent bath of absolute
alcohol. Oil of cloves can also be employed, but has the
disadvantage that it dissolves the anilin dyes, and the
colour of stained sections treated with it is apt to be less
permanent ; it has the advantage, however, of clearing
out of methylated spirit, absolute alcohol being unnecessary.
The alcohol and clearing agents are conveniently placed
in watch-glasses or small shallow glass capsules. The

1 Absolute alcohol may of course be employed instead of the first
bath of methylated (or rectified) spirit, but methylated answers just
as well and is less expensive (but see note, p. 86).

PARAFFIN SECTIONS 109

section is known to be cleared when it appears quite
transparent and almost invisible when the watch-glass
or capsule containing it is held over a dark surface. If
after two minutes in the clearing agent the section still
appears cloudy and opaque, it has not been sufficiently
dehydrated, and should be returned to a fresh bath of
absolute alcohol for a short time, and then transferred
again to the clearing agent. Care should be taken that
watch-glasses, etc., used for the absolute alcohol and
clearing agent are perfectly dry. The clearing agent,
especially clove oil, can be used many times before
becoming useless.

For transferring the sections from one solution to
another an ordinary needle, fixed in a light wooden handle,
suffices, or, better still, a piece of glass drawn out at one
end, the section being carefully lifted by one corner to
prevent crumpling ; but for the final process of mounting
it is necessary to use a section lifter or cigarette-paper.
The section, spread out with care, is raised by means of
the section lifter or cigarette-paper introduced under it,
and transferred to the slide, any crinkles are removed
by spreading with a needle, the superfluous clearing agent
is drained off, a drop of xylol balsam put on, and it is then
covered with a clean cover-glass. If clove oil has been used
as the clearing agent, the section, after draining, should
be blotted with two or three thicknesses of filter-paper
to remove as much oil as possible before putting on the
balsam. In blotting firm pressure should be used, and
the section will then adhere to the glass slide and not to
the blotting-paper. With delicate sections all the pro-
cesses of staining, dehydrating, clearing, etc., may be
carried out on the slide.

(b) Paraffin sections. — The section fixed on the slide
(p. 93) must be freed from paraffin before staining and
mounting. The slides with attached sections are treated

110 A MANUAL OF BACTERIOLOGY

as follows : Immerse in (1) xylol for one or two minutes,
drain ; (2) absolute alcohol one to two minutes to remove
the xylol, drain ; (3) distilled water. They are now
ready for staining, and are to be flooded with the staining
solution or immersed in it, and after staining they are
treated in the same manner, but in the reverse order, viz.
(1) distilled water; (2) methylated spirit; (3) absolute
alcohol ; (4) xylol. On being removed
from the xylol the slide is drained for a
few seconds, a drop of xylol balsam is then
put on, and the section covered with a
clean cover-glass. Glass pots (Fig. 20}
filled with the alcohol, xylol, etc., are
convenient for the treatment of paraffin
sections, the slide with the section upon it
being immersed in the fluid.

Section Staining

When Gram's method is applicable it
FIG. 20.— Glass pot , , .,

for clearing, etc. gives by far the best results, and should

always be employed. If, however, the
organisms are decolorised in Gram's process some other
method must be adopted. One of the best is to
stain for from ten minutes to six or eight hours in
Lomer's methylene blue. Fresh easily staining organisms
will be sufficiently stained in ten or fifteen minutes, but
when the organism is difficult to stain, as glanders, six to
eight hours may not be too long a time. Warming intensi-
fies the staining properties of all staining solutions ; for
frozen sections the watch-glass of stain may be warmed
on a sand-bath or asbestos cardboard, or in the blood-heat
incubator. Sections on the slide may be flooded with the
stain and warmed on a piece of asbestos cardboard placed
over a Bunsen flame, or a penny may be heated in the

SECTION STAINING 111

Bunsen and the preparation laid on it, the coin being
re-heated as often as required. The stain may be pre-
vented from flooding the slide by confining it between
grease-pencil lines as described for films (p. 107). After
staining, the sections are well rinsed in distilled water
and then slightly decolorised by rinsing for half a minute
or so in a watch-glass of 1 per cent, acetic acid in distilled
water. They are then again washed and passed as rapidly
as possible through alcohol, cleared in xylol, and mounted.
Carbol-methylene blue or carbol-thionine blue may be
used instead of the Lofner's solution, the staining taking
from a few minutes to half an hour. If a contrast stain
be desired the sections may be treated for a few seconds
with the eosin solution after the dilute acetic. If staining
be prolonged evaporation must be prevented. In the
case of a section mounted on the slide and flooded with
stain, the slide should be placed on a piece of wet
blotting-paper on a tile and covered with the lid of a
Petri dish.

The micro-organisms in sections stained with Loffler's
blue are very liable to become decolorised unless the
dehydration is expeditiously performed. To avoid this
Unna's method may be adopted. After staining and
decolorising with acidulated water as described, the
sections are placed on the slide (if not already mounted
thereon), gently warmed, and so dried ; they are then
treated with xylol and mounted in balsam. The tissue
elements, however, are apt to suffer.

A better method is, after decolorising with the dilute
acid, to dehydrate with anilin instead of with alcohol,
the section being treated with fresh anilin two or three
times, then with a mixture of equal parts of anilin and
xylol, and finally with two or three baths of xylol.

112 A MANUAL OF BACTERIOLOGY

Capsule Staining

Many organisms, especially in the tissues or body fluids,
are invested with a capsule of gelatinous matter, probably
derived from the membrane of the bacterial cell, and
differing in composition in different species. The capsule
may be as thick as the bacterial cell itself, and appears,
in the unstained state or after staining by the ordinary
methods, as a clear halo or zone surrounding the organism.
Organisms in films of albuminous matter often appear to
be surrounded by a clear halo, which must not be mistaken
for a capsule. As organisms frequently lose their capsules
on ordinary culture media, Moore recommends cultivating
in fluid serum to obtain the re- development of the capsule.
In order to stain the capsule one of the following methods
may be adopted.

1. Stain the preparations by just dipping in the following solution :

Carbol-fuchsin . 1 part

Distilled water . . 1 part

Rinse in water and then stain for fifteen seconds in a very weak

aqueous solution of gentian violet (0-1 per cent.). Rinse in water,

dry, and mount.

2. McConTcey's method. — The following solution is prepared :

Methyl green . . .1-5 grm.

Dahlia .... .0-5 grm.

Distilled water . . . 100 c.c.

When dissolved, 10 c.c. of a saturated alcoholic solution of fuchsin
are added, and the whole is made up to 200 c.c. with distilled
water. The stain should not be used for a fortnight, and should
be kept in a dark place. Specimens are stained for five minutes
or longer, then thoroughly washed in a stream of water, dried, and
mounted.

3. Friedlander's method (for tissues).— Mix,

Concentrated alcoholic solution of gentian

violet . 50 parts

Distilled water . 100 parts

Acetic acid . .10 parts

Stain the sections in this solution in the warm incubator for

SPORE STAINING 113

twenty-four hours. Rinse well in 1 per cent, acetic acid, pass
through alcohol and xylol, and mount in balsam.

Spore Staining

When spore- bearing bacteria are stained by the ordinary
methods the spores are just tinted, or remain uncoloured
with the outlines more or less stained. This seems to be
due to the fact that the spores are surrounded with a
slightly permeable membrane which inhibits the entrance
of the staining agent. By staining by some method which
causes the penetration of the stain, and then cautiously
decolorising, it is possible to remove the colour from every-
thing except the spores, the impermeable membrane of
which in the same way prevents the full action of the
decolorising agent.

(a) Simple method. — A film is prepared in the ordinary way. If
a cover-glass, it is floated on a watch-glass, or, if a slide, it is flooded
with carbol-fuchsin, and the stain is warmed for twenty minutes.
After being washed in water the preparation is rinsed for a second
or two in 1 per cent, sulphuric acid and again washed at once in
water. If there is still a good deal of the red colour remaining,
the film may be once more rinsed in the acid, but if nearly colour-
less it should be mounted in water and examined with the £-in.
objective. If the spores alone are well stained the preparation
may be counter-stained with Loffler's methylene blue for two to
five minutes, washed, dried, and mounted. If, however, the
bacilli as well as the spores retain the red colour, the preparation
must be further decolorised in the acid, while if everything has
been decolorised, it may be re-stained with warm carbol-fuchsin.

The spores sometimes stain better if the preparation be fixed
by passing through the flame twelve times instead of three, as is
usual. To obtain good preparations and ones showing the spores
in situ, the specimens should be made as soon as spores have
definitely developed in the cultures.

Spore staining often requires a good deal of patience, and in
many instances it is difficult to obtain a satisfactory preparation
by this simple method, in which case that of MoeJler should be
made use of, and rarely fails.

8

114 A MANUAL OF BACTERIOLOGY

(b) Moeller's method. — Prepare the cover-glass or slide specimen
in the ordinary way. Treat with absolute alcohol for two minutes,
and then with chloroform for two minutes. Wash in water and
treat with a 5 per cent, solution of chromic acid for two minutes,
wash, and then stain with warm carbol-fuchsin for ten minutes.
Wash, decolorise carefully in 1 per cent, sulphuric acid, again wash
and counter-stain with Loffler's methylene blue for one minute ;
wash, dry, and mount. Some organisms, such as the B. mesen-
tericus, stain better if treated with the chromic acid for five to ten
minutes.

Flagella Staining

Many organisms possess delicate protoplasmic processes
— flagella — in greater or less number ; but these are not
visible when the organism is examined in the living con-
dition (except by the use of dark-ground illumination),
nor when the ordinary staining methods are employed.
In order to demonstrate them it is necessary to make use
of some special method, in which a mordant is essential.
One of the earliest devised was that of Loffler, which with
care gave fair results. It is not, however, nearly so
satisfactory as some more recent ones, so is omitted.

For all methods of flagella staining the cover- glasses or
slides must be absolutely clean, the cultures recent, and
the growth sufficiently diluted to obtain the organisms in
an isolated condition.

(a) Stephens's method. — This is a modification of the well-known
Van Ermengem method,1 and has been communicated to the writer
by Dr. J. W. W. Stephens.

To clean slides. — Rub the slides with a clean cloth and place on
a piece of clean wire gauze and heat with a smokeless flame for
some minutes (by this means grease is completely removed).
Remove the slides when cool, not before.

To make the suspension. — All methods are unsatisfactory. Rub a
little of the culture in a small drop of tap-water in a watch-glass.
Then transfer a drop with the smallest possible platinum loop to
a minute drop of water on the slide. Mix and spread with the

1 Cenlr. f. Bakt., xv, 1894, p. 969.

FLAGELLA STAINING 115

platinum wire as quickly as possible. The film thus made should
dry immediately if a small drop only of water has been used.

Age of the culture. — A twenty -four hours' culture does quite well
(a younger one is perhaps better, but flagella can be shown for a
week or fortnight or more).
I. The mordant :

Osmic acid, 2 per cent. . . .1 part
Tannic acid, 20 per cent, watery solution 3 or 4 parts
II. Silver solution : Silver nitrate . . 1 per cent.

III. Gallic acid, 2 per cent, solution . . 1 part
Ammonia fort. ... .1 part

To be mixed before using and to be used immediately.

To stain. — Place the mordant on the film for one or two minutes
or less (time unimportant).

1. Wash in tap- water thoroughly.

2. Shake off as much water as possible.

3. Place a few drops of silver nitrate on the slide for a few seconds
or longer.

4. Shake off all excess.

5. Allow one drop of the ammonia-gallic solution to fall on the
middle of the slide from a small pipette. A wave spreads away
from the centre to each end of the slide. As soon as the film is
seen standing out clearly and black in the centre (in a few seconds),
wash off in tap-water.

6. Add again a drop or two of the silver solution and allow it to
act for half a minute or thereabouts.

7. Wash in tap-water, blot, and dry over the flame.

8. It is best not to mount in balsam or in cedar-wood oil, as the
preparations rapidly fade in these.

If done with any care, the film should now appear black and
distinct to the naked eye with no precipitate, and the flagella will
be found to be stained distinctly and intensely with hardly any
ground substance, or at least insufficient to interfere with a clear
view of them.

(6) Pitfield's method. — Two solutions are freshly prepared :

A. Saturated aqueous solution of alum . . 10 c.c.
Saturated alcoholic solution of gentian violet . 1 c.c.

B. Tannic acid ....... 1 grm.

Distilled water . . . . . .10 c.c.

The solutions should be made with cold water, filtered, and

116 A MANUAL OF BACTERIOLOGY

preserved in separate bottles. For use equal quantities are mixed
together. The specimens are flooded with the mixture and held
over the flame until it nearly boils ; they are then laid aside, with
the hot stain on them, for one minute, and are finally washed in
water. After washing, the preparations are flooded with anilin
gentian violet for one second, washed in water, dried and mounted,
(c) McCrories method x (modified by Morton 2). — Prepare the
following solutions :

A. Tannic acid ....... 1 grm.

Potash alum ...... 1 grm.

Distilled water . . 40 c.c.

B. " Night " blue . ... 0-5 grm.
Absolute alcohol . . . . . .20 c.c.

Mix and filter.

The prepared slides are stained with this solution (which should
always be filtered before use) for two minutes, the solution being
changed two or three times, washed gently in running water, and
then counter-stained in anilin gentian violet for one to two minutes,
washed, dried, and mounted.

Preservation of Cultures

Gelatin and agar cultures may be satisfactorily preserved by
submitting them to the action of formaldehyde vapour for some
hours by soaking the wool plug of the culture tube in formalin and
plugging the tube with it. The tube may then be sealed with
gutta-percha tissue, sealing-wax, or paraffin wax, or best of all in
the blowpipe flame. Plate cultivations may also be exposed to the
vapour and the lid of the dish afterwards cemented on, or the
cultures may be made in the flat bottles ("Soyka's bottles")
devised for the purpose, and after development treated like tube
cultures.

Preservation of Pathological Specimens

These may be preserved in the ordinary way in spirit, but a much
better method, by which the natural colour of the specimen is
retained, is the following. The specimens are first washed in water,

1 Brit. Med. Journ., 1897, vol. i, p. 971.

2 Trans. Jenner Inst. Prev. Med., vol. ii, p. 242.

PRESERVATION OF SPECIMENS 117

and then placed in the following solution for twenty-four to forty-
eight hours :

Formalin ...... 6 parts

Sodium chloride ..... 1 part

Sodium sulphate ..... 2 parts

Magnesium sulphate .... 2 parts

Tap- water .... .100 parts

After being taken from the formalin solution the specimens are
placed in methylated spirit for ten minutes, and then in a fresh
bath of methylated ; in this the colour to a large extent returns,
and they should be carefully watched and not allowed to remain
in it for more than an hour. They are then mounted in the
following mixture :

Glycerine ...... 400 c.c.

Potassium acetate ..... 200 grm.

Water 2000 c.c.

A trace of formalin should be added to this.

The writer has preserved meat infected with B. prodigiosus very
satisfactorily by the following method. Slices were cut off and
placed in the formalin solution given above for a few hours. They
were then well drained and placed in suitable glass capsules.
Ordinary nutrient gelatin was melted and sufficient poured in to
cover the specimens, and when it had set a little formalin was
poured on and allowed to remain for a few days. It was then
poured off and the glass top cemented down.

For further information on preparation of tissues, section cutting,
staining methods, etc., see The Microtomisfs Vade-Mecum, Bolles-
Lee ; Practical Histology, Schafer ; Methods of Morbid Histology
and Clinical Pathology, Walker Hall and Herxheimer ; and Lehrbuch
der Mikroskopischen Technik, Rawitz.

CHAPTER IV

METHODS OF INVESTIGATING MICROBIAL DISEASES—
THE INOCULATION AND DISSECTION OF ANIMALS-
HANGING -DROP CULTIVATION — INTERLAMELLAR
FILMS— THE MICROSCOPE

THE systematic study of a condition dependent on the
activity of micro-organisms is in many instances no light
matter. When only one or two forms are present and
these are readily cultivated it may be comparatively
easy, but when there are many the investigation may
become exceedingly complicated. The first step to be
taken is to ascertain by careful microscopical examina-
tion the general characters of any organisms that may
be present in the material, and their distribution both
in the fresh condition and in stained preparations, and
if possible at different stages of the disease. In disease
conditions, for example, the blood and secretions may be
examined both before and after death, but in the latter
it must be remembered that soon after the fatal event
adventitious organisms rapidly make their appearance,
gaining access from the air and from the intestinal tract.
If organisms be detected an attempt should be made to
determine whether there is any predominant form and
if this is constantly present at different stages. If
organisms are found, it simplifies matters, but if not, it
cannot therefore be said that they are absent, for they
may be few in number, and consequently be missed in a
microscopical examination ; or they may be confined to

118

INVESTIGATION OF MICROBIAL DISEASES 119

a particular locality or tissue, or are present only at one
stage of the infection. In addition to the microscopical
examination, cultures must be made on various media,
those media being chosen which will probably be suitable
for the growth of the organism present in the particular
condition ; for example, in the examination of animal
diseases, media rich in protein, such as blood-serum,
nutrient agar and gelatin, will be the most serviceable.
In the examination of plant diseases, vegetable infusions
prepared from the plant itself or from other sources, and
enriched by the addition of vegetable proteins, and carbo-
hydrates, should be chosen. In fermentations, beer-wort,
grape or fruit juice, and saccharine solutions should be
made use of ; while for the nitrifying organisms, solutions
containing nitrates and nitrites, salts of ammonia, urea,
and asparagin will have to be employed. In addition,
it will in most cases be advisable, and in all safer, in order
to isolate the various species, to make plate cultivations,
either in Petri dishes (p. 78), or by streaking several sloped
tubes of agar, etc. (p. 81). Having obtained pure cultiva-
tions it will be necessary to determine the species of
organism,1 if it has been previously isolated and described,
or to give a careful description of it, if it be a new one,
for the use of subsequent investigators. In the identifica-
tion or description of an organism all the following features
must be carefully noted :

1. The morphology of the organism under various conditions, its
size, form, and motility, the presence of flagella, and their number,
arrangement, and character.

2. The presence or absence of spore formation, its nature, the

1 The descriptions of a large number of species of bacteria have been
collected and tabulated in convenient form by Chester in A Manual
of Determinative Bacteriology (Macmillan and Co., 1901). The terms he
suggests for describing bacterial growths, etc., might well be adopted
by bacteriologists. A committee of the Society of American Bacterio
legists has drawn up an elaborate chart for the description of species
of organisms.

120 A MANUAL OF BACTERIOLOGY

conditions under which it occurs, and any peculiarities in the
germination of the spores, and their size and location in the cell.

3. The peculiarities of staining, and the staining reaction with
Gram's and the Ziehl-Neelsen methods.

4. The characters of the colonies in gelatin, agar, and other
media, both surface and deep.

5. The characters of the growth on a variety of culture media
at different temperatures — for example, for a pathogenic organism,
on blood-serum, agar, and gelatin (surface and stab cultures), in
broth and on potato ; liquefaction or not of the gelatin ; the growth
in milk, with or without curdling, and the reaction therein ; and
the fermentation reactions on carbohydrates, glucosides, alcohols,
etc. ; the nature of the gas, if any, formed therefrom, and the
H:C02 ratio.

6. The behaviour towards oxygen — is it aerobic or anaerobic ?

7. The range of growth at different temperatures.

8. The reducing power by growing in litmus broth which becomes
decolorised, or by the formation of nitrites in a solution containing
nitrates.

9. The production of indole with or without nitrites.

10. The production of pigment and the conditions under which
it occurs.

11. The pathogenic action on various animals if it be a disease
germ, or the changes which it produces if it be an organism connected
with other conditions.

12. The chemical changes which it induces.

13. The thermal death-point and the action of germicides and
antiseptics upon it (see Chapter XXII).

For descriptive purposes, " standard " culture media
should always be employed, and the acidity or alkalinity
of the medium stated (p. 64).

It must never be forgotten that under cultivation the
properties of organisms may be considerably modified, and
due allowance must be made for this. For example,
pathogenic organisms may lose their virulence more or
less completely, pigment production be lost, and fer-
mentive action modified (see also p. 6).

To obviate these difficulties the organisms should be
cultivated under as nearly natural conditions as possible
and sub- cultivation avoided so far as can be. No general

COLLODION SACS 121

rule can be given as to the duration of life of cultures on
artificial media. Most organisms will retain their vitality
for at least three or four weeks without being transferred
to a fresh soil, some for many months ; a few must be
sub- cultured every week, or they will die out ; while
there are still a small number which have so far rarely or
never been cultivated. On the whole, organisms retain
their vitality best on gelatin.

For an organism to retain its virulence it is, as a rule,
necessary to pass it through a susceptible animal at longer
or shorter intervals, and to enhance the virulence recourse
must be had to a succession of passages through susceptible
and then less susceptible animals. In this way the viru-
lence of organisms has been increased to a point far greater
than is ever met with naturally, as in the case of the
Streptococcus pyogenes. If an organism retains its virulence
even slightly, it is generally possible, by employing large
doses, to enhance this by passage through a susceptible
animal. Another method may also be adopted, namely,
to inject along with it some other pathogenic form, such
as the Streptococcus pyogenes ; the combination will kill
the animal, and the slightly virulent organism can be
recovered and will be found to have increased in virulence.
A third method is to inject the organism into a susceptible
animal together with a lethal dose of toxin obtained from
a virulent form of the same species, or with some substance,
such as lactic acid, which lowers the vitality of the tissues.
The slightly virulent organism will then be able to grow
under the more favourable conditions, and a form which
has become completely non-virulent can be made to
regain its lost virulence.

Collodion sacks are now frequently used to study the
action upon animals of the dialysable products produced
by micro-organisms which do not form any appreciable
amount of toxin in vitro, for cultivating species which

122 A MANUAL OF BACTERIOLOGY

are difficult to grow by ordinary methods, for studying
the phenomena of infection when the micro-organisms are
protected from the phagocytes, and for other purposes.
A glass rod or small test-tube, according to the size desired,
is dipped into a beaker containing the ordinary (not
flexible) collodion, is then withdrawn and allowed to dry,
and the process is repeated two or three times. In order
to detach the collodion from the glass, the whole is dipped
for a few seconds alternately into strong spirit and into
water, the collodion loosens, and may be easily peeled
off the glass. The sack may be sterilised by placing in a
test-tube and heating to 150° C. in the hot-air steriliser.

For the inoculation of animals various methods may be
adopted. Thus, after clipping the hair, the organism may
be introduced by rubbing into the skin after scarification,
or, a small incision having been made through the skin,
a small quantity of a culture may be introduced on a
platinum needle ; or a broth culture or an emulsion, made
with sterilised water or broth, may be injected with a
sterilised syringe subcutaneously, intra-peritoneally, or
into the muscular or other tissues or organs as required,
since the seat of inoculation may have to be varied for the
different species to produce their pathogenic effect. For
injection purposes a syringe like an antitoxin syringe,
i.e. with asbestos or metal piston and glass barrel that
can be boiled, may be used. Several sizes, 1 c.c., 2 c.c.,
and 5 c.c. at least, are required. An all-glass syringe is
a still better form, but is expensive. For accurate dosage,
the piston-rod should be graduated and have a nut
travelling on a screw up and down it. Before use the
syringe with the needle should be boiled for ten minutes
to sterilise it ; after use it may be well rinsed and again
boiled. The needles should be wiped dry and a wire
inserted, or they may be kept in a bottle of xylol.

Guinea-pigs and rabbits are usually inoculated in the

ANIMAL INOCULATION 123

thigh or abdomen ; mice in the dorsal region or at the
root of the tail (dorsally), the hair being clipped, and
the skin disinfected, but this is not generally necessary.
Numerous mechanical holders have been devised for
animals, but are not as a rule required. Rabbits may
be inoculated intra-venously by one of the large veins
in the ear. The ear is -shaved, and the skin is well washed
with a little alcohol with vigorous rubbing ; the base of
the ear is lightly pinched so as to obstruct the venous
but not the arterial circulation, and render the vein
prominent, and the injection is made with a small syringe
fitted with a fine needle, the needle being passed into the
vein towards the base of the ear. After the withdrawal
of the needle the wound is compressed for a little and may
be dressed with some antiseptic wool and collodion.

Guinea-pigs frequently eat the carcases of their dead
companions, so that the cages should be examined twice
daily, and, if the carcase is required, it may be advisable
to keep each animal in a separate cage.

The phenomena occurring after inoculation must be
noted. Usually these are not very obvious in the rodents,
but loss of appetite, sluggishness, staring coat, convul-
sions, etc., may be observed. The weight of the animal
is a good index of what is happening. If the infection is
serious, the weight rapidly falls ; if the animal is to
recover, its weight soon begins to increase after the pre-
liminary fall. The temperature in the rectum may also
be taken, but is not so valuable, as in the guinea-pig
variations occur from mere handling or other slight causes.
The temperature of the guinea-pig averages 38-6°, but
varies between 36° and 39° C. (Eyre).

The examination of the dead animal should be carried
out with as little delay as possible. For dissection, the
body should be pinned out on the back on a board, which
may stand in a shallow enamelled iron pan, by pins or

124 A MANUAL OF BACTERIOLOGY

nails through the feet, and the abdomen well soaked with
antiseptic solution, not so much to sterilise the skin as to
prevent the hair from getting into the incision ; to obtain
complete sterilisation of the skin, it is preferable to clip
or shave the hair and then sear with a red-hot iron.
Knives, forceps, scissors, etc., should be well boiled in an
enamelled iron mug or pie- dish, the water being kept
boiling during the progress of the dissection and the
instruments rinsed from time to time in it. A little sodium
carbonate may with advantage be added to the water.
A small enamelled iron fish-kettle with perforated strainer
forms an excellent steriliser for instruments, or a surgical
instrument steriliser may be used. An incision is made
and the skin well reflected and pinned out ; the knife and
forceps should then be re-sterilised, or fresh sterile instru-
ments taken, for the deeper incision and opening the body
cavities ; these again must be re-sterilised, or a third set
of instruments employed for incising the organs.

During the progress of the dissection the condition of
the tissues at the seat of the inoculation should be noted,
and likewise the conditions of the serous membranes and
the various organs. In many diseases the organism is
met with most abundantly in the spleen, in others in the
blood, and in some at the seat of inoculation. When a
systematic examination is made, film specimens and
cultures on two or three media, aerobic and anaerobic,
should be prepared from the seat of inoculation, the spleen,
liver, lungs, and heart-blood, and in some cases from the
serous membranes, muscles, or central nervous system
in addition, the carcase being in the intervals covered
with a bell- jar which has been rinsed in, or with filter-
paper moistened with, antiseptic solution. An assistant
is often useful or even necessary. The greatest care must
be taken to avoid dropping or splashing or otherwise
disseminating infective material, any stains being im-

POST-MORTEM EXAMINATION 125

mediately swabbed up with antiseptic solution ; and the
operator must exercise every precaution to prevent the
infection of himself and others. It is convenient to have
some efficient antiseptic solution near at hand ; it may be
kept in a large bottle on a wall bracket and drawn off as
required by a syphon tube provided with a tap or spring
clip. The most generally used antiseptics are 5 per cent,
carbolic, and 1-500 corrosive sublimate, but 2 per cent,
cyllin or kerol or 3 per cent, lysol is cheaper and more
efficient. The access of flies to the carcase must also be
prevented, as they might carry infection. When finished
with, the carcase should be efficiently disinfected and
disposed of without delay, preferably by burning it,
together with the board on which it has been pinned out.

If the carcase be left, especially in warm weather, for
even a few hours before the examination is carried out,
the tissues are liable to become invaded and infected by
organisms from the respiratory and digestive tracts. In
the post-mortem room, infection of the tissues is very
common ; out of fifty cases, Symes1 found only seventeen
to be sterile. Ford states that even in normal animals,
killed and immediately examined, bacteria are present
in 70 per cent, of the internal organs.2

When the blood of an animal is required several ex-
pedients may be adopted. From large animals, like the
horse, sheep, and goat, it may be obtained by passing
the needle of a large syringe into the external jugular
vein (which runs superficially on either side of the under
part of the neck) and then aspirating with the syringe.
In the case of small animals not again needed, the animal
may be decapitated or the throat may be cut, and the
blood collected in a porcelain dish ; but if a sample only
is wanted, and the animal has to be further treated, as

1 Lancet, 1899, vol. i, p. 365.

* Journ, of Hygiene, vol. i, No. 2, 1901, p. 277.

126 A MANUAL OF BACTERIOLOGY

in antitoxin work, it is generally possible to bleed from a
superficial vein. The needle of a syringe may be passed
into the heart of a guinea-pig and 2-3 c.c. of blood with-
drawn without injury to the animal. In the rabbit blood
may be obtained by passing the fine point of a piece of
glass tubing, drawn out and bent to a convenient angle,
or the needle of a syringe, into one of the ear veins
and aspirating the blood into it. Or the vein may
be punctured and the blood allowed to drip into a small
tube.

Blood may be obtained from a patient for the aggluti-
nation reaction, for microscopical examination, or for
culture experiments, by pricking the finger or the lobe of
the ear with a sterile needle, preferably a flat one of the
" Hagedorn " type, or with half a steel pen (nib) or a
glass point ; for disinfection, the skin may be rubbed with
a little alcohol or ether. After swinging the arm and
winding a piece of rubber tubing round the finger or
thumb and pricking 1-3 c.c. may generally be obtained.
The blood may be collected in a small test-tube, vaccine
tubes, small bulbous tubes (Fig. 7, p. 52), or Wright's
tubes (Fig. 35, p. 215).

If the tube with contained blood is sealed in the flame,
and is afterwards centrifuged to obtain clear serum, care
should be taken that one end is not wetted with the blood,
and this dry end should be sealed first so as to obtain a
perfect seal, When centrifuging, this sealed end should
be placed downwards in the centrifuge " bucket."

Organisms, in natural infections in man, are usually
present only in small numbers in the blood, and for
demonstrating them by culture methods it is necessary
to withdraw 2-5 c.c. from a superficial or deep vein by
means of a sterile syringe under aseptic conditions, and to
inseminate broth tubes or agar plates each with 0'5 c.c. of
the blood.

EXAMINATION OF LIVING ORGANISMS 127

Although the modern methods of isolation and cultivation have
rendered immense service to bacteriology, they have also had the
effect of diminishing the attention paid to the exact morphology
and biology of organisms. At the present time there is a tendency
to investigate bacteria en masse rather than to study them as
individual living forms. As the late Marshall Ward remarked :

" The introduction and gradual specialisation of Koch's method
of rapid isolation of colonies encouraged the very dangers they were
primarily intended to avoid. It was soon discovered that pure
cultures could be obtained so readily that the characteristic differ-
ences of the colonies in the mass could presumably be made use of
for diagnostic purposes, and a school of bacteriologists arose who
no longer thought it necessary to patiently follow the behaviour of
the single spore or bacillus under the microscope, but regarded it
as sufficient to describe the form, colour, markings, and physiological
changes of the bacterial colonies themselves on and in different
media, and were content to remove specimens occasionally, dry and
stain them, and describe their forms and sizes as they appeared
under these conditions. To the botanist, and from the point of
view of scientific morphology, this mode of procedure may be
compared to what would happen if we were to frame our notions
of species of oak or beech according to their behaviour in pure
forests, or of grass or clover according to the appearance of the
fields and prairies composed more or less entirely of it, or — and this
is a more apt comparison, because we can obtain colonies as pure
as those of the bacteriologist — of a mould fungus according to the
shape, size, and colour, etc., of the patches which grow on bread,
jam, gelatine, and so forth."

Examination of Living Organisms

One essential procedure in the investigation of an
organism is its examination in the fresh and living con-
dition. This may be done by placing a droplet of sterile
water, broth, or salt solution on the slide, inoculating
with a trace of the material or growth, and covering
with a cover-glass and examining microscopically. The
action of stains and reagents on the organisms may be
observed by the irrigation method. A drop of the stain
or reagent (c, Fig. 21) is placed on the slide, A, just in

128

A MANUAL OF BACTERIOLOGY

contact with one margin of the cover- glass, B, and is
drawn through the preparation by means of a small piece
of filter-paper, D, placed on the other side, a torn margin
touching the film of fluid at one edge of the cover- glass.

The filter-paper absorbs the fluid from under the cover-
glass, leaving the cells and other particles behind, and at
the same time the reagent on the opposite side flows under
the cover- glass to take the place of the absorbed fluid.
Afterwards the excess of the reagent or stain may be

A

B

FIG. 21. — Method of irrigation.

washed away by running in water under the cover- glass
in a like manner. Care must be taken that no fluid gets
on to the upper surface of the cover-glass, which must
always be kept dry. The advantage of this method is
that it may be applied while the specimen is being examined
under the microscope, and the action of the reagent on a
particular cell or granule can, with a little care, be watched.
If the cells be large and it is desirable to avoid pressure
of the cover-glass, a fine hair or bristle may be so placed
on the slide that when the cover- glass is lowered one
edge rests on it. If the specimen has to be kept for
any length of time, the film of fluid will before long
evaporate and the preparation become dry. To prevent
this a ring of oil or vaseline may be painted round the
margin of the cover-glass so as to seal it to the slide.

A simple method for keeping organisms under examina-
tion for a lengthened period of time, and of watching

HANGING-DROP PREPARATIONS 129

their growth and development, is by the use of hanging
drop preparations. To prepare a hanging drop, a ring of
vaseline is painted round the margin of the hollow of a
hollow-ground slide (or other cell, see below). A cover-
glass is sterilised by flaming in the Bunsen, care being
taken not to heat sufficiently to melt it. A droplet of
some sterile fluid medium — water, broth, wort, sugar
solution, etc. — is then placed in the centre of the cover-
glass with a sterile platinum loop. This droplet is then
inoculated with the organism which is to be observed,
care being taken not to add too many organisms — a few

FIG. 22. — Hanging-drop preparation.

isolated organisms and small groups in each field is what
should be aimed at. The vaselined cell is now taken and
turned over, so that the ring of vaseline is downwards,
and is then applied to the cover-glass, in such a way that
the droplet is situated in the middle of the hollow, but not
touching the slide at any point. The cover-glass adheres
to the slide by means of the vaseline, and on quickly
inverting the whole, so that the fluid has no time to run,
it will be found that the droplet is hanging from the
under surface of the cover- glass in a cell which is hermeti-
cally sealed by the vaseline, and evaporation is thus
rendered impossible (Fig. 22). Such a preparation, in
fact, can be kept for a week or ten days in a warm incubator
without drying up. Great care must be exercised in
examining a hanging-drop specimen microscopically,
especially with the immersion lenses, for the slightest
pressure breaks the unsupported cover-glass. It often
saves time first to centre the drop with the low power
before examining with the immersion lens ; an ink or
pencil dot at the margin of the drop aids focussing. The

9

130 A MANUAL OF BACTERIOLOGY

light must be diminished by closing the diaphragm, lowering
the condenser, etc. (p. 132), and artificial light is generally
preferable to daylight. The central parts of the drop
only should be examined, not the margin.

Instead of hollow slides, various devices may be em-
ployed to form the cell. Metal, glass, or vulcanite rings,
or rings cut out of thin sheet lead, tin-foil, cardboard, or
two or three thicknesses of paper or filter-paper may be
cemented on to slides with vaseline, Hollis's glue, gold
size, or Canada balsam, or a thick ring of vaseline, or
paraffin, or plasticine may be used.

The only certain method for ascertaining whether an
organism is motile or not — often an important clue to its
identification — is by the use of hanging-drops. Actively
motile organisms may frequently assume a non-motile
resting stage, although still alive, and various factors
may bring about this condition, such as old age, exhaustion
of nutriment, excessive heat or cold, electric shocks, and
the like. The absence of movement of an organism in a
specimen prepared from an ordinary culture, particularly
if more than a day or two old, does not necessarily prove
that it is non-motile. A hanging-drop should be prepared
with a nutrient medium (the best, perhaps, is glucose
broth) and placed under conditions of temperature, etc.,
favourable to the growth of the organism, and examined
after an interval of an hour or so, or better still at intervals
of half an hour for three or four hours. In this time the
old cells will revivify, and new ones will have been pro-
duced, and if the organism be a motile one, more or less
active movement of some of the cells is almost sure to be
observed. It is necessary to beware of two fallacies in
connection with motility — not to mistake for it the so-called
Brownian movement, which is a vibratory one back-
wards and forwards about one point, and common to
all fine particles suspended in a fluid ; and not to be

INTERLAMELLAR FILMS 131

misled by a flotation of the cells due to currents set up
in the fluid from some cause or other— all the particles
then tending to move in the same direction.

Another purpose for which the hanging-drop cultivation
may be employed is that of obtaining a permanent record
of the various phases through which an organism may pass
during its development. If a number of these cultivations
be made, say twenty, in an exactly similar manner, and
afterwards kept under identical conditions, and if at the
end of every half-hour one of the preparations be taken,
its cover-glass carefully removed, and the droplet dried
and stained, a permanent record of the life-history of the
organism is obtained extending over ten hours.

Various more elaborate forms of cells for hanging-drop
preparations can be obtained, some being provided with
inlet and exit tubes for the passage of various gases. For
anaerobic preparations cells are made having a groove
at the bottom into which a mixture of pyrogallic acid
and potash is introduced.

The observation of hanging-drop cultivations at
blood-heat can be carried out on some form of warm
stage.

Interlamellar films.1 — Another method of investigating the life-
history of organisms, especially moulds and protozoa, is by means
of interlamellar films. A glass slide 1| by 3 in. is sterilised in the
Bunsen flame, and while hot three small drops of sealing-wax are
placed on it, so arranged that they form the apices of an equilateral
triangle, the side of which measures about one inch, and a drop
of sterile nutrient medium is deposited between them. A cover-
glass of about 1^ in. in diameter is then sterilised in the Bunsen
flame, a droplet of a suitable nutrient medium is placed upon it
and inoculated with the organism to be observed, and the pre-
pared cover-glass is picked up with sterilised forceps, inverted,
and lowered on to the slide. The nutrient medium is thus contained
between the slide and the cover-glass, and by using a hot wire,
and so softening the sealing-wax, it can be spread out to form as

1 Delepine, Lancet, 1891, vol. i, June 13.

132 A MANUAL OF BACTERIOLOGY

thin a layer as desired. The preparation is kept in a moist chamber
to prevent evaporation, and can be studied when required.

The Microscope

A bacteriological microscope is generally of the mono-
cular form, and should be provided with a rack-and-pinion
coarse adjustment and an efficient fine adjustment. The
stage, preferably of vulcanite, should be large and roomy
and quite plain, with two or more holes at its margin to
receive spring clips for fixing the slide. For the ordinary
examination of specimens a mechanical stage is not needed ;
in fact it hampers that freedom of manipulation which
is so useful for the rapid examination of a specimen. For
some purposes a mechanical stage is very useful, and for a
critical survey of the whole of a specimen, e.g. a blood-
film, it is essential. A detachable form is to be preferred
(Fig. 23), so that, if required, the stage may be free for
the examination of plate cultivations, etc.

New forms of binocular microscopes have recently been
introduced by Messrs. Beck and by Messrs. Leitz which
possess marked advantages over the monocular instru-
ment.

A sub- stage condenser is essential for all work in which
high powers are employed, and also enhances the value of
low powers. It consists of a system of lenses below the
stage, by means of which the light is concentrated on
the object. It should have a rack-and-pinion, or a screw,
adjustment for focussing, and be provided with some
form of diaphragm for modifying the light, preferably
an " iris." The condenser must be centred — that is,
adjusted so that its optical axis corresponds with the
optical axis of the objective ; and for this purpose it ought
to be provided with two lateral screws working at right
angles to each other, by means of which its position

THE MICROSCOPE 133

relative to the optical axis can be altered. In order to
centre, a diaphragm with small aperture is used, and the
hole in the diaphragm is focussed with a low power ; then,
by means of the lateral screws, this hole is brought into
the centre of the field. Below the sub-stage condenser
a mirror with concave and plane surfaces should be fitted,
the plane surface being used with the condenser, as a

FIG. 23. — Swift's detachable mechanical stage.

general rule. The concave mirror may be used for illumi-
nation with low-power objectives, the condenser being
detached or swung out of position. The necessity for
careful illumination must be insisted upon ; in fact, to
obtain the best results the light should be readjusted for
every specimen by mirror, diaphragm, and condenser,
i.e. " critical " illumination should be aimed at. A good
specimen may be utterly spoilt, visually, by faulty illumi-
nation ; while an indifferent one may be made to look
passable by proper illumination. In the examination of
micro-organisms in the fresh or living and unstained
condition, it is necessary, as a rule, to diminish the light
by means of a small diaphragm, or by racking down the
condenser, or by both ; while for stained or opaque objects

134 A MANUAL OF BACTERIOLOGY

the full aperture of the diaphragm, or thereabouts, may
generally be employed. It must be remembered, however,
that the resolving power of a lens (see below) is diminished
by closing the diaphragm and by throwing the condenser
out of focus ; the illumination then becomes " non-critical."
For fine work, if the illumination is too intense, this should
be diminished by diminishing the source of light or by
interposing a coloured screen, such as Gifford's, which
consists of a cell containing a solution of malachite green
in which is inserted a piece of green signal glass. Coloured
glass may also be interposed. The microscopist should
accustom himself to examine specimens both by daylight
and by artificial light ; hanging-drop specimens are
usually best seen with the latter. For artificial light,
probably nothing surpasses a paraffin lamp with flat
wick, the edge of the flame being always used, while to
obtain the best results the mirror should be removed, and
the flame used direct by elevating and tilting the micro-
scope somewhat. For the finest work, daylight illumina-
tion is inadmissible. An admirable form of electric lamp
is the " Barnard," made by Messrs. Swift and Son, the
source of illumination being a Nernst lamp. For ordinary
routine work, an incandescent carbon or metal filament
electric lamp, a Nernst lamp, or an argand or incan-
descent gas burner may be used. Various devices have
been introduced for the employment of monochromatic
illumination, e.g. the quartz mercury vapour lamp by
Barnard.

With the filament, Nernst, or incandescent gas, lamps,
the image of the filament or mantle is troublesome when
the condenser is in focus ; this may be obviated to some
extent by the use of frosted bulbs or by interposing a
screen of fine ground glass, by the use of Gordon's glass rod
illuminator, or by interposing a spherical flask filled with
water or dilute copper sulphate solution. Incandescent

THE MICROSCOPE 135

bulbs may be frosted by dipping in a 15 per cent, solution
of caustic soda and allowing to dry.

Two eyepieces are sufficient, and the lower-power ones
are to be preferred, such as the B and c of the English,
or the 2 and 3 of the Continental, makers. Although
increased magnification can be obtained by the use of a
high-power eyepiece, it is at the expense of definition,
the image losing its sharpness, because the eyepiece mag-
nifies the image formed by the objective, and any imper-
fections in the latter are made more apparent, so that the
use of very high eyepieces is not to be recommended,
except with the finest lenses ; moreover, as will be pointed
out later, it is useless to increase the amplification beyond
a certain point.

With regard to the length of the tube of the microscope,
this differs in the English and Continental systems. The
standard English tube-length is 8-75 in., the Continental
is 6-3 in., and is usually adopted, but the longer tube gives
greater amplification. The tube of the microscope is
generally provided with an inner, or draw-tube, by means
of which its length can be nearly doubled ; this gives
increased amplification, but at the expense of definition,
at least with the higher powers which are corrected or
adjusted for a definite tube-length.

The lenses or objectives must next be considered.

For powers higher than the J-in., or thereabouts, it is
advisable, for many reasons, to employ the immersion
system of objectives. With these lenses a drop either of
water, in the water-immersion system, or of cedar oil, in
the oil- immersion one, is placed on the cover- glass, and
the objective is racked down so that its front lens touches
and is immersed in either the water or oil, as the case
may be. It is a good plan then to raise the objective very
slightly by means of the coarse adjustment, still, however,
keeping it in contact with the drop of water or oil. The

136

A MANUAL OF BACTERIOLOGY

observer then, looking down the microscope, very cautiously
and gradually racks down again with the coarse adjust-
ment until the object comes into view, and finishes the
focussing with the fine adjustment. The fine adjustment
should only be used after the object has been brought into
view by means of the coarse adjustment. After the
examination has been concluded for the day, the lens

Cl

y

FIG. 24. — Diagram to illustrate the refraction of light.

should be carefully wiped with a soft rag, or preferably
with a piece of soft Japanese paper, to remove the water or
oil. If the oil should happen to dry on the lens, it may be
removed by wiping with a soft rag or Japanese paper moist-
ened with xylol, quickly drying with another rag or paper.
Instead of cedar-oil, a liquid paraffin has also been used.

The T^ in. (2 mm.) oil-immersion lens is the one usually
selected. It combines sufficient magnification for most
purposes with adequate working distance for convenience
in using. If expense is not an object, the Zeiss J in.
(3 mm.) apochromatic oil-immersion lens is a very fine one
for general use. By means of the compensating oculars

THE IMMERSION SYSTEM

137

sufficient magnification can be obtained, while the working
distance is greater, the field is larger, and the penetrative
power is greater than with the ^ in. lens.

The immersion system of objectives has many advantages : the
loss of light is less, the distance between the cover-glass and the
front of the objective — the working distance, as it is termed — is
greater, and more can be seen with an immersion lens than with

S-.

FIG. 25. — Diagram to illustrate tlje course of rays of light
through an objective.

a dry lens of equal magnifying power. This can be best illustrated
by means of two simple diagrams.

In Fig. 24 let cd represent the surface of a fluid, either water or
oil, and let ab be drawn perpendicular to this surface, and cutting
it at y. Let ry represent a ray of light proceeding from a rarer
medium, such as air, into a denser one, water or oil. As is well
known, this ray when it enters either the water or the oil does not
continue in the same direction, but is " refracted " or bent nearer
the perpendicular ab, the bending being more marked with oil
than with water. Thus we may suppose that the direction of the
ray in water would be represented by the line yw, and in oil by the
dotted line yo. Conversely, a ray of light proceeding from a
denser medium into a rarer is bent away from the perpendicular,
and the rays wy in water, and oy in oil, would, on emerging into
air, proceed in the direction yr.

In Fig. 25 (which for convenience is drawn somewhat out of

138 A MANUAL OF BACTERIOLOGY

proportion) let s represent an ordinary glass micro-slide, x a layer
of Canada balsam in which the object is mounted, and covered
with the cover-glass G, while L is the objective with its front lens.
Let the object be illuminated by the ray of light Yy ; this on enter-
ing the glass of the slide and the Canada balsam will be refracted
or bent nearer the perpendicular and will proceed in the direction
yt. Canada balsam, and also cedar oil, produce about the same
amount of " refraction," or bending of a ray of light, as crown
glass, and hence these three substances — crown glass, Canada
balsam, cedar oil — are said to have the same " refractive index,"
and, consequently, the glass of the slide, the Canada balsam, and
the cover-glass act as one homogeneous medium, and the line yt
is a straight one. In the first place, let us suppose that the
objective L is a dry one, having a layer of air between its front lens
and the cover-glass ; then the ray of light, on emerging from the
cover-glass into the air, is now bent away from the perpendicular
and pursues a direction practically parallel to its former one,
represented by the line tw, and misses the lens altogether — the
lens is unable to take it up. If, however, we suppose that our
objective is an oil-immersion one, and that a drop of cedar oil
takes the place of the layer of air between the cover-glass and the
front lens in the foregoing example, then the glass slide, Canada
balsam, cover-glass, cedar oil, and the front lens of the objective
form practically one medium ; they all have the same refractive
index and produce the same amount of refraction or bending of a
ray of light. Therefore the direction of the ray forms a straight
line in all these, and the ray passes into the objective as is repre-
sented by the broken line t — v. More important still, however, is
that which happens to rays which fall on the slide at a very oblique
angle. In the same figure (Fig. 25) let ef represent such a ray ;
on entering the slide it will be refracted, and its passage through
the slide, balsam, and cover-glass may be represented by fk. As
before, let us suppose that in the first place our objective is a dry
one, and that we have a layer of air between the cover-glass and its
front lens. In this case, if the angle which fk makes with the
perpendicular is greater than about 39° or 40°, the ray, instead of
emerging from the cover-glass into the layer of air, is totally reflected
by the cover-glass and pursues a course roughly represented by kr,
so that it never enters the objective. If, however, we employ an
oil-immersion objective, with oil instead of air between the cover-
glass and its front lens, then, as before, the slide, balsam, cover-
glass, oil, and front lens of the objective form practically one
homogeneous whole, and the ray efk, instead of being totally

DARK GROUND ILLUMINATION 139

reflected, continues its course in a straight line, and is taken up by^
the objective, as is represented by the dotted line k — v'. Hence we
see that the same rays which are unable to enter a dry objective
are admitted by an oil-immersion one, and that an oil-immersion
lens can take up rays which fall on the slide at a very oblique angle.

In order that these oblique rays may be present, ready to be
taken up by the oil-immersion objective, it is necessary to employ
a sub-stage condenser. It is only by means of a sub-stage condenser
that a " wide -angled cone of rays," as it is termed, is obtained.
Hence to make full use of an oil-immersion objective — to " get
most out of it " — it is absolutely essential to employ a sub-stage
condenser, and for the finest work a special " oil -immersion con-
denser " is employed. It will be obvious also that although a
water-immersion objective admits more rays than a dry one, it does
not admit so many as an oil-immersion. It must be pointed out,
however, that Canada balsam, or some medium having the same or
a higher refractive index, must be used for mounting to obtain the
full advantage of the oil-immersion system. The oil-immersion
can of course be used for examining objects mounted in water, etc.,
cedar oil being still used between the cover-glass and the lens. It
is to be noted that a dry objective cannot be used as an immersion
one, nor an immersion objective dry, as the construction differs in
the two cases.

Of late " dark ground illumination " has been much employed,
particularly for the examination of living objects. In this special
condensers are used, the central rays passing through which are
" stopped out," so that the object is illuminated only by very
oblique rays and appears white on a dark background. A dry lens
is used, or if an oil-immersion one, a stop must be introduced to
reduce its aperture, and slides and cover-glasses of special thickness
together with brilliant illumination are necessary.

The lenses in the objective are formed by cementing
together different kinds of glass in order to correct for
" spherical " and for " chromatic " aberration. The rays
passing through the margin and the centre of a simple
lens are not focussed at the same point, and a distorted
image is the result ; this is known as " spherical aberra-
tion," while the violet and red ends of the spectrum,
being of different refrangibility, and a simple lens acting
like a prism, coloured fringes are observed ; this is termed

140 A MANUAL OF BACTERIOLOGY

"chromatic aberration." The apochromatic system of
objectives and eyepieces has these defects very perfectly
corrected by the use of special glass and fluorite, correction
being partly effected in the objective, and this is com-
pleted by combination with the special eyepieces. The
latter, termed " compensating oculars," are therefore
essential for perfect correction with apochromatic objec-
tives, but can also be used with ordinary lenses. For
photographic purposes apochromatic lenses are far superior
to achromatic ones. Apochromatic objectives are, how-
ever, expensive, and though advantageous are not really
necessary for ordinary bacteriological work.

In consequence of certain optical principles, the
" diffraction " theory, for details of which the reader must
refer elsewhere,1 it is useless to increase the magnifying
power of objectives beyond a certain point ; for, although
the object viewed appears larger, no more details of structure
can be made out.

The use of the immersion system increases the " re-
solving power," or the amount of detail which can be
seen. Thus, if a number of fine equidistant parallel lines
be ruled on a glass plate, it is impossible to see with a dry
lens, using white light, more than about 90,000 lines to
the inch as isolated lines. If more are ruled they will
not appear, and practically nothing is visible. With a
water-immersion objective it is possible to see about
120,000 lines to the inch, and with an oil-immersion as
many as 146,000 lines to the inch, as separate lines — a
clear gain in resolving power in the latter case of about
one half over a dry lens.2 As it is necessary, in order to
see such fine structures as lines ruled 50,000 or more to
the inch must be, to have considerable amplification in

1 See Carpenter on the Microscope, edited by Dallinger. (Churchill.)

2 These figures refer to lenses having a numerical aperture of 1-0
(dry), 1-33 (water), and 1-4 (oil).

ULTRA-MICROSCOPIC ORGANISMS 141

addition to resolving power, not much is gained, in
ordinary work at any rate, by adopting the immersion
system for the lower power objectives, such as the g-in.

By the physical theory of microscopical visibility, it can be
shown that objects having a diameter of less than about 0-16 p
cannot be seen with the best optical appliances. If, then, a micro-
organism is less in size than this it could not be seen microscopically,
and this fact may explain why it is that in certain undoubted
infective diseases no micro-organism has yet been isolated. Of the
existence of such " ultra-microscopic " organisms we have proof.
The finest porcelain filters, such as the Chamberland B, do not allow
visible particles to pass through, yet in several instances, if the
infective material be filtered through such a filter, the filtrate is
still infective. This is the case with the blood-serum in yellow
fever, Cape horse sickness, dog distemper, hog cholera, and swine
fever, in bird and cattle plagues, and with the juice of bird mollus-
cum. The organism of cattle pleuro-pneumonia is just on the limit
of visibility. The rabic and vaccine viruses also seem capable of
passing through a Berkfeld V. These experiments do not neces-
sarily prove that the organism in all stages is invisible. l Siedentopf
and Zsigmondy have devised a method whereby ultra-microscopical
particles may be rendered visible, but inasmuch as they appear
merely as luminous points, it is questionable whether the method
will be of great service in bacteriology. Some thirty ultra -
microscopic viruses are now known, including, in addition to those
mentioned above, those of anterior poliomyelitis, measles, mol-
luscum, and trachoma.

There is no real necessity in bacteriological work for the
immersion objective to be provided with a " correction
collar." The " correction collar " is an additional screw
in the objective by means of which the distance between
some of its constituent lenses can be altered to " correct "
for varying thicknesses of cover- glass, etc., and though
necessary with the higher power dry lenses, it is theo-
retically unnecessary with the immersion system. Never-
theless, as slight variations do occur in the various media,

1 See Roux, Bull, de VInst. Past., vol. i, 1903, pp. 1 and 49. Rem-
linger, ibid. vol. iv, 1906, pp. 337 and 385; Trans. XVIIth Internal.
Cong. Med. 1913, Sect. IV, Pt. I, pp. 35 (Loffler) and 49 (McFadyean).

142 A MANUAL OF BACTERIOLOGY

glass, oil, etc., and they may not form a truly homogeneous
whole, for the finest work the correction collar is still
desirable. So much for the high-power objectives. As
regards the lower powers, which, of course, are dry, a
f-in. and a |-in. are generally selected. The f-in. is a
more serviceable lens than the 1-in. which is often recom-
mended. A very useful accessory is a " double " or
" triple nosepiece." This consists of a light metal frame-
work, which is attached to the lower end of the tube of
the microscope, on to which two or three objectives can
be screwed. The framework can be rotated, thus bringing
each objective in succession into the optical axis of the
instrument, and the necessity for unscrewing and screwing
on each time an objective is changed is obviated. A
microscope such as described, with sub-stage condenser,
two eyepieces, a f-in. and a J-in. dry and a ^--in. oil-
immersion objectives, triple nosepiece, etc., complete in
case, can be obtained for about £15, and it is well to add
another sovereign or two for superior finish. Both British
and Continental firms supply microscopes arranged as
indicated, and in this department the English makers
hold their own.

The measurement of micro-organisms is carried out by
means of a stage micrometer, alone, or in combination
with an eyepiece micrometer. The former consists of
a scale of tenths and hundredths of a millimetre or
hundredths and thousandths of an inch ruled in fine lines
on a glass plate, by means of which the measurements
can be made by focussing the scale under the microscope.
The stage micrometer is placed in position on the stage
and the scale is focussed with the particular ocular,
objective, and tube length which are to be used. A
drawing of the scale is made with a camera lucida ; the
micrometer is then removed and the object placed in
position and a second drawing is made of the object on the

MEASUREMENT OF MICRO-ORGANISMS 143

scale already drawn. A simpler and less expensive arrange-
ment is to make use of a disc of glass ruled with equi-
distant fine lines, which can be placed in the eyepiece by
unscrewing the top lens and dropping it on the diaphragm
below. The value of the divisions in the eyepiece scale
is first ascertained by means of the stage micrometer.
The stage micrometer is then removed and the object to
be measured put in its place, and its dimensions are
determined by means of the eyepiece scale. With the
eyepiece micrometer, the value of the divisions is first
ascertained by means of the stage micrometer, which is
then replaced by the object. If the objective or the eye-
piece be changed the value of the divisions of the eyepiece
scale in both cases will be altered, and must again be
determined by means of the stage micrometer. The unit
for microscopical measurement is the micron (sometimes
erroneously termed a micro-millimetre), which measures
one thousandth of a millimetre, or approximately 0^375-
of an inch, and is designated by the sign /m.

If a micrometer is not available, rough measurements
may be carried out by comparison with a red blood-
corpuscle. The majority of the red corpuscles of normal
human blood measure 7-5 JUL in diameter.

CHAPTER V

INFECTION— VEGETABLE AND ANIMAL PARASITES—
THE INFECTIVE PROCESS — ANTI-BODIES — ANTI-
SERA AND ANTITOXINS— IMMUNITY

Infection

BY the term INFECTION is meant the invasion of the living
tissues by living micro-organisms which grow and multiply
at the expense of the host. A disease produced by the
growth and multiplication of micro-organisms is termed
an infective disease, and is transmissible in most instances
by inoculation. If the micro-organisms are from time to
time discharged from the body of the host, either with
the excreta, secretions, desquamated particles, or in some
other way, the disease becomes infectious or contagious,
according to the ease with which another individual
becomes infected, and the material which conveys the
infection is often termed the contagion. Thus, in scarlatina
and smallpox the contagion is very readily conveyed from
person to person even for a distance through the air, and
these are infectious diseases. Ringworm and syphilis,
as a rule, require more or less close contact for infection
to take place, and these are, therefore, contagious diseases ;
while malaria is neither infectious nor contagious, since
persons in the neighbourhood never directly contract the
disease, though it can be conveyed by inoculation, and
it is therefore infective only. But the distinction between
infectious and contagious is mainly one of degree, and these

144

INFECTION 145

terms have now to a laTge extent been discarded. Ex-
cluding individual susceptibility, the relative infectivity
of a disease probably depends on three factors : (1) the
contagion is freely given off aerially and is not destroyed
thereby ; (2) the contagion gains access by the respira-
tory tract ; and (3) the relative virulence of the contagion ;
in some instances the smallest amount of the contagion
is sufficient to infect. If the contagion can gain access
only through a wound or the digestive tract, the chances
of infection may be largely reduced. In certain instances
infection is conveyed by an intermediary, e.g. the mosquito
in malaria, and in such cases infectivity will obviously
depend on the presence and abundance of the intermediary.
Infection is manifestly a part of the whole subject of
parasitism, which includes the animal and vegetable
parasites which develop in the animal body. If, however,
the subject of parasitism is considered more closely, it
will be seen that there is a vast difference between, say,
a condition caused by the echinococcus or by the round
worm, in which the effects are largely mechanical and in
which relatively little poison is produced by the parasite,
and the disease diphtheria caused by the diphtheria
bacillus, in which the diphtheria bacilli have little or no
action mechanically, but elaborate virulent chemical
poisons which cause a general intoxication. Some parasites
also may produce a general infection, e.g. anthrax, others
only a local infection, e.g. ringworm.

Parasites may therefore be divided into infective and
non-infective, though there is a series of connecting links
between these, and the two groups cannot be sharply
separated. The infective parasites are : (1) vegetable
micro-organisms, chiefly bacteria, a few yeasts and some
moulds ; (2) many protozoa ; and (3) a few metazoa,
generally worms. The non-infective parasites are the
animal parasites generally, particularly many worms.

10

146 A MANUAL OF BACTERIOLOGY

The production of the phenomena of disease by patho-
genic organisms has been ascribed to (1) the using up of
the oxygen which should go to the tissues ; (2) the using
up of the proteins of the body and of the food ; (3) the
effects of plugging of the vessels by the microbes ; and
(4) the effects of substances or " toxins," having a poisonous
action, formed by the microbes. Of these, the first three
are quite subsidiary, embolism and thrombosis being
perhaps the most important, and the toxins are the chief
factors which induce the pathogenic effects. These toxins
are substances of a very complex composition, probably
allied to the proteins ; in some instances they seem to be
of the nature of enzymes or ferments, and they are direct
products of the bacterial cells. The toxins of most
pathogenic organisms, e.g. typhoid, cholera, plague, etc.,
are more or less integral parts of the bacterial cells ; they
are " endotoxins," and are not excreted to any extent
into the surrounding medium, but may gain access to it by
autolysis of some of the organisms. A few organisms,
notably Bacillus diphtherice and Bacillus tetani, produce
extra- cellular toxins which are found in the culture liquid.
The toxins are classified by Sidney Martin,1 as follows
(see also p. 39) :

(1) Poisons produced by the digestive or the destructive
action of bacteria on proteins in the culture medium.
Examples of these are the poisons of the Bacillus anthracis
and of the pus-producing staphylococci.

(2) Poisons which are the result of the digestive or
destructive action of bacteria on proteins, but formed as
an excretion (the toxin) of the bacterium. The Bacillus
diphtherice is the best example of this. A similar com-
bination of poisons is found in snake- venom.

(3) Poisons which are excretions only, such as those
produced by the tetanus bacillus.

1 Manual of General Pathology, p. 76.

THE INFECTIVE PROCESS 147

(4) Poisons which are typically intra-cellular, but which
may also be excretory. The poisons produced by the
typhoid bacillus, the Bacillus coli, the Bacillus enteritidis
of Gaertner, and the cholera vibrio belong to this group.

Thiele and Embleton1 suggest that the toxins of bacteria
are really cleavage products derived from their cellular
proteins under the influence of ferments present in the
body of the host. These cleavage products are, however,
toxic only at a certain stage of their disintegration. Given
the power of existing and multiplying in the body of the
host, the pathogenicity of a bacterium depends on the
quantity and consequent activity of the ferments of the
host. A certain degree of ferment activity renders the
cleavage products of the bacterio-protein toxic, a further
degree of ferment activity carries the disintegration so far
that the cleavage products are no longer toxic. A
bacterium may therefore be harmless to a host if the
latter (a) has no ferments capable of digesting its bacterio-
protein ; (b) has such a poor supply of ferments that the
bacterio-protein is so slowly disintegrated that toxic
products never attain a sufficient concentration to be
harmful ; (c) has such a plentiful supply of ferments that
the cleavage of the bacterio-protein rapidly passes beyond
the toxic stage. A harmless bacterium, e.g. B. megaterium>
may be rendered pathogenic if suitable ferments can be
produced in the host to bring about the necessary dis-
integration of its bacterio-protein.

The Infective Process

With regard to the pathogenic micro-organisms, or

disease germs, Koch laid down the following conditions,

which have been termed " Koch's postulates," which

must be complied with before the relation of an organism

1 Lancet, vol. i, 1913, pp. 234 and 332.

148 A MANUAL OF BACTERIOLOGY

to a disease process can be said completely to be demon-
strated :

(1) The organism in question must be present in the
tissues, fluids, or organs of the animal affected with, or
dead from, the disease.

(2) The organism must be isolated and cultivated out-
side the body on suitable media for successive generations.

(3) The isolated and cultivated organism, on inoculation
into a suitable animal, should reproduce the disease.

(4) In the inoculated animal the same organism must
be found.

To these may be added :

(5) Chemical products with a similar physiological
action may be obtainable from the artificial cultures of the
micro-organism, and from the tissues of man or animals
dead of the disease.

(6) Specific serum and other reactions, agglutinative,
bacteriolytic, complement fixative, etc., are generally
obtainable, under certain conditions, if the blood of the
infected person or animal be allowed to act on the specific
organism producing the infection.

It is true that one or more of these conditions may not
be fulfilled in all cases, but on general evidence the disease
is classed as infective.

The modes of infection, or entrance of the infective
agent into the body, are varied. The infective agent
may enter by (1) the gastro-intestinal tract, e.g. typhoid,
cholera, and glanders ; (2) the respiratory tract, e.g.
pneumonia and influenza, and occasionally typhoid,
plague, etc. ; (3) by inoculation, not necessarily only of
the skin, but also of the mucous membranes, e.g. the
septic diseases, glanders, tetanus, etc. The extreme infec-
tivity of some diseases — e.g. variola, scarlatina, influenza,
etc. — may be due to the fact that infection takes place
by the respiratory tract. In certain instances the

ANTI-BODIES 149

infection is conveyed in some special way, e.g. by
mosquitoes in malaria and in yellow fever. Nor is
infection necessarily confined to one mode of entrance ;
in plague, for example, infection by the skin is com-
monest in some epidemics, but it is not infrequent by
the respiratory, and may occur by the digestive, tract.
The infecting agent may remain localised, giving rise to
a local infection, or it may be widespread through the
body, a septiccemia1 or general infection. The absorption
of chemical products from a local site of infection may
produce general symptoms ; this is intoxication, as occurs
in cholera, in which the microbe is limited to the bowel,
in the early stage of diphtheria, in which the diphtheria
bacillus is limited to the membrane, and in a local abscess.
Fever is usually one of the results both of intoxication
and of general infection.

Infection, if recovery ensues, is usually followed by
remarkable alterations in the blood and tissues. One
of these is the production of immunity or insusceptibility
to the same infecting agent ; this will be considered later
(p. 195). Agglutinins, substances which cause clumping
of the infecting organism, are also generally produced
(p. 185).

Anti-Bodies 2

Another remarkable property, and one of considerable
importance in immunity, conferred by the injection into
an animal of complex substances, such as bacterial toxins,
bacteria, blood- corpuscles, cells and cellular proteins,
ferments, etc., is the development of anti-bodies. Thus

1 " Septicaemia " and " a septicaemia " have different meanings. The
former is applied to a general infection with the so-called septic
organisms, the latter to a general infection with any organism.

2 All the subjects dealt with in the subsequent portion of this chapter
are discussed in detail by Emery, Immunity and Specific Therapy,
1909.

150 A MANUAL OF BACTERIOLOGY

an animal injected with sub-lethal doses of a bacterial
toxin, e.g. diphtheria toxin, acquires a tolerance towards
the toxin, becomes immunised, and a substance is de-
veloped in the blood that antagonises the toxin which
was injected ; this substance is known as antitoxin. If
bacteria be injected, the fresh blood in vitro has a solvent
action on the bacteria (bacteriolysis) ; if blood- corpuscles
be injected, the fresh blood has a solvent action on the
same kind of blood-corpuscles (haemolysis) ; if cells be
injected, the blood has a solvent action on the cells
(cytolysis), and so on. If ferments be injected, anti-
ferments are formed and will prevent the specific action
of the ferment. With doubtful exceptions,1 it is only
complex bodies of protein nature, or allied to the proteins,
which give rise to the production of anti-bodies on inocula-
tion ; alkaloids, carbohydrates, mineral poisons, etc., do
not give rise to anti-bodies, though some insusceptibility
to them may be produced (see also p. 206). Any substance
which gives rise to an anti-body may be termed an anti-
gen. These anti-bodies, etc., may first be considered,
after which immunity will be discussed.

Anti-bodies are probably formed for the most part in the
spleen, lymph-glands and bone-marrow by leucocytes, or
by endothelial cells, or by both.

ANTITOXINS. — The anti-bodies produced by the inocu-
lation of an animal with bacterial toxins or toxic proteins
(e.g. ricin, abrin, and snake-venom) are known as anti-
toxins, and are of considerable practical importance. An
animal injected with increasing amounts of the toxin
acquires a high degree of immunity, and its blood-serum
injected into a second animal confers on the latter a
similar immunity against the same toxin, but not against
other toxins ; the serum is specific. The anti-serum

1 Ford has described the formation of an anti-body by the injection
of a poisonous glucoside derived from fungi.

ANTITOXINS 151

formed by the injection of toxin is antitoxic and not
anti-microbic, and the diphtheria bacillus will grow and
multiply in diphtheria antitoxin. Since, however, the
pathogenic effects of an organism such as the diphtheria
or the tetanus bacillus are caused by the toxin which it
forms, the antitoxin will counteract the effects of the
micro-organism as well as of its toxin. The neutralisa-
tion of the micro-organism, however, may not be quite
complete, a certain amount of local reaction or necrosis
ensuing.

Antitoxins are prepared by injecting animals — prefer-
ably horses, but goats, rabbits, etc., may also be employed
• — with bacterial toxins or with cultures.

With those organisms which produce potent toxins
such as diphtheria and tetanus, it is customary to grow
the organism in a fluid medium so that an active and
virulent toxin is obtained. The culture is then filtered
through a Berkefeld or Pasteur-Chamberland filter and the
toxic filtrate inoculated subcutaneously into an animal,
generally a horse, commencing with sub- lethal doses.

The dose of toxin can be gradually increased, and con-
currently with the increase in insusceptibility the blood-
serum acquires antitoxic properties. The treatment is
tedious, and the activity of the antitoxic serum is largely
dependent upon the amount and activity of the toxin
injected. The requisite degree of strength having been
attained, the horse is bled with aseptic precautions, the
blood is allowed to coagulate, and the serum is bottled
for use. Antitoxin may be obtained in a concentrated
form by " salting out " the globulin constituents of an
antitoxic serum (p. 167), and a dried product may be
prepared by evaporating the serum to dryness in vacuo
at 40° C. ( 10 c.c. serum = 1 grm. dry residue).

The mode of production of the antitoxin by the injection
of the toxin has been the subject of various theories. By

152 A MANUAL OF BACTERIOLOGY

some it has been supposed that the antitoxin is modified
toxin, the modification being brought about by the vital
activities of the cells. But the amount of antitoxin
produced does not necessarily bear any relation to the
quantity of toxin injected. Woodhead records instances
in which the amount of antitoxin formed amounted to
40,000 times the equivalent amount of toxin injected,
bleeding the animal only temporarily reduces the anti-
toxin content of the serum, and substances which increase
the secretive properties of glandular cells, such as pilo-
carpine, enormously increase the output, so to speak, of
antitoxin.

In view of these facts Ehrlich elaborated his " side-
chain theory," a theory which, whether it be the real
explanation or no, has received a considerable amount of
experimental support, and has had far-reaching effects
in stimulating research. Ehrlich believes that the chemical
activities which are the manifestations of the vital
activities of the living cell are due to a very large nucleus
or chemical molecule having a ring structure, analogous
to the benzene ring, and having attached to it a number
of atomic groups or " side- chains." A " side- chain " is
an atomic group, a carbon atom of which is linked to one
of the carbon atoms in a ring. These atomic groups or
side-chains are unstable in nature, and enter freely into
combination with other suitable groups should these be
presented to them, and thus the physiological activities
of the cell, assimilation, nutrition, etc., are carried out
(Fig. 26). Now Ehrlich supposes that antitoxin is merely
an excess of certain side-chains which are normally present
and subserve some of the ordinary functions of the cell
and which have become free in the blood. The antitoxins
being specific, by this assumption the difficulty is obviated
of supposing that special chemical groups or molecules
exist preformed ready to combine with a number of

SIDE-CHAIN THEORY

153

different toxins on the remote chance that some one of these
may at some time or other come within the particular
sphere of action of one of those groups. Moreover, small
amounts of anti-bodies, such as antitoxin, bacteriolysin,
agglutinin, etc., are met with in normal untreated animals
and in man. While some have supposed that the small
amount of diphtheria antitoxin (equivalent to half a unit

FIG. 26. — Diagram to represent
the cell with its various com-
bining groups or side-chains.
(After Ehrlich.)

FIG. 27. — First stage in anti-
toxin formation. (Black =
toxin molecule. (After Ehr-
lich.)

or so) present in human blood-serum is due to an infection
with the diphtheria bacillus (not necessarily an attack of
diphtheria), it seems more rational to suppose that this
antitoxin is due to a natural liberation of such side- chains
from the protoplasm and that artificial antitoxin pro-
duction is merely a very great stimulation of this natural
process.

The toxin molecule, according to Ehrlich, possesses at
least two fixative atomic groups or side-chains. One
of these, the " haptophore group," conditions the union
of the toxin molecule with cell-protoplasm ; the other,
the " toxophore group," conditions its toxic action.
Similarly, in order that the cell may suffer the full effect
of the action of the toxin, it also must possess two receptive

154 A MANUAL OF BACTERIOLOGY

groups or side-chains having a maximum affinity for the
haptophore and toxophore groups of the toxin ; these
may be termed the " receptor " and " toxophile " groups
respectively (see Fig. 31). The relationship of each
fixative group of the corresponding groups — viz. that of
the toxin and that of the side-chain of the cell— must be
most intimate, and analogous to the relations to each

FIG. 28. — Second stage in anti- FIG. 29. — Third stage in anti-
toxin formation. (After Ehr- toxin formation. Side-chains
lich.) beginning to be produced in

excess. (After Ehrlich.)

other of a male and a female screw (Pasteur) or of a lock
and its key (E. Fisher).

The genesis of antitoxin on the " side- chain theory "
takes place in the following manner : Toxin being intro-
duced, the haptophore groups of the toxin molecules unite
with the particular receptor side-chains of the proto-
plasm for which they have an affinity (Fig. 27). By this
combination the physiological activities of the cell are
interfered with, a defect is created, the cell is damaged
(it is only necessary to consider the case of one cell, or,
more strictly of one molecular group of the cell-protoplasm).
But through its recuperative powers the cell soon recovers
by the formation of new receptor side- chains to take the
place of those which have been put out of action. On
injecting more toxin, this combines with these new receptors

RECEPTORS

155

arid a defect is again created (Fig. 28). Once more the
cell responds, and a fresh series of receptors is developed
(Fig. 29). But by this continual stimulation, as it were,
the cell commences to form the particular receptors in
excess of that needed to repair the defect created, and ultimately
these receptors are reproduced in such numbers that
they no longer all remain attached to the cell but some be-
come free in the plasma (Fig. 30).
These receptor sidechains, detached
from the cell and floating free in
the blood-stream, constitute the anti-
toxin. This excessive production
of side-chains after stimulation by
repeated injections of toxin is not
a phenomenon confined to anti-
toxin formation, but is a general
physiological law enunciated by
Weigert ; as a result of repeated
stimulation, over-production or
hyper-compensation is the rule
and is met with in various

pathological processes. Ehrlich has termed the diverse
free receptors which occur in the body fluids in various
circumstances " haptines."

The existence of both haptophore and toxophore groups
in the toxin molecule is suggested by the following experi-
ments. Tetanus toxin injected into the blood-stream of
an animal rapidly disappears, within a few seconds of
the injection, and even if the animal be at once bled,
the blood withdrawn being replaced by fresh blood,
tetanus ensues, but not until after the lapse of an in-
cubation period of some hours. The tetanus toxin,
therefore, immediately becomes fixed or anchored to the
tissues of the central nervous system. Evidently the
toxin molecules enter at once into combination with the

FIG. 30. — Fourth stage in
antitoxin formation.
Side-chain, i.e. antitoxin,
free in the blood. (After
Ehrlich.)

156 A MANUAL OF BACTERIOLOGY

nerve-tissues by means of their haptophore groups ; this
after a time brings the cells within the sphere of influence
of the toxophore groups, and after a certain incubation
period toxic symptoms ensue. The affinity of tetanus
toxin for nerve tissues may be shown in another way. If
fresh guinea-pig brain be emulsified with tetanus toxin,
the emulsion will be found to be innocuous on injection,

FIG. 31. — Diagrammatic scheme to represent the union of toxin
(black) with the cell. In A the toxin is attached to the pro-
toplasm by the union of the haptophore and receptor groups.
In B the toxophore and toxophile groups have also united,
and poisoning now ensues.

owing to a combination between the two having taken
place. The cerebral cortex of a highly susceptible animal
(e.g. mouse) has a marked neutralising power, of a less
susceptible animal (e.g. rabbit, fowl) a feebler, and of an
insusceptible animal (e.g. frog, tortoise) no neutralising
power.1 Moreover, both diphtheria and tetanus toxins
may be converted into non- toxic modifications (" toxoids ")
which to some extent retain the power of immunising
and of producing antitoxin on inoculation, and of com-
bining with antitoxin : that is to say, according to Ehrlich,

1 The combination of brain matter with tetanus toxin seems to be
specific and of the same order as that between antitoxin and toxin.
See Noon, Journ. of Hyg., vol. vii, 1907, p. 101, and Besredka and
Bordet, Ann. de Vlnst. Past., xvii, 1903.

NEUTRALISATION OF TOXIN 157

the toxophore groups have been destroyed while the
haptophore groups remain unaffected. It is the presence
of the haptophore group which conditions the union of
toxin with antitoxin. Thus, if toxin be injected into
blood containing antitoxin, the haptophore groups of
the toxin unite with the free receptor groups, i.e. with
the antitoxin (Fig. 32), and therefore the toxophore groups
cannot exert their influence
because the toxin is now
unable to unite with the pro-
toplasm, its haptophore or
binding groups being already
occupied.

In a poisonous toxin, such
as diphtheria or tetanus toxin,
the toxophore group is more
readily destroyed than the FIG. 32.— Neutralisation of toxin
haptophore group, and by ^~h,in the "^
heating a toxin for some

time to 60°-70° C. its toxicity is destroyed, but it still
retains an affinity for antitoxin. If some antitoxin be
mixed with such heated toxin it will be found that the
capacity of the former for neutralising active toxin is
much diminished — in other words, although the toxophore
groups of the heated toxin have been destroyed, the
binding or haptophore groups still remain. Toxin which
has been kept for some time decreases in toxicity, but
retains the power of combining with antitoxin, again
showing that haptophore or binding groups are present
(such derivatives of toxin possessing haptophore groups
are termed " toxoids "). Wassermann and Bruck have
obtained presumptive evidence of the existence of the
second stage in antitoxin formation, viz. the increased
production of receptors by the cells. Using tetanus toxin
which had been kept for some time and had lost its

158 A MANUAL OF BACTERIOLOGY

toxicity, but which still combined with antitoxin — that
is, toxoids with haptophore groups still present — they
found that on injecting it into animals no antitoxin
was formed as a result of the injection. They then per-
formed some experiments based on the following line of
reasoning : If the old non-poisonous tetanus toxin con-
taining these toxoids be first injected into an animal, and
after a short interval, some fresh, actively poisonous
tetanus toxin, more of the active toxin ought to be required
to kill this animal than a normal one, because, owing to
the previous toxoid injection, part of the cell receptors
susceptible to tetanus toxin are already occupied. Pro-
vided Ehrlich's theory be correct, so that this binding
of the toxoid really occurs, the conditions should be entirely
different, when, instead of injecting the toxin shortly
after the toxoid, a longer time elapsed — one to three
days — before the injection of the active tetanus toxin.
For in that case Weigert's law should come into play
and the receptors should have increased in number — i.e.
the organism would now possess more sensitive groups
than before. This should be manifest by the fact that, in
contrast to the first experiment, the fatal dose of active
tetanus toxin ought now to be smaller than previously ;
in other words, a smaller dose should now tetanise the
animal in a shorter time. The experiments yielded results
which were exactly in accordance with these theoretical
considerations. A guinea-pig was injected with some of
the non-poisonous toxoid, and then, one hour later^ with
the active tetanus toxin. It was found that much more
toxin was required to kill this animal than a normal guinea-
pig of equal size. If, on the contrary, an interval of one
to three days were allowed to elapse, it was then found
that a dose of tetanus toxin which would not even tetanise
a normal guinea-pig was sufficient to kill this one.
The fact that no antitoxin is formed — i.e. no receptors

ABSORPTION OF TOXIN 159

are thrust off — by the single injection of the non-
poisonous toxin, or toxoid, Wassermann ascribes to the
lack of stimulus which he suggests resides in the toxo-
phore groups.

The slow combination of the haptophore and receptor
groups has been proved by Wassermann in another way.
The researches of Meyer and Ransom have shown that
tetanus toxin is absorbed by the nerve- trunks, not by
the blood and lymph-channels, while tetanus antitoxin
is absorbed by the latter — the blood and lymph- channels.
Adrenalin is a substance which strongly contracts the
capillaries, and thus tends to block absorption in a parti-
cular area. The following experiment was devised :
Tetanus toxin and antitoxin were mixed in such propor-
tions that the mixture was innocuous to animals, i.e. it
was just neutral. If this mixture be injected into the
hind paw of a guinea-pig no tetanus develops. When,
however, some adrenalin is injected into the hind paw of
a similar-sized guinea-pig, and a few minutes are allowed
to elapse so that the capillaries may contract, and then the
mixture of toxin and antitoxin is injected, typical tetanus
ensues. The explanation of this is that the channel of
absorption for the tetanus antitoxin, the vessels, is blocked
by the adrenalin, while that for the toxin, the nerve path,
remains open. The toxin and antitoxin had not yet
combined, or such combination as had occurred is a loose
one and becomes dissociated, and, therefore, the toxin
travelled along the nerves to the central nervous system
with the production of tetanus.

The experiment, however, succeeds only within a certain
period, not exceeding an hour after mixture of the toxin
and antitoxin, because after this the toxin-antitoxin
combination becomes a stable one.

If a longer time — say three or four hours — is allowed
to elapse, it will be found that, even in the adrenalin

160 A MANUAL OF BACTERIOLOGY

animal, no tetanus is produced, because by this time the
combination, previously a loose one, has become so stable
that the substances can no longer be dissociated. This
union can be hastened by employing more tetanus anti-
toxin, for with an excess of antitoxin, even after only half
an hour, it is impossible by means of adrenalin to free
the tetanus toxin. This experiment, therefore, shows that
the combination of tetanus toxin with antitoxin takes
place slowly and is at first a loose one, and that the union
becomes firmer and firmer with lapse of time. It also
suggests the possibility of hastening the combination by
increasing the amount of antitoxin — a point of consider-
able practical value in serum therapy.

The above considerations are of importance in the antitoxin
treatment of disease. Antitoxin, in the strict sense, is not anti-
microbic, and therefore antiseptic treatment of the throat in
diphtheria, and of the wound in tetanus, should be pursued. The
fact that the toxophore group of the toxin does not come into
action as a rule for many hours at least (an exception is snake -
venom) is a fortunate coincidence, for the antitoxin may, there-
fore, act before tissue damage has occurred. Antitoxin cannot
repair tissue damage if this has been produced by the toxin, but
it can, and does, prevent the occurrence of further damage by
neutralising any fresh amounts of toxin that may be absorbed.
Hence the necessity for early treatment. Toxin already anchored
to the tissues by its haptophore group may for some time be dis-
sociated from them if a multiple of the simple neutralising dose of
antitoxin be injected, and the quantity necessary to accomplish
this rises rapidly as the interval between the introduction of the
toxin and of the antitoxin increases ; hence the necessity for the
use of antitoxin in large excess. Probably the union between tissue
and toxin at first is a loose one, and a large amount of antitoxin
by mass action transfers the affinity of the toxin from the tissues
to itself. It must be clearly recognised that colloidal reactions
(to which category that between antitoxin and toxin, anti-body
and antigen, belongs) differ considerably from ordinary chemical
reactions.

An essential condition in antitoxic treatment is the administration
of a sufficient amount of anti-serum, and this does not depend on

NEUTRALISATION OF TOXIN 161

the actual volume of serum injected. The anti-serum may be
regarded as a solution containing a variable amount of the anti-
toxic or anti-microbic constituent, and for therapeutic use its
strength must be ascertained, and is for convenience described in
arbitrary unite.

The dose of antitoxin is dependent upon the gravity of the
disease, and not on the age of the patient, for evidently just as
much toxin may be formed in a child as in an adult. The anti-
toxins are strictly specific ; diphtheria antitoxin, for example, has
not the slightest influence in tetanus.

To obtain an immediate reaction to antitoxin it should be
administered intra-venously. A subcutaneous injection may not
be completely absorbed in less than thirty-six hours, an intra-
muscular injection is much more rapidly absorbed.

In cases of mixed infection, e.g. where diphtheria bacilli are
associated with streptococci or staphylococci, the diphtheria anti-
toxin will have no influence on the streptococcic or staphylococcic
infection.

The complications and accidents of antitoxin treatment are few
and usually unimportant. Abscess and other local troubles at the
seat of inoculation should not occur if proper antiseptic precautions
be taken. Urticaria or other rashes and joint pains are by far the
most troublesome complications. These are due to the injection
of foreign serum, and not to the antitoxin, for the serum of an
untreated horse produces a like effect. Repeated injections of
serum at short intervals may be continued for a long period without
inducing more disturbance than that caused by one or two or a few
injections, but if twelve days or more elapse between two injections
a condition of " supersensitation," due to anaphylaxis, ensues
(see p. 168). This consists in the rapid appearance of rashes, joint
pains, pyrexia, etc., or even of grave symptoms, faintness, vomiting,
dyspnoea, convulsions, collapse, etc.

Anti-sera may be used as prophylactics, but the immunity
produced by them does not last more than three weeks.

Various hypotheses have been advanced to explain
the manner in which toxin is neutralised by antitoxin.
Roux and Buchner suggested that the antitoxin in
some way renders the cells and tissues insusceptible
to the toxin, and Buchner performed experiments show-
ing that while mice are more susceptible than guinea-
pigs to tetanus toxin, a tetanus toxin-antitoxin mixture

IT

162 A MANUAL OF BACTERIOLOGY

which is just neutral for mice is distinctly toxic for guinea-
pigs.

To explain this Ehrlich suggested that there may be
present in a toxin solution several toxic substances, some
of which exert a toxic action on the guinea-pig but not
on the mouse. Madsen and Dreyer showed that a mixture
of diphtheria toxin and antitoxin which is innocuous to
guinea-pigs on subcutaneous inoculation is lethal to rabbits
on intra- venous injection, and in order to explain this
Ehrlich made a similar assumption. Morgenroth, how-
ever, found that the difference in the latter case depends
on the mode of injection. The reaction between the
toxin and antitoxin takes time to complete : there is an
interval probably of some hours at 20° C. before equilibrium
is reached (see also p. 163). When a recently prepared
mixture of toxin and antitoxin is injected subcutaneously,
absorption is slow, and in the meanwhile the toxin and
antitoxin combine, but when the mixture is injected into
the veins, the toxin is fixed by the tissues before it has
had time to combine with the antitoxin, and poisoning
ensues. If the mixture be kept for some hours before
injection, intravenous injection is then innocuous.

Ehrlich concluded that diphtheria toxin is neutralised
by diphtheria antitoxin much in the same way as a strong
base is neutralised by a strong acid, and that the course of
neutralisation suggests the presence in the toxin of several
toxic and atoxic substances (toxoids and toxones), all
of which combine with, though they have different affinities
for, the antitoxin.

Arrhenius and Madsen, however, believe that the toxin-
antitoxin reaction is analogous to the action of an acid
on an alcohol, and that the chemical laws of mass action
apply equally to the two. The chief reaction is considered
to be between two substances only, toxin and antitoxin,
that it is reversible, and that when the system has reached

THE TOXONE EFFECT 163

equilibrium, a fraction of toxin and also of antitoxin
remains free, this fraction of toxin producing the " toxone
effect " (see p. 165). If equivalent quantities of acetic
acid and alcohol are mixed, the reaction is never complete ;
the acid and alcohol never entirely disappear, because the
water formed reacts with the ethyl acetate, re-converting
it into acid and alcohol. Such a reaction is termed rever-
sible, and this particular case could be thus represented :

CH.COOH + CH.OHCH.COOCH H0.

Bordet has suggested that the fixation of toxin by
antitoxin is an adsorption phenomenon, similar to the
fixation of a dye by a tissue.

These hypotheses may now be examined more in detail.
Ehrlich's experiments l on diphtheria toxin seemed to
show that the neutralisation of toxin by antitoxin follows
the laws of simple chemical combinations, such as the
neutralisation of a strong base (NaOH) by a strong acid
(HC1). If so, it would be expected that antitoxin would
neutralise proportionate amounts of toxin ; but this is
not so, and Ehrlich was forced to the conclusion that toxin
is a complex mixture of proto-, deutero-, and trito- toxin,
and toxone, with different toxicities and different avidities
for antitoxin. Moreover, when toxin is kept it decreases
in toxicity, though still retaining much of its avidity for
antitoxin. Ehrlich assumed, therefore, that the toxin
becomes transformed into substances termed toxoids,
which are non-toxic but retain their affinity for antitoxin
(see also section on the standardisation of diphtheria anti-
toxin). This he explained as due to destruction of the
unstable toxophore groups, with the retention of the
more stable haptophore groups. That the neutralisation
of toxin by antitoxin is due to some sort of union between

1 l-'w Trans. Jcnncr Inst. Prcv. Mcd., vol. ii, p. 1 ; Croonian Led.,
Roy. tioc. Lond., 1900 ; and p. 293.

164 A MANUAL OF BACTERIOLOGY

the two, though not necessarily chemical combination in
the strict sense, seems to be proved by the work of Martin
and Cherry. Brodie,1 and Martin and Cherry,2 making
use of a Chamberland filter, the pores of which had been
rendered very fine by saturating with gelatin, found that
toxin would pass through such a filter but that antitoxin
would not, presumably because the molecule of the latter
is larger. By mixing diphtheria toxin and antitoxin in
such proportion that the latter was in sufficient quantity
to neutralise the toxin, and subjecting the mixture to
filtration through a gelatin filter, the filtrate was found
to be non-toxic. Now since toxin can pass through such
a filter, the inference is that the toxin has united with
the antitoxin. Using snake- venom and its anti-serum or
anti-venin, another method was employed. The anti-
venin is destroyed by heating to 68° C. for ten minutes,
while the toxic properties of the venom are unaltered by
this treatment. By making mixtures of venom and anti-
venin, and, after a certain time has elapsed for the inter-
action to take place, heating to 68° C. for ten minutes,
it was found that the mixture is non-toxic, pointing to the
union of the toxin (venom) with the antitoxin (anti-venin).
Calmette had performed the same experiment but with
a different result, finding his mixtures still toxic after
heating. Calmette, however, treated his solutions almost
immediately after mixing, and Martin and Cherry point
out that a certain time must be allowed to elapse for the
interaction to take place, and noted that moderate warming
hastens it, as is the case with all chemical interactions.
For instance, they found that one mixture of venom and
anti-venin allowed to interact for two minutes, five minutes,
and ten minutes before heating, killed the animals in
thirteen hours, fifteen hours, and twenty- three hours

1 Journ. of Path, and Bact., 1897, p. 460.

2 Proc. Roy. Soc. Lond., vol. Ixiii, 1898, p. 420.

TOXIN- ANTITOXIN REACTION 165

respectively (the control animal with the same dose of
venom died in nine hours), but after fifteen minutes the
same mixture rendered the animal ill but it survived, while
after thirty minutes no toxic symptoms ensued.

At one time it was stated that by electrolysis of toxin
small amounts of antitoxin are formed, but this is very
questionable. Electrolysis destroys the toxicity of toxins
by the production of acids, chlorine, and hypochlorites.

Ehrlich's views have been opposed, principally on physico-
chemical grounds. Thus, Danysz, observed that if ricin or diph-
theria toxin be brought into contact with its corresponding anti-
body, the degree of neutralisation depends on the manner of
mixture. If the toxin be added to the antitoxin in two fractions,
allowing a considerable time to elapse between the additions, the
mixture contains a much larger amount of free toxin than is the
case when the whole (and same) amount of toxin is added at once
to the antitoxin. This phenomenon, known as the " Danysz or
toxone effect," seems inexplicable if toxin and antitoxin have
relations the same as a strong base and a strong acid.

Arrhenius, Dreyer, and Madsen maintain that the phenomena
observed in the toxin-antitoxin reaction are explicable on the
hypothesis that the rate of reaction — avidity — of the toxin decreases
as antitoxin is added, that the interaction is a slow one, and that
different fractions of the toxin are progressively neutralised by the
added antitoxin, but more and more slowly. On these grounds
they consider that there is no reason to regard the diphtheria poison
as a highly complicated body. Whereas Ehrlich considers the toxin
and antitoxin to combine with great avidity, analogous to the
combination of a strong base with a strong acid, e.g. NaOH with
HC1, these critics believe the avidity of antitoxin for toxin to be
feeble, analogous to the combination of ammonia with boric acid,
in which as more and more acid is added, the amount of free
ammonia decreases, but more and more slowly, in correspondence
with a hyperbolic curve. The phenomena can be calculated accord-
ing to the law of " mass action," there being an equilibrium between

Free NH3. Free H3O3B = y(NH4H2O3B)2

vol. vol. vol.

where K is the constant of dissociation. The curve of the neutralisa-
tion of tetanolysin by anti-tetanolysin corresponds almost exactly
to the ammonia-boric-acid curve.

166 A MANUAL OF BACTERIOLOGY

Whereas on Ehrlich's views the combination of toxin and anti-
toxin would be represented by a straight line, and the crude toxin
seems to be composed of a whole series of different toxins and sub-
stances having an avidity for antitoxin, on this hypothesis, although
the greater part of the toxicity of toxin is removed by the anti-
toxin, the latter must be added in large excess before the toxicity
completely disappears, and the course of neutralisation would be
represented by a hyperbolic curve. In fact, as the antitoxin is
added, the amount of free toxin diminishes but never completely
disappears. There comes a point, of course, when the amount of
free toxin is so small as to be negligible and cannot be recognised
by the ordinary indicators (blood-corpuscles, animal tests, etc.).
This hypothesis would explain the fact that while a certain amount,
V, of a mixture of toxin and antitoxin is innocuous to an animal, a
multiple of the dose, n V, of the same mixture may be toxic ; it
would also explain Buchner's experiments alluded to above (p. 161),
and Roux's experiments in which a toxin-antitoxin mixture in-
nocuous to normal guinea-pigs was toxic to guinea-pigs whose
resistance had been reduced by injections of the Massowah vibrio.

Nernst has questioned from the mathematical standpoint the
validity of the views of Arrhenius, and so has Craw from much
experimental work on agglutination and on the interaction between
megateriolysin and anti-megateriolysin ; Craw also considers that
there is some doubt attaching to Arrhenius's calculations. Accord-
ing to Craw, the two substances most thoroughly investigated by
Arrhenius and Madsen, diphtheria toxin and tetanolysin, do not
admit of sufficiently exact determination, the former because of the
uncertainty attaching to animal experiments, the latter because
tetanolysin is a most unstable body. Working with a more stable
substance, megateriolysin, he holds that the Arrhenius and Madsen
equation does not apply. Again, on the addition of a small amount
of antitoxin to toxin there is no decrease in toxicity (as noted by
Ehrlich and attributed by him to the presence of toxoid) as there
should be, and Arrhenius was thus forced to the conclusion that a
second substance, epitoxonoid, is present with the toxin in diphtheria
toxin. Craw denies that the toxin-antitoxin reaction is reversible,
believes that antitoxin must be regarded as a colloid (and is not in
true solution), that the mixture therefore is heterogeneous, not
homogeneous, and that the chemical law of mass action is not
applicable.

On the other hand, Craw maintains that the phenomena of the
toxin-antitoxin reaction, including the Danysz effect, have their
counterpart in adsorption phenomena, such as occur in the staining

ADSORPTION 167

of paper, porcelain, etc., with anilin dyes, in the " adsorption " of
substances by colloids, etc.,1 and this view is supported by Bordet
and Gengou. Thus, when solutions of arsenious acid are shaken
up with colloidal ferric hydroxide, a portion of the arsenic is taken
up by the ferric hydroxide and a portion remains in solution.
Moreover, more arsenious oxide is taken up by the ferric hydroxide
from dilute than from concentrated solutions ; this has its counter-
part in agglutination. Again, when an antitoxin is added to a
toxin in just sufficient amount to produce a non- toxic solution, the
amount of toxin which must then be added to constitute a fatal
dose is greater than the minimum lethal dose without antitoxin.
This is also found to be the case with ferric hydroxide and arsenious
acid ; if ferric hydroxide and arsenious acid are mixed so as to form
just a non-toxic mixture, the amount of arsenious acid which must
then be added to render the mixture toxic is greater than the toxic
dose of arsenious acid.2

If pieces of filter-paper be placed in a dilute solution of stain at
sufficiently long intervals, the pieces first immersed will become
coloured while those last immersed will remain colourless. On the
other hand, if all the pieces be simultaneously placed in the solution
they all become coloured to the same degree. This is exactly
comparable to the Danysz effect. All the phenomena of the toxin-
antitoxin reaction seem best explained on the adsorption hypothesis
of Bordet. Specificity, it is true, is not completely explained
thereby, nor is it explained by any other hypothesis.3

The antitoxic constituent of antitoxin seems to be a protein
body, probably allied to globulin, and, as already mentioned, the
globulin content of the blood of an animal treated for antitoxin
production increases in some cases. Tizzoni, by precipitating the
antitoxic serum by saturation with magnesium sulphate at 30° C.,
obtained the antitoxin in the precipitate. By partial saturation of
antitoxic serum with ammonium sulphate, the antitoxin is carried
down with the second precipitate, that is, with the pseudo-globulin

1 " Adsorption " is physical in nature and mainly due to surface
condensation.

2 See Findlay, Physical Chemistry and its Applications in Medical
and Biological Science, 1905.

3 On the toxin-antitoxin reaction see Craw, Proc. Roy. Soc. Lond.,
B. vol. Ixxvi, 1905, p. 179 ; Journ. of Hyg., vol. vii, 1907, p. 501 ; and
ibid. vol. ix, 1909, p. 46 ; Arrhenius, Immuno-chemistry, 1907, and
Journ. of Hyg., vol. viii, 1908, p. 1 ; Madsen, Brit. Med. Journ., 1904,
vol. ii, p. 567 ; Bordet, Ann. de VInst. Pasteur, xvii, p. 161 ; McKendrick,
Proc. Eoy. Soc. Lond., B, vol. Ixxxiii, 1911, p. 493 ; Gengou, Journ. of
State Med^ xx, 1912, pp. 65 and 141 (Bibliog.)

168 A MANUAL OF BACTERIOLOGY

fraction. It is thus possible to concentrate antitoxic serum and to
make use of a weak serum, which would otherwise be inconvenient
on account of the volume necessary to inject in order to introduce
the requisite amount of antitoxin. For this purpose various salts
have been employed for saturation, ammonium sulphate (Pick and
others), magnesium sulphate (Dieudonne), mixtures of sodium and
potassium chlorides (Atkinson), etc.

Dzergowski and Predtechensky l have elaborated a very exact
method by which they state that the whole of the antitoxin can be
concentrated and recovered from a comparatively weak serum by
means of precipitation with ammonium sulphate.

ANAPHYLAXIS. — An animal usually becomes more and
more tolerant to injections of an antigen, e.g. to diphtheria
and tetanus toxins in the preparation of the corresponding
antitoxins. Sometimes, however, the opposite effect is
produced, viz. increased sensitiveness. This has been
noticed in the preparation of tetanus antitoxin ; after
the animal has received a few doses of the toxin without
ill-effect, a smaller dose of toxin may cause fatal tetanus.
The tuberculin reaction is, probably, another example ;
tubercle toxins circulating in the tuberculous individual
render him peculiarly sensitive to a minute dose of tuber-
culin (i.e. tubercle toxin) which in a normal person produces
no effect. Sensitisation may be obtained with difficulty
by administration by the mouth, and this may be the
explanation of the urticaria, etc., produced in some indi-
viduals by certain foods, e.g. shell-fish. This condition of
hypersensitiveness is known as " anaphylaxis " (i.e. the
opposite of " prophylaxis "). Probably any antigen under
particular conditions may induce anaphylaxis, but the
phenomenon has been especially studied in connexion with
serum injections, though any protein, e.g. egg-white or
bacterial cells, similarly causes it. The injection of an
anti- serum usually produces no ill- effect other than the
rashes, joint pains, and pyrexia already mentioned, even

1 .See Hewlett's Serum Therapy, 1910, p. 68.

ANAPHYLACTTC SHOCK 169

if large amounts of the serum be given extending over days
or even weeks, but a second injection of serum given after
a first injection with an interval of twelve days or more
between the two injections is liable to be followed by
effects which may be more or less serious, constituting
the so-called " anaphy lactic shock " or " serum disease,"
or immediate or accelerated reactions, " supersensitisation,"
may ensue (see p. 161).

In anaphy lactic shock, plain muscle contracts and Dale 1
has used the excised uterus of sensitised guinea-pigs to
give a graphic record of the action of the reacting dose.
Specificity is shown by the fact that the uterus of a guinea-
pig sensitised with sheep-serum contracts only when flooded
with a reacting dose of sheep-serum and not with any
other serum. The animal may be sensitised with two or
three different proteins and then the uterus contracts in
turn to each reacting dose of the different proteins. Once
the reacting dose has been given and the uterus has con-
tracted, the muscle is no longer sensitive to the protein.

The symptoms of anaphylactic shock are nausea and
vomiting, small and rapid pulse, faintness or more serious
heart failure, dyspnoea with rapid and shallow respiration
and feeling of suffocation, collapse, rigors, convulsions,
and even coma. The severity of the symptoms varies in
different cases, and the symptoms usually pass off in the
course of an hour or two ; but a few fatal cases have been
recorded. Death is easily produced experimentally, and,
post-mortem, scattered ecchymoses are found and a dis-
tended condition of the lungs due to spasm and contraction
of the bronchioles, to which the fatal event is due.

In the immediate reaction, rash, pyrexia, joint pains,
vomiting, rigors, and occasionally convulsions and collapse
occur, generally within six hours after the second injection
of serum. In the accelerated reaction, these phenomena

1 Jmirn. Pharmacd. and Exp*r. Therapeutics, IV, 1913-14, p. 167.

170 A MANUAL OF BACTERIOLOGY

appear between the eighteenth hour and the fifth day after
the second injection of serum.

The immediate and accelerated reactions may occur a
long time after the first course of serum treatment if more
serum be given. Goodall records one case in which over
four years elapsed between serum treatments for first
and second attacks of diphtheria, an accelerated reaction
occurring after the reinoculation for the second attack.

The amount of serum given does not definitely influence
the result. The remarkable features of the phenomenon
are — (1) they do not occur unless an interval of about
twelve days or more elapses between the two injections
of serum ; (2) the long period which may intervene between
the two injections of serum and still be accompanied by
symptoms ; (3) the serious nature of the condition in
some instances.

The explanation of the phenomenon is difficult. Un-
doubtedly the symptoms are due to some substance in
the serum which has a toxic action, and have nothing
to do with the antitoxic constituent, for normal serum
produces the same effects.

In experimental anaphylaxis produced in animals by
the injection of normal serum, it is found that the con-
dition only occurs if the two doses of serum are separated
by an interval of about twelve days or more ; the first is
termed the sensitising, the second the reacting, dose. The
larger the sensitising dose, the longer must the interval
be for the reacting dose to produce a maximum effect.
Moreover, the two injections must be of the same serum
or other protein ; thus a first injection of horse serum
followed by a second injection of rabbit serum would
not produce it. Extremely small doses of serum will
also bring it about ; and lastly, ansesthetisation, when
the second dose of serum is given, prevents the develop-
ment of the symptoms — a very extraordinary result.

ANAPHYLAXIS 171

The Arthus phenomenon occurs when a guinea-pig
receives several doses of normal horse serum at intervals
of some days. Another injection of horse serum then
causes an cedematous mass, an aseptic abscess, or an area
of necrosis at the site of the new inoculation, which may
be far removed from the region of the previous inoculations,
and the animal becomes cachectic and dies.

The Theobald Smith phenomenon occurs when a guinea-
pig has been sensitised by a very small single dose of normal
horse serum, 0-01 c.c., 0-001 c.c., or even 0-000001 c.c. ;
if, then, after an interval of twelve to fourteen days a
somewhat larger dose of serum, 0-1 c.c., be given, the serious
symptoms of hypersensitiveness develop within a few
minutes, viz. respiratory failure, paralysis, clonic spasms,
and frequently death. At one time it was believed that
a small sensitising dose is more effective than a large one
in producing anaphylactic shock, but it has been shown
that this is not the case, a large dose merely lengthens the
incubation period (up to, it may be, forty days). The
reason for this may be that the toxic substance slowly
formed by the sensitising dose combines as it is produced
with a part of the antigen injected, so that the ultimate
result is as though a small sensitising dose had been injected.

Various hypotheses have been advanced to account for
anaphylaxis. The fact that an interval or incubation
period is necessary for the development of the condition
clearly points to the formation of anti- bodies as a necessary
part of the phenomenon. Moreover, a " passive " anaphy-
lactic condition may be induced in an animal by injecting
it with the serum of a sensitised animal : this treated
animal suffers from anaphylactic shock on being injected
with the antigen. The substance which gives rise to the
anaphylactic shock is termed " anaphylatoxin " by
Friedberger and " apotoxin " by Richet.

Besredka believes that anaphylaxis is caused by the

172 A MANUAL OF BACTERIOLOGY

presence of two substances in the serum, one thermostable
and having the properties of an antigen (see p. 150), which
he terms " sensibilisogen/' and which on injection produces
its anti-body, " sensibilisin." The other substance is
thermolabile, and is termed " anti-sensibilisin," and
combines with sensibilisin whenever it meets with the
latter. Sensibilisin is particularly fixed by the cells of the
nervous system, and, according to Besredka, it is the
violent reaction between anti-sensibilisin and sensibilisin
in the nerve tissues which causes the serious disturbance
characteristic of anaphylaxis. When, therefore, a small
dose of serum (po,,-.-,1,, c.c.) is administered, the sensibili-
sogen slowly forms sensibilisin. If a second dose of serum
is given twelve days or more after the first injection, the
anti-sensibilisin in it combines with the sensibilisin formed
by the first injection, and disturbance results.

The reason why ana3sthetisation with ether when the
second injection is given prevents the symptoms of ana-
phylaxis developing is that the anaesthetic renders the
nerve cells insensitive to the reaction between the
sensibilisin and antisensibilisin.

According to Richet, a " toxigen " is formed in the blood
or cells at the end of the incubation period and persists
for a long period. A toxic apotoxin or precipitin is formed
as a result of the interaction of toxigen with antigen, the
toxicity of which is further increased by combination with
the alexin of the blood.

Bordet suggests that the union of anti-body and antigen
creates a complex which by adsorption monopolises certain
principles in the blood plasma which then becomes toxic.
Thus Wassermann and Reysser found that if guinea-pig
serum and kaolin, *an inert powder, be mixed and then
centrifuged, the intravenous injection of the fluid is
followed by symptoms closely resembling those of anaphy-
laxis. A weak agar jelly (0-05 per cent.) acts similarly.

ANTI-MICROBIC SERA 173

The serum must be fresh and active ; serum heated to
56° C. is inert.

Anaphylaxis, supersensitisation, or hypersensitisation
may be of considerable importance in serum treatment.

On the serum disease, supersensitisation, and anaphylaxis, see
Hewlett, Serum Therapy, ed. 2, 1910 ; Rosenau and Anderson,
Journ. Amer. Med. Assoc., 1906, p. 1007 ; Von Pirquet and Schick,
Die Serum-Krankheit, 1905 ; Richet, Ann. de Vlnst. Pasteur, xxi,
p. 497, and Anaphylaxis (Constable and Co., 1913. Bibliog.) ;
Besredka, Ann. de Vlnst. Pasteur, xxi, p. 950, and Bull, de Vlnst.
Pasteur, vii, 1909, p. 721 ; Currie, Journ. of Hygiene, vol. vii, 1907,
pp. 35, 61, and vol. viii, 1908, p. 457 ; Grunbaum, ibid. vol. viii,
1908, p. 9 ; Goodall, ibid. vol. vii, 1907, p. 607 ; Bordet, Journ.
State Med., 1913, p. 449.1

ANTI-MICROBIC SERA. — If an animal be injected with
increasing doses of bacteria, care being taken to keep
below a lethal one, the animal gradually becomes accus-
tomed to the microbe, and ultimately acquires a high
degree of immunity, so that it is unaffected by amounts
which would infallibly kill an untreated animal. More-
over, the blood-serum of such a treated animal, if injected
into a second animal, will protect the latter against a few
lethal doses of the microbe, but not against a large amount.
Nor is the protection afforded proportional to the amount
of serum injected ; for example, if 0-005 c.c. of anti-cholera
serum will protect against 5 mgrm. of living cholera culture,
three times as much, or 0-015 c.c. of the serum, will not
protect against 15 mgrm. of cholera culture, and when a
certain dose of the culture is reached no amount of serum
will save the animal. The mode in which the serum acts
may be studied microscopically. If cholera anti-serum
and cholera culture be injected into the peritoneal cavity
of a guinea-pig, and the peritoneal contents be examined
at short intervals afterwards, it will be found that the

1 Trans. XVIIthlntcrnat. Cong, of Medicine, 1913, Sect. IV, Pt. I, pp. 1
(Bobi-cdka) and 13 (Richet), and ibid. Pt. II.

174 A MANUAL OF BACTERIOLOGY

vibrios lose their motility, become distorted and globular,
undergo solution, and finally disappear. The protection
afforded by the anti-serum is therefore due to the
destruction of the microbes by solution, the process
being known as bacteriolysis,1 and the bodies which bring
it about being termed " bacteriolysins." The reaction is
known as " Pfeiffer's phenomenon " or reaction, from its
discoverer. If the serum and the microbes be mixed in
vitro the latter are unaffected ; apparently, therefore,
some constituent of the living body in addition to the
anti-serum is necessary for the solution of the microbes.
But in 1895 Metchnikoff showed that the reaction will
take place in vitro provided that some of the fresh peri-
toneal exudate of a normal guinea-pig be added to the
mixture of anti-serum and microbes. The same year
Bordet found that the addition of the peritoneal exudate
is unnecessary provided the anti- serum be perfectly fresh.
These experiments prove that the solution of the microbes
is brought about by the interaction of at least two sub-
stances, one of which is present in all fresh serum and in
the living body, but is unstable, disappearing on keeping
or heating the serum, the other is a relatively stable body
produced during the process of inoculation. The former,
the unstable normal body present in all animals, is usually
termed " complement " (Ehrlich and Morgenroth), " alexin "
(Buchner and Bordet), or " addiment " ; while the stable
constituent produced by immunisation is known as the
" amboceptor " (Ehrlich), " immune body," " interme-
diary," "preparer" (Gruber), " fixateur " (Metchnikoff),
or " substance sensibilisatrice " (Bordet).

These considerations suggest an explanation why anti-microbic
serum neutralises only a limited amount of living culture, viz. the
amount of complement present in the body at one time is limited,
and when this has been used up bacteriolysis ceases. Anti -micro bic
sera are relatively inefficient in practice, insufficiency of complement

1 See Gruber, " Harbcu Lectures," Journ. State Med., 1902.

AMBOCEPTOR AND COMPLEMENT 175

being suggested as the reason. Attempts have been made to supple-
ment the complement present by injecting fresh normal serum with
the anti-serum, but without success, and some anti-micro bic sera,
e.g. anthrax serum, are not bacteriolytic ; this explanation is, there-
fore, unsatisfactory. Deflection of complement (p. 178) may occur
in some instances, or the complement may not be of the right kind.
In other cases, the organism in certain situations may be inaccessible
to the blood-stream and to the anti -serum,
e.g. the vibrios, in the bowel in cholera.

Another reason advanced is the extreme
specificity of anti-serum and the variability
of bacteria so that many races or strains
of an organism may exist, e.g. of B. coli,
streptococci, pneumococci, etc. Hence the
anti-serum prepared with one race may not
neutralise another race. Attempts have
been made to overcome this factor by pre-
paring the anti-serum by the injection of
many races and so obtaining a " polyvalent
serum."

FIG. 33.— Diagram to

The amboceptor or immune body show the union be-
seems to link the complement to the *™ee" complement

(black) and proto-

bactermm (Fig. 33) ; complement re- piasm of cell by

mains free if the appropriate ambo- means of the ambo-

u j • ceptor (white). (After

ceptor or immune body is not present, Ehrlich )

and bacteriolysis does not ensue (see
also p. 174). Complement is thermolabik, i.e. it is destroyed
by heating to 56° C. for thirty minutes ; while the ambo-
ceptor is thermostable, i.e. it is not destroyed by this
treatment.

According to Ehrlich, fresh serum contains numerous
complements which are more or less specific for different
amboceptors (see also note, p. 182). When the comple-
ment is destroyed by heating it is converted into " comple-
mentoid " (analogous to toxoid). Both complement and
complementoid on injection give rise to anti- complement.
The amount of complement in different sera varies con-
siderably ; horse serum contains very little, guinea-pig
seruni much. Complement itself probably consists of two

176 A MANUAL OF BACTERIOLOGY

portions, as it is generally accepted that it can be split
into a " mid-piece " and an " end-piece " by the action of
dilute hydrochloric acid, carbon dioxide, and dialysis.
The mid-piece is thought to be in the globulin fraction,
the end-piece in the albumin fraction. Noguchi, however,
considers that the whole complement is present in the
albumin fraction and that inactivation of the complement
by acid, etc., is due not to splitting into two fractions, but
to inactivation of the whole complement.

Pfeiffer's reaction is of considerable value in practical
bacteriology for the exact recognition of bacterial species.
A mixture of a suspension of the organism to be tested with
a small quantity of serum from a highly immunised animal
is injected into the peritoneal cavity of a normal guinea-
pig. The fluid in the peritoneal cavity is then examined
microscopically half to one hour after the injection, and
if the reaction be positive the organisms will be found in
all stages of degeneration, being mostly converted into
spherules. In this case, according to Pfeiffer, the organism
is to be regarded as belonging to the same species as that
by means of which the immunisation of the animal, from
which the blood-serum was obtained, was carried out. If,
on the other hand, the reaction be negative, the organisms
are unaffected after being in the peritoneal cavity for an
hour or so, and the organism is then considered to be a
species different from that used for the immunisation.
Thus, Pfeiffer's reaction may be made use of to differentiate
the cholera-like vibrios from true cholera vibrios and the
members of the typhoid- colon group from one another.

The destruction of the bacteria by bacteriolysis is
regarded by some as being brought about by osmotic
changes, by others by processes analogous to digestion.
During bacteriolysis the specific immunising substances
and anti-bodies are used up, and for the lysis of a given
quantity of bacteria a certain amount of immune serum

PFEIFFER'S REACTION 177

is necessary, while after lysis has taken place the latter
loses the power of dissolving bacteria. The same holds
good for haemolysis, and the facts relating to bacteriolysis
and haemolysis are almost interchangeable.

Anti-endotoxic sera. — The comparative inefficiency of anti-
microbic sera, particularly typhoid, led Macfadyen to attempt to
prepare sera with microbial endotoxins, and the work has been
continued by Siidmersen and the writer. The method was to
immunise horses with the endotoxin obtained by the method
described on p. 40. With a typhoid serum so prepared Goodall
and the writer obtained promising results.1

Method of applying Pfeiffer's reaction. — For Pfeiffer's test, the
organism must be virulent, and a high-grade immune serum is
necessary. If the organism is not virulent, it is spontaneously
destroyed in the peritoneal cavity without the addition of immune
serum. The method may be best explained in the case of a vibrio
supposed to be the cholera vibrio. The cholera-immune serum
(obtained from a horse repeatedly injected with cholera culture)
should possess a titre of not less than 0-0002 c.c., i.e. this amount
of serum mixed with one loop (2 mgrm.) of an eighteen-hour agar
cholera culture (virulent), suspended in 1 c.c of broth, and injected
into the peritoneal cavity of a small guinea-pig should cause granular
degeneration and bacteriolysis of the vibrios within one hour.

Four mixtures are made — (a) one loop of an eighteen-hour agar
culture of the vibrio to be tested, 0-001 c.c. cholera-immune serum,
suspended in 1 c.c. of broth ; (&) the same as (a), but 0-002 c.c.
cholera serum ; (c) the same as (a), but 0-001 normal serum of an
animal of the same species as that furnishing the cholera serum ;
(d) one quarter loop of the vibrio in 1 c.c. of broth, as a control of
the virulence of the culture. These mixtures are then injected into
the peritoneal cavities of four guinea-pigs each of about 250 grin,
weight. At intervals of thirty and sixty minutes hanging-drop
preparations are made of the peritoneal fluid of each animal, the
fluid being obtained by inserting a capillary pipette through a
minute incision in the skin. In the guinea-pigs injected with (a)
and (6), if the organism be cholera, the vibrios should show marked
degenerative changes within sixty minutes, while (c) and (d) will
show plenty of active vibrios. If the organism be non-virulent,
two methods may be adopted for applying the Pfeiffer reaction.
The first, a microscopical or direct method, is carried out by micro-

1 Proc. Roy. Soc. Med., vol. ii, 1907-8, Med, Sect., p. 245 et seq.

12

178

A MANUAL OF BACTERIOLOGY

scopical examination of hanging-drop specimens of the organism
suspended in a drop of the immune serum to which a trace of fresh
peritoneal fluid (complement) is added. If the organism is homolo-
gous with the immune serum, the bacteria are soon transformed
into granules. Controls are put up at the same time with a known
strain of the organism with (1 ) its homologous immune serum -f- com-
plement ; (2) non-immune serum
of the same animal -f- comple-
ment ; also of the organism being
tested with non-immune serum of
the same animal -j- complement.
The peritoneal fluid may be ob-
tained by injecting 3-4 c.c. of
broth into the peritoneal fluid
of a guinea-pig and four hours
later withdrawing the fluid (now
turbid with leucocytes) and oen-
trifuging, or allowing it to
stand on ice for twenty-four
hours.

In the second, or indirect,
method, the organism is used to
prepare an immune serum by
injecting an animal (e.g. a rabbit)
with it, and the immune serum
so prepared is tested on a known
virulent stain in the peritoneal

cavity of guinea-pigs in order to ascertain whether or no it brings
about bacteriolysis, i.e. the Pfeiffer phenomenon.

Deflection, deviation,1 diversion or blocking of complement. —
Pfeiffer in 1895 observed that a large amount of immune serum
might not protect an animal from the cholera vibrio, while a smaller
amount with the same dose of vibrio did so. In 1901 Neisser and
Wechsberg demonstrated an analogous reaction in vitro. They
studied the effect of a bacteriolytic immune serum when varying
amounts of the inactivated serum were employed. The quantity
ranged from 0-0005 c.c. to 1 c.c. To each of these amounts constant
volumes of normal serum and bacterial suspension were added. No
bacteriolysis occurred when large and small amounts of immune
serum were used, but with medium amounts bacteriolysis was
complete. Theyj^explained this anomalous reaction, the absence

1 " Fixation of complement " (p. 183) is frequently erroneously
termed " deviation of complement."

FIG. 34. — Diagram to represent
the condition of the blood in
which there is an excess of
amboceptors. The ambocep-
tors (white) unite with both
complement (black) and re-
ceptors (dotted), so that the
receptors cannot combine with
the amboceptor-complement
groups.

AGGRESSItfS 179

of bacteriolysis with large amounts of immune serum, as follows :
When the amboceptors are in large excess, a portion combines
with the complement, leaving some amboceptors free, and these
free amboceptors then unite with the receptors before the activated
amboceptors (amboceptors + complement) do, and thus the comple-
ment-amboceptor groups are rendered inert. The reaction is
represented diagrammatically in Fig. 34. Arrhenius, however, does
not accept this explanation. He says : "If we have the compounds
ea and ab which may combine to form the compound eab, the
formation of the latter depends wholly upon whether e has a greater
affinity for ab than for a. If not, then eab is not formed, even if a
is not present in excess." (a = amboceptor, e = microbe, b = com-
plement.) The phenomenon may be quite analogous with the
inhibition met with in agglutination (p. 188).

Aggressins

Bail has discussed the question of the relationship between
bacteriolysis and immunity. He argues that there is apparently
little relationship between the bactericidal properties of the body
fluids and the immunity of an animal to infection through bacterio-
lytic processes ; and points out that in rabbits immunised against
anthrax there is no bacteriolytic power, the bacteria disappearing
gradually as the result of phagocytic action of cells, chiefly marrow-
cells ; that a comparison of the sera of sheep, rabbits, and cattle
shows great variation in their content of immune body, though the
animals are almost equally susceptible to anthrax ; and that in
test-tube experiments a bacteriolytic serum is blocked when the
conditions are approximated to those in the body by the addition
of body cells to the mixture ; the bactericidal properties of the
serum disappear or are greatly inhibited. Kruse suggested that for
infection to take place the invading bacteria must elaborate chemical
substances which so act on the cells and fluids of the invaded animal
that they overcome its natural resistance against infection. These
substances are considered by him and Bail to be distinct from the
toxins, and are termed by these writers " aggressins. " x The
aggressins are supposed to be secreted by the living uninjured
bacteria and not to be extracts, nor derived by solution, of the
bacteria ; they occur particularly in the fluids of pathological

1 See Cenlr. f. Bakt., Orig., xlii, 1906, pp. 51, 139, 241, 335, 437, and
546. Also an excellent summary by Marshall, Philip'pine Journ,
of Science, vol. ii, 1907, p. 352>

180 A MANUAL OF BACTERIOLOGY

oedemas and cxudates, and may be obtained from these by centri-
fugation and sterilisation at low temperatures. Bail believes
that the aggressins cannot be anti-complements, anti-immune
bodies, etc., but are substances heretofore unrecognised and the
active substances of the infection, and he considers that in order to
produce true immunity in disease anti-aggressin sera must be
prepared. The following are some of the properties of these sup-
posed aggressins : (1) Sterilised aggressin with a non-lethal dose of
the corresponding organism renders the latter fatal ; (2) aggressin
alone is only slowly toxic, producing a prolonged illness with
emaciation preceding death ; (3) inoculation of aggressin with
bacteriolytic serum into the peritoneal cavity suspends the action
of the latter ; (4) aggressin with bacteria blocks phagocytosis.
Bail believes that the aggressins promote infection by interfering
with the protective mechanism of the infected animal, particularly,
if not solely, by inhibiting phagocytosis. Upon the power to pro-
duce aggressin Bail has classified bacteria into (1) true parasites
which always produce aggressin, e.g. anthrax and chicken cholera ;
(2) half -parasites, the aggressin-producing power of which is variable,
e.g. typhoid, cholera, dysentery, and plague ; (3) saprophytes. The
virulence of an organism does not coincide with aggressivity, and
extremely virulent bacteria may be half -parasites.

Bail's hypotheses have been much criticised, and Wassermann
and Citron believe that the supposed aggressins are derivatives of
the bacterial protoplasm which have the power of combining with
the specific protective substances of the animal and so inhibit
the action of the latter ; they are, in fact, endotoxins of feeble
toxicity.

HAEMOLYSIS.1 — Some blood sera possess marked powers
of dissolving the red blood- corpuscles of another species,
and of setting free their contained hemoglobin (e.g. goat
serum dissolves rabbits' and guinea-pigs' corpuscles, and
ox and human sera usually dissolve sheep's corpuscles),
and if an animal be injected with the blood-corpuscles of
another species its blood-serum generally acquires the
property of dissolving the blood- corpuscles with which

1 See Bulloch, Practitioner, December 1900, p. 672, and Trans. Path.
Soc. Lond., vol. Hi, Part 3, 1901, p. 208 ; Gruber, " Harben Lectures,"
Journ. State Med., 1902, February, March, and April ; Ehrlich, Collected
Studies on Immunity ; Muir, Studies on Immunity.

HAEMOLYSIS 181

it has been injected. For example, the serum of a normal
rabbit has no haemolytic action upon the red corpuscles
of the sheep ; but if a rabbit receive a few injections of
defibrinated sheep's blood, its blood-serum acquires
haemolytic properties and dissolves the red corpuscles of
the sheep. This solution of the blood- corpuscles is termed
" haemolysis," and the substances which produce haemo-
lysis are " haemolysins." If the active serum be heated to
56° C. it is " inactivated " and loses its haemolysing
power, but can again be rendered haemolytic or " acti-
vated " by the addition of fresh normal serum ; normal
serum, however, rapidly loses its activating properties
on keeping. It will thus be seen that there is an almost
complete analogy between bacteriolysis and haemolysis,
the latter being brought about by the interaction of two
substances, one specific and stable produced by the injec-
tions, the haemolytic " amboceptor " or " immune body,"
and the other an unstable body present in fresh normal
serum, the " complement " or " alexin."

Haemolysin formed by the injection of corpuscles of
another species is termed " heterolysin." If corpuscles
of the same species be injected, haemolysin is formed
(" isolysin "), but the injection of the animal's own cor-
puscles does not give rise to haemolysin, i.e. " autolysin "
is not formed.

Blood- corpuscles are more tangible entities than bac-
teria, and are far easier to work with than the latter, and
haemolysis has been the subject of a large amount of
experimental work by Bordet and Gengou, Ehrlich, Mor-
genroth, Gruber, Bulloch, Muir, and others, and the results
obtained have shed considerable light upon the complex
phenomena of immunity and of the actions of anti- bodies
in general. Moreover, the globulicidal material in haemo-
lysis seems to be identical with the bactericidal one in
bacteriolysis — that is to say, it is the complement or

182 A MANUAL OF BACTERIOLOGY

alexin. x According to Ehrlich's view, whether it be normal
or " immune " serum (i.e. serum of a treated animal),
bacteriolysis or haemolysis takes place only when the
complement and amboceptor unite (Fig. 33, p. 175),
complement by itself having little affinity for the bacterium
or erythrocyte, the combination forming the " lysin,"
which then acts. According to Gruber, however, neither
bacteriolysin nor hsemolysin exist as a chemical entity,
the specific bacteriolytic or hsemolytic action being due
to the fact that the cells first absorb the amboceptor and
so become accessible to the complement, for the two
substances do not combine in definite proportions — the
more the blood- corpuscles are laden with the amboceptor
the smaller the quantity of complement required to bring
about their solution.

Many bacteria — e.g. B. pyocyaneus, B. typliosus, staphy-
lococci and streptococci — produce hsemolysins, and the
haemoglobin staining occurring in septic diseases, etc., is
probably partly due to the action of bodies of this nature
elaborated by the infecting organisms.

Practical Uses of Haemolysis, etc.

1. Haemolysis test. — Some micro-organisms produce non-specific
hsemolysins, others do not ; this may constitute a difference between
allied organisms. For instance, as a rule true cholera vibrios do
not haemolyse, while many cholera-like vibrios do. The test can be
applied in two ways : (a) Defibrinated rabbits' blood may be mixed
with melted agar cooled to 45° C. The mixture is poured into Petri
dishes, allowed to set, and when cool inoculated with the organism

1 As previously stated (p. 175), numerous complements undoubtedly
exist, yet bacteria will absorb both bacteriolytic and haemolytic com-
plements. Bordet and Gengou suppose that while a particular ambo-
ceptor has a maximum avidity for its homologous complement (which
may be termed dominant), it is also able to take up other " non-
dominant " complements, and thus bacteriolytic amboceptor is able to
absorb both bacteriolytic (dominant) and hsemolytic (non-dominant)
complements.

FIXATION OF COMPLEMENT 183

to be tested in such a manner that separate, well-defined colonies
are obtained. After twenty-four hours' incubation at 37° C.,
colonies when haemolytic are surrounded with a clear, well-defined
halo contrasting sharply with the dark opaque colour of the agar.
If blood-agar is not available, a substitute may be devised by smear-
ing some sterile human or rabbits' blood on a sterile agar plate.
(b) A young agar culture is emulsified in 4-5 c.c. of physiological
salt solution ; 0-1 c.c. of this suspension is mixed in a tiny test-tube
with O9 c.c. of sterile salt solution and one drop of a sterile suspen-
sion of well- washed rabbit or other corpuscles. After twelve to
twenty -four hours haemolysis will be apparent if the organism forms
haemolysins.

2. Fixation or absorption test.1 — A haemolytic serum may be used
as a delicate reagent for complement, and may thus serve as a
test for an organism or an immune serum. As an example take the
case of a supposed cholera vibrio. If an immune serum (previously
heated to 56° C. so as to destroy complement) — haemolytic for the
corpuscles of an animal, or bacteriolytic for a given micro-organism,
e.g. cholera vibrio — be mixed with the red corpuscles of the same
animal, or with the cholera vibrio, the corpuscles or the vibrios
respectively absorb the corresponding amboceptor or immune body.

Bordet showed that if corpuscles or microbes that have absorbed
the corresponding amboceptor be added to fresh non-heated comple-
ment (e.g. fresh guinea-pig serum), the corpuscles or the microbes
absorb the complement, so that none remains free in the liquid.

But if fresh guinea-pigs' serum be added to cholera vibrios which
have not absorbed any cholera amboceptor, the complement will
not be absorbed and remains free in the liquid. The proof of this
is that if " sensitised " corpuscles (i.e. corpuscles which have taken
up haemolytic amboceptor) be added to such a mixture, the globules
are quickly haemolysed. If, on the other hand, vibrios which have
already taken up the cholera amboceptor be added to the same
quantity of fresh serum, the microbe-amboceptor complex absorbs
the complement ; and, provided the amount of fresh serum is not
too great, the complement is absorbed so completely that " sensi-
tised " corpuscles when added to the mixture are not dissolved.
If vibrios other than cholera be added to cholera serum, the ambo-
ceptor is not fixed, the complement added remains free, and the
sensitised corpuscles are dissolved. These facts constitute the
" Bordet-Gengou phenomenon." The mixture of an inactivated
haemolytic serum (i.e. heated to 56° C.) with the homologous
corpuscles (i.e. those with which the haemolytic serum was prepared)

1 Often termed " deviation of complement " test.

184 A MANUAL OF BACTERIOLOGY

is known as a " haemolytic system." The following example illus-
trates the method of carrying out the test: The cholera-immune
serum is heated to 56° C. for half an hour. An eighteen hours old
agar culture of the organism to be tested is suspended in 2 c.c. of
sterile physiological salt solution. The complement is fresh guinea-
pig serum ; a portion of this is also heated to 56° C. (= non-immune
serum). The following mixtures are prepared in three small
test-tubes :

Tubes 1 and 2 each contain 0-2 c.c. microbic suspension + 0-6 c.c.

heated immune serum + 0-1 c.c. complement.
Tube 3 contains 0-2 c.c. microbic suspension +0-6 c.c. heated

non-immune serum + 0-1 c.c. complement.

These are well shaken to mix their contents, and are kept for
half to one hour at 37° C. At the end of this time 0-1 c.c. of the
following mixture is added to tubes 1 and 3 : two volumes of heated
(to 56° C. for half an hour) serum hsemolysing sheep's red corpuscles
+ one volume of washed sheep's corpuscles. To tube 2 is added
0-1 c.c. of a mixture of two volumes of physiological salt solution +
one volume of washed sheep's corpuscles. The tubes are kept for
a further hour or so at 37° C., and at the end of that time the
occurrence of haemolysis is noted. If the organism is homologous
with the immune serum, the immune body will fix the complement
in tube 1 and no haemolysis will occur ; in tube 3 haemolysis will
occur because the complement remains free. Tube 2 serves as a
control, and should show no haemolysis in three hours (though if
kept for eighteen to twenty-four hours haemolysis will occur if the
organism produces hcemolysins, apart from any action of comple-
ment). If the organism is not homologous with the immune serum,
haemolysis will occur in tube 1, because the complement docs not
become fixed, tubes 2 and 3 being the same as before.

It is not even necessary to use the living organism ; the dead
organism or extracts thereof, and, in cases where the organism
cannot be cultivated, a dried and pulverised organ or an extract
thereof, has been employed. Certain non-specific substances may
sometimes be used as in the Wassermann reaction for syphilis (see
" Syphilis ").

The haemolytic serum may be obtained by injecting rabbits with
a 10 per cent, suspension of well-washed sheep's red corpuscles.
The sheep's blood should be obtained as ascptically as possible
from the slaughterhouse ; the blood, as it runs, is caught in a
sterile wide-mouthed bottle containing a coil of fine wire witli which
it is defibrinated by shaking. The iiuid blood is then mixed with

CYTOTOXINS 185

sterile physiological salt solution (0-9-0-95 per cent.) and centri-
fuged, and the deposited corpuscles are again washed with salt
solution two or three times. Three doses of 1 c.c., 2 c.c., and 3 c.c.
respectively are given intravenously on successive days, and after
an interval of 5-7 days the rabbit's serum should be strongly h«3mo-
lytic. Very active hsemolytic sera may be purchased. The serum
may be collected aseptically, inactivated by heating to 56° C. for
half an hour, and preserved in sealed ampoules. The activity of
the hoemolytic arnboceptor must be tested and the appropriate dose
of it, complement, and corpuscles ascertained. (For manner of
testing, see " Syphilis.")

CYTOTOXINS. * — Anti-sera, analogous to the hsemolysins or hasmo-
toxins, may be prepared which have a destructive action upon
cellular elements ; these are termed " cytotoxins." If a rabbit be
injected with bull's semen, its serum (" spermo toxin ") acquires
the property of immobilising the spermatozoa of the bull. The
reaction is specific, but spermatolysis does not seem to occur.
Similarly, by injecting ciliated epithelium into the peritoneum of a
guinea-pig an anti-epithelial serum, or " trichotoxin," is developed.
With liver, kidney, and nerve cells anti-bodies having a destructive
action upon these cells are developed as a result of their injection.
Nephrotoxin, the serum of an animal inoculated with an emulsion
of kidney, when injected into a second untreated animal, produces
albuminuria and urajmia with disintegration of the epithelium of
the convoluted tubules ; hepatotoxin, the serum of an animal
treated with emulsions of liver, produces fatty and inflammatory
changes in the liver resembling phosphorus poisoning ; neurotoxin,
the serum of an animal treated with emulsions of nerve tissues,
produces paresis, paralysis, depression, convulsions, etc. ; a leuco-
toxic serum obtained by injecting leucocytes agglutinates and dis-
solves the leucocytes, and so on. The formation and mode of action
of these cytotoxins resemble those of the haemolysins. It was
hoped that the study and preparation of cytotoxins would open up
possibilities in the way of treating such diseases as carcinoma and
sarcoma, but so far this hope has not been realised.

AGGLUTINATION. — If an animal be injected with cultures
of typhoid or cholera bacilli, its serum soon acquires the
property of agglutinating or of aggregating into clumps the
typhoid bacilli or cholera vibrios respectively when mixed
with a broth culture of these organisms. The reaction may
1 Sue Uulloch, Pracllltoiier, May 1901, p. 499 (Bibliog.)

186 A MANUAL OF BACTERIOLOGY

be observed microscopically in a hanging- drop preparation ;
the organisms first lose their motility and soon become
aggregated into large masses or clumps. Macroscopically,
the reaction may be followed in a narrow test-tube into
which the mixture of culture and serum has been intro-
duced ; after some hours the micro-organisms become
aggregated into masses so large as to form visible flocculi.
The substances which bring about this agglutination are
known as agglutinins. Agglutinins seem to be present
in small amount in normal serum ; for instance, most
normal human sera up to a dilution of 1 in 2 or 1 in 4 will
agglutinate the typhoid bacillus and still more powerfully
the glanders bacillus. They are also present in bacterial
cultures ; if an old broth culture of typhoid be filtered,
the filtrate agglutinates the bacilli in a fresh broth culture ;
hence young cultures should always be used for agglutina-
tion tests. Agglutinin is formed by the action of antigen
derived from the bacterial cell, but may also be naturally
present. Agglutination is brought about by the action
of the agglutinin on the antigen ; the agglutinin first
unites with the antigen, and this may occur at 0° C., and
afterwards exerts its specific action, which takes place
only at higher temperatures and in the presence of certain
salts. The agglutinable substance is known as aggluti-
nogen. Agglutinin is converted into agglutinoid at 70°-
75° C. ; the latter does not agglutinate, though it unites
with bacteria and then prevents the subsequent action of
agglutinin.

The agglutination of organisms by anti-sera, though
hardly specific, is usually very special ; given proper
precautions as to dilution, time-limit, condition of test
culture, etc., an anti-serum will generally only agglutinate
the homologous organism or closely allied species — that is,
it is a group reaction. The anti- serum may agglutinate
both the organism with which it has been prepared, and

AGGLUTINATION 187

also allied species, though usually not to the same extent ;
anti- typhoid serum, for example, may agglutinate not
only the typhoid bacillus, but also, though to a less degree,
members of the paratyphoid group. As the result of
infection or of inoculation with an organism, agglutinins
may, however, be produced which agglutinate not only
the organism of the infection, but also other organisms —
e.g. typhoid serum may agglutinate the B. coli as well
as the B. typhosus and typhus serum B. typhosus and M.
melitensis. The agglutinins acting on the infecting organ-
ism may be termed primary or homologous, those acting
on other organisms secondary or heterologous. In a case
of double infection each organism may produce its own
primary agglutinin, so that the agglutination of two
species by a serum may be due to the presence either of a
primary and a secondary agglutinin or of two primary
agglutinins. Castellani,1 by applying the saturation test
(p. 193), found that an organism would absorb both its
primary and secondary agglutinins, but would not absorb
two different primary agglutinins. This test, therefore,
would distinguish a double infection from a single one.
Thus, if a typhoid serum agglutinated both the B. typhosus
and the B. coli, and the serum after saturation with typhoid
culture still agglutinated the B. coli, this would point to
an infection with the latter as well as with typhoid. The
formation of primary and secondary agglutinins may be
brought about as follows : In the bacterial cell there are
several substances, each of which forms its own agglutinin.
The cells of two bacterial species we can imagine both
contain three or four substances capable of producing
agglutinins, and it may happen that one of these in each
species is the same and will produce the same agglutinin —
the secondary agglutinin — and, therefore, the serum
produced by each bacterium will agglutinate the other.
1 Zeitschr. /. Hy<j., xl, 1902, p. 1.

188 A MANUAL OF BACTERIOLOGY

The agglutination reaction is made use of in bacterio-
logical tests and in clinical diagnosis. The " Bordet-
Durham " reaction consists in testing an unknown organism
with a specific anti- serum prepared by injecting an animal
with a known microbe ; if the organism tested becomes
agglutinated, it is regarded as being of the same species
as that with which the anti-serum was prepared. With
certain precautions the " Bordet-Durham " reaction is
one of the most delicate and certain for the recognition of
bacterial species. The converse of this is the agglutination
reaction proper (frequently termed the Widal reaction),
and consists in testing an unknown serum upon a
known microbe. It is especially used in the diagnosis of
microbial diseases ; for example, in typhoid fever the
blood of the typhoid patient powerfully agglutinates the
typhoid bacillus, that of Malta fever the Micrococcus
melitensis, that of bacillary dysentery the dysentery
bacillus, etc.

A remarkable phenomenon observed in connection with
agglutination, which the writer has particularly noticed
in the case of Malta fever, is the occurrence of what may be
termed a zone of no reaction or of inhibition with some
particular dilution. Thus, dilutions of 1 in 10 and 1 in 20
may agglutinate strongly, a 1 in 30, however, may hardly
agglutinate at all, while dilutions of 1 in 40 and upwards
to 1 in 100 or more may agglutinate well. A similar
phenomenon has been observed with non-specific agglu-
tinating agents, and also in the action of coagulating agents
on colloid emulsions. Thus orthophosphoric acid agglu-
tinates a certain volume of a suspension of B. coli when
present to the extent of between 118 cgrm. and 4 cgrm.,
and between 1-1 mgrm. and 0-001 mgrm., but not in
intermediate amounts between 40 and 1-1 mgrm.

Anti-serum, prepared by injecting erythrocytes, also
agglutinates the red blood- corpuscles, and in certain

THEORIES OF AGGLUTINATION 189

diseases, e.g. pneumonia, chromocyte clumping may be a
marked feature.

Various theories have been propounded to account for
the phenomena of agglutination :

1. Pfeiffer and Emmerich and Loew regarded agglutina-
tion as a vital paralysis of the bacilli due to the action of a
bacteriolytic enzyme. Agglutination, however, is not a
vital phenomenon, for dead bacilli agglutinate, and bac-
teriolytic enzymes seem to be destroyed by temperatures
at which agglutinins remain unaffected.

2. Gruber, Dineur, and Nicolle supposed that a glutinous
substance, " glabrificin," is absorbed from the serum by
the bacilli causing the cell membranes or the flagella to
become adhesive ; but this explanation will hardly account
for the aggregation of non- motile organisms.

3. Paltauf and Duclaux considered that a precipitate
is produced in the medium, which during flocculation
mechanically carries the bacilli with it ; but there is no
demonstrable evidence that such precipitation occurs.

4. Bordet separated the mechanism of agglutination
into two stages — (1) fixation of agglutinin, and (2) aggre-
gation. The fixation of agglutinin by the organisms he
considers to be analogous to the adsorption of a dye by a
tissue ; and once the agglutinin is fixed, the organisms
obey the laws of inert particles, aggregation being caused
by changes in surface tension, in the molecular attraction,
between the organisms and the surrounding medium, a
view supported by Craw.1 Ohno,2 however, believes
that the union of agglutinin and agglutinable substance
is not analogous to the fixation of a dye by a tissue, but
that it is a chemical combination, as maintained by Ehrlich.

Agglutinated bacteria are not injured by agglutination ;

1 Journ. of Hygiene, vol. v, 1905, p. 113. See also Joos, Zeitschr. f.
Hyg., xxxvi, p. 422, and ibid, xl, p. 203.

2 Philippine Journ. of Science, vol. iii, 1908, p. 47.

190 A MANUAL OF BACTERIOLOGY

they will, in fact, grow and multiply in an agglutinating
serum. The amount of agglutination does not bear any
constant ratio to the intensity of an infection ; on the
whole, if the patient is reacting satisfactorily to an infec-
tion, the agglutination reaction tends to be marked ; if
not, it may be feeble or absent. Thus, in severe typhoid
infections with fatal issue, agglutination may be absent.
RufEer and Crendiropoulo * regard the agglutinins as
being formed in the polymorphonuclear leucocytes.

The Agglutination Reaction

A. For Clinical Diagnosis (" Widal " Reaction)

This is principally made use of in typhoid and paratyphoid fevers,
Malta fever, and bacillary dysentery.

Collection of blood. — Blood is collected (p. 214), preferably in a
Wright's capsule (Fig. 35, d, p. 215), or in a capillary bulbous pipette
(Fig. 7, p. 52), or in a vaccine tube. The ends of the tube are sealed,
the dry end always being sealed first ; the blood is allowed to
coagulate (which may be hastened by placing in the blood-heat
incubator), and then centrifuged to separate the serum, care
being taken that the dry sealed end of the tube, which will be
perfectly sealed, is distal when spinning.

If tubes are not available, the blood may be spotted on to a
piece of glass, cover-glass, or slide, glazed paper, tinfoil, etc., and
allowed to dry. For use, a drop of distilled water is placed on the
dry blood to dissolve it, and the solution used like serum.

The culture. — For the microscopic test a young broth culture is
to be preferred. A hanging drop should be examined to ascertain
that clumps are absent ; this specimen is kept as a control. If
clumps are present they may be removed (in the case of typhoid)
by filtering the culture through filter-paper. A suspension of an
agar culture may also be used, likewise dead cultures : a broth
culture or suspension of an agar one being heated to 65° C. for ten
minutes and preserved in sterilised glass pipettes ; dead cultures
are, however, unsatisfactory in tropical climates. For the macro-
scopic test a thick suspension of an agar culture in salt solution is
to be preferred, the suspension being allowed to sediment for half

1 Brit. Med. Journ., 1902, vol. i, p. 821 (Bibliog.).

THE AGGLUTINATION REACTION 191

to one hour before use. Some strains of an organism are better
than others, and old laboratory strains are generally much more
sensitive to agglutination than recently isolated ones.

Dilution of the serum. — This may be carried out in various ways,
with the haemocytometer pipette, with a pipette with rubber teat
as used for opsonin work (Fig. 35, a, p. 215), or with a platinum
loop. With the pipette a little serum is aspirated up so as to
occupy 1^-2 cm. of the stem, and the upper limit is marked with a
grease pencil or ink. A bubble of air is then admitted so that an
air-space is left between the end of the pipette and the lower end of
the column of serum. The end of the pipette is then immersed in
a watch-glass of salt solution, and the salt solution is aspirated up
to the mark, another bubble of air is admitted, and the process is
repeated again and again ; so that, finally, the pipette contains
1 volume of serum and 4-14 volumes of salt solution, each volume
being separated from the next one by an air-bubble. The contents
of the pipette are then expelled into a watch-glass and thoroughly
mixed, and further dilution of this dilution is performed in the same
manner. Two or three dilutions are usually made — e.g. 1 in 15,
1 in 25, and 1 in 50. A platinum loop may also be employed as a
measure ; a loopful of the serum is deposited in a watch-glass, and
by spotting round it nine or fourteen loops of salt solution a dilution
of 1 in 10 or 1 in 15 is prepared, or any other dilution in a similar
manner.

The microscopic test. — Two or three hanging-drop slides are
vaselined, and two or three cover-glasses cleaned. One loopful of
a dilution of serum is placed on each cover-glass, and to each is
added a loopful of the broth culture of the organism — e.g. typhoid —
and well mixed up, and the specimens are mounted as hanging
drops. Starting with three dilutions of serum — e.g. 1 in 15, 1 in 25,
and 1 in 50 — the dilutions in the specimens will be 1 in 30, 1 in 50,
and 1 in 100 respectively. Should only one dilution of serum have
been made — e.g. 1 in 15 — if on each cover-glass one loopful of this
be placed, and to the first be added one loopful, to the second two
loopfuls, and to the third three loopfuls of typhoid culture, then the
final dilutions in the three specimens will be 1 in 30, 1 in 45, and
1 in 60 respectively.

Care should be taken that the hanging-drop cultures are quite
sealed with the vaseline, so that evaporation is prevented. The
hanging drops are then examined microscopically, a £-in. objective
sufficing for typhoid. In the case of typhoid the following phenomena
will be observed : The motility of the majority of the bacilli is
instantaneously or very quickly arrested, and in a few minutes they

192 A MANUAL OF BACTERIOLOGY

begin to aggregate together into clumps, and by the end of the half
hour there will be very few isolated bacilli visible. In less marked
cases the motility of the bacilli does not cease for some minutes,
while in the least marked ones the motility of the bacilli may never
be completely arrested, but they are always more or less sluggish
as compared with the control hanging drop made from the culture,
while clumping ought to be quite distinct by the end of one hour
(with a 1 in 30 to 1 in 50 dilution).

The central portions of the drop should be examined, not the
margins. With blood which has been dried and dissolved, organisms
may become entangled in debris, and must not be mistaken for
clumps.

In all cases two or three different dilutions should be made to exclude
the possibility of a " zone of no reaction " with some particular dilution
(see p. 188).

Macroscopic, or sedimentation method. — The serum, having been
diluted by means of a pipette with four times its volume of salt
solution, is mixed with five to twenty times its volume of culture
suspension containing plenty of micro-organisms in the same manner
as described in the previous section. The mixture is sucked up into
a fine, but not capillary, bore tube. This is sealed at the lower end
and allowed to stand in the upright position for eight to twenty-four
hours at 20° C., or six hours at 37° C. ; the reaction is often distinct
within an hour at 37° C. When the reaction is posit i ve t he organisms
become agglutinated, and form flocculi, which are easily seen wilh
the naked eye or with a hand-lens and stick to the sides or sink to
the bottom of the tube. The dilution usually employed is 1 in 30
to 1 in 50. Whole blood is not suitable for the sedimentation test ;
clear serum should always be used. It is well to set up at the same
time a control tube with saline solution, or, preferably, with normal
serum.

If sufficient serum is available the mixture may be put up in
little test-tubes, such as the inner tubes of Durham's culture-tubes
(p. 83).

B. For the Recognition of Bacterial Species

1. Bordet-Durham reaction. — This is carried out in much the
same manner as for clinical diagnosis, but an immune serum of
high agglutinating value or high " titre " (at least 1 : 1000) is
required, and the serum from a patient is not applicable. The
immune serum may be obtained from a horse or other animal
immunised with killed cultures (and living also if a high titre is
required). In the laboratory the serum may be prepared by giving

THE MEIOSTAGMIN KEACTION 193

a rabbit three to five intravenous injections at intervals of seven
days of killed culture of a virulent strain of the organism, e.g.
typhoid or cholera. The culture is killed by heating to 60°-65° C.
for half an hour, and the dose is increased from one loop to ten
loops of an agar culture. Seven days after the last dose the animal
is bled from an ear vein, and the serum obtained. The agglutinating
value of the serum must be determined, and controls should always
be put up with normal serum of an animal of the same species as
that from which the immune serum has been obtained. A series of
dilutions of both sera is made with salt solution and a twenty-four
hour agar culture of the organism to be tested used. Both the
macroscopic and microscopic methods should be employed. The
dilutions may be made with a 1 c.c. pipette graduated in hundredths,
with the haemocytometer pipettes, or by the method used clinically.

2. Saturation test. — Castellani noticed that a suspension of a
microbe added to the homologous agglutinating serum absorbs
most, if not all, the specific agglutinin, whereas an organism not
homo]ogous with the serum absorbs little or only a portion of the
agglutinin. The test may be carried out as follows :

Ten loopfuls of a young agar culture of the organism to be tested
are mixed with 10 c.c. of a 5 per cent, solution of a highly aggluti-
nating serum. After incubating for two or three hours, the mix-
ture is centrifuged, the clear supernatant fluid decanted, and
the agglutinating power of the decanted liquid is then tested on the
organism with which the serum was prepared. If the organism
tested is homologous with the organism with which the agglutinat-
ing serum was prepared, the decanted fluid will have lost most,
or a considerable proportion, of its agglutinating power for the
latter.

THE MEIOSTAGMIN REACTION. — Ascoli has found that if an
immune serum be mixed with an alcoholic extract of the homologous
antigen and the mixture incubated at 37° C. for two hours the
surface tension is reduced ; if the serum and antigen extract are
not homologous the surface tension is unaltered. For example, in
the case of typhoid the following is the procedure. An alcoholic
extract of typhoid bacilli is prepared ; this is diluted with saline
solution to 1-1000 — 1-1,000,000. The typhoid serum is similarly
diluted, 1-10. To 9 c.c. of the diluted serum 1 c.c. of the diluted
antigen extract is added. By means of some form of viscosimeter
or stalagmometer the number of drops yielded by a given volume
of the mixture is ascertained, immediately after the mixture is
made and after the mixture has been incubated at 37° C. for two
hours. If the surface tension has been reduced, the number of

13

194 A MANUAL OF BACTERIOLOGY

drops counted in the second determination will be greater than in
the first.1

ANTI-FERMENTS. 2 — By the injection of rennin or other enzyme
the blood-serum of the treated animal acquires the property of
neutralising the action of the enzyme with which the inoculation
has been performed. Thus if rennin and anti -rennin (the serum of
an animal injected with rennin) be mixed with milk no curdling
takes place. Similarly, the serum of an animal inoculated with
pancreatin inhibits the action of this ferment, and if coagulated
egg-albumen, pancreatin, and anti -pancreatin be mixed, the egg-
albumen undergoes no digestion.

PnECiPiTiNS.3 — Kraus was the first to demonstrate the presence
of specific precipitins in blood by adding typhoid, cholera, and plague
anti-sera to filtrates of the cultures of the corresponding microbes.
If to such a filtrate in a test-tube a little of the corresponding
anti -serum be added by running in carefully, so that it forms a layer
at the bottom, an opalescent ring makes its appearance at the line
of junction of the two fluids. So also if an animal be injected with
milk, its serum, when added to milk of the same kind as that with
which it has been injected, causes precipitation of the casein. This
reaction is specific, and it is thus possible to distinguish various
milks from one another. Similarly, anti-sera which produce pre-
cipitates, each with the homologous substance, are obtained by
the injection of peptone, of egg-albumen, blood-serum, and other
proteins. The latter reaction has an important medico-legal
application, for by means of it the blood and flesh of different
species of animals can be distinguished. Thus the presence of
horseflesh in sausages can be detected. The method employed is
to inject a rabbit intraperitoneally with four to six injections of
defibrinated blood or of blood-serum (or with a solution of the
particular substance, e.g. horseflesh), commencing with about 5 c.c.
and increasing to 10 c.c. at intervals of a few days. After treat-
ment the animal is bled from an ear vein, and the serum is obtained.
The blood to be tested may be dried on filter-paper, pieces are then
cut up, a solution is made in 1-6 per cent, sodium chloride solution,
and to this the specific serum is added. Tested in this way human
blood anti-serum reacts — i.e. forms a precipitate — markedly with

1 Ascoli and Izar, Munch, med. Woch., Ivii, 1910, pp. 62, 182, 403.

2 See Dean, Trans. Path. Soc. Lond., vol. lii, 1901, Part 2, p. 127.

3 See Nuttall, Journ. of Hyg., vol. i, 1901, p. 367 (Bibliog.), also Brit.
Med. Journ., 1902, vol. i, p. 825 ; Welsh and Chapman, Journ. of
Hygiene, vol. x, 1910, p. 177 ; ibid. Australasian Med, Gazette, December
12. 1908 (hydatid disease).

IMMUNITY 195

human blood, less so with ape's blood, not at all with other blood ;
ox blood anti-serum reacts with ox blood, less so with sheep, feebly
with horse, hardly at all with dog. Mixtures of bloods may also
be tested. Precipitins are also formed naturally in vivo. Thus
the serum of a patient the subject of hydatid disease gives a precipi-
tate with hydatid fluid, and the reaction may be used diagnostically.
The production of the anti-body seems to be due to the globulin
constituent of the injected serum.

It will thus be seen that the anti-bodies which result
from the injection into an animal of different substances
are extremely numerous and have varied properties, their
most notable characteristics being their extreme specificity
and the extraordinary delicacy of the interactions produced
by them. It is important to note that these anti-bodies
are produced only as the result of inoculation with complex
compounds allied to the proteins. The tolerance estab-
lished by the ingestion or inoculation of simpler com-
pounds, such as arsenious acid and morphine, is of a different
nature, and is not coincident with the development of
anti-bodies. According to Ehrlich, the latter kind of
tolerance may be due to the exhaustion or using up of
certain receptors (" chemo-receptors ") of the protoplasm
(see p. 206).

Immunity *

No fact in biology is more striking than the differences
in susceptibility to infection exhibited by different races
and different animals. For example, the natives in many
parts of the world are comparatively insusceptible to yellow
and typhoid fevers and malaria, the dog and goat are rarely
affected with tuberculosis, and tetanus is never met with
in the fowl ; and to come nearer home, while some indi-
viduals are lucky enough to escape most of the commoner

1 See Metchnikoff, Immunity in Infective Diseases, 1905. Also Brit.
Med. Journ., 1902, vol. i, p. 784 ; 1904, vol. ii, pp. 557-582 ; and 1907,
vol. ii, pp. 1409-1425 ; Journ. of Hygiene, vol. ii, 1902 ; Emery, Im-
munity and Specific Therapy, 1909.

196 A MANUAL OF BACTERIOLOGY

infectious fevers, others seem to contract them on every
possible occasion, and to suffer from all the ills that flesh
is heir to. These instances show that there is often a
natural insusceptibility to infective disease, or a natural
immunity, as it is termed. This may be complete or
partial, or it may appertain only to a race — " racial
immunity " ; or, varying in different individuals and at
different ages, it constitutes " individual immunity," as
in the case of diphtheria and scarlatina, which become
more and more rare as age advances.

Still more striking, perhaps, is the fact that an insus-
ceptibility may be acquired after an attack of infective
disease or be conferred in certain instances by inoculation.
Thus second attacks of smallpox and scarlatina are rare,
inoculated smallpox and vaccinia protect against variola,
and bacterial vaccines confer considerable protection.

With regard to the immunity of native races to certain
diseases, this is probably due to natural selection and
heredity ; during long periods of time, the individuals
being all exposed to the same risks, the susceptible ones
are weeded out, while the survivors transmit their insus-
ceptibility to their descendants ; but this, of course, does
not explain the reason for the relatively greater immunity
of the insusceptible individuals. Immunity is generally
not absolute either to infection or intoxication ; that is,
infection can usually be induced under certain conditions.
Thus fowls, which are highly refractory to tetanus and
tolerate considerable doses of tetanus toxin with impunity,
can be tetanised with large doses of an active toxin ; white
rats, which are insusceptible to anthrax, become susceptible
after fatigue, or when fed on an exclusively vegetable diet.
Immunity is therefore either (1) natural, or (2) acquired,
and it is evinced against either (a) toxins, or (6) micro-
organisms, and these different phases must be con-
sidered.

IMMUNITY 197

1. Natural immunity against toxins. — There are various
non-specific reactions in the body by which toxins may
be eliminated or destroyed. Thus the dilatation of the
vessels and the acceleration of the blood-stream which
take place in an inflamed area dilute and eliminate the
toxin, and the proteolytic enzymes produced by the
organisms and as a result of tissue disintegration may
have a destructive action on the toxins. Oxidation,
hydration and dehydration, and various analytic and
synthetic processes which go on in the body, and particu-
larly in the liver, are other agencies whereby toxins may be
destroyed. These non-specific processes by which toxin
is destroyed or eliminated, though of the greatest impor-
tance, can probably deal with only small amounts of toxin ;
if large amounts are present, specific reactions have to be
evoked.

Another cause of natural immunity to toxins may be
the absence of suitable receptors for the toxin. As already
stated (p. 153), in order that a bacterial toxin or endotoxin
may produce intoxication, it must become anchored to
the cells by its haptophore group, and that this may occur
the cell molecules must possess atomic groups or side-
chains (" receptor groups ") which have a special affinity
for the haptophore groups of the toxin. Should these be
wanting the toxin cannot become anchored to the cells,
its toxophore groups cannot exert their influence, and
natural immunity is the result.

This has been proved to be the case in several instances.
Thus in the lizard and turtle, if tetanus toxin be injected
no effect is produced, but the toxin is not eliminated and
remains in the body for months, as may be proved by
withdrawing a little of the blood and injecting it into a
mouse ; the animal dies of tetanus.

In other instances, for some reason or other, the cells
of the animal are insusceptible to the toxophore group of

198 A MANUAL OF BACTERIOLOGY

the toxin. Thus, if an alligator be injected with tetanus
toxin, no effect is produced, but the toxin rapidly disappears
from the blood. If the animal be kept at ordinary tem-
perature (20° C.), although the toxin disappears, antitoxin
is not formed, but if it is kept at 30°-37° C. antitoxin is
rapidly produced. The two experiments together suggest
that the toxin is fixed by the cells, but has no effect upon
them ; if the. toxin were not fixed, it would be possible to
detect it, and presumably it would not produce antitoxin.
2. Natural immunity against micro-organisms. — A number
of factors are doubtless concerned in preserving the body
from invasion by micro-organisms, and while non-specific
reactions may suffice when the number of organisms is
small, specific reactions have to be evoked if the number
of organisms is large. The unbroken surfaces of the
body have a considerable protective action in preventing
the entrance of micro-organisms. The flushing-out action
of accelerated circulation will exert some action in elimina-
ting organisms from a localised focus of infection just as
it does with toxins. The body temperature may be of
some importance, and the febrile condition so generally
induced by infection is probably to some extent protective
and curative. Thus frogs, fish, and chickens are naturally
immune to anthrax. In the one case the body tempera-
ture is low, 18° C. or thereabouts ; in the other it is high,
40° to 41° C., and this may influence the growth of the
anthrax bacillus, preventing the full and rapid development
which may be necessary for the production of the disease.
That such is the case would seem to be shown by experi-
ments in which when the temperature of the medium is
raised or lowered, infection takes place ; frogs and fish
kept in water raised to a temperature of 35° C., and chicken
refrigerated so as to reduce their temperature, all perish
from anthrax after inoculation. It is clear, however, that
this is not necessarily the only factor, for sparrows, which

IMMUNITY 199

have a temperature as high as that of the chicken, can
be infected with anthrax without refrigerating. Behring
would ascribe the immunity of white rats to anthrax to
the high alkalinity of their blood, and claims to have shown
experimentally that a vegetable diet reduces this, and
fatigue is said to act similarly.

In some cases the animal, after invasion by the organism,
becomes gradually tolerant to its presence (immunitas non
sterilisans). This is particularly the case in protozoan
infections, e.g. piroplasmosis. The animal, after a period
of ill-health, gradually recovers, though the organisms
may still be present, as can be demonstrated by injecting
some of its blood into a susceptible animal. Conceivably
the receptors necessary for the intoxication become
gradually used up, and when this state is attained the
animal becomes insusceptible.

The blood, lymph, and other fluid and tissue juices
undoubtedly exert a more or less germicidal action on
bacteria experimentally in vitro, and to some extent
probably also in the body. But in this respect there is
often a marked difference between the circulating blood
and the blood in vitro.

Lewis and Cunningham (1872), Traube and Gscheidlen
(1874), Fodor (1877), and Wysokowicz showed that bac-
teria injected into the circulation rapidly disappear, and
were inclined to attribute this result to the bactericidal
properties of the blood. In the main, however, this dis-
appearance is due to lodgment in the capillaries, phago-
cytosis, and excretion by the excretory glands.

Halliburton prepared from the lymphatic glands a
protein, cell-globulin /3 (really a nucleo-protein). Hankin
found that this substance had marked germicidal properties,
and concluded that it was probably the germicidal con-
stituent of the blood-serum. Bitter, who repeated Hankin's
experiments, failed, however, to confirm them. To the

200 A MANUAL OF BACTEKIOLOGY

germicidal constituents of the cells and body fluids Buchner
gave the name " alexins."

Grohmann performed the first experiments with extra-
vascular blood. He found that anthrax bacilli, after being
kept in plasma, became less virulent. Fodor, adding
anthrax bacilli to blood and plating at intervals, found
there was a progressive diminution in the number of
organisms.

Nuttall, in 1888, used the defibrinated blood of several
animals, rabbits, mice, pigeons, sheep, and found that it
destroyed the B. anthracis, B. subtilis, B. megaterium, and
M. pyogenes var. aureus. He confirmed Fodor's results,
which also showed that after a while the blood loses its
germicidal properties and becomes a suitable culture
medium. The blood or serum similarly loses its bactericidal
properties on heating, and serum that has once been used
loses its bactericidal properties. Nissen continued this
work, and also found that fresh serum is germicidal for a
variety of organisms.

In 1890, Buchner with Voit, Sittmann, and Orthen-
berger came to the conclusion that the germicidal action
of cell-free serum is due to the protein constituents.

Christmas prepared a germicidal substance from the
spleen, and Bitter, who examined the method, in the main
confirmed Christmas.

Behring and Nissen, however, found that the serum
of the white rat, dog, and rabbit destroys the Bacillus
anthracis, but serum from the mouse, sheep, guinea-pig,
chicken, pigeon, and frog has no action. Thus, while the
rabbit is highly susceptible to anthrax, its serum is germi-
cidal ; the chicken, on the other hand, is immune to
anthrax, but its serum is inactive. Hence there is a
considerable difference between the action of circulating
and of extra-vascular blood.

Vaughan, Novy and McClintock, in a series of papers,

IMMUNITY 201

ascribed powerful bactericidal properties to the nucleins,
and surmised that in serum the nucleins set free by the
disintegration of leucocytes and other cells are the germi-
cidal agents. Forrest and the writer 1 found, however,
that all the germicidal properties ascribed by Vaughan
to the nucleins are probably due to the weak alkali in
which the nucleins were dissolved, and came to the con-
clusion that Vaughan's results are at least not proven.

Gengou also found that the plasma collected in vaselined
tubes is often almost devoid of bactericidal power, whilst
the corresponding serum may be capable of destroying
large numbers of micro-organisms.

We therefore see that while the blood, lymph, and
other fluids and tissue juices undoubtedly exert more or
less germicidal action on bacteria experimentally in vitro,
there is often a marked difference in this respect between
the circulating blood and the blood in vitro and it may be
doubted if this factor is of great importance in the produc-
tion of natural immunity. At the same time, it is to be
noted that directly infection has started more or less cel-
lular disintegration and serous exudation occur, and thus
the germicidal action of the body fluids and tissues may
be exerted in vivo, though such substances may act rather
by stimulating the leucocytes or by rendering the bacteria
more phagocytosable, as will be referred to later (p. 209).
Thus Kanthack and Hardy found that the coarsely
granular oxyphile leucocytes in the frog are first attracted
to the site of a bacterial invasion, there discharge their
oxyphile granules, the bacteria then show signs of degenera-
tion, and polymorphonuclear leucocytes and other " phago-
cytic " cells now approach and ingest the degenerate
bacteria. The observations, however, do not seem to have
been confirmed. Wooldridge also protected animals from
anthrax by injections of " tissue fibrinogen " (nucleo-

1 Journ. Roy. Army Hed. Corps.

202 A MANUAL OF BACTERIOLOGY

protein). For some micro-organisms a bacteriolytic
mechanism exists, the amboceptor- complement complex,
whereby they may be digested and got rid of. Thus
normal serum has a marked bacteriolytic action on B.
typhosus and B. coli. In many cases, however, e.g. for
staphylococci, such a bacteriolytic mechanism does not
naturally exist, but may be evoked as a result of infection.

The hypothesis which ascribes immunity to the germi-
cidal and bacteriolytic action of substances in the fluids
of the body has been termed the " humoral theory."

Another important theory of immunity is the doctrine
of phagocytosis, so ably supported by MetchnikofL This
is the " cellular " theory of immunity. It has as its basis
the following fundamental facts : Firstly, the leucocytes
in the circulating blood ingest and destroy any foreign
particles present therein ; secondly, an injury to the
tissues is immediately followed by an inflammatory reac-
tion, in which the leucocytes emigrate from the vessels
and congregate at the injured spot. Similarly, in many
instances the leucocytes rapidly congregate at the seat of
a bacterial infection, and approach and engulf the bacteria
in the same manner as they do other foreign particles, and
so rid the body of the unwelcome guests (Plate I., a and b).

The migration of the leucocytes towards the scene of
action is explained by MetcrmikofT on the hypothesis that
the chemical substances elaborated by the bacteria attract
the latter and exert what he termed " positive chemo-
taxis." In this case the bacteria are removed by the
leucocytes, and general infection and death do not occur.
But, unfortunately, in other cases the bacterial chemical
products repel, or perhaps it is more correct to say do not
attract, the leucocytes, and " negative chemotaxis " occurs,
so that the bacteria are free to grow and multiply, and
general infection ensues. Positive and negative chemo-
taxis can be shown to occur by a simple experiment. If

IMMUNITY 203

a fine capillary tube containing some peptone solution be
introduced into a suspension of bacilli, e.g. B. fluorescens
liquefaciens, under a cover- glass, and watched microscopi-
cally, the bacilli will be attracted to the tube and soon
invade its lumen. If, however, a weak acid be substituted
for the peptone water, the bacilli will be repelled. The
process by which the bacteria are ingested by the leucocytes
can be similarly watched. The leucocytes which act in
this manner are termed phagocytes, and they are of two
classes — the macrophages, the large mononuclear leuco-
cytes, and the smaller microphages, or polymorphonuclear
leucocytes. Certain of the tissue cells and endothelial
cells also possess phagocytic properties. The importance
of phagocytosis is also shown by the fact that, while in
ordinary susceptible rabbits infection with anthrax is
followed by a feeble phagocytosis and the animals succumb,
in rabbits vaccinated against anthrax phagocytosis is very
active. Moreover, in an animal refractory to anthrax,
such as the frog, anthrax bacilli grow and multiply if they
be enclosed in paper or collodion sacs, so as to prevent the
access of the phagocytes.

Phagocytosis, in vitro, and probably also in the normal
body, is extraordinarily active, so that it might be expected
always to be sufficient to deal with any number of bacteria
that might be introduced. If, however, the bacteria be
virulent, negative chemotaxis will occur. Moreover, the
presence of substances which render the bacteria phago-
cytosable, " opsonins," is necessary, and it seems likely
that the amount of opsonin becomes diminished in infection
(see p. 211).

Metchnikoff admits that the destruction of bacteria in
phagocytosis is brought about by chemical bacteriolytic
substances, which he terms " cytases," and which he
regards as being derived from the leucocytes, and as
identical with the alexins. He believes that there are two

204 A MANUAL OF BACTEEIOLOGY

kinds of cytases, one " macrocytase," obtainable from
tissues, such as the spleen and lymph- glands, rich in
macrophages, which acts specially on elements of animal
origin, the other " microcytase," derived from the micro-
phages, and which acts principally on micro-organisms.
He considers the alexic action to be of the nature of a
digestive process (but this is doubtful), and as regards the
complex nature of a cytolytic serum, which contains ambo-
ceptor and complement, believes that the amboceptor is
formed within the macrophages in intra-cellular digestion,
and that a portion of it escapes from them into the serum.
All the facts point to the leucocytes and leucocytic tissues
being the great defensive mechanisms against parasitic
invasion, either by the production of alexins, or of bacterio-
lysins, or by phagocytosis, or probably by a combination
of these (the " cellulo-humoral " hypothesis of immunity).
It is probable that the greater part of phagocytosis takes
place in the spleen. This organ acts as a sort of filter,
and phagocytosis may be active in it when none can be
discerned in the blood. Phagocytosis is also active in the
bone-marrow.

Experiments by Tizzoni and Cattani seemed to show
that rabbits could not be rendered refractory to tetanus
by injection of tetanus antitoxin after extirpation of the
spleen ; and although Benario and other observers have
not confirmed this, the manner in which the spleen is
attacked in such diseases as tuberculosis, plague, etc.,
points to this conclusion. The discordant results obtained
after splenectomy may be due to the rapid regeneration
of spleen tissue, and to other structures, such as the
hsemolymph glands, taking on its functions after ablation.

Although small amounts of antitoxin may occasionally
be met with in the normal animal (e.g. diphtheria anti-
toxin in man and in the horse, see pp. 153 and 274), this
substance plays little or no part in natural immunity

IMMUNITY 205

against either toxin or micro-organism. Thus the blood-
serum of the fowl, which is highly refractory to tetanus
does not exert the slightest antitoxic or neutralising action
on tetanus toxin.

3. Acquired immunity. — Acquired immunity may be
induced in several ways :

(1) By an attack of the disease ending in recovery.

(2) By vaccinating with a modified and less virulent
form of the living infective agent (Pasteur's method).

(3) By treatment with sterilised cultures, or with
bacteria-free toxins.

(4) Occasionally by treatment with sterilised cultures
or toxins of a different species. Thus, B. pyocyaneus
protects from anthrax (p. 238), and Klein1 showed that
an injection of one of the six following organisms —
(1) Koch's comma, (2) Finkler-Prior's comma, (3) B. coli,
(4) Proteus vulgaris, (5) B. prodigiosus, (6) B. typhosus —
will protect an animal against any one of the remaining
five. He therefore concluded that there is an immunising
agent common to all these six organisms, and that this
substance is intra- cellular and a constituent of the bacterial
cells themselves. In this case, however, the immunity is
probably one against certain bacterial proteins and not
against the specific endotoxins of the organisms.

(5) By injection of the blood-serum derived from an
animal treated or immunised by method (3) — that is to say,
antitoxins or other anti-bodies (e.g. amboceptors) are
introduced.

The immunity acquired by methods (l)-(4) is known as
" active immunity," because the animal's cells and tissues
are altered by the process, so that they are no longer
susceptible to the microbe or its toxin. The immunity
conveyed by method (5) — the injection of an immune
serum, is known as " passive immunity," because the

1 Trans. Path. Soc. Lond., 1893, p. 220.

206 A MANUAL OF BACTERIOLOGY

immunity lasts only so long as the anti-bodies remain ;
there is no active participation of the animal's cells and
tissues in the process. Active immunity is generally of
long duration — some months at least — and is not trans-
missible to the fetus ; but passive immunity is of short
duration — two to four weeks — and is transmissible to the
fetus and nursling. Acquired immunity to toxins may be
due to the elimination of the receptors concerned in the
fixation of the toxin by the cells, or to the production of
the neutralising antitoxin. The leucocytes are probably
the active agents in destroying and eliminating toxin,
whether neutralised by antitoxin or not.

Various explanations have been given of the production
of acquired immunity against the organisms. Pasteur
suggested that the organism, by its growth in the body,
exhausts some specific pabulum necessary for its develop-
ment, so that it cannot again grow in the animal which
has been attacked. This hypothesis, therefore, pre-
supposes that in the body there is some nutrient material
necessary for the growth of each species, which is difficult
to believe, and is negatived by the fact that an organism
will grow in the blood and tissues removed from an animal
vaccinated against, and insusceptible to, the disease
produced by itself.

Pasteur's '" exhaustion " theory has been revived by Ehrlich 1
in a modified form, under the name of " atrepsy," to explain certain
cases of immunity. Thus, for a chemical poison to act, Ehrlich
assumes that particular receptors in the protoplasm for binding
the poison are necessary ; these he terms " chemo -receptors."
Bird-pox, virulent for both fowl and pigeon, if passed through the
pigeon becomes completely avirulent for the fowl. To explain this
Ehrlich suggests that the parasite in passing through the pigeon
has to assimilate substances different from those assimilated during
its passage through the fowl ; therefore that part of the receptors
which deals with the nutritive substances of the fowl's organism is

1 " Harben Lecture," ii, Journ. Roy. Inst. Public Health, 1907.

IMMUNITY 207

not in use during the passage through the pigeon, and may become
atrophied, so that on the parasite being transferred back to the
fowl it will not be able to thrive owing to the loss of the receptors
necessary to assimilate the fowl's nutritive substances. Ehrlich
suggests that the majority of non -pathogenic micro-organisms, if
introduced into the animal body, perish by this mechanism. In
the case of mouse carcinoma inoculated into rats, the tumour-cells
proliferate for a few days, then atrophy and disappear. Ehrlich
suggests that some specific substance is necessary for the prolifera-
tion of mouse carcinoma-cells which is not present in the rat, and
as soon as the traces of this specific substance carried over by the
inoculation are used up, the cancer-cells cease to proliferate and
finally atrophy and disappear. These are examples of Ehrlich's
" atrepsy " and " atreptic immunity."

Chauveau, in his retention theory, suggested that the
bacteria during their growth in the tissues form substances
which ultimately inhibit their growth, and, if the animal
recovers, prevent a subsequent development of the organ-
ism. The same objections may be urged against this
hypothesis as against Pasteur's exhaustion hypothesis.

Bacteriolysis and phagocytosis are probably the two
main factors which bring about the refractory condition
in acquired immunity against bacteria, as well as recovery
from an infection. After immunisation it may be shown
that phagocytosis is increased, and that positive chemotaxis
takes place towards the organism, whereas previously
negative chemotaxis occurred ; the leucocytes have been
" educated," as it were, to be attracted, instead of repelled,
by the bacterial invasion. According to Andrewes,1 the
defence against the pyogenic cocci is not only essentially
phagocytic, and dependent upon the polynuclear leuco-
cytes, but is also, in the main, opsonic. In tuberculosis
and syphilis the polynuclear leucocyte takes little part in
bodily defence, which is essentially a function of the endo-
thelial and fixed tissue- cells. With the colon group of
organisms certain humoral responses, notably agglutination

1 " Croonian Lectures," Lancet, June 25 et seq., 1910.

208 A MANUAL OF BACTERIOLOGY

and bacteriolysis, are better marked than with most other
bacteria, and polynuclear phagocytosis seems subsidiary.

Antitoxin formation probably plays little or no part
in acquired immunity, or even in recovery from infection.
In diphtheria, for instance, antitoxin is not found until
the disease has subsided. Possibly, in chronic infections,
antitoxin formation does play a subsidiary role in recovery.

To sum up, natural immunity is probably due to a
number of factors, some or all of which may be operative
in particular instances, and it is impossible to state with
certainty any general law. In most cases phagocytosis
is the principal means of defence, the germicidal, inhibi-
tory, or bacteriolytic actions of the body-fluids aiding,
though of subsidiary importance ; in others the cells and
tissues are unaffected by the bacterial toxins, sometimes
because the cells are lacking in the particular side- chains
or receptors which fix the toxin ; sometimes because, for
some unknown reason, the cells are unaffected by the
toxophore group of the toxin.

As regards the immunity acquired after an attack of
disease, this may be due to the " education " of the leuco-
cytes, whereby they are attracted, whereas formerly
repelled, by the products of bacterial development, or to
substances which stimulate the action of the leucocytes.
The germicidal, inhibitory, and bacteriolytic actions of the
body-fluids may also be enhanced. It seems probable
also in certain instances that the side- chains or receptors
having an affinity for the toxin become in some way
destroyed or used up, so that further fixation of the
particular toxin cannot take place.

It is to be noted, as Metchnikofl: has pointed out, that
immunity is much more rapidly acquired against micro-
organisms than against their toxins. In Nature, it is
principally against micro-organisms that the body requires
protection.

PHAGOCYTOSIS 209

Adaptability seems to be one of the innate properties
of protoplasm, and immunity is but an instance of adapta-
bility. It might be expected, therefore, that immunity
towards infection will become established, more or less
completely, when the need for it arises ; and we find that
this is the case, however difficult it may be to explain the
mechanism by which it is attained.

The Role of the Serum in Phagocytosis

The fact that in an immunised animal, no sooner does
the virulent organism gain access than the leucocytes
migrate to the site of infection, surround the invaders,
ingest and so destroy them, was at one time ascribed by
Metchnikoff to " education," i.e. modification, of the
leucocytes ; but since the serum of the immunised animal
injected into a non- immunised one causes the leucocytes in
the latter to behave in the same manner as they do in the
immunised animal, the effect must be due to something
in the plasma or serum, and Metchnikoff ascribed the
action to substances, " stimulins," which heighten the
activity of the leucocytes. Later work has not confirmed
this view, and no certain proof of the existence of stimulins
is forthcoming, although Leishman attributed a stimulin
action to thermostable substances in the serum in typhoid
and Malta fevers. Subsequently Metchnikoff conceived
the serum as acting, not on the leucocytes, but on the
microbe, causing it to become positively chemotactic and
no longer to repel, but to attract the phagocytes. Con-
siderable support was given to this view by the work of
Wright and Douglas, who, by a modification of Leishman's
ingenious method for quantitatively estimating phago-
cytosis, emphasised the importance of the serum in the
mechanism of phagocytosis.

Neufeld and Eimpau also concluded that substances,

210 A MANUAL OF BACTERIOLOGY

" bacteriotropines," are produced in the course of immu-
nisation which promote the phagocytosis of bacteria.

Leishman's method for estimating phagocytosis* — A thin suspen-
sion of some micro-organism, e.g. M. pyogenes, is mixed with an
equal volume of blood from the finger ; a droplet of this mixture
is placed on a clean slide, and covered with a cover-glass, and the
preparation is at once placed in a moist chamber in the incubator
at 37° C. for half an hour. At the end of this time it is taken out,
the cover-glass slipped off, and the films on slide and cover-glass
are driea, fixed, stained, and examined microscopically, and the
number of microbes ingested by the polymorphonuclear leucocytes
is counted.

Wright and Douglas 2 found that washed leucocytes
without serum are non-phagocytic, but become so on the
addition of normal serum. If, however, the serum be
first heated to 60°-65° C. before being added to the mixture
of leucocytes and microbes, phagocytosis does not take
place ; but if the unheated serum is mixed with the bac-
teria, the mixture kept at 37° C. for fifteen minutes and
then heated to 60° C. for fifteen minutes, phagocytosis can
still take place, thus demonstrating that the serum acts
in some way on the bacteria, rendering them suitable prey
for the phagocytes. This thermolabile serum feast pre-
parer is called by Wright and Douglas " opsonin " (from
a Greek word meaning " to cater for ").

They have also shown that during the process of active
immunisation the opsonic value of the serum is increased,
and they have succeeded in demonstrating this opsonic
immunity for a number of infections, such as the staphy-
lococcic, Malta fever, pneumococcic, and tuberculous. If
it be desired to measure the quantity of opsonins present,

1 Brit. Med. Jcurn., 1902, vol. i, p. 73.

2 Prcc. Roy. See. Lond., B. Ixxii, 1903, p. 357 ; B. Ixxiii, 1904, p. 128 ;
B. Ixxiv, 1905, pp. 147, 159 ; B. Ixxvii, 1907, p. 211. Also in Practitioner,
May 1908 ; various papers in Lancet and Brit. Med. Journ. ; Wright;
fitudies in Immunity, 1909.

OPSONINS 211

say in a case of furunculosis, which is almost always caused
by the M. pyogenes, the following are required : (1) a
drop or so of the patient's serum ; (2) a drop of serum
from a normal person ; (3) a suspension in salt solution,
of a culture of M. pyogenes preferably derived from the
furuncle ; (4) leucocytes washed free from the plasma.
Equal volumes of the patient's serum, leucocytes, and
suspension are mixed, draw^n up in a capillary tube, incu-
bated for fifteen minutes at 37° C., and films are then
prepared and stained. As a control a similar mixture is
prepared and treated in the same way, but using the normal
serum instead of that of the patient. The films are then
examined, and the number of cocci taken up by, say,
fifty leucocytes is counted in the two specimens, and a
ratio obtained. Taking the figure for the normal serum
as 1, that for the patient's serum will probably be 0-5 or
0-6, and this is termed the " opsonic index " (see below,
p. 219).

In subacute and chronic local infections the opsonic
value of the serum is usually diminished, occasionally
increased. In acute infections the index will, as a rule,
below ; in chronic infections which are not strictly localised,
e.g. tuberculosis, the index will sometimes be low, some-
times high. A low index generally indicates an infection,
or a low power of resistance to the particular organism,
or that a chronic but quiescent infection exists ; a high
index may indicate that the person has had an infection
but has overcome it, or has a quiescent infection. The
normal index for healthy persons varies only within
narrow limits, from about 0-8 to 1-2 as extremes ; an index
above or below these values is therefore probably patho-
logical.

By injecting small quantities of a vaccine consisting of
a killed culture, tuberculin, etc., the opsonic index can
be raised, and the infection thereby tends to be cured.

212 A MANUAL OF BACTERIOLOGY

The first effect of the injection is to cause a fall in the
opsonic index, the " negative phase " of Wright, which
is usually afterwards followed by a rise, and by properly
spacing the injections a considerable rise in the opsonic
value may ultimately result. If too much vaccine be
given the effect may be to permanently depress the index
and cause harm instead of good, hence the desirability of
controlling all injections by determinations of the opsonic
index. This, however, renders the treatment very labo-
rious, and generally by employing small doses and allowing
at least a week to elapse between the doses, determina-
tions of the opsonic index are unnecessary (for dosage, etc.,
see p. 221). By movement, massage, etc., applied at or
about the seat of a local infection, bacterial products are
disseminated which may alter the index ; a process of
auto-inoculation may thus result.

The opsonic index may be used for diagnostic purposes ;
a low or high opsonic value towards a particular organism
suggests that an infection by this organism exists or has
recently existed.

Bulloch came to the conclusion that the blood contains
a number of specific opsonins, one for tubercle, another for
M. pyogenes, and so on. Simon, Lamar, and Bispham,1
however, from a number of carefully devised experiments,
conclude that specificity of opsonins does not exist, and
suggest that opsonins may be a constant quantity, and
that the number of organisms taken up by the leuco-
cytes is influenced by a second unknown and variable
factor.

Russell 2 also concludes that in normal serum the opsonins
are " common " and not specific, \nnd can be removed by
a number of bodies. In immune serum, on the other
hand, both "common" and "immune" opsonins are

1 Journ. Exper. Med., vol. viii, 1906, p. 651.

2 Johns Hopkins Hosp. Bull., vol. xviii, 1907, p. 252.

THE OPSONIC METHOD 213

present, the latter being quite specific. That is to say, in
the process of immunisation specific opsonins are formed
and the increase of opsonins following injection of a vaccine
is probably due to the formation of immune opsonins which
react specifically.

Muir and Martin 1 believe that in immune serum a specific,
immune, thermostable opsonin is present, and also a
normal, thermolabile opsonin.

Wright considers the opsonins to be substances distinct
from all others, but MetchnikofT, Dean, and other observers
suggest that they are identical with the " substance
sensibilisatrice."

It is doubtful if opsonins are present in more than
traces in the unaltered blood plasma : like alexins, they
seem to develop as a result of coagulation. The role of
opsonins in immunity and in recovery from infection is
therefore a complex problem.

The opsonic method has been criticised of late. Thus Moss2
says : " None of the present methods of estimating the opsonic
content of the blood seems sufficiently accurate to be of practical
value " ; Fitzgerald, Whiteman, and Strangeways,3 in an elaborate
investigation, concluded that the method is unreliable. Whereas
Wright takes into account the serum only, Shattock and Dudgeon 4
state that " the cells (i.e. the phagocytes) vary in value like the
serum." It may be granted that the whole truth respecting the
opsonic reaction and method is not yet fully known, but many of
the criticisms have been based on an imperfect technique. On the
whole, it may be said that Wright's method, with careful technique
and in practised hands, gives information previously impossible to
obtain, and the proper dosage of, and treatment by, vaccines has
been largely elaborated by means of it.

1 Proc. Roy. Soc. Lond., B. Ixxix, 1903, p. 187.

2 Johns Hopkins Hosp. Bull., vol. xviii, 1907, p. 237.

3 Bull. Committee, for the Study of Special Diseases (Cambridge), vol. i,
1907, No. 8.

4 Proc. Roy. Soc. Med., vol. i, 1908, " Medical Section," p. 169.

214 A MANUAL OF BACTERIOLOGY

Method of Determining the Opsonic Index

The requisites are :

1. Several Wright's pipettes with india-rubber teats.

2. The serum of the patient to be tested.

3. The serum of a healthy person for a control.

4. A suspension of the organism for which the deter-
mination is to be made.

5. A suspension of living leucocytes.

1. Wright's pipettes with india-rubber teats. — These are
made of glass tubing of the form shown in a, Fig. 35,
which is about two- thirds full size. Glass tubing must be
chosen which properly fits the teats. A piece of glass-
tubing about 4 inches in length is taken, heated in the
blowpipe flame until quite soft, then it is taken out of
the flame and the two ends are drawn steadily apart ;
the more they are drawn apart, the finer will be the bore
of the tube — about ^ in. is a suitable size. The middle
of the capillary part should then be introduced into a
small white gas- flame and drawn apart so as to form
two pipettes. By filing off the sealed end at a suitable
spot the open extremity may be slightly contracted as
shown in b ; this prevents the column of fluid in the tube
moving so quickly.

2 and 3. The sera. — These two specimens should be
taken at about the same time, and the determination
should be made as soon as possible.

The blood is preferably collected in a Wright's capsule
(Fig. 35, d). Both ends of the pipette are broken off,
and the blood is collected by immersing the bent end in
the blcod as it luns from a prick with a Hagedorn or
triangular needle in the ear or finger. The capsule, which
should be at least one-third filled, is then sealed in the
flame, the dry or straight end being sealed first. After

THE OPSONIC INDEX

215

coagulation, which may be hastened by placing in the
warm incubator for half an hour, the capsule is hung Ly
the curved end in the centrifuge and centrifuged to obtain
clear serum. Little change in the serum ensues for two
to three days if the capsules are kept sealed. The capsules
may be stuck into a lump of plasticine until required.

FIG. 35. — a. Glass pipette, with india-rubber teat for opsonic
determinations, etc. ; 6 shows (enlarged) the contracted
extremity of the pipette ; c shows the stem of the pipette,
containing the equal volumes of serum, leucocytic suspension,
and bacterial suspension, before mixing ; d is the Wright's
capsule for collecting blood.

Plasticine is useful for many such purposes, for temporarily
plugging tubes, etc.

4. Suspension of the organism. — In the case of tubercle,
suitable dead cultures can be purchased. To prepare
the suspension from these, a small portion of the growth
(about as big as a grain of rice) is ground up in a small
agate mortar, 1-5 per cent, salt solution being added drop
by drop up to 2 c.c. This suspension will still contain
clumps, which must be got rid of by centrifuging for three
or four minutes. With the tubercle bacillus and gono-
coccus spontaneous phagocytosis is apt to occur if ordinary
(0-8 per cent.) salt solution is used.

A staphylococcic suspension is prepared by taking an
agar culture not more than twenty- four hours old, adding
salt solution (0-8 per cent.), and shaking gently so as to
wash off the growth. When the suspension is made it

216 A MANUAL OF BACTERIOLOGY

must be pipetted off into a small tube and centrifuged for
a few minutes. The suspension must not be too thick,
otherwise the leucocytes will take up an unaccountable
number of cocci ; the proper density can be judged by
experience alone, but the suspension should be only faintly
opalescent. Suspensions of pneumococci and other organ-
isms are made in the same way. Variations in the number
of bacteria ingested may occur according as recently isolated
or old strains are employed.

Instead of centrifuging, the suspensions may be filtered
through a double thickness of filter-paper.

5. Suspension of living leucocytes. — To prepare this,
take about 10 c.c. of physiological salt solution containing
J per cent, of sodium citrate, to prevent the coagulation
of the blood. This must be freshly prepared (or kept
sterile, which is inconvenient), and the simplest method
is to use " soloids " prepared for the purpose by Burroughs
and Wellcome ; one of these dissolved in 10 c.c. of distilled
water will yield the solution required. This is put into
a centrifuge tube and warmed to blood-heat. A healthy
person is then pricked in the ear or finger, and his blood
is allowed to drop into the fluid until 1 c.c. or more has
been collected. The tube is then centrifuged until all the
corpuscles have come to the bottom and the supernatant
fluid is left clear. If the deposit is closely examined the
red corpuscles will be seen to be at the bottom, whilst
above them there is a thin grey layer of leucocytes. The
whole of the clear fluid is then pipetted off, as close as
possible to the leucocyte layer, but without disturbing
the latter, with a pipette armed with an india-rubber teat,
or with a syringe. The tube is again filled with saline
solution, the blood and fluid are mixed, the mixture is
centrifuged, and the clear fluid pipetted off, and this
process of washing is repeated. Next, the leucocyte layer
with the upper layer of red corpuscles (which also contains

THE OPSONIC INDEX 217

leucocytes) is pipetted off into a small tube, and the whole
is thoroughly mixed by repeatedly sucking into, and
expelling from, the pipette. The result is a suspension of
living leucocytes mixed with red corpuscles.

The process.- — (1) Make a pipette and place an india-
rubber teat on the thick end. With a grease pencil or
with ink, make a transverse line about an inch from the
point ; the volume of fluid contained between the point
and this mark is spoken of as the unit.

(2) Having the patient's serum and the suspensions of
leucocytes and of bacteria ready to hand, take the pipette
between the index ringer and thumb of the right hand
and compress the nipple. Immerse the point beneath
the surface of the suspension of bacilli, and relax the
pressure on the nipple until the suspension has risen exactly
to the mark so that one unit has been drawn up ; then
remove the point from the fluid and relax the pressure
again so that a small volume of air is sucked up. This
will be quite easy if the point is a good one, otherwise
it will be difficult or impossible, as the column of fluid
will either refuse to stir or will oscillate violently. Next
immerse the point in the suspension of leucocytes and draw
up one unit. This will be separated from the suspension
of bacteria by the bubble of air. Kemove the point from
the suspension and draw up a second volume of air.

Lastly, draw up one unit of the serum. There will now
be in the pipette (counting from the nipple towards the
point) one unit of bacterial suspension, a bubble of air, a
unit of leucocytes, a bubble of air, and lastly a unit of
serum (c, Fig. 35).

(3) Put the point of the pipette on to a clean hollow-
ground slide or an artist's porcelain sunk palette, and
express the whole of its contents, and mix well together,
aspirating them repeatedly into the pipette and expelling
without causing bubbles. If bubbles form, a hot wire

218 A MANUAL OF BACTERIOLOGY

brought near will quickly dispel them. When thoroughly
mixed, aspirate the mixture into the pipette, suck up a
short volume of air, and seal the tip in the flame.

Then place the pipette point downwards in the incubator,
or better, in a water-bath at 35° to 37° C., noting the
time exactly, and proceed to prepare a second pipette in
precisely the same way, using the same suspensions of
bacteria and leucocytes, but the control serum instead of
the patient's. Place this in the incubator or water-bath,
by the side of the other, noting the time at which this is
done. When each pipette has been incubated for a quarter
of an hour it is removed from the incubator or water-bath,
the end broken off and the nipple fitted to the thick end ;
then the contents are expelled on to a hollow slide or
porcelain palette and mixed thoroughly together. Films
are then prepared. This may be done by depositing a
drop in the middle of a large cover- glass (1-inch squares,
No. 2), dropping on to it another cover- glass and drawing
the two apart. Or the films may be made on slides, for
which Wright recommends roughing the slides with fine
emery paper and spreading the film with the sharp edge
of a broken slide (see next page). The films then have to
be stained. For staphylococci, streptococci, pneumococci,
B. coli, etc., the films may be fixed with formalin and
stained with carbol-thionine blue or borax-methylene blue
(see " Malaria "), or they may be stained without previous
fixing with the Leishman stain. For tubercle, the films
may be fixed in a saturated solution of mercuric chloride
(one or two minutes), stained in warm carbol fuchsin,
decolorised with 2J per cent, sulphuric acid in methylated
spirit, and counterstained with methylene blue.

Wright now uses the whole blood instead of the leuco-
cyte layer only. After the blood has been drawn into the
titrated salt solution it is centrifuged, washed twice with
salt solution, the fluid is pipetted off, and finally the

PREPARATION OF VACCINES 219

corpuscles are well mixed. The various mixtures — washed
corpuscles, bacterial suspension, and serum — are made
and incubated as previously described. In order to make
the film for staining and counting, the contents of the
pipette are discharged on to one end of a slide roughed with
fine emery paper and the mixture is spread by means of a
slide which has been broken across after notching with
a file or glass cutter. The object is to obtain a broken edge
having a very slight concavity, and many slides may
have to be sacrificed to attain this. The film is spread by
drawing (not pushing) along, the leucocytes adhere to
the edge of the spreader, and finally are deposited mostly
at the end of the preparation, the red corpuscles being
left behind.

Lastly, the films after staining are examined with the
oil-immersion lens, preferably with the aid of a mechanical
stage, and the number of organisms contained in not less
than fifty polymorphonuclear leucocytes is counted. Parts
of the film in which the cells are broken down or not well
stained, or cells containing obvious clumps of organisms,
should be avoided. The ratio between the number in
the control and the number in the specimen prepared with
the patient's serum gives the opsonic index. Thus, if in
the control there are 125, while in the patient's specimen
there are 75, the index would be T7^5- = 0-6, i.e. not much
more than half the normal.

Preparation of vaccines for treatment, etc. — The vaccine used for
treatment is a sterilised, standardised suspension of the infecting
organism, except in the case of tuberculosis, for which tuberculin
(TR or BE) or an analogous preparation is employed. In certain
instances a mixture of organisms is used — e.g. M. pyogenes, var.
aureus and var. albus, with or without the acne bacillus in some
cases of acne — and the strain of organism isolated from the lesion
is generally to be preferred.

The vaccine is prepared by growing the organism under appro-
priate conditions, , the staphylococcus on agar, the streptococcus,

220 A MANUAL OF BACTERIOLOGY

pneumococcus, and gonococcus on blood-agar, etc. The growth is
then made into a suspension by adding a few drops of sterile 0-1 per
cent, sodium chloride solution and well rubbing up with a sterile glass
or aluminium rod. Two or three tubes are treated in this way ;
the suspension is poured into a sterile tube or small flask of stout
glass, the culture tubes are rinsed out with a little more of the salt
solution, and the washings added to the suspension, two or three
sterile glass beads are added, and the vessel, sealed or corked, is
shaken vigorously for some time, preferably in a shaking machine,
so as thoroughly to break up the masses of organisms. The con-
tents of the vessel, which should measure 5 c.c. or thereabouts, are
then centrifuged for some minutes, the suspension is poured off from
the deposit into a second sterile flask and is now ready for
standardisation.

Standardisation is carried out by Wright's method. Two or
three volumes of citrate solution are sucked up into a pipette such
as that used for opsonic determinations, the finger is pricked and
one volume of blood is taken up in the pipette, separated from the
citrate solution by an air-bubble, and finally one volume of the
bacterial suspension, also separated from the blood by an air-bubble,
is taken up. The whole contents of the pipette are then well
mixed by expelling on to a clean slide and sucking up three or four
times. About one-third of the mixture is then transferred to each
of three clean slides, and the drops are spread with the edge of a
slide so as to obtain thin uniform smears. These are allowed to
dry, stained with Leishman's stain, and the number of red corpuscles
and bacteria is counted in a number of microscopical fields. Assum-
ing that there are 5,000,000 red cells in a cubic millimetre of blood, it
is easy to calculate approximately the number of bacteria contained
in the suspension. Suppose that 500 red cells have been counted,
and with these 1500 bacteria are admixed. Since equal volumes
of blood and suspension have been taken, one cubic millimetre of

bacterial suspension will contain 5'000>0^Q X 15Q° = 15,000,000

£>00

bacteria. But one cubic centimetre contains 1000 cubic milli-
metres, therefore the suspension contains 15,000,000 x 1000 =
15,000,000,000 bacteria per cubic centimetre, and by appropriate
dilution any bacterial content of the suspension may be obtained.
Thus, if 1,000,000,000 organisms per cubic centimetre is desired,
1 c.c. of the suspension must be diluted with 14 c.c. of salt solution.
To the prepared dilution of the bacterial suspension 0-5 percent, of
carbolic acid, or 0-2 per cent, of trikresol, is added, and the flask is
placed in a water-bath at 56° to 60° C. for one or one and a half

DOSAGE OF VACCINES

221

hours, according to the resistance of the organism. The stock
solution may subsequently be introduced into small sterile glass
ampoules of 1-2 c.c. capacity, which, after sealing and standing for
twenty -four hours, may again be sterilised for an hour at 60° C.
to ensure the destruction of the organisms ; cultures may be made
from the sterilised vaccine to ascertain that this is the case. The
lower the temperature and the less the heating, consistent with
sterilisation, the more active will be the vaccine.

The annexed Table x gives an idea of the doses of vaccines, their
toxicity, and frequency of inoculation.

Vaccine

Relative toxicity

Doses

Frequency of
inoculation

Tuberculin

Very toxic

looooo ~~ foooo ~

Every 10-14 days.

B. coli

Very toxic

5-15 millions

Every 2, 5, or 10

days.

Pneumococcic

Less toxic

10-50 millions

Every 36-48

than B. coli

hours in pneu-

monia ; every 10

days in chronic

infections.

Streptococcic

More toxic than

20-60 millions

Every 7-14 days.

pneumococcic

Staphylo-

Less toxic than

100-1000

Every 10 days.

coccic

streptococcic

millions

M. melitensis

—

Y^Q- sq. cm. of

Every 7-14 days.

surface agar

culture (because

very difficult to

count)

Gonococcic

Slightly toxic

100-500 millions

Every 7-14 days.

The smaller doses are given at the commencement of the treat-
ment, and the doses are gradually increased.

The writer has employed endotoxin solutions as vaccines and
believes they are very efficient.

Prophylactic vaccines. — In addition to the therapeutic vaccines
for the treatment of the declared disease, vaccines are also employed
for prevention of disease. The preventive or prophylactic vaccines
may be :

(1) Living, but attenuated, cultures, e.g. anthrax and cholera.

1 See Harris, Practitioner, May 1908, p. 647.

222 A MANUAL OF BACTERIOLOGY

This method has also been proposed for plague, and vaccinia must
be regarded as being of this nature (this is the " Pasteurian
method ").

(2) Killed cultures, autolysed cultures, and endo-toxins. — The
first and second are used for typhoid, plague and dysentery, and
Hewlett has suggested endo-toxins for typhoid, cholera, plague and
diphtheria.

(3) Immune sera give protection for a limited time.

(4) Besredka has suggested " sensitised vaccines," i.e. living
cultures saturated with the homologous immune body derived from
an immune serum. He claims that the organisms being unaltered
by heating, etc., the vaccine gives better results than a dead vaccine,
while the saturation with the immune serum prevents infection
although the organisms are living.

(For further particulars, see Hewlett's Serum Therapy, ed. 2,
J. and A. Churchill, 1910.)

CHAPTER VI
SUPPURATION AND SEPTIC CONDITIONS

THE subjects of septic infection and of suppuration are of
great practical importance, and a knowledge of their
etiology is one of the main factors which have conduced
to the great advances that were made during the Victorian
era in the treatment of wounds, whether accidental or
made by the surgeon's knife.

Ogston in 1881 and Rosenbach in 1884 demonstrated
that micro-organisms are almost invariably present in
the pus of acute abscesses, and these observations were
repeatedly confirmed by subsequent investigators. A
number of experiments were then initiated in order to
ascertain whether these organisms bear a causal relation
to the phenomena of suppuration or are merely accidenta ly
present. These experiments showed that a large number
of organisms can produce suppuration, and render it
certain that in ninety-nine cases out of a hundred the
suppurative and septic conditions met with spontaneously,
or occurring after surgical interference, are due to the action
of micro-organisms. The chief of these are several micrococci
(commonly known as staphylococci, and the infections which
they produce, as staphylococcic infections) and streptococci.

Under the terms " suppuration " and " septic diseases "
are included such varied conditions as abscesses, boils and
carbuncles, cellulitis, osteomyelitis, erysipelas, gonorrhoea,
infective endocarditis, pyaemia, septica3mia and saprsemia,
puerperal fever, and hospital gangrene.

223

224 A MANUAL OF BACTERIOLOGY

As will be gathered from the descriptions of the individual
organisms, suppuration may be set up by inoculation with
several species, and a number of experiments by various
observers, carried out by inunction, subcutaneous inocu-
lation, and inoculation in the serous cavities and circula
tion, have conclusively proved that this is the case, not
only in animals, but also in man.

A problem of great importance is whether micro-
organisms are usually the cause of suppuration, or whether
mechanical injury, chemical agents, etc., can also produce
it. Mechanical injury alone does not seem to be capable
of inducing pus production, but it is otherwise with regard
to chemical agents. For a long time considerable differ-
ence of opinion existed and discordant results were
published. These discrepancies have now been explained,
and are found to depend upon the method of experiment
and the particular animal and chemical agent employed
That chemical agents should produce suppuration might
be expected, for it would be against analogy, derived from
all other bacterial diseases, if the pyogenic organisms do
not produce suppuration through the chemical substances
formed by, or present within, their cells, and if these
chemical substances act thus, why should not other
chemical substances be found to act in a similar way ?

In experiments with chemical agents the greatest care
has to be taken to exclude the entrance of micro-organisms.
This is best done by sealing the sterilised substance in
sterilised fusiform glass tubes and introducing these under
the skin or into the tissues with strict aseptic precautions.
When the wounds have completely healed the tubes are
broken by pressure and their contents allowed to diffuse
nto the surrounding tissues.

Sterilised cultures (above a certain amount) of the
Micrococcus pyogenes and a crystalline body, phlogosin,
obtained by Leber from its cultures, produce abscesses on

SUPPURATION 225

inoculation. Mercury produces suppuration in the dog,
but not in the rabbit ; silver nitrate (5 per cent, solution)
has a similar action. Ammonia fails to produce pus ; it
is either absorbed without damage, or if in stronger solution
produces necrosis of the tissues. Turpentine produces
large sterile abscesses in carnivora, and Brieger's cadaverine
is likewise stated to set up suppuration.

Buchner was also able, by warming various bacteria
with 0-5 per cent, caustic potash, to obtain a solution
containing protein which was powerfully pyogenic, and
Nannotti found that sterilised pus had a similar property.
It thus seems certain that a number of chemical sub-
stances can set up suppuration. At the same time, it
must be clearly recognised that suppuration and sup-
purative complications, as they occur naturally, are to be
regarded as due to the activity of micro-organisms in
almost every instance.

Of so-called " septic " diseases, sapraemia, septicaemia,
and pyaemia must be mentioned. By " sapraemia " is
meant the constitutional condition arising from the
absorption of the toxic products elaborated by micro-
organisms, the latter being localised and absent from
the general circulation. In the acute form it is not a
common condition, the best example being that which
occurs after parturition ; by simply clearing and washing
out the uterus the symptoms rapidly abate. In septicaemia
not only is there usually (though not necessarily) a local
site of infection, but in addition micro-organisms are
present in the general circulation. It is true they are not
abundant in the latter situation, and Cheyne1 believes
that they are to a large extent arrested in the capillaries.
Micrococci and streptococci are the commonest forms.
Pyaemia is characterised by the presence of micro-organisms,
most frequently streptococci, in the general circulation,

1 System of Medicine, Clifford Allbutt, ed. 2, vol. i, p. 876.

15

226 A MANUAL OF BACTERIOLOGY

and in addition by the formation of abscesses in various
situations. These arise usually from suppurative phlebitis
with the formation of septic emboli and thrombi. The
sequence of events, according to Cheyne,1 is (a) phlebitis
in direct connection with the wound ; (b) a thrombus
impregnated with micro-organisms is formed in the vein ;
(c) this softens and disintegrates, and particles or emboli
are carried to distant parts ; (d) these lodge in the
capillaries, with the formation of infarctions and abscesses.
Suppurative pylephlebitis is a pyaemia affecting the portal
system of vessels. As regards the so-called chronic
pyaemia or multiple abscesses, Cheyne considers that it
differs from true pyaemia in that embolism plays no part.
Organisms gain access to the blood- stream, settle in any
spot where the vitality of the tissues is depressed, grow
and multiply, and there produce an abscess.

The mere presence of micro-organisms does not always
suffice, however, for they may be present without pro-
ducing suppuration ; and the same organism, for example,
the Streptococcus pyogems, may at one time produce a
localised abscess, at another diffuse cellulitis. and at a
third pyaemia ; a number of factors control and modify
the occurrence and the particular form of septic disease.

As already mentioned (p. 199), many micro-organisms
when injected into the blood-stream are rapidly disposed
of ; so when moderate quantities of the Micrococcus
pyogenes are injected into the circulation of a rabbit,
abscesses, as a rule, form only in the kidney. If, however,
the organisms be attached to gross particles, so that they
cannot pass through the capillaries, embolism occurs and
abscesses form about the embolic foci. The virulence of
.the infecting organism, which varies much, is another factor
of great importance. The effect of inflammation and
injury in making a part " susceptible " is also very marked.
1 Loc. cit. p. 881.

MICROCOCCUS PYOGENES 227

Inject the M. pyogenes into animals in which the endo-
cardium or a bone has been damaged, and in all probability
an endocarditis or an osteomyelitis will ensue. The dose
and concentration of the organisms are also important
factors. Watson Cheyne found that 250,000,000 cocci
(M. pyogenes) injected into the muscles of a rabbit pro-
duced a circumscribed abscess, but 1,000,000,000 caused
a general septicaemia and death. So, probably, while the
cells in a healthy wound can dispose of a few organisms,
if the latter are abundant or in masses they may gain the
mastery.

Micrococcus pyogenes, var. aureus (Staphylo-
coccus pyogenes aureus)

Morphology and biology. — A minute spherical organism
measuring about 0'75 //. in diameter. It generally occurs
in more or less irregular groups, but may be met with
singly or in pairs (Plate I. c). It is non-motile, does not
form spores, and stains well with all the anilin dyes and
also by Gram's method. It is aerobic and facultatively
anaerobic, will develop in vacuo, and grows well and
rapidly on all the usual culture media at temperatures
from 18° to 37° C. On agar-agar it forms a thickish,
moist, shining growth, cream-coloured at first, but after
a day or two developing a characteristic orange-yellow
colour. It grows in the same manner on blood-serum
without liquefaction of the medium. Gelatin is rapidly
liquefied, the liquefied gelatin being at first somewhat
turbid from yellowish masses of organisms ; these later
on subside and form an orange-yellow sediment (Plate
I. d) In gelatin plates the colonies form at first small
whitish, granular points, developing in two or three days
into circular areas of liquefaction with yellowish masses
of the organism floating in them. On potato it forms a

228 A MANUAL OF BACTERIOLOGY

growth similar to that on agar. When grown in milk it
produces coagulation. Acid production (lactic and butyric
acids) can be demonstrated by growing on a neutral litmus
glucose-agar. When grown in broth or peptone water
it gives the indole reaction with the addition of a nitrite,
but not without.

The rate of liquefaction of gelatin and the pigment
production vary ; the latter is sometimes much deeper
than at others, recently isolated cultures show it better
than old ones, and the presence of oxygen also seems
to be necessary. The amount of acid production appears
to vary directly with the virulence, which is likewise very
variable.

Pathogenicity . — The Micrococcus pyogenes, var. aureus,
is by far the commonest of all organisms met with in
suppurative processes. Ogston found it alone in thirty-
four, and associated with the Streptococcus pyogenes in
sixteen, out of sixty-four cases of abscess. It occurs in
acute abscesses, boils, and acne, in some cases of puer-
peral fever and infective endocarditis, and is almost
invariably found in osteomyelitis, but only occasionally
in pyaemia. The organism injected under the skin of
man or animals produces an abscess, and injection into
the blood-stream under certain conditions is followed by
infective endocarditis or pyaemia. Impetigo pustules are
produced by inunction into the skin.

It may be said to be universally present on all parts
of the skin, and in the mouth, and is frequently met with
in the air. According to Sternberg, recent cultures in
gelatin are destroyed by an exposure to a temperature of
56° to 58° C. for ten minutes ; but when dried much
higher temperatures, 90° to 100° C., are required, and in
the dried state (on a cover-glass) it retains its vitality for
more than ten days. According to different experimenters,
from five to fifteen minutes are required to destroy it

PLATE I.

'.-

Phagocytosis by polymorphonuclear leucocytes, a. M. pyogenes, var.
aureus. b. B. tuberculosis.

c. M. pyogenes, var. aureus in pus. Smear
preparation, x 1000.

d. M. pyogenes, var aureus.

Gelatin stab-culture,

four days old.

MICROCOCCUS PYOGENES 229

with a 1-1000 mercuric chloride solution ; but it is evident
that much depends on the state of aggregation of the
organisms, and Abbott has shown that while most of the
cocci in a culture are destroyed in five minutes, a few may
survive much longer.

Toxins. — In a case of infective endocarditis examined
by Sidney Martin, due to the M. pyogenes, var. aureus, a
large amount of an albumose and of a basic body was
extracted from the blood and spleen. The albumose
produced fever and wasting, and retarded the coagulation
of the blood.

Leber extracted a crystalline body, which he termed
phlogosin, from cultures of the M . pyogenes, var. aureus,
and Brieger also obtained a crystalline base.

The decomposition products of the action of the M.
pyogenes, var. aureus, on egg-albumen are, according to
Emmerling, phenol, indole, and skatole, many volatile and
non- volatile acids, betaine, and trimethylamine.

Anti-serum. — Attempts have been made to prepare an
anti-serum by the injection of cultures, but the serum is of
no practical value. A vaccine, prepared by heating a
suspension of an agar culture to 65° C. for half an hour
and standardising, has been used with much success in
chronic staphylococcic infections, such as acne and boils.

Micrococcus pyogenes, var. albus, and var.
citreus. Micrococcus epidermidis. Micro-
coccus cereus

These organisms are of rarer occurrence than the
preceding one. In morphology and cultural characteristics
the first two agree with the Micrococcus pyogenes, var.
aureus, except that the albus produces a white, shining,
porcelain-like growth, and the citreus a lemon-yellow
growth, on agar. They are said to be less pathogenic than

230

A MANUAL OF BACTERIOLOGY

the aureus, and are only occasionally found alone, being
usually associated with the aureus. Cheyne, however,
states that in his experience the albus is more virulent
than the aureus, and mixed infections with the aureus are
regarded as more severe than infection with the aureus
alone. The albus has been found in some cases of pan-
ophthalmitis, and is said by Fliigge to be commoner than
the aureus in the lower animals.

Chief Types of Human Micrococci

Acid formation

^

3

"o •

"o

from

CO

Organism.

Broth
culture.

Pigment
on agar.

ot in mil

5 g

eduction
eutral rei

sduction
nitrate.

to

1

I

i

|

1

I

'S

G

s

C< B

P3

*s

1

o

(5

Micrococcus

Turbid

Orange,

+

+

0

+

+

+

+

+

+

pyogenes

yellow,

or white

Micrococcus

Turbid

White

+

+

_l_

+

+

+

+

0

Feeble.

epidermis

Micrococcus

Clear

White

0

0

0

4

+

0

+

0

0

salivarius

Scurf micro-

Turbid

White

0

0

0

+

0

0

0

+

0

coccus

or clear

Andrewes and Gordon1 regard the aureus, albus, and
citreus merely as variants of a single species, the Micro-
coccus pyogenes. They found that every variety of colour,
from orange, through yellow to white, might be obtained
by cultivation. The Micrococcus fiavescens, met with by
Babes in abscesses, may probably be placed in the same
category. On the other hand, the Micrococcus epidermidis
(albus), first described by Welch as occurring on the skin,
in stitch abscesses, etc., and feebly pathogenic compared

1 Rep. Med. Off. Loc. Gov. Board for 1905-06, p. 543.

MICROCOCCUS ZYMOGENES 231

with the M. aureus, is stated by these authors to be
perfectly distinct from the foregoing. Other organisms
which are occasionally met with in abscesses, the Staphylo-
coccus cereus albus and S. cereus flavus of Passet, form
shining waxy growths on agar, and do not liquefy gelatin,
and are probably variants of another species, which may
be termed the Micrococcus cereus. There may be many
other varieties of micrococci not yet properly differentiated.1
Well-defined micrococci occur in the saliva (M. salivarius),
and in the scurf from the scalp. Andrewes and Gordon
give a differential Table (see p. 230) of some of these
micrococci.

Micrococcus zymogenes

Isolated by MacCallum and Hastings2 from a case of
acute endocarditis. A minute micrococcus, non- motile,
and staining by Gram's method. On surface agar it forms
a thin, slightly elevated, moist, glistening, greyish-white
growth. In gelatin stab-cultures the growth is somewhat
opaque and granular, with slow liquefaction. Blood-
serum is slowly liquefied. On potato a thick, moist,
dirty-white growth develops, becoming dry and brownish
after three days. Broth becomes slightly clouded after
twenty-four hours' growth, but in three to four days the
organisms settle to the bottom, leaving the medium clear.
Neither indole nor gas is formed. In neutral litmus milk
the litmus is decolorised after a few hours, and in twenty-
four hours the milk is firmly curdled. Somewhat later
liquefaction of the curd ensues from above downwards ;
at first the turbid fluid is reddish in the superficial layer
and yellowish below ; ultimately the whole curd is trans-
formed into a turbid liquid with a reddish colour through-
out. These changes in milk are characteristic of the

1 See Gordon, Rep. Med. Off. LOG. Gov. Board for 1903-04, p. 388.

2 Journ. Exp. Med., vol. iv, 1899, p. 521.

232 A MANUAL OF BACTERIOLOGY

organism. It is pathogenic to white mice, hardly so to
guinea-pigs and white rats, and moderately so to rabbits ;
intra-venous inoculation into the latter sometimes sets
up an endocarditis. Harris and Longcope1 have reported
five more instances of the occurrence of this organism
(once from a cesspool, four times as secondary invasions
at autopsies), and Birge2 has isolated a similar but less
virulent organism from the larynx of crows. Braxton
Hicks3 has also isolated this organism from a case of
malignant endocarditis.

Micrococcus neoformans

This organism was isolated by Doyen from malignant
growths, and was supposed by him to be the causative
organism of malignant disease. It is a typical Gram-
positive coccus, giving a white growth on agar and
liquefying gelatin in three to four days. According to
Dudgeon and Dunkley,4 it gives all Gordon's fermenta-
tion tests for the M. pyogenes, var. albus, except that it
does not acidify mannitol.

The serum of patients suffering from malignant disease does not
give any marked agglutination with the M. neoformans, nor does
it contain opsonins specific for the organism. The M . neoformans
is non-pathogenic for rats and mice.

The Streptococci

Many streptococci of very variable virulence occur in
man and animals. Formerly only one pathogenic species
was described, Streptococcus pyogenes, now several varieties,
if not species, are recognised.

1 Centr. f. Bakt. (]> Abt.), vol. xxx, 1901, p. 353.

2 Johns Hopkins Hosp. Bull., vol. xvi, 1905, p. 309.

3 Trans. Eoy. Soc. Med., vol. v, 1912, Path. Sect., p. 126.
Journ. of Hygiene, vol. vii, 1907, p. 13.

PLATE II.

a. Streptococcus pyogenes in pus. Smear preparation, x 1000.

6. Streptococcus pycgenes. Film preparation
of a broth culture, x 1500.

c. Streptococcus pyogenes.

Pure culture on glycerin

agar.

STREPTOCOCCI 233

Morphology. — The streptococci are non-motile cocci,
each cell measuring about 1 JUL in diameter. They stain
well with anilin dyes and are Gram-positive.

Fission takes place in one direction only, so that chains
of cocci are formed. A cell here and there in a chain is
often somewhat larger than its fellows, and some authors
have considered these enlarged individuals to be arthro-
spores.

The length of the chains is very variable and may be
modified by cultivation, and occasionally branch- chains
form.

Von Lingelsheim distinguished two varieties, brevis and
longus, the former rendering broth turbid, growing in
short chains, and being non-pathogenic to mice and
rabbits, the latter leaving the broth clear, growing in long
chains, and always pathogenic to these animals.

Gordon1 divided the streptococci into four varieties, viz.
(1) the S. longus, isolated from the mouth, restricted to
an organism forming exceptionally long chains ; (2) S.
medius, including the majority of streptococci from pus,
sepsis, and erysipelas, and Lingelsheim's longus ; (3) S.
brevis, including Lingelsheim's brevis and the Diplococcus
pneumonia ; (4) S. scarlatince or conglomerate, isolated
from scarlatinal angina.

Cultural reactions. — The streptococci can be cultivated
on the ordinary culture media, and usually grow both
aerobically and anaerobically. On agar, or better, glycerin
agar, minute whitish, semi-transparent, more or less
isolated colonies form in twenty-four to forty-eight
hours (Plate II. c). On gelatin the growth has much the
same characters, and is better seen, as this medium is
clearer than agar, but it takes some days to attain the
maximum. In stab-cultures minute spherical colonies
develop all down the line of the stab, but without invading
1 Rep. Med. Off. Loc. Gov. Board for 1898-99, p. 482.

234 A MANUAL OF BACTERIOLOGY

the surrounding medium ; the gelatin is not liquefied.
In broth a flocculent deposit forms, the fluid sometimes
remaining clear, sometimes becoming turbid. There is
no growth on potato. Litmus milk is usually acidified
and sometimes coagulated, and acid is generally produced
from glucose. The indole reaction can be obtained in
broth cultures in seven to fourteen days on the addition
of a nitrite, but not without. It is the only organism with
which the writer is acquainted that does not reduce a weak
solution of methylene blue.

The thermal death-point of the streptococci is 53° to
55° C., the time of exposure being ten minutes, and they
are destroyed by weak solutions of disinfectants, e.g.
1-100 phenol, in ten minutes.

Considerable attention has been directed to the dif-
ferentiation of streptococci by Houston,1 Andrewes,2
Andrewes and Horder,3 Gordon,4 and Besredka. Con-
siderable differences are found in the fermentation re-
actions of various strains of streptococci, and Andrewes
and Horder distinguish (1) Streptococcus pyogenes from
pus, erysipelas, cellulitis, pya3mia and septicaemia, endo-
carditis, etc. (2) S. salivarius, the common type in the
saliva. Also met with, probably as a " terminal " infec-
tion, in endocarditis and septicaemia. Shades into the
S. fcecalis and S. anginosus. (3) S. anginosus, from
inflamed and scarlatina throats, endocarditis, and rheu-
mati^m. (4) S. fcecalis, abundant in faeces, air, and dust.
Met with also in endocarditis, meningitis, cystitis, and
suppuration. Two strains of the Diplococcus rheumaticus
proved to be this organism. (5) The pneumococcus.

1 Rep. Ned. Off. Loc. Oov. Board for 1902-03, p. 511, and 1903-04,
p. 472.

2 Lancet, November 24, 1906.

3 Ibid. 1906, vol. ii, pp. 708, 775, 852.

4 Ibid. November 11, 1905, and Rep. Med. Off. Loc. Gov Board
for 1903-04, p. 388.

STREPTOCOCCUS PYOGENES

235

(6) S. equinus, present in the intestine of herbivora. They
do not assert that these are absolutely denned species ;
at the most they seem to be species in the making, and are
connected by transitional forms. Walker1 does not con-
sider that these reactions afford a means of distinguishing
definite varieties among human streptococci.

Andrewes and Horder give the following Table sum-
marising the characters of the various streptococci :

r^

O-ri

.

>>

^

.

0

•

A

i

='

J

q

1

. 0

2 g

1

Name.

H

i

0

|

£
~

=
d

"3

1

~

la

1

Is

'

1«

a

a

V.

.

1

°1

i

I2

8
B

Streptococcus pyogenes
Streptococcus salivarius
Streptococcus anginosus
Streptococcus fcecalis
Streptococcus equinus
Streptococcus pneumonice

4-
±

±

++++++ |

-!-

-;-

.:.

:::

:;

±

r::
-

-:

5

+-H-H++ 1 |

longus
brevis
longus
brevis
brevis
brevis

-

0
0
0*

+ = Positive or acid-production. — = Negative or no acid-production.

± = Acid-production sometimes present, sometimes absent.

(These differences are not constant ; with various strains one or other reaction

may be lacking.)

Crowe2 makes use of Dorset's egg- medium with the
addition of 0-005 per cent, of neutral red for the purpose
of differentiating streptococci.

The Streptococcus pyogenes is found in some 16 per cent,
of acute circumscribed abscesses. It is, however, especially
frequent in spreading inflammations, lymphangitis, cellu-
litis, and progressive gangrene, and is a common cause of
septicaemia, pyaemia, and puerperal fever, It is met with
in about one-third of the cases of infective endocarditis,
occasionally in acute osteomyelitis, and seems to be the
cause of the septic pneumonia so often observed after
operations about the mouth and throat.

1 Proc. Roy. Soc. Lond., B. vol. Ixxxiii, 1911, p. 541.
3 Proc. Eoy. Soc. Med., vi, 1913 (Path. Sec.), p. 117.

236 A MANUAL OF BACTERIOLOGY

A streptococcus (S. viridans) producing a green growth
on blood-agar and belonging to the S. salivarius group has
been isolated by Major1 and others from several cases of
sub- acute infective endocarditis. It is probably not a
distinct form but only a variant of the S. salivarius.

In erysipelas, streptococci are present in the lymphatics
at the margin of the zone of redness. These were first
isolated by Fehleisen, who described the organism as the
Streptococcus erysipelatis, and by inoculation experiments
on man and animals demonstrated its causal relation to
the disease. The experiments on man were made in cases
of extensive and inoperable carcinoma and sarcoma, as it
had been noticed that malignant tumours were frequently
benefited after an attack of erysipelas. Several cases were
inoculated, and in all but one typical erysipelas developed
(see Coley's fluid, p. 250). Jordan,2 however, produced
typical erysipelas in a rabbit's ear not only with the
streptococcus, but also with staphylococci, pneumococci,
and B. coli, and although human erysipelas is generally
caused by the streptococcus, this disease may, therefore,
occasionally be produced by staphylococci, and possibly
by the pneumococcus, B. coli, and even the B. typhosus.

At one time the Streptococcus erysipelatis was considered
to be different from the Streptococcus pyogenes, but the
two organisms are now regarded as identical, the differences
in cultural characters being slight and not constant. A
typical erysipelas in the human subject may be induced
by inoculation with a pure culture of a streptococcus
derived from a case of suppurative peritonitis, and an
animal immunised against a streptococcus derived from
a case of erysipelas is also immune against a streptococcus
isolated from an abscess.

The different effects produced by the Streptococcus

1 Johns Hopkins Hosp. Bull., xxii, 1912, p. 326.

2 Munch, med. Woch., August 27, 1901.

ANTI-STREPTOCOCCIC SERUM 237

pyogenes, abscess in one case, erysipelas in another, cellu-
litis or pyaemia in a third, are attributable partly to real
differences in virulence, partly to the site of infection and
mode of entrance into the body, partly to real differences
existing between different races of streptococci. Strepto-
cocci have been described in a number of diseases about
which we know little, such as variola, scarlatina (S.
scarlatince or conglomeratus), and vaccinia, but it is un-
certain what causal relation they bear to these conditions.
Strangles, a disease of horses, seems to be due to strepto-
cocci.

Anti-serum. — The important lesions due to the strepto-
coccus and their grave nature have led to the attempt to
prepare an anti-serum, but many and great experimental
difficulties have to be overcome to do this. The virulence
of the streptococcus has to be increased by passing it
through a series of rabbits, and it is only by growing it in
serum media that satisfactory cultures for the inoculation
of the horses can be prepared. Human serum is the best,
but is difficult to obtain ; a mixture of asses' serum and
peptone beef-broth comes next. The cultures are grown
for about a fortnight and are then inoculated into horses,
first killed and then living cultures being used, and after
a time the blood acquires anti-microbic properties. It
is customary now to make use of a " polyvalent " serum,
i.e. one prepared by the injection of many strains of
streptococci. The streptococcus anti-serum has been
employed in erysipelas, cellulitis, puerperal fever, and
pyaemia, in many cases with success. Cheyne suggested
its use before operations about the mouth and throat as
a preventive of septic pneumonia, but a vaccine would
probably be better for this purpose.

A vaccine prepared by sterilising cultures with heat has
been used with benefit in streptococcic infections, which
do not run too rapid a course, e.g. infective endocarditis.

238 A MANUAL OF BACTERIOLOGY

Bacillus pyocyaneus

This is the organism found in green and blue pus, and it
also occurs on the surface of the body. Its presence in
wounds greatly retards healing, and occasionally a general
toxaemia may result from it. It has been met with in
otitis media and in the green pus of the pleural and peri-
cardial cavities. It is a slender bacillus measuring 3 to
4 jULt frequently united in pairs and forming filaments. It
is actively motile, does not form spores, and is aerobic
and facultatively anaerobic. It does not stain by Gram's
method. On gelatin it grows freely with rapid liquefaction,
a greenish, fluorescent colour developing in the liquid,
while whitish flocculi of growth sink to the bottom. On
agar a whitish, moist layer develops, and the medium is
stained a greenish colour. On potato the growth is a
dirty brown or sometimes greenish.

Milk is coagulated and a greenish colour develops.
Broth becomes turbid, and there is a slight film formation
with a greenish colour. Oxygen is necessary for the
development of the pigment, which is generally a mixture
of a blue pigment, pyocyanin, and a yellow one,
pyoxanthose. Pyocyanin (C14H14N20) is said to be an
anthracine derivative ; it is soluble in chloroform, and
on oxidation yields pyoxanthose.1 Various races of the
organism exist, differing in their pigment production.

Subcutaneous inoculations of a small amount of a culture
produce local abscesses ; larger amounts cause oedema
with purulent infiltration of the tissues and death. Animals
can be vaccinated by means of small quantities of living
cultures or by sterilised cultures. Sterilised cultures will
prevent infection (experimentally) by anthrax if used
early — that is to say, if an animal be inoculated with

: 1 See Centr. f. BaU., xxv, p. 897 Journ. Exp. Med., September-
November 1899.

BACILLUS PYOCYANEUS 239

anthrax, and shortly afterwards injected with a broth
culture of the Bacillus pyocyaneus, a fatal result is averted.
Emmerich and Loew1 claim to have isolated from cultures
a ferment- like body, " pyocyanase," which they state
has preventive and curative properties towards anthrax
and diphtheria infections. Emmerich2 has employed the
dry pyocyanase as an application in diphtheria to dissolve
the false membrane.

Williams and Cameron3 describe four cases of diarrhoea
with green stools, wasting and death in infants in which
the B. pyocyaneus was obtained, and suggest that many
cases of marasmus may be due to it. A form of epidemic
dysentery seems occasionally to be caused by this organism
(see " Dysentery "). A few cases of general infection with
this organism have also been recorded. It has also been
isolated from conditions of dermatitis and bullous erup-
tions.4 The B. pyocyaneus has been found in water,
dung, soil, and in the effluent from filter beds. Lehmann and
Neumann state that, with the exception of pathogenicity,
there is no essential difference between this organism and the
B.fluorescens liquefaciens so frequently met with in water.

The B. pyocyaneus seems to be of more frequent occur-
rence and of greater pathogenicity in the tropics than in
this country. A disease bearing a remarkable similarity
to rabies may be caused by it (see " Kabies ").

Clinical Examination

In many cases some idea can probably be formed as to the
organisms likely to be present in the pus or discharge, etc., from
the clinical characters of the disease, in which case the examination
may be more particularly directed towards the isolation of the
suspected organism. For example, in a urethral discharge the

1 Zeitschr. /. Hyg., 1899 ; Centr. f. Bakt., xxxi (Originate], p. 1.

2 Munch, med. Woch., November 5 and 12, 1907.

3 Journ. Path, and Bact., vol. iii, 1896, p. 344 (Refs.).

4 See Fernet, Brit. Wed. Journ.. vol. ii. 1904, p. 992.

240 A MANUAL OF BACTERIOLOGY

gonococcus will be especially looked for, in an empyema following
pneumonia the Diplococcus pneumonia, and in a tropical abscess
of the liver the Amoeba coli. In all cases the pus or discharge should
be collected with aseptic precautions in sterile capillary pipettes
or in sterile test-tubes at the time of operation. The discharge from
opened abscesses and from wounds is liable to become contaminated
and the original infection to be masked. In septic wounds the
infection may be a mixed one.

In all cases the examination should be commenced as early as
possible.

(1) Make several smears from the pus or discharge.

(2) Stain one or two of these with L6 filer's blue and one or two
by Gram's method. Mount and examine microscopically.

(a) If staphylococci only are detected, the presence of the
ordinary pyogenic cocci may be suspected. Proceed as in 3, 4, and 5.

(b) If encapsuled diplococci are detected, suspect the presence of
the Diplococcus pneumonice, and proceed as in 3, 5, and 7.

(c) If diplococci and tetracocci are present, note whether they
are in groups within the pus-cells ; if so, suspect the presence of
either the gonococcus or Diplococcus intracellularis meningitidis,
and proceed as in 6.

(d) If free tetracocci are detected, suspect the presence of the
Micrococcus tetragenus, and proceed as in 3 and 4 (rare).

(e) If streptococci are present the Streptococcus pyogenes is
probably the species. Proceed as in 3, 4, and 5.

(/) If bacilli are present they may be the colon bacillus, the
Bacillus Welchii (aerogenes capsulatus), the bacillus of malignant
oedema, the tetanus bacillus, the typhoid bacillus, the Bacillus
pyocyaneus, or putrefactive bacilli of the Proteus group (which see).
The result of Gram-staining and the clinical history of the case will
be some guide.

a. The colon bacillus, especially frequent in suppurative peri-
tonitis and in diseases of the urinary organs. (See p. 387).

/3. The Bacillus Welchii (aerogenes capsulatus}, especially met with
in foul wounds and gangrenous conditions, with much development
of gas. (See Chapter XIII.)

y. The bacillus of malignant oedema occurs in septic wounds
with septicsemic complications. (See Chapter XIII.)

c). The tetanus bacillus is found in the wound in cases of traumatic
tetanus. (See Chapter XIII.)

f. The typhoid bacillus is rare ; it may occur in suppurative
conditions complicating or following typhoid fever. Proceed as in
3 and 4. (See also p. 355.)

MICROCOCCUS MENINGITIDIS 241

£. When the Bacillus pyocyaneus is present the pus or discharge
may be blue. Proceed as in 3 and 4.

(g) If yellow granules, having a rosette-like structure micro-
scopically, are present, actinomycosis may be suspected and
examined for by the methods given in Chapter XV.

(h) If thread forms be present, streptothrix or aspergillar infection
may be suspected (see Chapters XV and XVII) : if large round or
ovoid cells or yeast -like forms, Blastomycetes or Sporotrichon
(Chapter XVI).

(i) If a mixture of organisms be present, agar and gelatin plate
cultivations should be prepared and further examined by sub-
cultures from the colonies.

(j) If no organisms can be detected microscopically, proceed as
in (3), (7), or (9). In the pus of ordinary abscesses micro-organisms
can generally be detected, unless caused by the tubercle or glanders
bacillus, the pneumococcus, or the Amoeba coli. In broken-down
granulomata, e.g. gummata, if unopened, no organisms may be
present.

(3) Make several cultivations on agar and gelatin (anaerobic if
required), and examine microscopically and by subcultures when
the growths have developed.

(4) Make two or three sets of agar and of gelatine plate cultiva-
tions. Examine the colonies microscopically and by subcultures.

(5) Stain two or three of the cover-glass preparations by Gram's
method, and counter-stain with Bismarck brown.

(6) The gonococcus and Diplococcus intracellularis may be identi-
fied and distinguished by the methods detailed at pp. 247 and 242.

(7) Inoculate guinea-pigs or mice subcutaneously and intra-
peritoneally with the material.

(8) Organisms can rarely be detected in the blood by a micro-
scopical examination of stained films. Therefore 2-5 c.c. of blood
should be withdrawn and cultivated (p. 126).

(9) If the abscess be probably a tropical abscess of the liver, the
pus or scrapings from the wall of the abscess should be examined
for the presence of the Amoeba coli. (Chapter XVIII.)

Micrococcus meningitidis 1

Weichselbaum in 1887 isolated from cases of epidemic
cerebro-spinal meningitis (spotted fever) a coccus which

1 See Gordon, Rep. Loc. Gov. Board, 1907 (Bibliog.) ; Arkwright,
Journ. of Hygiene, vol. vii, 1907, p. 193 ; vol. ix, 1909, p. 104.

16

242 A MANUAL OF BACTERIOLOGY

he named the Diplococcus intracellularis meningitidis, and
further research has confirmed the accuracy of Weichsel-
baum's discovery and the etiological relationship of the
organism to the disease.

Morphology, etc. — The meningococcus, as it may be
termed, occurs as single cocci and diplococci in groups
within the leucocytes (Plate III. a) ; in grouping and
general appearance, in fact, it closely resembles the gono-
coccus, and, like the last-named, is Gram- negative, though
staining well with the ordinary anilin dyes and with the
Leishman stain. In cultures it occurs as cocci, diplococci,
and occasionally as tetrads.

Cultural characters. — The meningococcus is an obligatory
aerobe, and does not grow at a temperature below 25° C.
It will occasionally grow in primary culture on glycerin
agar, but frequently not, though when acclimatised it
grows fairly well on agar and in broth. The organism
develops best on agar smeared with blood, on ascitic-fluid
agar or broth, or on the nutrose ascitic agar of Wassermann
(termed by Gordon " nasgar ") :

Ascitic fluid . . . . .16 c.c.
Distilled water . . . . .35 c.c.
Nutrose ...... 1 grm.

The mixture is placed in a flask, brought to the boil with constant
shaking, and filtered. It is then mixed with double the volume of
ordinary nutrient agar, steamed for thirty minutes, filtered, and
filled into tubes.

The colonies of the meningococcus on this medium after
twenty-four hours' incubation at 37° C. appear as moist,
grey, translucent, circular or oval discs with regular
outline ; after a further twenty hours' growth they may
attain a diameter of 3 to 4 mm. The colonies never
exhibit any yellowish coloration as do those of some other
Gram-negative cocci. Ascitic fluid broth (ascitic fluid
1 part, broth 9 parts) is also a good culture medium, and

MICROCOCCUS MENINGITIDIS 243

it grows in milk without clotting or change in reaction.
Arkwright found that grown in gelatin at 37° C. the
meningococcus causes liquefaction, while the M. catarrhalis
does not. The organism needs constant transplantation
to maintain vitality in culture. The fermentation re-
actions, which are somewhat variable, are given in the
table on p. 248.

Symmers and Wilson1 examined the fermentation
reactions of a number of strains of the meningococcus.
Glucose, maltose, and dextrin were fermented with the
production of acid, laevulose, galactose, lactose, mannitol,
dulcitol, anda number of glucosides were never fermented.

Pathogenesis. — In man the organism causes epidemic
cerebro- spinal meningitis, and is occasionally met with
in sporadic cases of cerebro-spinal meningitis. It is also
capable of producing a hsemorrhagic septicaemia without
meningitis. It occurs in the cerebro-spinal fluid (obtained
by lumbar puncture) in the blood in 25 per cent, of the
cases provided quantities of 5 to 20 c.c. be cultured,
sometimes in the upper respiratory passages, particularly
the nose, in the middle ear, eye and joints. Park states
that the organism is usually present in the nose in the
early days of the illness. The meningococcus is patho-
genic to mice and guinea-pigs by intraperitoneal or intra-
pleural, but not by subcutaneous, injection. Intraspinal
injection into monkeys produces a typical meningitis.

An agglutination reaction is given in some cases, but
is neither constant nor marked enough to form a sure
means of diagnosis.

Symmers and Wilson1 have found that the blood of
epidemic cerebro-spinal meningitis cases may occasionally
agglutinate the B. typliosus and B. coli in comparatively
high dilutions.

1 Journ. of Hygiene, vol. ix, 1909, p. 9.
1 Ibid. vol. viii, 1908, p. 314.

244 A MANUAL OF BACTERIOLOGY

Vaccine and anti-serum. — Cases have been reported of
remarkable benefit derived by vaccinating with killed
cultures.

Flexner has prepared an anti- serum with which suc-
cessful results have been obtained.

Still observed in simple posterior basic meningitis of infants a
diplococcus closely resembling the meningococcus but growing more
freely on agar, etc. By some it is regarded as an attenuated form
of the latter. According to Arkwright it does not liquefy gelatin,
and grows on this medium at 22° C., fails to produce acid from
glucose, maltose, and galactose, and is not agglutinated by a
meningococcus serum. It is in these respects very like the
M . cinereus of Lingelsheim. Wollstein 1 failed to find any reliable
criteria of difference between strains of the D. intracellularis and
several cultures obtained from cases of posterior basic meningitis.
Houston and Rankin 2 found that ten Gram-negative cocci isolated
from cases of sporadic cerebro -spinal meningitis differed from the
D. intracellularis in respect of their opsonins and agglutinins,
though eight of them were identical with the meningococcus in fer-
mentative power. Diplococcus crassus (Gram-positive), D. mucosus
(grows on gelatin), D. flavus (produces yellow pigment), and
M. catarrhalis, the three latter Gram-negative, may occur in the
naso-pharynx. (See Arkwright, loc. cit., also p. 248.)

Gram-positive cocci and other organisms may occasionally cause
a sporadic cerebro-spinal meningitis, e.g. the pneumococcus, typhoid
and Gartner bacilli, and streptococci (S.fcecalis and S. salivarius,
Symmers and Wilson, loc. cit. 1909).

Micrococcus gonorrhoeas

The Micrococcus gonorrhcece was discovered by Neisser in
1879 in cases of gonorrhceal urethritis. In gonorrhoeal
pus it occurs usually in pairs, occasionally in tetrads, the
elements of which are somewhat ovoid in shape, their
opposed surfaces being flattened. The organism has a
characteristic arrangement : it occurs in groups within

1 Studies from the Rockefeller Inst., vol. x, 1910, No. 13.

2 Brit. Med. Journ., 1907, vol. ii, p. 1414.

PLATE III.

a. The meningococcus. Smear of cerebro-spinal fluid. x 1000.

b. The gonococcus. Smear of gonorrhoaa pus. x 1500.

MICROCOCCUS GONORRHCE^E 245

the pus-cells (Plate III. &). The individual cocci vary
somewhat in size, the average being about 0*7 //, in the
long and 0-5 JUL in the short diameter. It stains readily
with the ordinary anilin dyes, Loffler's blue being perhaps
the best, but is decolorised by Gram's method — an im-
portant practical distinction from many other cocci.

Cultural characters. — The gonococcus is difficult to
cultivate, and usually soon dies out under cultivation —
within a week, unless transferred to fresh soil — but it
does not seem to lose its virulence. Growth takes place
between 25° and 38° C., but the optimum temperature
is between 35° and 37° C. It is aerobic, and possibly
facultatively anaerobic, and will develop on a feebly
alkaline or acid soil. The ordinary agar and gelatin
media are useless for the cultivation of the gonococcus ;
it will grow only on a medium containing " native "
protein. Blood- serum agar gives fair results, but the
ordinary Loffler's blood-serum is of no use. The best
medium is agar smeared with blood. Ordinary sloping
agar tubes or small agar plates may be employed. Blood
obtained by pricking the finger, with antiseptic precautions,
is taken up in a sterile capillary tube and deposited on
the agar. A trace of gonorrhoeal pus, collected with
aseptic precautions, is taken up on a small sterile camel's-
hair brush, and is rubbed up with the drop of blood and
smeared over the surface of the agar. The cultures are
incubated at 37° C., and in twenty-four hours the colonies
of the gonococci appear as transparent greyish specks,
which increase in size up to the end of three days. At this
stage the colony measures 1 to 2 mm. in diameter, is
raised, brownish, and finely granular in appearance, and
roundish with a crinkled margin. The cocci from cultures
resemble those in the pus, but tetrads are more frequently
met with. Egg-broth also gives good results. The fer-
mentation reactions and comparison with other Gram-

246 A MANUAL OF BACTERIOLOGY

negative cocci will be found in the Table, p. 248. The
specific virulence of gonorrhoeal pus is destroyed by
exposure to a temperature of 60° C. for ten minutes.

Paihogenicity . — The gonococcus is a strictly parasitic
organism, and seems exclusively to attack man. From
inoculation experiments on the human subjects it appears
to be the specific organism of gonorrhoeal urethritis and
vulvitis. In the female it is most frequent in the urethral
or vulvar discharge, less so in that from the cervical canal,
and is rarely or never seen in a purely vaginal one. It is
generally, even at an early stage, associated with other
organisms, particularly other diplococci (see Table, p. 248)
which have to be distinguished from the gonococcus.
The features which serve to identify the latter are its
shape and size, its non-staining by Gram's method, its
arrangement in groups within the pus- cells, absence of
growth on ordinary media, the characters of the colonies,
and the fermentation reactions.

The gonococcus is associated with a variety of lesions
besides those already mentioned, viz. epididymitis,
ovaritis, salpingitis, cystitis, peritonitis, arthritis, and
conjunctivitis. It has been met with in the blood, and
occasionally produces endocarditis, pericarditis, and
meningitis. The gonococcus is fatal to guinea-pigs and
mice by intraperitoneal inoculation.

Toxin, anti-serum, and vaccine. — Christmas1 found that
the blood- serum of the rabbit, fluid or coagulated, is an
excellent culture medium for the gonococcus. By culti-
vating the gonococcus for ten days in an ascitic bouillon
mixture he succeeded in obtaining a toxin which, when
injected intravenously into rabbits in large doses, caused
death, in smaller doses fever and loss of weight, while
precipitated with alcohol and injected into the anterior
chamber of the eye it produced severe inflammation. By

1 Ann. de Vlnst. Pasteur, xi, 1897, p. 609.

MICROCOCCUS GONORRHOLE 247

injecting rabbits with small doses of the toxin immunisa-
tion was produced, and the blood acquired antitoxic
properties. A vaccine may be prepared by sterilising
cultures with heat, and has proved of service in chronic
gonorrhoeal infections.

Clinical Diagnosis

The diagnosis of gonorrhoea is very important, not only in clinical
but also in medico-legal cases. For this purpose microscopical
examination and culture methods are made use of. In a chronic
gleet the material must be examined carefully and repeatedly.

(1) Microscopical examination. — Several thin smear specimens of
the pus or discharge should be prepared. If the best results are
desired the films should be air-dried, and then fixed by placing in a
mixture of equal parts of alcohol and ether for fifteen minutes.
After fixing, a couple of the films are stained in Loffler's blue for
five to ten minutes, washed in water, dried and mounted.
Leishman's stain also gives good results, the films being merely
air-dried and not fixed. The preparations are then examined with
a TV-inch oil-immersion ; a lower power lens is useless. The ovoid
cocci in pairs, and occasionally in tetrads, occurring within the
pus-cells in groups of not less than four pairs are very characteristic.
Diplococci situated outside the pus-cells should be neglected (it is
to be noted that the nuclei of the pus-cells are deeply, the cytoplasm
only faintly, stained with methylene blue). The next step is to
ascertain the staining reaction by Gram's method. Stain two more
films for fifteen minutes in anilin gentian violet, dip in water, place
in Gram's iodine solution for two minutes, decolorise in absolute
alcohol until the drainings fail to stain white filter paper, and
counter-stain for forty-five seconds in a saturated aqueous solution
of Bismarck brown diluted with three times its volume of
distilled water. The gonococci are decolorised, and take up the
brown stain. In chronic urethritis the urine may be centrifuged,
and preparations are made from the deposit and threads and
stained.

(2) Culture methods. — Whenever a diagnosis is of great importance
an attempt should be made to cultivate the organism. Plate
cultures of agar smeared with blood as described (p. 245) and another
set with agar only should be prepared and incubated at 37° C. In
forty-eight hours colonies of the gonococcus should be recognisable
on the blood-agar, but not on the plain agar.

248

A MANUAL OF BACTERIOLOGY

If cultures are obtained, the fermentation tests (see below) may
be applied.

N.B. The greatest caution must be exercised in declaring a case
free from infection on the ground of NEGATIVE results of the examina-
tion.

The Characters of the Chief Gram-negative Cocci (Gordon)

Growth on

Growth on

0

i

0
0

I

Organism or source.

nutrose ascitic

gelatin at

Pathogenicity.

8

c*

£

c3

agar at 37° C.

20° C.

0

1?

C3

o

0

M. catarrhalis. Nasal

Opaque,

Positive (grows

Mice and

0

0

0

0

and pharyngeal dis-

granular

on ordinary

guinea-pigs by

charge

agar at 37° C.)

intraperitoneal

inoculation

only

M. intracellularis

Clear, smooth

Negative

In some cases

4-

-|-

4-

0

(meningococcus) .

mice and

Cerebro-spinal menin-

guinea-pigs by

gitis

intraperitoneal

inoculation

only

M. gonorrhoea (gono-

No growth

Negative

Ib.

_l-

-f-

0

0

coccus). Urethral

unless blood

discharge

added

From nasal discharge

Clear, smooth,

Negative at

Mice and

_l_

0

_l_

0

from Hertford case
of influenza-like epi-

later becomes
yellowish

first, positive
later (grows on

guinea-pigs by
intraperitoneal

demic (see " Influ-
enza ")

ordinary agar
at 37° C.)

inoculation

Ib. . .

Opaque, granu-

Negative

Ib.

+

-(-

+

+

lar

From urethra

Opaque, some-

Positive

—

-f

-J-

+

-f

what granular,

smooth edges

M. melitensis. Malta

Creamy and

Positive

Monkeys. Also

_

0

0

0

fever

slightly

rabbits and

yellowish

guinea-pigs by

intracorcbral

inoculation

+ = acid.

= alkali.

no action.

Micrococcus catarrhalis 1

This organism occurs in the nose and throat in cases of catarrh,
and particularly in the " influenzal cold " (see " Influenza "), in
bronchial catarrh, and occasionally in other conditions and in well
people. Morphologically it occurs in pairs and tetrads, often

1 See Gordon, Brit. Med. Journ., 1905, vol. ii, p. 423 ; Arkwright,
Journ. of Hygiene, vol. vii, 1907, p. 145.

MICROCOCCUS CATARRHALIS 249

within the polymorphonuclear leucocytes. It is Gram-negative.
The primary generation develops feebly on agar, but subsequent
generations grow fairly well, forming whitish translucent colonies.
Blood or ascitic media should be used for isolation. Some of the
fermentation reactions and a comparison with other Gram-negative
cocci are given in the table on page 248.

Micrococcus tetragenus

This organism is frequently met with in phthisical cavities and
may be expectorated in the sputum, and has also been found in
the pus of acute abscesses. The cells occur singly (diameter 1 /*),
in pairs, or in fours, and are enclosed within a capsule. It stains
with the ordinary anilin dyes and also by Gram's method. On
gelatin it develops slowly, with the formation of a thick, white,
shining growth without liquefaction. On agar the growth has much
the same characters, and on potato is white and viscous. Inoculated
into animals, particularly mice, a local abscess may form, but usually
a fatal general infection ensues, and the organism is found in the
blood and organs.

A few cases of general infection in man have been described.

Sarcina ventriculi

An organism occurring in the contents of the stomach, especially
in cases of dilated stomach. Originally described by Goodsir in
1842.

It occurs as a large ovoid cell, several of which are grouped
together quadrilaterally so as to form more or less cubical masses,
the so-called " woolpacks." According to Falkenheim, it forms on
gelatin in thirty-six to forty-eight hours roundish, prominent
colonies of a yellowish colour, and in neutral hay infusion a brownish
film and flocculi. It produces an acid reaction.

Other sarcinse also occur in the stomach.

Clinical examination. — 1. The organism can be detected in the
vomit, etc., most readily by examination in the fresh state, a little
of the material being placed on a slide, diluted with water if neces-
sary, irrigated or not with iodine solution, covered with a cover-
glass, and examined.

2. Film preparations may be stained with weak carbol fuchsin,
or by Gram's method.

250 A MANUAL OF BACTERIOLOGY

Other Organisms met with in Suppurative
and Septic Conditions

Many other organisms may be met with in various suppurative
and septic processes, e.g. :

a. The B. coli in cystitis and pyelitis, ischio-rectal abscess,
peritonitis associated with perforation and intestinal obstruction,
and puerperal fever (see Chapter X).

&. The Diplococcus pneumonia in abscesses, empyema, arthritis,
meningitis, pericarditis, peritonitis, etc. (see Chapter XII).

c. The B. typhosus in abscesses, cholecystitis, empyema, and
osteomyelitis (see Chapter X).

d. The B. cedematis and B. Welchii in foul, gangrenous wounds
(see Chapter XIII).

e. The B. tuberculosis and B. mallei (see Chapter IX).

/. The actinomyces and streptothrix forms (see Chapter XV).
g. Blastomycetes, Sporotrichon (see Chapter XVI) and Hypho-
mycetes (see Chapter XVII).

h. The Amoeba coli (see Chapter XVIII).
». Capsulated bacilli (see note, p. 258).

Coley's Fluid

This preparation consists of the toxins of the streptococcus of
erysipelas and the B. prodigiosus. It was devised by W. B. Coley,
of New York, as a cure for inoperable malignant tumours, particu-
larly sarcoma. The treatment is based on the undoubted fact that
malignant growths may decrease or even disappear completely
after an attack of erysipelas (p. 236). Originally prepared by grow-
ing a virulent streptococcus obtained from a fatal case of erysipelas
in bouillon for about ten days ; the culture is then inoculated
with the B. prodigiosus and the two are allowed to grow together
for another week or ten days. The culture is finally heated to from
58° to 60° C. for one hour, and a piece of thymol added to preserve
it. The fluid is now prepared by growing the organisms separately
and then mixing the two sterilised cultures in proper proportions.

The fluid is injected subcutaneously in the vicinity of the tumour.
The primary dose recommended is J minim of the fluid. The
dose is gradually increased each day until there is a temperature
reaction of 103° to 104° F.

Full particulars will be found in Coley's paper (Proc. Roy. Soc.
Med., vol. iii, 1909-10, Surg. Sect., p. 1).

CHAPTER VII
ANTHRAX

ANTHKAX is essentially a disease of cattle known as splenic
fever, and though occurring in England only sporadically,
or in small outbreaks, in some parts of the world it assumes
serious proportions — as in Siberia, where it has been
termed the Siberian plague. In France also at one
time it ravaged the sheep to such an extent as to
threaten them with extinction. Man is also occasionally
attacked.

Anthrax was the first disease to be definitely associated
with a specific micro-parasite, for the organism was
observed as glassy homogeneous rods and filaments in the
blood of infected animals so long ago as 1849 by Pollender
and 1850 by Davaine, and the latter also claimed in 1863
to have demonstrated by inoculation experiments the
causal relation of the organism to the disease. Davaine's
experiments were made by inoculating an animal directly
with the blood from an infected animal, and were, there-
fore, hardly conclusive, as they did not comply with the
second and third of Koch's postulates, which declare that
the micro-organism must be cultivated outside the body,
and the cultivated organism must produce the disease on
inoculation, and the objection was raised that infection
was due, not to the bacillus, but to something else in the
blood. This objection was subsequently removed by the
work of Pasteur and of Koch, who obtained pure cultures
of the organism, the Bacillus anihracis, and with these

251

252 A MANUAL OF BACTERIOLOGY

produced results the same as had previously been obtained
by inoculation with the blood of an infected animal.

Morphology. — The Bacillus anihmcis is a rod-shaped
organism varying slightly in size in different animals and
under cultivation ; in the blood it measures from 5 to 20 //.
in length and 1 to 1-25 /x breadth (Plate IV. a), but in
cultures long filaments develop. Examined in the fresh
and living condition in a hanging- drop preparation, these
rods and filaments appear homogeneous or slightly granular ;
in stained preparations, however, they are seen to be made
up of a series of segments with unstained interspaces,
each segment measuring about 4 to 5 /x in length, and the
ends of the segments appear cut off square, provided care
has been taken not to overheat in fixing and to stain with
an aqueous solution ; they also appear to be encapsuled
(p. 263). In the blood the filaments never exceed about
five or six segments in length, except perhaps in swine, in
which animals they may be somewhat longer. In cultures,
however, the filaments may be of almost unlimited length,
and lie parallel to one another or in more or less tangled
masses. In the animal body during life, and for some
hours after death, spores never occur ; but in cultures
more than a day or so old, and from which oxygen has not
been excluded, they are always present, almost every
segment containing one. The spores are ellipsoidal,
measuring about 1 ^ by 1-25 /x, and are centrally placed
in each segment, the long axis corresponding with the
long axis of the segment.

Cultural reactions. — The anthrax bacillus is aerobic and
facultatively anaerobic ; it is non-motile, and stains well
with the ordinary anilin dyes, and especially so by Gram's
method. It grows readily on all culture media at from
20° to 37° C., the latter being the optimum. Develop-
ment ceases at temperatures below about 15° and above
5° C. Small, cream-coloured, granular colonies develop

PLATE IV.

'/ V

i£l

>mli

a. Bacillus anthracis. Smear of blood of inoculated guinea-pig.
X 750.

W ' I

6. Anthrax. Section of kidney through glomerulus. x 500.

ANTHRAX

253

in a gelatin plate in about thirty hours, and in two to
three days appear as small, roundish, cream-coloured
pasty masses in little pits in the gelatin, due to its lique-
faction. Microscopically the colonies are somewhat char-
acteristic ; each consists of a mass of wavy, tangled
filaments like a tiny wad of
cotton- wool. In gelatin streak-
cultures development is slow,
and in four or five days a
creamy, pasty growth forms
in a trough of liquefaction. In
a gelatin stab- culture (prefer-
ably 5 per cent, gelatin) lateral
branches spread from the cen-
tral growth, longer in the upper
layers, shorter below, so that
at the end of a week the cul-
ture is like an inverted fir tree
(Fig. 36), and the gelatin be-
comes gradually liquefied from
above downwards. The colonies
on an agar plate develop in
twenty hours at 37° C. as cream-
coloured points. The surface
colonies microscopically consist
of little masses of wavy, tangled
filaments (Plate V. a and b) ;
" they are not circular but run to a point in two or three
directions, with gracefully curved margins " (Reichel), and
the growth is sticky. The young deep agar colonies, which
Eurich x considers most characteristic, consist of interlacing
knotted coils of fine filaments. On an agar surface culture
at 37° C. there is a copious development in eighteen hours
of a thick, cream-coloured, slimy growth, which at this
1 Journ. Path, and Bact., xvii, 1912, p. 249.

FIG. 36.— Anthrax. Gelatin
stab-culture. Seven days
old.

254 A MANUAL OF BACTERIOLOGY

early stage has a finely granular, ground-glass appearance.
On blood-serum a thick creamy layer forms, with slow
liquefaction of the medium. On potato the organism
grows freely as a dry greyish layer, with an abundant
formation of spores. In broth it forms a somewhat scanty
flocculent deposit, the broth remaining clear and giving
the indole reaction.

In old cultures various involution forms are met with ;
the rods lose their regular shape and become swollen,
producing the so-called torula forms, while the homo-
geneous appearance of the protoplasm changes and
becomes granular. Ultra-violet rays are stated by Mme.
Henri to produce marked mutations of the anthrax
bacillus (see p. 6). Spores are found in all culture media
when there has been free access of oxygen, as in surface
cultures on potato and agar ; but in a deep broth culture,
where the supply is limited, spore-formation is absent
or very scanty. Spores are never met with in the living
animal ; they only appear some hours after death, or
when matter containing the bacilli comes in contact with
air, as in the bloody discharge from the nostrils. It has
therefore been supposed that oxygen is necessary for
spore- formation to take place, but this does not seem to
be the whole explanation, for spores form in an atmosphere
of nitrogen, though they do not do so in one of hydrogen.
The life-history of the organism and the development of
spores can be well watched in a hanging- drop specimen
prepared by inoculating a droplet of broth with the blood
of an infected animal. The preparation can be observed
on a warm stage, or examined at stated times, being kept
in the intervals in the blood-heat incubator. At the end
of twenty-four hours the short filaments, which alone are
present in the blood, will have grown so long that they
stretch across the field, while the protoplasm has become
granular, and minute shining points are visible here and
there. In another twenty-four hours the filaments extend,

PLATE V.

a. Bacillus anthracis. Impression preparation of a surface colony.

X 40.

b. Bacillus anthracis. Impression preparation of a surface colony.
X 750.

ANTHRAX 255

the protoplasm becomes still more granular, and the
shining spots are now well-marked ovoid, highly refractile
bodies — the mature spores. In old cultures the rods and
filaments almost disappear, numbers of spores alone
remaining. These spores, when placed under favourable
conditions of moisture, warmth, and nutriment, again
produce rods and filaments ; a little bud appears at the
extremity of the long diameter, which grows in length and
ultimately becomes a mature rod, often with the empty
spore capsule embracing one end. Sporeless varieties of
the anthrax bacillus have been • obtained by cultivating
under unfavourable conditions, as at a high temperature
(44° C.) or in the presence of minute quantities of anti-
septics (1 : 1000 carbolic acid).

The spores are of considerable practical importance,
for they are highly resistant forms, requiring at least some
minutes' boiling and three hours in dry air at 140° C. for
their destruction, whereas the bacilli without spores are
destroyed in ten minutes in the moist condition by a
temperature of 54° C. The same resistance occurs towards
various germicidal substances. While 1 per cent, carbolic
acid solution quickly destroys bacilli without spores, the
spores resist 5 per cent, carbolic for days, and at least
5 per cent, solutions of high-coefficient phenoloid dis-
infectants, acting for not less than twenty-four hours
at 20° C., are required to kill the spores. The resistance
of the spores is stated to increase with their age, but the
writer has not found this always to be the case. Formalin
and a formalin- containing disinfectant, " Bacterol," seem
to have a selective action on anthrax spores and are
efficient disinfecting agents for them. Reichel and Gegen-
bauer recommend for the purpose a mixture of 10 per
cent, salt and 1 per cent, hydrochloric acid at 30° C.,
acting for twenty- four hours. Anthrax spores retain their
vitality and pathogenic power unimpaired for years in a
dried condition.

256 A MANUAL OF BACTERIOLOGY

Certain anthrax-like bacilli have been described and
have to be distinguished from B. anthracis, e.g. B. pseudo-
anthracis, B. anthracoides, B. anthracis similis. These are
non-pathogenic and are haemolytic for rabbit, sheep, horse,
and ox corpuscles, while the B. anthracis is non-hsemolytic.1
The former form no capsule in the animal nor when
cultivated in an inactivated serum, anthrax forms a
capsule in such circumstances.

Pathogenicity. — The anthrax bacillus is pathogenic for
man, cattle, sheep, goats, rabbits, guinea-pigs, and mice.
The horse and the pig are also susceptible ; but adult
white rats are partially,2 and dogs, cats, and Algerian
sheep are completely, immune.

Inoculated anthrax is rarely fatal to cattle in India
(Holmes).

Young white rats, or rats fatigued by muscular work,
can be infected, and frogs and fish, though immune under
ordinary conditions, can be rendered susceptible by raising
the temperature of their environment. Birds, such as
fowls and pigeons, are also almost insusceptible, but may
be rendered susceptible by lowering their temperature ;
smaller birds, such as sparrows, are more susceptible.
The virulence varies considerably and may be artificially
modified in many ways : by passing through a series of
susceptible animals it is heightened, by growing in the
body of an insusceptible animal it is lowered, and the
latter result is also obtained by cultivating for two or
three weeks at a temperature of 42° to 45° C., or by the
addition of certain chemical substances to the culture
medium — for example, O01 per cent, of potassium bi-
chromate. These methods of " attenuation," as it is
termed, are practically applied in the preparation of the
anthrax vaccine.

1 Jarmai, Centr. f. Bakt., Abt. I (Orig.), Ixx, 1913, p. 72

2 Hall, Ibid. Ixvi, 1912, p. 293.

SYMPTOMS OF ANTHRAX 257

Symptoms of the disease in cattle are not very marked.
A beast may appear a little out of sorts and the next
day be found dead, or after suffering for a day or two
with general malaise, fever, and rigors, and with a san-
guineous discharge from the nostrils and bowel, it dies
suddenly. Post-mortem, the chief feature that attracts
attention is enlargement of the spleen ; the organ may
be two or three times larger than normal, is highly con-
gested, and very soft and friable. Microscopically, the
bacillus is found in enormous numbers in the spleen,
somewhat less numerously in the blood, and still less so
in the liver, kidney, and other organs.

Swine do not often suffer from this disease, unless fed
with the offal of an infected animal, in which case the chief
clinical sign is great enlargement about the throat ; this
is almost pathognomonic, while the chains of bacilli tend
to be somewhat longer than in other animals.

Mice inoculated subcutaneously usually die in about
twenty-four hours, and enlargement and congestion of
the spleen are very noticeable. An infected guinea-pig
generally dies in about thirty-six hours and usually shows
no symptoms until the last, when it may suffer from
rigors, with high temperature, convulsions, and staring
coat. Post-mortem, the muscular tissue is found to be
pale and cedematous, the spleen is enlarged to two or three
times its normal size and is highly congested and very
soft, and minute haemorrhages may occur in the serous
membranes. Microscopically, bacilli are found throughout
the spleen, and are often so numerous that in a stained
preparation there appear to be more bacilli than tissue.
Large numbers are also present in the blood and lungs,
fewer in the liver and kidney ; in the latter organ they are
almost confined to the glomeruli (Plate IV. 6). Imme-
diately after death, however, comparatively few bacilli may
be met with in the blood, the heart, and great vessels.

17

258 A MANUAL OF BACTERIOLOGY

The spread of the disease in nature seems to result
from the ingestion of spores while the animals are feeding.
Although the bacilli without spores would be destroyed
by the acid gastric juice, this is not the case with the
spores, which are probably generally developed from the
organisms present in the bloody discharges of a stricken
animal, and are distributed by wind and flood, and in this
way may infect large tracts of pasture. Crows and foxes
may also serve to spread the disease by feeding on infected
material and disseminating the spores by the excreta.1
Pasteur suggested that earthworms might bring the spores
to the surface in their casts from the buried carcases of
infected animals, but some experiments by Koch negatived
this. The non-sporing bacilli rapidly degenerate and die
in a buried carcase.

Man seems to be relatively insusceptible to anthrax.
The disease is generally met with among butchers,
veterinary surgeons, shepherds, etc., and among those
who sort wool or hair or work with, or carry, hides, e.g.
glove-makers, tanners, porters, etc. The disease occurs
in two forms : the so-called " malignant pustule," a
cutaneous infection, not unlike an angry carbuncle,
occurring at the seat of inoculation, on exposed parts of the
body, such as the back of the neck, the face, wrists, and
hands ; and " wool-sorters' disease," a general infection,
severe and fortunately rare, through the lungs or stomach.
Rag-sorters are likewise sometimes attacked by anthrax,
but there is also a distinct " rag-sorters' disease " which
is stated to be due to a non-motile, non-sporing, non-
liquefying, capsulated bacillus, the Proteus capsulatus
hominis 2 of Bordoni Uffreduzzi.

1 MoUet, Centr. f. Bakt., Abt. I (Orig.), Ixx, 1913, p. 19.

2 Capsulated bacilli have been met with in many septic processes.
This group includes Friedlander's pneumo-bacillus, P. capsulatus
hominis, B. mucosus capsulatus of Fricke, and the B. coli immobilis.
They are met with in conditions associated with sepsis, pus production,

INDUSTRIAL ANTHRAX 259

Under the Factories and Workshops Act 1895 all
cases of anthrax contracted in connection with various
industries have now to be reported to the Home Office.
In 1909, 56 cases, in 1910, 51 cases were thus reported,
with mortalities of 21*5 and 17' 6 per cent, respectively.
In addition, in 1910 there were 31 other cases in England
and Wales. 101 cases of Anthrax occurred in 1913 with
10 deaths as follows :

Industries Cases Deaths

Wool 43 4

Horsehair .... 5 1

Hides and Skins . . .19 2

Other Industries ... 3
Not^ reportable ... 31 3

101 10

Industrial anthrax has also been exhaustively dealt
with by Legge.1 It is particularly Persian wool, Chinese
hides, and Russian hair which are dangerous, while
Argentine, Australian, and New Zealand wools are
almost innocuous. The sorting and exclusion of wool
derived from infected animals seem to be impracticable,
and the efficient sterilisation of the thousands of bales
that are imported an impossibility. As regards hides
and skins, Legge points out that it is doubtful if there is
any way in which hides to be afterwards tanned can be
effectively disinfected, and to be of real benefit it would
have to be done before the material is opened in the
warehouse ; but to secure this would be impossible. A
method introduced by Seymour Jones has been favourably
reported on 2 ; it consists in soaking the skins for twenty-
broncho-pneumonia, ulcerating stomatitis, etc. They are shortish, non-
motile, non-sporing rods, usually Gram-negative, easily cultivated and
not liquefying gelatin, and in the tissues surrounded with a capsule.

1 Brit. Med. Journ., 1905, vol. i, pp. 529, 589, and 641.

2 Ponder, Report to the Worshipful Company of Leathersellers, 1911.

260 A MANUAL OF BACTERIOLOGY

four hours in a mixture consisting of 1 per cent, formic
acid and 1 in 5000 mercuric chloride. After this treatment
the skins are soaked in a strong brine solution. The
writer, however, has found that for horse-hair the solution,
to be efficient, must be two or three times stronger
than this. As regards horse-hair, Webb and Duncan1
carried out a number of experiments on its disinfection,
from which it would seem that, leaving out of considera-
tion white or grey hair, which is liable to change colour,
no injurious effect is produced on hair by steam disinfection
provided the temperature does not exceed 218° F. ; but
this is a comparatively low temperature for efficient
disinfection, and success can then be obtained only with
minute care in the construction and regulation of the
apparatus. Legge concludes that to secure certain de-
struction of all anthrax spores in horsehair absolute
reliance cannot be placed on either steam disinfection
(within the limits in which it can be applied) or simple
boiling. Adoption of one or the other is a very material
safeguard, but risk must always be run by those who
prepare the hair for disinfection. Disinfection has been
attempted by subjecting the material to the action of
certain phenoloid disinfectants, but from experiments by
Hall and the writer, a modified Seymour-Jones method
or formalin or bacterol seem to be the only efficient ones.2

Steam disinfection at 215°-230° F. can be applied to
wool, but the fibres are materially damaged by the process.3

A number of cases of anthrax, resulting in many deaths,
have been reported in various parts of the United States
from tanneries dealing with hides imported from China.

1 Ann. Rep. of Chief Inspector of Factories, 1900, p. 472, and 1902,
p. 278.

2 In disinfection experiments with anthrax, agar should be used
for the subcultures, broth for some unexplained reason being inefficient
See Hewlett and Hall, Journ. of Hygiene, xi, 1911, p. 473.

3 See Eighth Rep., Anthrax Investigation Board.

ANTI-ANTHRAX SERUM 261

Also a number of cattle have been infected as the result
of drinking water from rivers and creeks receiving the
waste liquors from these works.

Houston1 detected the anthrax bacillus in a catch-pit
in a hide factory at Yeovil, and in sewage and effluents
and in the mud of the Yeo. It has also been met with
in linseed cake and oats.

Toxins. — From pure cultures of the Bacillus anthracis
Hoffa obtained small quantities of a ptomine, which pro-
duced fall of temperature and haemorrhages, and Hankin
isolated a proteose which in large amounts was fatal, but
in small amounts conferred immunity to subsequent
inoculation with living bacilli. Brieger and Frankel
obtained a tox-albumin from animals dead of anthrax.
Marmier, by growing the anthrax bacillus in a solution of
peptone, glycerin, and salts, and subsequently precipitating
with ammonium sulphate, obtained a toxin which he states
is neither protein nor basic, and is contained within the
bacterial cells.

Sidney Martin,2 by growing the anthrax bacillus in
alkali albumen for ten days, obtained from the culture
albumoses and an alkaloidal substance. From the bodies
of animals which had died of the disease, chiefly from the
spleen and blood, he obtained similar substances, the
amount of alkaloid being more than double that of albu-
mose. The mixed products produced fever in animals
followed by coma and death. The albumose was proved
to be the fever, and the alkaloid the coma, producer ;
the latter also caused a spreading oedema at the seat of
inoculation.

Anti-serum. — An anti-serum for anthrax was prepared
by Marchoux by immunising sheep by vaccination and
then inoculating with progressively increasing doses of

1 Second Rep. Commis. on Sewage Disposal, 1902, p. 31.

2 Brit. Med. Journ., 1892, vol. i, p. 641.

262 A MANUAL OF BACTERIOLOGY

virulent anthrax cultures. Sclavo has prepared an anti-
serum by first immunising asses with a vaccine and then
inoculating them with increasing doses of virulent cultures
over a prolonged period. This serum has been used
successfully in a number of cases of anthrax in man, and
should always be employed, 60-80 c.c. being injected
intravenously. Salvarsan also seems to be an efficient
drug for the treatment of anthrax. As already mentioned
(p. 239) B. pyocyaneus, and pyocyanase obtained there-
from, is antagonistic to anthrax infection. Louis and
Fortineau l state that they have treated 50 cases of
anthrax infection in man by injections of 10 c.c.-20 c.c.
of sterilised broth cultures of B. pyocyaneus with a
mortality of 10 per cent.

Vaccine. — An attenuated virus has been extensively
employed for the prophylactic vaccination of cattle and
sheep. Cultures are attenuated by growing at 42°-43° C.
(Pasteur, Chamberland, and Roux). A weak vaccine is
first injected, followed after ten to twelve days by an
injection of a stronger vaccine. The mortality as a result
of the vaccination is small and the animals are subsequently
protected for some months against the virulent disease.
Sobernheim has applied a combined method, 5-15 c.c. of
anti-anthrax serum being inoculated on one side of the
animal, and the vaccine on the other. This practically
eliminates all danger from the vaccine.

Clinical Examination

(1) In veterinary practice. — If an animal is suspected to have
died from splenic fever, an extensive post-mortem is inadvisable
because of the risk of distribution of material containing bacilli
with subsequent development and dissemination of spores, with
infection of pasture, etc. The abdomen should be opened and the

1 Comp. Rend. Acad. Sc., vol. 158, No. 14, 1914, p. 1035.

MALIGNANT PUSTULE 263

spleen examined. If this is found to be much enlarged, and so
soft that it can hardly be handled without rupture, there is a high
probability of splenic fever, which the history of sudden death,
with or without symptoms, coupled with a sanguineous discharge,
increases. To confirm the diagnosis, some smear preparations
should be made from the spleen and blood, and can be stained and
examined on arriving home. If slides or cover-glasses are not
available, the ear or a small piece of the spleen may be removed
and taken home, where the specimen may be examined. When
material is sent from a distance for examination the ear should be
forwarded.

The smears may be stained with Loftier 's blue and by Gram's
method with eosin. Methylene-blue staining gives the most
characteristic appearances, according to McFadyean. A smear
preparation is made, not too thin, is air-dried, and then fixed by
passing once through the Bunsen flame. The film is stained in a
1 per cent, aqueous solution of methylene-blue for ten minutes
and then lightly rinsed and dried. The anthrax bacilli appear as
blue rods surrounded by a pale violet capsule. If the post-mortem
has been made shortly after death no spores are visible. Unless the
material be quite fresh large saprophytic bacteria somewhat resembling
anthrax are always present and must not be mistaken for that
organism ; by the McFadyean method of stain these saprophytes
do not show the violet capsule. If a hanging-drop preparation can
be made, a characteristic is the non-motility of the bacilli.

The stained preparations can be kept and produced in a court of
law if necessary. Cultivations can also be made from the spleen,
but the necessary culture media are not of course usually forth-
coming. Finally, a guinea-pig or mouse may be inoculated sub-
cutaneously in the abdomen with a particle of the spleen, and after
death examined microscopically and by culture methods.

As regards the disposal of the carcase of an animal dead from
anthrax, this should be burned if possible, but, failing this, it may
be buried in a deep pit, preferably with plenty of lime. All traces
of blood and discharge must be carefully mopped up with a strong
lime-wash or solution of chloride of lime, or other reliable
disinfectant.

(2) In man. — In malignant pustule, smear specimens should be
prepared from the fluid of the vesicles or with the scrapings from the
incised pustule, or sections of the excised pustule may be made,
and stained, some with Loffler's blue, others by Gram's method
with eosin. The bacilli are not often met with in the blood, except
shortly after death. Examination of the blood-serum of the case

264 A MANUAL OF BACTERIOLOGY

by the opsonic method, using anthrax spores, may be of value.
At the same time cultivations on agar and gelatin should be pre-
pared, and may yield positive results when the microscopical
examination has been negative. In the later stages of the disease
the bacilli may be difficult to find, even in sections.

In all cases of doubt a guinea-pig or mouse should be inoculated
subcutaneously with the material, and if the animal dies the diagnosis
of anthrax may be confirmed by the characteristic appearances, by
a microscopical examination, and by cultivation. The animal
experiment is by far the most certain method of diagnosis, a nega-
tive result being nearly as valuable as a positive one.

N.B. It must be noted that both cultivation and inoculation
experiments may fail to give positive results if the material be old
or putrid.

(3) In -wool, hair, etc. — Eurich (loc. cit.) recommends a suitable
quantity of the material to be placed in a flask with 50 c.c. to
100 c.c. of boiled water to which 3-5 c.c. of 5 per cent, solution of
caustic potash is added. If much blood-stained, the mixture is
allowed to stand at 37° C. for several hours. It is then poured into
a flat dish and the wool or hair is well teased. The mixture is then
heated to 80° C. for 2-3 minutes. Tubes of melted agar (6-9 c.c.)
at 80° C. are then inoculated with ^ c.c. of the wash and poured into
Petri dishes (4 inch). The characteristic deep-lying colonies
(p. 253) should then be searched for after twenty hours' incubation.
Animals may also be inoculated.

CHAPTER VIII
DIPHTHERIA J

Diphtheria in England — The Diphtheria Bacillus — The Pseudo-
Diphtheria Bacillus — Clinical Diagnosis — The Xerosis Bacillus
— Diphtheritic Affections of Birds and Animals

DIPHTHERIA seems to have been known from the earliest
ages, being recognised by the classical (medical) writers,
and it was epidemic in England and on the Continent
during the Middle Ages. Bretonneau2 experienced an
outbreak at Tours, 1818-1821, and gave to the disease
the name " Diphterite " (afterwards changed to " Diph-
terie ") from the formation of membranes which is so
marked a feature in it. In England the diphtheria deaths
have only been separately scheduled since 1855. Since
1881 until recently there has been a steady increase in the
prevalence of diphtheria, particularly in the large towns,
but latterly the prevalence seems to be decreasing.

As regards croup, it is universally admitted that the
vast majority of cases of membranous croup are cases of
diphtheria.

Diphtheria is distinctly a disease of the young, especially
at the ages from two to ten, and this holds good both for
London and for England and Wales.

That diphtheria is an infective disease is amply proved
by the history of epidemics, and by the recorded cases

1 See The Bacteriology of Diphtheria, Cambridge University Press,
1908.

2 See Memoirs on Diphtheria, New Sydenham Soc., 1859.

265

266 A MANUAL OF BACTERIOLOGY

where the disease has been conveyed from one individual
to another.

The disease occurs in all grades of severity, from the
classical ones with wash-leather-like membrane and great
prostration, to those which present a mild tonsillitis or
angina.

The bacteriological study of diphtheria was commenced
as long ago as 1882 by two German investigators, Klebs
and Loffler. Klebs especially investigated the pathological
histology, and ascribed the disease to small rod-shaped
organisms, which he observed in the membrane. It was
reserved for Loffler to place this observation of Klebs on
a firmer basis by the isolation and cultivation of the
bacillus from the membrane, and by the production of
certain phases of the disease by inoculation with the
isolated organism. The cause of diphtheria is, therefore,
this diphtheria bacillus, which, from its discoverers, is
frequently known as the Klebs-Loffler bacillus.

The isolation of the specific organism was by no means
an easy matter, as a number of other species of bacteria
is frequently associated with it in the membrane, but was
accomplished by Loffler by the use of a special culture
medium now known as Loffler's blood-serum, which
consists of a mixture of blood-serum (ox serum was that
originally used) 3 parts and glucose bouillon 1 part, the
whole being coagulated (see p. 61). On this medium the
diphtheria bacillus grows and multiplies exceedingly well,
while the other organisms associated with it in the mem-
brane are to a large extent inhibited in their growth. By
rubbing a small piece of membrane from a case of
diphtheria over the surface of two or three tubes, or
of a plate of Loffler's serum, and incubating at 37° C.
for twenty to twenty-four hours, colonies of the diph-
theria bacillus will be found more or less isolated according
to the number of organisms present in the membrane,

DIPHTHERIA 267

and by subculturing from these pure cultures may be
obtained.

Characters of the Diphtheria Bacillus

Morphology. — The B. diphtheria is a small, delicate
bacillus, with rounded ends, measuring 3 JUL or 4 //. in
length. It is non-motile and does not form spores.
The size varies somewhat even on the same medium,
and three varieties of the bacillus have been described,
viz. long, medium, and short, according to the length.
These varieties tend to be constant and to breed true.
Some of the rods both in cultures and in the mem-
brane have a swollen end, the so-called clubbing,
and parallel grouping, both in the membrane and in
cultures, is almost universal, the bacilli lying parallel
side by side (Plate VI. a). This parallel arrangement
arises from the peculiar mode of division of the bacillus.
If a cell be observed upon a warm stage it first elongates,
then becomes constricted at about its middle, and then
suddenly one side of the cell-membrane seems to rupture
and one half of the cell bends over to the other, so that
the two halves form a V . This mode of division, occurring
in contiguous cells and being repeated, and the cells thus
becoming more and more crowded together, leads to the
arrangement in parallel series. The bacilli are generally
joined end to end in pairs, and distinct thread and branch-
ing forms, though of rare occurrence, may be met with.
On different media the same strain exhibits considerable
variation in size. On blood-serum and on gelatin the
bacilli are of medium length and on the whole fairly regular
in shape ; in broth they tend to be short and stunted ;
while on agar, especially glycerin agar, they are much
larger than on the former media, and long club-shaped,
spindle-shaped and barred or segmented involution forms

268 A MANUAL OF BACTERIOLOGY

are abundant ; on blood-serum club-shaped involution
forms also occur, but sparsely in a young, eighteen to
twenty hours' culture, in a forty- eight hours' culture
more numerously.

Staining reactions. — The B. diphtheria stains well with
the ordinary anilin dyes and is Gram-positive. With
Loffler's methylene blue the coloration is usually some-
what irregular, more deeply stained portions alternating
with paler intervals, the so-called segmentation, and
especially marked with agar cultures. The ends of the
organisms are also frequently deeply stained, the so-called
polar staining, while the phenomenon known as " meta-
chromatism " is often marked both at the poles and also
in the rod, appearing as granules of a purplish tint and
contrasting with the blue of the methylene blue. With
Neisser's stain (p. 294) deep inky coloured dots, appearing
somewhat larger in diameter than the rods, occur at the
poles of the organism and occasionally at the centre.

Cultural reactions. — The diphtheria bacillus is an aerobic
and also a facultatively anaerobic organism, and grows
well on all the ordinary culture media, forming cream-
coloured growths or colonies, the latter on serum tending
to be somewhat flattened, with regular margins. It grows
slowly on gelatin, forming a raised whitish growth without
liquefaction of the medium, and flourishes in milk, with
the production of an acid reaction, but without curdling.
In broth some strains give a granular growth on the sides
and at the bottom of the tube, the broth remaining clear,
sometimes with a thin surface pellicle ; other strains may
render the broth turbid throughout. On potato the
growth is slight and invisible.

The indole reaction can be obtained in peptone-water
cultures either with or without a nitrite, but the writer
has shown that this reaction is due, not to indole, but to
skatolecarboxylic acid (see below, p. 288).

PLATE VI.

a. The Klebs-Loffler or diphtheria bacillus. Film preparation
of a serum culture, x 1500.

* w -

•-**,".>' ":-N

h^PviS.

6. Section of diphtheritic membrane with Klebs-Loffler bacilli. Gram
and eosin. x 750.

DIPHTHERIA 269

The diphtheria bacillus attacks glucose and lactose
with the formation of acid only, no gas (see Table, p.
292). As regards the production of acid, Neisser found
that during the first nine hours there is little or none ;
at the end of twenty- four hours a considerable quantity
has been formed, and the amount increases until the end
of the second day, after which the production ceases.

The B. diphtheria is agglutinated by the serum of
patients and by a diphtheria serum, but the test is difficult
to apply on account of the coherence of the growth, is
somewhat erratic with different strains, and is of no
practical value in the diagnosis of the disease. For the
same reasons, the agglutination reaction is of little use
for the recognition of the organism and for distinguishing
it from the so-called " pseudo-diphtheria " bacilli.

The Klebs-Loffler bacillus retains its vitality in cultures
for a month, and when dried for three or four weeks.
According to Welch and Abbot, it is destroyed in ten
minutes by a temperature of 58° C. It is readily destroyed
by antiseptics when in culture, but in the membrane it is
difficult to find an agent which will penetrate and kill
the bacilli beneath the surface.

The diphtheria bacillus and its characters under culti-
vation have been described somewhat fully, because of
the importance of the identification of the organism as a
means of clinical diagnosis. As mentioned at the com-
mencement of this chapter, the clinical diagnosis of diph-
theria presents many difficulties, and considerable assist-
ance may be derived from a bacteriological examination.
The diagnosis is based on the presence or absence of the
Klebs-Loffler bacillus, either in smears, or in cultivations,
made from the membrane or secretion (see p. 292). This
method is of very real assistance in doubtful, and especially
in mild, cases, which clinically it may be very difficult to
decide whether they be diphtheritic or no. The mild

270 A MANUAL OF BACTERIOLOGY

cases are those which it is of the greatest importance to
identify, especially in schools, for if not recognised the
patients may go about and prove a source of infection
to all around. The method also affords valuable evidence
as to when a case can be considered free from infection ;
so long as bacilli are present in the throat infection must
be possible, and the length of time for which they may
occasionally persist is remarkable. In half the cases the
bacilli disappear within three days of the disappearance
of the membrane, in a few cases they linger for as long
as three weeks, but occasionally they persist much longer.
The writer isolated them for so long as five months (and
virulent to the last) ; and a case is recorded in which they
persisted for no less than fifteen months after the attack.
In all cases two or three examinations should be made at
short intervals with negative results before the bacilli
can be pronounced to be absent, and no case should be
discharged from hospital until the absence of bacilli has
thus been proved. When bacilli persist, treatment with
antiseptic sprays or gargles, combined with syringing the
nose, may be tried. Syringing the nose is important, for
the bacilli probably extend to the post-nasal space, where
they are untouched by a throat spray or gargle. Another
mode of treatment has also been adopted. A polyvalent
anti-microbic agglutinating anti-diphtheria serum has been
prepared, dried, and compressed into tablets, one of which
is dissolved in the mouth every two hours, and fifteen
minutes after solution the naso-pharynx is flushed with
physiological salt solution. While this treatment some-
times succeeds, it often fails. The writer has tried the
use of subcutaneous inoculations of diphtheria endotoxin
(2-0-5-0 mgrm.) at intervals of seven to ten days. About
half the cases seem to clear after one to three injections.

With regard to the value to be attached to the bacterio-
logical examination for diphtheria, while the finding of

DIPHTHERIA 271

the bacilli is proof positive of the diphtheritic nature of
the affection and its infective nature, their apparent absence
is not of so much value, as various circumstances modify
the result. For example, an unskilled person may not
happen to touch the right spot with the swab, or from
struggling, etc., on the part of the patient even a skilled
operator may fail to reach any but a small portion of the
mucous membrane, instead of obtaining a good mop from
all over, especially when there are no definite patches of
membrane. The use of antiseptic gargles or paints shortly
before the swabbing is taken will likewise prevent the
growth of the bacilli. It sometimes happens that a very
mixed growth is obtained in the cultures, and in such
cases the Klebs-Lofner bacillus may be missed. Bearing
such sources of fallacy in mind, and making due allowances
for them, the negative result of a bacteriological examina-
tion may have considerable value in those cases which
clinically are doubtful. In no case where there is a reason-
able suspicion of diphtheria should treatment with antitoxin
be delayed until the bacteriological report is obtained.

The bacilli from the throat are frequently associated
with other organisms, especially micrococci and torulae ;
and those cases in which the temperature tends to be high
and the throat fetid are usually a mixed infection of
diphtheria bacilli with the Streptococcus pyogenes or Micro-
coccus pyogenes, var. aureus. The fact of such mixed
infection cannot, however, be definitely decided from the
cultures, as these organisms may be present in the mouth
or throat without necessarily taking part in the infective
process. Nor can the severity of the disease be gauged
from the characters or numbers of the diphtheria bacilli
and other organisms present, though perhaps in a number
of cases those which yield practically pure cultures will
probably be more severe than the cases which yield cultures
with few bacilli. It has been stated that the long form

272 A MANUAL OF BACTERIOLOGY

of the diphtheria bacillus is the most, and the short form
the least, virulent, the medium being intermediate, but
this is by no means a universal rule. Westbrook1 has
divided all forms of the diphtheria bacillus into three
groups, distinguished by their staining reactions with
methylene blue. Those with deeply staining granules he
calls " granular forms" those with transverse bands
" barred forms" and those staining evenly " solid forms"
Each group is further divided into seven types according
to shape and size, the types being designated by the letters
A to G and being progressively smaller from A to G.

It is sometimes stated that a microscopical examination,
unless controlled by inoculation of the isolated bacteria, is
unreliable. Such a statement is extremely misleading.
If the bacilli which have been cultivated from a suspicious
throat possess all the characters of diphtheria bacilli,
inoculation experiments are not needed, and if they were
performed with a negative result (i.e. the bacteria are not
virulent) would prove little, for the bacilli from different
parts of a culture from a throat often possess different
degrees of virulence. Occasionally, it is true, even the
expert may be in doubt about a particular bacillus, but
such cases are the exception. Here an inoculation experi-
ment may help, but would be of no value if a negative
result were obtained. It is absolutely essential in the
microscopical examination for diphtheria to use a good
lens, proper illumination, and sufficient amplification, not
less than 800-1000 diameters.

Paihogenicity . — The diphtheria bacillus is pathogenic
for man, the horse, ox, rabbit, guinea-pig, cat, chicken,
pigeon, and finches, all of which are more or less susceptible,
while mice and rats are immune. In man the respiratory
tract is usually affected, though the conjunctiva and other
mucous membranes, as of the vagina and stomach, and

1 Rep. Minnesota State Board of Health, 1899-1900.

DIPHTHEKIA 273

wounds may be attacked. A pseudo-membrane usually
forms, consisting of laminae of fibrin entangling a few
leucocytes and other cells, and here and there small effusions
of blood, together with coagulative necrosis of the under-
lying mucous membrane, and the bacilli are for the most
part located in the superficial layers of this pseudo-mem-
brane (Plate VI. 6), though in all cases in which the disease
has lasted for any time they are found in the lungs, spleen,
and kidneys, and may occur even in the blood. If the
patient recovers from the diphtheritic attack, paralytic
sequelae are not uncommon and are due to a peripheral
neuritis. Pseudo-membranes may be formed by other
organisms, e.g. by the streptococcus and pneumococcus
also by the pneumobacillus, and occur in Vincent's angina
(p. 296), but it is doubtful whether paralytic sequelae
follow any but a diphtheritic infection.

Some remarkable skin affections of an eczematous or
ichthymatous nature have been found by Hare1 and
others to be due to the diphtheria bacillus.

Another affection which seems to be generally diphtheritic
is membranous rhinitis. Whereas true nasal diphtheria is
a serious condition, membranous rhinitis is seldom, if
ever, attended with any risk to life, sequelae do not occur,
and it is rare to obtain a history of infection from cases
of it. This is extraordinary and very difficult to explain,
for virulent diphtheria bacilli are abundant in the nose
and nasal secretion.

Diphtheroid organisms can occasionally be isolated from
well people and those not known to have been in contact
with diphtheria cases. The Klebs-Loffler bacillus can be
isolated from the throats of nearly 7 per cent, of the
presumably healthy population ; 2 in the throats of con-
tacts the percentage rises to 33 or more. Murray and

1 Lancet, 1908, vol. i, p. 282.

2 See Eyre, Brit. Med. Journ., 1905, vol. ii, p. 1104.

18

274 A MANUAL OF BACTERIOLOGY

the writer x found diphtheria-like bacilli in 58 out of 385
children (15 per cent.) admitted into the Victoria Hospital,
Chelsea.

Ford Robertson believes that diphtheroid organisms —
possibly the Klebs-Loffler bacillus itself — may play an
important part in the production of general paralysis of
the insane. His views have not gained general acceptance,
and Eyre (loc. cit.) found that the percentage incidence of
all diphtheroid organisms and of the Klebs-Loffler bacillus
in the throats of the insane was not greater than in well
persons, and was unable to isolate the B. diphtheria post-
mortem from cases of general paralysis.

Traces of antitoxin can be detected in the blood after
an attack of diphtheria, usually at the end of the first week
of convalescence : this antitoxin has probably little to do
with the actual recovery from the disease (see p. 208).
A small amount of antitoxin has also been occasionally
found in well people and in untreated horses. It has
been suggested that in such cases there has been a latent
infection with the B. diphtheria, but on Ehrlich's side-
chain hypothesis it seems more likely that in such cases
there happens to be an excess of the receptors which
constitute antitoxin naturally free in the blood.

Guinea-pigs are the animals generally employed for
experimental work on diphtheroid organisms. In order
to compare the effects and virulence of various bacilli it
is customary to make the inoculation with a measured
volume of a forty-eight hours' broth culture. From Ol
c.c. to 2 c.c. of such a culture, according to the virulence,
inoculated subcutaneously, is usually required to kill a
250-grm. guinea-pig within three days. At the seat of
inoculation hsemorrhagic oedema forms, haemorrhages
occur in the serous membranes, and especially in the

1 Brit. Med. Journ., 1901, vol. i, p. 1474. See also Graham-Smith,
Journ. of Hygiene, vol. iii, 1903, p. 216.

DIPHTHERIA 275

adrenals, while the renal epithelium and the liver-cells
undergo cloudy degeneration.

Inoculated into the trachea of the guinea-pig, rabbit,
and chicken, pseudo-membranes form, and the same occurs
with the superficially injured conjunctiva and vagina. It
is stated by some that the diphtheria bacillus does not
develop on a normal mucous membrane — this must first
be injured, and the staphylococcus and streptococcus, so
often associated with the diphtheria bacillus in the human
subject, may play a part in preparing the way for infection
by damaging the cells and tissues. Rabbits usually live
somewhat longer than the guinea-pig after inoculation
and paralysis frequently develops if life is prolonged,
simulating the post-diphtheritic paralysis of man.

The question of the occurrence of the Klebs-Lofner
bacillus in the lower animals is of considerable importance
with regard to the spread of the disease and the conveyance
of infection. The so-called diphtheritic affections of
pigeons, poultry, and calves (referred to more in detail
below, p. 298) are as a rule diseases quite distinct from
human diphtheria, and are not communicable to man.
A number of observers assert, however, that cats may
suffer from the disease, which in these animals runs a
chronic course, and is associated with bronchitis, lobular
pneumonia, nephritis, and wasting. Klein 1 points out
that not only are cats liable to the disease in houses where
diphtheria has occurred, but that a similar infectious disease
exists naturally among cats, and symptoms similar to this
natural disease may be produced by inoculating healthy
cats with the Klebs-Loffler bacillus. The diphtheria bacillus
has also been isolated from the horse.2

Several epidemics of diphtheria have been traced to an
infected milk supply. In some instances the infection

1 Rep. Med. Officer LOG. Gov. Board for 1889, p. 162.

2 Cobbett, Centr. f. Bakt., xxviii, No. 19, p. 631.

276 A MANUAL OF BACTERIOLOGY

has undoubtedly been derived from contamination from
a human source, e.g. in an outbreak in Lambeth, Priestley
traced the infection to the ulcerated thumb of an employe
in a particular dairy which had become infected with
virulent diphtheria bacilli, but in others this mode of
infection has not been demonstrated, and it has been
suggested that certain eruptive conditions on the teats
and udder of the cow may be caused by the Klebs-Lofner
bacillus and the milk become infected therefrom. Klein l
made experiments with a view of determining this point.
He inoculated healthy cows in the shoulder with a bouillon
culture of the diphtheria bacillus. This caused fever and
local swelling, and in about a week a papular and vesicular
eruption appeared on the udders and teats. The B.
diphtheria was isolated from the contents of the vesicles
and also from the milk on the fifth day, but not subse-
quently. The cows died in two to four weeks, and the
B. diphtheria was obtained from the local lesions. Abbott 2
obtained somewhat different results, but Klein 3 points
out that these experiments were not performed under
exactly the same conditions as his own.

Klein, Eyre, Dean, and Marshall4 have isolated the
diphtheria bacillus from milk. It is to be noted that
diphtheria-like, but non-pathogenic, bacilli are often to be
found in milk and cheese (see section on " Milk ").

Toxins. — Diphtheria toxin has not been obtained in a
state of purity and its exact chemical nature is unknown.
LofHer first investigated the chemical products formed by
the diphtheria bacillus, and by precipitating bouillon
cultures with alcohol obtained a white toxic substance
which he classed among the enzymes.

Roux and Yersin precipitated the toxin from filtered

1 Hep. Med. Officer Loc. Gov. Board for 1889 and 1890.

2 Journ. Path, and Bact., vol. ii, 1894, p. 35.

3 Ibid. p. 428.

* Jmirn. of Hygiene, vol. vii. 1907, p. 32 (Refs.).

DIPHTHERIA 277

broth cultures by means of absolute alcohol, and also by
the addition of calcium chloride. They found that O4
mgrm. was sufficient to kill eight guinea-pigs or two rabbits,
and considered it to be an enzyme.

From the blood and spleen of cases of diphtheria Sydney
Martin 1 isolated albumoses (chiefly deutero-albumose) and
an organic acid, but no basic body. Injected subcuta-
neously the albumose produces much oedema and irregu-
larity of temperature ; in larger doses depression of tem-
perature with paralysis and coma. Small multiple doses,
not sufficient to destroy life, may give rise to some fever,
and in two or three days to paralysis of the hind legs in
rabbits, with general weakness and loss of weight. Post-
mortem, the nerves are found to have undergone degenera-
tion— breaking up and disappearance of the myelin and
interruption of the axis cylinder, while the heart is fatty.
The organic acid is also a nerve poison, but is not so toxic
as the albumose. From diphtheritic membrane, extracted
with a 10 per cent, salt solution, only traces of albumose
and organic acid were obtained, but the extract was
highly toxic, producing fever and paralysis. Sidney
Martin suggests that a substance of the nature of a ferment
may be present, and that the ferment in the membrane
on absorption may perhaps form the albumose in the body.
From cultures of the diphtheria bacillus in alkali-albumin,
albumose and organic acid, with similar actions to those
isolated from the body, were obtained.

Brieger and Frankel (1890) were unable to find any
basic substance in cultures, and concluded that the toxic
substance was a protein body, which they designated a
" tox-albumin." It was destroyed by a temperature of
60° C. but not by one of 50° C., even in the presence of an
excess of hydrochloric acid, and hence is probably not an
enzyme. The tox-albumin is non-dialysable, is precipitated

1 Brit. Med. Journ., 1892, vol. i, p. 641.

278 A MANUAL OF BACTERIOLOGY

by saturation with ammonium sulphate but not with
magnesium sulphate, and hence is neither a peptone nor
a globulin, contains a large amount of sulphur, and gives
the biuret and Millon's tests. A curious property of this
substance is that small quantities (2'5 mgrm. per kilo-
gramme of the body-weight) do not produce their effects
until the lapse of weeks. Brieger and Boer in a later
research prepared the diphtheria tox-albumin by precipi-
tating a bouillon culture with a 1 per cent, solution of
zinc sulphate or chloride. The precipitate of the zinc
double salt was washed with slightly alkaline water and
decomposed with a stream of carbonic acid gas. The
purified tox-albumin gives the xanthoproteic, biuret, and
Adamkiewicz's reactions, and the red coloration on heating
with Millon's reagent.

According to Ehrlich the toxin broth is a complex
mixture of toxic constituents belonging to the proteins,
but this is denied by Madsen and Arrhenius (see p. 165).
Its poisonous property gradually diminishes on keeping,
and is destroyed by boiling in five minutes, at lower
temperatures more slowly, and also by light.

Diphtheria antitoxin. — By the injection of sub-lethal
and increasing doses of the toxin into an animal an anti-
toxin is generated. For the preparation of a potent
antitoxin for therapeutic use the first essential is a highly
toxic toxin, and for obtaining this a diphtheria bacillus of
high virulence is required, and few strains possess the
necessary virulence. The virulent bacillus is grown in an
alkaline broth (rendered alkaline to the extent of about
5*7 c.c. of normal caustic soda solution per litre beyond
the neutral point of litmus) in Erlenmeyer flasks containing
half to one litre for eight to twelve days at 37° C. Various
small details have to be attended to in order to obtain toxin
of maximum toxicity ; it is important that growth should
occur upon the surface of the broth. The use of meat some

DIPHTHERIA ANTITOXIN 279

days old has been advocated, or of acid beef-broth in which
B. coli has been grown for twenty-four hours, in order
to eliminate the glucose (p. 27). L. Martin makes use of
" peptone " prepared by the auto-digestion of a pig's
stomach with dilute hydrochloric acid. The cultures are
then filtered through a Berkefeld or Pasteur-Chamberland
filter to remove the bacilli. The filtrate is germ-free and
very toxic, and a little carbolic acid may be added to
preserve it. In New York 10 per cent, of a 5 per cent,
solution of carbolic acid is added to the culture, the bacilli
are allowed to deposit by standing for forty-eight hours,
and the culture is filtered through paper ; in this way
filtration through a filter-candle is dispensed with. Less
than O'Ol c.c. of the toxin should kill a 250-grm. guinea-
pig in three to four days. Selected horses which have been
tested with mallein and tuberculin, and kept under obser-
vation for some time to ensure that they are healthy, are
then inoculated with this filtrate, commencing with a dose
of O01 to Ol c.c., according to the toxicity of the toxin,
or 20 c.c. of the toxin together with 10,000 units of anti-
toxin may be given for the first three doses. Individual
horses vary very much in their susceptibility to the toxin,
so that care has to be exercised with the first injections.
The injections are given subcutaneously over the shoulder,
and produce a local swelling and some rise of temperature
and general disturbance, lasting two or three days. When
this has passed away the inoculation is repeated, a larger
dose being administered provided the reaction due to the
former one was not too severe. The treatment is con-
tinued for five to six months, the dose of toxin administered
being gradually increased until it may attain 500 c.c. or
more. Cartwright-Wood found that by growing virulent
diphtheria bacilli for three or four weeks in ordinary
peptone broth, with the addition of 10 or 20 per cent, of
blood- serum or plasma, subjecting the culture to a tern-

280 A MANUAL OF BACTERIOLOGY

perature of 65° C. for an hour and filtering before injection,
much, larger initial doses can be given and some degree of
immunisation attained, and subsequently the ordinary
broth cultures may be injected in large doses. Individual
horses vary much in their capacity to yield antitoxin : on
the whole those that are moderately sensitive to the
toxin seem to produce most antitoxin ; a horse to be of
value should after three months' treatment yield an anti-
toxic serum containing not less than 300 units per c.c.
The required potency having been attained, as shown by
the test described below, the horse is bled with aseptic
precautions, the blood is allowed to coagulate, and the
serum is drawn off and filled into sterile bottles each
containing a dose of the antitoxic serum. A small amount
of antiseptic, such as trikresol, is generally added as a
precautionary measure to prevent the multiplication of any
stray germs that may have gained access during the
various manipulations.

Standardisation of antitoxin. — The potency of diphtheria
antitoxin is always described in " units " and is estimated
by ascertaining the quantity of antitoxin required just
to neutralise a certain amount of a standardised toxin
when both are injected into a 250-grm. guinea-pig. For-
merly, by Roux's method, the minimal lethal dose of the
toxin is first ascertained, and then the number of grammes
of guinea-pig which 1 c.c. of antitoxin will protect against
this minimal lethal dose is determined. If 0*01 c.c. of
antitoxin protects a 300-grm. guinea-pig against the
minimal lethal dose, 1 c.c. will protect 300 x 100 = 30,000
grm. of guinea-pig, and the immunising value of the anti-
toxin would be described as 30,000. This method is open
to the fallacy that if only a portion of the lethal dose be
neutralised the guinea-pig may survive, and a fictitious
value be given for the potency of the antitoxin. Behring
later adopted ten minimal lethal doses as the test dose

STANDARDISATION OF ANTITOXIN 281

of toxin, and he termed ten times the amount of antitoxin
which protects a guinea-pig against the ten minimal lethal
doses a unit (the Behring unit, which therefore = 100
minimal lethal doses of toxin), from which the Ehrlich
unit, now universally adopted, is derived. Though this
method eliminates to a large extent the objections to the
Roux method, Ehrlich found that by it the same antitoxin
tested with different toxin broths yielded different values.
This he explained by assuming that diphtheria toxin broth
contains not only toxin but also other substances which
combine with antitoxin. These substances, though non-
toxic, or comparatively so, vary in amount in different
toxin broths, and variable results, therefore, may be
obtained by the simple method of testing. These sub-
stances, having an affinity for antitoxin, are toxoids
and toxone. There are several varieties of toxoids, viz.
(1) those having a greater affinity for antitoxin than toxin
itself, protoxoids ; (2) those having the same affinity,
syntoxoids ; (3) and those having a less affinity, epitoxoids.1
Toxoids are probably derivatives of toxin ; they increase
in quantity in old toxin broth which has been kept, and
which at the same time decreases in toxicity. The toxones
also combine with antitoxin, having a less affinity for it
than toxin, are primary secretory products of the diphtheria
bacillus, and while not acutely lethal, induce induration,
necrosis, and paralysis. The toxoids are comparatively
scanty in a fresh toxin broth and are negligible, but it is
otherwise with the toxone, which is always present in
appreciable quantity. Owing to the fact that toxone has
less affinity for antitoxin than toxin has, if an exactly
neutral mixture of toxin broth and antitoxin be prepared,
considerably more than the minimal lethal dose of the toxin
broth must be added to render the mixture acutely toxic,

1 See pp. 165-168 for other views on the constitution of diphtheria
toxin.

282 A MANUAL OF BACTERIOLOGY

because the first portion of the added toxin simply dis-
places the toxone from its combination with the antitoxin,
and is neutralised by the antitoxin so set free.

Thus, suppose a certain amount of a toxin broth contains
90 units of toxin and 10 units of toxone, and to this amount
100 units of antitoxin are added so as to form a physiolo-
gically neutral mixture, the combination which occurs is
shown by the following " equation " : 90 toxin-antitoxin -f-
10 toxone-antitoxin = L0 (i.e. neutrality). If an amount
of the toxin broth be now added, corresponding to 11 units
of toxin, the effect will be as though only one unit of toxin
has been added, as is shown by the following " equation " :
90 toxin-antitoxin + 10 toxone-antitoxin + 11 toxin =
100 toxin-antitoxin + 10 toxone (free) + 1 toxin (free) =
L+ (i.e. just acutely lethal). Thus although the equivalent
of eleven minimal lethal doses of toxin has been added to
the physiologically neutral mixture of toxin broth and
antitoxin, only one minimal lethal dose of toxin remains
free and active, because ten toxin units displace the ten
toxone units from the toxone-antitoxin complex and are
neutralised by the antitoxin thus set free. Ehrlich, there-
fore, devised a method of standardisation which eliminates
irregularities due to the variable proportions of toxone
and toxin in the toxin broth by adopting antitoxin and
not toxin as the standard. In order to standardise an
antitoxin, a virulent toxin broth is employed and its
minimal lethal dose is approximately ascertained — i.e.
that amount which is just sufficient to kill a 250-grm.
guinea-pig on the fourth or fifth day. A solution of accu-
rately standardised antitoxin, which can be obtained from
the Serumspriifung Institut, Frankfort-on-Maine, is then
prepared, containing one " unit " of the antitoxin in 1 c.c.,
and the toxin is standardised with this by mixing with
one unit various quantities above and below one hundred
minimal lethal doses. It is required to ascertain the

STANDARDISATION OF ANTITOXIN 283

amount of the toxin broth which, when mixed with one
unit of antitoxin, just suffices to kill a 250-grm. guinea-
pig on the fourth or fifth day after the injection of the
mixture ; this amount of toxin is known as the L+ dose.
The L+ dose may be defined as that amount of a given
diphtheria toxin broth which is not completely neutralised
by one " unit " of standard antitoxin to the extent that
exactly one simple lethal dose of toxin remains unneutralised ;
it corresponds usually to 105-120 minimal lethal doses.
For example, suppose 0'003 c.c. of the toxin was found
to be the minimal lethal dose, with separate " units " of
standard antitoxin, 0*2, 0'3, 0*4, and 0'5 c.c. respectively
of the toxin might be mixed, and each mixture injected
into a guinea-pig ; probably the guinea-pigs receiving the
" unit " of antitoxin plus 0'2 and 0'3 c.c. of toxin would
remain alive, while the animal receiving the 0'4 c.c. of
toxin would die in twenty-four to forty-eight hours. The
death in the last case is too rapid ; more than a simple
lethal dose has remained unneutralised, and therefore the
L+ dose of toxin lies between 0'3 and 0'4 c.c., and further
experiments would have to be performed with amounts
of toxin between these limits in order to ascertain the
exact dose. Death of the guinea-pig on the fourth or
fifth day has been chosen because it has been found that
if the dose of toxin be diminished ever so little below that
producing this result, death does not ensue under nine or
ten days. That is to say, an acute intoxication is fatal
at the latest on the fourth or fifth day, a fatal result after
then being due to a chronic intoxication. The amount of
toxin which is exactly neutralised by one " unit " of the
standard antitoxin is known as the L0 dose. By exact
neutralisation is meant absence of any reaction, general 01
local, at the seat of inoculation, in the inoculated guinea-
pig. If toxin broth were a single substance, containing
only toxin, then L, - L0 = D, the simple lethal dose,

284 A MANUAL OF BACTERIOLOGY

would be equal to the minimal lethal dose. But because
of the presence of toxone, which also has an affinity for
antitoxin, D, the difference between the L+ and the L0
doses, is usually a multiple (8-12) of the minimal lethal
dose.

From these considerations we are now in a position
to define the unit of antitoxin : a " unit " is that amount
of antitoxin which will neutralise about 100 minimal lethal
doses for the guinea-pig of diphtheria toxin. From certain
considerations Ehrlich considers that the unit would
exactly neutralise 200 minimal lethal doses of a theoretical
toxin, containing only toxin and neither toxoid nor toxone,
but, inasmuch as such a toxin is unknown practically, the
unit corresponds usually to 105-120 minimal lethal doses
of a toxin broth, the extremes which Ehrlich has found
being 16 and 136 lethal doses. Having standardised a
specimen of toxin by means of standard antitoxin, this
standardised toxin is in its turn used to standardise the
antitoxic serum which has been prepared for therapeutic
use. The toxin is preserved by the addition of toluol,
and is kept in a cool, dark place ; it needs to be restand-
ardised every few weeks.

In standardising antitoxin, the L+ dose of the stand-
ardised toxin is mixed with varying amounts of the
antitoxin, the mixtures are injected into guinea-pigs, and
the amount of the antitoxic serum which neutralises the
L+ dose of toxin is thus ascertained. If, for example, it
were found that 0'05, OO4, and 0'03 c.c. of the antitoxic
serum neutralised the L+ dose of toxin, but that the guinea-
pig receiving 0'025 c.c. suffered from some local necrosis,
wasted, and died in a few days, and the animal receiving
0*02 c.c. died in two or three days, 0'03 c.c. of this anti-
toxin would be about equivalent to one unit of standard
antitoxin, and the antitoxic serum therefore contains 33
units per c.c. For all the experiments the conditions

ANTITOXIN TREATMENT 285

must be kept as constant as possible, guinea-pigs weighing
250 grm. or thereabouts employed, and to eliminate irregu-
larities a number of animals must be used. The antitoxic
constituent of diphtheria antitoxin is globulin in nature,
or is intimately associated with the globulin content of
the serum. Thus Atkinson found that if the serum is
precipitated by saturation with magnesium sulphate, the
whole of the antitoxin is carried down with the precipitate,
and also that the globulin content of the blood- serum of
antitoxin horses is increased. His results were confirmed
by Ledingham.1

There can now be no doubt as to the value of the antitoxin
treatment of diphtheria. Since the introduction of antitoxin
treatment, which was commenced about the middle of 1894, there
has been a steady decline in the case mortality from diphtheria,
especially in London, where probably the majority of the cases
are injected with antitoxin. From 1891 to 1894 the case mortality
from diphtheria in the hospitals of the Metropolitan Asylums Board
averaged about 30 per cent, in 1895 it was 22-8 per cent., and after-
wards steadily fell, until during the last eight years it has ranged
between 8-3 and 10 per cent.

Not less than 2000 units should be injected for a dose, and early
treatment is of paramount importance. As soon as there is a
reasonable probability that the case is one of diphtheria the anti-
toxin should be used, and treatment should not be delayed for the
result of the bacteriological examination. The statistics show that
in cases treated on the first day of the disease the case mortality
is 3-3, on the second day it is 6-5, on the third day 10-6, on the
fourth day 12-9, and on the fifth day and afterwards 14-8 per cent.

In bad cases, and in those coming under treatment at a late
stage of the disease, the dose may be increased to 10,000, 20,000,
or even 30,000 units with advantage, and to bring the patient under
the influence of the antitoxin as rapidly as possible the first dose
may be administered intravenously. The dose may have to be
repeated once or twice in mild cases, in bad cases perhaps every
six or twelve hours until several doses have been given, the guide
being the general condition of the patient and the rapidity of the
separation of the membrane. In addition to antitoxin, the recum-

1 Journ. of Hygiene, vol. VA» * 007, p. 65.

286 A MANUAL OF BACTERIOLOGY

bent posture and general and local treatment should be pursued
as usual.

In cases of mixed infection, in which the diphtheria bacilli are
associated with streptococci or staphylococci, diphtheria antitoxin
may prove of less value, as it has no influence on the streptococcic
or staphylococcic infection, and injections of anti-streptococcic
serum may be given in addition.

Diphtheritic paralysis seems to be rather more frequent after the
use of antitoxin than in the cases not treated with it, probably
because a greater number of cases survive.

The antitoxin has also been employed as a prophylactic in schools
or other places where susceptible individuals are congregated together,
and where cases of diphtheria have occurred, with excellent results.

The procedure in such circumstances should consist of a bacterio-
logical examination of the throats of all the inmates in the institu-
tion, isolation of those in whom the B. diphtheria is found, and the
injection of every one, or at least of all the young contacts, with a
prophylactic dose, repeated if considered desirable, ten days later.
For this purpose a dose of about 500 units should be given. The
immunity so produced does not last for more than three weeks.

The objection to the use of antitoxin for prophylaxis is that
should the patient subsequently develop diphtheria, treatment with
antitoxin may induce serious symptoms due to supersensitisation or
anaphylaxis. To obviate this, an antitoxin prepared in the ox has
been placed on the market for prophylactic use. The writer
believes that all the advantages of antitoxin without its disadvan-
tages may be obtained by the use of a vaccine consisting of diph-
theria endotoxin, and that it is of service in the treatment of carrier
cases. x Behring 2 has suggested the use of a toxin-antitoxin mix-
ture for prophylactic use and the treatment of carrier cases. This,
although non-toxic for the guinea-pig, engenders the formation of
a large amount of antitoxin in the recipient which persists for a
long time.

Some clinicians assert that antitoxin exerts its effect when
administered by the mouth or the rectum. Hewlett was unable to
detect any absorption of tetanus antitoxin from the stomach or
rectum, nor Sternberg of diphtheria antitoxin from the rectum, of
rabbits. Blumenau and Dzerzhgovsky could in no instance secure
immunity in animals by oral administration of diphtheria antitoxin,
nor could any antitoxin be detected in the blood of animals so
treated (Roussky Vratch, March 9, 1913).

1 Lancet, July 20, 1912, and June 28, 1913.

2 Deut. Med. Woch., May 8, 1913.

PSEUDO-DIPHTHERIA 287

Pseudo-diphtheria and Diphtheria-like Bacilli

Diphtheria-like bacilli are not uncommon in wounds
and in pathological exudates, etc., and in connection with
diphtheria an important question must be discussed, viz.
the occurrence and nature of the so-called pseudo-diphtheria
bacilli. The term was originally used by Loffler, and by
the rule of priority should be reserved for the organism
described by him under this name. The pseudo-diphtheria
bacillus of all authors is an organism occurring in the
throat in various anginal conditions, scarlet fever, etc.,
and occasionally in the throats and noses of well persons,
and is non-pathogenic to guinea-pigs. Park and Beebe
met with it in twenty-seven out of 330 healthy throats
examined by them. Roux and Yersin, Abbott and Frankel
describe it as morphologically resembling the Klebs-
Loffler bacillus, while Loffler, von Hofmann, Koplick,
Park and Beebe, Peters, and Hewlett and Miss Knight,1
consider that an organism differing somewhat from the
Klebs-Loffler bacillus should alone be termed the pseudo-
diphtheria bacillus ; to avoid confusion it is best to
designate it the Hofmann bacillus.

Morphology. — Typically, the Hofmann bacillus is a
shortish rod tapering towards the ends, which are rounded,
the average length being from 1'5 ^ to 2 /*, and it occurs
in pairs, resembling two suppositories placed base to base.
It is non-motile, does not form spores, is arranged in a
parallel grouping like the Klebs-Loffler bacillus (due to the
same mode of division), and involution forms are, as a
rule, not met with (Plate VII. a). It is Gram-positive,
and stains deeply and regularly with Loffler' s methylene
blue, segmentation and polar staining usually being absent.
With Neisser's stain no inky granules are perceptible, as is
the case with the diphtheria bacillus.

1 Trans. Brit. Inst. of Prev. Med., vol. i, 1897.

288 A MANUAL OF BACTERIOLOGY

Cultural reactions. — The Hofmann bacillus develops
well at temperatures from 20° to 37° C., and is almost a
strict ae'robe ; there is no growth anaerobically in hydrogen.
On serum, agar, and gelatin it forms cream-coloured colonies
or growths, barely distinguishable from those of the Klebs-
Loffler bacillus ; gelatin is not liquefied. On ordinary
potato it hardly grows at all, what growth there is being
quite invisible. On alkaline potato,1 however, it forms
distinct cream-coloured colonies, usually visible by the
second day. Tn stab-cultures in gelatin and glucose-agar
no gas is formed, and the growth is confined to the upper
part of the stab. In broth it forms sometimes a granular
deposit, sometimes a general turbidity. On neutral litmus
glucose-agar and in litmus milk a blue colour is developed,
indicating the production of alkalinity ; milk is not
curdled. Cultivated in peptone water an indole-like
reaction with sulphuric acid alone can be obtained after
a variable time, three to four weeks, while the diphtheria
bacillus gives it in about a week ; with a nitrite and
sulphuric acid the indole-like reaction can be obtained with
both the pseudo- and diphtheria bacilli in about a week.
The substance giving this indole-like reaction is not indole,
but skatole-carboxylic acid.2 A broth culture reduces a
weak solution of methylene blue. The Hofmann bacillus
is non-pathogenic to guinea-pigs in doses of 5 c.c. or more
of a forty- eight hours' broth culture, but is virulent to
certain birds (see below, p. 290). Mandelbaum and Heine-
mann 3 state that if a glycerin- agar plate be smeared with
human blood and inoculated, the diphtheria bacillus
produces colonies surrounded by a yellow zone, while the
colonies of the Hofmann and xerosis bacilli do not change

1 Ordinary potato rendered alkaline with a 10 per cent, solution of
sodium carbonate before sterilisation.

2 Hewlett, Trans. Path. Soc. Land., vol. li, 1900, p. 187 ; vol. lii,
1901, p. 113.

3 Centr. f. Bakt. (Orig.), liii, 1910, p. 356.

PLATE VII.

a. The pseudo-diphtheria or Hofmann bacillus. Film
preparation of a serum culture, x 1500.

b. Vincent's angina. Smear from exudation showing fusiform
bacilli (dark) and spirilla (light), x 2000.

THE HOFMANN BACILLUS 289

the red colour of the blood. In addition, the Hofmann
bacillus does not ferment any sugar, etc. (see Table, p. 292).

The histories of several cases investigated by Miss
Knight and Hewlett seemed to show that the Hofmann
bacillus is associated with mild anginal conditions, which
are free from complications, end in recovery, and are not
followed by sequelae. In many of the cases the anginal
condition was associated with distinct patches of mem-
brane, and in two symptoms were present suggestive of
the toxaemia which is met with in diphtheria.

In a long series of experiments Hewlett and Miss Knight
believed that some evidence was obtained of the conversion
of the Hofmann into the Klebs-Loffler bacillus and vice
versa. Moreover, the Hofmann bacillus seemed in many
instances to replace the Klebs-Loffler bacillus in the throat
during convalescence, and it is possible in a large series of
cultures to obtain connecting links between the Klebs-
Loffler bacillus on the one hand and the Hofmann bacillus
on the other. Cobbett,1 however, suggests that these facts
are capable of another explanation, viz. that during the
acute stage, diphtheria bacilli being readily found, the
Hofmann bacillus is likely to be overlooked, whereas at a
later stage a more careful search may be necessary to
detect the diphtheria bacillus, and in the course of that
search the Hofmann bacillus is therefore more frequently
seen.

Miss Knight and Hewlett came to the conclusion that
in some cases, at least, the Hofmann bacillus is a modified
Klebs-Loffler bacillus, and the view taken of its relation
to the Klebs-Loffler bacillus was, that it is a very attenuated
Klebs-Loffler bacillus, i.e. one far removed from virulence.
It would therefore seem wise to treat anginal cases in which
the pseudo-diphtheria bacillus is found as possibly infective,
though it would probably be inexpedient to admit to a

1 Journ. of Hygiene, vol. i, 1901.

19

290 A MANUAL OF BACTERIOLOGY

general diphtheria ward (unless a prophylactic dose of
antitoxin or of an endotoxic vaccine be given), nor would
antitoxin be needed in the majority.

Most authorities have been unable to convert the pseudo-
bacillus into a virulent Klebs-Loffler bacillus, or vice versa,
and many are of opinion that it has probably nothing to
do with diphtheria (Park and Beebe, Peters, Washbourn,
Cobbett, Clark). A few fatal cases have been recorded
(e.g. by Stanley Kent) in which a careful search has failed
to reveal any but Hofmann bacilli. Boycott 1 found that
the seasonal prevalence of the Klebs-Loffler and Hofmann
bacilli does not correspond, the former prevailing during
September, October, and November ; the latter is more
frequent from May to August.

Priestley records an outbreak of what he terms " pseudo-
diphtheria," in which the Hofmann bacillus seemed to
be the causative organism, and expresses the opinion
that this bacillus is not related to the Klebs-Loffler
bacillus.2

Salter 3 claimed to have found that the Hofmann bacillus
is virulent to many small birds (goldfinch, chaffinch, canary,
etc.), and that by successive passages it becomes converted
morphologically into a Klebs-Loffler form with feeble
virulence for the guinea-pig. He also found the filtered
broth culture of the Hofmann bacillus, though harmless to
guinea-pigs, to be toxic to small birds, and that it contains
a non-toxic substance (toxoid) which has the power of
combining with, and neutralising, diphtheria antitoxin.
Salter concluded, therefore, that diphtheritic organisms
are to be met with of every grade of virulence, the weakest,
known as Hofmann's or the pseudo- diphtheria bacillus,
representing the most attenuated form of the Klebs-Loffler

1 Journ. of Hygiene, 1905, vol. v, p. 223.

2 Public Health, July 1903.

3 Trans. Jenner Inst. Prev. Med., vol. ii, p. 113. (Bibliog.)

THE HOFMANN BACILLUS 291

bacillus. The writer,1 Cobbett,2 Petrie,3 Williams,4 and
Clark 5 have, however, quite failed to confirm Salter's
results. Thiele and Embleton also claim to have effected
the transformation of a typical Hofmann bacillus into a
virulent Klebs-Loffler bacillus by massive intra-peritoneal
inoculation of guinea-pigs with Hofmann culture suspended
in 30 per cent, gelatin and after death of the guinea-pig,
injection of the peritoneal exudate with a smaller amount
of living bacilli into a second guinea-pig, and repeating
this method of inoculation. Finally the bacillus became
Klebs-Loffler in morphology and 1 c.c. of its toxin killed a
guinea-pig in forty-eight hours, and this toxin was
neutralised by diphtheria antitoxin.

To sum up : the Klebs-Loffler-like avirulent bacilli met
with in the throat, the pseudo-diphtheria bacilli of Roux
and Yersin, are probably modified and avirulent diphtheria
bacilli. As regards the Hofmann bacillus, the general
trend of opinion at present is to consider it as quite distinct
from the Klebs-Loffler bacillus. Another view is to regard
it as in reality including several species, of which one
may be a modified Klebs-Loffler bacillus, the others having
no relation with this organism. The Klebs-Loffler-like
avirulent bacilli might, therefore, be regarded as true
diphtheria bacilli slightly removed from virulence, the
Hofmann bacillus, if derived from the Klebs-Loffler, as a
diphtheria bacillus far removed from virulence.

In determining the fermentation reactions of the diphtheria-like
bacilli, the organisms should first be grown in broth until they
become acclimatised to this medium, or should be grown in a
medium which suits them, e.g. broth with the addition of serum or
of ascitic fluid. Hiss's serum- water medium is satisfactory — serum

1 Brit. Med. Journ., Sup., July 9, 1904.

2 Journ. of State Med., vol. xi, p. 609.

3 Journ. of Hygiene, vol. v, p. 134.

* Journ. Med. Research, 1902, p. 83.

5 Journ. Infect. Diseases, vol. vii, 1910, p. 335.

292

A MANUAL OF BACTERIOLOGY

1 part, water 3 parts, with 1 per cent, of carbohydrate or other
substance, tinged with litmus and sterilised in the steamer on three
consecutive days. Graham-Smith * gives the following Table of
fermentation tests :

Hiss's medium (10 days' growth).

Organism.

C

1

§'

I

1

_•

O

1

i

03

1

1

|

'5

c

1

Q

0>

£

0

B. diphtherice, virulent

C

C

C

C

C

C

C

and avirulent .

A

A

A

A

A

A

A

Hofmann bacillus *

0

0

0

0

0

0

0

0

0

Xerosis bacillus *

C
A

0

0

0

0

C
A

0

0

C
A

C

C

r

B. coryzce *

A

0

0

A

0

\j

A

0

0

0

Diphtheria-like bacilli :

From the ear *

0

0

0

0

0

0

0

0

0

From the urethra *

A

0

0

A

A

A

0

0

0

From the throat *

C
A

0

0

C
A

A

C
A

0

0

0

From the fowl *

A

0

0

A

A

0

0

0

(* Avirulent to the

guinea-pig)

C = coagulation ; — = no coagulation ; A = acid ; 0 = no reac-
tion. Slight variations were occasionally noted : for example, four
out of twenty diphtheria bacilli gave no acid with lactose, and the
amount of acid production and of coagulation was somewhat variable.

Clinical Diagnosis

(A) In man and animals : — I. In some cases the diphtheria
bacillus can be identified in the membrane or discharge, and the
diagnosis established thereby.

Films are made with the exudation, or with a fragment of the
membrane teased up as finely as possible on a slide, a droplet of
water being added if necessary. One of these films should be

1 Journ. of Hygiene, vol. vi, 1906, p. 286.

DIAGNOSIS OF DIPHTHERIA 293

stained with Loffler's methylene blue, another by Gram's method.
The bacilli will be found lying parallel to one another in larger or
smaller groups, together with involution forms. Films stained
with Neisser's or Pugh's stain (see below) may also be of assist-
ance. Another method is to stain the films for five seconds in
dilute carbol-methylene blue (seven drops to 10 c.c. water),
rinsing and drying, and counter-staining in dilute carbon-fuschin
(ten drops to 10 c.c. water) for one minute, rinsing and drying
(Higley).

* II. Frequently the membrane is so crowded with different forms
of organisms that it is extremely difficult to recognise the diphtheria
bacilli with any degree of certainty. Recourse must then be had to
cultivation.

For this purpose sloping blood-serum tubes, or tubes of serum-
agar, must be employed ; simple agar is unsuitable. *

A piece of membrane or a swabbing from the throat is rubbed
over the surface of one or two serum tubes, care being taken not
to break up the medium. The tubes are incubated at 37° C. for
eighteen to twenty hours, and are then examined microscopically
whether there is any visible growth or not. If there be no visible
growth a scraping is taken by means of a sterilised platinum needle
from the whole surface and a film is prepared. If there is a visible
growth the film should be prepared from the most likely colonies,
or, if the growth be confluent, from the upper half inch or so.
A microscopical examination must always be made, for some
colonies — certain staphylococci and torulae, for example — simulate
those of the diphtheria bacillus very closely. The films may be
stained with Loffler's methylene blue for five to ten minutes, or
by Pugh's method, then washed and dried. If the films are made
on a slide, after staining, washing, and drying, a drop of cedar oil
may be put on the stained patch, which is then examined directly
without a cover-glass. If, however, there is very little growth, it
is better to make a cover-glass specimen, as the position of the
material is so much more easily located. The preparations are
examined with a iVm- oil-immersion lens magnifying not less than
800-1000 diameters, and the Klebs-Loffler bacillus is identified
from the description given in the text.

Prausnitz considers that if negative results are obtained after
eighteen to twenty -four hours' incubation the tubes should be incu-
bated for a further twenty to twenty -four hours and re-examined,

1 Various selective media have been devised, e.g. potassium-sulpho-
cyanide, neutral-red, glucose- blood-serum (Rankin, Journ. of Hyg.
xii, 1912, p. 60).

294 A MANUAL OF BACTERIOLOGY

and undoubtedly occasionally a positive result may be obtained
by this longer incubation.

Loffier's methylene blue gives much more characteristic prepara-
tions than Gram's method.

Although eighteen to twenty hours is recommended for incubating
the cultures, a microscopical examination will sometimes reveal
the bacilli at a much earlier period. The writer has found them in
as short a time as six hours, but if bacilli are then not found the
tubes must be incubated for the longer period.

Neisser's method of staining is as follows :

(a) One gramme of methylene blue (Griibler's) is dissolved in
20 c.c. of 96 per cent, alcohol, which is then mixed with 950 c.c.
of distilled water and 50 c.c. of glacial acetic acid.

(b) Two grammes of Bismarck brown are dissolved in one litre
of boiling distilled water and the solution is filtered.

The preparations are stained in (a) for one to three seconds,
rinsed in water, and stained in (b) for three to five seconds, washed
in water, dried, and mounted. The bacilli are stained brown, and
contain two, rarely three, inky-blue dots. This is a valuable con-
firmatory stain for the diphtheria bacillus, but staining for a longer
time than that recommended by Neisser is advisable, viz. half a
minute in the blue and one minute in the brown. Tanner treats
with Gram's iodine solution for half a minute after the blue. The
staining solutions seem to keep well but occasionally fail to act, so
should be controlled on an undoubted diphtheria culture.

Pugh's stain is also a very good one. It is a mixture containing
1 grm. of toluidine blue dissolved in 20 c.c. of absolute alcohol
and added to 1000 c.c. of distilled water and 20 c.c. of glacial
acetic acid. The mixture is applied for two minutes. The proto-
plasm of the bacilli is stained a pale blue and the polar bodies are
deeply stained and stand out in marked contrast ; by artificial
light they appear a reddish purple.

In the majority of cases, after a little experience, the Klebs-
Loffler bacillus will be readily recognised if present. Occasionally,
however, bacilli may be present which resemble the Klebs -Loftier
very closely, and of which it is difficult to be certain. In such a
case the following points should be noted in attempting to arrive
at a decision :

1. The character of the growth on the medium.

2. The depth of staining with Loffler's blue, and the presence or
absence of segmentation or polar staining : the Klebs-Loffler
bacillus usually stains somewhat deeply, while the bacilli resembling
it stain but feebly.

DIAGNOSIS OF DIPHTHERIA 295

3. The presence or absence of involution forms, clubbing, etc.

4. The presence or absence of thread forms : the Klebs-Loffler
bacillus does not form threads.1

5. The presence or absence of spores : the Klebs-Loffler bacillus
does not form spores.

6. Motility in a hanging drop : the Klebs-Loffler bacillus is non-
motile.

7. Gram's method of staining : the Klebs-Loffler bacillus stains
well.

8. The grouping of the organism : the parallel grouping of the
Klebs-Loffler bacillus is somewhat characteristic. The bacilli when
lying side by side do not seem quite to touch, while the bacilli
which resemble the Klebs-Loffler and show a parallel grouping
frequently lie much closer together than the Klebs-Loffler bacillus
ever does.

9. The reaction with Neisser's or Pugh's stain (the culture must
be a young serum one) : the pseudo-bacillus and other bacilli do
not give the diphtheritic reaction (polar staining).

10. The final test of virulence may be applied. For this pur-
pose the organism must be isolated in pure culture by plate cultiva-
tions. Two guinea-pigs, of 250-300 grm. weight, are each inocu-
lated with 2 c.c. of a forty-eight hours' broth culture, one receiving
at the same time 1 c.c. of diphtheria antitoxin. If the guinea-pig
inoculated with culture only dies, while the one receiving culture
and antitoxin lives, this is complete proof that the organism is the
diphtheria bacillus ; if both live no inference can be made except
that the organism is non-virulent ; if both die it shows that
the organism is virulent, but that it is not neutralised by
antitoxin, and therefore is not the diphtheria bacillus. In cases in
which bacilli persist, the test of virulence is frequently applied. If
the organism proves to be non-virulent, presumably the patient
is non-infective. Such a presumption, in the writer's opinion,
however, is not necessarily true.

11. Agglutination tests are unsatisfactory and not of service.

It occasionlly happens that a conclusion cannot be arrived at
without an extended investigation.

If serum tubes are not available an egg may be used. It is
boiled hard, the shell chipped away from one end with a knife
sterilised by heating, and the inoculation made on the exposed white ;
the egg is then placed, inoculated end down, in a wine-glass of such

1 Klein and others have described thread and branched forms in
cultures of the Klebs-Loffler bacillus in certain circumstances, but
these are not likely to be observed under the conditions mentioned.

296 A MANUAL OF BACTERIOLOGY

a size that it rests on the rim and does not touch the bottom. A few
drops of water may with advantage be put at the bottom of the
glass to keep the egg-white moist. The preparation is kept in a
warm place for twenty-four to forty-eight hours and then examined.
Antitoxin itself may be used as a culture medium, provided it con-
tains no antiseptic (this is now rarely the case). A test-tube is
sterilised by heating, or with boiling water or steam from a kettle,
antitoxin to the depth of about an inch is poured in, and is coagulated
by holding the tube very obliquely in boiling water or steam. After
coagulation and cooling the medium is inoculated. If no incubator
is available, the culture may be kept in a warm place, or in an
inside pocket.

Many laboratories now undertake the examination of material.
Culture outfits are supplied by some, consisting of a sterilised tube
containing a sterilised swab. Failing this, a piece of membrane
may be forwarded in a tube or bottle which has been sterilised by
heating, or with boiling water or steam. If there be no membrane,
a swab can be readily extemporised by wrapping a little wool or
lint (non-antiseptic] round the end of a piece of wire, knitting
needle, hair-pin, penholder, or splinter of wood. The wood may
be sterilised by moistening with water and then holding in a flame.
Membrane or secretion may also be forwarded on pledgets of wool,
pieces of lint or calico, and even on paper, but these are not so
suitable.

(B) In milk.— See section on " Milk."

Vincent's Angina

An infective malady characterised by sore throat, fetor,
dysphagia, and ulceration and membrane simulating diphtheria,
The diphtheria bacillus, however, is not present, and the affection
is caused by an apparent association of a bacillus and a spirochaete.
The bacillus (B. fusiformis) measures 6-8 p. to 10-12 p. in length,
has pointed ends and is usually somewhat bent, not straight, often
appears feebly motile, and does not stain by Gram. It can be
cultivated anaerobically on the ordinary media to which human
blood-serum, ascitic or hydrocele fluid has been added. The
spirochaete is long and sinuous and very motile, but cannot be
cultivated, and is stated to be developed from the fusiform bacillus,
Smears may be stained with methylene blue or dilute carbol-fuchsin,
and the appearance of the associated organisms is so characteristic
that a diagnosis is easily effected (Plate VII. b).

THE XEROSIS BACILLUS 297

Fusiform bacilli have been met with in various necrotic pro-
cesses, e.g. noma (see Chapter XX).

The Xerosis Bacillus

The xerosis bacillus was isolated by Neisser from cases of xerosis
conjunctives, and is met with in follicular conjunctivitis. Lawson
and also Griffith isolated it from nearly 50 per cent, of all normal
conjunctival sacs. In morphology and staining reactions it re-
sembles the Klebs-Loffler bacillus very closely. It differs from the
Klebs-Loffier bacillus in the following particulars : (1) Usually,
but not always, in the primary cultivations from the eye on blood-
serum, colonies do not appear under about thirty hours, while
those of the Klebs-Loffler bacillus are visible in sixteen to twenty
hours. This does not apply to the secondary cultivations, in which
the colonies appear as soon as those of the Klebs-Loffler bacillus.

(2) Upon agar it will seldom or never grow in primary culture, and
in secondary cultures it forms only a thin, translucent, dry film.

(3) Upon gelatin it will never grow in primary culture and seldom
in secondary culture. (4) It does not give rise to acid production
in milk or glucose broth. (5) It is non-pathogenic to guinea-pigs.
(6) The Neisser stain is negative. The fermentation reactions will
be found in the Table on p. 292.

In all probability the organism is not causative of xerosis con-
junctivse.

To isolate the organism, blood-serum tubes are inoculated with
a looped platinum needle from cases of follicular conjunctivitis or
xerosis and incubated at 37° C. for forty to forty-ei ght hours.
Half the tubes will usually show a growth. Preparations may be
stained with Loffler's blue and by Gram's method.

Bacillus coryzae (segmentosus)

An organism first described by Cautley, of frequent occurrence
in the nasal secretion in cases of " influenza " cold. It bears a
striking resemblance morphologically to the B. diphtheria when
stained with methylene blue, and is Gram -positive, but does not
show granules either with Loffler blue or with Neisser's stain. On
agar it grows more slowly than B. diphtherice, and in glucose broth
and litmus milk acid production is slow and feeble. It is non-
pathogenic to guinea-pigs. The fermentation reactions will be
found in the Table on p. 292.

298 A MANUAL OF BACTERIOLOGY

Other Diphtheria-like Bacilli

As already mentioned, diphtheria-like bacilli are not infrequent
in wounds, pathological discharges and secretions. Some of them
may be positive with Neisser's stain. They are always non-virulent.
The fermentation reactions of some of these organisms will be found
in the Table on p. 292.

Bacillus diphtherias columbarum

Pigeon diphtheria is an infectious disease of pigeons, charac-
terised by the formation of diphtheritic -like membranes on the
tongue, fauces, and corners of the mouth ; occurs in extensive
epizootics from time to time. Loffler isolated a bacillus to which
he gave this name. It is short, with rounded ends, non-motile, does
not form spores, and does not stain by Gram's method. On gelatin
it forms a whitish growth without liquefaction, on agar a creamy
growth, and on potato a thin grey film. Milk is not curdled and is
unchanged in reaction. It is pathogenic for the mouse and pigeon,
but only slightly so for the fowl and guinea-pig. It is possible to
prepare a vaccine, and an anti-serum for the disease.1 Recent
research has, however, suggested that the disease may be due to a
filter-passer. 2

Diphtheritic roup of poultry is a different disease, and is stated
to be due to a protozoan parasite.3 Macfadyen and the writer 4
found Klebs-Loffler-like organisms to be present in the mouths and
throats of healthy pigeons and fowls. These organisms resembled
the true Klebs -Loffler bacillus in their cultural reactions, but were
quite non-virulent to guinea-pigs (see Table, p. 292).

The so-called diphtheria of calves is produced by an anaerobic
streptothrix.

1 See Ann. de rinst. Pasteur, xv, 1901, p. 952.

2 Dean and Marshall, Journ. of Path, and Bact., vol. xiii, 1908, p. 29.

3 See also Gordon Sharp, Lancet, 1900, vol. ii, p. 18.

4 Trans. Path. Soc. Lond., vol. li, 1900, p. 13, and Brit. Med. Journ.,
1900, vol. i, p. 994.

CHAPTER IX

" ACID-FAST " BACILLI

TUBERCULOSIS— LEPROSY— THE SMEGMA BACILLUS-
GLANDERS

11 Acid-fast " Bacilli

AN important characteristic of the tubercle, leprosy, smegma, and
certain other bacilli is the property they possess when stained
with fuchsin of retaining the red colour after treatment with a
strong solution of a mineral acid (25 per cent, sulphuric or 30 per
cent, nitric). They are therefore termed " acid-fast." Most other
organisms are rapidly decolorised even by 1 or 2 per cent, sulphuric
acid, but it must be recognised that several apparently saprophytic
bacilli are also " acid-fast." The retention of the fuchsin colour in
spite of treatment with the acid seems to be due to the presence of
substances of a fatty or waxy nature within the organisms with which
the fuchsin either combines or is protected from the action of the acid.

Moreover, by cultivating many saprophytic bacilli in media
containing butter, Bienstock and Gottstein converted them into
" acid -fast " forms.

" Acid-fast " bacilli are also present in Johne's disease, occasion-
ally in rats, in butter (Petri, Rabinowitsch, Rubner), on certain
Graminaceae (the " Timothy -grass bacillus " of Moeller), and in
dung (the " Mist bacillus "). It has been suggested that these
saprophytic acid-fast bacilli may be derived from the tubercle
bacillus, but Panisset's work gives no confirmation of this.

The StreptotricheaB occasionally exhibit " acid-fast " properties.
All the acid-fast bacilli seem to be Gram-positive.

Tuberculosis

Tuberculosis is, unfortunately, only too common in the
human subject, and most of the domestic animals and wild
animals in a state of captivity may be attacked by it.

299

300 A MANUAL OF BACTERIOLOGY

The conception of tuberculosis was originally a purely
anatomical one, the name being given to a condition in
which the organs were studded with little yellowish points
or nodules, which were termed tubercles. Laennec was
the first to indicate the characters of these nodules or
tubercles, and traced with considerable accuracy their
development from minute lesions, the miliary tubercles,
up to the large cheesy masses which may be met with in
the glands and lungs.

Microscopically, the structure of a young and typical
tubercle is characteristic. At the centre one or more giant-
cells are found — large protoplasmic masses, each containing
ten to twenty nuclei arranged round the periphery (Plate
IX. 6). They are of the nature of plasmodia, similar to
the masses of fused cells which surround a foreign body in
the lower animals (Adami). Around the giant- cells are
well-defined epithelial- like cells with large and distinct
nuclei, which are known as epithelioid, or more properly
endothelioid, cells. A zone of smaller cells with scanty
protoplasm and small nuclei surrounds the endothelioid
cells ; they are known as lymphoid cells from their likeness
to the cells of lymphoid tissue. This is the structure of a
typical tubercle, but one or other of the components may
be wanting, and none can be said to be absolutely charac-
teristic of the tubercle. The nodule possesses no blood-
vessels, and as its size increases by growth at the periphery
the central parts undergo degenerative changes, and may
become either structureless or hyaline, or be converted
into a soft yellowish material somewhat like cheese and
termed caseous. More or less extensive inflammatory
reaction ensues in the tissues surrounding the tubercle,
and the cellular elements so produced often become spindle-
shaped and ultimately fibrous, so that the tuberculous
nodule becomes enclosed by a capsule of fibrous tissue
which may contract and convert it into a fibrous nodule.

THE TUBERCLE BACILLUS 301

After caseation has occurred calcification may ensue —
that is, lime-salts are deposited and the nodule is converted
into a calcareous mass.

So far back as 1865 Villemin showed that inoculation
of rabbits with human caseous material was followed by a
development of nodules similar in all respects to the miliary
tubercles in man. Cohnheim, Burdon Sanderson, and
Wilson Fox confirmed this observation, but they also
showed that the development of tubercles apparently
followed the introduction, not only of tuberculous material,
but also of setons, pieces of putrid muscle, and gutta-
percha. It was pointed out, however, that in all proba-
bility these results were due to accidental contamination
or inoculation with tuberculous matter, and, by adopting
suitable precautions in order to prevent such sources of
error, it was conclusively shown that non-tuberculous
matter is unable to set up tuberculosis. Tuberculosis is
therefore inoculable, and is an infective disease, and as
such must be due to a specific infective agent, to the
discovery of which observers then directed their attention.
In 1882 Koch announced that he had discovered a special
bacillus, the tubercle bacillus, in tuberculous tissues, which
could be isolated and cultivated, and which reproduced the
disease on inoculation.

The Tubercle Bacillus

Morphology. — The tubercle bacillus (B. tuberculosis) is
a slender rod with rounded ends, often slightly curved, and
averaging 2-3 /m in length, though the length varies in
the tissues from 1*25 /JL to 6'5 JUL ; in cultures it tends to be
short, on serum being about 1 yu. In stained preparations
one or more unstained intervals are often seen in the rods
(Plate VIII. a) ; these have been considered by some
observers to be spores, but there are many objections to

302 A MANUAL OF BACTERIOLOGY

this view. Spores are usually single and not multiple,
and are regular spherical or ovoid bodies, whereas the
unstained spaces in the tubercle rods are irregular. More-
over, in the same specimen of sputum a varying amount
of " beading," as it is termed, may be brought out by
different staining methods (Plate VIII. 6) ; 'in a prepara-
tion stained by Gram's method it is usually more pro-
nounced than in one stained with carbol-fuchsin. In class
work also it will be found that one student's specimen will
show beading much more markedly than another's. These
considerations render it probable that the beading is partly
due to segmentation of the protoplasm, and partly, perhaps,
is an artifact due to the staining process, and is not a
spore formation. The tubercle bacillus, however, probably
does form spores, though this is a debated point. Some
observers have described clear, regular, unstained spaces
in bacilli from old cultivations, and consider these to be
true spores.

The tubercle bacillus is a non-motile, strictly parasitic
organism (it has been described as being both motile and
flagellated). It usually occurs singly, occasionally linked
in twos or threes so as to form short chains, and under
certain conditions, especially in old cultures, filamentous
forms develop, and Foulerton1 and others include it
among the Streptotrichece. The bacillus is agglutinated
by the blood-serum of a tuberculous animal (see p. 324).
There are several varieties of the tubercle bacillus (see
pp. 315 and 319).

Staining reactions. — The tubercle bacillus stains in-
differently with the ordinary watery solutions of dyes,
prolonged treatment with, or warming, the solution being
required. It stains well by Gram's method. It also
stains well and deeply with carbol-fuchsin, particularly on
warming, and when so stained is markedly resistant to
1 " Milroy Lectures," Lancet, 1910, vol. i, p. 551, et seq.

PLATE VIII.

<^ -

v?mfi "s*i

The tubercle bacillus. Film preparation of a pure
culture, x 1000.

b. Tubercle bacilli in sputum, x 1500.

THE TUBERCLE BACILLUS 303

the decolorising action of 25-30 per cent, mineral acid ;
that is to say, it is strongly " acid-fast," and this property
is made use of for demonstrating its presence in tissues,
etc., and for diagnostic purposes. This " acid-fastness "
is due to the chemical constitution of the bacillus (see
p. 309). In old and particularly healing lesions red-
staining granules may take the place of definite bacilli :
these are the " splitter " forms of Spengler.

Cultural characters. — The tubercle bacillus is aerobic
and facultatively anaerobic, and thrives best at a tem-
perature of 37° C. or thereabouts, but development even
then is slow, four weeks at least being required for an
appreciable growth. Primary cultivations from the lesions
cannot be obtained on ordinary culture media but should
be made on (a) Dorset's egg medium, (6) glycerinated
potato in Roux's tubes (Fig. 9), the bulb being filled
with 5 per cent, glycerin in physiological salt solution,
(c) glycerin brain agar, or (d) glycerinated serum (preferably
dogs'). Twort1 has successfully isolated the bacillus
from sputum by direct cultures in an ericolin medium.
Dorset's egg medium is prepared thus : the contents of
four eggs are well beaten up, 25 c.c. of water are added,
and the mixture is strained through muslin. The fluid is
then tubed, and the tubes are heated in the sloping position
to 70° C. for four hours. At the time of inoculation, a
drop or two of sterile water should be added. Brain agar
is prepared by making a 3 per cent, nutrient agar of -f 20
reaction, adding an equal volume of pounded ox-brain,
and sufficient glycerin to make 5 per cent, in the mixture,
and sterilising. Egg broth is also a good culture medium.

After culture on these media for some generations, the
tubercle bacillus will develop on 5 per cent, glycerin agar
(reaction + 15 or 20), and in 5 per cent, glycerin broth
(veal is best) ; it will also grow, though very slowly, on

1 Proc. Roy. Soc. Lond., B vol. Ixxxi, 1909.

304 A MANUAL OF BACTERIOLOGY

glycerin gelatin at 22° C. Gelatin and blood-serum are not
liquefied. On glycerin agar the growth forms a dry, crinkled
and wrinkled, cream-coloured or brownish-yellow film, which
has been well described as resembling the patches of lichen
met with on trees (Fig. 37). The growth,
however, varies considerably, both in
colour and in the amount of wrinkling,
though retaining more or less the char-
acteristics just mentioned. In broth it
forms soft, cream-coloured, flaky masses,
which increase slowly both in size and
number, the broth remaining perfectly
bright and clear. Sometimes a dry crink-
led film forms on the surface of the broth,
and may spread all over it, and tends to
creep up the sides of the vessel. This film
formation seems to be essential for the
preparation of a satisfactory old tuber-
culin, but it is necessary in order to start
it that some of the inoculated particles
should float and form nuclei from which
FIG. 37. — Tubercle the film spreads. The virulent organism
bacillus Glycerin- from the primary cultivations is difficult

agar culture three , .

months old. to grow on anything but glycermated

potato or serum, or brain agar.

TUBERCULINS. — Extracts of, and suspensions of tritu-
rated, tubercle bacilli, human or bovine, are employed
in treatment and for the diagnosis of tuberculosis. The
preparations are known as tuberculins.

Old tuberculin, T.A. — This is prepared by growing the
tubercle bacillus in glycerin veal broth in a shallow layer
in flat flasks (Fig. 38), so that there is a free supply of
oxygen. After some weeks an abundant growth with
copious film formation develops ; the latter seems to be
essential, but it does not appear to matter whether the

OLD TUBERCULIN 305

bacilli be virulent or non-virulent, or whether they be
of human or of mammalian origin. The cultures, bacilli
and all, are heated at 115° C. in the autoclave for half an
hour, then concentrated over a water-bath to about one
tenth of their volume, and finally are filtered through
porous porcelain ; the resulting fluid is thick, owing to
the concentration of the glycerin by the evaporation, is
of a dark amber colour, and possesses a curious charac-

FIG. 38. — Flask for growing tuberculin.

teristic smell. The large proportion of glycerin preserves
the fluid, which keeps indefinitely in a cool dark place.

This old tuberculin possesses remarkable properties.
Relatively large amounts (O1-O5 c.c.) may be injected
into a healthy animal or individual without effect, but
in a tuberculous one a minute dose, O001 c.c. or less, gives
rise to a marked reaction — elevation of temperature with
constitutional disturbance more or less severe, and swelling
and tumefaction of tuberculous lesions (glands, ulcers,
etc.), and this reaction is made use of for diagnostic pur-
poses (see p. 330). By cautiously increasing the amount
a toleration is gradually induced, so that considerable
doses cause little or no disturbance. Injections of tuber-
culin tend to produce marked changes in the tuberculous
parts, leading to necrosis and exfoliation, with subsequent
healthy reaction and repair. This is especially seen in
cases of lupus ; by continued injections a marvellous

20

306 A MANUAL OF BACTERIOLOGY

improvement results, so much so that a cure is apparently
effected ; but, unfortunately, when the treatment is
discontinued the scar usually breaks down and the disease
returns. Nevertheless, a few cases have remained perma-
nently healed.

For treatment, the dose to commence with should not
be more than O0001 c.c., dilutions being made with O5
per cent, carbolic solution, and the dose is repeated when
all reaction has passed away and is gradually increased.
Tuberculin R, or tuberculin BE, is now generally employed
(see below).

Healthy guinea-pigs bear considerable injections of
tuberculin without harm ; but if they be tuberculous, if
the disease is advanced (eight to ten weeks after inocula-
tion), doses of O01 c.c. produce death ; if less advanced
(four to five weeks after inoculation) a larger dose, O2
to O3 c.c., is required ; but O5 c.c. always proves fatal.
The post-mortem appearances are congestion of the
lymphatics and viscera, and dark red spots, from mere
points to the size of a hemp-seed, on the liver and spleen.
These are due to enormous engorgement of the capillaries
in the immediate neighbourhood of tuberculous deposits,
actual extravasations of blood being rarely found. The
hsemorrhagic-like spots on the liver are almost pathogno-
monic of death from tuberculin.

Absolute alcohol precipitates the active principle of
tuberculin in the form of a whitish flocculent precipitate
which chemically consists of proteoses. This precipitate,
re-dissolved, is made use of in the ophthalmic reaction
(p. 330). Tuberculin applied to the scarified skin also
gives a cutaneous reaction in tuberculosis (p. 330).

Tuberculin R, or TR, new tuberculin, is prepared from
young and virulent cultures of the tubercle bacillus. The
growth is collected, dried in vacuo, and triturated by
machinery. Of the triturated material, 1 grm. is treated

TUBERCULINS 307

with 100 c.c. of distilled water, and centrifuged. The
supernatant liquid is rejected, and the residue is collected,
dried, again triturated and centrifuged. The supernatant
liquid is carefully pipetted off and kept, while the residue
is again submitted to the same treatment, and the process
is repeated until no solid residue is left. The fluids are
then mixed, the solid content is estimated gravimetrically,
some glycerin is added, and the liquid is diluted to the
correct volume, so as to contain 2 mgrm. of solid matter
per cubic centimetre (not 10 mgrm. as formerly stated),
and for use is diluted with 20 per cent, sterile glycerin
solution.

Tuberculin R, according to Koch, possesses distinct
immunising properties, and causes neither reaction nor
suppuration.

For treatment of tuberculosis in man the initial dose is
equivalent to not more than TTT oV o o ~ TTTO TOO ~ 5 uW
mgrm. of solid matter, according to the nature of the case.
The doses are given subcutaneously at intervals of ten
to fourteen days, and the treatment may be controlled
in the earlier stages by opsonic determinations. According
to Latham, tuberculin may also be given by the mouth.
Cases of cutaneous or localised tuberculosis, and those in
which the opsonic index to tubercle is moderately reduced,
react best.

Tuberculin, bacillary emulsion (BE), is an emulsion of
the powdered bodies of tubercle bacilli in 50 per cent,
aqueous glycerin. The mixture is allowed to sediment
until all heavy particles have deposited, the milky super-
natant fluid is pipetted off, and standardised so as to
contain 5 mgrm. of solid matter per c.c. The dosage is
similar to that of tuberculin R.

Behring has prepared another tuberculin, tulase or TC,
by treating tubercle bacilli with chloral, which he states
has a marked curative action, and is better administered

308 A MANUAL OF BACTERIOLOGY

by the mouth than by subcutaneous inoculation. By
giving tulase to cows, the milk is said to acquire immu-
nising and curative properties which are transmitted to
those consuming it. Rosenbach's tuberculin is prepared
by growing the tubercle bacillus with the ringworm or-
ganism, Friedmann's is derived from a turtle tubercle
bacillus. Other tuberculins are also on the market, and
any tuberculin may be prepared with a human or with a
bovine strain of bacillus.

Chemical products. — The tubercle bacillus produces no
extra-cellular toxin. Crookshank and Herroun obtained
from glycerin broth cultures of the tubercle bacillus a
proteose and an alkaloidal body. The proteose was also
obtained from " perlsucht." Both the alkaloid and the pro-
teose (from both sources) produced a rise of temperature in
tuberculous guinea-pigs, while in healthy animals the former
caused a slight, and the latter a marked, fall in temperature.

De Schweinitz and Dorset x described chemical products
isolated from the tubercle bacillus grown in a special
glycerin-asparagin mixture. From the bacilli themselves
an acid body was isolated, probably teraconic acid, an
unsaturated acid of the fatty series. A certain amount
of the same body was also obtained from the special culture
medium, but only a trace from glycerin broth, in which
the bacilli had been cultivated, in the latter case not
because it was not formed, but because of the difficulty
of isolation. This acid seemed to produce on injection
depression of temperature and necrosis of the tissues
locally, possessed some immunising power, and may be
the substance producing caseation in the tuberculous
nodules. The bacilli extracted with hot water yielded an
albuminoid, which gave the tuberculin reaction. This
they regard as the fever-producing substance.

1 Med. Journ. N. Y., 1897, July 24, p. 195. Also Fifteenth Annual
Rep. Bureau of Animal Industry, U.S. A,, 189C

ACTION OF HEAT 309

Bulloch and Macleod l state that the acid-fast substance
of the tubercle bacillus is an alcohol. Hot xylol will
remove this substance from the tubercle bacillus, and
ether or 5 per cent, caustic soda that from the smegma
bacillus ; the organisms after this treatment are no longer
" acid-fast."

Maragliano states that toxic bodies are present in the
blood and urine of tuberculous individuals. Cellulose also
seems to be present in small amount in the bacilli (it has
also been found in tuberculous nodules).

Tubercle bacilli, living or dead, are with great difficulty
absorbed when in any quantity. The dead bacilli when
injected under the skin invariably cause suppuration, and
several months later it is still possible to detect in the pus
numerous bacilli which stain well ; introduced into the
circulation of rabbits they give rise to nodules in the lungs
similar to the tuberculous nodules produced by living
bacilli (Koch).

Action of heat and antiseptics on the tubercle bacillus.—
The thermal death-point of the bacillus has been the
subject of some controversy. Sternberg found that tuber-
culous sputum exposed for ten minutes to a temperature
of 90°, 80°, and 66° C. failed to infect guinea-pigs in inocu-
lation, while another specimen of the same sputum heated
for ten minutes to a temperature of 50° C. produced tuber-
culosis in a guinea-pig, so that from these experiments the
thermal death-point lies between 50° and 66° C.

Yersin in 1888, by culture methods, failed to obtain
any growth from bacilli which had been heated to 70° C.
for ten minutes, while those heated to 55° C. and 60° C.
gave growths in glycerin broth in ten days and twenty-two
days respectively. Macfadyen and the writer, in the
course of some experiments on the sterilisation of milk
found that milk to which powdered dried sputum had been

1 Journ. of Hygiene, vol. iv, 1904, p. 1.

310 A MANUAL OF BACTERIOLOGY

added was rendered innocuous by a momentary heating
to 67°-68° C. These experiments indicate that a tem-
perature of 65° C. and over is probably rapidly fatal to
the tubercle bacillus, so that milk which has been pas-
teurised (i.e. heated to 68°-70° C. for twenty to thirty
minutes) may be regarded as quite safe. Experiments
by the Royal Commission on Tuberculosis with virulent
tuberculous milk gave somewhat irregular results ; in
one instance heating to 65° C. for two and a half minutes
rendered the milk innocuous, in another instance after five
minutes at 70° C. it was slightly virulent, but twelve minutes
at the same temperature rendered it inert (see also section
on " Milk "). Foulerton found that emulsified tuberculous
material from tuberculous guinea-pigs did not lose its power
of infecting unless heated to 70° C. or over for ten minutes.
The tubercle bacillus offers considerable resistance to
the action of antiseptics and germicides. Yersin found
that it was killed by 5 per cent, carbolic acid in thirty
seconds, by 1 per cent, in one minute, by absolute alcohol
in five minutes, and by mercuric chloride, 1-1000, in ten
minutes. Crookshank found that tuberculous sputum
mixed with an equal volume of 5 per cent, carbolic was
rendered innocuous in a few minutes, and this without
any special precautions as to breaking up the masses.
For disinfecting sputum mercuric chloride is unsuitable.
(See also Chap. XXI.)

Pathogenesis, etc. — Man is, unfortunately, only too fre-
quently attacked with tuberculosis, the manifestations
of which tend to differ somewhat at different age periods.
Thus, in the very young, general miliary tuberculosis,
tuberculous meningitis, and tuberculous disease of the
peritoneum, intestine, and mesenteric glands (tabes mesen-
terica) are the commonest ; in older children, up to the
age of puberty, the lymphatic glands, especially in the
neck, joints and bones, and the skin (lupus) are mostly

PLATE IX.

a. Tubercle bacilli in sputum, x 1000.

i

*v

• <^k • ..r

*•% ^r

il+ t

« • ^ &

&. Giant-cell in a tubercle containing tubercle bacilli, x 1000.

DISTRIBUTION OF BACILLI 311

attacked ; young adults suffer from disease of the lung
(consumption, phthisis), and older people from chronic
disease of the lung and tuberculous disease of the urinary
organs and testes, and of the suprarenal capsules (Addi-
son's disease). Scrofula and struma were terms formerly
much employed ; both denote a swollen neck, and were
applied to cases suffering from chronic tuberculous inflam-
mation with enlargement of lymphatic, especially of the
cervical, glands, with which other conditions, such as
inflammations of the ear, throat and eye, and implication
of bones and joints, are frequently associated.

The distribution of the bacillus in the tissues varies
considerably. In young and active tubercles the bacilli
are more plentiful and more easily demonstrated than in
older and more chronic ones. They tend to be more
numerous in some animals than in others — in the ox and
horse than in man, for example. In man the bacillus is
difficult to demonstrate (by staining) in enlarged and
caseating glands, in pus, in synovial membranes, and in
lupus. In some animals, especially the ox and horse,
bacilli can usually be readily demonstrated, and may be
present in large numbers, and frequently have the typical
distribution, viz. within and at the periphery of the giant-
cells, though they are by no means confined to this locality
(Plate IX. b).

It was asserted, particularly by Rosenberger and For-
syth, that tubercle bacilli can be detected in the blood in
the majority of cases of pulmonary tuberculosis. Hewat
and Sutherland,1 however, made twenty-two blood exami-
nations on twenty patients in all stages of the disease and
in only one detected two acid-fast bacilli. Schroeder and
Cotton tested the blood of forty-two cattle in all stages of
tuberculosis by inoculation into guinea-pigs with negative
results.

1 Brit. Med. Journ., 1909, vol. ii, p. 1119 (References).

312 A MANUAL OF BACTERIOLOGY

Tuberculosis in animals. — The majority of the domestic
animals are subject to tuberculosis. It is most common
in the ox, pig, and horse, much less so in the sheep and
goat, cat and dog. Wild animals, both mammals and
birds, in a state of captivity are also specially prone to be
attacked, and a large number of the deaths in Zoological
Gardens, particularly among the apes, are due to this
disease.

In the ox the tuberculous lesions are most frequently
met with in the lymphatic glands and serous membranes,
particularly the pleura, and in the lungs and liver, while
the fat and muscular tissues, which constitute the major
part of " meat," are very rarely affected. On the pleura
the growths take the form of nodular masses, which from
their arrangement are popularly termed " grapes " or
" angle berries."

In carp, tubercle-like nodules are occasionally met with
in which a bacillus resembling the tubercle bacillus in
morphology and staining reactions is present. It grows,
however, much more freely than the true tubercle bacillus,
and though inoculable into fish and frogs, is non-inoculable
into warm-blooded animals. But it yields a tuberculin
which reacts with mammalian tuberculosis, and by feeding
carp on the mammalian tubercle bacillus this can apparently
be transformed into the piscian variety.1

Bird or avian tuberculosis undoubtedly differs in many
respects from mammalian tuberculosis. The tuberculous
new formations may be very large, but do not show nearly
such a disposition to caseation or suppuration as the
human lesions. Epithelioid cells form the major part of
the growth, and giant-cells are very infrequent. One
remarkable feature is the enormous numbers of bacilli
which may be present in the tissues ; in places they may
be so numerous and closely packed as to form distinct

1 See Himmelberger, Centr. f. Bakt., Abt. I (Orig.), vol. 73, p. 1.

AVIAN TUBERCULOSIS 313

masses or nodules. The bacilli of avian have the same
staining reaction as those of mammalian tuberculosis, but
on cultivation and inoculation various differences between
the two races become evident. Rats, guinea-pigs, and
rabbits are practically insusceptible to inoculation with
the avian bacillus.

The mammalian bacilli flourish best at about 37° C.,
and growth ceases at 41° C., whereas the avian bacilli
thrive luxuriantly at 43° C., and the growth of the latter
on glycerin agar is much moister and more wrinkled, and
often more pigmented, than that of the former. Fowls
and dogs are with difficulty infected with human bacilli,
but dogs are susceptible to infection with avian bacilli.
By cultivation on boric-acid agar and on eggs, etc., the
mammalian bacilli are stated to assume the characters of
the avian.

Avian tuberculosis is of practical importance not only
as attacking poultry, but also in human pathology, as
several cases have been recorded in which the bacilli
cultivated from human cases seemed to be of the avian
type, and were therefore probably derived from an avian
source of infection. Two types of tuberculosis also occur
in the horse — one in which the lesions are chiefly abdominal,
in the other the lungs and bronchial glands are most
affected. Nocard states that the bacillus obtained from
the pulmonary variety is generally of the ordinary mam-
malian type, while that of the abdominal one belongs to
the avian.

Relation of human and bovine tuberculosis. — It was
noticed long ago that there are certain differences between
the bacilli of human and of bovine tuberculosis, the latter
tending to be shorter and thicker and less readily culti-
vated than the former ; also, whereas human tuberculous
material injected into a rabbit generally produces small
discrete lesions which tend to retrogress, bovine material

314 A MANUAL OF BACTERIOLOGY

induces a progressive disease with large caseating masses.1
These distinctions were regarded as being due to variations
in the bacilli as a result of growing upon a different soil
and not to any fundamental difference between the two
strains of bacilli. In 1901, however, Koch stated 2 that
young cattle and swine cannot be infected with human
tuberculous material, and he therefore concluded that
human and mammalian tubercle bacilli are essentially
different. As a result of his experiments he made the
statement that " though the important question whether
man is susceptible to bovine tuberculosis at all is not yet
absolutely decided, if such a susceptibility really exists,
the infection of human beings is but a very rare occurrence."
This view met with considerable opposition, and a
second Royal Commission was appointed to investigate
the question, and the following summarises the results
obtained up to the present, from which it will be gathered
that while there is no justification for assuming that man
is infected from human sources alone, infection from
human sources is probably vastly more frequent than
from any other. Thirty different viruses isolated from
cases of tuberculosis occurring spontaneously in bovines
have been studied, and the results of introducing them
into a number of different animals by feeding and inocula-
tion are recorded. In calves, inoculation usually results
in generalised progressive tuberculosis, but the effect is
somewhat dependent on the dose, i.e. the number of bacilli,
administered. Thus whereas 50 mgrm. of culture always
induced a fatal generalised progressive tuberculosis, in
two instances much smaller doses — 0'01-0'02 mgrm. —
produced only limited retrogressive tuberculosis. Feeding,
on the other hand, usually produced lesions limited to

1 The bacill derived from tuberculosis of the sheep, pig, and horse
(pulmonary lesions) are also of the bovine type.

2 See Brit. Med. Journ., 1901, vol. ii, p. 189.

BOVINE TUBERCULOSIS 315

the neighbourhood of the digestive tract, which generally
retrogress and become calcareous. The bovine bacillus,
when introduced into rhesus monkeys or chimpanzees,
either by inoculation (even in so small a dose as 0-001
mgrm.) or by feeding, induces rapid generalised tuberculosis,
and, considering the close relation that exists between the
anthropoid apes and man, these results are of the highest
importance. In pigs, generalised progressive tuberculosis
is readily set up both by feeding with, and by the inocula-
tion of, bovine bacilli. Goats, dogs, and cats are relatively
less susceptible, but more or less tuberculous infection can
similarly be produced in them. On this part of the inves-
tigation the Commissioners remark that the bacillus of
bovine tuberculosis is not so constituted as to act on
bovine tissues only, and the fact that it can readily infect
the anthropoid apes, and, indeed, seems to produce this
result more readily than in the bovine body itself, has an
importance so obvious that it need not be dwelt on. The
viruses isolated from sixty cases of the disease in man
were also studied, and the results obtained show that they
may be divided into two groups, subsequently referred to
as Group I and Group II. The bacilli of Group I com-
prised fourteen viruses, one obtained from sputum, three
from tuberculous cervical glands, and ten from mesenteric
glands of primary abdominal tuberculosis in children.
The results produced by introducing these viruses into
animals are identical with those produced by the bovine
bacillus. The bacilli of Group II comprised forty viruses
obtained from various forms of human tuberculosis — cer-
vical glands, mesenteric glands (8), lungs and bronchial
glands (10), joint and bone disease (9), testis, kidney, etc. — •
grow more luxuriantly in culture than those of Group I,
and inoculated into calves and rabbits do not produce
the generalised and fatal disease caused by the bovine
bacillus, but in rhesus monkeys and in the chimpanzee

316 A MANUAL OF BACTERIOLOGY

set up a general tuberculosis. Certain human viruses,
differing in certain respects from those of Groups I and
II, were also met with and are classed as Group III, but
an opinion on their significance is reserved for a future
report.

The Commissioners conclude that the tubercle bacillus
in its nutritive and reproductive powers resembles other
simple organisms, and that the essential difference between
one strain and another depends on variations in these
factors, and they classify those bacilli that grow with
difficulty on artificial media as dysgonic, and those that
grow readily on media as eugonic.

As regards the histological appearances of the tuber-
culous process in different animals, Eastwood states that
there is an underlying unity of the morbid processes
produced experimentally by infection with every variety
of bovine and human tubercle bacillus.

In their final Report, the Commissioners conclude that
an appreciable amount of human tuberculosis is caused
by bacilli of the bovine type, and that tuberculosis may be
communicated to man from infected cow's milk, and from
tuberculous meat, either beef or pork.

So far, therefore, from any relaxation of the existing
supervision of milk-production and meat-preparation being
possible, the Commissioners press upon the Government
the enforcement of food regulations, " planned to afford
better security against the infection of human beings
through the medium of articles of diet derived from tuber-
culous animals." More particularly they urge such action
" in order to avert or minimise the present danger arising
from the consumption of infected milk."

Of young children who died of wasting disease of the
intestine, the bovine bacillus was present in nearly half
the cases. Further, a large proportion of cases of tuber-
culous cervical glands in both children and adults was

CHANNELS OF INFECTION 317

due to the same bacillus. The wording of the report is :
" Whatever, therefore, may be the animal source of tuber-
culosis in adolescents and in adult man, there can be no
doubt that a considerable proportion of the tuberculosis
affecting children is of bovine origin, more particularly
that which affects primarily the abdominal organs and
the cervical glands. And further, there can be no doubt
that primary abdominal tuberculosis, as well as tubercu-
losis of the cervical glands, is commonly due to ingestion
of tuberculous infective material. The evidence which
we have accumulated goes to demonstrate that a con-
siderable amount of the tuberculosis of childhood is to
be ascribed to infection with bacilli of the bovine type
transmitted to children in meals consisting largely of the
milk of the cow.

" We are convinced that measures for securing the
prevention of ingestion of living bovine tubercle bacilli
with milk would greatly reduce the number of cases of
abdominal and cervical gland tuberculosis in children,
and that such measures should include the exclusion from
the food supply of the milk of the recognisably tuberculous
cow, irrespective of the site of the disease, whether in the
udder or in the internal organs."

Eber,1 in an extended investigation, succeeded in infect-
ing calves from three cases of human pulmonary tuber-
culosis. The bacilli isolated from the human material were
of the human type, but after passage through the calf
became transformed into the bovine type. He affirms,
therefore, the essential identity of the human and bovine
types of tubercle bacilli.

With regard to the channel of infection in human tuber-
culosis opinions differ. Koch insisted that inhalation of
air- borne bacilli derived from dried human sputum is the
principal source of infection ; Von Behring, on the other

1 Centr.f. BakL, Abt. I (Orig.), lix, 1911, p. 193.

318 A MANUAL OF BACTERIOLOGY

hand, expressed the opinion that tuberculous milk fed to
children is the main source of infection both of children
and of adults ; in the latter case he suggested that bacilli
are ingested in childhood and lie dormant for years before
becoming active.

Calmette similarly believes that in the young infection
by the digestive tract, especially by tuberculous milk, is
the more frequent, and attaches little or no importance to
dry dust containing tubercle bacilli as a source of infection.
Ravenel considers that the alimentary tract, particularly
in children, is a frequent portal of entry for the tubercle
bacillus, which he believes is able to pass through an intact
mucous membrane. Of sixty cases of human tuberculosis
investigated by the Royal Commission on Tuberculosis,
twenty- eight possessed clinical histories indicating that
in them the bacillus might have been introduced by the
alimentary canal. Eraser has also directed attention to
the frequency of the bovine type of bacillus in the tuber-
culous lesions of bone and joints in children.

Fliigge, on the other hand, states that his experiments
show that tuberculosis can be communicated to animals
by inhalation, and that the dose of bacilli required to
infect by the respiratory tract is far less than that required
to infect by the alimentary canal. The mode of infection
in man doubtless varies, and he believes that children
may be infected by the digestive tract, by tuberculous food,
particularly milk, but the most extensive source of infection
is the number of droplets of tuberculous expectoration
coughed up by consumptives ; these float in the air and
serve as sources of infection to others. Ribbert and
Schrotter, also, from the evidence of autopsies, considered
inhalation as the chief mode of infection in man.

Bulloch,1 from a careful survey of the literature, con-
cludes that pulmonary tuberculosis is invariably caused
1 " Horace Dobell Lecture," 1910.

COMMISSION ON TUBERCULOSIS 319

by bacilli of the human type, and, therefore, is presumably
due to inhalation of human bacilli.

McFadyean,1 also, from a critical survey of the experi-
mental evidence, concludes that (1) inhalation of tubercle
bacilli suspended in the air is a very certain method of
infecting susceptible animals ; (2) experimental infection
by the digestive tract is comparatively difficult to realise ;
(3) inhalation is probably the commonest natural method
of infection, both in man and in animals ; (4) infection
by the digestive tract can be inferred only when the
lesions are confined to the abdomen. He finally states
that " the whole of the experimental evidence on which
the theory of the intestinal origin of pulmonary tuber-
culosis in man was built up has been swept away."

While the death-rate per 1000 living from all forms of
tuberculosis is about 1-64, that from phthisis is 1-14, so
that the greater part of the mortality from tuberculosis
must be ascribed to infection from human sources. There
still remains the residuum of glandular, abdominal, bone
and joint tuberculosis which is ascribed to infection with
the bovine bacillus. The experiments of the Royal
Commission on Tuberculosis favour this view, but an
alternative explanation is possible. Thus Spengler, Klem-
perer, and Baumgarten from direct experiments on man
assert that the bovine bacillus is not pathogenic to man,
and Spengler distinguishes two types of human tubercle
bacilli, (a) the " humanus brevis," the ordinary human
type, and (b) the " humanus longus " type. The latter
is very like the bovine bacillus in pathogenic action, but
after carefully weighing all the facts, Spengler considers
the true " bovinus " and the " humanus longus " types are
not identical. It may be then that the bacillus found in
certain human lesions and considered to be the bovine
variety is really this humanus longus variety. Even

1 Jaurn. Roy. Inst. Public Health, vol. xviii, 1910, p. 705.

320 A MANUAL OF BACTERIOLOGY

admitting that the bovine bacillus does infect man, it by
no means follows that all such cases of infection are derived
from a bovine source, for humans might infect one another
with the bovine bacillus ; this possibility never seems to
be considered.

The occurrence of tuberculosis in the domestic animals raises
points of practical importance, especially the occurrence of infection
from the consumption of meat and milk from diseased animals.
There can be no doubt that the carcase of an animal extensively
affected with tuberculosis, especially if wasting has occurred, should
be condemned as unfit for food, and likewise all parts in which
there are tuberculous deposits. But it becomes an important
question for the community, financially as well as from a hygienic
point of view, as to the method of procedure with the meat from a
beast comparatively slightly affected with tuberculosis — an enlarged
gland or two, and a few nodules on the pleura. No doubt the ideal
method in such a case is the condemnation and destruction of
the whole carcase, be the amount of tubercle ever so little ; but
from financial considerations this procedure is hardly practicable
on account of the large amount that would have to be paid in
compensation. Experiment has demonstrated that the tubercle
bacilli are practically confined to the tuberculous areas and are
extremely rarely met with in the muscular tissue, and these portions,
therefore, it might seem, could be eaten with impunity, especially
as they would be cooked before consumption. As regards swine,
however, it is generally held that tuberculosis anywhere condemns
the whole carcase.

The report of the first Royal Commission on Tuberculosis, how-
ever, indicated two dangers. Firstly, in cutting up a carcase the
butcher will most likely use the same knife throughout, and in
this way may infect the meat with tuberculous matter by smearing
with the knife. Secondly, cooking cannot be depended upon to
destroy the bacilli unless the joints are under 6 Ib. in weight ; when
the weight is above this the temperature in the interior may not
rise sufficiently high. Evidently one of the first measures to be
taken is the abolition of private slaughter-houses and the establish-
ment of municipal abattoirs where the meat would have to be
passed by competent inspectors. In this way all badly affected
carcases would be condemned, and those only slightly affected could
be separately dealt with and special precautions taken to eliminate
tuberculous pieces, etc.

TUBERCULOUS MILK 321

Tuberculous milk also raises many important points. Probably
some 10-15 per cent, of all samples are infective to guinea-pigs, but
this does not necessarily indicate that this proportion would be
dangerous to man, for the material is introduced into the guinea-
pigs by inoculation after concentration by centrifuging (see also
section on " Milk "). Tubercle bacilli are present in milk not only
when the udder is tuberculous, but also when the cows are suffering
from tuberculosis elsewhere which is clinically recognisable. Thus,
when the lungs are affected, bacilli are disseminated from the air-
passages and also by the faeces. It is noteworthy that the incidence
of abdominal tuberculosis in young children occurs just when cow's
milk is the staple article of their diet. At the same time this
incidence does not seem to fall on those who consume most milk.

Much might be done by the registration of all dairy premises,
the use of selected cows, the elimination of all tuberculous animals,
and by enforcing the inspection of dairy cattle by competent
inspectors at suitable intervals. The notification of all forms of
udder disease is now compulsory. In the absence of inspection
and the use of selected cows, treatment of milk intended for the
food of infants and young children by pasteurisation or sterilisation
has been recommended, but has disadvantages (see section on
" Milk "). The ideal method, and one which commends itself at
first sight as being the most satisfactory, is the elimination by
slaughter of all animals which are tuberculous. This was adopted
in the State of Massachusetts ; under an order of the Board of
Cattle Commissioners all beasts in the State were tested with
tuberculin, and every animal that reacted was slaughtered, and
strict quarantine combined with the tuberculin test imposed on all
imported cattle. Even in this small State such a plan was found
to be unworkable, the expense of compensation becoming formid-
able. A middle course seems to be the only practicable one, viz.
all manifestly tuberculous animals, especially where wasting or a
tuberculous udder is present, to be slaughtered ; other animals to
be tested with tuberculin, and those which react to be separated
from the healthy and to be disposed of (foi slaughter) as soon as
convenient, and in the meanwhile kept as much as possible in pasture.

Avian tubercle bacilli have occasionally been met with in man. x

Tuberculosis is diminishing among the white races ; it is, how-
ever, spreading among many coloured races. It is to be noted that
the decline began long before the germ origin had been demonstrated,
and, what is more, the rate of decline was almost as great before
any administrative measures were taken against it as since. Never -

1 Lowenstein, Wien. Klin. Woch., May 15, 1913.

21

322 A MANUAL OF BACTERIOLOGY

theless, it can hardly be doubted that measures should be adopted
by local authorities and others to prevent the spread of tuberculosis.
All forms of tuberculosis have now been made notifiable in this
country. Patients should be warned of the danger of disseminating
their expectoration, and should use pocket-spittoons containing an
antiseptic, or handkerchiefs (such as the Japanese paper ones)
which can be destroyed. Rooms which have been inhabited by
tuberculous patients should be disinfected, for which purpose
Delepine recommended spraying with a 1-100 solution of chloride
of lime. Although the occurrence of direct infection can rarely be
proved, the possibility of this cannot be ignored. Not only should
the dissemination of infection be prevented, but the resistance of
the individual should be raised by providing a healthy environment
and by inculcating the importance of fresh air.

Serum therapeutics and vaccine. — Many sera have been
introduced for the treatment of tuberculosis, e.g. Marag-
liano's, Marmorek's, Spengler's, Mehnarto's, etc. Speng-
ler's I.K. serum is of considerable value in many cases :
it is prepared by immunising rabbits by intramuscular
injections and contains the laked red- corpuscles as well
as the serum.1 Mehnarto's is stated to be a mixture of
sheep and snake serums and is reported favourably on by
Barcroft.2

For vaccine treatment, tuberculins R and BE are usually
employed (p. 306). Latham has found that tuberculin
given per os produces its characteristic effects.

Immunity. — Attempts have been made from time to
time to produce immunity against the B. tuberculosis,
particularly in cattle. Thus McFadyean 3 found that
heifers which had previously been subjected to repeated
doses of tuberculin (old) in some cases resisted infection
with virulent bacilli. Behring 4 also employed human
tubercle bacilli for the vaccination of cattle with satis-

1 See Treatment of Tuberculosis by Immune Substances (I.K.) Therapy.
Fearis (John Murray, 1912).

2 British Journ. of Tuberculosis, 1913.

3 Trans. Path. Soc. Lond., vol. liii, 1902, p. 20.

4 Brit. Med. Journ., 1906, vol. ii, p. 577.

COMPLEMENT FIXATION IN TUBERCULOSIS 323

factory results. His tulase likewise confers immunity
when given either by the mouth or by the stomach.

Theobald Smith x also concludes that vaccination of
calves with the human type of bacillus is harmless, and
that the procedure leads to a relatively high resistance to
fatal doses of the bovine bacillus.

Clinical Examination

I. The " complement-fixation " test was first used in tuberculosis
by Wassermann and Briick. The method has been further elaborated
by Emery.2 He makes use of a standard emulsion of tubercle
bacilli in salt solution, containing about 4 per cent, by volume of
solid bacillary substance. This is sterilised by intermittent sterilisa-
tion and keeps for four to six weeks. Bacilli from various sources
vary somewhat, so that the emulsion should be standardised so as
to give an absorption-time with normal sera of about 20 minutes,
i.e. the complement of normal serum should be just completely
absorbed in about 20 minutes. A water-bath kept at a constant
temperature of 38° C. is used to warm all the constituents and mix-
tures. One part of the serum to be tested is mixed with four parts
of the bacillary emulsion in a small tube (e.g. a Durham's tube) in
the water-bath, the time of mixing being accurately noted. After
2^ minutes' incubation, 4 volumes of the mixture are removed by
means of a capillary pipette with teat (Fig. 35, p. 215), into which
also a single volume of sensitised corpuscles (i.e. a hsemolytic
system, p. 184) is taken up and the whole is expelled into a small
tube already standing in the water-bath. The process is repeated
after 5, 10, 15, and 20 minutes, and longer if necessary. By the
occurrence or absence of haemolysis in the various tubes, the time
taken for the absorption of complement is ascertained, the comple-
ment used being that contained in the serum itself, which therefore
should be fresh. A control with normal serum should always be
performed at the same time. With normal serum complete absorp-
tion should take place in about 20 minutes ; with tuberculous sera
it is often complete in 2| minutes. If, then, absorption of comple-
ment is complete in much less than the time necessary for absorption
with a normal serum, presumably the serum is derived from a
tuberculous individual. (But see Emery's paper for limitations.)

1 Journ. Med. Research, vol. xviii, 1908, p. 451.

2 Lancet, 1911, vol. i, p. 485.

324 A MANUAL OF BACTERIOLOGY

II. Precipitin reaction. — Spengler has devised a precipitin
reaction for the diagnosis of, and prognosis in, tuberculosis. The
reagents are the blood-serum or the laked whole blood, or both,
very highly diluted and mixed in different dilutions with tuber-
culin.1

III. Agglutination reaction. — The method of agglutination was
proposed by Arloing and Courmont for the diagnosis of tuber-
culosis, but is difficult to carry out and is not much employed.
A special method has to be employed to obtain homogeneous
cultures of the tubercle bacillus or a powder of pulverised or
ground-up bacilli may be used : this powder may be purchased.
The reaction may be carried out either microscopically or macro-
scopically ; for the latter small sterile test-tubes may be employed.
For each test three dilutions of the serum are made, a 1 in 5, a
1 in 10, and a 1 in 20, and the tubes filled with these dilutions are
allowed to stand in an inclined position (45°) for five to ten hours.
In man the serum of normal individuals may agglutinate up to a
dilution of 1 in 5, while in animals this is variable — imperceptible
in the guinea-pig, rabbit, and calf ; feeble in the goat ; in the adult
ox up to 1 in 5, but in the dog it may be up to 1 in 10 or even 1 in 20.

A positive serum reaction in a suspected subject is a sign of great
value in establishing the diagnosis ; a negative serum reaction is
of less value.

IV. The examination of sputum, etc., for the tubercle bacillus is
a routine procedure of the greatest value in forming a diagnosis.
Fortunately, owing to the peculiar staining reaction of the tubercle
bacillus, the method is comparatively simple.

If it is inconvenient to examine the sputum for a day or two a
little 1-20 carbolic should be added. This preserves the sputum
and the tubercle bacilli retain their staining power for some time.

1. Sputum. — Film specimens are prepared by smearing a little
of the sputum on to a slide with a needle so as to form a thin film
covering two-thirds of the surface, or by placing a particle of the
sputum on one slide, applying another slide, pressing together, and
then drawing apart so that a thin film is left on each slide. The
thick portion of the sputum should be used, the thin mucoid portion
being rejected. If the sputum is thin and watery, the thicker
portion can be obtained by covering the bottom of a Petri dish
with filter-paper, placing a large drop of the sputum on this, and
working it over the paper with a bent steel needle. The paper
absorbs the water, leaving the thicker material on the surface. If
there are any small yellow caseous particles present these should
1 See Fearis, Practitioner, i, 1913.

ZIEHL-NEELSEN METHOD 325

be chosen, and sufficient material should be used so as to form a
distinct but not too thick film ; a little experience will soon decide
the right amount ; too thin a film should be avoided. Preparations
may also be made by smearing the sputum on a cover-glass or
between two cover-glasses instead of using slides. Whichever plan
is adopted, the film is dried and fixed in the usual manner (generally
by heat), and then stained by one of the following methods :

(a) Ziehl-N eelsen mzthod. — Film specimens on slides are most
conveniently stained by flooding with filtered, undiluted carbol-
fuchsin and warming for 2 to 5 minutes on a piece of asbestos
cardboard supported on a tripod, or on a heated penny (p. 110), or
slides or cover-glasses flooded with the stain may be held in the
forceps and carefully warmed over a flame, or the preparations
may be immersed in a watch-glass or dish of the stain, covered, and
placed in the warm incubator for half an hour. In no case must
the stain bs allowed to boil, or the bacilli may lose their staining
power ; it should only bs warmed sufficiently to steam (50°-60° C.),
and with slides or cover-glasses as evaporation takes place more
stain (always filtered), or better, 5 per cent, carbolic, should be added.
After staining, the preparations are rinsed in water and are then
decolorised by treating with 25 per cent, sulphuric or 30 per cent,
nitric acid. The preparation may be flooded with the acid, but a
better method is to immerse the preparation in a pot (Fig. 20,
p. 110) containing the acid. In the acid the colour changes after
a few seconds to a yellowish brown, the preparation is then rinsed
in water, and some of the pink colour returns. The treatment with
acid and with water alternately is repeated until the preparation
is nearly colourless when rinsed in water. With sputum this is
usually the case after three or four rinses in the acid, but it varies
with the thickness of the film and with the number of tubercle
bacilli present ; when these are absent the film often decolorises
more readily than when there are many. The presence of blood
renders the decolorisation difficult. After decolorising and washing,
the preparations are stained for one minute in Loffler's methylene
blue, washed in water, and mounted in water, or, better, dried and
mounted in Canada -balsam or cedar oil. When the preparation is
made on the slide, after washing and drying, it can be examined
directly without a cover-glass with the oil-immersion after applying
a drop of cedar oil, unless a permanent specimen is desired, in which
case it should be mounted in Canada-balsam.

The tubercle bacilli appear as delicate red rods, often beaded or
segmented, on a blue background composed of cells, mucus, and
putrefactive or other bacteria. Occasionally here and there a little

326 A MANUAL OF BACTERIOLOGY

red colour may be present in addition to the tubercle bacilli. Hair
and keratinised material generally, such as horny epithelium, and
red blood-corpuscles, retain the red colour after the foregoing
treatment, and the spores of bacteria are also liable to retain the
red somewhat persistently. These exceptions are not, however,
likely to prove a source of error, for the tubercle bacilli should be
recognised not only by their red colour, but also by their charac-
teristic size, shape, and general appearance. It is conceivable that
acid-fast bacilli not tubercle might be present in sputum, but such
an event is a very unlikely one. For the microscopical examination,
a ^-inch with good illumination is sufficient when the tubercle bacilli
are present in any number. When they are scanty it is necessary
to use a yV'inch oil-immersion, and this is the better lens in any
case. (See" Plate IX, b, and Plate X, a.)

If tubercle bacilli are not found, other specimens should be pre-
pared and examined. It is only by repeated examinations on different
occasions that the negative evidence, the absence of tubercle bacilli,
becomes of any value.

The tubercle bacillus is occasionally not acid-fast ; 1 probably
the bacilli in such cases are degenerate, and, like all degenerate
bacteria, fail to stain well. Spengler claims that the following
method will stain these and "splitter" forms: (1) Stain with
warm carbol-fuchsin by the ordinary method, avoiding overheating ;
(2) pour off the stain without washing and treat with picric acid
alcohol (equal parts of saturated aqueous picric acid and absolute
alcohol) ; (3) after 3 seconds rinse with 60 per cent, alcohol ;

(4) treat with 15 per cent, nitric acid until yellow (about 30 seconds) ;

(5) rinse again with 60 per cent, alcohol ; (6) counter-stain with
the picric acid alcohol until yellow ; (7) wash with distilled water.
This is an excellent method, and thick films may be used. In
material which has been preserved a long time, e.g. sputum with
carbolic, or tissue in spirit, the bacilli may be much less acid-fast
than in fresh material.

Various methods have been recommended for the solution of
the sputum and the examination of the sediment of the bacilli.
In one method 5 c.c. of sputum are mixed with 50 c.c. of normal
KOH solution ; the mixture is warmed in a water-bath to 60°-65° C.
until the sputum is dissolved (about 3 hours) ; 50 c.c. of cold water
are next added, the whole is well shaken, and again warmed for
^ hour. Petroleum ether, 2 c.c., is next added, the whole is well
shaken, and is then kept at 60° C. until the ether has separated.
The bacilli will be concentrated in the fluffy layer at the junction
1 See Lancet, 1908, vol. i, p. 1222.

MUCH'S METHOD 327

of the ether and water ; this is pipetted off and films are made with
it and stained. Antiformin (a mixture of sodium hypochlorite and
sodium hydrate) has also been recommended. Into a boiling- tube
or small flask of 50 c.c. capacity, 5 c.c. of the sputum are introduced.
To this are added 25 c.c. of antiformin solution (10-20 per cent,
aqueous solution) diluted with 10-20 c.c. of water according to the
density of the sputum. The mixture is well shaken until homo-
geneous (about 15 minutes), then centrifuged, the deposit is washed
three times with salt solution by centrifuging, and films are made
with the washed deposit and stained by the Ziehl-Neelsen or
Spengler method.

If the tubercle bacillus cannot be detected microscopically after
repeated examinations, and a certain diagnosis is important, the
inoculation method may be employed. A couple of guinea-pigs are
inoculated subcutaneously in the thigh or abdomen with 0-5 to 1 c.c.
of the sputum. If tubercle bacilli are present the animals will
show signs of tuberculosis in three to six weeks (see below,
" Urine ").

(6) Other methods have been devised for staining the tubercle
bacillus, but do not seem to be better than the Ziehl-Neelsen or the
Spengler. The following may be useful for those who are colour-
blind to red :

a. Muck's method. — Prepare the following solution : 10 c.c. of
a saturated alcoholic solution of methyl violet B.N. in 100 c.c. of
2 per cent, aqueous carbolic ; (1) stain the film with this, warming
over the flame, or for 24-48 hours at 37° C. ; (2) treat with Gram's
iodine solution, 1-5 minutes ; (3) treat with 5 per cent, nitric acid
for 1 minute ; (4) treat with 3 per cent, hydrochloric acid for
10 seconds ; (5) treat with a mixture of equal parts of acetone and
absolute alcohol.

/3. Herman's method. — Prepare shortly before use the following
solution : 3 parts of a 1 per cent, aqueous solution of ammonium
carbonate, 1 part of a 3 per cent, solution of krystal violet in 95 per
cent, methyl alcohol. (1) Flood the film with this, warm until it
steams, and stain for 1 minute ; (2) decolorise with 10 per cent,
nitric acid for a few seconds, and then with 95 per cent, alcohol
until the film assumes a pale blue colour, then rinse in tap-water
followed by distilled water ; (3) counter-stain with 1 per cent,
aqueous eosin.

By both these methods the tubercle bacilli appear blue-black.
2. Tissues. — The histological appearance of the tubercle is usually
sufficient for diagnostic purposes without the demonstration of the
tubercle bacilli, which in many instances may be difficult in human

328 A MANUAL OF BACTERIOLOGY

material, as the bacilli may be very scanty, or practically impossible
to find, e.g. in lupus. Sections should be prepared either by the
freezing or the paraffin method, stained with haematoxylin, and
counter-stained with eosin, or orange-rubin, or with the Ehrlich-
Biondi mixture.

In order to demonstrate the tubercle bacillus in fresh tissue
smears may be made and stained like sputum, or sections prepared
and stained in warm carbol-fuchsin for about ten minutes. For
frozen sections the stain may be contained in a watch-glass or small
glass capsule, and is warmed until it steams, but not boiled, on a
piece of asbestos cardboard or a sand-bath. Paraffin sections
should be fixed to the slides with glycerin albumin, and may be
stained by flooding with the carbol-fuchsin and warming on asbestos
cardboard, or a heated penny, for ten minutes. After staining, the
sections are washed in water and are then decolorised in 25 per
cent, sulphuric acid. This is a longer process than with sputum,
and the sections after being in the acid for a few seconds are washed
in water and then returned to the acid, and this alternate rinsing in
acid and in water is repeated until they are nearly colourless when
placed in water. It is not necessary to remove the colour absolutely ;
a faint pink remaining does not matter. After rinsing in fresh water
to remove all the acid, the sections are counter-stained in Loffler's
methylene blue for two minutes, rinsed in methylated spirit, passed
through absolute alcohol somewhat rapidly to avoid removing too
much of the blue, cleared in cedar oil or xylol, and mounted in
balsam. The sections may also be counter-stained with haema-
toxylin or Bismarck brown.

Instead of using the strong acid solution for decolorising, an
acid alcohol solution may be used with advantage, or 2 per cent,
aqueous hydrochloride of anilin may be employed.

Gram's method may also be used, but is, of course, not distinctive
for the tubercle bacillus.

Sections may also be first stained with Ehrlich's or other haema-
toxylin solution, then stained with warm carbol-fuchsin, washed,
treated with 2 per cent, aqueous anilin hydrochloride for a few
seconds, decolorised with 75 per cent, alcohol until the red colour
is no longer apparent (15-30 minutes), and counter-stained with an
aqueous solution of orange.

Where a positive diagnosis is important, a small piece of the
tissue may be inserted under the skin of the thigh or abdomen of
a guinea-pig. If tuberculous, the animal will show signs of tuber-
culosis in two or three weeks (see below, " Urine ").

Films of pure cultivations of the tubercle bacillus may be stained

TUBERCLE BACILLUS IN URINE 329

in warm carbol-fuchsin for two to five minutes, rinsed in the
sulphuric or nitric acid solution, washed, dried, and mounted. They
can also be stained by Gram's method, which usually brings out
the beaded appearance very markedly, or by any of the other
methods mentioned under Sputum. Differentiation from the
leprosy bacillus will be found at p. 337, and from the smegma
bacillus and other acid-fast organisms at p. 339.

3. Urine. — The tubercle bacillus is often very difficult to
demonstrate in urine. The urine must be allowed to stand in a
conical glass for twenty-four hours or centrifuged, and film
specimens are prepared with the sediment and treated by one of
the methods for sputum given above. Several specimens should be
made and must be very carefully examined. The sediment may
also be treated by the antiformin method. It is important to
exclude the smegma bacillus, and the urine is preferably drawn off
by a catheter. Staining may be carried out by Housell's method,
by which the smegma bacillus is decolorised, viz. after staining in
warm carbol-fuchsin the specimen is washed and dried. It is then
immersed in acid alcohol (alcohol + 3 per cent, hydrochloric) for
ten minutes, washed in water, counter-stained for a few seconds in
a saturated alcoholic solution of methylene blue, washed, dried,
and mounted (see also p. 339). An electrolytic method for the
concentration of the tubercle bacilli has been devised by Russ.1

If a diagnosis is of importance inoculation should be resorted to.
Two guinea-pigs are inoculated subcutaneously in the thigh or
abdomen with 0-5 to 1 c.c. of the deposit from the sedimented or
centrifuged urine, or one may be inoculated subcutaneously, the
other intra-peritoneally. If tubercle bacilli are present the animals
may show signs of tuberculosis as early as two to three weeks after
inoculation. Sometimes, of course, the animals may die from some
intercurrent infection before the tuberculous infection has had time
to develop Delepine 2 recommends the inoculations to be made
on the inner aspect of the leg about the level of the knee. The
order of infection after inoculation is as follows : the popliteal,
superficial and deep inguinal, and sub-lumbar glands, the retro-
hepatic, mediastinal and bronchial, deep servical, and subscapular
glands, the spleen, liver, and lungs. The inoculated animals are
killed in two to three weeks, dissected, and the lesions examined
microscopically. Others inoculate two guinea-pigs, one sub-

1 Proc. Roy. Soc. Lond., B. 1909.

2 Brit. Med. Journ., 1893, vol. ii, p. 664. The results only apply to
ordinary forms of tuberculosis, and not to certain modified forms such
as lupus and the avian variety.

330 A MANUAL OF BACTERIOLOGY

cutaneously in the abdomen, the other intra-peritoneally. Negative
results are nearly as valuable as positive ones.

In fceces, if definite yellow caseous particles can be found, these
should be picked out, and films made and stained. Antiformin
may also be used. About 5-6 c.c. of faeces are mixed with 20 c.c.
of 15 per cent, aqueous antiformin in a conical glass, well agitated
and broken up, and an equal volume of the dilute antiformin is
then added. The mixture is allowed to stand for an hour, and films
are prepared from the white curdy layer which forms, stained, and
examined.

4. M ilk. — See section on milk (Chapter XXI).

V. The opsonic method. — The general mode of carrying this out
is described at pp. 314-319, the tubercle bacilli being suspended in
1-5 per cent, salt solution.

VI. Tuberculin reactions. — The old tuberculin is used for diagnostic
purposes ; it is not perhaps very safe. A dose of 0-0002 c.c. is
injected subcutaneously, and the temperature taken four-hourly
during the succeeding thirty-six hours. A rise of 2°-3° F. or more
ensues a few hours after injection in tuberculous subjects. If no
reaction occurs another dose of 0-0005 c.c. may be given after the
lapse of some days.

This method has now almost completely been superseded by the
cutaneous or by the ophthalmo reaction.

The cutaneous tuberculin reaction. — Von Pirquet 1 discovered that
when tuberculin is introduced into the superficial layers of the skin
of tuberculous individuals, as in vaccination, a reaction occurs
consisting of the formation of a papule with redness, slight swelling
and exudation, and sometimes small vesicles. This reaction is
usually at its height twenty-four to forty-eight hours after inocula-
tion. In healthy individuals no reaction follows the inoculation.
The method is to scarify a small spot on the forearm through a drop
of a dilution of the old tuberculin, and protect the patch with a
simple dry dressing. Moro has modified the method by applying
the tuberculin to the skin in the form of ointment.

The ophthalmo-tuberculin reaction. — Calmette transferred the site
of inoculation from the skin to the conjunctiva. He makes use of
material prepared by precipitating the old tuberculin with alcohol,
of which a 1-100 solution is prepared in distilled water. One drop
of this is instilled into the inner half of the conjunctiva of one eye.
In tuberculous individuals a reaction follows, usually in six to sixteen
hours after medication, consisting of a conjunctivitis, ranging in
intensity from a local redness to a redness extending over the whole
1 Wien. msd. Woch., July 6, 1907.

PSEUDO-TUBERCULOSIS 331

eye and having the appearance of an acute conjunctivitis. The
reaction soon passes off, generally without leaving ill effect. On
the whole the reaction appears to be fairly constant in tuberculous
individuals, but absence of reaction is not certain proof that the case
is not tuberculous.1

VII. Tuberculin for veterinary use. — The dose of the various
preparations in the market varies according to their strength ; it
corresponds to 0-1 c.c. or 0-2 c.c. of Koch's original tuberculin.

The appropriate dose is injected subcutaneously in the neck and
the reaction consists of a rise of temperature of from 1-5° to 6° F.
above the average normal, commencing 8-12 hours after injection
and lasting 12-14 hours, the temperature being taken at the
twentieth hour after injection, or, if it can be done, at frequent
intervals from the twelfth to the twentieth hour. The temperature
should be taken just before inoculation, and, if possible, morning
and evening for two or three days previous to inoculation.

A healthy animal is unaffected by the injection, and if an animal
be extensively affected with tuberculosis the reaction may not be
given, or may be masked by the fever present.

An ophthalmo-reaction may also be employed in cattle.

Johne's disease,2 a bovine enteritis, is due to an acid-fast bacillus
closely resembling the tubercle bacillus in morphology. It occurs
in scrapings of the affected mucous membrane of the bowel, and also
in sections of the intestinal wall. The Johne bacillus is inoculable
into the goat, but not into the guinea-pig or rabbit, and does not
grow on any of the ordinary laboratory media. Twort states that
it can be cultivated on the medium employed by him for growing
the leprosy bacillus (p. 335, and from the cultures a diagnostic)
vaccine may be prepared.3

Pseudo-Tuberculosis

The term " pseudo- tuberculosis " (which is not a good
one, and should be discarded) has been applied to a number
of different conditions which have as a common character
the presence of tubercle-like nodules, but which are not
caused by the tubercle bacillus. Such are produced by

1 See Brit. Med. Journ. and Lancet, 1907, vol. ii, and 1908, vol. i.

2 See MacFadyean, Journ. Comp. Path, and Therap., vol. xx, 1907,
p. 48.

3 Twort, Veterinary Record, Sept. 14, 1912.

332 A MANUAL OF BACTERIOLOGY

certain parasitic worms, by Blastomycetes, Streptothrix and
Aspergillus, Protozoa, and by several bacteria.

PfeifFer's Bacillus pseudo-tuberculosis produces nodular
deposits in the organ, accompanied by wasting, very like
true tuberculosis. The disease, however, runs a more
rapid course, death ensuing in the guinea-pigs two to three
weeks after inoculation. Guinea-pigs, rabbits, mice and
monkeys can be readily infected. The nodules consist of
masses of round cells which undergo necrosis and caseation.
The bacillus in the tissues is not readily stained, carbol-
methylene blue being the best solution, as it is not acid-
fast, nor does it stain by Gram's method. Morpho-
logically it is a small rod 1-2 /m in length, usually non-
motile, although, according to Klein, it possesses a single
flagellum or two flagella at one end. On gelatin it forms
a whitish growth without liquefaction, like that of the
colon bacillus, but confined to the needle-track. It pro-
duces alkali, forms no gas, and does not curdle milk.
Broth remains clear, with a whitish stringy flocculent
deposit. The bacillus grows readily and rapidly.

MacConkey has found that the fermentation reactions
of this organism and of the plague bacillus are practically
identical (see " Plague," p. 395), and sterilised cultures of
either will protect against the other.

Ovine caseous lymphadenitis, a disease of sheep simu-
lating tuberculosis, is due to a short pump bacillus with
rounded ends which stains well by Gram's method, and
grows best on blood-serum, on which it forms greyish
colonies.1

Much finds in the glands in Hodgkin's disease anti-
formin- resistant bodies, non-acid-fast, and similar to the
non-acid-fast tubercle bacilli which he has described.

1 Sixteenth Ann. Rep. Bureau of Animal Indust. U.S.A., p. 638.

LEPROSY 333

Leprosy

Leprosy, elephantiasis Graecorum or true elephantiasis
is a disease of which we have records from the earliest
times. It was undoubtedly somewhat prevalent in the
British Isles from the twelfth to the fifteenth centuries,
as the many leper houses and enactments against lepers
testify, though no doubt other skin diseases, psoriasis,
lupus, etc., were at that early period of medical diagnosis
confounded with it. At the present day leprosy, although
extinct in the British Isles, may be said to have a world-
wide distribution, for it is met with in Iceland and Scan-
dinavia, Russia and the Mediterranean coasts ; in Persia,
India, China, Siberia, and Japan ; in Africa from north to
south ; in many districts of the American continent ; and
in the Pacific Islands. Three varieties of leprosy are
described — the tuberculated or nodular, the anaesthetic,
and the mixed.

The mode of spread is probably by personal contact
(though possibly insects play some part), and throughout
ancient and mediaeval times leprosy was considered to
be a contagious and communicable disease, as witness the
stringent regulations in the Mosaic and other laws for the
segregation of lepers. J. Hutchinson supposed that fish
in the diet, particularly if stale, decomposed, or badly
cured, in some way is a causative factor ; but he is
practically alone in this view.

A bacillus, the Bacillus leprce, is abundant in the tissues
and was discovered by Hansen in 1879. In form it
resembles the tubercle bacillus, but is slightly more slender ;
it probably does not form spores, though in stained pre-
parations the same irregularity in staining — namely, the
occurrence of unstained intervals, the so-called " beading *'
— is met with as in the tubercle bacillus, and is assumed
by some to be due to the presence of spores. The organism

334 A MANUAL OF BACTERIOLOGY

as obtained from the tissues is non-motile, stains readily
with the ordinary anilin dyes, and by Gram's method,
which brings out the beaded appearance very well, and is
markedly acid-fast, thus closely resembling the tubercle
bacillus, and the methods used to demonstrate it are the
same as for the latter organism.

The Bacillus leprce is found in enormous numbers,
usually crowded together in bundles or masses, in the
leprous nodules in the skin (Plate X. a), liver, spleen, and
testicles, in the affected nerves in the anaesthetic form
and even in the ganglion cells of the central nervous system
— in fact, any viscus may be affected ; it has also been
found in the blood, but only in the febrile paroxysms
which set in when the disease is approaching a fatal
termination. The exact situation of the leprosy bacilli
in the tissues has been a matter of controversy. By some
it has been held that they are contained within certain
round cells, the so-called leprous cells, and this may be
the case, but to an inconsiderable extent. Unna has
always regarded these leprous cells as really being trans-
verse sections of lymphatic vessels containing bacillary
thrombi, and this seems to be usually the case. Giant-
cells are occasionally present in the leprous nodules. One
of the most constant and earliest situations in which the
B. leprce is found is the nasal mucous membrane.

Lepers react to the old tuberculin and also give the
Wassermann reaction.

Although the organism is present in such enormous
numbers and is so readily demonstrable, to cultivate it
on artificial media and to infect animals with it are both
difficult matters. Babes, Bordoni-Uffreduzzi, Czaplewski,
are some of those who in the past believe that they have
cultivated the leprosy bacillus. Van Houten1 claimed to
have succeeded by growing it in glycerin fish broth. The

1 Journ. Path, and BacL, vol. viii, 1903, p. 260.

LEPROSY BACILLUS 335

bacillus cultivated was acid-fast, and agglutinated with,
and was sensitised by, lepers' serum.

Deycke,1 by taking fragments of leprosy tissue and
incubating for several weeks in physiological salt solution
at 37° C., obtained a growth of a semi-acid-fast strepto-
thrix, S. kproides. He is uncertain if this is a true growth
of the leprosy bacillus. Injected into leprosy patients it
seemed to produce a beneficial effect. The acid-fast
property resides in a fatty substance which can be extracted
with solvents, particularly benzoyl chloride. The fatty
substance Deycke terms " nastin " ; it is a neutral fat,
the glycerin ester of a fatty acid of high molecular weight.
Injected into leprosy patients it sometimes produces
marked reaction, sometimes not. In solution in benzoyl
chloride it is much more active, and Deycke hopes that it
will act as a curative vaccine in leprosy. On the whole,
the results obtained with nastin have been disappointing.
Twort2 claimed to have cultivated the B. leprce on a
medium consisting of eggs, glycerin, and ground-up
tubercle bacilli. Clegg states that the leprosy bacillus
will grow in symbiosis with amoebae, and Duval that it
grows in 1 per cent, human serum in symbiosis with some
bacteria. Kedrowsky and Bayon claim to have grown the
organism on a placental-juice agar, and Bayon has obtained
complement fixation with his cultures with leper serum.
Kedrowsky's organism is a non-acid-fast diphtheroid,
Clegg's an acid-fast chromogenic bacillus, Duval's and
Bayon's are acid-fast leproid bacilli.

In 1904 Rost announced that he had obtained cultures
of the leprosy bacillus in a chlorine-free medium, but this
was not confirmed. In 1909 he again claimed success by
cultivating in a medium consisting of the fluid obtained
by the steam distillation of rotten fish to which is added

1 Brit. Med. Journ., 1908, vol. i, p. 802.

2 Proc. Roy. Soc. Lond., B., 1911.

336 A MANUAL OF BACTERIOLOGY

a little Lemco broth and milk, and Bannerman believes
that he is correct.1 Williams has grown a non-acid-fast
streptothrix in ordinary broth, and has also cultivated
acid- fast bacilli in a modified Rost medium (substituting
distilled water for the fish distillate). The writer has also
grown a non- acid- fast streptothrix from a case of leprosy
on brain agar containing the juice from disintegrated
B. megaterium. As a result of these alleged positive
cultural results, it has been surmised that the B. leprce is
really a streptothrix, that it is acid- fast only under certain
conditions, viz. in the body or in media containing fat,
and that under cultivation the streptothrix may break up
into non-acid-fast diphtheroid bacilli or into acid-fast
leproid bacilli. On the other hand, Fraser and Fletcher2
have made 373 inoculations from 33 non- ulcerating cases
of leprosy on a variety of culture media with entirely
negative results. More work is therefore required before
it can be definitely stated that the leprosy bacillus has
been cultivated.

A certain number of positive results of the inoculation
of leprous material into the lower animals have been
reported by Ortmann and others. Nicolle3 has reported
the successful inoculation of a macaque monkey, but
most of the attempts have ended in failure ; positive
results are open to criticism and may be fallacious, for
lepers not infrequently suffer from coincident tuberculosis,
and the animals therefore may have been infected with
tuberculosis. Japanese dancing mice are also stated to be
slightly susceptible. The local lesion induced in animals
may be simply inflammatory, produced by the leprous
material acting as a foreign body, and the bacilli may be
diffused without proliferating. Human beings have also

1 See Sc. Mem. Gov. of India, No. 42, 1911.

2 Lancet, Sept. 27, 1913.

3 Comp. Rend. Acad. Sc., 1905.

DIAGNOSIS OF LEPROSY 337

been inoculated, but the positive results obtained are all
open to objection.

The differentiation of leprosy from tuberculosis, although
the bacilli are so similar, does not in the majority of cases
present much difficulty. The large number of bacilli
present in the lesions, and particularly in the skin, forms
a marked distinction from tuberculosis. The Bacillus
leprce also stains more readily, and with watery solutions
in a shorter time, than does the Bacillus tuberculosis,
though this distinction is hardly marked enough for
diagnostic purposes.

Cases of leprosy, both of the nodular and anesthetic
varieties, have been treated with injections of Koch's
tuberculin, which has been found to produce a certain
amount of reaction followed by some amelioration in
their condition. Rost and Williams with their cultures
have prepared vaccines with which treatment is being pur-
sued. Nicholls and others have used extracts of leprous
tissue as a vaccine, and Bayon states that a filtered extract
of the Kedrowsky culture is of service for treatment.

Dean x and others have met with a leprosy-like disease in the rat.
Marchoux found about 5 per cent, of the sewer rats in Paris infected
with it. Nodules are found in the tissues which contain large
numbers of an acid-fast bacillus closely resembling the B. leprce^
Material from infected rats inoculated into healthy rats reproduces
the disease after some months, but has no effects on guinea-pigs.
The disease is probably conveyed by contact.

Dean cultivated a diphtheroid non-acid-fast bacillus from this
disease ; Bayon an acid-fast leproid bacillus which he finds to be
very similar to that obtained by him from human leprosy.

Clinical Examination

(1) If cutaneous nodules be present, one is clamped, pricked, and
films are prepared with the juice that exudes and stained as for

1 Journ. of Hyg., vol. v, 1905, p. 99 ; Marchoux and Sorel, Ann. de
Vlnst. Pasteur, xxvi, 1912, p. 778.

22

338 A MANUAL OF BACTERIOLOGY

tubercle. The occurrence of large numbers of bacilli, having the
same staining reactions as the tubercle bacillus and obtained from
the cutaneous structures, is diagnostic of leprosy (the smegma
bacillus may be present on, but not in, the skin).

(2) In the tissues, sections of which are stained in the same
manner as tuberculous material, the diagnosis must be based on the
presence of the bacilli in large numbers in the so-called leprosy -cells.

(3) Leprosy is not inoculable in guinea-pigs.

N.B. — It must be remembered that lepers not infrequently suffer
from coincident tuberculosis.

(4) The differentiation of the leprosy from the tubercle bacillus
by staining methods cannot be said to be satisfactory. By staining
in a saturated aqueous solution of fuchsin in the cold for five to
seven minutes, and subsequently decolorising with acid alcohol
(nitric acid 1 part, alcohol 10 parts), it is stated that the leprosy
bacillus is stained, the tubercle bacillus not.

The Smegma Bacillus1

The smegna bacillus is an organism found in the smegma
praeputii, between the scrotum and thigh, and between the
labia. It also occurs in the cerumen, occasionally on the
skin, and possibly in the sputum.

It is a small bacillus resembling the tubercle bacillus
in size and appearance, and, like the latter, is difficult
to stain, but when stained with carbol- fuchsin, retains
the colour after treatment with a 25 per cent, mineral
acid (Plate X. 6) ; it is also Gram-positive. It has,
therefore, to be distinguished from the tubercle bacillus
in certain localities, viz. in urine and about the external
genitals. It is non-inoculable on animals, and does not
usually grow in primary cultures on ordinary media,
but can be isolated by the use of blood- serum or nutrose-
agar, on which it forms delicate, ropy colonies. After
isolation it grows freely on agar as a thin, slightly brownish,
creamy layer, in which the bacilli may be very short but

1 See Neufeld, Arch. f. Hygiene, xxxix, p. 184; Zeitschr. /. Hyg.,
Xxxix, 1901 ; and Moeller, Centr.f. BakL, xxxi, 1902 (Originale), p. 278.

PLATE X.

a. Leprosy. Section of skin, x 1500.

b. The smegma bacillus. Smear preparation of smegma.
X 1500.

THE SMEGMA BACILLUS 339

retain their acid- fast properties ; on potato it forms
minute (0-5-1 mm.) greyish colonies. It has been sug-
gested that the syphilis bacillus of Lustgarten is identical
with the smegma bacillus ; neither is decolorised by
Lustgarten's permanganate method, but while the smegna
bacillus after staining is with difficulty decolorised by
acid, and is easily decolorised by alcohol, the reverse is
the case with Lustgarten's bacillus.

Staining and Differentiation

Film preparations of smegna may be stained in exactly the same
manner as for tubercle, after treating the preparations with ether
to get rid of fatty material.

The urine should be drawn off with a catheter when it is to be
examined for the tubercle bacillus ; this will generaUy exclude the
smegma bacillus. Young and Churchman l conclude that the
smegma bacillus is a scant invader of the male urethra, and that
by washing the glans and irrigation of the urethra it may be
eliminated from the urine.

If there is reason to suspect the presence of the smegma bacillus
when staining for tubercle, Bunge and Tranteroth 2 recommend
that the film specimens should be treated as follows :

(1) Immerse in absolute alcohol for three hours.

(2) Immerse in 5 per cent, chromic acid for fifteen minutes.

(3) Stain in warm carbol-fuchsin.

(4) Decolorise in 25 per cent, sulphuric acid for two to three
minutes.

(5) Counter-stain in a concentrated alcoholic solution of methyl-
ene-blue for five minutes.

The smegma bacillus will be decolorised by this method (see also
p. 329).

Coles recommends (Journal of State Medicine, vol. xii,
1904, p. 225) the following staining method :

(1) Spread thin and even films on slides, and fix by heat, in the
ordinary way.

1 Johns Hopkins Hospital Rep., vol. xiii, 1906, p. 15.

2 Fortschrit. der Med., xiv, 1896, Nos. 23 and 24. See also ibid.
No. 9.

340 A MANUAL OF BACTERIOLOGY

(2) While still warm from the heat fixation flood with filtered
carbol-fuchsin for half a minute. Again warm for a few second
over the flame without actual boiling. Allow it to stand and stain
for seven minutes.

(3) Wash thoroughly in running water, and then decolorise in
either of the following solutions :

(a) In Pappenheim's solution.1 — Place the preparation in a wide-
mouthed bottle containing the solution for not less than four, and
not longer than twelve, hours. Wash, dry, and mount. Tubercle
bacilli are the only organisms stained red.

(6) In Pappenheim's solution without methylene-blue. — Proceed as
in (a) ; wash in water and counter-stain for a minute in weak
aqueous methylene-blue solution. The tubercle bacilli are biilliantly
red.

(c) In 25 per cent, sulphuric acid. — Pour on a few drops of the
acid and allow it to act for half a minute. Pour off, and then place
the preparation in a wide-mouthed bottle containing the acid for
not less than sixteen hours and not more than twenty-four hours.
Wash thoroughly, counter-stain with weak aqueous methylene-blue.
Tubercle bacilli are the only bacilli which retain the red.

Acid-fast bacilli in milk and butter. — Numerous acid-fast bacilli
have been obtained from milk and butter. They usually grow
freely and quickly on agar and on gelatin without liquefaction,
sometimes as a creamy layer, sometimes as a dry, crinkled film,
which may be pigmented (yellow, orange, pale brown or brick red).
Some are pathogenic to guinea-pigs by massive intra-peritoneal
inoculation only, producing a plastic peritonitis, but not nodules
in the organs. In culture, the bacilli are acid-fast and occasionally
resemble B. tuberculosis, but are generally thicker. (See Petri
Arb. a. d. Kais. Gesundheitsamte, xiv, 1897 ; Rabinowitsch, Zeitschr.
f. Hyg., xxvi, 1897 ; Grassberger, Munch, med. Woch., 1899, Nos. 11
and 12 ; Tobler, ibid, xxxvi ; Swithinbank and Newman, Bacteri-
ology of Milk [Murray, 1903].)

Grass bacilli and mist bacillus. — Moeller isolated from a grass
(Phleum arvense) an acid-fast bacillus which he termed the Timothy-
grass bacillus ; other grasses also yield acid-fast bacilli (Grass
Bacillus II). They grow readily on culture media, and are not so
acid-fast as the tubercle bacillus. The Mist bacillus was isolated
from dung, and is considered by Pettersson to be identical with
the Timothy-grass bacillus. (See Moeller, Deutsch. med. Woch.,

1 Pappenheim's solution consists of one part of corallin (rosolic acid)
in 100 parts of absolute alcohol, to which methylene-blue is added to
saturation ; 20 parts of glycerin are then added.

GLANDERS 341

1898, p. 376 ; Herr, Zeitschr. f. Hyg., xxxviii, 1901 ; Pettersson,
Berl klin. Woch., 1899, p. 562.)

Glanders l

Glanders is a disease which has been known from the
earliest times, being recognised by the Greek and Roman
writers, by whom it was termed yuaX*? and malleus respec-
tively. It is distinctly a disease of the horse, mule, and

FIG. 39. — Nasal septum of glandered horse, showing ulceration of
Schneiderian membrane (McFadyean).

ass, but is also communicable to man and to certain
other animals. It is caused by a small bacillus discovered
by Loffler and Schiitz in 1882.

In the horse the lungs are always affected, and fre-
quently the nasal mucous membrane (Fig. 39). Nodules
form which afterwards break down and ulcerate, and a
muco-purulent discharge appears ; in the older writings
the name " glanders " covered only these advanced cases
of the disease. In " farcy " the lymphatic vessels and

1 See McFadyean, Journ. of State Med., vol. xiii, 1905, pp. 1, 65, and
125.

342

A MANUAL OF BACTERIOLOGY

glands are affected, the enlarged glands being known as
" farcy buds " (Fig. 40).

In man the disease is rare, an average of four deaths
per annum being caused by it in this country. It occurs
in two forms — the acute and the chronic. The former
is a very serious affection, accompanied by high fever,
prostration, and delirium, and almost invariably fatal in
from two to three weeks. The seat of infection is usually

FIG. 40. — Horse affected with farcy (McFadyean).

the hand or arm, the nasal mucous membrane being
sometimes subsequently involved, and deposits may form
in the lymphatic glands, internal organs, and muscles.
In the chronic form intramuscular abscesses are frequent,
from the breaking down of which indolent ulcers may
result ; the disease runs a prolonged course of weeks or
even months, and about half the cases end in recovery.
In the early stage an eruption may develop on the
forehead and face simulating very closely that of
smallpox,

BACILLUS MALLEI 343

The Glanders Bacillus

The glanders bacillus (B. mallei) is an obligatory parasite
with the equine species for its normal host. It hardly
grows on artificial media below about 20° C., and probably
cannot maintain a saprophytic existence outside the
animal body.

Morphology. — The glanders bacillus occurs in the
tissues as a cylindrical rod with rounded ends, varying
between 2 //, and 5 /m in length, and generally straight,
though sometimes slightly curved. The bacilli are usually
irregularly scattered, and do not tend to form colonies.
In stained preparations they often appear more or less
beaded, or may exhibit bipolar staining, but some stain
uniformly. The bacilli from young cultures not more
than twenty- four hours old are almost always short rods,
a little thicker than those found in the lesions (Plate XI. a).
In old broth cultures the surface growth is largely com-
posed of filaments, which do not show any regular seg-
mentation, but may exhibit lateral branching, and may
have club-shaped extremities. From these features some
have inferred that the glanders organism belongs to the
Streptothricce. The bacillus does not form spores, and is
probably non- motile, though in a hanging- drop prepara-
tion a very active Brownian movement is present.

Staining reactions. — The bacillus is Gram- negative, and
is not acid-fast, but from young cultures stains readily
with the ordinary anilin dyes. In smears of glanders or
farcy material, a simple staining with any of the basic
anilin dyes, with subsequent decolorisation with dilute
acetic acid, suffices to demonstrate it if it is present in any
number, a difficulty in recognising the organism being the
presence of deeply staining nuclear detritus. In sections,
methylene-blue staining with decolorisation in dilute
acetic and mordanting with tannin gives the best results

344 A MANUAL OF BACTERIOLOGY

(p. 350). The bacillus shows dark staining dots when
treated with osmic acid, suggesting fat- globules (Shattock).

Cultural characters. — The Bacillus mallei is an aerobic,
and facultatively anaerobic organism. The growth on
gelatin at 22° C. is scanty and pale brownish in colour
without liquefaction. On glycerin agar it forms a thick
cream- or slightly brown- coloured growth, and on blood-
serum a somewhat amber- coloured growth, which after-
wards becomes brownish. The growth on potato at
37° C. is most characteristic, and practically diagnostic.
If the surface of the potato is inoculated with a loopful
of farcy pus or material from the centre of a glanders
nodule, the resulting growth is usually not distinctly
visible until the third day, when raised, translucent,
viscid, amber- yellow coloured growth or colonies appear.
With continued incubation the colonies coalesce, the
growth becomes thicker and fawn-coloured, then reddish-
brown, and finally generally chocolate-brown. The
growth is also odourless, limited to the site of implanta-
tion, and does not stain the potato. Broth or glycerin
broth becomes uniformly turbid, and after a week or so
patches of a whitish surface scum form, and after three
weeks the broth is nearly covered with this surface growth,
which is slimy and easily broken up on shaking. Broth
cultures give the indole reaction. Litmus glucose agar
becomes pink. Milk is not coagulated.

Resistance to Germicides., etc. — The glanders bacillus is
but little resistant,* and cultures frequently die out in a
month or so. Complete desiccation at 37° C. of nasal
discharge, farcy pus, or bacilli from cultures, is frequently
fatal in twenty- four to forty- eight hours. Young broth
cultures are soon destroyed by bright sunlight, and an
exposure of ten minutes to a temperature of 55° C. is fatal
to the cultivated bacilli. A 3 per cent, solution of carbolic
acid, a 1 per cent, solution of potassium permanganate,

PLATE XI.

a. The glanders bacillus. Film preparation of a
pure culture, x 1000.

Section of a glanders nodule, showing giant -cells (after
McFadyean).

PATHOGENICITY OF GLANDERS BACILLUS 345

and a 1 in 5000 solution of corrosive sublimate are fatal
in two to five minutes.

Pathogenicity , etc. — The glanders bacillus varies con-
siderably in virulence, and under continued cultivation
may become almost non-pathogenic.

Glanders is met with exclusively among horses, asses,
and mules, and man is infected from these animals, nearly
all cases of human glanders being among ostlers, grooms,
and coachmen, and the usual mode of infection is by
farcy pus or nasal discharge coming into contact with a
cutaneous wound or abrasion. A remarkable immunity,
however, is enjoyed by the slaughterers, who have to deal
with the carcases of glandered animals, and who might
be supposed to run the greatest risk. But it must be
remembered that Babes frequently found at the post-
mortem on persons who had to do with horses, and who
died from diseases other than glanders, encapsuled glanders
nodules in the lungs and internal organs, suggesting that
the disease may often be latent in man, who appears to
be relatively insusceptible, and that infection may be
possible by inhalation. In the horse glanders is readily
transmissible experimentally both by ingestion and by
inoculation, and ingestion is probably the common mode
of infection naturally, infection by inhalation occasionally
occurring. Even when glanders bacilli are administered
experimentally by the mouth in the horse, the lesions may
be most prominent in, or even be confined to, the lungs.
In the horse, the disease has periods of epidemic prevalence,
and is particularly frequent in London. In 1892 there
were 3000 equine cases in Great Britain, in 1903 there were
2499 cases, and nearly 90 per cent, of all cases occur in the
Metropolitan area. These, it is to be noted, were cases
in which the disease was well developed and manifest, but
there are also numerous others in which it is latent.
Guinea-pigs and field mice are highly susceptible to the

346 A MANUAL OF BACTERIOLOGY

disease, which may also be contracted by some of the
Carnivora, such as the cat, lion, and tiger, by inoculation
or by feeding on diseased carcases. The rabbit, sheep,
and dog are but slightly susceptible, while cattle, swine,
and house mice are stated to be immune. Shattock1 found
that the white mouse is somewhat susceptible, and suggests
that in all probability the house mouse is similarly so.

In the horse the most constant seat of glanders lesions
is the lung, and McFadyeaii states that no case of glanders
with lesions elsewhere than in the lungs, and with these
organs unaffected, has ever been recorded. In nearly
every case of farcy, also, nodules are present in the lungs.
The lung lesions have the form of rounded, firm, or shotty
nodules. The number present is variable, rarely less than
a dozen ; exceptionally there are hundreds, fairly evenly
distributed throughout the lung tissue. The nodule
commences as a collection of polymorphonuclear leucocytes,
around which a zone of congestion is present. Later, the
alveolar walls undergo necrosis, and the leucocytes necrose
and disintegrate, but their chromatin persists as rounded
fragments which retain their affinity for nuclear stains
(chromatotaxis). The nodule may become surrounded
with a layer of thin fibrous tissue, between which and
the necrotic central area a zone of endothelioid cells with
giant- cells may be present (Plate XI. 6).

The lesions of farcy are at the onset histologically
identical with the glanders nodule, but by the progressive
liquefaction of the tissues actual abscesses form.

The lesions set up in an inoculated guinea-pig are very
characteristic, and can be used for diagnostic purposes.
With a very virulent culture, such as can be obtained by
several passages through a susceptible animal, a guinea-pig
may die in four or five days, and the post-mortem lesions
are slight, consisting of some caseation at the seat of
1 Trans. Path. Soc. Lond., vol. lix, 1898, p. 333.

STRAUS'S TEST 347

inoculation and slightly enlarged spleen, which contains
a few small yellowish nodules resembling miliary tubercles.
The material from human cases as a rule seems more
virulent than that from the horse, and death of the guinea-
pig often ensues a few days after inoculation.

The culture or material from a glandered horse does not
usually produce death of a guinea-pig until a lapse of
two or three weeks. A male guinea-pig being chosen, the
changes observed are caseation followed by ulceration
at the seat of inoculation, when this is done subcutaneously,
and great enlargement of the testicles ; on cutting into
these they are found to be partially or almost entirely
converted into a pasty caseous material, while the skin
covering them is so adherent that it can only be detached
by cutting, and the spleen is very much enlarged and
studded with small yellowish nodules. In a female
guinea-pig the ovaries are attacked. These appearances
are of importance in the diagnosis of the disease. The
difficulty of finding the bacillus in the discharges by
microscopical and staining methods is so great that these
cannot be employed with any certainty. Loffler and
Straus therefore recommend the inoculation of a male
guinea-pig intraperitoneally with the discharge or other
material. If the glanders bacillus is present the lesions
thus described rapidly ensue, and the diagnosis is estab-
lished in four or five days (Straus's test1). At the present
time the inoculation method has been almost entirely
superseded by the introduction of mallein, the former
being reserved for clinical diagnosis in man.

McFadyean found that the blood of a glandered animal
produces agglutination or clumping of the glanders bacillus
similar to that obtained in the agglutination (Widal) test
for typhoid, and has suggested this reaction as a means
of diagnosis. As an aid to the clinical diagnosis of the

1 See also Nicolle, Ann. de Vlnst. Pasteur, xx, 1906.

348 A MANUAL OF BACTERIOLOGY

disease in man it is doubtful if the method of serum diagnosis
can be applied, for Foulerton found that typhoid and diph-
theria sera also produce agglutination of the glanders bacillus.

Toxins.' — Mallein, a preparation analogous to tuberculin,
is prepared by growing a virulent glanders bacillus for a
month or six weeks in glycerin veal- broth in flat flasks
such as are employed for tuberculin (Fig. 38), so that there
is free access of oxygen. The culture is then autoclaved
for fifteen minutes at 115° C., filtered through a Berkefeld
filter, concentrated to one fourth of its volume, and mixed
with an equal volume of a J per cent, solution of carbolic
acid. This yields an active mallein, 1 c.c. of which is a
dose, and gives a good reaction. Like tuberculin, it
possesses feeble curative properties, though a few cases
of cure by prolonged use have been reported by Babes
and others, but is used for diagnostic purposes ; the
veterinary authorities are unanimously agreed that it is
one of the most certain means we possess for diagnosing
glanders in the horse. Injected into an unglandered
horse little or no effect is produced, but in a glandered
animal, about twelve hours after injection, the tempera-
ture rises 1-5° to 3° C. above the normal, a large and
painful swelling forms at the seat of inoculation (it may be
as large or even larger than half a cocoanut), while any
affected lymphatic vessels or farcy buds become swollen.
Reaction may, however, be produced in the absence of
glanders if the horse is being treated with bacterial products,
toxins, etc.1

Epizootic lymphangitis has a superficial resemblance to
farcy in the horse, and must not be mistaken for the latter
(see " Sporotrichosis," Chapter XVI).

The greatest care should be exercised when working with
glanders material or cultures, several fatal laboratory
accidents having unfortunately happened.
1 See Sudmersen and Glenny, Journ. of Hygiene, vol. viii, 1908, p. 14.

DIAGNOSIS OF GLANDERS 349

Whit more1 describes a glanders -like disease occurring in man in
Rangoon. A non-Gram-staining bacillus is present, morphologically
like the glanders bacillus, but killing guinea-pigs with septicaemic
symptoms and not affecting the testes, growing well and luxuriantly
on culture media, liquefying gelatin slowly, growing well on potato
with at first a cream-coloured, and subsequently a yellowish growth,
curdling milk and not fermenting any sugar.

Clinical Examination

(1) Prepare and stain film preparations of the pus or discharge
in Loffler's blue, with subsequent partial decolorisation in 4 per
cent, acetic. The ordinary pyogenic cocci will not be found unless
a secondary infection has occurred, and the material may appear
sterile, for the glanders bacilli may be very scanty.

(2) Several tubes of glycerin-agar and potato should be inoculated
and incubated at 37° C. for seventy-two hours. On the agar,
colonies of the glanders bacillus will develop in twenty-four to
thirty-six hours, but the potato will not show the characteristic
amber-yellow growth under forty-eight to seventy-two hours.

(3) It will usually be necessary (in man, at least) to confirm the
diagnosis by an inoculation experiment. A fully developed male
guinea-pig is chosen, and a little of the discharge, or an emulsion
of the material (O5 to 1 c.c.) is injected intraperitoneally, if the
material be fairly sterile, but if not, subcutaneously. In three to
five days the animal should show the characteristic swelling of the
testicles if the material be glandered.

(4) An ophthalmo-reaction is stated to be reliable both in man
and in animals.

(5) In animals the mallein test may be applied. The dose is
injected subcutaneously in the neck over the vertebrae, and midway
between the jaw and the shoulder.

(a) The temperature of the animal should be taken if possible
morning and evening for two or three days previous to inoculation ;
in any case at the twentieth hour after inoculation, or, better, at
frequent intervals from the twelfth to the twentieth hour.

(6) A complete reaction comprises (i) a rise of temperature of
more than 2-5° F., (ii) an extensive hot and painful swelling at the
seat of inoculation. Systemic disturbance, such as prostration, loss
of appetite, shivering, etc., majr occur.

(c) The temperature reaction is unreliable in all cases in which

1 Journ. of Hyg., xiii, 1913, p. 1.

350 A MANUAL OF BACTERIOLOGY

the temperature at the time of inoculation is 2-5° F. above the
normal. In such cases, if there be any suspicious clinical signs to
assist, reliance may be placed upon the local swelling.

(6) In animals the agglutination reaction is stated by Moore
and Taylor l to give accurate results. In man this test might
give an inconclusive result (see ante).

(7) In the tissues the glanders bacillus is difficult to demonstrate.
Sections may be stained for half an hour with carbol methylene-
blue, treated with 4 per cent, acetic for a few seconds, washed, and
rapidly dehydrated with alcohol, cleared and mounted. McFadyean
recommends, after treating with acetic and washing, flooding with
a saturated solution of tannic acid in water for fifteen minutes,
washing, counter-staining in a 1 per cent, aqueous solution of acid
fuchsin for fifteen to thirty seconds, washing, dehydrating, and
clearing in cedar oil.

Twort's method may also be employed (see section on Amoeba
coli, " Clinical Diagnosis ").

1 Journ. of Infect. Diseases, Sup. No. 3, May 1907, p. 85.

CHAPTER X

TYPHOID FEVER— PARA-TYPHOID FEVER— BACILLUS
ENTERITIDIS AND THE GARTNER GROUP— SWINE
FEVER— BACILLUS DYSENTERIC— BACILLUS COLI

THE organisms considered in this chapter form a natural group or
family, the " Typhoid-Colon " group, and pass as it were by grada-
tions in cultural characters from the typhoid bacillus to the colon
bacillus. Loffler classes them together in a family, the Typhacese,
which is divided into sub-families : (a) Typheae, which includes the
B. typhosus and B. dysenteries ; (b) losarceae,1 which includes the
Gartner group of organisms ; and (c) Colese, the B. coli group of
organisms.

The group can be divided into lactose fermenters and non-lactose
fermenters. The former includes B. coli and its variants. There
is also a group of late lactose fermenters (after six days) which occur
in the intestine, e.g. B. coli mutabilis. The non-lactose fermenters
are classified by Henderson -Smith 2 as follows :

I. Certain groups of no known pathogenic importance. Frequent
in the intestine.

II. The Typhoid group, B. typhosus.

III. Paratyphoid-Enteritidis (Gartner) group.

1. Atypical members.

a. Saccharose fermenters. Not agglutinated with Gartner

or paratyphoid serum.
6. Dulcitol non-fermenters.

c. B. paratyphosus A.

d. Salicin fermenters. Frequent in animals*

2. Typical members.

a. B. enter itidis of Gartner.

b. B. paratyphosus B.

c. B. suipestifer.

1 From los, poison, and <rap£, flesh.

2 Centr.f. Bakt. Abt. I (Orig.), 68, 1913, p. 151 (Bibliog.).

351

352 A MANUAL OF BACTERIOLOGY

IV. Dysentery group.

1. Mannitol non-fermenters. B. dysenterice, Shiga.

2. Mannitol fermenters.

a. B. dysenterice, Strong.

b. Sorbite fermenters.

(a) Dextrin non-fermenters.

(b) Dextrin fermenters.

c. Sorbite non-fermenters.

(a) Dextrin non-fermenters.

(b) Dextrin fermenters.

a. Maltose fermenters. B. dysenterice, Flexner.
/3. Maltose non-fermenters. B. dysenterice Y.

The typhoid bacillus is a remarkably stable and well-defined
organism showing little or no variation, unlike most other members
of the group.

All the foregoing are non-liquefiers ; for convenience certain
liquefying forms, e.g. B. cloacce, may be placed in this group.

Typhoid Fever

The specific organism of typhoid fever is a bacillus origi-
nally isolated by Eberth in 1880, and more closely studied
by Gafiky in 1884.

The Eberth- Gaff ky bacillus, or Bacillus typhosus, is best
observed in sections of the spleen, in which it occurs in
groups or colonies consisting of short rods with rounded
ends, each measuring about 3 jm in length. It has also
been demonstrated in the mesenteric glands and liver, in
the swollen Peyer's patches before ulceration, and in
other situations.

Pure cultivations may be obtained from the spleen
during life by puncture (p. 370), from the blood (p. 369),
sometimes from the urine, or from the spleen of a cadaver.
In the latter case the organ is washed, and then cauterised
lineally by means of a red-hot iron, in order to destroy
the saprophytic bacteria on and near the surface. An
incision is made with a sterilised knife through this

BACILLUS TYPHOSUS 353

cauterised area, and a little of the splenic pulp is taken with
a sterilised platinum needle and inoculated on to tubes
or plates, preferably of litmus lactose, Conradi-Drigalski,
or malachite-green, agar. These are incubated at 37° C.
for twenty-four to forty-eight hours, and the growths
which develop are examined microscopically and are tested
by agglutination and by cultural methods. The Bacillus
typhosus has the following characters :

Morphology. — Bacilli with rounded ends averaging 3 u.
in length, and 0-6 //. broad. It is, however, in cultivation
a markedly pleomorphic organism, and very short rods,
long rods, and thick filaments 10 to 30 /u. in length occur ;
the latter are known as involution forms (Plate XII. a).
It does not form spores, but granulation and vacuolation
may be observed in the protoplasm, particularly in old
cultures.

It is actively motile, and possesses a number of flagella,
arranged peritrichically both at the poles and sides (Plate
XII. c). The flagella are long and wavy, and average
eight to twelve in number, a point of differentiation from
the Bacillus coli, which usually has only three or four. It
stains by the ordinary anilin dyes, but not by Gram's
method.

Cultural characters. — The B. typhosus is aerobic and
facultatively anaerobic, and grows well on the ordinary
culture media. On agar it forms a thick, moist, greyish
layer. On gelatin it grows slowly, and the growth, which
is usually scanty and confined to the needle-track, is white
and shining, and somewhat irregular (Plate XII. 6). The
colonies in gelatin are visible in about forty-eight hours,
and form small roundish- white points, which are granular
and brownish in colour by transmitted light. In broth
it produces a general turbidity, without film formation.
The growth on potato acid in reaction is somewhat charac-
teristic ; it forms a moist, grey, shining layer, which is

23

354 A MANUAL OF BACTERIOLOGY

almost invisible. If, however, the reaction of the potato is
neutral or alkaline, the growth may be yellowish. The
B. typhosus grows well in milk, with slight permanent
acidity, but without coagulation.

Acid is formed in small quantity during its growth in
many media (volatile fatty acids, and lactic acid), which
can be demonstrated by cultivating in litmus milk, or in
litmus glucose media, and the organism will grow in slightly
acid media. Neither gas nor indole 1 is formed in cultures ;
acid is produced from glucose, but no gas ; lactose is
unacted upon. The fermentation reactions on various
media are given in the Table on p. 381, and are there con-
trasted with those of the B. coli and other organisms (see
also p. 384). Chatterjee 2 finds that agar on which the
typhoid bacillus has been grown contains substances which
inhibit further development of the organism if it be inocu-
lated on to an agar culture which has been scraped so as
to remove all growth.

Pathogenicity. — In cases of typhoid fever in man the
Bacillus typhosus is widely distributed in the body, in the
various tissues, and in the blood, from which it may be
obtained by cultivations made from at least 0-5 c.c. (see
" Clinical Diagnosis," p. 369). The bacillus is constantly
present in the blood from the commencement of the disease,
though not in large numbers, and cultures from the blood
in competent hands result in the recovery of the organism
in approximately 100 per cent, of the cases ; in the later
stages of the disease it is less frequently recovered.3 In
addition to being present in the Peyer's patches, mesenteric
glands, and spleen, the B. typhosus has been found in the
rose-spots of the eruption, in the sweat, in the sputum

1 Occasionally a feeble indole reaction may be obtained by careful
testing.

2 Trans. Fourteenth Internal. Cong, of Hygiene (Berlin, 1907), Bd. iv,
p. 34.

3 Coleman and Buxton, Amer. Journ. Med. Sci., June 1907.

PLATE XIL

a. Bacillus typhosus. Film preparation of a
pure culture. X 1500.

6. Gelatin culture of B.
typhosus, six days old.

c. Bacillus typhosus. Film preparation showing
flagella. x 1500.

BACILLUS TYPHOSUS 355

and lungs in the pulmonary complications, and in the
urine. In the urine it is so frequently present that special
disinfection should be practised, more particularly during
convalescence, and in some cases it may be so abundant as
to produce a turbidity (typhoid bacilluria) and cystitis.
It is also pyogenic, and occurs (usually in pure culture) in
concurrent or post-typhoidal complications, e.g. empyema,
abscesses, osteomyelitis, suppurating ovarian cysts,1 etc.
Clumps of bacilli in the gall-bladder have been suggested
as the nuclei of gall-stones, and the bacilli may be so
numerous in the gall-bladder and bile-ducts as to cause
cholecystitis and cholangitis. It is not easy to isolate the
organism from the stools, and plate cultivations on special
media must be employed, e.g. Conradi-Drigalski, malachite-
green, or brilliant- green, agar (see " Water ").

Injected intraperitoneally into mice and guinea-pigs
the B. typhosus usually produces death, and the same
result follows from intravenous injections in rabbits, but
the pathogenic effects so obtained are not specific. By
continuous cultivation it loses its pathogenic properties.
Given by the mouth no result follows, and the same is
the experience of most observers who have fed animals on
typhoid stools ; a disease process analogous to typhoid
fever in man has rarely been induced experimentally.
Remlinger 2 states that by feeding young rabbits on
vegetables, cabbage, etc., soaked in water, to which had
been added some culture of the typhoid bacillus, he has
succeeded in inducing a condition resembling typhoid fever
in man. The charts which accompany the paper show
a typical rise of temperature, a period of pyrexia with
morning remission, followed by a typical fall of tempera-
ture. The animals suffered from diarrhoea, and their
blood gave the agglutination reaction. Post mortem, the

1 Taylor, Jcurn. Obstet. and Gyncecol. Brit. Empire, Nov. 1907.

2 Ann. de rinst. Pasteur, xi, 1897, p. 822.

356 A MANUAL OF BACTERIOLOGY

intestine was congested and filled with yellow diarrhoeic
matter, the Peyer's patches were swollen and in some
places commencing to ulcerate. The spleen was increased
to two or three times its normal size, and cultures of the
typhoid bacillus were obtained from it. MetchnikofE 1
has infected the chimpanzee per os with typhoid faeces.

The proof of the causal relation of the Bacillus typhosus
to enteric fever is based on the following facts. It is met
with in the tissues in cases of enteric fever, can be obtained
from the spleen during life by puncturing with a hollow
needle, and may be isolated from the urine and blood
during the course of the disease, and is not met with in
other diseases. The writer has had under his care three
cases, and knows of several others, in which the disease
was almost certainly contracted in the laboratory from
working with pure cultures. The blood and blood-serum
of an animal immunised against the B. lyphosus are found
to bring about cessation of movement and agglutination
or aggregation of the bacilli in a broth culture of the
organism. A similar result occurs when the serum of a
patient, in the second week of an attack of typhoid fe^er,
acts on the B. typhosus, the reaction not occurring with
healthy individuals or in other diseases (Plate XIII. a).
This indicates that in the body of an individual suffering
from typhoid fever the same substances are formed as
in an animal artificially immunised by cultures of the
B. typhosus. This reaction is now recognised as a valuable
clinical test in doubtful cases of enteric fever (the " Widal "
or agglutination reaction 2).

The agglutination reaction. — For the method of carrying
out the agglutination reaction see p. 190. Normal serum
will generally agglutinate the typhoid bacillus in a dilution

1 See Ann. de VInst. Pasteur, xxv, 1911, p. 193.

2 Some controversy has arisen as to the discoverer of this reaction.
Griinbaum claims to have first observed it.

THE AGGLUTINATION REACTION 357

up to 1 in 3 or 4, but occasionally is more active. Dead
bacilli may be used. The reaction is not obtained before
the sixth or seventh day of fever, occasionally not until
much later. Very rarely the reaction seems to be inter-
mittent. The blood may retain its agglutinating power
for years after an attack, and inoculation with anti-
typhoid vaccine also confers agglutinative properties.
Cases do occur in which agglutination is absent throughout,
but they are rare and often tend to be severe and to ter-
minate fatally. Usually, if the blood during the course
of an attack fails to give a reaction when tested on three
occasions at intervals of three to four days, it is improbable
that the case is one of typhoid fever. Moreover, cases
occur, simulating typhoid closely, due to infection with
the so-called para-colon or para-typhoid bacilli. These
" para " bacilli belong properly to the Gartner group of
organisms (see p. 371). If a positive reaction be obtained, yet
the case does not seem to be one of typhoid, a previous attack
or inoculation with typhoid vaccine must be excluded. The
previous injection of a typhoid anti-serum into the patient
might induce a non-typhoid infection to give the reaction.

Gwyn 1 found that out of 265 cases diagnosed as typhoid
and accurately studied, only one persistently failed to
give the reaction. The blood of this case, however, reacted
typically with a Gartner-like organism obtained from the
blood (a case, therefore, of para- typhoid infection).

Johnson and McTaggart 2 found that typhoid blood dried
for sixty days still gave a typical agglutination reaction.
An incomplete reaction was occasionally obtained as early
as the end of the second day, and the complete reaction
was rarely delayed beyond the fifth day. They also
noticed that the blood of the horse often produced clump-
ing, etc., of typhoid bacilli, indistinguishable from an

1 Johns Hopkins Hosp. Bull., vol. viii, 1900, p. 387.

2 Brit. Med. Journ., 1896, vol. ii, p. 629.

358 A MANUAL OF BACTERIOLOGY

agglutination reaction with typhoid blood ; but the same
agglutinating effect was also produced on the colon bacillus.
Many chemical substances also produce agglutination of
typhoid bacilli, so that it is necessary to exclude them in
making a diagnosis. For example, corrosive sublimate
(0-7 : 1000), alcohol, salicylic acid, vesuvin, and safranin
(1 : 1000) agglutinate, while carbolic and lactic acids,
chloroform, caustic soda, and ammonia do not, the two last
only provided the test typhoid emulsion be made with dis-
tilled water. Safranin has a powerful agglutinating action
on the typhoid bacillus, but not on the colon bacillus.

While there is no constant connection between the
activity of agglutination and the severity of the disease,
active agglutination tends to go with cases which recover,
and cases in which agglutination is feeble or absent tend
to be severe.

Toxins. — From cultures of the typhoid bacillus Brieger
isolated a base which he termed typhotoxin, and which is
isomeric with gadinine. In animals it produced salivation,
profuse diarrhoea, paralysis, and death. Brieger and
Frankel isolated from cultures a toxic protein body.
Fen wick and Bokenham l extracted from spleens of
typhoid fever patients a proteose, an alkaloid, and a fatty
residue. The proteose produced fever, anorexia, and loss
of weight in guinea-pigs and rabbits, but the alkaloid and
fatty matter were without effect.

The toxins of the typhoid bacillus, however, seem to
be largely intra- cellular, and filtered broth cultures are
usually almost non-toxic. Sidney Martin 2 by cultivating
in a protein medium was able sometimes to obtain a
toxic nitrate, a few c.c. of which produced lowered tem-
perature, diarrhoea and death. Macfadyen and Rowland,3

1 Brit. Med. Journ., 1895, vol. i, p. 801.

2 Ibid. 1898, vol. ii, pp. 11 and 73.

3 Centr. f. Bakt., xxx, p. 753.

TYPHOID CARRIERS 359

by disintegrating large quantities of typhoid bacilli, filter-
ing, and so obtaining the intracellular constituents in
the filtrate, found that small doses of the latter produced
a transient rise of temperature in guinea-pigs and a loss
of weight which was soon recovered from. Animals so
treated were protected against a certain lethal dose of
typhoid bacilli, and their blood exhibited agglutinative
and bacteriolytic properties towards the typhoid bacillus.
Macfadyen * later obtained the intra- cellular juice of
typhoid bacilli by disintegration after freezing with liquid
air, and found it to be very toxic to guinea-pigs by intra-
peritoneal, and to rabbits by intra-venous inoculation.
The writer found that cultures of the Bacillus typliosus
do not give the " diazo " reaction.

Survival of the typhoid bacillus in the body. — Bacilli may
persist in the spleen for weeks, in the gall-bladder fol
years, and in suppurative lesions for six years or more.
Foster and Kayser obtained pure cultures irom the gall-
bladders of seven out of eight cases, and in 2 per cent, of
the cases this " cholecystitis typhosa " becomes a chronic
process, and typhoid bacilli may be discharged into
the bowel for long periods. Dean 2 found this to be
the case in a patient who had had enteric fever
twenty- nine years previously. Such "typhoid carriers"
have been the subject of much investigation recently.3
A. and J. Ledingham record three instances met with in
an asylum in which mysterious cases of typhoid had
occurred — 31 cases during fourteen years. Davies and
Walker Hall 4 relate similar outbreaks, the carrier in this
case being a woman who had suffered from enteric fever
in 1901, milk serving as the vehicle of transmission, and

1 Proc. Roy. Soc. Lond., B. Ixxi, 1902, p. 77.

2 Brit. Med. Journ., 1908, vol. i, p. 562.

3 See Ledingham, Rep. Med. Off. Loc. Gov. Board for 1909-10
(Bibliog.) ; ibid, for 1912-13, p. 336.

4 Proc. Roy. Soc. Med., vol. i, 1908, Epidemiolog. Sect., p. 175.

360 A MANUAL OF BACTERIOLOGY

a number of other instances have been recorded. Three-
fourths of the cases are women (and three-fourths of the
cases of gall-stones occur in women), and usually the
serum of the carriers gives a marked agglutination reaction,
and their stools frequently contain such large numbers of
typhoid bacilli that these largely replace the natural
bacterial flora of the intestine and may often be recovered
from the stools by simple plating. Firth's statistics give
an idea of the frequency of the development of the carrier
state. Of 1229 cases of enteric fever among the British
troops in India bacteriologically examined, 13 cases of
chronic carriers and 13 cases of temporary carriers were
detected. Obviously the typhoid carrier is a source of
serious risk to the community, and mysterious outbreaks
of enteric fever, ascribed by some in the past to a " de
novo " origin of the specific organism, become explicable.
Typhoid convalescents should be bacteriologically examined
before discharge from hospital and the negative cases
may with reasonable safety be allowed to resume their
civil life (Ledingham). The typhoid bacillus may occur
in the contents of ovarian cysts, usually causing suppura-
tion, and may survive for months — twelve in a case
recorded by Taylor l — after the attack of typhoid.

Survival of the typhoid bacillus outside the body. — The
Bacillus typhosus has been isolated in a few instances from
WATER SUPPLIES which have become infected, and have
given rise to epidemics, as in the case of the Lincoln
epidemic in 1905. 2 This is the exception, however, and
the isolation of the typhoid bacillus from an infected water
is a very difficult matter on account of the fact that the
bacillus may have died out before the investigation is
commenced, that it is generally in a small minority and
admixed with numbers of coliform organisms, and that

1 Journ. Obstet. and Gyncecol. Brit. Empire, November 1907.

2 Rep. Med. Off. Loc. Gov. Board for 1905-06.

SURVIVAL OF BACILLUS TYPHOSUS 361

until recently no medium was available which, inhibited
the growth of the coliform organisms without at the
same time inhibiting the growth of the B. typhosus. By
the use of malachite or brilliant green media, the last-
named difficulty seems to have been overcome (see section
on " Water ").

In sterilised waters, including distilled water, the Bacillus
typhosus maintains its vitality for upwards of a month,
and in some cases for much longer. The survival is not
necessarily longer in an organically polluted water than
in a pure water. Infecting sterilised Thames water (from
the Temple Embankment) and sterilised tap-water of the
Chelsea Water-works with typhoid cultures, the writer
found that, examining small quantities (1 c.c.) of the water,
the bacillus appeared to die out in the former in two to three
weeks, in the latter in four to five weeks.

The survival of the typhoid bacillus in natural waters
must be influenced by many circumstances — temperature,
chemical composition, struggle for existence with the
natural bacterial flora, etc., of the water. Experiments
by Russell and Fuller,1 in which the organism, suspended
in collodion sacs, was subjected to the action of lake water,
indicated that the maximum was eight to ten days.
Houston,2 using raw Thames, Lee, and New River waters
artificially infected with varying quantities of ordinary
laboratory typhoid cultures, and examining quantities of
100 c.c. of the water, found that in none of eighteen
experiments was a negative result obtained in four weeks,
and it was only after nine weeks that the typhoid bacillus
could not be isolated from this quantity in all the experi-
ments. But in subsequent experiments,3 in which typhoid
bacilli, obtained directly from the urine of a carrier case

1 Jown. Infect. Diseases, Sup. No. 2, February 1902, p. 40.

2 First Rep. on Research Work, Metropolitan Water Board, 1908.

3 Sixth Research Report, Metropolitan Water Board, 1911.

362 A MANUAL OF BACTERIOLOGY

by centrifuging and without culturing, were added to
the water, the number of bacilli was reduced by 99'99 per
cent, after a week, and after ten days the organism
could not be isolated from 100 c.c. of the infected water,
indicating that the uncultured bacillus rapidly dies in a
natural water, and that even a week's storage of water
affords enormous protection against water-borne typhoid.
In aerated (C02) waters the B. typhosus does not survive a
fortnight. The methods of isolation from water are given
in Chapter XXI.

The Bacillus typhosus may gain access to shell- fish,1
oysters, mussels, cockles, etc., particularly if obtained
from sewage-polluted laying. Such polluted shell-fish
may give rise to typhoid epidemics — as at Winchester
and Southampton in the case of oysters, and in the case
of cockles, derived from the Thames Estuary and imper-
fectly cooked, to typhoid cases. Buchan found that out
of 855 primary cases of typhoid fever occurring in house-
holds in Birmingham, 124, or 14-5 per cent., had a history
of mussel eating, and in seventeen instances the histories
were conclusive of mussel infection. Mussels, under
certain conditions (which are not well understood), are
liable to develop mytilotoxin, etc. (p. 38), which gives rise
to gastro- enteritis. Shell-fish from sewage-polluted layings
contain B. coli in varying numbers, but from uncontami-
nated layings are free from this organism, which may
therefore serve as an index of pollution (see " Examination
of Shell-Fish," Chapter XXI). Contaminated shell-fish,
removed to pure water, gradually cleanse themselves —
probably after two to three weeks' sojourn. Klein obtained
the typhoid bacillus from artificially infected oysters, kept

1 On pathogenic organisms in shell-fish see Reports by Bulstrode to
the Local Government Board, 1894 and 1911 ; Rep. Med. Off. Loc. Gov.
Board for 1899-1900, p. 574 ; Houston, Fourth Report of the Sewage
Commission., vol. iii, 1904 ; McWeeney, Loc. Gov. Board, Ireland, 1904 ;
Buchan, Journ. cf Hygiene, vol. x, 1910, p. 569.

SURVIVAL OF BACILLUS TYPHOSUS 363

in tanks of sea- water, after nine, sixteen, and even eighteen
days from the commencement of the experiment, the
oysters showing no abnormal condition.

As regards the vitality of the Bacillus typhosus in sewage
we have little certain information ; probably it tends to
die out within a few days. In sterilised sewage inoculated
with it the B. typhosus hardly multiplies at all, and at
the end of ten days has died out. Certain organisms in
sewage seemed to have a deleterious action on the B.
typhosus, hastening its extinction, viz. the B. fluorescens
Uquefaciens and B. fluorescens stercoralis. Russell and
Fuller, subjecting the bacillus to the direct action of
sewage, found the survival to range from three to five
days.

In dry garden earth, according to Dempster,1 the Bacillus
typhosus does not live longer than eighteen days (Firth
and Horrocks recovered it up to twenty- five days), and
in peat it dies within twenty-four hours. In moist soil,
however, the bacillus still survived on the forty-second
day. In an artificially dried soil it was not found alive
after the seventh day.

Sidney Martin found that in moist sterilised soil kept at
temperatures from 3° to 37° C., the B. typhosus maintains
its vitality for upwards of fifteen months, but that in
unsterilised soil it rapidly dies.2

Mair 3 concludes that the typhoid bacillus can survive
in natural soil in large numbers for about twenty days, and
is still present in a living condition after seventy to eighty
days, but that there is no evidence that it is capable of
multiplying and leading a saprophytic existence in ordinary
soil. He suggests that Martin's result (the rapid extinc-
tion of the bacillus in unsterilised soil) may be explained

1 Med.-Chir. Trans., vol. Ixxvii, 1894, p. 263.

2 Reps. Med. Off. Loc. Gov. Board for 1896-1901.

3 Journ. of Hygiene, vol. viii, 1908, p. 37.

364 A MANUAL OF BACTERIOLOGY

by the use of broth cultures for infection, the broth added
causing a multiplication of the saprophytes. Firth and
Horrocks l similarly conclude that the typhoid bacillus
displays no tendency to increase in numbers, nor to grow
upwards or downwards in soil, though it may be washed
by water through a thickness of 18 inches. Neither virgin
nor sewage- polluted soils differed much in these respects.

Vitality of B. typhosus in dust, fomites, etc. — Firth and
Horrocks found the B. typhosus to be alive in soil dry enough
to form dust for as long as twenty-five days, and consider
that infective material can be readily transmitted from
dried soil and sand by means of winds and air- currents.
Doubtless much depends on the degree of dryness of the
substratum. From khaki drill and serge inoculated with
cultures the bacillus was recoverable for from ten to twelve
weeks, and for from ten to seventeen days from the same
materials fouled with enteric faeces.

Semple and Grieg,2 with cloth and blanket infected with
typhoid urine, failed to obtain the bacillus after seventeen
days. This, however, was in India, and the survival of
the typhoid bacillus on fomites probably greatly depends
on the degree of drying of the material. A striking instance
of the conveyance of infection by fomites was that of the
blankets used in the South African War and brought to
this country, which gave rise to many cases of typhoid
fever.

Firth and Horrocks demonstrated that house-flies can
convey enteric infective material from specific excreta
or other polluted material to objects on which they settle
or feed, and the Commission which investigated the preva-
lence of enteric fever in the Spanish- American War ascribed
to flies the principal part in the dissemination of the disease
(see also p. 389).

1 Brit. Med. Journ., 1902, vol. ii, p. 936.

2 Sc. Mem. Gov. of India, No. 32, 1908.

SEWER GAS AND DISEASE 365

There has always been considerable discussion on the
exact relation of " sewer-gas " to disease. It is generally
held that sewer-gas is at least a predisposing cause to
enteric fever, diphtheria and tonsillitis. Some have
considered that the specific organisms are present in the
emanations from sewers, and this may occasionally be
the case. Thus Horrocks,1 in some experiments performed
at Gibraltar, by pouring sewage artificially infected with
typhoid culture down drains, showed that specific bacteria
present in sewage may be ejected into the air of ventilation
pipes, inspection chambers, drains and sewers by (a) the
bursting of bubbles at the surface of the sewage, (6) the
separation of dried particles from the walls of pipes,
chambers and sewers, and probably by (c) the ejection of
minute droplets from flowing sewage. " Sewer- gas " may
also lower vitality and increase susceptibility. Thus Alessi
found that animals exposed to drain emanations are at
first more susceptible to infection, but after a month or
so acquire tolerance and are no more susceptible than
animals kept under ordinary conditions. Exposure to
the gaseous emanations from putrefying matter is stated
by Trillat to increase the virulence of pathogenic bacteria.
There is no evidence that sewer-men or those employed at
sewage-works suffer from ill-health.

Action of heat, germicides, etc. — The B. typhosus in broth
culture is killed by a temperature of 53°-54° C. in half an
hour, and of 56°-60° C. in ten minutes. It is readily
destroyed by antiseptics. (See Table, Chap. XXII.)

Semple and Grieg (loc. cit.) found bright sunlight to be
germicidal in from two to six hours.

Wines and spirits have some germicidal action on the
typhoid bacillus. Champagne destroys the bacillus in
ten minutes, white wines in fifteen to twenty minutes,
red wines in thirty minutes or thereabouts. If diluted

1 Journ. Roy. San. Inst., May 1907, j>. 176.

MO A MANUAL OF BACTERIOLOGY

with water the germicidal action takes much longer to
accomplish, and the acidity, not the alcohol content, seems
to be the active factor. 1 Spirits, such as whisky or brandy,
if diluted with not more than one to two times the volume
of water, kill in ten to twenty minutes.

Anti-typhoid serum. — Attempts have been made to
prepare an anti-typhoid serum by inoculating horses with
increasing doses of typhoid bacilli, first killed (by heat,
chloroform, etc.) and then living, but such sera have
proved quite useless.

Macfadyen 2 prepared an endotoxic serum by treating
horses with the endotoxin obtained by triturating the
bacilli in the presence of liquid air. The writer continued
the work, and obtained a serum which gave promising
results.3

Chantemesse,4 by cultivating a virulent strain of the
typhoid bacillus in a special broth made with ox spleen,
heating the culture to 55° C., centrifuging and injecting
horses with the fluid, obtains a serum which he claims has
marked curative properties, the mortality being 4-3 per
cent., as against 17 per cent, for those subjected to ordinary
treatment. The patients receive very small doses of the
serum — five or six drops — and the dose is repeated only
two or three times. This dosage is quite different from
that of an ordinary antitoxic or antimicrobic serum, and
Wright suggested that toxins (and not anti-bodies) in the
serum may be the active agents. Chantemesse has
accepted this view, and the treatment, therefore, seems
to be a vaccine one.

1 Sabrazes and Marcandier, Ann. de Vlnst. Pasteur, 1907.

2 Proc. Ray. Soc. Lond., B, vol. Ixxi, 1903, pp. 76 and 351 ; Brit.
Mcd. Journ., 1906, vol. i, p. 905.

3 See Hewlett, Goodall and Bruce, Proc. Roy. Soc. Med., vol. ii,
1907-08 (Medical Section), p. 245 et seq. ; and Hewlett's Serum Therapy,
p. 220.

4 Trans. Fourteenth Internat. Cong. Hygiene and Demography, 1907.

ANTI-TYPHOID VACCINE 367

The disease has also been treated with a vaccine (con-
sisting of a killed culture) with promising results by Semple,
Smallman, Leishman, and others. The initial dose is
40-100 millions, and. the amount is cautiously increased
up to 300-400 millions.

Anti-typhoid vaccine. — Wright first prepared an anti-
typhoid vaccine by the following method.1 A typhoid
culture of moderate virulence (the virulence being kept
up by intraperitoneal passage through guinea-pigs) is
grown in peptone beef broth in flasks at 37° C. for from
fourteen to twenty-one days. The flasks are then so
heated that their contents attain, and remain at for a
few minutes, a temperature of 60° C. To obtain uniform
toxicity, the contents of several flasks should be mixed,
and to safeguard the vaccine from contamination one
twentieth of its volume of 10 per cent, lysol is added.
Various ingenious devices have been adopted by Wright
and Leishman to prevent contamination and for stan-
dardisation.

The immunising power of a typhoid vaccine depends
upon the number of bacilli it contains, and on the particular
strain of bacillus used. The vaccine is standardised by
counting the number of bacilli it contains by Wright's
method (p. 220). Leishman 2 now cultivates for about
forty-two hours, and the bacteria are killed by heating to
53° C. for one hour, the higher temperature having proved
to be deleterious, and after cooling 0*25 per cent, of lysol
is added ; it is not necessary to employ a virulent bacillus.
In the early days the symptoms produced by the inocula-
tion were often severe, but with more moderate methods
are now hardly appreciable. Two doses of the vaccine
should be given, with an interval of about ten days between
the two, the doses being 500 and 1000 millions respectively.

1 Wright and Semple, Brit. Med. Journ., 1897, vol. i, p. 256.

.2 gee Journ. Roy. Inst. Pub. Health, vol. xviii, 1910, pp. 385, 440, 513.

368 A MANUAL OF BACTERIOLOGY

The vaccine deteriorates on keeping. Emulsions of agar
cultures and autolysed cultures have also been used for
preparing vaccines.

Inoculation is now being extensively practised, and
Leishman (loc. cit.) gives the following statistics of its
value : total number under observation, 18,483-19,314 ;
average period under observation, twenty months ; number
inoculated, 10,378 ; number uninoculated, 8936 ; case-
incidence of enteric per 1000, inoculated 5'39 ± 0*48,
uninoculated 30*4 ± 1-23 ; case-mortality per 100, inocu-
lated 8'9, uninoculated 16'9. In the French navy Chante-
messe states that during nine months in 1912, among
67,843 unvaccinated persons 542 cases of typhoid fever
occurred, while among 3107 vaccinated ones not a single
case of typhoid occurred.

Variation of the B. typhosus. — Allusion has already been
made to T wort's work on the " education " of B. typhosus
to ferment lactose, and on the apparent conversion of
B. typhosus into B. alkaligines by Horrocks (p. 6). Penfold
also records variations in the fermentive powers of B.
typhosus (Journal of Hygiene, vol. xi, 1911, p. 30).

Relapses

Various hypotheses have been advanced to account for the
relapses which occur in typhoid and other diseases (e.g. Malta and
relapsing fevers). Chantemesse and Widal x showed that if the
B. typhosus is injected into an animal together with toxins of the
streptococcus, B. coli, or Proteus, its virulence is enhanced, or the
animal's resistance may be lowered. If, then, immunising and
bactericidal properties of the blood and tissues are but slightly
acquired during the attack, an absorption of toxic substances from
the alimentary tract may be sufficient to give the typhoid bacilli
still present a fresh start, and so produce a relapse. This Sanarelli 2
was able to do experimentally. Wright and Lamb formulated

1 Ann. de VInst. Pasteur, vi, 1892, p. 755.

2 Ibid, vi, 1892, p. 721 ; and ibid, viii, 1894, p. 193.

RELAPSES 369

another hypothesis.1 The organisms in typhoid, Malta, and re-
lapsing fevers, are deposited in the spleen and internal organs,
multiply and form colonies there, which become protected from
the bactericidal substances by the formation of a non-anti-bacterial
envelope. When the anti-bacterial substances in the blood and
lymph have increased to such an extent as to penetrate and abolish
the non-anti-bacterial envelopes which surround these colonies,
the production of toxins will be so diminished that the temperature
will fall. If, however, for some reason or other, even a single
colony escapes the full anti-bacterial power of the lymph, owing, it
may be, to being shut off in a capillary which has become blocked,
or in some other part not freely infiltrated by the blood- or lymph-
streams, the bacteria of this colony will go on multiplying until the
blood has become modified in such a manner as to bring about a
diminution of the anti-bacterial substances, and thus render a
relapse possible.

A third theory has been suggested by Durham.2 He regards a
given infection as due to the " result of the action of a sum of a
number of infecting agents, each of which is similar but not identical
in its nature," the apparently simple infection being " in reality a
complex phenomenon brought about by a number of varieties and
sub-varieties of the given microbe." He suggests, therefore, that
in a typhoid infection a particular race of typhoid bacilli is in excess,
and when the anti-bodies for this particular race have been formed
in sufficient quantity, the disease process comes to an end. There
may, however, be present at the same time other races which have
produced little of their specific anti-bodies ; these then begin to
grow and multiply, and a relapse ensues.

In the case of relapsing fever the organism may be a protozoon,
and in protozoal diseases relapses coincide with developmental cycles
of the parasite, e.g. in malaria.

Clinical Diagnosis

(1) Blood cultures. — Three to 5 c.c. of blood are withdrawn
from a superficial vein with a syringe with aseptic precautions,
and 0-5 c.c. of the blood so obtained is sown into each of several
tubes containing 15 to 20 c.c. of sterile broth. The tubes are
incubated at 37° C., and if organisms develop these are isolated and

1 Lancet, 1899, vol. ii, p. 1727 ; Sc. Mem. Med. Officers of Ind. Army,
pt. xii.

2 Journ. Path, and Bact., vol. vii, 1901, No. 2, p. 240.

24

370 A MANUAL OF BACTERIOLOGY

examined culturally for the typhoid bacillus. Coleman and Buxton
recommend the following culture medium : Ox-bile, 90 c.c., glycerin
10 c.c., and peptone 2 grm. Distribute in small flasks, 20 c.c. in
each, and sterilise. Each flask is inoculated with 2 to 3 c.c. of
blood, incubated for eighteen to twenty-four hours, then streaks
from each are made on to litmus lactose agar plates, which are
incubated for a few hours. If the growth does not redden the medium
and a typhoid-like bacillus is present, it is tested for agglutination
with typhoid-immune serum.

(2) Agglutination reaction. — This is carried out by the micro-
scopic or the macroscopic (sedimentation) method described at
p. 190. Dilutions of 1 : 30, 1 : 50, and 1 : 100 should be made.
The microscopic method is the more rapid. Various apparatus
(agglutinometers) can be obtained, consisting of measuring devices
and a supply of dead culture, with which the sedimentation test
can be carried out by any one, but are unsatisfactory in the tropics.

(3) Ophthalmo-diagnosis. — Chantemesse (loc. cit.) has devised a
method analogous to the ophthalmo-diagnosis for tuberculosis
(p. 330). The material is prepared from agar cultures of typhoid
which are emulsified, dried, triturated, and extracted, and the
extract is precipitated with absolute alcohol and dried (for details
see Hewlett's Serum Therapy (p. 382). The dry substance is
powdered in an agate mortar, and for use 8 to 10 mgrm. are dissolved
in 1 c.c. of sterile water. Of this solution a drop is instilled into the
conjunctival sac ; in a case of typhoid, after a lapse of two to three
hours the conjunctiva becomes red and there is a sensation of heat,
after six to ten hours there is a marked conjunctivitis, which may
persist for one to three days and then passes off. In healthy
persons and in other diseases no conjunctivitis ensues. A cutaneous
reaction has also been devised.

(4) Puncture of the spleen with a sterilised hypodermic needle and
syringe. — A little of the blood and pulp is withdrawn with the
syringe, and cultivations are made as in (1). This method seems
hardly justifiable, and now that the blood-culture method and
agglutination reaction have been introduced should be discarded.

(5) Examination of pus. — Cultivations may be made as in (1)
if the bacillus is present, apparently in pure culture. If not, plate
cultivations, preferably on litmus lactose agar, Conradi-Drigalski,
malachite- or brilliant-green agar, may be prepared (see " Water ").

(6) Examination of the stools. — This is hardly practicable for
clinical diagnosis ; it takes too long, is tedious and uncertain.
Plate cultivations from the dilated stools are made on Conradi-
Drigalski, malachite- or brilliant-green, agar (see "Water").

THE GARTNER GROUP 371

The Gartner or Enteritidis Group of Bacilli

The Gartner group of bacilli, of which the type is the B. enteritidis
of Gartner, are bacilli morphologically resembling the B. typhosus,
i.e. they are pleomorphic, actively motile, • multi-flagellate, non-
sporing, and non-Gram-staining, but culturally are intermediate
between B. typhosus and B. coli. Thus, like B. coli, they ferment
glucose with the production of gas and acid and change neutral
red ; like B. typhosus they do not attack lactose and do not curdle
milk. In litmus milk they usually first produce slight acidity,
followed after three to four days by a change to alkalinity, and the
milk ultimately becomes limpid. The fermentation reactions of
some members of the Gartner group are given in the Table on p. 381.
The organisms of the Gartner group may be divided into four sub-
groups :

1. Enteritidis group. — Produce acute gastro-intestinal disturbance
in man. The cause of epidemic meat-poisoning, e.g. the B. enteritidis
of Gartner.

2. Pneumonic group. — Produce pneumonic symptoms in man.
The cause of some outbreaks of epidemic pneumonia, e.g. B. psitta-
cosis.

3. Paratyphoid group. — Produce a disease resembling typhoid
fever in man. May also produce "food-poisoning" with gastro-
enteritis. Subdivisions A or a and B or /3.

4. Group non-pathogenic to man, e.g. B. typhi murium.

The Bacillus enteritidis

A number of outbreaks of what has been termed " epi-
demic meat poisoning " have been traced to infection
with the B. enteritidis. (See also " Food Poisoning,"
Chap. XXL) The disease takes the form of an acute
gastro- enteritis — urticaria, abdominal pain, vomiting, diar-
rhoea, nervous symptoms and collapse — occurring from
eight to thirty-six hours after partaking of a meat meal,
usually pork (sausage, pork-pie, ham), occasionally beef
and tinned meat. The principal outbreaks of this nature
have been those at Jena, in 1888, investigated by Gartner,
and from which he isolated the type form of the B. enteri-

372 A MANUAL OF BACTERIOLOGY

tidis ; Welbeck in 1880 ; Middlesborough in 1888 ; Mans-
field in 1896 ; and Derby in 1902. A small outbreak
occurred at Bedford in 1907.1 These outbreaks are usually
caused by varieties of the B. enteritidis having the general
characters of the group, which usually do not ferment
lactose, and are distinguishable by agglutination reactions
and fixation tests, the organism isolated as a rule agglu-
tinating well with the patient's serum.

The B. enteritidis in morphology, motility, and staining
reactions resembles the B. typhosus, forms no, or only
traces of, indole, and changes neutral red to a fluorescent
yellowish colour. Litmus milk after a faint acidity becomes
alkaline, and is converted into a thin watery translucent
fluid, without coagulation. It does not attack either
salicin or glycerin. The fermentation reactions are given
in the Table on p. 381. Savage2 obtained this organism
from only one out of fifty-three specimens of human
excreta examined. A number of variants were isolated
from various materials, some fermenting salicin, some
glycerin, and some both these substances (see " Meat,"
Chap. XXI).

Swine Fever or Hog Cholera3

Swine fever, or hog cholera (to be distinguished from swine
erysipelas, which see), is an infective disease of pigs, highly con-
tagious, and causing considerable mortality. The duration of the
affection is usually three to four weeks ; the animals lie about,
their temperature is raised, and they may suffer from cough and
frequent respiration, and some lameness in the hind legs. Towards
the end mucous diarrhoea is a prominent symptom. Post mortem,
the large intestine is found to be ulcerated, the ulcers much
resembling the typhoid ulcers of man, and according to Klein,

1 PuUic Health, vol. xx, 1907-8, p. 310.

2 Rep. Med. Off. Loc. Gov. Board for 1909-10, p. 446.

3 See Uhlenhuth, Trans. Fourteenth Internal. Cong, of Hygiene
(Berlin, 1907), Bd. iv, p. 50 ; Journ. Roy. Inst. Pub. Health, 1911.

SWINE FEVER 373

pneumonia is commonly present, whence he termed the disease
" pneumo -enteritis." McFadyean, however, from his own experi-
ence and that of the Board of Agriculture, considers pneumonia
very infrequent. The ulcers occur mainly in the caecum and colon,
and are due to a well-defined circular necrosis involving the whole
thickness of the mucous membrane and occasionally extending to
the wall of the bowel. A diffuse diphtheroid lesion also occurs, due
to a superficial necrosis with deposition of a thin layer of fibrinous
exudate on the surface of the mucous membrane. All gradations
are found between the well-defined circular necrosis and the diffuse
diphtheroid lesion.

An organism constantly present is a member of the para- typhoid
sub-group of the Gartner group (B. suipestifer oisuicholerce, apparently
identical with B. aertryck), but it seems to be a terminal infection
and not the true ctiological agent, as the blood and tissues filtered
through a porcelain filter are still infective — i.e. the organism is
probably ultra-microscopic. Some confusion exists in the nomen-
clature of the disease. Swine fever is the British, and hog cholera
the American, name. In addition, a disease of swine was formally
described under the designation "swine plague" ("Schweine-
seuche," Schiitz). This clinically much resembles swine fever, but
pneumonia is a prominent lesion, and a non-motile, stumpy,
bi-polar staining bacillus belonging to the group of the haemorrhagic
septicaemia bacilli is present (see under "Chicken Cholera").
This is now regarded as a secondary infection and the disease as
being really swine fever. The B. suipestifer is apparently identical
with the B. ictero'ides of Sanarelli. (See also Chap. XIX.)

Although the lesions are very similar, swine fever has nothing to
do with typhoid fever of man, nor with ulcerative colitis.

Other organisms belonging to the Gartner group are :

1. The Danysz bacillus, used as a virus for exterminating rats
(the Danysz virus).

2. The B. icteroides of Sanarelli, supposed by him to be the cause
of yellow fever, but apparently identical with the B. suipestifer (see
" Yellow Fever," Chap. XIX).

3. The B. typhi murium of Loffler, used as a virus for exterminating
mice.

4. The B. psittacosis of Nocard, causing an infective disease of
parrots and transmissible to man (bird-fanciers, etc.), in whom it
produces a severe and often fatal broncho-pneumonia.

5. Summer diarrhoea. — Morgan x concluded that the summer or

Ri * Brit. Med. Journ., 1906, vol. i, pp. 908 and 1131 ; ibid. 1907,
vol. i, p. 16.

374 A MANUAL OF BACTERIOLOGY

epidemic diarrhoea of infants is not caused by the dysentery bacillus
(see p. 379). In 50 per cent, of the cases he isolated a motile
bacillus producing acid and gas from glucose which appears to be
most closely allied to the hog-cholera bacillus, differing from the
latter by producing alkalinity in litmus milk (without previous
acidity) and much indole, and by failing to produce acid and gas
from mannitol, arabinose, maltose, and dextrin. It does not fer-
ment dulcitol, saccharose, salicin and sorbite. There are two
variants, designated as No. 1 and No. 2. Eyre and Minett 1
examined the normal faeces of sixty young children, and in four
only isolated a bacillus allied to the Morgan bacillus. The method
of isolation was by means of plates of bile-salt agar containing
1 per cent, of mannitol and coloured with neutral red. (See also
Chap. XX.)

Para-typhoid Fever 2

The name " para-colon " bacillus was given by Gilbert in 1895
to races of bacilli intermediate in type between the typhoid bacillus
and the colon bacillus, and this designation was also applied by
Widal and Nobecourt to a bacillus isolated by them from an
abscess in the neighbourhood of the thyroid. The name " para-
typhoid " bacillus appears first to have been used by Archard and
Bensaude in 1896, and was reintroduced by Schottmiiller in 1901,
and would seem to be the preferable designation for those micro-
organisms that produce typhoidal symptoms.

Para-typhoid fever may be defined as a disease much
resembling typhoid fever in its clinical aspect, which is,
however, caused, not by the typhoid bacillus, but by
organisms belonging to the para-typhoid sub-group of the
Gartner group of bacilli. Para-typhoid infections some-
times occur in epidemics, may be spread by drinking-
water and by " carriers," and occur in all parts of the
world.

Para-typhoid bacilli are also occasionally the pathogenic
agents in cases of " food poisoning " with gastro- enteritis,
particularly B. suipestifer (or aertryck).

1 Brit. Med. Journ., 1909, vol. i, p. 1227.

2 See Savage, Rep. Med. Off. Loc. Gov. Board for 1COS-9, p. 316 ;
JBainbridge and O'Brien, Journ, of Hygiene, vol. xi, 1911, p. 68 (Bibliog,).

PARA-TYPHOID FEVER 375

The para-typhoid bacilli are morphologically like the
typhoid bacillus and are actively motile, but they ferment
glucose with the production both of acid and of gas. A
number of races have been isolated differing from one
another in their source, rate of fermentation of glucose,
action on milk, action on neutral red, and agglutination
reaction, and are distinguished by the names of those who
isolated them.

Two groups of para- typhoid bacilli may be distinguished
which have been termed A and B by Buxton. Group A
produces less gas in glucose media than group B ; with
group A milk remains permanently acid ; with group B
it becomes alkaline after a transient acidity ; and though
group A changes neutral red to yellow, the red colour tends
to return after three weeks or so, while with group B the
yellow colour is permanent. That is to say, in its reactions
group A is more closely allied to the typhoid bacillus than
is group B.

B. para-typhosus A or a is rarely found, the vast majority
of cases of para-typhoid fever being associated with the
presence of the B or p type. The fermentation reactions
of some of the para-typhoid bacilli are given in the Table
on p. 381.

As regards the agglutination reaction, the blood of the
para-typhoid fever patient either does not agglutinate
the typhoid bacillus or agglutinates it only in low dilution
— e.g. 1 in 10 to 40, while it agglutinates the para-typhoid
bacilli in far higher dilution — e.g. I in 100 or 200, or even
higher ; thus in Cushing's case the patient's serum agglu-
tinated the para-typhoid bacillus isolated from it up to
1 in 8000.

The diagnosis of para-typhoid fever would be based
on (a) the agglutination reaction ; (6) the isolation of a
para-typhoid bacillus by cultures from the blood (p. 369).
Prophylactic vaccines for para-typhoid fever may be

376 A MANUAL OF BACTERIOLOGY

prepared with para-typhoid bacilli in the same manner as
for typhoid fever and Castellani has made use of a mixed
typhoid-para-typhoid vaccine.

Bacillus dysenteriae l

In one type of dysentery, the so-called epidemic or
bacillary form (see " Dysentery," Chap. XX), a bacillus
B. dysenteries, is the causative agent. The B. dysenteries
includes a group of closely allied organisms.

The dysentery bacillus was first isolated in 1897 by
Shiga in Japan. Somewhat later Kruse isolated an almost
identical bacillus in Germany, and this type is known as
the Shiga-Kruse type. Later, Flexner and Strong isolated
another type of the dysentery bacillus, and during the
last few years similar organisms, but differing from the
Shiga-Kruse and Flexner types in some of their fermenta-
tion and other reactions, have been isolated ; these are
sometimes termed " pseudo-dysentery " bacilli.

The Shiga-Kruse and other types of dysentery bacilli
have been isolated by Flexner and Strong in the Philip-
pines, Park, Duval, Bassett, Martini, Hiss, Russell and
others in the United States, Castellani in Ceylon, Rogers
and others in India, RufTer and Willmore in Egypt (El
Tor), and Eyre, McWeeney and others in the British Isles.

Morphology. — The B. dysenterice are small slender bacilli
much resembling the colon bacillus. They are non-motile,
but Brownian movement is often active,2 Gram-negative,
and non-sporing, and are readily destroyed by heat (58°
60° C.) and antiseptics.

Cultural characters. — The dysentery bacilli are aerobic
and facultatively anaerobic. On agar a thinnish creamy

1 See an excellent summary by P. H. Bahr, Dysentery in Fiji
(Witherby & Co., London, 1912).

2 Flagclla have been described by some observers, but cannot usually
be demonstrated.

BACILLUS DYSENTERIC 377

growth develops ; on gelatin a white growth nearly limited
to the inoculation track, and without liquefaction. The
colonies on a gelatin plate closely resemble those of the
typhoid bacillus. On potato the growth is either thin,
grey and slightly visible, or thicker and yellowish or
brownish. The colour of neutral red media is unaltered.
Litmus milk first becomes faintly acid, then markedly
alkaline ; no clotting. Indole is generally not formed
(never by the Shiga type) ; occasionally a trace may be
detected. All strains ferment glucose with the formation
of acid only, no gas ; none ferment lactose. Some strains
(the Flexner type) ferment mannitol with the formation
of acid only, no gas ; other strains (the Shiga-Kruse type)
have no action on this alcohol. The principal fermenta-
tion and other reactions are given in the Table on p. 381.
These reactions are very variable with different stains,
but differentiation may be accomplished by agglutination,
saturation, and complement fixation, tests. Shiga x dis-
tinguishes five groups of dysentery bacilli as follows :

1. Fermenting dextrose alone [Shiga, Kruse, Flexner

(Newhaven)].

2. Fermenting dextrose and mannitol (Hiss and Russell's

Y bacillus, Ferran, Seal Harbour bacillus).

3. Fermenting dextrose, mannitol and saccharose

[Flexner, Strong (Manila)].

4. Fermenting dextrose, mannitol, maltose and saccha-

rose (Harris, Gay, Woolstein).

5. Fermenting dextrose and maltose, and giving a feeble

acid reaction with mannitol (Shiga).

Bahr found occasional variations in fermentive power
after sub-culturing and after a sojourn in flies.

Agglutination reaction. — The agglutination reaction is
given by the blood of patients suffering from the bacillary

1 Zeitsch. f. Hyg., Ix, 1908, pp. 75, 120.

378 A MANUAL OF BACTERIOLOGY

form of dysentery, but not by the amoebic form (unless a
double infection be present, which occasionally is the case).
The agglutination reaction is obtained in dilutions of 1
in 10 to 1 in 100, but may occur only with the particular
strain causing the infection.1 Thus by the agglutination
reaction variations between different strains of the B.
dy sentence may be detected.

Pathogenic action. — The organism seems limited to the
bowel and its mucous membrane and does not gain access
to the blood. No characteristic lesions are produced in
animals by administration of the dysentery bacillus per os.
In man, cultures given by the mouth are stated to have
induced a typical dysentery. Animals such as rabbits,
guinea-pigs and mice are very sensitive to injections of
living and killed cultures ; in fact, it is very difficult to
immunise animals against the organism. Amounts of
0*1-0'2 mgrm. of an agar culture given intravenously
or intraperitoneally are fatal to these animals.

In man the organism is abundant in the bloody mucoid
discharge from the bowel, and at an early stage is easy to
isolate by means of Conradi-Drigalski agar plates, on which
it forms small transparent blue colonies ; at a later stage
(after two to three days) the other organisms in the bowel
multiply to such an extent that isolation may become
very difficult. " Carriers " occur and help to spread the
disease, which may be conveyed by infected water and
food and by flies.

Toxins. — The nitrate of dysentery cultures (four to six
weeks old) in a somewhat highly alkaline broth (broth
just alkaline to litmus + 7 c.c. normal NaOH per litre) is
markedly toxic, O'l c.c. being a fatal dose for a large
rabbit.2

Anti-serum and vaccine. — The serum of horses immu-

1 See Hewlett, Trans, Path. Soc. Lond., vol. Iv, 1904, p. 51.

2 Todd, Journ. of Hygiene, vol. iv, 1904, p. 480 (Bibliog.).

THE COLON BACILLUS 379

nised with the toxin, or with dead and then with living
cultures, possesses marked antitoxic properties, and the
use of this antitoxic serum has been successful in cases
of acute bacillary dysentery. Shiga obtained a reduction
in mortality from 22 to 7 per cent, by the use of serum
in a severe epidemic, and striking results were obtained
by Buffer and Willmore1 in Egypt and by Bahr in Fiji.
It is necessary, however, to employ a serum prepared with
the particular strain of the disease.

When the disease has become chronic the use of a
vaccine, consisting of a culture sterilised by heat, is some-
times beneficial. Castellani also suggests the use of a
vaccine for prophylactic purposes.

Para-dysentery bacilli. — In the dysenteries of Ceylon,
Castellani 2 has sometimes isolated dysentery bacilli nearly
related to the Shiga-Kruse type, but showing differences
from it in agglutination, persistence of acid reaction in
litmus milk, and virulence ; these he has termed " para-
dysentery " bacilli.

Asylums dysentery and summer diarrhoea of infants.—
Both in America and in England some cases of summer
diarrhoea of infants are found to be associated with the
B. dysenteries (see above, p. 374). The asylums or insti-
tutional dysentery, or ulcerative colitis, is also due to
this organism, and the blood of patients gives the agglu-
tination reaction.3 In both instances the B. dysenteries
present is of the Shiga-Kruse type.

Bacillus coli

The Bacillus coli, or colon bacillus (B. coli communis),
is an organism of considerable importance, both in con-

1 Brit. Med. Journ., 1909, vol. ii, p. 862, and 1910, vol. ii, p. 1519.

2 Journ. of Hygiene, vol. iv, 1904, p. 495.

3 Hewlett, Trans. Path, Soc, Lond,, vol. Iv, 1904, p. 51.

380 A MANUAL OF BACTERIOLOGY

nection with the Bacillus typhosus, in pathological pro-
cesses, and in water supplies as an indication of pollution.
As its name implies it is a constant inhabitant of the
intestinal tract in man and animals (except perhaps in
certain arctic animals), and is one of the most widely
distributed organisms in nature. While the term " colon
bacillus " is applied to a fairly well-defined organism (the
" typical B. coli "), there are a number of allied organisms
differing from the type in one or more characters — e.g.
motility, indole production, fermentation reactions, rate and
extent of milk curdling, etc. — and these varieties are said to
belong to the " colon group," or are termed " coliform."

The B. coli may be readily isolated by inoculating litmus
lactose bile-salt peptone-water tubes with a trace of a
suspension of fresh faeces, growing for from twenty-four
to forty- eight hours at 42° C., and plating the culture on
litmus lactose agar, on gelatin, or on Conradi-Drigalski
agar, or by direct plating of the faeces suspension on the
last-named medium (see also " Water ").

Morphology. — The B. coli is a short rod with rounded
ends, 2 or 3 /x long and 0*5 /m broad, frequently linked in
pairs or more. It is often so short that it is merely ovoid
in shape ; and, on the other hand, longer individuals and in-
volution forms occur 10 /UL or more in length (Plate XIII. 6).
It is feebly motile, and possesses lateral flagella to the
number of three or four on an average, which are
shorter and straighter than those of the typhoid bacillus.
It is sometimes met with in diplococcoid form, which by
cultivation in ascitic fluid may become fixed. Capsulated
forms have been described.

Spore-formation does not occur, but vacuolation may
sometimes be observed. The organism stains well by
the ordinary anilin dyes, but is Gram-negative.

Cultural characters. — The B. coli is aerobic and faculta-
tively anaerobic, and grows readily on the ordinary culture

FEKMENTATION REACTIONS

381

I 1 + + 1 +

g

o osoonio

+ +1 +

+ 1 1 + +

a.

<2

+ H- :

I I I I I I I I I I I I +

>IIIK snui^ii

* ^*

+ i i

I 1 I I

I I I I I I + + I + +

I I

I i + + + + +

SHI I

-

S 1

i

i 8

1 ^ I 1 II-

B.
B.

ftj flq cq oq Bq cq eq

PQ pq

382 A MANUAL OF BACTERIOLOGY

media from 20° to 42° C. In gelatin plates the colonies
are visible in twenty-four to forty-eight hours. The deep
colonies are spherical, granular, and of a pale brownish
colour, darker at the centre than at the periphery. The
superficial colonies are at first punctate, round and almost
transparent, but subsequently spread on the surface and
may attain a diameter of 3 mm., their margins become

FIG. 41. — Colonies of the colon bacillus, superficial and deep.

irregular, the surface is smooth, they are finely granular,
opalescent in appearance, and are thicker at the centre
than at the periphery (Fig. 41). On a gelatin streak a
copious white, shining, smooth growth develops, the mar-
gins of which are irregular and crenated (Plate XIII. c),
and in old cultures the medium becomes opalescent. In
a gelatin stab-culture a white growth develops along the
line of inoculation with one or more gas-bubbles. The
gelatin is not liquefied. On agar and on blood-serum a
thick, moist, shining, greyish layer forms. There is
abundant formation of gas in a stab-culture in glucose-agar

PLATE XIII.

a. The agglutination reaction. A clump of typhoid bacilli.
X 1500.

b. Bacillus coli. Film preparation from a pure c Gelatin culture of
culture, x 1000. B. coli, six days old.

THE COLON BACILLUS 383

and in gelatin shake cultures (Fig. 42), provided the latter
medium be made with meat ; " lemco " gelatin, however,
generally fails to give gas. On acid potato it forms a
straw-yellow or brownish-yellow, moist, thick growth, but
if the potato is not fresh and acid in reaction the growth
may be colourless. Milk is a good culture medium, and
is curdled in twenty- four to seventy- two hours. This

FIG. 42. — Colon bacillus. Gelatin shake culture showing gas
production.

curdling is principally due, not to an enzyme, but to the
formation of a considerable amount of lactic acid, though
a milk-curdling enzyme has been described by Savage 1
as being formed under certain conditions. The gas which
is produced in culture media under anaerobic conditions
consists of hydrogen and carbon dioxide. Under aerobic
conditions marsh gas is stated to be also formed. The
ratio of H to C02 is about 2 : 1 for dextrose and lactose.
In broth it produces a general turbidity without film
formation, and the culture gives the indole reaction on

1 Journ. Pathol. and Bact., November 1904.

384 A MANUAL OF BACTERIOLOGY

the addition of a nitrite in twenty-four to forty-eight
hours.

The fermentation reactions are given in the Table,
p. 381 . It will be seen that the B. coli is an active fermenter
of many carbohydrates, alcohols, and glucosides,1 e.g.
glucose, lactose, galactose, mannitol and dulcitol, but not
of adonit. Cane-sugar may or may not be fermented ;
sometimes only acid is formed, sometimes both acid and
gas are produced. To the variety producing both acid
and gas from cane-sugar Durham gave the name B. coli
communior. Prescott and Winslow consider that the term
B. coli should be applied only to an organism that does not
attack ketonic sugars. Neutral red in glucose broth is
changed to a fluorescent yellow, and Houston describes
a typical B. coli as " flaginac," i.e. producing fluorescence
in neutral red glucose pep tone- water (fl), acid and gas from
glucose (ag), indole in pep tone- water (in), and acid and
curd in milk (ac). The colonies on Conradi-Drigalski agar
are large and red (see " Water "). The B. coli does not
give the Voges-Proskauer reaction (p. 389).

The differentiation of the B. coli from the B. typhosus
should present no difficulty if the morphology and motility
of the organisms and their fermentation and agglutination
reactions be compared. Bacteriologists usually make use
of the following tests for the differentiation of B. coli :
(1) Morphology, (2) motility, (3) Gram staining, (4) char-
acter of growth and colonies on gelatin, (5) non-lique-
faction of gelatin, (6) action on milk, (7) indole formation,
(8) fermentation of glucose, (9) fermentation of lactose
and saccharose, (10) action on neutral red. MacConkey
suggests that instead of tests Nos. 4, 6, 7, 8, and 10, the
following should be substituted : (a) fermentation of
dulcitol, but not of adonit and inulin ; (6) the Voges-
Proskauer reaction.

1 Sec Twort, Proc. Roy. Soc. Lond., B, vol. Ixxviii, p. 329.

THE COLON BACILLUS 385

Other media which have been recommended for the differentia-
tion of B. coli from JB. typhosus are the Proskauer-Capaldi media
and Petruschky's litinus whey, but are not now much used.

The Proskauer-Capaldi medium No. 1 is an asparagin-mannitol
solution with certain salts ; medium No. 2 is a peptone-water -
mannitol solution. Both solutions are carefully neutralised and
tinged with litmus.

If these media be inoculated with B. typhosus and B. coli respec-
tively and incubated at 37° C. for twenty-four hours, the following
changes will be noted :

Medium No. \. Medium No. 2.

B. typhosus No growth or change Growth with strongly

in reaction. acid reaction.

B. coli . Growth with acid Growth with neutral

reaction or faintly alkaline

reaction.

Petruschky's litmus whey is prepared as follows : Fresh milk is
warmed and the casein precipitated by the addition of a minimal
amount of hydrochloric acid. It is filtered, and the filtrate of clear
whey is carefully neutralised with dilute caustic soda solution.
The fluid is then steamed for two hours and filtered ; the filtrate
should be clear, colourless, and neutral in reaction. Enough neutral
litmus solution is then added to render it well coloured, and the
mixture is distributed into test-tubes and sterilised. This medium
is rendered slightly acid (represented by 6-10 c.c. N/ 10 caustic soda
per cent.) by B. typhosus, very acid (40-50 c.c. ditto) by B. coli.

The thermal death-point of the organism, according
to Weisser and Sternberg, is 60° C. with an exposure of
ten minutes. The B. coli will grow freely in a slightly
acid medium, and in media containing as much as 0*15
per cent, of carbolic acid. In this respect it is a more
resistant organism than the B. typhosus.

Chemical products. — The acids produced are mainly
laevo-lactic acid with some dextro-lactic acid from glucose,
laevo -lactic acid only from mannitol ; also acetic, formic
and succinic acids, and alcohol. According to Harden,
B. coli attacks glucose in a characteristic manner, each
molecular proportion of sugar yielding half a molecular
proportion of acetic acid and of alcohol, and one molecular

25

386 A MANUAL OF BACTERIOLOGY

proportion of lactic acid, together with a small amount of
succinic acid, and gaseous carbonic acid and hydrogen.1
Nitrates are reduced to nitrites.

No toxin, or a trace only, is formed in cultures, but
the dead bacilli are toxic and pyogenic, and a toxin is
obtained by autolysis of cultures or by triturating the
bacilli with liquid air (Macfadyen).

Vaughan,2 by washing large quantities of colon and
typhoid bacilli, extracting the bacterial cells first with
alcohol, then with ether, and then digesting the ground
residue with alcohol containing 2 per cent. NaOH, states
that two constituents are obtained, one soluble in alcohol
and toxic, the other insoluble in alcohol and non-toxic.
The latter confers a certain degree of immunity on animals
injected with it.

Pathogenicity. — The pathogenic action and pathogenicity
of the B. coli are very varied. Introduced into the circu-
lation or into the peritoneal cavity in guinea-pigs or rabbits
it usually causes death in from one to three days with a
general septicaemia. Some varieties are, however, non-
virulent to animals.

In man the colon bacillus is associated with a number
of important pathological processes. It is usually the
organism causing the peritonitis which is due to infection
from the intestine, as in hernia with obstruction or per-
foration, in ulceration of the bowel and enteritis, in can-
cerous growths, and affections of the appendix, biliary
canals, and gall-bladder. The exudation in these cases
is often characteristic ; at first it is clear and greenish,
it then becomes greenish-yellow, thin, semi-opaque and
foul-smelling, and finally purulent. The colon bacillus
may pass through the intestinal wall where it has been
damaged, but not yet perforated, as in strangulation.

1 See also Eevis, Cenlr. f. Bakt. (2t° Abt.), xxvi, 1910, p. 161.

2 Trans. XIV Internal. Cong. Hygiene (Berlin, 1907), Bd. iv, p. 28.

PATHOGENICITY OF COLON BACILLUS 387

The B. coli is a pyogenic organism, and has been met
with in ischio-rectal abscesses (probably the B. pyogenes
fetidus of Passet). Possibly it causes in some instances
the pneumonia and pleurisy occurring after peritonitis,
for it has been obtained from the lung and pleura in these
conditions, but it must be recognised that the B. coli
is a common secondary or terminal infection. B. coli
sometimes induces puerperal fever and other forms of
septicaemia and it is a common cause of cystitis and other
infections of the urinary tract.

In the Pictou cattle disease, characterised by extensive
hepatic cirrhosis, Adami found a minute diplococcus or
short bacillus. A similar form was afterwards isolated
by him in hepatic cirrhosis in man. Miss Abbot,1 from
a study of several such cases, came to the conclusion
that this organism is a variety of the B. coli. It has been
suggested that hepatic cirrhosis is produced by poisons
or toxins, e.g. of the B. coli, and that alcoholism, the usual
cause assigned, is but an exciting or secondary agent.

Anti-serum and vaccine. — Attempts have been made to
prepare an anti-serum for B. coli infections, but they have
met with little or no success.

A vaccine prepared by sterilising cultures by heat and
standardising has been used successfully in the treatment
of chronic B. coli infections, e.g. cholangitis, cholecystitis,
pyelitis, and cystitis. The B. coli vaccine is more toxic
than most vaccines, and small doses must therefore be
given (see p. 221).

Clinical Examination

(1) The appearance and odour of the pus are often characteristic.
Smears of the pus show small bacilli, which are decolorised by
Gram's method.

1 Journ. Path, and Bad., vol. vi, 1900, No. 3, p. 315 (Bibliog.).

388 A MANUAL OF BACTERIOLOGY

(2) The organism may be isolated by plating on gelatin, agar,
litmus lactose agar, Conradi-Drigalski agar, or by the use of neutral
red or bile -salt media (see " Water "). The isolated organism must
be tested as to its morphology, motility, non-Gram staining, non-
liquefaction of gelatin, indole production, curdling of milk, and
fermentation of glucose, lactose, dulcitol, mannitol, etc.

(3) An agglutination reaction may likewise be tried, but if nega-
tive is of little value, as there are so many varieties of the colon
bacillus, and one variety may not be agglutinated by the specific
serum obtained with another variety. A positive reaction must
also be carefully controlled, as the colon bacilus is much more
readily agglutinated by normal serum than is the typhoid bacillus.

Varieties of Bacillus coli

Organisms are frequently met with in faeces, manure, sewage and
polluted water which resemble the typical B. coli in many of their
characters, but which differ from it in certain particulars. Thus
the colonies on gelatin, instead of being smooth, may be wrinkled ;
milk may be but slowly curdled (three to eight days) ; acid or gas
production, or both, in sugars may be less marked than usual.
These organisms are generally regarded as varieties of the B. coli,
and are perhaps dervied from typical B. coli. There is, however,
little evidence that B. coli can be transformed into such varieties,
or that these varieties can be reconverted into typical B. coli ;
Revis (loc. cit.) has produced considerable alterations of fermentive
power, and in the characters of the colonies of certain coliform
organisms.

Organisms that have been Regarded as
Variants of B. coli

A number of organisms have been regarded as being varieties of
the B. coli (consult Table of fermentation reactions, p. 381).

(1) Bacillus cavicida (Brieger). — This resembles B. coli in most
of its characters, but was stated to be non-motile. MacConkey says
it is motile.

(2) Bacillus neapolitanus (Emmerich). — Isolated from the bowel
in cases of cholera. It differs from B. coli in not being motile, and
in fermenting cane sugar.

(3) Gas-forming bacilli of Laser and Gartner.1

1 Oentr.f. BaJct (lte Abt.), xiii, 1893, p. 217 ; xv, 1894, pp. 1, 276.

FLIES AS CARRIERS OF INFECTION 389

(4) Aerobic bacillus of malignant oedema (Klein).

(5) Bacillus lactis aerogenes of Escherich. — Found in the intestine
of nurslings and in milk. Much like B. coli, but is non-motile.
It differs from B. coli by not fermenting dulcitol, by fermenting
saccharose and adonit, and by giving the Voges-Proskauer reaction
(see Table, p. 381). According to Harden and Walpole,1 its action
on glucose differs from that of B. coli, more alcohol being produced
and formed at the expense of that part of the molecule of the sugar
which in the B. coli fermentation yields acetic and lactic acids.

The Voges-Proskauer reaction is obtained by growing the
organism in 2 per cent, glucose broth in a fermentation tube (Fig. 17,
p. 84) for three days and adding some strong caustic potash solu-
tion ; on standing exposed to the air a pink colour develops.
According to Harden and Walpole 2 the reaction is probably due
to acetylmethyl-carbinol, which in the presence of air and potash
is oxidised into diacetyl, which then reacts with some constituent
of the peptone in the medium, giving the pink colour.

The B. lactis aerogenes (which may be classed among the capsu-
lated bacilli, see p. 258) is occasionally pathogenic, causing peri-
tonitis. 3 In these circumstances, it is capsulated, but the capsule is
difficult to stain. It seems probable that the B. capsulatus of
Pfeiffer is identical with this organism.

(6) B. cloacae, (Jordan). — Met with in sewage. In general char-
acters it much resembles B. coli, but produces more gas (75 per cent.)
from glucose and liquefies gelatin in four or five to thirty days. Like
B. lactis aerogenes, saccharose is always fermented and the Voges-
Proskauer reaction is positive, but neither dulcitol nor adonit is
fermented. (See Table, p. 381.)

Flies as Carriers of Infection

Flies and other "insects " may convey infection (1) by acting as
" porters " and infecting food, etc., (2) by direct inoculation, (3) by
inoculation after a cycle of development — in which case the carrier
is more or less specific ; e.g. anopheline mosquitoes in malaria. In
the first method the organisms are generally bacteria, occasionally
ova of worms ; in the second, bacteria or protozoa ; in the third,
invariably protozoa, filaria, etc., i.e. animal organisms.

1 Journ of Hygiene, vol. v, 1905, p. 488 ; Proc. Roy. Soc. Lond., B,
vol. Ixxvii, 1906, p. 399.

2 Proc. Roy. Soc. Lond., B, vol. Jxxvii, 1906, p. 399.

3 See Churchman, Johns Hopkins Hosp. Bull., vol. xxii, 1911, p. 116.

390 A MANUAL OF BACTERIOLOGY

The ordinary domestic fly, the blue-bottle and other similar flies
(of which there are many) have no biting proboscis, but undoubtedly
directly convey infection to food, etc., by carrying organisms upon
various parts of their body, or by the organisms passing through
the digestive tract and infecting the food with the faeces. In tin's
way, typhoid, bacillary dysentery, B. enteritidis, summer diarrhoea,
cholera, and possibly anthrax, and also the ova of certain worms,
may be conveyed.

The ordinary house-fly breeds in dung and garbage containing
dung, and it has a possible range of flight of about a mile. The
house-fly experimentally infected remains grossly infected for at
least three days, and a smaller degree of infection persists for ten
days or even longer. 1

1 Sec Reports to the Loc. Gov. Board on Flies as Carriers of Infection,
Nos. 1-4, 1910 and 1911. Martin, Brit. Med. Jovrn., 1913, I., p. 1.

CHAPTER XI

BUBONIC PLAGUE— CHICKEN CHOLERA-
MOUSE SEPTICAEMIA

Bubonic Plague

PLAGUE was epidemic throughout Europe during the
Middle Ages ; in England in the fourteenth century it
appeared as the Black Death, and in the seventeenth
century as the Great Plague of London, while numerous
other lesser visitations have been recorded. For some
years plague has been practically pandemic. The disease
seems always to have been endemic in certain centres,
e.g. in Asia Minor, on the Persian Gulf, in Yunnan, in
Uganda, etc. A characteristic of plague is the manner
in which it appears and remains prevalent for a time in
a district and then disappears, to reappear again after a
considerable interval ; this has happened not only in
Europe, but also in Persia, Syria, India, and China.

Three principal types of the disease are recognised, the
bubonic in which the femoral (rarely the inguinal), axillary
and other glands become enlarged (whence the disease
derives its name), the septicaemic, and the pneumonic.
In India the disease has been mainly bubonic (70 per cent,
of the cases). Occasionally the majority of the cases are
pneumonic, as was the case in Accra, in China in 1910-11,
and in the small outbreak in Suffolk in 1910. Septicsemic
cases are the exception, but any form tends to become
septicsemic on the approach of death.

391

392 A MANUAL OF BACTERIOLOGY

At the commencement and at the end of an epidemic
the disease may assume an extremely mild type, the
so-called " pestis minor."

Bacilli were first observed in this disease in the blood,
buboes, and organs by Kitasato in 1894. In the same
year (1894) Yersin investigated the outbreak of bubonic
plague at Hong Kong, and described the bacillus met

FIG. 43. — Smear preparation from spleen of inoculated
guinea-pig, x 1000.

with in the buboes and its cultural and pathogenic properties
very fully. This organism is known as the Bacillus pestis.
Morphology. — The B. pestis belongs to the group of
hsemorrhagic septicsemic bacilli (chicken cholera, rabbit
and ferret septicaemia, swine plague, etc., see p. 404),
and is a markedly pleomorphic organism. In the animal
body it occurs for the most part as a short, plump, non-
sporing rod, measuring 2-3 /m by 1-2 /u., but longer forms
may be seen here and there measuring as much as 5 ju,
(Fig. 43). Polar staining is a marked feature (Plate XIV.
a and 6), and swollen involution forms occasionally occur.

PLATE XIV.

a. Bacillus pestis. Smear preparation from a bubo, x 1000.

b. Bacillus pestis. Smear preparation of sputum, x 1000.

THE PLAGUE BACILLUS 393

The typical form of the organism, the bi-polar staining,
short, stumpy bacillus, is met with in smears from the
buboes, in the sputum in the pneumonic form, and in the
blood in the septicsemic variety, but only in the earlier
stages of the disease. Later the typical forms tend to
disappear, their place being taken by a few large, rounded,
ovoid, or pear-shaped involution forms. Under cultiva-
tion the bacilli in young cultures (twenty-four to forty-
eight hours) are so short as to be almost coccoid or slightly
ovoid ; on agar their size is about the same as that in the
animal body, on gelatin they are somewhat smaller, but
a few well-marked rods and even threads are always present.
In older cultures, rod, thread and involution forms occur
more numerously ; on agar containing 2-3 per cent, of
salt the latter are swollen and yeast-like.

In broth chains of slightly ovoid organisms occur
resembling streptococci (Plate XV. a).

The organism is non-sporing and non-motile, although
Gordon described the presence of one or two fine spiral
terminal flagella (others have not found flagella).

Sometimes in hanging- drop cultivations a capsule is
apparently present, but the writer has failed to verify this
by staining methods.

The B. pestis stains well with Loffler's blue and anilin-
gentian violet, polar-staining being a marked feature,
especially in smear preparations. It does not stain by
Gram's method. With old laboratory strains polar stain-
ing may be completely absent, but in such cases may
sometimes be obtained by first treating the preparations
with alcohol or by the Gram method, and subsequently
staining with Loffler's blue or weak gentian violet. Sections
are best stained with carbol methylene or thionine blue.

Cultural characters. — The B. pestis is aerobic and facul-
tatively anaerobic. On blood-serum it forms moist,
smooth, shining, cream-coloured colonies or growths,

394 A MANUAL OF BACTERIOLOGY

slightly raised above the surrounding medium. The
blood-serum is not liquefied.

On agar the colonies are raised, round and cream-
coloured, finely granular, denser at the centre than at
the margins, which are regular. Size 0-25 to 0-5 mm. in
two days at 37° C.

On surface agar the B. pestis forms a thick, opaque,
moist, smooth, cream-coloured growth, the margins of
which are usually markedly crenated ; the growth is
very sticky and tenacious. Haffkine states that when
grown on dry agar (agar which has been kept in the warm
incubator for two to three weeks) and viewed from behind
the growth has an appearance like that given by the
back of a mirror — i.e. a dull, silvery appearance.

On a salt agar (2-5-3-5 per cent, of sodium chloride)
Hankin describes the development of remarkable spherical
or pear-shaped involution forms.

On gelatin the colonies are whitish, filmy, finely granular
with regular margins. Size, 0-1 to 0-25 mm. in five days
at 22° C.

On surface gelatin the organism forms a thin, white,
granular growth, with slightly irregular surface and margins,
and nearly confined to the inoculation track (Fig. 44).
The growth does not penetrate into the medium, nor does
it render it cloudy. The growth is very adherent.

In a stab gelatin culture a delicate whitish, finely granular
growth develops to the end of the stab, with little tendency
to spread from the needle track. The gelatin is not
liquefied. Both in agar and gelatin cultures fresh punctate
growths sometimes develop in the original growth, simu-
lating a contamination. No growth occurs on ordinary
potato, and milk is not coagulated.

In broth the growth is somewhat characteristic. For
two or three days the broth remains perfectly clear, but
a flocculent growth forms and gradually increases in

PLATE XV.

a. Bacillus pestis. Film preparation from a 72-hours' broth
culture, x 1000.

6. Chicken cholera. Film preparation of blood of fowl.
X 1000.

THE PLAGUE BACILLUS

395

amount on the bottom and sometimes upon the sides of
the tube. After some days the broth may become a little
cloudy. A delicate flocculent film develops if the tube
be kept absolutely at rest. In broth to which a little
butter-fat or ghee has been added little islands of growth
appear on the surface, and from
these flocculent tapering depen-
dent growths form in about a
week, provided the tubes or
flasks be kept absolutely at rest,
the bulk of the broth remaining
clear. This is the stalactite
growth of Haffkine, and is very
characteristic (B. pseudo-tuber-
culosis also gives it). Broth
cultures reduce a weak solution
of methylene blue.

With sulphuric acid alone a
feeble indole reaction can be
obtained with week- old broth
cultures. With sulphuric acid
and a nitrite a well-marked
indole reaction can be obtained
under the same conditions.

The fermentation reactions of
the B. pestis, which MacConkey
has pointed out are practically identical with those by the
B. pseudo-tuberculosis, are as follows : Acid production,
but no gas, in glucose, laevulose, galactose, maltose,
mannitol, and dextrin, no change in lactose, cane-sugar,
and dulcitol.

Action of antiseptics, etc. — The plague bacillus is readily
destroyed by antiseptics ; a 1 : 1000 corrosive sublimate
or 1 : 100 chloride of lime solution being efficient. An
acid solution of corrosive sublimate is preferable, and for

FIG. 44. — Plague, surface cul-
ture on gelatin four days
old.

396 A MANUAL OF BACTERIOLOGY

the practical disinfection of native houses a 1 : 250 solution
of sulphuric acid may be employed. A temperature of
65° C. kills the organism in about fifteen minutes. Desic-
cation over sulphuric acid at 30° C. is also rapidly fatal.

Vitality and virulence of cultures. — Cultures retain their
vitality for at least a month. As regards virulence, the
organism varies much according to the source from which
it is obtained. Under cultivation it gradually loses its
virulence unless subcultured in the following manner :
The cultures are made every week on surface agar, are
placed in the blood-heat incubator for twenty-four hours,
and are then removed and kept at room temperature. If
inoculated into animals the virulence may be heightened
for a particular species by successive passages, but in
so doing is diminished for other species.

Pathogenic action. — In addition to man, the following
animals are liable to contract plague under natural con-
ditions— the monkey, cat, rat, mouse, squirrel, ground
squirrel, ferret, bandicoot, and marmot. The guinea-
pig and rabbit are also susceptible to inoculation. The
horse, cattle, sheep and goat are relatively insusceptible,
though Simpson * stated that calves and poultry may be
infected by feeding, and suffer from a chronic form of the
disease (this observation of Simpson's has not been con-
firmed by other workers). Birds are not easily susceptible,
and vultures feeding on the corpses of the plague- stricken
do not seem to contract the disease. The mouse, rat, and
guinea-pig are the animals chiefly used for experimental
purposes in the laboratory ; the first two are highly
susceptible, a simple prick in the thigh with an infected
needle being sufficient to induce the disease.

A guinea-pig inoculated with plague material or with
a pure cultivation usually dies in from two to seven days,
the symptoms being sluggishness and loss of appetite,

1 Report on the Plague in Hong Kong.

PATHOGENICITY OF PLAGUE BACILLUS 397

sometimes a discharge from the eyes, and towards the
end, staring coat and perhaps convulsive and paralytic
attacks. The post-mortem appearances are extensive
haemorrhagic oedema at the seat of inoculation, enlarge-
ment and congestion of the spleen, and enlargement of,
and hemorrhages into, the inguinal and axillary lymphatic
glands. If the animal live six or seven days, the glands
may be as large as small nuts (see some admirable prepara-
tions in the College of Surgeons Museum). The spleen

FIG. 45. — Spleen of guinea-pig inoculated with plague.
(Nat. size.)

may be enormous, six times its natural size, and studded
with small yellowish nodules resembling miliary tubercles,
consisting of necrotic areas with masses of bacilli (Fig. 45) ;
the lungs also may be more or less inflamed, and contain
small and large necrotic foci. The bacilli are extremely
numerous at the seat of inoculation, in the glands, and
in the spleen, less so in the peritoneal fluid, liver, and
blood ; if the death of the animal is delayed the exudation
in the bronchi may contain considerable numbers. Some
bacilli may generally be found in the duodenum, trachea,
and larynx. Mice usually die in from two to three days,
and rats in from three to seven days after inoculation. In
rats and mice the post-mortem appearances are similar
to those in the guinea-pig. A very small dose of a pure
culture may fail to kill an inoculated animal. Rabbits
are much less susceptible to plague than guinea-pigs, and
may be injected with considerable doses of living cultures
without showing marked illness. Rats can be infected

398 A MANUAL OF BACTERIOLOGY

by feeding on the corpses or carcases of men or animals
dead from the disease.

In man the bacilli are found in large numbers in the
fluid in the buboes, either alone or mixed with streptococci
or micrococci, and in the sputum in the pneumonic form.
They are not usually found in any number in the blood
except in the septicaemic variety, or shortly before death,
and in stained preparations appear as short plump bacilli,
often in pairs, with polar staining and unstained centres
(Plate XIV. a and 6). If the organisms are found to be
free and numerous in the buboes the prognosis tends to
be grave, but if they are largely present within the
phagocytic polymorphonuclear leucocytes the prog-
nosis is better and the disease will probably remain
localised. •

Toxins. — The plague bacillus forms but little toxin, the
minimal fatal dose of the most active filtered broth culture
for a mouse being about 0-02 c.c. In order to prepare a
vaccine or an anti-serum it is necessary, therefore, to
employ unfiltered cultures -4.e. the microbes themselves.

Macfadyen obtained an endotoxin by triturating the
bacilli frozen with liquid air.

Vaccines and immunity. — Of the plague vaccines, that
of Haffldne, the Haffldne prophylactic, is the best known,
and has been extensively employed. It consists essen-
tially of a four to six weeks old butter- fat broth culture
of the plague bacillus, killed by heating to 65° C. for an
hour, with a small addition of antiseptic. As to the
value of Hafikine's prophylactic a mass of figures is
available. By its use both the incidence of, and mortality
from, plague are markedly diminished. Wilkinson col-
lected the following data of the efficiency of the vaccine :
Among the inoculated the case incidence was 1*8 and the
case mortality 23-9 per cent. ; among the uninoculated
the figures were 7-7 and 60-1 respectively. The immunising

PLAGUE VACCINES 399

products seem to be mainly intracellular, but the broth
itself is not without action.

Other vaccines have also been devised. Lustig and Galeotti
prepared one by digesting the growth from agar cultures with 1 per
cent, caustic soda solution, filtering through paper, and precipi-
tating with very dilute acetic or hydrochloric acid, or by saturation
with ammonium sulphate. The precipitate is dissolved in a 0-5 per
cent, solution of sodium carbonate, and filtered through a Chamber-
land filter ; this forms the vaccine fluid. Calmette prepared a
vaccine by emulsifying an agar growth in water, well washing the
organisms with sterile water to remove adherent toxin, emulsifying
again in sterile water, heating to 70° C. for an hour, and finally
drying in vacuo. The dry substance can be kept for a considerable
time without change. For use 1-2 mgrm. are emulsified in 2-3 c.c.
of sterile salt solution and injected.

Yersin proposed vaccinating with living culture of feeble viru-
lence, which has been done by Strong in Manila. Though such a
method might be used in a plague -stricken district, it is obviously
one that can be used only with the greatest caution.

Klein 1 has prepared a prophylactic by drying the organs of a
guinea-pig dead of plague for three days at 46° C., rubbing the
material to a powder, and further drying at 37° C. for three days.
Of this dry powder 15-16 mgrm. protected a rat, and 25 mgrm. a
monkey.

With reference to experimental immunity and protection in
plague, Klein 2 found that a guinea-pig which had been three times
injected with an amount of living culture insufficient to kill was
still capable of being infected ; that the blood of a guinea-pig which
had twice passed through an attack of plague did not contain an
appreciable amount of germicidal substances ; and that the im-
munisation of guinea-pigs by sterilised cultures is an extremely
slow and difficult process. Calmette also found that the guinea-pig
was extremely difficult to immunise.

Calmette, from laboratory experiments, surmised that protection
with a vaccine is not attained for some days, and that in the interval
susceptibility to infection is increased. These observations are not
borne out in practice, for Bannerman 3 found that so far from there
being an increase in mortality among those who have been inocu-
lated and who develop plague within ten days of inoculation the

1 Rep. Med. Off. Loc. Gov. Board for 1905-06.

2 Ibid. 1896-97, App. B., p. 2.

3 Centralbl. f. Bakt- /lte Abt.). Bd. xxix, p. 873 (Bibliog.).

400 A MANUAL OF BACTERIOLOGY

reverse is the case, and that in a small community where the
population had been partly vaccinated and partly not vaccinated,
the incidence of plague during the week following vaccination was
less among the vaccinated than among the unvaccinated, pointing
to the rapid production of protection.

Anti-plague serum. — This is prepared by growing the
B. pestis on the surface of agar in plate bottles, washing
off and emulsifying the growth, and for the earlier injec-
tions the emulsion is heated to 65° C. for one hour, and
the commencing dose is ^ part of a flask. The injections
are given intravenously at intervals of a week. At the
end of three months the bactericidal power of the blood
will have become very marked, and living cultures are
then injected for a further period of about three months
until a whole flask- culture is given at a dose. An interval
of a fortnight is allowed to elapse between the last dose
and the bleeding of the animal. The serum is tested
upon mice.

The anti-plague serum, which is mainly anti-microbic,
is not very potent, and to be of service large amounts and
early treatment are essential.1

Epidemiology. — The mode of infection in man has been
a matter of controversy. The pneumonic form arises
generally from aerial infection by the respiratory tract.
It is extremely fatal and infectious, while the bubonic and
septicaemic varieties are hardly ever contagious. Although
a gastric and intestinal form of the disease has been
described, and there is evidence to show that food or drink
may be the vehicle of infection, this must be a rare mode
of infection. Yersin claimed to have isolated the bacillus
from the dust and earth of a native dwelling, and Hankin
from the brackish water in a field. The observations of
Hankin and others indicate, however, that contagion is
likely to occur only from immediate contact with man or

1 See Hewlett's Serum Therapy, 1910.

TRANSMISSION OF PLAGUE 401

animals, or their excretions, infected with plague, and
not from a saprophytic form of the organism.

Certain animals, especially the rat (Mus rattus and Mus
decumanus), are important agents in spreading the disease.
The association of sickness and of death among rats
with an epidemic of plague has been established by a
number of observations, and in some instances the epizootic
among rats has been definitely shown to precede the
epidemic in man. The epidemics at Sydney are perhaps
the most striking instances of rat-borne plague ; discussing
the first one Tidswell says : " The one clear fact in our
epidemic was that human beings were not becoming infected
from one another." In the first epidemic the mode of
introduction of the disease was never traced to any human
source. During an epidemic rats may be found in
all stages of illness and plague bacilli can be found in
large numbers in their carcases. In the various epidemics
at Sydney, cases of plague first occurred among the rats
and mice, followed after an interval of days or weeks by
human cases. Other animals may also occasionally be
the means of disseminating the disease. The experiments
of the Advisory Committee on Plague Investigation in
India have conclusively shown the important part played
by rats in the dissemination of the disease, though the
origin of the primary infection in rats is doubtful. They
may possibly become infected from the dust of earthen
floors of the native houses soiled with excreta or discharges
of plague patients, or from their clothing, poultices or
dressings, but the readiest method is probably by feeding
on the dead. Once the epizootic has started, further
infection is simple ; rats fight, and so may directly inocu-
late one another ; the sick rats may soil grain or other
food-stuffs, and the dead rats are eaten by their fellows.
Moreover, parasitic insects, especially fleas, undoubtedly
may transmit the disease from one animal to another.

26

402 A MANUAL OF BACTERIOLOGY

Thus it is found that if guinea-pigs be placed in a plague-
infected compound, many of the animals contract plague ;
but if the animals be placed in cages of wire-gauze, the
mesh of which is small enough to prevent access of fleas,
the animals do not contract plague. The transmission
of the disease from rats to man is similarly due to trans-
mission by fleas (except in the pneumonic forms in which
infection is direct from the sick to the healthy). The great
majority of rat fleas are Xenopsylla cheopis, Ceratophyllus
fasciatus, Cer. anisus, Ctenopsylla musculi, and Ctenoph-
ihalmus agyrtes, of which the first is most prevalent in the
tropics and subtropical regions, the second in cooler regions
and in this country.1 Walker 2 has found that bed-bugs
may occasionally transmit plague. The bacilli multiply
in some of the fleas to such an extent as to occlude the
entrance to the stomach. Such fleas will still bite, but
on ceasing to suck, some of the blood with numerous
bacilli in it regurgitates into the wound and thus infects.3
The seasonal prevalence of plague coincides with the
prevalence of rat-fleas. The manner in which the periods
in the year when human plague does not occur are bridged
over is unknown. In such periods rats suffering from
plague have been found, but these are regarded as having
a retrogressive form of the disease rather than a chronic
infection. The destruction of rats, either by trapping,
poisoning, or asphyxiating, or by the use of the Danysz
rat virus (see p. 373), is, therefore, one of the means to be
adopted in fighting the disease. The extermination of
rats seems quite impossible, but by rat destruction there
is a likelihood of destroying infected animals and the
subsequent development of a healthy race. On the other
hand, objection has been taken to rat-destruction, it being

1 See Chick and Martin, Journ. of Hygiene, vol. xi, 1911, p. 122.

2 Ind. Med. Gaz., May 1910.

37Bacot and Martin, Journ. of Hygiene, XIII, Plague Supp. Ill
914, p. 423.

DIAGNOSIS OF PLAGUE 403

surmised that if the epizootic be allowed to proceed, the
susceptible rats will be exterminated and a race of rats
relatively insusceptible to plague will ultimately be
established.

On Plague, see Simpson, Treatise on Plague (Cambridge Univer-
sity Press) ; Klein, Bacteriology of Oriental Plague ; " Reports on
Plague Investigations in India," Journ. of Hygiene (extra numbers),
vols. vi-xiv.

Clinical Examination

If it cannot be examined immediately, plague material may be
placed in a solution containing glycerin 20 c.c., distilled water
80 c.c., calcium carbonate 2 grm. The bacilli retain their vitality
and virulence in this for thirteen days (Albrecht-Ghon method).

(1) Withdraw a little of the fluid from the bubo by means of an
antitoxin syringe. Make smears and stain with methylene or
thionine blue. Search for short plump bacilli, often in pairs, with
polar staining and unstained centres. They are not stained by
Gram's method.

N.B. — There may be a mixture of organisms in the buboes.

(2) Make agar plates and broth cultures. Incubate the cultures
at 25°-27° C., not at 37° C. From colonies on the agar plates the
organism may be isolated and its cultural and pathogenic characters
ascertained. The appearance of the broth cultures, if charac-
teristic, would be very suggestive of plague, but if uniform turbidity
develops this may be due to contaminating organisms, e.g. micrococci.

(3) Inoculate mice, rats, or guinea-pigs subcutaneously with the
fluid or with the culture. Some of the animals should be inoculated
by the cutaneous method — rubbing a little of the material on the
shaved abdomen, and also as in (4). Inoculation of rats serves to
distinguish the B. pseudo-tuberculosis from the B. pestis. If the
animals die, investigate for the Bacillus pestis by staining and
culture methods.

(4) In the pneumonic form, dilute the sputum with a little
boiled water, inoculate several agar tubes, and incubate at
25°-27° C. Examine in two to three days. Also daub the
nostrils of a guinea-pig or rat with a brush or pledget of wool
dipped in the diluted sputum, avoiding wounding the mucous
membrane. Smears of the sputum may also be made, stained, and
examined. Gram's method will distinguish the B. pestis from the
Streptococcus pneumonice ; the latter stains well by Gra.m,

404 A MANUAL OF BACTERIOLOGY

(5) Agglutination reaction. — The Indian Plague Commissioners
state that in their opinion no practical value attaches to the method
of serum diagnosis in plague, but a modified method is considered
by Dunbar x to be of considerable value. The method is carried
out as follows :

A small quantity of peptone solution, inoculated with the tissue
juice from the suspected organ, is mixed with an equal quantity
of plague-serum of such a strength that the dilution reduces it to
1 : 200 (approximately). A second dilution of 1 : 400 and a third
of 1 : 800 are also prepared.

As a control, an equal quantity of the inoculated peptone water
is mixed with normal serum (rabbit or horse serum), the dilution
being 1 : 100.

In a few minutes a distinct difference is observable. The
" control " shows with the oil-immersion lens a few isolated non-
mobile bacteria, while the plague-serum dilution 1 : 200 shows
larger and smaller masses of agglutinated bacteria.

After two hours' incubation the same result is obtained with
the plague-serum dilution of 1 : 400. No agglutination, however,
is observed after incubation for twenty -four hours of the dilution
of 1 : 800. This agglutination reaction, in conjunction with other
suspicious phenomena, justifies an official notification of suspected
plague.

In the examination of rats suspected to be suffering from plague
infection, it is essential not only to take the naked-eye characters
into account, but to make microscopical preparations and cultures,
and to test the cultures by animal inoculations. Care must be taken
not to mistake Jicemorrhagic septiccemic bacilli (see pp. 392, 405) and
other organisms for the plague bacillus. The B. coli, B. proteus, and
other organisms are recorded by Klein (loc. cit.) as simulating the
B. pestis.

Chicken Cholera

Chicken cholera is a disease of poultry characterised by profuse
diarrhoea ; its course may be very rapid, and the bird found dead
without having shown signs of illness. The organism is a very
short rod, non-motile, so short that it is almost ovoid, 0-6 to 0-8 p
in length, and 0-4 to 0-5 p. in diameter. It stains by the ordinary
anilin dyes, but not by Gram's method, and the staining tends to
be polar, so that Pasteur, who first investigated the disease,
described it as a diplococcus (Plate XV. b). The organism grows

1 Centralbl.f. Balct,, xli (Originate), 1906, p. 860,

CHICKEN CHOLEKA 405

freely on the various culture media from 20° to 38° C., on agar
forming a thick, moist, cream-coloured layer, on gelatin a shining,
white, expansive growth without liquefaction. In broth a general
turbidity forms, but growth on potato is indifferent. It produces
acid, does not ferment glucose or lactose, is aerobic and faculta-
tively anaerobic, does not form spores, and is killed by a tempera-
ture of 60° C. in fifteen minutes. If dried it dies in a few days, but
retains its vitality for a considerable time in damp earth or in water,
and so infection is readily conveyed. Fowls die after subcutaneous
intramuscular or intravenous inoculation and by feeding, the
organisms being found abundantly in the blood. Post-mortem, the
serous membranes may be inflamed and haemorrhagic, the liver
large and soft, and the intestine shows haemorrhagic spots, and is
sometimes ulcerated and contains a mucoid fluid stained with blood.
Other birds, pigeons, pheasants, sparrows, wild and domestic ducks
are also susceptible to the disease, and rabbits and guinea-pigs can
be successfully inoculated ; in the latter animal a local abscess
sometimes forms instead of a general infection. By continuous culti-
vation with free access of oxygen the virus becomes attenuated, and
Pasteur was able thus to prepare a vaccine which protected fowls.

The bacillus of chicken cholera belongs to the group of hcemor-
rhagic septiccemic bacilli (p. 392), and seems to be identical with
Koch's bacillus of rabbit septicaemia, and with the bacillus of swine
plague (see p. 373). These organisms tend to form a stalactite
growth in butter broth.

Organisms have been described by Klein in fowl enteritis, grouse
disease, etc., differing somewhat from the bacillus of chicken
cholera.

Mouse Septicaemia

This disease may be conveniently described here. Koch first
obtained a minute bacillus by injecting putrefying material sub-
cutaneously into mice. It seems to be identical with the bacillus
found in swine erysipelas. The organisms are met with in large
numbers in the blood and tissues of mice. They measure only
1 p in length, and occur in considerable numbers in the leucocytes.
The bacillus stains well by Gram's method, and is stated by some
writers to be motile. It grows readily, forming on agar extremely
delicate, almost invisible colonies ; in stab gelatin cultures after
some time a delicate cloudiness radiates from the central puncture.
From an agar culture the bacilli are somewhat larger than those
found in the animal body, and form filaments. It is pathogenic for
swine, rabbits, and mice.

CHAPTER XII
PNEUMONIA, INFLUENZA, AND WHOOPING-COUGH

Pneumonia

PNEUMONIA is of two types, lobular, catarrhal, or broncho -
pneumonia, and lobar or croupous pneumonia. The former may
be primary, or may be secondary and arise in connection with many
of the specific fevers, as in measles, whooping-cough, diphtheria,
enteric fever, influenza, plague, etc. The broncho -pneumonia
occurring in the course of other diseases may be due to the causative
organism of the disease, or may be due to other organisms. Eyre 1
examined 62 cases of broncho -pneumonia occurring in the course
of other diseases and 102 cases in which the broncho-pneumonia
was the primary lesion. Of these 164 cases, 52-4 per cent, yielded
pure cultivations of some one or other of six bacteria — pneumo-
coccus, Strep, longus, M. pyogenes var. aureus, M. catarrhalis,
B. pneumonias, and B. inftuenzce ; whilst 47-5 per cent, gave a mixed
growth of one or more of these six in association with one or more
of five other bacteria — M. tetragenus, B. pertussis, B. pyocyaneus,
B. typliosus, B. diphtherice. The B. coli also occurs in broncho-
pneumonia. Acute croupous or lobar pneumonia in many of its
characters resembles an acute specific infection, and while frequently
a primary disease, may also occur secondarily in almost any con-
dition, and occasionally in epidemic form.

Friedlander in 1882-83 first described organisms in cases of
pneumonia.

In 1883-85 Talamon, Klein and Sternberg each described in
pneumonic sputum an oval encapsuled organism, which induced
pneumonia in animals ; it was termed by the former the Micro-
coccus lanceolatus, and by Sternberg the Micrococcus Pasteuri.
This and Friedlander's organisms were at first believed to be
identical, but Frankel and Weichselbaum subsequently showed that

1 Journ. Path, and Bact., vol. xiv, 1910, p. 160.
406

THE PNEUMOCOOCUS 407

they are quite distinct, and that the former is the etiological agent
of acute croupous pneumonia.

The majority (95 per cent.) of cases of acute croupous pneumonia
are caused by the Streptococcus pneumonice, and Friedlander's
organism, now termed Friedlander's pneumo-bacillus, or B. pneu-
monice, is of etiological significance in only a small minority, if at
all. The latter is, however, associated with certain pathological
processes which will be referred to below.

From pleuro-pneumonia of cattle, Nocard and Roux succeeded
in cultivating in broth in collodion sacs in the peritoneal cavity of
rabbits an organism just visible as minute granules with a magnifica-
tion of 2000 diameters. Bordet x states that it may be grown on
the medium employed by him for the cultivation of the B. pertussis
(p. 417), and then appears as fine, straight, curved, undulating, or
even spirillar filaments not unlike spirochaetes.

The Streptococcus (Diplococcus) pneumoniae

Synonyms, Frankel's pneumococcus, Micrococcus Pasteuri (Stern-
berg), Micrococcus lanceolatus (Talamon), Micrococcus pyogenes
tenuis (Rosenbach).

Morphology. — The Streptococcus pneumonice in the sputum
and tissues occurs as an oval or lance- shaped coccus united
in pairs, occasionally in chains of three or four elements,
and then often almost spherical, and is generally surrounded
by a well-marked capsule (Plate XVI. a). In order to
isolate the organism several tubes of glycerin agar, serum
or serum- agar may be inoculated with rusty sputum and
incubated for forty- eight hours ; in some a pure culture
may be obtained. A more certain method is to inject
a drop or two of the rusty sputum into the peritoneal
cavity of a mouse or young rabbit. The animal will die
in from twenty-four to thirty-six hours, and the organism
will be found in considerable numbers in the lung and
blood, from which cultures may be obtained. It is non-
motile, stains with the ordinary anilin dyes and by Gram's
method.

1 Ann. de VInst. Pasteur, xxiv, 1910. March.

408 A MANUAL OF BACTERIOLOGY

Cultural characters.- — The S. pneumonice is aerobic and
almost facultatively anaerobic. On glycerin agar at 37° C.
it forms minute, transparent, almost invisible colonies like
droplets of fluid ; on serum the growth has much the same
characters, but is somewhat more abundant. It hardly
grows on gelatin at the ordinary temperature, but in a
20 per cent, gelatin at 25° C. minute white colonies develop
without liquefaction. In broth it produces a' slight cloudi-
ness ; it does not grow on potato but develops in milk,
which is usually coagulated ; neutral litmus glucose- agar
becomes red during growth, indicating the production of
acid. The fermentation reactions are given in the Table
on p. 235. Hiss's medium (p. 291) with inulin is fermented
and coagulated ; most other streptococci fail to ferment
inulin. On the ordinary culture media it retains its
vitality for a short time only, not more than about a week ;
but if a little blood be smeared over the surface of the
agar the vitality may be prolonged for a month or even
longer. Washbourn recommended an agar rendered alka-
line to the extent of 4 c.c. of normal caustic soda per litre,
after neutralisation, rosolic acid being the indicator. This
medium is smeared with blood, placed in the incubator
for twenty-four hours to ascertain whether it be sterile,
then inoculated, capped, and kept at 37° C. Foa's method
for keeping Frankel's pneumococcus alive and virulent is
to receive the infected blood of an inoculated animal into
a small glass tube 5 mm. in diameter and 20 cm. long, so
that the blood completely fills the tube, which is then
sealed and kept away from the light at the ordinary tem-
perature. If inoculated on to ordinary gelatin, which is
then kept in the blood heat (37° C.) incubator, the organism
retains its vitality for a month or six weeks.

Under cultivation the S. pneumonice usually assumes
the form of a short streptococcus (Plate XVI. b) (included
by Gordon in his S. brevis class) and the capsule is lost.

PLATE XVI.

Diplococcus pneumonia. Film preparation of blood of
inoculated animal. X 1000.

b. Diplococcus pnzumonice. Film preparation of a pure
culture. X 1500.

THE PNEUMOCOCCUS 409

but is regained on passage through a susceptible animal,
or by growing in fluid serum. A good deal of variation
occurs in the morphology of the organism obtained from
different sources and under cultivation. The thermal
death-point of the S. pneumonia according to Sternberg
is 52° C., the time of exposure being ten minutes, and it is
readily destroyed by the ordinary germicides, by light, and
by desiccation ; but in dried sputum it may retain its
vitality and virulence unimpaired for weeks.

Pathogenic action. — The S. pneumonia is pathogenic for
a number of animals, the most susceptible being mice,
then in decreasing order, rabbits, rats, guinea-pigs, and
dogs. Pigeons and fowls are immune. Death follows
after subcutaneous, intravenous, intraperitoneal, or intra-
thoracic injection of a virulent culture, or of rusty pneu-
monic sputum, into mice and rabbits in twenty-four to
forty-eight hours. The virulence of the organism varies
considerably ; under cultivation it may be completely
lost, while by a series of passages through a susceptible
animal it may be much increased. The less virulent it is
the longer it tends to retain its vitality under cultivation.
Except when injected into the lung or into the trachea,
pneumonia does not result, but the disease runs the course
of a septicaemia with high temperature and dyspnoea,
death being generally preceded by a subnormal temperature
and often convulsions. The post-mortem appearances are
much oedema and inflammatory infiltration at the seat of
inoculation, hemorrhages in the serous membranes, enlarge-
ment and congestion of the spleen, and congestion of the
lungs. The organisms occur in large numbers in the
blood, lungs, and spleen, usually in the form of oval
diplococci with well-marked capsules (Plate XVI. a), but
sometimes as short chains of streptococci. When injected
into the lung or trachea a typical fibrinous or croupous
pneumonia results.

410 A MANUAL OF BACTERIOLOGY

The S. pneumonia is the cause of acute croupous pneu-
monia in man, and occurs in large numbers in the rusty
sputum and hepatised lung, and in 20 per cent, of the
cases can be isolated from the blood if 5-10 c.c. be cultured.
The production of a typical pneumonic process experi-
mentally and the presence of the diplococcus in a large
proportion of cases of acute croupous pneumonia point
to its specific relationship to the disease. With regard
to the latter observation, Weichselbaum obtained it in
94 cases out of 129 examined, Wolf in 66 out of 70 cases,
and Netter in 75 per cent, of the cases examined. In
America the disease has of late been much on the increase,
in Chicago the mortality having reached as high as 20
per 10,000 inhabitants. Acute croupous pneumonia some-
times occurs in epidemic form and has decimated the
native labourers in the Rand mines.

The organism is frequently present in the saliva of
healthy individuals, as shown by Netter, Sternberg, and
others, and the generally accepted idea of the relationship
of " catching cold " to an attack of the disease is explicable
on the theory that the action of cold lowers vitality, and
renders the tissues vulnerable to the attacks of the organism
already in close proximity to them.

Besides acute croupous pneumonia, more than half
the cases of broncho-pneumonia, both primary and
secondary in the course of other diseases, are due to the
S. pneumonia, which is also associated with a number of
other important pathological conditions in man. It is a
pyogenic organism, producing abscesses when inoculated
into a relatively insusceptible animal such as a dog, and
has been met with in abscesses, empyema, suppuration
in the antrum, and purulent arthritis. It is also found
in about half the cases of purulent meningitis, sometimes
in cerebro-spinal meningitis, in about a third of the cases
of otitis media and infective endocarditis, sometimes in

ANTI-PNEUMOCOCCIC SERUM 411

purulent pericarditis, and occasionally in peritonitis. The
pneumococcus is also frequent in chronic bronchial catarrh.
An agglutination reaction with patient's serum on the
pneumococcus is only very irregularly obtained and normal
serum rarely exerts any bactericidal effect upon the
organism.

As regards opsonic determinations, freshly isolated
strains frequently fail to give any phagocytosis, and every
strain of pneumococcus gives a different amount of phago-
cytosis. For the control, the pooled serum of several
individuals should be used, and the culture should be
emulsified in distilled water. The serum of the Rand
native seems to have a very low opsonic content for the
pneumococcus compared with that of the European.1

Toxins. — Auld separated a proteose and an organic acid
from the blood and organs of infected animals, and from
cultivations of the S. pneumonia in alkali- albumin the
same products were apparently obtained, the alkaline
medium soon becoming permanently acid. The proteose
on subcutaneous or intravenous injection produced some
fever ; on intra- thoracic injection fever and dyspnoea, and
post-mortem pleurisy and consolidation of the lung were
found. The organic acid produced slight rise of tempera-
ture, but no other symptom. Macfadyen 2 obtained an
endotoxin by triturating cultures with liquid air.

Anti-serum. — Immunity can be conferred on susceptible
animals by treating them with attenuated cultures, or
by inoculation with increasing doses of filtered broth
cultures of the virulent organism followed by doses of the
living organism. The blood-serum of such immunised
animals will protect other animals when injected, and an
anti-pneumococcic serum has been prepared by the fore-
going method. This anti-serum has been used in the

1 Wright, Lancet, 1914, i, p. 1 et seq.

2 Brit. Med. Journ., 1906, vol. ii, p. 776 (Refs.).

412 A MANUAL OF BACTERIOLOGY

treatment of pneumonia and other pneumococcic infec-
tions, but the results have not been very encouraging.
The protective serum seems to produce aggregation of
the cocci when added to a culture of the diplococcus.
Klemperer and Washbourn found that the serum of con-
valescent patients possesses some degree of protective
power. The serum, however, withdrawn during the
pyrexial stage of the disease rather increases the suscepti-
bility of animals to pneumococcic infection.

Vaccine. — A vaccine prepared from cultures killed by
heat and standardised has been found of service in chronic
pneumococcic infections, and has also been employed in
acute croupous pneumonia.1 Wright (loc. cit.) has also
recommended a vaccine for prophylactic inoculation against
pneumonia on the Rand, a dose of 1000 millions apparently
being the optimum for this purpose.

Friedlander's Pneumo-bacillus

This organism, already referred to above in the general
discussion of pneumonia, and originally believed by Fried-
lander to be the cause of the disease, has been obtained
by recent observers in only a small proportion of cases
of pneumonia.

Morphology. — The B. pneumonia is a very pleomorphic
organism, occurring in sputum or in the blood of an inocu-
lated animal generally as a short rod with rounded ends
surrounded by a marked capsule. It is non-motile, does
not form spores, and is readily stained with the ordinary
anilin dyes, but not by Gram's method — an important
distinction from the S. pneumonia. In cultivations it
forms short rods, long rods, chains, and even filaments,
the capsule being absent, but this is regained on passage
through a susceptible animal.

1 Willcox and Morgan, Brit. Med. Journ., 1909, vol. ii, p. 1050.

THE PNEUMO-BACILLUS

413

Cultural characters. — The B. pneumonice is aerobic and
facultatively anaerobic, and may produce indole. It grows
readily on the various culture media from 20° to 37° C.,
on agar and blood- serum forming a copious, viscid, greyish
growth ; on gelatin, a thick, white, shining, porcelain- like
growth without liquefaction ;
and in stab- cultures in gelatin
a so-called nail-shaped growth
is developed (Fig. 46), consist-
ing of a white growth along
the needle- track, tapering from
above downwards, and at the
surface heaped up and ex-
panded, forming the " head " of
the nail. On potato a copious
whitish growth develops, while
milk is curdled and gas- bubbles
frequently form in stab- gelatin
cultures. It is an active fer-
menter of carbohydrates ; the
fermentation reactions are given
in the Table, p. 381.

Pathogenic action. — The

pneumo- bacillus of Friedlander FIG. 46. — Friedlander's pneumo-

is pathogenic to mice and bacillus' Gelatin stab-cul-

^ . & . ture, seven days old.

guinea-pigs, but rabbits are

immune. Post-mortem, the spleen is enlarged, the lungs
are congested and consolidated in patches, and the organism
is found in large numbers in the blood. In a small per-
centage of cases of croupous pneumonia Friedlander's
bacillus may be associated with the S. pneumonice. Fried-
lander's bacillus may sometimes set up a broncho-pneu-
monic or bronchitic process, and is occasionally associated
with anginal conditions, which are characterised by the
formation of a false membrane, with an absence of any

414 A MANUAL OF BACTERIOLOGY

general symptoms. A microscopical examination of the
membrane will show the organisms surrounded with a
capsule and unstainable by Gram's method. If a culture
be made on serum, the large, round, greyish colonies of
the bacillus will be recognisable in fifteen to twenty hours,
and should be examined microscopically. To obtain a
pure culture a white mouse should be inoculated from a
colony ; it will die in twenty- eight to sixty hours. Fried-
lander's pneumo- bacillus has also been met with in water
by Grimbert. According to him, it is identical with the
B. capsulatus of Mori.

Clinical Examination (Pneumonia)

1. Make smear specimens from the rusty sputum, and stain
some with Loffler's blue, and others by Gram's method with eosin.
By a microscopical examination the oval diplococci will be readily
recognised, the B. pneumonia and B. pestis being distinguished
from the S. pneumonia by being decolorised by Gram's method.
The latter organism is the only one, moreover, which is likely to
be ordinarily met with.

2. If the diplococci are found to be fairly abundant in the sputum,
and other organisms nearly absent, an attempt may be made to
cultivate by inoculating several glycerin-agar and serum tubes
and incubating at 37° C. for forty-eight hours.

3. If the diplococci are scanty, or so mixed with other organisms
that it is difficult to identify them, and probably impossible to
obtain a pure culture, a drop or two of the sputum should be injected
into the peritoneal cavity of a mouse or rabbit. The animal will
die in from twenty -four to thirty-six hours, and the S. pneumonice
will be found plentifully in smears prepared from the blood or
lung-juice, and pure cultures can be readily obtained by inoculating
glycerin-agar tubes with the blood or lung-juice.

4. The culture or inoculation method, preferably both, will
probably have to be adopted for the recognition and isolation of
the S. pneumonice in pus from empyemata, abscesses, etc.

5. Friedlander's pneumo-bacillus can be readily isolated by
making gelatin-plate cultivations, in which its colonies form white,
shining, heaped-up points.

INFLUENZA 415

Influenza

A minute bacillus was first described in this disease
by Pfeiffer in 1892, who found it in large numbers in the
bronchial secretion. In order to isolate the organism a
patient with bronchial expectoration should be chosen ;
he rinses his mouth and gargles his throat with hot water
several times, and then, after coughing, the expectoration
is obtained. A little of this expectoration is washed by
shaking in a test-tube with sterile salt solution, then
repeating the washing with sterile salt solution in a second
and finally in a third test-tube. By means of a platinum
needle a number of glycerin- agar and blood- agar tubes
are inoculated with the sputum after the last washing, and
incubated at 37° C.

Morphology. — The influenza bacillus is one of the smallest
bacilli with which we are acquainted. It is a minute rod
0-5-1-5 jji in length, and is non-motile and non-sporing.
It does not stain by Gram's method, and not very readily
with the ordinary dyes, dilute carbol-fuchsin or prolonged
staining with Loffler's blue yielding the best results, the
poles tending to stain more deeply than the centre. In
the sputum it occurs singly, in short chains, in small groups,
or in larger masses, being most numerous early in the
acute stage of the disease.

Cultural characters. — The bacillus is strictly aerobic,
and no growth occurs on media at 22° C. On glycerin-
agar and blood-serum at 37° C. it forms very small, trans-
parent, drop-like colonies in from twenty-four to forty-
eight hours, which, according to Kitasato, never became
confluent. There is no growth on potato. The organism
grows best on media containing blood, such as agar smeared
with sterile human, rabbit's, or pigeon's blood. In broth
it grows at the surface in fine white flakes which subse-
quently sink.

416 A MANUAL OF BACTERIOLOGY

It soon dies out in cultivation, but according to Klein
can be kept alive for some weeks in gelatin incubated at
37° C. The melted gelatin remains clear, the growth
forming a delicate flocculent precipitate at the bottom.
Preparations from cultures show long twisted chains and
threads of bacilli, aggregated so as to form dense networks
and convolutions. These chains or threads are composed
of bacilli placed end to end, and united by a continuation
of the cell- membrane. Involution forms occur. It is
stated to grow better in association with the M. pyogenes
var. aureus than alone. The organism does not seem to
be able to live outside the body for any length of time,
and is readily destroyed by desiccation, weak antiseptics,
and by a temperature of 60° C. acting for five minutes.

Pathogenic action. — Canon stated that he obtained this
bacillus from the blood in a number of cases, but many
other investigators have failed to find it. Klein also
obtained it in six cases out of forty-three examined.
According to Pfeiffer the bacillus is pathogenic only to
monkeys and rabbits. Klein, however, was unable to
obtain any definite effects in these animals by the injection
either of sputum rich in bacilli or of pure cultures.

The influenza bacillus is met with in all uncomplicated
cases of influenza in the nasal and bronchial secretions,
often almost in pure culture, and in the bronchial tubes
and lung in the pneumonic complications accompanying
the disease. The organisms disappear with convalescence,
and are not met with in other diseases. Klein x appears
to consider that the pneumonia often complicating the
disease is probably directly due to the bacillus. The
typical influenza pneumonia is of the lobular type with a
cellular rather than a fibrinous exudate. True lobar
pneumonia, due to the S. pneumonia, may, however,

1 " Further Report on Epidemic^ Influenza," 1889-92, Loc. Gov.
Board Report, 1893, p. 85.

PERTUSSIS 417

often complicate the influenzal attack. The organism also
occurs in bronchitis, broncho-pneumonia, and whooping-
cough.

Although the typical influenza may be due to the B. influenzce,
many febrile conditions attended with pulmonary catarrh and
frequently termed " influenza " are not due to this organism. In
an epidemic simulating influenza occurring in Essex in 1905, the
examination was negative as regards streptococci, B. diphtherice
and B. influenzce, but the M. catarrhalis was present in number in
most cases (twenty-two out of twenty-four). This organism was
originally isolated by Seifert in a small epidemic of infectious
bronchitis, afterwards by Pfeiffer in cases of broncho-pneumonia in
young children (see p. 248). Two other Gram-negative cocci were
also isolated from three other cases (see Table, p. 248).

Clinical Examination

In cases of influenza, accompanied with bronchitis or pneumonia,
the influenza bacillus may be met with in large numbers in the
sputum, and their presence may aid in confirming the diagnosis.
Film preparations may be stained with carbol-methylene blue.

Whooping-cough (Pertussis) l

An influenza-like bacillus has been isolated by Koplik, Czaplewski
and Hensel, Davis and others in this disease, but the researches of
Bordet and Gengou have shown that it is distinct from the influenza
bacillus.

The B. pertussis is a minute baciUus, very like the B. influenzce,
non-motile, non-sporing, and Gram-negative. It is scanty in the
bulk of the expectoration, but is abundant in the viscid exudate,
rich in leucocytes, coming from the depth of the bronchi, and voided
at the end of a paroxysm of coughing.

It is best isolated on a medium consisting of defibrinated blood
(human or rabbit), thoroughly mixed with an equal volume of 3 per
cent, agar containing a little extract of potato made with 4 per
cent, aqueous glycerin. It forms on this a fairly thick whitish
streak, the subjacent blood being hsemolysed. It may also be

1 See Bordet, Brit. Med. Journ., 1909, vol. ii, p. 1062.

2?

418 A MANUAL OF BACTERIOLOGY

grown in serum or blood broth in shallow layers. After acclimatisa-
tion to artificial media it will develop on the ordinary laboratory
media.

The B. pertussis is agglutinated feebly by the blood of patients,
but complement-fixation is marked.

Monkeys are stated to develop a typical whooping-cough on
inoculation, but the ordinary laboratory animals are susceptible
only to massive intraperitoneal or intravenous inoculation, death
ensuing from a septicaemic process.

Attempts have been made to treat the disease with a vaccine.

CHAPTER XIII
ANAEROBIC ORGANISMS

TETANUS— MALIGNANT (EDEMA— BLACK QUARTER-
BACILLUS WELCHII (AEROGENES CAPSULATUS, EN-
TERITIDIS SPOROGENES) — BACILLUS CADAVERIS
SPOROGENES— CLOSTRIDIUM BUTYRICUM

Tetanus

THE causation of tetanus was for a long time involved in mystery.
No obvious or characteristic changes being met with after death,
the disease was regarded by many as "functional." Others
believed that a primary lesion of the central nervous system might
be the cause of the affection, while a few classed it with the specific
diseases.

It had long been noticed that wounds soiled with earth were
specially prone to be followed by tetanus, and Sternberg in 1880,
and Nicolaier in 1884, produced tetanus in rabbits by introducing
a little garden earth beneath the skin. The latter observer found
at the seat of inoculation and in his impure cultures — for he was
unable to obtain pure ones — a distinctive bacillus, and he was able
with these cultures, and with the pus from the seat of inoculation,
to induce tetanus in other animals. Carle and Rattone subsequently
showed that the bacillus of Nicolaier was present in the tissues of,
and secretions from, the wound, in cases of traumatic tetanus in
man, and that inoculation with the pus from such a wound pro-
duced tetanus in the lower animals — observations which were con-
firmed by Rosenbach in 1885. The bacillus was isolated in pure
culture by Kitasato in 1889 by taking the impure cultures obtained
from the wound in a case of traumatic tetanus, heating to 80° C.,
and plating the heated cultures, the plates being incubated
anaerobically in hydrogen.

419

420 A MANUAL OF BACTERIOLOGY

The Bacillus tetani

Morphology. — The Bacillus tetani is a straight, slender
rod with rounded ends, but under cultivation the rods
may grow into longish filaments. It is somewhat motile
and possesses a large number of flagella, three or four of
which are generally thicker than the rest.1 Spores are
freely formed ; they are spherical and develop at one
extremity of the rod, and their diameter being much
greater than that of the rod, the spore-bearing organism
has been likened to a " pin " or " drum-stick " (Plate
XVII. a). It stains with the ordinary anilin dyes, and
also by Gram's method. " Drum-stick " bacilli are not
necessarily tetanus ; other anaerobic bacilli, e.g. B. putri-
ficus (coli), may also have large terminal spores.

Cultural characters.- — The B. tetani is a strictly anaerobic
organism, and will not grow in the presence of a trace of
free oxygen, nor in an atmosphere of carbon dioxide. It
can be cultivated in deep stabs in glucose agar and gelatin,
or in broth by Buchner's method, or in an atmosphere of
hydrogen (p. 73). In a gelatin stab-culture at 22° C. the
growth radiates from the central puncture, and the gelatin
is slowly liquefied. In a glucose agar stab- culture it forms
feathery, radiating outgrowths from the central puncture,
a small amount of gas being formed (Fig. 47). Broth
becomes turbid with the formation of some gas and the
development of a foul odour ; there is no film formation.
The colonies have a central opaque portion surrounded by
diverging rays. It grows on serum without liquefaction
and in milk without curdling. The tetanus bacillus remains
alive for some time, possibly indefinitely, in cultures, and
the spores retain their vitality for years in the dried state,
withstand a temperature of 80° C. for an hour, but are

1 Kanthack and Connell, Journ. Path, and Bact., vol. iv, 1897, p. 452.

THE TETANUS BACILLUS

421

killed by boiling for five minutes. Carbolic acid (1 : 20)
does not destroy the spores under about fifteen hours.

Occurrence and pathogenic action. — Man and the horse
are most subject to tetanus ; cattle and sheep are rarely
affected, while the fowl, frog, triton, snake and tortoise
are immune. Mice, guinea-pigs and
rabbits are all very susceptible. The
bacillus is present in the superficial
layers of the soils in many localities, but
not in all, and this accounts for the fact
that tetanus is rare in some places and
frequent in others. The natives of the
Solomon Islands have made use of this
fact for the preparation of poisoned
arrows. The arrows are tipped with a
viscid fluid, then rubbed in the soil from
a mangrove swamp containing tetanus
spores, and afterwards dried. Individuals
wounded with these arrows generally de-
velop tetanus.

Tetanus spores are frequently present
in the dejecta of cattle, horses, and

other animals, and occasionally of man
, .90v FIG. 47.— Tetanus

bacillus. Stab-

The bacillus is confined to the seat culture in glucose
of inoculation, or at most is met with afar' seven days
in the nearest lymphatic glands, so
that the general symptoms are due to the absorption of
toxin. The researches of Ransom and Meyer have shown
that the tetanus toxin is mainly absorbed by the nerve-
trunks (see also p. 159). The organisms associated with
the tetanus bacillus in earth are probably of considerable
importance in the production of the disease, for it has been
shown that if the tetanus bacilli and their spores be care-
fully washed so as to remove all adherent toxins, they fail

422 A MANUAL OF BACTERIOLOGY

to set up tetanus on inoculation, while if the same washed
bacilli be injected, together with a little lactic acid, tetanus
follows, the explanation being that the bacilli are unable
to multiply unless the surrounding tissues are damaged
and phagocytosis is prevented. The associated organisms
in the wound probably effect this, and do not act by
producing a condition of anaerobiosis as has been suggested.
Semple x has recently found that tetanus spores are occa-
sionally present in the human intestinal tract (Hamilton
suggested that tetanoid organisms in the intestinal tract
might be the cause of the so-called idiopathic or rheumatic
tetanus). He injected guinea-pigs with washed spores,
and tetanus did not ensue, but the tissue at the site of
inoculation, examined five to seven months later, still
contained the living spores. Semple suggests that such
latent spores may in some instances be disturbed and
become active by the hypodermic or intra- muscular injec-
tion of quinine, owing to the tissue necrosis and inhibition
of phagocytosis produced by the drug.

Toxins.- — Cultivated anaerobically in broth, the tetanus
bacillus forms a most potent extra- cellular toxin, so that
if the culture be filtered through a porcelain filter, 0-001 c.c.,
0-0001 c.c., or even 0-00001 c.c. of the filtrate is a fatal dose
for a guinea-pig.

Tetanus toxin broth contains a tetanising substance,
termed tetano-spasmin, and also a haemolysin, tetano-
lysin. The toxin has a special affinity for nerve-tissue
(see p. 159). Injected into animals such as the mouse,
guinea-pig and rabbit, the toxin broth produces tonic,
not clonic, spasm and with small doses the muscles at or
near the seat of inoculation tend first to be affected, so
that the spine may be curved, the leg paralysed, etc.
(Fig. 48).

By treatment with carbon disulphide, tetanus toxin
1 Sc. Mem. Gov. of India, No. 43, 1911.

TETANUS

423

broth becomes practically non-toxic, though it still retains
its power of immunising on inoculation and of combining
with antitoxin — that is to say, bodies are formed analogous
to the toxoids of diphtheria toxin.

Brieger, from impure cultures of the tetanus bacillus,
obtained two basic bodies which he termed " tetanine "
a,nd " tetano- toxin," the former producing tetanic symp-

FIG. 48. — Guinea-pig inoculated with a small dose of tetanus toxin,
showing paralytic condition of right hind leg due to spasm.

toms in mice, and the latter tremor, paralysis, and finally
convulsions. Brieger also isolated tetanine from the
amputated limb of a tetanic patient. Brieger and Frankel
obtained a tox- albumin from bouillon cultures which
induced tetanus in guinea-pigs. Brieger and Cohn subse-
quently investigated the tetanus poison obtained by preci-
pitating veal-broth cultures with ammonium sulphate
added to saturation, and purifying by re- dissolving, preci-
pitating the protein with basic lead acetate, and removing
other soluble impurities by dialysis. The purified product
forms yellow flakes, soluble in water, but not giving the

424 A MANUAL OF BACTERIOLOGY

Millon and xanthoproteic reactions. It is not precipitated
by most metallic salts, and is not carried down by Roux
and Yersin's method of precipitation with calcium phos-
phate. It contains no phosphorus and only traces of
sulphur. Of the most active preparation 0-00000005 grm.
killed a mouse.

In a case of tetanus examined by Sidney Martin, an
albumose, chiefly deutero-albumose, was extracted from
the blood. Injected into an animal, it produced depression
of temperature, followed by progressive wasting, but no
spasm or paralysis.

Antitoxin. — If an animal is cautiously injected with
tetanus toxin, commencing the treatment with a weakened
toxin, and increasing the dose very gradually, a high
degree of immunity is ultimately obtained, and the blood-
serum acquires marked antitoxic properties. The toxin
is obtained by growing the tetanus bacillus in bouillon in
an atmosphere of hydrogen for about three weeks, and
filtering through porous porcelain. To obtain an active
serum treatment has to be prolonged, a horse immunised
by the writer requiring six months. The antitoxic serum
so obtained is by far the most active of any of the sera,
and is now recognised as the proper remedy to use in cases
of tetanus in man. The antitoxic treatment of tetanus
is not nearly so successful as that of diphtheria, and for
this reason : in diphtheria, in a large proportion of the
cases, a local manifestation is present to aid diagnosis
before any serious absorption of the toxin has taken place,
whereas in tetanus the disease is only recognisable by the
symptoms induced by such absorption. Nevertheless,
tetanus antitoxin should always be employed not only
in the fully developed disease, but also in certain cases as
a prophylactic. As the toxin is at once fixed by the nerve-
tissue, the antitoxin should be injected into the central
nervous system in order to obtain immediate action.

MALIGNANT (EDEMA 425

The antitoxin may be standardised by the Roux or by
the Behring method (see p. 280). Recently a method
analogous to that used for standardising diphtheria anti-
toxin has been introduced.1

Clinical Examination

The symptoms of tetanus are usually so obvious that a bacterio-
logical examination is not needed to establish the diagnosis, and
unless there is an evident wound it will be difficult, if not impossible,
to detect the tetanus bacillus.

(1) Prepare several smears of the pus or discharge, and stain
by Gram's method. Examine microscopically, looking for the
spore-bearing rods or " drum-sticks." A " drum-stick " bacillus is,
however, not necessarily the tetanus bacillus (see p. 420).

(2) If " drum-sticks " be found, an attempt may be made to
isolate the bacillus by making anaerobic plate cultivations from
the discharge, after heating it in capillary pipettes to 80° C. for
half an hour.

(3) Inoculate mice and guinea-pigs with the heated discharge.
If they die with tetanic symptoms, treat the pus at the seat of
inoculation as in (2).

Malignant (Edema

Malignant oedema is met with in man in connection
with wounds soiled with septic matter, compound frac-
tures, contused and lacerated wounds, etc. Usually there
is a putrefactive and oedematous condition of the tissues
with subcutaneous emphysema. Animals also occasionally
suffer from the disease, which can be produced artificially
by inoculation with dust, dust from straw, the upper
layers of garden earth, and decomposing animal and
vegetable matter.

If a guinea-pig be inoculated subcutaneously with a
little garden earth, it will very likely die in forty- eight

1 On the standardisation and therapeutic use of tetanus antitoxin,
see Hewlett's Serum Therapy, 1910.

426 A MANUAL OF BACTERIOLOGY

hours. Post mortem, the subcutaneous tissues around
the seat of inoculation will be found to be cedematous and
blood-stained, with more or less development of gas. The
internal organs are only slightly altered, but the spleen
may be somewhat enlarged. The juice from the seat of
inoculation will be found to contain a mixture of organisms,
but in the blood and organs few will be found. Under the
capsule of the spleen, however, long slender rods may
be seen ; these are the bacilli of malignant oedema.

Morphology. — The bacillus of malignant oedema is a long
and slender rod, several of which may be united into a
thread. It is motile, possesses several flagella, and is
readily stained by the ordinary anilin dyes, but not by
Gram's method. It spores freely at temperatures above
20° C., the spores being large and central.

Cultural characters.- — The bacillus of malignant cedema
is strictly anaerobic. In a deep stab in glucose-agar it
forms a thick line of growth in the needle track, with
irregular outline and greyish-white in colour. There is
profuse development of gas, accompanied by a foul odour,
and attended with disruption of the medium into several
portions.

The bacillus of malignant oedema is an organism which has to
be distinguished from anthrax, and there should be no difficulty in
doing this. Post mortem, the spleen is rarely found much enlarged
in malignant oedema, the organism is not very abundant, is almost
entirely absent from the blood, and is only found under the capsule
of the spleen, not at its centre. If, however, several hours have
elapsed since death occurred, the organism may have wandered
into the blood and the centre of the spleen. The bacillus of
malignant oedema is motile under anaerobic conditions, the anthrax
bacillus non-motile ; the former occurs as a long slender filament,
which on staining is seen to consist of two or three long segments ;
it does not stain by Gram's method (except by Claudius's modifica-
tion), and is strictly anaerobic.

EMPHYSEMATOUS GANGRENE 427

Bacillus botulinus

In certain forms of meat poisoning (see Chap. XXI) van
Ermengem isolated an anaerobic bacillus, the B. botulinus. It
is chiefly met with in ham and sausage, and the symptoms are
caused by the absorption of toxin, which has a special effect on the
nerve centres.

The organism is a large Gram-positive sporing anaerobic bacillus,
often occurring in pairs or in short chains. In glucose gelatin it
forms a whitish streak in the line of the stab, with lateral out-
growths, liquefaction of the medium, and gas-formation. The
cultures have a rancid odour, due to butyric acid production. The
colonies in gelatin are semi-transparent spheres. The optimum
growth is from 20°-30° C. The source of the organism is unknown,
but it has once been isolated from the excreta of a healthy pig.

The B. botulinus in broth cultures forms a potent extra-cellular
toxin, which is toxic both by injection and by ingestion. The toxin
is also produced in the infected ham, sausage, etc. With the toxin
an antitoxin can be prepared.

Bacillus Welchii *

Probable synonyms. — B. aerogenes capsulatus (Welch and Nuttall),
Granulo -bacillus saccliaro-butyricus immobilis liquefaciens (Grass-
berger and Schattenfroh), B. enteritidis sporogenes (Klein), B. per-
fringens (Veillon and Zuber), gasphlegmon bacillus (Frankel).
bacillus of acute rheumatism (Achalme : see " Rheumatism ").

This organism was originally described by Welch and
Nuttall under the name B. aerogenes capsulatus, and
occurs in conditions accompanied by much development
of gas in the tissues, as in cases which might be described
either as phlegmonous erysipelas or as emphysematous
gangrene, especially after injuries. It is also met with

1 See Welch and Nuttall, Bull. Johns Hopkins Hosp., vol. iii, 1892,
p. 81 ; Welch, ' Shattuck Lecture,' ibid. vol. xi, 1900, p. 185 ; Dunham,
ibid. vol. viii. 1897, p. 68 ; Welch and Flexner, Journ. Exper. Med.,
vol. i, 1896, p. 5 ; Herter, Bacterial Infections of the Digestive Tract,
1907 ; Kamen, Centr. f. Bakt., Orig. xxxv, 1904, pp. 554, 6S6 ; Archiv.
f. Hyg., vol. liii, 1905, p. 128 ; and Blake and Lahey, Journ. Amer.
Med. Assoc., vol. liv, 1910, p. 1671.

428 A MANUAL OF BACTERIOLOGY

occasionally in perforative peritonitis and in various
septicaemic and pysemic conditions, in the puerperal state,1
complicated stricture, etc.

The B. Welchii is widely distributed, and has been
cultivated from the soil, dust, and contents of the intestine.
It has either been described under a variety of names, or
a group of closely related bacilli may exist. Gas-bubbles
found in the blood and internal organs (" foamy organs ")
at an autopsy seem generally to be due to this organism,
but may occasionally perhaps be caused by other putre-
factive bacteria.

Morphology. — The B. Welchii is a non-motile, sporing,
anthrax-like bacillus, variable in size, being 3 to 6 /x in
length (Plate XVII. b). It occurs singly, in short chains,
or in clumps, and occasionally in long threads. It stains
well with the ordinary anilin dyes and also by Gram's
method. A capsule is often present, but spores are only
formed in blood- serum cultures.

Cultural characters. — The B. Welchii grows well on all
the ordinary culture media, slowly at 20° C., rapidly at
blood-heat, but is strictly anaerobic. It forms greyish-
white colonies on agar, and gelatin is liquefied. In glucose-
broth it produces at first a diffuse cloudiness, but later
the fluid becomes clear and a whitish viscid sediment
settles. Milk is coagulated, the casein forming a thick,
stringy, honeycombed mass on the surface of a clear watery
whey. On potato the growth is almost invisible. There
is abundant formation of gas in culture media, the gas
both in dextrose media and in milk, according to Theobald
Smith, consisting of hydrogen and carbon dioxide in the
ratio 2 : 1 or 3 : 2.

Pathogenicity. — The B. Welchii is pathogenic for guinea-
pigs and mice, but slightly so for rabbits. The whey of a
milk culture in quantities of 0-5-2 c.c. per 100 grm. of
1 See Little, Bull. Johns Hopkins Hosp., vol. xvi, 1905, p. 136.

PLATE XVII.

a. Bacillus tetani. Film preparation of a pure culture.
X 1500.

b. Bacillus Welchii. Film preparation of a milk culture.
X 1000.

BACILLUS WELCHII 429

body-weight produces death in a guinea-pig within forty-
eight hours. Post mortem, if injected subcutaneously,
the hair strips readily from the skin, which may be green
and gangrenous ; the subcutaneous tissue may also be
green and gangrenous, or more or less digested, so that the
skin hangs loose, and the sac formed contains gas and
exudation, sometimes scanty, sometimes abundant, thin
and sanguinolent, and containing numbers of bacilli.
If the post-mortem be delayed, or if the heart-blood be
taken up into tubes, and these are sealed and incubated
for some hours, many of the bacilli will spore. Pigeons,
by intra- muscular inoculation, are also susceptible. Injected
intravenously into a rabbit, the animal killed immediately
and the carcase incubated at 37° C. for twenty-four hours
and examined, there is an abundant formation of gas,
particularly in the liver, which is riddled with gas-
bubbles. This is a very characteristic test ( Welch- Nuttall
test).

The B. Chauvcei also produces this " foaming " condition
of organs when similarly treated, but spores freely, whereas
the B. Welchii does not spore under such conditions.
Monkeys fed with considerable numbers of B. Welchii are
unaffected. In the human intestine the organism is
almost absent or scanty in nurslings and children, but
becomes more and more abundant as age advances. It is
probable that it is capable of producing necrotic changes
in the intestinal mucous membrane. Different strains
seem to vary much in virulence.

Products and toxins. — The gas production has already
been mentioned. Butyric and allied acids are freely
formed, but lactic acid is scanty. Indole may or may not
be produced. Hsemolytic substances can be readily
detected in blood-bouillon cultures, and the organism is
abundant in the intestine in some cases of primary anaemia
and possibly may have some relation to the condition.

430 A MANUAL OF BACTERIOLOGY

In some cases of infection the blood- serum agglutinates
the organism.

Under the name B. enteritidis sporogenes, Klein 1 isolated a bacillus
similar to the B. Welchii from the evacuations of and from milk
consumed by, patients suffering from an epidemic diarrhoea which
occurred in St. Bartholomew's Hospital ; as did Andrewes, 2 from
cases of diarrhoea admitted into the same hospital. Klein believed
this organism to be the cause of the diarrhoea, and stated that it
could not be found in the intestinal evacuations of healthy
individuals. Klein also found it in water, sewage, manure, and
milk. The writer, however, showed that it could generally be
found in the normal dejecta also in road and laboratory dust and
frequently in milk, and the opinion he formed was that it was
probably a ubiquitous organism and had little to do with the
diarrhoea.3 Glynn also found the organism to be very widely
distributed, and fed guinea-pigs with, and himself ingested, cultures
without result. 4

The B. enteritidis sporogenes in its morphology, staining reaction,
and cultural characters is almost, if not quite, identical with the
preceding organism, the B. Welchii or B. aerogenes capsulatus of
Welch. The only point of difference between them is that the
former, according to Klein, is motile and flagellated, while the latter,
according to Welch, is non-motile and non-flagellated. Spores are
only formed in serum or gelatin, not on agar. It is abundantly
present in sewage and sewage -contaminated water (see Chap. XXI).

The Clostridium butyricum of Botkin, an energetic butyric -acid-
forming anaerobic bacillus (p. 432), produces in milk changes
similar to those of the B. Welchii, but is non-pathogenic.

Clinical Examination (Malignant (Edema and
B. Welchii)

The character of the wound and discharge will probably give
some indication of the existence of infection with malignant oedema
or with B. Welchii. The tissues are softened, cedematous, and dis-
coloured, and soaked with a foul-smelling, sanguineous fluid, which

1 Rep. Ned. Off. Loc. Gov. Board, 1895-96, p. 197 ; ibid. 1897-98.
p. 225.

2 Ibid, for 1896-97, p. 225.

3 Trans. Jenner Inst. Prev. Med., vol. ii, 1899, p. 70.

4 Thomson Yates Lab. Rep., vol. iii, Pt. ii, 1901, p. 131.

BLACK QUARTER 431

may be frothy from the development of gas. Other bacilli will
probably be present.

(1) Make films from the discharge. Stain some with Loffler's
blue, and others by Gram's method. Examine microscopically, and
look for bacilli of the forms described. B. Welchii stains, malignant
oedema does not stain, by Gram.

(2) Inoculate two guinea-pigs subcutaneously with the discharge
or with portions of the tissues. It the animals die, look for the
characteristic organism.

(3) An attempt may be made to isolate the bacillus by anaerobic
cultures and plate cultivations, prepared from unheated, and
heated (80° C. for ten minutes), material.

Bacillus cadaveris sporogenes

This is another organism isolated by Klein,1 and has to be dis-
tinguished from the B. Welchii. The two organisms are morpho-
logically very similar and both stain by Gram's method, but the
B. cadaveris sporogenes does not produce the typical changes in
milk. In a culture two or three days old the milk below the cream
layer commences to clear, and later this change proceeds rapidly,
so that at the end of a week three layers are apparent — an upper of
unchanged cream, a middle, yellowish and watery, and a lower of
precipitated casein. Its colonies on agar are also different, sending
out ramifying, anastomosing threads from their margins, and it
spores freely on agar in two to three days.

Black Quarter

Syn. : Black Leg, Quarter Evil, Symptomatic Anthrax, Rausch-
brand.

Black quarter is a disease affecting sheep and oxen, and is un-
known in man. The names black quarter, black leg, and quarter
evil are derived from the dark discoloration of the muscles of the
leg and flanks or quarters of the affected animals. When the muscles
are cut into, a thin sanguineous fluid exudes, and in this fluid slender
bacilli are present, some of which are swollen or club-shaped from
the presence of spores. The muscles are dark, slightly crepitant
owing to the presence of gas, and have a rancid odour.

The organism, the B. (Clostridium) Chauvcei, is a slender rod

1 Centr.f. Bakt. (lte Abt.), xxv, p. 278.

432 A MANUAL OF BACTERIOLOGY

never forming long threads, is strictly anaerobic and motile, but
loses its motility in the presence of oxygen. Some of the rods are
cylindrical throughout, others form slender spindles, others are oval
or lemon-shaped. It stains with the ordinary anilin dyes but not
by Gram's method (except by Claudius's modification). Occasion-
ally in the tissues it seems to stain by Gram. The organism forms
endogenous spores, the spore-bearing rods being enlarged or club-
shaped, and therefore should be termed a " clostridium."

It can be grown in deep stabs in gelatin and agar. Gelatin is
rapidly liquefied. In glucose -agar it forms a thick, irregular,
greyish growth, with much development of foul-smelling gas. The
writer has found extreme difficulty in isolating and in maintaining
cultures of the organism. The guinea-pig is susceptible if inoculated
subcutaneously or into the muscles, the bacilli being found at the
seat of inoculation, but not in the blood or internal organs. Artificial
immunity can be induced in various ways : by bacilli attenuated
by heat or by successive cultivations, or by heating the dried
muscle to 85° to 90° C. for six hours (Kitt), also by inoculating the
susceptible animal at the tip of the tail. Hanna,1 by growing the
organism in a mixture of blood-plasma and broth, obtained toxins
which, by careful injection, conferred immunity on rabbits, the
animals after injection yielding an antitoxic serum.

Hamilton has described specific anaerobic bacilli in braxy,
louping-ill, and other diseases of sheep and deer.2

Clostridium butyricum

An anaerobic organism occurring in milk, in which it produces
a marked butyric acid fermentation with changes like those of the
B. Welchii. It forms short rods, and also long ones 3 to 10 ^ in
length, and filaments are met with. Spore -formation takes place
freely in enlarged segments. It forms a whitish growth on agar,
and gelatin is rapidly liquefied, a scum forming on the surface. It
is non-pathogenic (p. 430).

1 Journ. Path, and Bact., vol. iv, 1897, p. 383.

2 Rep. Louping-ill and Braxy Com., Board of Agriculture and
Fisheries. 1906.

CHAPTER XIV

ASIATIC CHOLERA— SPIRILLUM METCHNIKOVI— SPIRIL-
LUM OF FINKLER AND PRIOR— SPIRILLUM TYRO-
GENUM— SPIRILLUM RUBRUM

Asiatic Cholera

THE bacteriological study of Asiatic cholera may be said to date
from the researches of Koch, who in 1884 was sent by the German
Government to investigate the disease in Egypt and India. He
described an organism present in the intestine and in the dejecta
which he believed to be the specific contagium, and termed it the
" comma bacillus " .from its curved shape. This name is a mis-
leading one, for the organism is not shaped like a printer's comma,
but is a curved rod or vibrio, by some placed in the genus spirillum ;
however, it is commonly known as " Koch's comma bacillus."

Spirillum (Vibrio) choleras asiaticae

Morphology. — Curved rods with rounded ends 1 to 2 JUL
in length, sometimes forming half a circle, sometimes
united in pairs forming an S-shaped curve (Plate XVII I. a).
It is present in the intestine and in the alvine discharges,
especially in the rice- like flakes, but is not found in the
blood, organs, or tissues. (Greig has twice isolated the
organism from pneumonic patches in the lungs and
suggests that in a certain percentage of cases blood-
infection may occur.) In the rice-like flakes it is fre-
quently so numerous that in a film the " commas " are
arranged in " ranks and files " parallel to one another ;
this is known as the " fish-in-stream " arrangement. The

433 28

434 A MANUAL OF BACTERIOLOGY

vibrio stains well with the ordinary anilin dyes, especially
with dilute carbol-fuchsin, but is decolorised by Gram's
method. It is actively motile, and typically possesses a
single terminal flagellum at one end only, but there is some
variation in this respect. Spores are not formed, though
in old cultures Hueppe described bodies which he believes to
be arthrospores. In such cultures the bacilli lose their regular
shape, and swollen and distorted involution forms are seen.

The majority of the organisms in a young agar culture
assume the vibrio form, but in broth or peptone water
cultures two or three days old they are longer and there
is a tendency for them to become somewhat spirillar.

Cultural characters and biology. — The Koch vibrio is
aerobic and facultatively anaerobic, and grows well on
the ordinary culture media from 20° to 37° C. It grows
readily in an atmosphere of hydrogen, but does not develop
in one of carbonic acid gas.

In gelatin plates at 22° C. small cream-coloured colonies
appear in about twenty- four hours, soon accompanied by
liquefaction, so that in two or three days the plate becomes
pitted. Microscopically, the young colonies are rounded
with irregular margins, cream-coloured, and coarsely
granular. In stab- cultures development occurs all along
the stab as a whitish, opaque, punctate growth, thicker
above than below. Liquefaction commences about the
second day and progresses slowly ; in the early stage
it is confined to the surface, and looks like a little bead or
air-bubble (Plate XVIII. 6), but in a fortnight or so the
greater part of the gelatin may be liquefied. Liquefaction
varies greatly both in rate and in extent in different
cultures and stocks ; in some old laboratory cultures it
may be almost absent. On surface agar a thick, moist,
shining, greyish growth quickly develops with more or less
crenated margins, often becoming brownish when old. On
blood-serum much the same growth occurs with slow

PLATE XVIII.

a. Spirillum chohrce. Film preparation of a pure culture.
X 1500.

6 c d

Gelatin stab- cultures, two days old, of (b) Sp. cholerce,

(c) Sp. Metchnikovi, (d) Sp. Finkleri,

434 A MANUAL OF BACTERIOLOGY

vibrio stains well with the ordinary anilin dyes, especially
with dilute carbol-fuchsin, but is decolorised by Gram's
method. It is actively motile, and typically possesses a
single terminal nagellum at one end only, but there is some
variation in this respect. Spores are not formed, though
in old cultures Hueppe described bodies which he believes to
be arthrospores. In such cultures the bacilli lose their regular
shape, and swollen and distorted involution forms are seen.

The majority of the organisms in a young agar culture
assume the vibrio form, but in broth or peptone water
cultures two or three days old they are longer and there
is a tendency for them to become somewhat spirillar.

Cultural characters and biology. — The Koch vibrio is
aerobic and facultatively anaerobic, and grows well on
the ordinary culture media from 20° to 37° C. It grows
readily in an atmosphere of hydrogen, but does not develop
in one of carbonic acid gas.

In gelatin plates at 22° C. small cream-coloured colonies
appear in about twenty- four hours, soon accompanied by
liquefaction, so that in two or three days the plate becomes
pitted. Microscopically, the young colonies are rounded
with irregular margins, cream-coloured, and coarsely
granular. In stab- cultures development occurs all along
the stab as a whitish, opaque, punctate growth, thicker
above than below. Liquefaction commences about the
second day and progresses slowly ; in the early stage
it is confined to the surface, and looks like a little bead or
air-bubble (Plate XVIII. b), but in a fortnight or so the
greater part of the gelatin may be liquefied. Liquefaction
varies greatly both in rate and in extent in different
cultures and stocks ; in some old laboratory cultures it
may be almost absent. On surface agar a thick, moist,
shining, greyish growth quickly develops with more or less
crenated margins, often becoming brownish when old. On
blood- serum much the same growth occurs with slow

PLATE XVIII.

a. Spirillum cholerce.

Film preparation of a pure culture.
X 1500.

6 c d

Gelatin stab-cultures, two days old, of (b) Sp. cholerce,

{c) Sp. Metchnikovi, (d) Sp. Finkleri,

THE COMMA BACILLUS 435

liquefaction. A thin brownish layer is formed on potato
at 37° C. ; and broth becomes turbid, a delicate film
forming on the surface. Peptone water, or Dunham's
modification of it (1 per cent. NaCl), is a good cultivating
medium, and a delicate film forms on the surface. In
milk it multiplies rapidly without curdling ; neutral
litmus glucose- agar is reddened from the development of
acid, but no gas is produced under cultivation. Acid,
but not gas, is produced from glucose, maltose, saccharose,
lactose, and starch.

An important characteristic of the cholera vibrio is
the rapid formation of indole in considerable quantity,
and the reduction of nitrates to nitrites, especially in
peptone water. This forms the basis of the important
cholera-red reaction ; a few drops of pure sulphuric or
hydrochloric acid added to a pep tone- water culture, eight
to twelve hours old, give a pink colour, and the colour is
intense when the culture is two to three days old, and of
a purplish-red colour, like that of potassium permanganate.
Some specimens of " peptone " are unsuitable for pre-
paring the peptone water used for obtaining the reaction,
either on account of the absence of a tryptophane nucleus,
or of nitrates and nitrites. The medium should be sugar-
free, and the addition of 0-01 per cent, potassium nitrate
to it is an advantage. Some believe that two pigments
are formed in the reaction, a cholera- red and the nitroso-
indole pigment.1 The reducing action of the cholera
vibrio can also be shown by growing in litmus broth, which
becomes decolorised (Cahen's test).

Kraus and PrantschofI 2 noticed that certain vibrios
dissolved red blood- corpuscles, but came to the conclusion
that no true recently isolated cholera vibrio is hsemolytic
(see also p. 441).

1 Wherry, Bureau of Government Laboratories, Manila, Bulls. 19
and 31, 1904 and 1905.

2 Wien, klin. Woch., 1906, p. 299.

J.50 A MANUAL OF BACTERIOLOGY

Strong,1 in the Philippines, found that all vibrios which
agglutinated well with a cholera serum were genuine
cholera vibrios and that none of them was haemolytic. On
the other hand, Baerthlein 2 found that seven freshly
isolated strains of the cholera vibrio were definitely hsemo-
lytic in suspensions of sheep's corpuscles in from twenty-
four to forty-eight hours. Van Loghem 3 employs goat's
blood in haemolytic tests for the cholera vibrio. He
asserts that goat's blood is quickly haemolysed by haemo-
lyshig cholera-like (e.g. El Tor, p. 441) vibrios, but that
recently isolated cholera strains, if they haemolyse at all, do
not do so for some time — twenty- four to forty- eight hours.

With regard to this important question of haemolysis
and the cholera vibrios, Van Loghem 4 distinguishes two
types of blood solution, viz. haemolysis proper and haemo-
digestion. He asserts that the apparent haemolysis on
a blood- agar plate occasionally occurring with the true
cholera vibrio is really haemo- digestion. He distinguishes
the two conditions by the tint of the haemolytic zone —
red in true haemolysis and greenish in haemo- digestion —
and spectroscopically the affected zone shows oxyhae-
moglobin in haemolysis but not in haemo-digestion. The
blood agar used for the plates is composed of ordinary
nutrient agar with an addition of 11-12 per cent, of
defibrinated goat's blood.

The cholera vibrio retains its vitality in cultures for a
month. It can multiply in water and on the surface of
moist linen, but rapidly dies on drying. Its thermal
death-point, according to Sternberg, is 52° C. with an
exposure of four minutes.; according to Kitasato, 55° C.
in about ten minutes. ' It is easily destroyed by the
ordinary germicides.

1 Philippine Journ. of Science, vol. v, 1910, p. 403.

2 Arb. aus dem kaiserl. Gesundheitsamte, xxxvi, 1911.

3 Centr.f. Bakt., Abt. I (Originate), Ivii, 1911, p. 289.

4 Ibid. Ixx, 1913, p. 70.

SURVIVAL OF THE COMMA BACILLUS 437

In some experiments by Dempster1 it was found that
the comma bacillus lived from three to five days in dry
soil, but only one day in an artificially dried soil, while
in moist soil it lived from twenty- eight to sixty-eight days.
In peat, however, it was invariably dead within twenty-
four hours. In sterilised salt solution (0-75 per cent.) the
comma bacilli were alive on the 159th day, and in fresh
urine (sterilised) they lived fourteen days at 37° C. and
twenty-nine days at 22° C.

In sterilised distilled water the cholera vibrio usually
rapidly dies, as a rule within twenty-four hours. The
addition of sodium chloride greatly increases the length of
time it may remain alive, a survival of five or six weeks
having been recorded. In ordinary sterilised potable
waters it may survive many months. In unsterilised
potable waters its survival is greatly influenced by the
presence of salts ; in some cases it dies out rapidly ; in
others, especially in those containing a large proportion
of salts, it may remain alive for some time. Houston 2
found that cholera vibrios die very rapidly in raw Thames,
Lee, and New River waters as the result of storage in the
laboratory. At least 99-9 per cent, perish within one
week, and it was not possible to isolate any, even from
100 c.c. of the water, three weeks after infection. Klein 3
found that the cholera vibrio could retain its vitality for
at least fourteen days in unsterilised sea- water, while
from the interior of oysters, kept in water infected with
the vibrios, it was obtained up to nine days after infection.
In sterilised sewage the cholera vibrio multiplies and
survives for months ; in unsterilised sewage it may survive
for two to four weeks (Houston).

Pathogenicity. — The disease is spread mainly by infected

1 Med.-Chir. Trans., vol. Ixxvii, 1894, p. 263.

2 Metropolitan Water Board, Fifth Rep. on Research u-o)k, 1910.

3 Rep. Med. Off. Loc. Gov. Board for 1896, p. 135.

438 A MANUAL OF BACTERIOLOGY

water ; milk, salads, vegetables and flies are other
sources of infection. The organism has been found in
the dejecta of contacts not suffering from the disease, and
it may sometimes persist for long periods after convales-
cence. In these cases the vibrio may sometimes be located
in the biliary tract. Crendiropoulo examined the stools
of 34,461 persons on ships coming from cholera-infected
ports. Cultures of vibrios were obtained from 63 of these,
of which 23 were agglutinated, and 40 were not agglu-
tinated, by a high-titre cholera serum.

The relation of the cholera vibrio to the disease has
been a very vexed question in the past, but the outcome
of the voluminous researches which have been made is
to confirm Koch's work. The organism is found in all
cases of cholera, and several instances of laboratory
infection from cultures have been recorded.

None of the lower animals suffers from or contracts a
disease in any way comparable to Asiatic cholera, so that
the test of animal experiments cannot be applied except
in the case of young suckling rabbits (see below, " Anti-
serum "). By first neutralising the acidity of the gastric
juice by an injection of sodium carbonate solution into
the stomach, then diminishing peristalsis by an injection
of tincture of opium into the peritoneal cavity, and finally
injecting a broth culture of the cholera vibrio into the
stomach, Koch succeeded in inducing in guinea-pigs a
condition somewhat similar to cholera in man — namely,
indisposition with falling temperature, weakness of the
extremities, and death in forty-eight hours. Post mortem,
the small intestine was congested and filled with a watery
fluid containing large numbers of the vibrios. Injected
into the peritoneal cavity of mice, guinea-pigs and rabbits,
the vibrio produces death from a general septicffimia, and
intra-muscular inoculation into pigeons is sometimes fatal.
The virulence varies much and is lost under cultivation.

OCCURRENCE OF VIBRIOS 439

Metchnikoff x ascribes the immunity of animals to intes-
tinal cholera as largely due to the inhibitory action of
the other organisms present in the digestive tract. In
man digestive disturbances are often an important pre-
disposing cause of an attack. The acidity of the gastric
juice is also probably a means of defence (see " Water ").

The blood-serum of an animal immunised by injections
of the cholera vibrio gives a typical agglutination reaction
with recent cultures of the organism. The reaction can
also be obtained with the blood-serum of cholera patients,
sometimes as early as the first day of the disease, but it is
usually of little use for diagnostic purposes, as the disease
generally runs such a rapid course.

Occurrence of the vibrio. — That the cholera vibrio is
etiologically associated with the disease seems to be
beyond any doubt, and so constant is its presence in true
cholera that all investigators, even those who at one time
opposed Koch's views, rely on its detection for the bac-
teriological diagnosis. The matter, however, has become
complicated owing to the detection in various natural
waters of pathogenic vibrios which, although not identical
with the cholera vibrio of Koch, resemble it so closely that
it is difficult to classify them as anything but varieties
of the cholera vibrio. In certain epidemics in India varia-
tions have also been noted in the cholera vibrios that have
been isolated. Sanarelli 2 isolated from the Seine and
Marne thirty-two vibrios, of which four were almost indis-
tinguishable from cholera, except that they were only
slightly pathogenic, but by passage through a series of
animals their pathogenic power was much enhanced.
Sanarelli believed that these were the descendants of true
cholera vibrios that had gained access to the rivers during
some previous epidemic of cholera. At the same time it is

1 Ann. dc I Inst. Pasteur, vii, pp. 403, 562 ; vol. viii, pp. 257, 520.

2 Ibid, vii, p. 693, and ix, p. 129.

440 A MANUAL OF BACTERIOLOGY

to be noted that vibrios may also be present in the normal
intestinal tract of man and animals, and may therefore
gain access to streams (Sanarelli). D unbar similarly, from
the Elbe, Rhine, and other rivers, isolated a number of
vibrios which could not be distinguished from the cholera
vibrios (Spirillum Elwers). It was afterwards noticed that
some of these under certain conditions of oxidation and
temperature became phosphorescent,1 but Rumpel 2 has
also found that cultures of the genuine cholera vibrio may
exhibit phosphorescence, so this cannot be used as a
differential character for the separation of non- choleraic
forms. Neisser isolated a vibrio, which he termed Vibrio
Berolinensis, which agreed with the cholera vibrio in every
particular except that the colonies in a gelatin plate in
forty-eight hours were invisible to the naked eye. Heider
found in the Danube a spirillum, named by him the Vibrio
Danubicus, which resembled the cholera vibrio closely,
but its colonies were somewhat different, and it was more
actively pathogenic to mice. Ivanoff similarly obtained a
vibrio which could only be distinguished from cholera by
the finer granulation of its colonies and more distinct
spiral form. Lastly, there is the Vibrio Massowah, isolated
from an epidemic of cholera at Massowah, which differs
from the Koch vibrio in having two terminal flagella at
each end. Cunningham has also described several vibrios
differing but slightly from the cholera vibrio.

Applying the Pfeiffer and agglutination tests to the
vibrios in question, the following results were obtained.
In the first place, each of the organisms gives a complete
positive reaction to both tests with its own serum ; this,
of course, is only to be expected. Pfeiffer found that,
using his reaction, the variety Ivanoff gave a positive
reaction with cholera serum, and Durham found that

1 Centr.f. Bakt. (ltc Abt.), xviii, 1895, p. 424 (Kutscher).

2 Munch, med. Wochensch)., 1895, No. 3.

EL TOR VIBRIOS 441

Ivanoff and Berolinensis reacted completely with cholera
serum. Conversely, positive reactions with cholera vibrios
were obtained with Massoivah, Danubicus, and Elwers sera,
while Massowah and Elwers react completely to each other.
From these considerations it would therefore seem probable
that some of these vibrios — Sanarelli, Berolinensis, and
Ivanoff — may be varieties of the Koch vibrio. The Massowah
vibrio is usually considered not to be a true cholera vibrio.

Ruffer 1 in 1905 at El Tor isolated vibrios, which may
be distinguished as "El Tor vibrios," from the intestine
of pilgrims returning from Mecca and suffering from various
diseases (dysentery, diarrhoea, pneumonia, rheumatism),
but among whom there had been no cholera, and who had
not been in contact with cholera. These vibrios were sub-
j ected to detailed examination by the agglutination, satura-
tion and fixation tests, and Pfeiffer's reaction with Berlin
cholera-immune serum, and also by the haemolysis test.
Vibrios isolated from a previous epidemic of cholera (re-
ferred to as Group 1), and other vibrios isolated from cholera
and other stool (Groups 3 and 4), were also compared
with the El Tor vibrios. RufTer's results were as follows :

Group 1 (undoubted cholera vibrios). — Those which
react positively to the four principal tests with cholera
serum — namely, the agglutination, saturation, and fixation
tests, and Pfeiffer's reaction. They do not haemolyse,
even when remaining in contact with red corpuscles for
three days at the temperature of the laboratory.

Group 2. — The second group contains the vibrios agglu-
tinated by, and giving the saturation and PfeifTer's reactions
with, cholera serum, but not fixing the cholera-immune
body. These vibrios are strongly hsemolytic. This group
consists of the El Tor vibrios only.

1 Researches on the Bacteriological Diagnosis of Cholera. Sanitary,
Maritime, and Quarantine Council of Egypt, Alexandria, 1907. (Also
Brit. Med. Journ., 1907, vol. i, p. 735.)

442 A MANUAL OF BACTERIOLOGY

Group 3. — The third group is formed by vibrios which
are not agglutinated by immune serum, nor give the
saturation or Pfeiffer's reaction, but fix the cholera-immune
body. These vibrios also hsemolyse, but feebly and late,
often only after thirty-six to forty-eight hours.

Group 4. — The last group is formed by strongly hsemo-
lytic vibrios not reacting at all to cholera-immune serum.

Buffer concludes that the El Tor vibrios are not genuine
cholera vibrios. He says : " The only possible classifica-
tion is to group together all the vibrios reacting in the
same way to all tests, separating them from those which,
under the same conditions, behave in a different way. If
this method be applied to the vibrios found at El Tor,
there is no difficulty in distinguishing them from the true
cholera vibrios, in spite of several of the reactions of both
being similar. And it follows also that the agglutination,
saturation and Pfeiffer's tests are not in themselves of
absolute diagnostic value for cholera vibrios."

Neufield and Haendel,1 however, after a re- examination
of some of these vibrios, consider that they are true cholera
vibrios. The matter therefore remains undecided.

Klein found that the cholera vibrio kept in sea- water
showed marked variation from the original strain. In
the East many cases of cholera are mixed " vibrionic "
infections ; the stools may contain several varieties of
vibrios, some agglutinating with cholera serum, others
not ; some monociliate, others multiciliate.

It may be that, like the B. dysenteries, the cholera vibrio
is not a single definite organism, but that cholera may be
caused by any one of a group of closely allied vibrios.

Toxins. — Brieger in 1887 obtained cadaverin and pu-
trescin and two other basic bodies from cholera cultures.
Brieger and Frankel isolated a tox-albumin, and Gamaleia
a ferment-like body. Hueppe believes that the cholera

1 Arbeit, a. d. Kais. Gesundheitsamte, xxvi, 1907, p. 536.

CHOLERA TOXINS 443

poison is a tox- albumin formed in the culture medium,
but that immunising substances are derived from the
bacterial cells.

Rontaler compared the chemical products of the ordinary
cholera and of the Massowah spirilla, and could find little
difference between them.

Wesbrook 1 obtained albumoses and other bodies from
alkali-albumin, egg, and Uschinsky medium, cultures.
This observer also found aerobic cultures of the cholera
vibrio to be much more toxic than anaerobic ones.

Pfeiffer found that cholera cultures killed with chloro-
.form vapour contained a toxic substance fatal to guinea-
pigs in small doses, with extreme collapse. He believed
the substance to be an integral part of the bacterial cells.

Metchnikoff,2 Roux and Salimbeni demonstrated the
existence of a soluble cholera-poison in a very ingenious
manner. Collodion sacs of 2 c.c. to 3 c.c. capacity were
sterilised, filled with peptone solution, inoculated with
the cholera spirillum, and closed. The closed sac was
then introduced into the peritoneal cavity of a guinea-pig,
which died in three or four days from the effects of the
soluble toxins dialysing through the walls of the sac (see
also next page).

Brau and Dernier 3 obtained a toxic filtrate by culti-
vating the cholera vibrio in a medium consisting of horse-
serum with an addition of 10 per cent, of defibrinated
horse-blood.

Macfadyen obtained a highly toxic endotoxin by tritura-
ting cholera cultures with liquid air.4

Emmerich 5 strongly supports the view that the cholera
intoxication is not a toxin intoxication, but is due to

1 Journ. of Path, and Bact., vol. iv, 1896, p. 1.

2 Ann. de VInst. Pasteur, x, 1896, p. 257.

3 Ibid, xx, 1906.

4 Lancet, 1906, vol. ii, p. 494.

6 Munch, med. Wochenschr., 1911, No. 18, p. 942.

444 A MANUAL OF BACTERIOLOGY

nitrite poisoning, the nitrites being produced by the
reducing action of the vibrios on nitrates present.

Anti-serum. — By growing the cholera vibrio in a
shallow layer with free access of oxygen in a peptone
gelatin-salt medium, Metchnikoff and his co-workers
obtained a toxic fluid after three or four days growth.
During incubation the fluid becomes concentrated to about
one-eighth by evaporation. After filtration, 0-25 c.c.
killed a 300-grm. guinea-pig in eighteen hours. Goats,
inoculated with increasing doses of this toxin, com-
mencing with 10 c.c. and reaching 200 c.c. in six months,
become immunised and yield an antitoxic serum, 1 c.c.
of which will neutralise four times the lethal dose of toxin.
Metchnikoff had previously found that young suckling
rabbits suffer from an intestinal cholera when fed with
cultures, so that the effect of the cholera antitoxin in
preventing intestinal cholera could be tested on these
animals. Experiment showed that of the treated rabbits,
51 per cent, survived, of the untreated only 19 per cent.
Salimbeni employed a serum prepared in this manner in
the treatment of cases of cholera in the Russian epidemic,
1910.

Animals may be inoculated with dead and living cultures
and an immune serum so prepared, but no practical value
has yet attended the use of anti-sera in the treatment of
cholera. Macfadyen immunised a goat with cholera- cell
juice, and obtained a serum of which 5 -J-^- c.c. protected a
guinea-pig against three lethal doses of cholera culture.

The writer prepared an anti-endotoxic serum in this
manner, with which a few cases of cholera were treated
in Russia.1

Vaccine. — Ferran in 1885 first prepared a vaccine by
making cultures (mixed) in broth from cholera stools and
injecting 0-3-0-5 c.c. subcutaneously, but the reports of

1 Lancet, 1910, vol. ii, October 22.

CHOLERA VACCINE 445

the commissions sent to investigate the method were
unfavourable.

Haffkine subsequently prepared a vaccine against
cholera from cultures of the Koch vibrio, which seems
to be efficacious in preventing the disease. For example,
a number of labourers were inoculated during an epidemic,
and among the inoculated the mortality was only 2-25,
whereas among the uninoculated it was nearly 19 per cent.
In another instance amongst 654 uninoculated there were
71 deaths, a mortality of 10-86 per cent., while among
402 inoculated there were only 12 deaths, a mortality of
2-99 per cent., and a reduction in mortality of 72-47 per
cent.

In the Haffkine method two vaccines are made use of.
The first or weak vaccine is prepared from cultures of the
cholera vibrio attenuated by growing on the surface of
agar, with free aeration, for several generations. The
second or strong vaccine is prepared by enhancing the
virulence of a cholera culture by a succession of passages
through the peritoneal cavity of guinea-pigs. The viru-
lence of this culture must be maintained in the same
manner.

For making both vaccines, " standard " agar cultures
are employed. These are tubes in which the sloping
surface of agar measures 15 cm. in length, and the cultures
are incubated for twenty-four hours. The whole growth
on such a tube is emulsified in 8 c.c. of broth or salt solu-
tion ; the dose of this is 1 c.c., and the living vaccines are
injected into the flank, the second or strong being given
seven to ten days after the first or weak. Haffkine 1 in
a recent study on cholera inoculation suggests the use
of the strong vaccine " devitalised." The devitalised
vaccine may be prepared by two methods, (a) prolonged
cultivation in broth and treatment of the culture with

1 Preventive Inoculation against Cholera (W. Thackcr & Co., 1913).

446 A MANUAL OF BACTERIOLOGY

heat and carbolic acid, (6) cultivation on agar and treat-
ment with carbolic acid.

Besredka l claims that an immediate and lasting (six
months) immunity may be produced by making a mixture
of cholera culture and cholera-immune serum, allowing this
to stand for twelve hours, heating to 56° C. for one hour
and then injecting subcutaneously.

Strong 2 prepares a vaccine from autolysed cultures.
The cholera vibrio is grown on surface agar for twenty-
four hours at 37° C. ; the growth is then washed off with
sterile water, the suspension is kept at 60° C. for twenty-
four hours, and then at 37° C. for two to five days, and is
finally filtered through a porcelain filter.

Clinical Diagnosis

Some of the rice-like flakes should be picked out of the stool and
well rinsed in sterile salt solution.

1. From one of the whitish, slimy, rice-like flakes in the evacua-
tions or the intestine films are prepared, stained with Loffler's blue,
washed, dried, and mounted. If on examination large numbers of
curved rods lying in groups parallel to one another are observed,
the diagnosis of Asiatic cholera may be made with some degree of
certainty. Koch states that this is so in quite half the cases,
especially the acute ones. (Single, or a few, vibrios are of no
diagnostic significance ; they may occur in normal and diarrhrea
stools. The presence of numbers of vibrios having the " fish-in -
stream " arrangement is also not absolutely characteristic.)

2. Gelatin and agar plates should be prepared from an emulsion
of rice-like flakes. Agar plates are best prepared by smearing the
flake over the surface of the solidified agar. The plates are incubated
at 22° C. and 37° C. respectively. In the gelatin plates the charac-
teristic colonies of the cholera vibrios should be recognisable in
about twenty-four hours, in the agar plates in from twelve to
sixteen hours. The likely colonies should be examined microsco-
pically and peptone-water and other cultures prepared from them.

A better medium to employ is Dieudonne's blood alkali agar.

1 Ann. de VInst. Pasteur, 1902, p. 918.

2 Bureau of Gov. Laboratories, Manila, Bull. No, 16, 1904 (Bibliog.).

SPIRILLUM METCHNIKOVI 447

Equal parts of defibrinated ox-blood and normal caustic potash
solution are mixed and sterilised in the steamer. Of this 30 c.c. are
mixed with 70 c.c. of 3 per cent, peptone-agar (neutral to litmus),
previously melted. Plates are poured and kept at 60° C. for half
an hour, and are then allowed to stand for twenty -four hours for
ammonia to evaporate. On this medium few organisms except the
cholera vibrio develop (but cholera-like vibrios develop equally
well).

3. With other rice-like flakes several peptone-water cultures
should be prepared and incubated at 37° C. This is best done in
the small Erlenmeyer flasks containing a shallow layer (1-2 cm.
deep) of Dunham's peptone -water, without wool plugs, but capped
with a piece of sterile filter-paper. In eight to ten hours the upper
layers of the fluid should be examined microscopically for the
presence of vibrios, and gelatin, agar or Dieudonne agar plates and
subcultures in peptone-water are also made by inoculating from the
surface layer of fluid. The peptone -water culture may then be
tested for the presence of indole by carefully adding a few drops
of pure concentrated sulphuric acid. In cases of Asiatic cholera
the indole reaction can be obtained as early as eight hours after
inoculation.

If vibrios are found in the peptone -water or other cultures, they
should be tested for agglutination with a high-titre cholera-immune
serum ; if positive results are obtained, the diagnosis is practically
certain. The haemolysis test should also be applied, as it is com-
paratively simple (p. 182).

4. The saturation and fixation tests and Pfeiffer's reaction may
also be applied.

5. If the case has lasted some time, the agglutination reaction
may be applied, testing the patient's serum on a known strain of
cholera vibrio, but this is of doubtful value.

Spirillum Metchnikovi

Isolated by Gamaleia from the intestinal contents of chickens
dead of an infectious gastro -enteritis which occurred in certain parts
of Russia. The disease, although resembling chicken cholera in
some respects, is quite distinct from the latter. This spirillum forms
curved rods and spiral filaments, generally slightly shorter, thicker
and more curved than the Koch vibrio. It is decolorised by
Gram's method, and is best stained with weak carbol-fuchsin. It is
readily cultivated, and is aerobic and facultatively anaerobic. In

448 A MANUAL OF BACTERIOLOGY

gelatin plates it forms small whitish colonies, visible within twenty
hours, which grow more rapidly than the cholera vibrio, and in
two or three days produce marked areas of liquefaction. In a
stab-culture in gelatin a whitish granular growth occurs along the
line of puncture with liquefaction, much like that of the Koch
vibrio, but the rate of growth and the liquefaction are more
rapid (Plate XVIII. c). Grown in eggs by Hueppe's method
typical appearances are produced. After ten days the white becomes
transformed into a yellowish limpid liquid, while the yolk, though
retaining its form and consistence, is quite black. On surface agar
a thick cream-coloured layer develops ; on potato the growth is
brownish, and milk is coagulated. It grows freely in broth and
peptone-water, the fluid becoming uniformly turbid, and a slight
film forms on the surface, and these cultures give a marked indole
reaction on the addition of sulphuric acid alone, in this respect
resembling the Koch vibrio. The S. Metchnikovi is pathogenic
to chickens, pigeons and guinea-pigs, but not to rabbits or mice
except in large doses. It is, however, more pathogenic to guinea-
pigs than the cholera vibrio. Pigeons are killed by intra -muscular
inoculation, and fowls are susceptible to feeding, whereas the
cholera vibrio is not pathogenic to fowls by feeding. It is not
agglutinated with cholera-immune serum. Abbott * isolated a
pathogenic spirillum from the Schuylkill River, Philadelphia, which
resembles the S. Metchnikovi closely, and is probably identical with it.

Spirillum Finkleri (of Finkler and Prior)

Isolated from the stools in certain cases of cholera nostras, but
its etiological significance is doubtful. It occurs as short, thickish,
curved or straight rods, and sometimes as spiral filaments. It is
aerobic and facultatively anaerobic, does not form spores, and does
not stain by Gram's method. In a gelatin stab-culture a yellowish
growth forms with rapid liquefaction (Plate XVIII. d). On agar a
thick, slightly brownish, moist layer develops. Serum is rapidly
liquefied. On potato a slimy brownish growth occurs even at
room temperature. It grows in broth and peptone -water, pro-
ducing a general turbidity. It does not as a rule give the indole
reaction with sulphuric acid alone, but the ordinary laboratory
cultures after three to four days' growth occasionally give a slight
reaction. It is stated to be pathogenic to guinea-pigs by intra-
peritoneal inoculation.

1 Journ. of Exper. Med., vol. i, 1896, p. 419.

SPIRILLUM TYROGENUM 449

Spirillum tyrogenum

Obtained by Deneke from old cheese, and frequently spoken of
as Deneke's spirillum. It forms curved rods and spiral filaments
somewhat closely resembling the Koch vibrio. It grows well on
the ordinary culture media at room temperature, but development
is usually slight or absent at 37° C. In a gelatin stab -culture a
yellowish growth occurs with liquefaction, which is much more
rapid than that of the Koch vibrio, but less so than that of the
Finkler-Prior spirillum. On agar a thinnish, brownish, somewhat
membranous and coherent layer slowly develops at room tempera-
ture. On potato a yellowish growth occurs. It is stated to be
slightly pathogenic to guinea-pigs by intra-peritoneal inoculation.

Spirillum rubrum

A chromogenic spirillum obtained by Koch from the putrefying
tissues of a mouse. In a gelatin stab-culture a dark red growth
slowly develops along the line of puncture without liquefaction ;
at the surface, however, the growth is colourless. In broth at
37° C. it grows freely, producing a general turbidity with a red
deposit at the bottom of the tube ; there is no film formation. In
such a broth culture large numbers of typical spirillar filaments
can be seen, which are thin and delicate, of varying length, and
actively motile. It is non- pathogenic.

Vibrios are common in the mouth, and may be met with in the
discharge of septic ulcers.

CHAPTER XV

STREPTOTHRIX INFECTIONS— ACTINOMYCOSIS— MY-
CETOMA— LEPTOTHRIX BUCCALIS— CLADOTHRIX DI-
CHOTOMA— MYCOSIS TONSILLARIS

Streptothrix Infections (Streptothricosis) l

THE Streptotrichese are a group of thread-forming organisms showing
true, but not dichotomous, branching. Their exact position in the
botanical scale is uncertain ; by some they are considered to belong
to the higher Schizomycetes, forming a connecting link between
these and the Hyphomycetes ; others place them among the latter,
and others make them a separate and distinct group.

The Streptotrichese form a filamentous network, or mycelium,
the individual threads of which show branching, while their terminal
portions undergo segmentation, with the formation of rounded
bodies regarded as spores. The mycelial network, unless old,
stains by Gram's method, and occasionally possesses " acid-fast "
properties. The leprosy bacillus apparently sometimes grows as a
streptothrix, and the tubercle, glanders, and perhaps diphtheria,
bacilli may belong to this group.

Pathogenic streptothrix forms are not uncommon, the best
known being those causing actinomycosis of the ox and other
animals and of man, the white variety of mycetoma, the S. Eppingeri,
more or less acid-fast, originally isolated from a cerebral abscess,
and also causing a variety of madura foot, S. Nocardii of the ox,
and S. canis of the dog. Doubtless cases of streptothrix infection
in man may occasionally be missed, as the clinical characters closely
resemble those of tuberculosis.

Pinoy 2 distinguishes "Actinomycosis," in which the grains in
the pus are formed by very thin, unsegmented mycelial filaments,

1 See Musgrave, Clegg and Polk, Philippine Jomn. of Science, vol. iii,
1908, p. 447 ; Foulertori, Lancet, 1910, vol. i, p. 551, et seq.

2 Actinomycosis and Mycetoma, Bull, de VInst. Pasteur, xi, 1913,
pp. 929, 977.

450

ACTINOMYCOSIS 451

and " Mycetomata," in which the grains are formed by thicker
mycelial filaments, segmented, and with a well-defined membrane.

Actinomycosis

Actinomycosis in man clinically and pathologically
closely resembles tuberculosis, with which in the past it
was frequently confounded.

Actinomycosis in cattle has long been known, but its
exact pathology was involved in considerable doubt until
the researches of Bellinger in 1876. It forms tumours
chiefly affecting the tongue, jaw, face, and throat, and
was described under such varied names as wen, scrofula,
scirrhus, osteo-sarcoma, cancer, wooden tongue, etc.

The tumours after a time break down and discharge,
the tongue often protrudes from the mouth, the saliva
drips, and the animal becomes much emaciated.

On cutting into a " wooden tongue," or wen, a grating
sensation is felt, such as that experienced in cutting a
turnip or unripe pear ; on examining the section little
rounded, yellowish, frequently almost caseating areas will
be noticed, resembling old tubercles. On making sections
and examining with a low power, these rounded areas are
found to be composed of masses of small round-cells, with
occasionally giant-cells, surrounded by a capsule of fibrous
tissue. The growth may be so soft as to be practically
purulent, and abscesses varying in size from a pin's head
to that of an orange may be present in the affected areas.
Like tubercles, the growths may become caseous, calcified,
or fibrous. In the growth or in the pus from abscesses,
when examined fresh with a low power, yellowish or
yellowish- white granules will be found here and there,
which may be very minute, or as large as a small pin's head,
are somewhat soft in consistence, and on slight pressure
flatten out. Examined with a high power, these granules

452 A MANUAL OF BACTERIOLOGY

are found to contain round, ovoid, or reniform bodies
which have a rosette-like appearance, a more or less
structureless centre with club-shaped bodies radially
arranged around the periphery (Plate XIX. a). These
peculiar structures are the cause of the disease, and are
the form assumed in the animal body by an organism
belonging to the streptothrix group termed the Actino-
myces, or Streptothrix bovis (Nocardia bovis), or, from its
appearance, the ray fungus.

Sections of the diseased tissues show the structure of
the organism still better. Gram's method usually gives
good results, and it will generally be found that the fol-
lowing appearances can be observed : Surrounded by the
round- cells are the reniform or ovoid bodies, situated at
the periphery of which are radially arranged, club-shaped
structures deeply stained with the gentian violet, while
the central portion is unstained and structureless, or
contains granular matter or calcareous particles. Various
appearances may be met with in different parts of the
section, according as the actinomycotic nodules are cut
through their centre or periphery ; when the latter is the
case, the clubs are shown in transverse section and appear
as closely packed, deeply stained dots. Sometimes, how-
ever, in addition to the clubs, the centre of the rosette is
occupied by numerous interlacing filaments, also stained
by the gentian violet.

In man, actinomycosis is usually associated with sup-
puration. If a little of the pus be examined it will probably
contain tiny yellowish or sulphur- yellow granules, which,
microscopically, are found to consist of tufts of fine tangled
filaments, the ends of which may be continued into little
swellings or clubs. In teased-up specimens, or in sections
stained by Gram's method, an appearance is observed
very different from that of the bovine variety, viz. tufts of
interlacing filaments stained by the gentian violet, but a

PLATE XIX.

• • ^;'-^--1

^^"">"^-ik

M6&ft&&.

a. Actinomycosisbovis. Section of tongue. Gram. X 350.

• *- ' * * jt**£

6. Mycetoma. Section of tissue," white variety. Cram. X 350.

ACTINOMYCOSIS 453

complete absence of purple clubs (Plate XX. a). Clubs,
however, are frequently present around the periphery of
the filamentous tufts in a stunted condition, although they
do not usually stain by Gram's method. These clubs are
often seen better in fresh specimens of the pus or in
unstained sections, or by staining with orange-rubin, or
the Ehrlich-Biondi reagent (Plate XX. 6). The con-
ditions in cattle and man, at first sight so very different ,
are thus seen to be similar, a similarity which is further
established by the occasional occurrence in cattle of
filamentous tufts, staining by Gram's method, within the
rosettes, and by the clubs in man now and then taking
on the gentian- violet stain.

Cultural characters. — The cultivation of the Actino-
myces can be performed by collecting the pus from a case
of the disease in sterilised tubes, and subsequently turning
it out into a sterilised capsule and picking out the actino-
mycotic granules with sterilised needles, planting these
on the surface of glycerin agar, and incubating at 37° C.
A certain number of the tubes will probably be uncon-
taminated, but in others a growth of the Micrococcus
pyogenes var. aureus or other pyogenic organism, which is
not unfrequently associated with the Actinomyces, may
occur. In the uncontaminated tubes a growth begins to
appear in a few days in the form of little colonies of a tough
membranous consistence, somewhat wrinkled, greyish, and
shining, while the agar beneath them becomes stained
brownish. The growth increases and the colonies coalesce,
forming a brownish, wrinkled, membranous expansion,
sticking firmly to the agar^and difficult to remove or
break up, while the agar becomes stained brown through-
out ; later on the membranous growth may become
dappled with yellow as though powdered with flowers of
sulphur, but occasionally remains whitish. In gelatin
little spherical feathery tufts develop, and sink to the

454 A MANUAL OF BACTERIOLOGY

bottom as liquefaction progresses. On potato a remark-
able growth develops ; at first brownish, it afterwards
becomes almost black, and is very thick or heaped up with
a much wrinkled surface, while later on it has the appear-
ance of being sprinkled with flowers of sulphur (Fig. 49).
In broth delicate woolly flocculi form.
Films from young agar cultures show
masses of tangled filaments, which appear
to be more or less branched, and stain
well with the ordinary anilin dyes and by
Gram's method ; with the latter the fila-
ments often appear somewhat beaded,
but no trace of rosette formation or even
of clubs is ever found in cultures (Fig. 50).
In pus, especially human, the filaments
can sometimes be seen if stained by
Gram's method with orange-rubin. Inocu-
lated into the peritoneal cavity of rabbits
" ^ and guinea-pigs the cultivated organism
reproduces the disease, numerous actino-
mycotic nodules forming in the peritoneum
and elsewhere. There is much doubt as
to the mode of spread of, and the infection
FIG. 49.— Actino- of man with, the disease. It does not seem
myoes. Potato to be particularly contagious, and diseased
months old. * anc^ healthy animals are often placed to-
gether without bad result ; it can, however,
be conveyed by direct inoculation, for calves inoculated
intraperitoneally with portions of diseased tissues die after
some weeks or months, with an abundant development
of actinomycotic nodules, as shown by the experiments
of Jone and Ponfick. Crookshank also infected a calf with
the material from a human case. Feeding experiments
give negative results. The view generally held is that
the organism occurs on cereals, straw, or roots, and gains

PLATE XX.

a. Actinomycosis hominis.
mycelial mass.

Section of liver showing
Gram, x 500.

b.

clubs. Gram, x 350.

Section showing a ring of stunted
Same material as Fig. a above.

THE ACTINOMYCES

455

access to the system through slight scratches or wounds
in the mucous membrane of the mouth, pharynx, or larynx.
In man no source of infection has been traced, though
cases have been reported where the disease has occurred
after eating grains of barley, etc. The disease is met with
not only in cattle, but also in horses and swine. In the
last-named animals considerable calcification may be

FIG. 50. — Actinomyces. Film preparation.
Gram. X 750.

present in the nodules, and it may be necessary to decalcify
with dilute nitric or hydrochloric acid before the rosettes
can be stained.

It is important to note that tuberculin may cause a
reaction in actinomycosis, similar to that which occurs
in tuberculosis, and as the actinomycotic lesions are very
like those which are found in the latter disease, mistakes
may easily be made, and can only be avoided by a micro-
scopical examination. It is of considerable practical

456 A MANUAL OF BACTERIOLOGY

importance to distinguish actinomycosis from tuberculosis,
for in many cases of the former, both in man and in animals,
iodide of potassium exerts a specific curative action.
Vaccine treatment has also been employed with a certain
amount of success.

By some several species of Actinomyces are believed to exist,
but Homer Wright x considers that but one species of micro-
organism is the etiological agent, both in man and animals, the
A. bovis. Pinoy regards Actinomycosis in man as caused by several
fungi (Nocardia, Indiella, Colmistreptothrix).

" Farcin des bceufs," a disease of cattle occurring in Guadeloupe,
and characterised by infection first of the skin and afterwards of
the lymphatic glands and viscera, is due to the S. Nocardii.

Clinical Examination

1. Pour out the pus or discharge into a large capsule or Petri
dish so that it forms a thin layer, look for any yellowish or other
granules, pick them out with a needle, and place on a clean slide
in a drop of 50 per cent, glycerin. If no granules can be found, a
little of the discharge may be spread on a slide with a drop of 50 per
cent, glycerin. Cover with a cover-glass, and apply a little pressure.
Examine with a f-in. objective. If any actinomycotic tufts are
present they will be seen as yellowish or pale brownish, spheroidal,
ovoid, or reniform masses, and with a ^-in. objective will be found
to have a radiating structure from the presence of the clubs.

2. Stain films of the discharge, by Gram's method, with eosin.
The actinomycotic tufts will generally be found to consist of little
masses of tangled filaments stained violet, and surrounded by a
pink zone which has an indistinct radiating structure.

N.B. — In most instances the clubs in Actinomycosis hominis do
not stain by Gram's method. The reverse is the case in Actino-
mycosis bovis.

3. Sections of actinomycotic tissue are best prepared by the
paraffin method. If frozen, the actinomycotic nodules are very
apt to fall out. Sections may be stained by any of the following
ways :

(a) By Gram's method, with eosin or orange-rubin.

(6) With the Ehrlich-Biondi triple stain. Stain for from half an

1 Journ. Med. Research. 1905.

MADUKA DISEASE 457

hour to two hours. Place in methylated spirit until the sections
appear greenish, then pass through absolute alcohol and xylol.
The clubs are stained yellowish-brown, and are sometimes shown
in human cases when unstained by Gram's method.

(c) By Plant's method. Stain in warm carbol-fuchsin for ten
minutes, rinse well in water, stain in a saturated solution of picric
acid in methylated spirit for five to ten minutes, rinse well in water,
place in 50 per cent, alcohol for ten minutes, pass through absolute
alcohol and xylol.

(d) Good preparations may be obtained by staining in Ehrlich's
haematoxylin and counter-staining with orange rubin. This may
also show the clubs when they are unstained by Gram's method.

Madura Disease or Mycetoma

Madura disease, otherwise known as madura foot, mycetoma, or
the "fungus disease of India," is a chronic local affection generally
attacking the foot, occasionally the hand, sometimes extending up
the leg, but rarely to the trunk. The disease occurs in certain
districts in India, and full descriptions of it have been given by
Vandyke Carter and by Lewis and Cunningham. A " madura "
foot appears enlarged, and numerous sinuses with raised mammilated
apertures open on the surface (Fig. 51). On making a section into
the diseased tissues the bones are found to be more or less carious,
while the soft structures are tough and hypertrophied from the
occurrence of chronic inflammatory changes. Numerous small
cavities are present, sometimes filled by yellowish granules resem-
bling fish-roe, and hence termed " roe -like particles," at others
containing black particles of irregular shape, coal -like consistence,
and variable size, exceptionally as large as a marble or walnut.
The presence of the white or black granules, which may be dis-
charged from the sinuses before mentioned, divides the disease into
two classes — the so-called white and black varieties. Lewis and
Cunningham have also described a third variety, in which the
granules are red like cayenne pepper.

Vandyke Carter l first called attention to the similarity between
the white variety and actinomycosis in their microscopical characters.
In sections stained by Gram's method more or less crescentic or
reniform bodies are noticeable, divided into wedge-shaped areas,
which contain masses of fine filaments stained purple. Surrounding

1 Bombay Med. and Phys. Soc., vol. ix, 1886 (new series), p. 86. Also
Hewlett, Trans. Path. Soc. Lond., vol. xlii, 1893.

458 A MANUAL OF BACTERIOLOGY

the crescentic bodies is a zone of radially arranged elements, many
of which are fan-shaped owing to branching ; they are indistinct,
as they do not stain with the gentian violet, but they are very
suggestive of the club-shaped structures present in actinomycosis,
and they resemble the Actinomycosis hominis inasmuch as they do
not stain by Gram's method (Plate XVIII. 6). By staining with
haematoxylin and orange rubin, or with the Ehrlich-Biondi triple
stain, here and there in the radial zone well-defined clubs can be
demonstrated. It seems, therefore, that the radial zone is composed
of degenerate club-shaped structures, and the disease evidently

FIG. 51. — A foot affected with madura disease. (White variety.)

closely resembles actinomycosis, but seems to be due to a different
species of streptothrix.

From a case of the white variety 1 Boyce cultivated a streptothrix
which differed somewhat from the Actinomyces, as it grew slower,
produced no pigment, and on agar formed white raised colonies
with radial grooves, not unlike the tiny barnacles found on wooden
piles in the sea. Vincent 2 also isolated a streptothrix, perhaps
identical with that of Boyce, which differed from the Actinomyces
in growing feebly in broth, in not liquefying gelatin, and in not
being inoculable in the rabbit. He describes it as forming on
glycerin agar umbilicated colonies, first white and afterwards red.
Shattock 3 suggests that the red, cayenne -pepper-like grains occa-
sionally met with in mycetoma may be due to colonies of the strepto-

1 Hygienische Rundschau, 1894, No. 12.

2 Ann. de VInst. Pasteur, 1893.

3 Trans. Path. Soc. Lond., vol. xlix, 1898, p. 294.

MADURELLA 459

thrix which have produced their pigment. Microscopically, this
organism (Streptothrix madurce, Nocardia madurce) is identical with
the Actinomyces. Musgrave and Clegg in a case of the white
variety isolated a streptothrix (S. freeri) differing from the S.
madurce, but identical with the S. Eppingeri (Nocardia asteroides).

The relation of the black to the white variety of madura disease
has been somewhat debated. Kanthack x described the black
variety as being probably a late stage of the white. It seems,
however, that the co -existence of the two conditions in the same
specimen is very rare, and Boyce and Surveyor,2 after a critical
examination of a large number of specimens, came to the conclusion
that the black variety is a distinct disease, and due to an organism
belonging to the group of the higher fungi, the black particles or
masses being the lignified mycelium or " sclerotium " such as is
met with in ergot.

Pinoy regards the white variety as an Actinomycosis, the black
variety as a Mycetoma.

It is difficult experimentally to reproduce mycetoma in animals,
but Pinoy has succeeded in doing so with an Aspergillus, and
Xicolle with Madurella tozeuri (North Africa), both in pigeons.

By planting out the granules from an early case of the black
variety Wright succeeded in cultivating a hyphomycete. 3 It
formed long branching hyphse, but no spore-bearing organs were
produced, and inoculation experiments on animals were negative.
It grew on potato as a dense, widely spreading, coherent, velvety
membrane, in colour pale brown with white periphery. Small
drops of brown, coffee-coloured fluid appeared on the surface, and
the potato became brown throughout. On agar the growth formed
a meshwork of widely spreading greyish filaments ; in old cultures
(also in potato infusion) black hard granules, or " sclerotia," were
observed. In broth little balls of radiating filaments developed.

It would seem that there are several conditions, both in actino-
mycosis and in mycetoma, having a general resemblance but
differing slightly, and dependent upon different species of parasitic
organism.

According to Pinoy (loc. cit.}, the Mycetomata are caused by fungi
belonging to the genera Madurella, Aspergillus, and Sterigmato-
cystis. The common form in the Indian and African Mycetoma is
Madurella mycetomi (Laveran).

1 Journ. Path, and Bact., 1892.

2 Proc. Roy. Soc. Lond.. 1893, and Phil. Trans. Roy. Soc. Land.

3 Journ. Exp. Med., vol. iii, 1898, p. 421.

460 A MANUAL OF BACTERIOLOGY

Mycosis tonsillaris (Mycosis pharyngis lepto-
thricia)

A chronic disease attacking young adults, resistant to treatment,
and characterised by the presence of small, white, tough, adherent
excrescences on the mucous membrane of the pharynx. Micro-
scopically, the patches consist of collections of epithelial cells and
debris, infiltrated with leptothrix filaments and bacteria. The
disease, however, seems to be a keratosis, infection with the
organisms being secondary.

But occasionally a true " mycosis " apparently occurs, readily
amenable to treatment, and due to a leptothrix.1

Leptothrix buccalis

Four somewhat similar thread forms occur in the mouth, viz.
Leptothrix racemosa, L. buccalis maxima, L. innominata, and
Bacillus maximus buccalis. The first is very common, forms large
threads, shows a peculiar beaded appearance on staining which has
been regarded as sporulation, and may be a fungus form. L. buccalis
maxima and L. innominata differ from each other in that the former
gives a blue granulose reaction when treated with iodine and dilute
sulphuric acid, while the latter does not. All these three organisms
are very similar, and the filaments are either unsegmented, or the
segments are of considerable length. The B. maximus buccalis is
very like the L. buccalis maxima, but does not give the granulose
reaction, and its segments are shorter. It is motile, flagellated, and
sporing, and stains by Gram's method.

Some confusion exists respecting the thread forms of the mouth.2

Cladothrix dichotoma

An organism not unfrequently met with in natural waters. It
forms long threads, straight, or sometimes slightly undulating, or
even spiral and apparently branched, though the branching is not
dichotomous. It can be cultivated on the ordinary laboratory
media at room temperature, forming on agar a brownish, wrinkled,
tough, membranous layer, very adherent, and staining the medium
beneath it a pale brown, not unlike the Actinomyces in these respects.
It is non-pathogenic.

1 See Glasgow Medical Journal, No. 2, 1896, p. 81 et seg. (Brown
Kelly). 2 See Goadby, Mycology of the Mouth.

CHAPTER XVI

THE SACCHAROMYCETACE.E

The Pathogenic Blastomycetes — Yeasts and Fermentation

The Yeasts

THE Saccharomycetacese or Yeasts are characterised by a vegeta-
tive reproduction by budding or gemmation. If a cell of ordinary
brewer's yeast be watched under conditions favourable to growth
and reproduction, it will be found that a slight protuberance makes
its appearance at one pole of the organism ; this increases in size,
and ultimately a daughter-cell resembling the parent is reproduced
and separates off.

The true yeasts also reproduce by spore -formation by ascospores
(p. 465) ; in some there is a fusion of cells before sporulation, in
others the first cell formed by germination of the spore undergoes
fission, forming what is known as a pro-mycelium, after which the
cells multiply by gemmation. The Saccharomycetaceae may there-
fore be divided into :

1. Zygosaccharomyces, in which pairs of cells fuse before sporula-
tion.

2. Saccharomyces, in which there is no fusion of cells before
sporulation, and in which the spores germinate by ordinary budding.

3. Saccharomycoides, in which the spores germinate by means of
a promycelium.

Besides the true yeasts, there are a number of budding forms
known which do not spore. These have been termed " Torulse "
(any yeast-like cell is frequently called a " torula "). Some form
films on saccharine li quids and are known as Mycoderma. Organisms
are also known having a yeast-like form and multiple spores but
multiplying by fission ; these have been termed Schizosaccharo-
myces. The position of these forms is uncertain and they are
classed by the botanist among the Fungi Imperfecti (p. 470).

In addition to reproduction by gemmation, the Saccharomyce-

461

462 A MANUAL OF BACTERIOLOGY

tacese are also distinguished from the Bacteria by their larger size,
and in those forms in which endospores occur by the spores being
multiple and not single in each cell and by having a cellulose cell-
wall. From the Hyphomycetes, or moulds, the Saccharomycetaceae
are distinguished by being unicellular, and by the reproduction
being generally asexual. The Saccharomycetacese, however, are
probably much more nearly allied to the Hyphomycetes than are
the Bacteria, for many of the moulds have a stage in which the
mycelium (see next chapter) resembles an aggregation of yeast-cells,
and the yeasts in old cultures form films in which the cells become
much elongated, like those in the mycelium of a mould. Jorgensen
and others have attempted to show that some of the yeasts are
stages in the development of a fungus, but it cannot be said that
this has yet been satisfactorily demonstrated.

Pathogenic Yeasts l

Organisms apparently belonging to the Saccharomyce-
tacese and termed Blastomycetes have been isolated from
certain tumours, and have been regarded as having an
etiological significance in connection with malignant
disease. Sanfelice cultivated yeast forms from fermenting
fruits, which, on inoculation into guinea-pigs, produced
death in about a month with the formation of a tumour
at the seat of inoculation and embolic growths in the
spleen and liver. He also obtained a similar yeast from
an ox affected with carcinoma, which on subcutaneous
inoculation killed guinea-pigs in about two months, and
inoculated into the peritoneum in a month, with multiple
embolic growths in the lungs, spleen, and mesenteric glands.
A good deal of calcification was present in the growths,
from which fact Sanfelice named this yeast Saccharomyces
litogenes. Rabinowitch and also Foulerton 2 have found
that some of the ordinary yeasts give rise to tumour
formation on inoculation, especially in the rabbit. These

1 See Le Count and Myers, Journ. of Infectious Diseases, vol. iv,
1907, p. 187.

2 Journ. Path, and Bact., vol. vi, 1899, p. 37.

PATHOGENIC YEASTS 463

tumours produced by yeasts are probably granulomata and
not true malignant tumours.

Curtis l obtained a yeast from an apparently myxo-
matous tumour in a young man. The organism was met
with in two forms — free and encapsuled. The free form
appeared in young agar cultures as round or ovoid cells
measuring 3 to 6 /x in diameter, often showing budding.
The encapsuled form was met with in the original tumour
and in the tissues of inoculated animals, and occurred
as a large sphere 16 to 20 /u. in diameter, enclosing the
yeast cell, the capsule being hyaline and 4 to 6 M in thick-
ness. On agar at 37° C. the organism formed whitish,
opaque, creamy colonies in two to three days, becoming
a thick creamy growth at the end of a week, on gelatin
white colonies or growth in four to five days without
liquefaction, and in broth a flocculent deposit, the broth
remaining clear. It was aerobic, did not grow on serum,
and formed a small quantity of acetic acid and alcohol
when grown in beerwort and sugar solutions. It was not
pathogenic for guinea-pigs, but inoculated into rabbits,
rats, mice, and dogs it produced tumours and caused death.
The tumours to the naked eye appeared to be myxo-
sarcomata, and in them the yeasts were found.

Busse also obtained a pathogenic yeast from a young
woman who suffered from a tumour of the tibia, and
ultimately died with diffused growths in the bones and
organs. The yeast-like cells were observed in the affected
parts, and were isolated by cultivation, and the cultures,
inoculated into mice and rabbits, produced death with
growths in the organs. As in Curtis's case, the cells in
the tissues appeared to be encapsuled.

Gilchrist described a case of blastomycetic dermatitis.
Small miliary abscesses were present in the rete and
corium, in the pus of which the parasitic cells were
1 Ann. de Vlnst. Pasteur, x, 1896, p. 449 (Refs.).

464 A MANUAL OF BACTERIOLOGY

observed. These were usually in pairs of unequal size,
the largest measuring about 16 /*, surrounded by a well-
defined capsule, and containing a granular protoplasm in
which a vacuole was present. Clinically, the case had
been regarded as one of scrofuloderma, but no tubercle
bacilli could be found.

Numerous cases of blastomycetic dermatitis have now
been recognised, and several instances of general systemic
blastomycetic infection have been recorded.

Granulomatous tumours occurring in epidemics among
horses in Japan, France, and Italy are also caused by
Blastomycetes.

Clinical Examination (Pathogenic Yeasts, etc.)

The cells can be well seen in the fresh state in the teased-up
tissues mounted in water or glycerin.

Curtis recommends staining in carbol-thionine blue, and for
sections, picro- carmine.

Busse's method for sections is as follows :

1. Haematoxylin solution for fifteen minutes.

2. Wash in distilled water.

3. Counter-stain in weak carbol-fuchsin (1 : 20) for thirty minutes
to twenty -four hours.

4. Decolorise in 95 per cent, alcohol for fifteen seconds to one
minute.

5. Absolute alcohol, xylol, mount in Canada balsam.
Gilchrist recommends treating the sections with 10 per cent.

caustic potash solution and examining in 50 per cent, glycerin
without staining.

Brayton recommends that small pieces of the tissues should be
excised from the growing margin, treated with ether for two to
five minutes, macerated in 20 to 30 per cent, caustic potash solution
for five to ten minutes, and then examined without staining.
Cultures may be readily obtained, with a little care, preferably on
beer-wort gelatin or maltose agar.

FERMENTATION 465

Fermentation

The yeasts are of great importance in inducing many chemical
changes, especially alcoholic fermentation, beer and wine being
almost exclusively due to their activity.

Taking brewer's yeast, Saccharomyces cerevisice, as a type, the
yeast cell is observed to be slightly ovoid in shape, measuring 8 to
9 p. in diameter. The protoplasm is granular, contains one or more
clear spaces or vacuoles, frequently bright, refractile globules of
fatty matter, and is surrounded by a cell wall of cellulose. It has
been repeatedly stated that a nucleus is present, but this is doubtful.
When the yeast-cell is freely supplied with nutriment, reproduction
by gemmation proceeds rapidly, and a whole string of cells may
form owing to the daughter-cells budding again before they have
separated from the parent. When the cell is starved, gemmation
ceases, fat-globules and vacuoles increase in number, and the cell
may finally become little more than a large vacuole, the protoplasm
forming a thin coating over the inside of the cell wall. Within the
vacuoles are often seen minute spherical bodies of a doubtful nature
in rapid movement. In ordinary circumstances endospore forma-
tion does not occur, but by deprivation of nutriment, as by growing
on a block of plaster-of -Paris, the cells develop spores. First the
cell becomes divided by the development of membranes, the so-
called " partition- wall formation," into several chambers in which
the spores form. In the different yeasts the number and arrange-
ment of the spores vary ; in the S. cerevisice the typical number is
four, arranged close together, three on one plane and one resting
on these, like a pyramid of billiard balls.

Although the reproduction of yeasts by gemmation or ascospore
formation is usually asexual, ascospore formation is sometimes
preceded by conjugation of sister-cells, or conjugation may occur
between neighbouring cells at the moment of germination (Guillier-
mond, Nadson, and Marchand).

The spores are of considerable importance in the identification
of species of Saccharomyces, as the form of the cells alone and the
growths on culture media are not sufficiently distinctive. In fact
so little can these two characters be relied upon that in order to
isolate in pure cultivation it is necessary to grow from a single cell.
This can be done by making a miniature plate cultivation with
wort-gelatin on a large sterilised cover-glass, and, after the layer of
gelatin has set, mounting, gelatin downwards, on a large cell on a
glass slide. The cover-glass should be divided into small squares

30

466 A MANUAL OF BACTERIOLOGY

by cross-lines etched on the glass and numbered. The preparation
is carefully examined with a J or J inch objective, and the positions
of single isolated cells are noted. This is not a difficult matter on
account of the comparatively large size of the yeast-cells, and their
position is determined by the numbered squares on the cover-glass.
The preparations are kept in a moist chamber in a warm place,
and when visible colonies have developed, those which are derived
from a single cell can be inoculated into tubes or flasks of a suitable
culture medium.

It is found that the various yeasts form spores in different periods
of time when grown under similar conditions, and on this fact is
based what is known as the analysis of yeast — a most valuable
method, which we owe to Hansen. The chief " diseases " of beers
and yeast — i.e. abnormal fermentations giving rise to inferior pro-
ducts— are due to admixture of certain " wild yeasts," as they are
termed, with the brewer's yeast, chiefly the S. ellipsoideus and
S. pastor ianus ; and, in order to detect these " disease " species,
the analysis consists in determining at what time ascospores appear.
The mode of procedure is as follows :

The yeast is sown in a flask of sterile wort, and incubated at
25° C. for twenty -four hours. The yeast revives, and from the
deposit of young cells two cultures are made on plaster-of-Paris
blocks. These cultures are kept, one at 25° C., the other at 15° C.,
and are examined twice daily. In an uncontaminated brewing
yeast ascospores should not be detected in less than thirty hours
in the culture kept at 25° C., and seventy-two hours in that kept at
15° C. The plaster-of-Paris blocks are sterilised by careful flaming
in the Bunsen, and are then placed in sterile glass capsules with
lids, containing sufficient sterilised water thoroughly to moisten the
whole of the blocks ; unless this is done no growth occurs. By
this method of analysis as little " wild yeast " as one two-hundredth
of the whole can be detected.

Besides the distinct species of yeasts, there are also a number of
varieties employed in brewing, etc., differing but slightly in
morphological and cultural characters, yet giving rise to varied
products. These varieties may be divided into two groups — the
surface, high or top, and the sedimentary, low or bottom, fermenta-
tion forms. In this country beer is brewed by fermenting an
infusion of malt (" wort ") with yeast, which, during fermentation,
rises to the surface, and belongs to the first group ; while the German
beers are obtained by yeast, which sinks to the 'bottom, and belongs
to the second group. The floating of the yeast in the high fermenta-
tion process seems to be due to the attachment of minute bubbles of

FERMENTATION 467

carbonic acid gas to the cells, and it has not yet been possible to
convert the one form into the other.

Characters of some of the more important yeasts. — Hansen divides
the important yeasts into groups having the same general characters,
and distinguishes the varieties in each by Roman numerals (I,
II, etc.).

CEREVTSLS: GROUP. — These are the yeasts producing the normal
fermentations resulting in beer, etc. They are round or slightly
ovoid cells, and four ascospores are produced. In old cultures long
sausage-shaped or even filamentous cells may be met with.

8. cerevisice I. and II. — These are bottom fermentation forms in
use at the Old Carlsberg Brewery ; the cells of No. II are rounder
and slightly larger than those of No. I, and ascospore formation is
more abundant.

There is also a top fermentation form described by Hansen
(S. cerevisice I top), which is the yeast employed in the breweries of
London and Edinburgh.

The yeasts of the cerevisice group can invert cane sugar, select
dextrose from Isevulose, and ferment maltose, but they cannot
ferment lactose, nor decompose malto-dextrin.

PASTORIANUS GROUP. — These are wild yeasts. The cells are
elongated or sausage-shaped, and six or eight ascospores are pro-
duced in a cell.

8. pastorianus I. — A bottom fermentation yeast producing a
bitter taste in beer.

8. pastorianus II. — A feeble top fermentation form. Surface
cultures on yeast-water gelatin have smooth edges, which dis-
tinguishes it from the next species.

8. pastorianus III. — A top fermentation form producing turbidity
in beer. Surface cultures on yeast-water gelatin have woolly margins.

ELLIPSOIDEUS GROUP. — These are wild yeasts. The cells are
usually ovoid, or pear-shaped, sometimes round, rarely elongated.

Five or six ascospores are produced in a cell.

8. ellipsoideus I. — A bottom fermentation yeast occurring on
ripe grapes.

S. ellipsoideus II. — A bottom fermentation yeast causing turbidity
in beer.

Both the pastorianus and ellipsoideus groups resemble the cerevisice
group in their chemical actions, but they are able in addition to
decompose malto-dextrin.

8. anomalus is a yeast forming small ovoid cells. It is curious
in that the spores are hemispheres with a projecting rim at the base
like a bowler hat.

468 A MANUAL OF BACTERIOLOGY

Another point in the identification of species of yeasts is the
period of formation of films. If the yeast is grown in wort with
free access of air and is undisturbed, e.g. in a beaker capped with
filter-paper, after a varying period a film composed of a zooglosal
mass of cells appears on the surface.

If yeast, or disintegrated yeast-cells, be injected into animals,
the blood acquires specific agglutinative properties, agglutinating
the yeast-cells of the species with which the inoculation has been
carried out.1

On the yeasts of fermentation, see Jorgensen, Micro-organisms
and Fermentation, 4th ed., 1911 (C. Griffin and Co.), (full bibliog.)
Klocker, Fermentation Organisms.

Examination of Yeasts

The yeasts can be readily examined in the fresh state in hanging-
drop preparations. The cells should be young or they will not be
of the typical form ; a two or three days' old culture in wort or
grape-sugar solution may be used. The yeasts grow well at 20°-
30° C. on the ordinary gelatin, agar, and potato, but wort gelatin
or wort agar is to be preferred. The elongated cells, common to all
old cultures of yeasts, may be obtained from the films which form
on wort cultures in wide flasks or beakers after two or three weeks.

In order to stain yeasts, a dilution of the culture should be made
in a watch-glass of water, so that the cells may be isolated, as they
become distorted if groups form in the preparations.

If the yeast has been grown in wort, it is best, before staining,
to pour off the fluid from the deposit of cells at the bottom of the
flask or test-tube, add some physiological salt solution and shake,
then allow the vessel to stand for an hour for the cells to sediment,
or centrifuge, and the process of washing may be repeated once.
Films may be prepared in the ordinary way and stained for five
minutes in Loffler's methylene blue, washed in water, dried, and
mounted. Or the films, after air-drying, may be fixed by immersion
in equal parts of alcohol and ether for ten minutes, dried in the air,
and stained as before. The preparations can also be stained in
gentian violet or fuchsin, or by Gram's method.

Ascospores may be double stained by preparing films of a sporing
culture in the ordinary way, staining with carbol-fuchsin for two
minutes, rinsing in water, decolorising with 5 per cent, sulphuric
acid and with alcohol, rinsing in water, counter -staining with
Loffler's blue for five minutes, washing, drying, and mounting.
The spores are red, the remainder of the cells blue.

1 See Macfadyen, Centr. f Bakt. (lte Abt.); xxx, 1901, p. 368.

CHAPTER XVII
THE HYPHOMYCETES— ASPERGILLOSIS— RINGWORM

The Hyphomycetes

THE moulds are, for convenience, collectively termed the Hypho-
mycetes, but this is not a strict botanical group. They are Fungi
having as a common character a plant body made up of hyphse.
They are multlccllular individuals, composed of filaments, simple
or branched, jointed or unjointed, which are termed hyphce, and are
formed by the end-to-end union of elongated cells. When the
hyphaa project upwards into the air they are known as aerial
hypha3, and when downwards into the fluid or medium on which
the organism is growing as submerged hyphse, and the compact
tufts or masses resulting from interlacing hyphae are termed mycelia.
A mycelium may form a hard lignified mass or pseudo-parenchyma,
which is known as a sclerotium, such as is met with in ergot and in
the black variety of mycetoma.

Any piece of the mycelium will grow, but in addition moulds
reproduce by multiple spores, which may be asexual or sexual.
Practically all moulds produce asexually formed spores ; some pro-
duce sexually formed spores by the fusion of two cells or gametes.
The two principal sexually formed spores are zygospores and asco-
spores. Zygospores occur in Mucor (see p. 470). In ascospore
formation, after conjugation of the gametes, instead of immediately
developing into a spore, the fertilised cell grows into a mass of branch-
ing hyphse, some of the cells of which produce spore sacs or asci, each
of which contains two or more ascospores (see Penicillium, p. 471).

Asexual spores are either free, borne at the ends or sides of hyphae
— conidia — as in Penicillium, or are formed in specialised spore
cases — sporangia — as in Mucor.

Usually the spore-bearing hyphse are specially differentiated, and
one bearing conidia is known as a conidiophore, one bearing a
sporangium as a sporangiophore. Some moulds produce spores by
segmentation of hyphee, these conidia being known as oidiat

469

470 A MANUAL OF BACTERIOLOGY

The Fungi are divided into the Phycomycetes, Ascomycetes,
Basidiomycetes, and Fungi Imperfecti. The Phycomycetes are
distinguished by non-septate or slightly septate hyphae and zygo-
spore-formation, as in the Mucors. The Ascomycetes are charac-
terised by the development of the cell resulting from fertilisation
into cells, some of which become spore sacs or asci containing several
spores. Asexual spores are usually produced as well. The Basidio-
mycetes have spore-bearing structures known as basidia ; the rusts,
smuts, toadstools, puff-balls, and mushrooms belong to this group.
All fungi which do not fall into one of these three groups are placed
among the Fungi Imperfecti ; most of them probably belong to the
Ascomycetes. Mucor muiedo, Penicillium glaucum, and Asper-
gillus niger may be taken as types and more fully described.

Mucor mucedo

The Mucoracice belong to the Phycomycetes, and are divided into
some eighteen genera.

Mucor mucedo, the common white mould which appears like
tufts of cotton-wool on various substances, may be obtained by
exposing some moistened bread or horse-dung to the air for a short
time, and then keeping it moist under a bell- jar. It consists of a
mycelium composed of hyphae, and its fluffy appearance is caused
by aerial hyphas. The aerial hyphae are at first of even diameter
throughout, but later on their free ends become swollen and
ultimately form spherical bodies, which become filled with spores,
the sporangia. In the early stage the whole organism forms but
a single cell, the protoplasm of which is granular and contains
vacuoles and numerous small nuclei. As it grows, and the sporangia
form, these become separated by a septum from the hyphae, and
when it becomes older stil] the mycelial hyphae may be divided into
elongated cells. The development of a sporangium takes place as
follows : The distal end of an aerial hypha swells, and immediately
below the swollen part a division occurs in the protoplasm and a
cellulose septum is formed, so that the swollen part is separated
off from the rest of the hypha, forming the rudimentary sporangium.
The sporangium continues to grow, and its protoplasm undergoes
multiple fission into numerous ovoid masses, the spores, each of
which becomes surrounded with a cellulose capsule. The septum
separating the sporangium from the hypha projects upwards into
the interior of the sporangium as a club-shaped knob known as the
columella. When the sporangium is ripe the slightest touch causes
its wall to rupture, so liberating the spores. When placed under

PENICILLIUM 471

favourable conditions the spore germinates, and the buds increase
in length and ultimately form hyphse.

Occasionally a process of conjugation occurs. Two adjacent
hyphse send out lateral branches which come in contact with one
another, and a septum forms in each, separating a small portion of
protoplasm from the rest of the hypha. The apposed walls of the
two cells become absorbed and the contents mingle. The mass of
protoplasm so formed becomes surrounded with a thick cell-wall,
giving rise to an inactive spore-like body, the zygospore, which
under favourable conditions develops like an ordinary spore. Some
Mucors form thick-walled resting cells, known as chlamydospores,
in the vegetative mycelium.

Certain Mucors form appreciable amounts of alcohol from carbo-
hydrates, and M. rouxii has been used for the commercial production
of alcohol.

Penicillium glaucum

Penicillium belongs to the Ascomycetes, and bears conidiophores.
Penicillium glaucum forms the bluish-green mouldy patches familiar
to every one. It is by far the commonest of all species, and may
be obtained from moist bread or jam or by exposing a gelatin plate
to the air for a short time. If the mouldy patch be rubbed a fine
greenish dust comes away. This dust consists of myriads of spores ;
if a little of it be transferred with a moistened needle to a gelatin
plate, or, better still, to a hanging-drop preparation, the growth of
the organism can be studied. After two or three days little white
specks will be observed, which microscopically are found to consist
of tufts of delicate interlacing hyphse ; these, becoming interwoven,
ultimately form a tough mycelium. The patches of growth are
circular, and the hyphse will be found to radiate from the centre.
As the patch increases in size it changes in colour, becoming bluish-
green, though the margin for some time still remains white. From
the upper surface of the mycelium delicate aerial hyphse grow
upwards, and from the under surface short submerged ones project
downwards.

The hyphse are composed of elongated cells arranged end to end,
the cell-walls of which consist of cellulose enclosing a more or less
vacuolated protoplasm containing several nuclei.

The aerial hyphse are unbranched filaments, but as development
proceeds the distal ends branch dichotomously, the branches
remaining short and nearly parallel to one another, so that a kind
of brush is produced. The ultimate branches are known as sterig-
mata. The ends of the sterigmata become constricted so that little

472 A MANUAL OF BACTERIOLOGY

globular masses, the spores, are formed ; this process is repeated
until a chain of spores results, the proximal one being the youngest.
A spore when placed under favourable conditions germinates, a little
bud appearing, elongating, and forming a hypha, just as in Mucor.
Brefeld, by sowing spores on moist bread, inverting the bread,
and examining at intervals, observed a sexual method of repro-
duction in Penicillium. Two sets of spiral cells develop on a thick
hypha, they intertwine, their contents probably mingle, and from
the union or carpogonium a tube-like hypha develops, which
becomes surrounded and enclosed by branching hyphse from the
mother cell. By further development and thickening of the cell-
walls a sclerotium forms ; it is a hard solid body, yellowish in colour,
and resembles a grain of sand, the carpogonium being at the centre.
If placed in favourable conditions the sclerotia germinate after some
time. Two forms of hyphse are produced, one thick, the other
thin ; the latter become much twisted. The thick hyphce become
branched, and ultimately a number of pear-shaped bodies are pro-
duced. The contents of these bodies then become broken up and
form spores ; the bodies are known as asci and the spores as asco-
spores. From the ascospores the ordinary mycelial form again
develops.1

Aspergillus niger

Aspergillus also belongs to the Ascomycetes, and representatives
of this genus are common on damp and decaying vegetable matter.
The asci occur as golden-yellow bodies in the mycelium. It forms
conidiophores which are unbranched and are swollen at the tip.
Short unbranched stalks (sterigmata) grow on this swelling and on
the tips ot these the spores develop. A process of sexual reproduc-
tion occurs very like the one observed in Penicillium. Aspergillus
niger grows well on the ordinary laboratory media, producing on
potato a powdery, sooty growth after a time. Aspergillus gla.ucus
is a common green-spored species.

With the exception of the ringworm and allied fungi,
which produce parasitic skin affections, the Hyphomycetes
are not of very great pathological importance. In the
ear and nose mucors and aspergilli may be met with, but
in these situations they are epiphytes rather than parasites,
and the same species occur in bronchiectases and pulmonary

1 See Brefeld, Quart. Journ. Microscop. Soc., vol. xv, p. 342.

SPOROTRICHOSIS 473

vomicge. Occasionally, however, a pneumono-mycosis
has been met with, the mycelium of the fungus ramifying
in the lung tissue and setting up irritative and other
changes. " Pneumono-mycosis " or " pulmonary asper-
gillosis " is especially a trade disease among bird-rearers.
Grain is taken into the mouth and the bird is fed with it,
and in the course of this operation the mould spores are
inhaled. The course of the disease is much like chronic
bronchitis or pulmonary tuberculosis. The species met
with in this condition seems generally to have been the
Aspergillus fumigatus.

The black variety of madura disease, as already stated
(p. 459), is due to a fungus form, and varieties of mycetoma
may be caused by fungi belonging to Aspergillus.

Sporotrichosis l

A rare disease clinically resembling syphilis or tuber-
culosis, characterised by indurated granulomata like
gummata, which subsequently break down, suppurate and
ulcerate. Potassium iodide has a curative action on the
condition.

In the pus of the lesions large ovoid refractile bodies
suggestive of yeasts or of large spores may be detected,
but no mycelium.

Cultures are best obtained on maltose agar (p. 477)
from non-ulcerated lesions ; agar and potato may also
yield growths. The organism (Sporotrichon Beurmanni)
grows as small raised woolly colonies,, at first white, after-
wards becoming brown. The growths consist of a felted
mycelium of filaments with spores and yeast-like cells.
It produces granulomata in inoculated mice. The botanical
position of the organism is uncertain ; by some it is regarded

1 See Walker and Ritchie, Brit. Med. Journ., 1911, vol. ii, p. 1 ;
Gougerot, Journ. of State Med., xxi, 1913, p. 614 et seq.

474 A MANUAL OF BACTERIOLOGY

as a true fungus. It is stated to occur on decaying vege-
table matter and to be the cause of epizootic lymphangitis
in the horse — a disease having a superficial resemblance to
farcy — in the pus of which oat-shaped bodies are found,
the " cryptococcus " of Rivolta.

Thrush

Thrush is due to an organism (O'idium or Monilia albi-
cans) which is usually classed among the Hyphomycetes.
It forms the whitish patches so frequently seen on the
mucous membrane of the mouth and pharynx in children
and in those suffering from wasting diseases but a general
infection has occasionally been produced by it. If one
of these patches is removed and teased up, it will be found
to consist of masses of tangled mycelial threads with yeast-
like budding. The organism can be readily cultivated on
all the ordinary laboratory media, and will also grow on
slightly acid media such as wort gelatin. It produces
whitish, membranous, adherent growths, in which it
appears morphologically under two forms — as masses of
tangled filaments or hyphae and as yeast-like cells. On
aoid media the latter exclusively occur, on alkaline the
former predominate. It liquefies gelatin, stains by Gram's
method, produces an alkaline reaction by the formation
of ammonium carbonate, and does not ferment lactose.
Inoculated on to a damaged mucous membrane the
" thrush " patches appear, subcutaneously it produces an
abscess, and injected into the peritoneum a general infec-
tion, followed by death and accompanied by a sero-purulent
peritonitis.

Cultivation and Examination

The Hyphomycetes can be cultivated on the ordinary laboratory
media, but wort-agar, or wort -gelatin, potato, bread, or maltose agar
is to be preferred.

RINGWORM 475

They can be examined by removing a portion of the growth,
teasing up gently with needles in a little 50 per cent, alcohol con-
taining a trace of ammonia, removing the surplus fluid with blotting-
paper, and mounting in Farrant's solution or in glycerine jelly.*
If desired, they may be stained by the irrigation method with f uchsin.
Thrush may be examined in this way.

In the tissues they may be stained with hsematoxylin or methylene
blue, or by Gram's or by Weigert's method.

Ringworm

The ringworm fungi must probably be included in the
group of the Hyphomycetes. Human ringworm, formerly
regarded as a single disease, has been proved to comprise
at least two affections through the researches of Sabouraud.
These two forms are distinguished from each other clini-
cally and by differences in the parasitic organisms.

The first variety is an affection of early childhood,
forming 80 to 90 per cent, of the ringworms met with in
London ; it never attacks the scalp of adults, never affects
the beard or nails, is very intractable, and frequently
epidemic. The parasite is characterised by small round
or ovoid spores measuring 3 /x to 4 /x in diameter. Affected
hairs are generally broken off, forming relatively long
stumps, greyish in colour, and possessing a whitish sheath.
When suitably prepared in potash this sheath is seen to
be composed of the spores agglomerated together without
apparent order, and the hairs themselves are filled with
delicate parallel mycelial threads (Fig. 52). The fungus is
named the Microsporon Audouini.

The second variety comprises the ringworms with large
spores, and is divided into two groups by Sabouraud. The
first of these groups is exclusively of human origin, and
has a marked tendency to affect the interior of the hairs
only, and hence the organism has been termed the Tricho-
phyton megalosporon endothrix. The other group is of
animal origin, and the spores are met with chiefly on the

476

A MANUAL OF BACTERIOLOGY

outside of the hairs, and the fungus is hence termed the
Trichopliyton megalosporon ectothrix.

9 The endothrix form occurs later in childhood, is not so
persistent as the Microsporon, and does not attack the
nails or beard. Microscopically, the fungus is seen to
consist of beaded threads, which are rounded or ovoid
spores arranged end to end. The ectothrix form rarely

FIG. 52. — Ringworm in a hair. X 350.

attacks the scalp, but is responsible for all the tinea sycosis
and ringworm of the nails and half the cases of tinea
circinata. Suppuration is common in this form. Micro-
scopically appearances differ ; generally the spores are
arranged in chains, but the sporulation is less regular than
in the endothrix. The spores in the endothrix and ectothrix
varieties measure 4 /m to 12 JUL in diameter.

The ringworm fungi can be readily cultivated on all
the ordinary media — beer- wort agar and beer- wort gelatin
being especially favourable. They form whitish fluffy
growths with rapid liquefaction of gelatin. In order to
obtain cultivations the diseased hairs or stumps are

RINGWORM 477

removed by forceps and placed on a sterile glass slide.
The aerial portion of the hair is then cut away by means
of a sterile scalpel, and the diseased portion is divided into
small fragments. These can be picked up with a moistened
platinum needle and transferred to the culture media,
preferably beer-wort agar. In some cases a pure culture
is thus obtained, but in others further treatment is neces-
sary. When the Trichophyton or Microsporon has thrown
up its aerial hyphaa the plug of wool is removed from the
tube and the mouth well flamed ; the tube is then held
inverted over a Petri dish containing solidified maltose
agar. A sharp tap or two is given to the tube, sufficient
to cause the spores to drop, and the dish is re-covered.
A growth of the organism from single isolated spores thus
ensues, and pure cultures can be obtained (Blaxall).

The various forms of the ringworm fungi can be differen-
tiated by cultures, but it is necessary when comparing them
to employ media of identical composition, because slight
differences in the latter are liable to induce marked changes
in the characters of the cultures. A favourite medium,
used by Sabouraud and by Blaxall, is maltose agar :

Peptone 0*5 grm.

Maltose 3-8 grm.

Agar-agar 1-3 grm.

Water 100 c.c.

Blaxall found that different samples of maltose materially
influenced the characters of the cultures.

Characters of the cultures. — Cultures are incubated at
30° C. The colonies of the Microsporon do not show any
growth until about the seventh day ; little white downy
tufts then appear. The fully developed growth on maltose
agar forms a large white downy patch with a small central
boss ; on potato white downy patches appear with brown
discoloration.

478 A MANUAL OF BACTERIOLOGY

The endothrix variety commences to grow in six or
seven days, and on maltose agar in about a month forms
a rounded patch with a central crateriform depression,
the whole being dusted with fine white powder (Fig. 53) ;
on potato, powdery stars develop tinged with yellow and
usually without discoloration of the medium.

The cultures of the ectothrix form are variable. They
commence on the third or fourth day ; some develop

FIG. 53. — Culture of the ringworm organism. Endothrix form.

whitish smooth, or wrinkled growths ; others, from the
dog, form dry, brown, wrinkled, powdery growths ; others,
of bird origin, form purplish growths.

Microscopically, all the fungi show masses of mycelial
threads with spores. They stain with the ordinary anilin
dyes and also by Gram's method, and can be mounted in
glycerin jelly in the manner described at p. 475.

Macfadyen found that the ringworm organism produces
an active peptonising enzyme, and seems to increase the
solubility of keratin when grown on it ; no inverting
enzyme could be isolated.

RINGWORM 479

Clinical Examination

The hairs should be treated first with ether and then with caustic
potash solution of about 7 per cent, strength. In this reagent they
may remain for from a few hours to a few days ; they are then
floated on to a slide and carefully covered with a cover-glass.
Permanent preparations may be mounted in Farrant's solution or
in glycerine jelly.

Hairs, after treatment with ether, may be stained by the following
method :

(1 ) Stain in anilin-gentian violet for one to two minutes, and blot,

(2) Treat with Gram's iodine solution for one to two minutes,
and blot.

(3) Decolorise carefully (watching microscopically) with anilin
oil containing 1 per cent, of hydrochloric acid.

(4) Treat with anilin oil and then with anilin oil and xylol.

(5) Clear in xylol, and mount in Canada balsam.
ERYTHRASMA. — Due to infection with a fungus (Microsporon

minutissimum), very difficult to cultivate, which occurs as extremely
long, fine filaments.

FAVUS. — Favus is due to a fungus discovered by Schoenlein in
1839 — the Achiorion Schoenleinii. It is seen as a mycelial growth
with spores in the patches. The organism grows well on maltose
agar, forming fluffy, woolly, moss-like colonies with radiating out-
growths, first grey and then yellowish. It occurs on mice and other
animals.

DHOBIE ITCH. — Caste! lani has isolated three trichophyton-like
organisms in this disease.

PITYRIASIS ALBA. — In this disease Unna's " bottle bacillus " is
invariably present. It occurs as large round or oval bodies like
yeast-cells, which may occasionally show budding.

PITYRIASIS VERSICOLOR. — In the epidermal scales of this skin
affection a fungoid organism (Microsporon furfur) is present. It
occurs as short and thick curved hyphse between which are masses
of large coarse spores. It has not been cultivated (or very rarely).

PINTA. — A skin disease met with in South America. In the scales
short mycelial filaments with large (8-12 ^u) spores are seen. Various
organisms have been cultivated belonging to the genera PeniciHium
and Aspergillus.

PIEDRA. — A disease of the hairs met with in South America.
The nodosities on the hairs are composed of masses of very large
refractile spores.

CHAPTER XVIII
THE PROTOZOA x

The General Structure of the Protozoa — Pathogenic Amoebae
— Trypanosomata — Leishman-Donovan Body — Spirochaetac —
Coccidia — Malaria

THE Protozoa are an important group of unicellular organisms,
regarded as animal in nature, and sharply and definitely distin-
guished from the rest of the animal kingdom, to which the names
of metazoa and enterozoa are applied. The latter consists of many
cells, differentiated to perform different functions, and arranged in
two layers — endoderm and ectoderm — around a central cavity, the
enteron.

" It is true that some protozoa consist of aggregates of cells, and
should therefore be entitled to be called multicellular ; yet an
examination of the details of structure of these cell -aggregates and
of their life-history establishes the fact that the cohesion of the cells
in these instances is not an essential feature of the life of such multi-
cellular protozoa, but a secondary and non-essential arrangement.
Like the budded ' persons ' forming, when coherent to each other,
undifferentiated ' colonies ' among the polyps and corals, the
coherent cells of a compound protozoon can be separated from one
another and live independently ; their cohesion has no economic
significance. Each cell is precisely the counterpart of its neighbour ;
there is no common life, no distribution of function among special
groups of the associated cells, and no corresponding differentiation
of structure. As a contrast to this, we find in the simplest enterozoa
that the cells are functionally and structurally distinguishable into
two groups — those which line the enteron or digestive cavity, and
those which form the outer body wall. The cells of these two layers

1 See Lankester's Treatise on Zoology, Part I, first and second
Fascicles, 1907 and 1909 ; Minchin in' Clifford Allbutt's System of
Medicine, ed. 2, vol. ii, pt. ii ; Hartog in Cambridge Natural History,
vol. i.

480

SARKODINA 481

are not interchangeable, but are fundamentally different in proper-
ties and structure " (Ray Lankester). It is true that in some
instances there may be a difficulty in deciding whether an organism
is vegetable or animal, and Haeckel proposed to include all
indeterminate unicellular organisms in a distinct kingdom, the
Protista.

The cytoplasm of a protozoon is commonly differentiated into
an outer, clearer, denser layer or ectosarc, and an inner, granular,
more fluid portion, the endosarc. The cytoplasm is sometimes
naked, or may be covered with a cuticle, usually protein in nature.
The cytoplasm contains a well -marked nucleus, sometimes a
secondary nucleus, and occasionally subsidiary chromatin particles
or chromidia. A contractile vacuole, which is an excretory organ,
is frequently present.

In most protozoa reproduction takes place by simple division or
fission, and by a process of spore-formation ; in others reproduction
is exclusively by spores, which are often formed by a complicated
process of development. In many of the protozoa a simple form
of sexual reproduction by conjugation occurs. Two dissimilar
cells (gametes) are produced, the larger comparable to female cells
or ova and termed macrogametes, the smaller comparable to male
elements or spermatozoa and termed microgametes. The cells
from which the gametes are derived are known as gametocytes. The
gametes conjugate and form a zygote, which usually divides into
a number of spores from which the adult is reproduced.

In certain cases sexually differentiated individuals reproduce
by fission without conjugation ; this phenomenon is termed parthe-
nogenesis.

Various classifications of the Protozoa have been suggested.
Biitschli divides them into four classes : I. The Sarkodina (p. 481) ;
II. The Mastigophora (p. 487) ; III. The Infusoria (p. 507) ; and
IV. The Sporozoa (p. 508).

Class I. — Sarkodina

There are Protozoa in which the cell protoplasm is naked, and
locomotion and ingestion of food are performed by means of tem-
porary protoplasmic processes or pseudopodia.

The Sarkodina includes a number of forms of very varied mor-
phology and habits, such as the Amoebae, Heliozoa, Radiolaria, and
Foraminifera, the three latter groups being characterised by the
presence of a siliceous or calcareous skeleton or shell.

31

482 A MANUAL OF BACTERIOLOGY

Pathogenic Amoebae l

Three species of Amcebce seem to be parasitic in man,
and the generic name of Entamceba has been given to
them. One, the E. buccalis, occurs in the mouth in dental
caries, the other two inhabit the intestine. One of the
latter, the Entamceba coli (Amoeba coli, Losch), occurs in
the upper part of the large intestine and appears to be

FIG. 54. — Amceba histolytica. (After Councilman and Lafleur.)

harmless ; the other, the Entamceba histolytica, is regarded
as the cause of amoebic or tropical dysentery.

The Entamceba histolytica is met with in the faeces in
these cases, and also in the pus of the so-called tropical
abscess of the liver. It is especially abundant in the
mucoid material during the acute stage. The E. histoly-
tica is a large protoplasmic mass measuring 25 to 35 /x. in
diameter, possessed of slow amoeboid movement, and
having a clearer outer zone or ectosarc and a granular
endosarc. The pseudopodia are always blunt, never

1 Councilman and Lafleur, Johns Hopkins Hosp. Reps., vol. ii, 1891 ;
Schaudinn, A.K. Gesundheitsamte, xix, p. 547 ; Strong, Musgrave,
Clegg, Thomas and Woolley, Bureau of Gov. Laboratories, Manila
Bulls. 18 and 32.

PATHOGENIC AMOEBA 483

pointed (Fig. 55). Tn the endosarc highly refractile granules
occur, and it often contains blood- corpuscles and a vacuole
(Fig. 54, 6). A nucleus can also be demonstrated, but
being poor in chromatin, it stains with difficulty (Fig. 54, a).
According to Schaudinn, the E. coli differs from the E.
histolytica in that the ectoplasm is not distinctly seen
except during the formation of a pseudopodium, and the

FIG. 55. — Changes in form of an Amosba histolytica observed on
a warm stage, and drawn at intervals of one minute. (Semi-
diagrammatic by the writer.)

nucleus stains deeply. The development of the two forma
is also different. E. coli multiplies by simple binary fission,
and also by multiple fission into eight small amoebae.
Encystment may also occur, with repeated binary division
of nucleus and protoplasm, part of the nucleus being cast
off and ultimately the cyst contains eight nuclei around
which the protoplasm collects, so that, if swallowed, eight
small amoebae are set free.

The E. histolytica multiplies by binary fission, and also
by irregular gemmation, so that an indefinite number of

484 A MANUAL OF BACTERIOLOGY

small amoebae is formed. Instead of encystment, as in
the E. coli, resistant spores are formed. The nucleus gives
off chromidia, some of which, together with portions of the
ectoplasm, are extruded and become spores surrounded
by tough capsules. Infection of a fresh host apparently
occurs only with material containing these spores.

The presence of the amoeba in the pus, and especially
in the walls, of tropical abscesses is of considerable diag-
nostic significance, and the parasite is considered to be
the etiological agent in amoebic or tropical dysentery
(see " Dysentery "). The amoebae are not usually observed
in the abscess pus at the time of operation, but make
their appearance in the discharge about the third day,
i.e. when the wall of the abscess- cavity is contracting.
In the true tropical abscess the ordinary pyogenic organ-
isms are absent, unless a secondary infection has occurred,
which is the exception. The abscess is usually single, and
Rogers suggests that the amoebae reach the liver through
adhesions between it and the bowel. The amoebae may
be cultivated on ordinary or on water agar provided some
bacterium is present at the same time, e.g. B. coli, cholera
vibrio, etc. Material rich in amoebae may be smeared
over agar plates, which are grown at 25°-30° C. for twenty-
four to forty- eight hours, and are then examined with a
low power. At any spot where isolated amoebae are
observed, with a little dexterity the organism may be
lifted up with a fine needle and transferred to a fresh
plate, and by a repetition of the process pure cultures
may be obtained. The cultivated amoebae are pathogenic
for monkeys, and induce abscess on inoculation into the
liver. Musgrave and Clegg (loc. cit.) are of opinion that
all amoebae are, or may become, pathogenic.

PATHOGENIC AMOEBA 485

Clinical Diagnosis

1. A drop of the dysenteric discharge (the mucous portions
should be chosen from the stools), pus, or, better, a scraping from
the wall of the abscess, diluted, if necessary, with a little warm
(37° C.) physiological salt solution, is placed on a slide, covered
with a cover-glass, and examined microscopically with a j- or J-inch
objective. The amoebae will be readily recognised, and may be
examined more critically with a ^L-irich oil-immersion. To be
certain that the bodies are amoebae, the amoeboid movements must
be observed by keeping the preparation on a warm stage.

The stools should be fresh, unmixed with urine, collected in a
warmed bed-pan, and kept at blood-heat until examined, which
should be done as soon as possible.

2. The living amoebae in the stools may be stained by the irriga-
tion method with a weak (|-1 per cent.) aqueous solution of neutral
red. Preparations may also be stained by irrigation with methyl-
ene-blue and Beale's carmine ; the latter stains the nucleus, the
former does not. The preparation may be rendered permanent
by washing away the excess of stain, and running in some 50 per cent,
glycerin by irrigation.

3. Probably Heidenhain's iron-haematoxylin method is the best
for staining this and other protozoa :

(a) Make smears of the material and drop while wet into the
fixative — two parts of saturated aqueous mercuric chloride solution,
one part of alcohol, with a few drops of glacial acetic. They remain
in this for ten minutes.

(6) Wash in weak spirit and then in weak spirit coloured with
iodine, and finally wash in distilled water.

(c) Treat with 4 per cent, iron-alum solution for six to ten hours.

(d) Stain in Heidenhain's hsematoxylin for at least six hours.

(e) Differentiate in 1 per cent, iron-alum, watching microscopi-
cally.

(/) Wash well in tap-water, pass through alcohol and xylol, and
mount.

4. Twort's stain may be used for sections. . The stain (which is
a compound neutral red and light green preparation) is best made
by rubbing up 0-25 grm. of the stain (Griibler's) with some clean
sharp sand in a mortar ; this prevents the stain going into a sticky
mass when the alcohol is added. To the powder so obtained is
now added some purest methyl alcohol (Merck's), acetone-free,
containing 5 per cent, by volume of glycerin. Rub up well to

486 A MANUAL OF BACTERIOLOGY

obtain a saturated solution ; then pour off and add a further
quantity of alcohol -glycerin solution, and repeat the trituration ;
about 100 c.c. stain can be made from the quantity given.

The solution, when filtered, should be kept in a well-stoppered
bottle (and if a completely saturated solution has been obtained,
add 10 per cent, more alcohol-glycerin mixture). The stain may
be purchased ready for use.

Tissues to be examined should be fixed in Miiller's fluid con-
taining 10 per cent, of formalin, but on no account should 10 per
cent, formalin alone be used.

Paraffin sections (after xylol, alcohol and distilled water) are
stained for about five minutes with the stain made up by mixing
one part of distilled water with two parts of the glycerin-alcohol
stain solution. Sometimes in staining such organisms as glanders
ten minutes may be necessary, especially if insufficient stain is in
solution and the room temperature is low. Rinse in distilled
water.

Fix for half to one minute in Unna's glycerin-ether mixture —
2 per cent, in distilled water. Rinse in distilled water.

Differentiate and dehydrate in absolute alcohol. Should there
be much precipitate, this can easily be removed by a few drops
of methyl alcohol, or better by a mixture of equal parts of absolute
alcohol and xylol. Pass through xylol and mount.

Various elements stain different colours, viz. chromatin of nuclei,
purple red ; mucoid and colloid degenerations, bright orange red ;
fetal cartilage, orange red ; fibrous tissue, blue-green ; erythrocytes,
light grass -green. Micro-organisms stain bright red and stand out
in marked contrast to the green connective tissue containing
them.

Animal parasites, e.g. amcebaa, also stain well. The stain has
the advantage of leaving all the tissues sharply differentiated.

Allusion may here be made to the Mycetozoa (Myxomycetes).
These are masses of protoplasm resembling huge amoebae, which
are found on decaying vegetable matter. By some they are regarded
as vegetable, by others as animal, in nature, and belonging to the
Amoebae of the Sarkodina.1 Some important plant diseases, such
as the " finger-and-toe " of cabbage roots, are due to their activity.
The finger-and-toe disease is due to an amoaboid parasite (Plasmo-
diophora brassicce, by some included among the Amcebce), the cycle
of which begins with spores from which small flagellulae are set
free. Similar organisms have been supposed to be present in cancer.

1 See Lankester's Treatise en Zoology, Pt. 1, First Fascicle, p. 37.

MASTIGOPHORA 487

Class II. — Mastigophora

These are protozoa in which one or more permanent organs
serving for locomotion or food capture are present in the form of
flagella. As a rule the body is limited by either a cuticle or a
differentiation of the protoplasm into a firmer external portion or
periplast. One, two, or more flagella may be present, and when
multiple are arranged in various ways. Food-vacuoles may occur
in the protoplasm, also contractile vacuoles, but not in the parasitic
forms. Various other granules, including chromatophores, which
generally contain chlorophyl, may be present. The nuclear appa-
ratus is usually double, consisting of a large principal or macro -
nucleus, and a small or micronucleus or blepharoplast ; the latter
is not, as in the Infusoria, composed of generative chromatin, and
is in relation with the locomotor apparatus. An undulating mem-
brane, a thin protoplasmic membrane attached to one aspect of
the body like a dorsal fin, may be present. Euglena is a common
form in ditches, and Noctiluca is the chief cause of phosphorescence
in the sea ; both are uniflagellate. Volvox and Protococcus are
also placed by some in this group. The chief parasitic genera
are :

Trypanosoma and Trypanoplasma, both of which have an undu-
lating membrane, but the former has one flagellum, the latter two
flagella, one at each end of the body, but both starting from the
blepharoplast. Spirochaeta (see p. 493).

Herpetomonas, like Trypanosoma, has a single flagellum, but no
undulating membrane.

Crithidia has a pear-shaped body with single flagellum.

Trichomonas, also somewhat pear-shaped, with three short
flagella and an undulating membrane.

The trypanosomes and other forms living in the blood are known
as haemoflagellates.

Trypanosomata *

The trypanosomes are all parasitic in the blood of vertebrates,
and a blood-sucking invertebrate is almost invariably concerned
in their transmission. In the case of each pathogenic trypanosome,
some indigenous wild animal, tolerant to that form, serves as a
reservoir from which infection is derived.

1 For current literature on Trypanosomes and trypanosome diseases
see The Tropical Diseases Bulletin.

488 A MANUAL OF BACTERIOLOGY

A trypanosome has a slender, flexible, flattened body, one
extremity of which is pointed, the other passes into a single flagellum.
A delicate undulating membrane passes along one edge of the body.
The organism lives in the plasma, in which it is actively motile, the
flagellated end being usually anterior, and measures 15-30 /*, or
even 40-50 /*, in length. The protoplasm of the organism is finely
granular, and near the centre of the body is a large macronucleus,
and generally between it and the non-flagellated end is a smaller
micronucleus or blepharoplast. From the latter a chromatin
filament starts, runs along the free edge of the undulating membrane
and passes into the flagellum. Reproduction takes place by longi-
tudinal division, occasionally probably by transverse division, and
amoeboid and plasmodial masses may be found in the internal
organs and bone -marrow. The trypanosomes have great mor-
phological similarity, which renders them practically indistinguish-
able by structural characters. They can usually be differentiated
into three forms — indifferent, male, and female — which in some
cases may all occur together, but only become fully differentiated
in an invertebrate host. The males are slender, active, only slightly
granular, and with an elongated nucleus ; the females are bulky,
sluggish, granular, and have a rounded nucleus ; the indifferent
forms are intermediate. The males usually soon die off unless they
conjugate ; the indifferents are more hardy, the females most so.
The sexual forms conjugate in an invertebrate host, but if the males
have died off, both male and female forms may be reproduced from
the females by a process of parthenogenesis.

Trypanosoma Gambiense

In human trypanosomiasis and sleeping-sickness of
West and Central Africa, a trypanosome Tr. Gambiense
is the causative agent (Plate XXI. a). It is usually
present, though scanty, in the blood, but can often be
found in numbers in the fluid aspirated from the enlarged
cervical glands. In the later stages, when cerebral
symptoms ensue, it is found in the cerebro-spinal fluid,
but scantily, centrifuging being necessary in order to
demonstrate the parasites. The Tr. Gambiense is patho-
genic to monkeys, and to a less extent to white rats and
guinea-pigs. Cattle and certain antelopes and other wild

TRYPANOSOME GAMBIENSE 489

game may act as reservoirs of the parasite, and it has
been seriously suggested to kill off all the big game in
the affected areas. It is conveyed by a tsetse-fly (G.
palpalis), possibly by other tsetses.

The tsetse (and possibly other biting flies) may rarely convey
the disease by direct inoculation. Generally a cycle of development
is passed in the tsetse. The stages of this are not known with
certainty, but Roubaud has observed multiplication of the parasites
in the fly and the development of Herpetomonas forms. According
to the observations of Kleine and Bruce, the flies become infective
about thirty-four days after feeding and remain infective for at
least 70-80 days, and probably for the rest of their lives.

In Rhodesia, a human trypanosome (Tr. Khodesiense) has been
found which is probably distinct from Tr. gambiense, and the
O. palpalis does not occur in the district. The macronucleus of the
parasite is situated between the blepharoplast and the posterior end.

In Brazil another human trypanosome-like parasite has been
discovered by Chagas (Tr. or Schizotrypanum cruzi), which is
conveyed by a bug (Conorhinus megistus).

Tr. Brucei is the causative parasite of nagana or tsetse-
fly disease of horses in Africa.

Nagana is met with in large tracts of country in Zululand
and West Africa. It especially attacks the equines —
horse, mule, and ass — in which it is very fatal. The
animals become anaemic and emaciated, there is a discharge
from the eyes and nose, staring coat, swelling of the legs
and neck, and fever. The animal dies two to six weeks
after infection. Oxen are also attacked, but a small
proportion recover. The dog, cat, rabbit, guinea-pig,
mouse, and rat may be infected by inoculation with the
fresh blood of a diseased animal. In infected animals
the trypanosome is generally abundant in the blood and
spleen. The Tr. Brucei can be cultivated, though with
difficulty, on rabbit-blood agar — melted sterile agar cooled
to 45° C. -f sterile defibrinated rabbit's blood warmed to
45° C., mixed and allowed to solidify in the sloping position
(Novy and McNeal). The disease is conveyed through the

490 A MANUAL OF BACTERIOLOGY

bites of a tsetse-fly (Glossina morsitans). The trypanosome
is believed to live in the big game, from whence it is trans-
mitted to horses entering the infected localities. The
blood loses its infective properties usually within twenty-
four hours after being withdrawn.

Surra attacks horses in Burma, Mauritius, and the
Philippines, and is pathogenic to the same animals as
nagana, and in the blood a parasite (Tr. Evansi) similar
to that in nagana. but more active, was observed by
Evans. Surra is probably spread by certain biting flies
belonging to the Tabanidce.

The tsetse flies (Glossina) belong to the house-fly order (Muscidse)
and have a general resemblance to a house-fly, but when at rest
the wings fold completely over each other. The proboscis is long
and straight and the wing venation is characteristic, especially the
fourth longitudinal vein, which makes two bends. Instead of laying
eggs, the female extrudes a single full-grown larva. They are
confined to Africa and Arabia ; some sixteen species have been
differentiated, and they occur in the vicinity of water on the edge
of forest land (" fly-belts ").

Tr. equinum attacks horses in South America, causing weakness
and paresis of the hindquarters ("ma/ de caderas"). Cattle are
immune, most other animals susceptible.

Tr. Theileri, the largest trypanosome known (50-60 p in length),
is found in cattle in ISouth Africa, and is not pathogenic to any
other animal.

Tr. dimorphum occurs in two forms, large and small, in horses
in Africa. Is pathogenic to most animals.

Dourine, a venereal disease of the horse met with in North Africa,
Spain, and Hungary, is due to the Tr. equiperdum, which is conveyed
by direct contact, and is mainly confined to the lesions, being
scanty in the blood. It is pathogenic to the ordinary laboratory
animals.

In rats a non-pathogenic trypanosome was found by Lewis
(Tr. Lewisi). It is especially met with in sewer-rats, but also
occurs in field-rats (Crookshank). It is somewhat shorter and
thinner than the Tr. Brucei, and there are other small differences
between the two forms. With the exception of rats and mice,
and to a less Extent guinea-pigs, other animals cannot be infected
with the Tr. Lewisi. It may be kept alive for long periods in the

LEISHMANIOSIS 491

blood placed in a refrigerator, whereas the Tr. Brucei soon dies
under the same conditions. The two forms do not protect against
each other. The Tr. Lewisi is readily cultivated on rabbit-blood
agar and is transmitted by the rat-flea, in which it seems to pene-
trate into the epithelial cells of the gut and there undergoes a process
of multiplication.1 It is passed in the faeces of the flea and a rat
ingesting the infected faeces becomes infected.

A number of other trypanosomes have been found in the lower
animals, birds, fish, reptiles, and amphibians. A large and charac-
teristic one is generally present in the blood of the eel.

The trypanosomes are usually agglutinated when mixed with
the serum from an infected animal.

Hewlett was unable to obtain any toxic or immunising substance
from ground-up trypanosomes (Tr. Brucei).2

Levaditi and Twort 3 have found that the filtrate of broth cultures
of B. subtilis is markedly trypanocidal in vitro but not in vivo.

Examination of Trypanosomes, etc.

The trypanosomes, if numerous, are readily observed in the
fresh blood. A very shallow cell may be formed on a slide by
ringing with melted paraffin. For stained preparations theLeish-
man stain (see " Malaria ") or the Heidenhain method (p. 485)
may be employed.4

Leishmaniosis

This term is applied to a group of diseases, caused by
a similar parasite, and widely distributed in tropical
and sub-tropical countries of the old and new world.5

In kala-azar or tropical splenomegaly, a disease met
with in India, Assam and the East, a small parasite, the
Leishman-Donovan body, occurs in large numbers in the
spleen and liver, also in the lymphatic glands, lungs, and
intestinal submucosa, and in large rnononuclear leucocytes

1 Minchin and Thompson, Brit. Med. Journ., 1911, vol. ii, p. 361.

2 Proc. Roy. Soc. Lond., B., vol. Ixxxiv, 1911, p. 56.

3 Ccmp. Rend. Soc. Biol, vols. Ixx and Ixxi, 1911.

4 For a special method of staining, see Plimmer, Proc. Roy. $oc..
Lond., B. vol. Ixxix, 1907, p. 102.

5 See Hewlett, Practitioner, 1911, July, p. 109.

492 A MANUAL OF BACTERIOLOGY

and endothelial cells. The bodies are small (2-3 //), round,
ovoid, or oat-shaped masses of protoplasm, apparently
encapsuled, and contain two chromatin masses, one large
and oval, staining pale red with Leishman's stain, the
other small and rod-shaped, and staining deep red with
Leishman (Fig. 56, a). They sometimes occur in masses
(Fig. 56, c). Leishman considered the bodies to be degene-
rate trypanosomes, but the organism is now considered

FIG. 56. — a. The Leishman-Donovan body. b. The flagellated
form developing in citrated blood, c. Seven parasites in a
farge mononuclear leucocyte. (After James. Patton, and
Rogers.)

to belong to a distinct genus, and is termed Leishmania
Donovani. Rogers succeeded in cultivating it in citrated
blood at 20°-25° C., in which it develops into a flagellated
form like Herpetomonas (Fig. 56, 6).1 The parasite is not
inoculable into animals, and it is probably transmitted
to man by a bug (? a Conorhinus).

The bodies are well shown in smears stained with the
Leishman stain.

In Oriental sore, or Delhi boil, a parasite practically
identical with the Leishman-Donovan body is present,
but as the two diseases run a totally different course, it is

1 Brit. Med. Journ., 1907, vol. i, p. 427 et seg.

SPIROCHAETOSIS 493

probably a distinct species (L. tropica). On cultivation
it develops a flagellated form. The disease has a seasonal
prevalence, and Wenyon suggests that it is conveyed by
a mosquito, a species of Stegomyia.

In N. Africa Nicolle has observed a Leishmaniosis of
children due to another species (L. infantum). It is trans-
missible to the dog and monkey, and can be cultivated.
The disease has recently been found all along the Mediter-
ranean littoral.

Spirochaetosis l

Diseases caused by infection with spirochaetes. — The spirochaetae
are delicate, undulating, or somewhat spirillar, filiform parasites
occurring in the blood of man, mammals, birds, shell-fish, etc. The
filaments taper to a point at the ends, are flexible and motile,
coiling and uncoiling, are described as having two nuclear masses,
and some possess an undulating membrane, like trypanosomes,
but in the smaller forms no definite structure can be made out.
They are now generally regarded as protozoa, but some still con-
sider them to be bacteria. Bacterial cells are never pointed, nor
do they show the coiling movements of spirochaetes ; motility is
produced by flagella, which are absent from most spirochaetes
(statements to the contrary are due to errors of observation and
technique), and periodicity is not exhibited by bacteria. Spiro-
chaetes are said to multiply by longitudinal fission, while fission
in bacteria is transverse (Dobell states that multiplication is always
by transverse, but multiple, fission. See p. 12) ; they react in
some cases to drugs (e.g. salvarsan) like trypanosomes, are much
more sensitive to the action of immune sera than bacteria are, and
are transmitted by insects. Noguchi has cultivated certain spiro-
chaetes of the mouth and relapsing fever by a method similar to
that which he employed for syphilis (p. 497). For the saprophytic
spirochaetes a small quantity of oxygen is required, for the blood
spirochaetes absolute anaerobiosis is necessary as in the case of
syphilis.

Schaudinn believed that many so-called spirochaetes may be
connected with the trypanosomes. In 8. plicatilis he described
the presence of a thread-like nucleus and of chromidia, and of
an undulating membrane, but flagella are absent. In the little

1 See Nuttall, Jaurn. Roy. Inst. Pub. Health, xvi, 1908, p. 449.

494 A MANUAL OF BACTERIOLOGY

owl minute slender trypanosomes occur ; these later penetrate
leucocytes, and develop into relatively very large trypanosome
forms (which have been termed Leucocytozoa). These intra-
oorpuscular forms are male and female gametocytes, the male being
smaller and more slender than the female. If taken into the gnat's
stomach, the male gametocytes give rise to eight microgametes by
a process of sporulation, which fertilise the macrogamete, and the
resulting zygote ultimately forms by sporulation an immense
number of spirochaetes.

In the case of a Halteridium parasite of the little owl (Athene
noctua), Schaudinn claimed to have shown that it is a stage of a
trypanosome (jP. noctuce) which is disseminated by the common gnat.
His observations have not been confirmed, and Novy and McNeal
believe that Schaudinn was dealing with a double infection of both
a trypanosome and a Halteridium, not that one was transformed
into the other.

Spirochaeta recurrentis (Obermeieri). — Found in the
blood-plasma, not in the corpuscles, in relapsing fever
during the febrile paroxysms. It is very slender and
delicate, measuring 12 to 16 //. in length, and actively
motile (Plate XXI. 6). Bugs were formerly supposed
to transmit this parasite, but Nicolle, Blaizot and Conseil
have established the body louse as the agent of trans-
mission. Infection is however not due to the bite of the
louse, but to lice being crushed by the victim's scratching
and the contents of the lice rubbed into the abrasions.
The lice not only retain the infection for the rest of their
lives, but the spirochaetes pass into their eggs, and these
eggs and the larvae hatched from them may similarly be
infective to man. It is inoculable into monkeys, and,
less readily, into rats.

Noguchi and Hata l have cultivated this form : the
latter in a medium consisting of one part of horse- serum
and two parts of saline. This mixture is placed in tubes
to a depth of 7 cm., which are then heated slowly in a
water-bath from 58° C. to 70° C., at which they are kept

1 Centr.f.'Bakt., Abt. I (Originate), vol. Ixxii, 1913, p. 107.

I

PLATE XXI.

a. Trypanosoma Gambiense. Smear of blood of inoculated rat.
X 1500.

6. Spirochaeta recurrentis (Obermeieri). Smear of blood.
X 1500.

SPIEOCHAETA PERTENUIS 495

for thirty minutes. Small pieces of rabbit kidney are then
pushed to the bottom of the tubes and the incubation
must be carried out anaerobically.

It is probable that the spirachaetes of relapsing fever
in different countries are distinct species.

Spirochaeta Duttoni. — Found in the blood-plasma in
African relapsing, or tick, fever. It closely resembles
the S. recurrentis, but is more readily inoculable into
rats, mice, and guinea-pigs, and the one does not protect
against the other. It is conveyed by a tick, Ornithodoros
moubata, the malpighian secretion of which is the principal
infective agent. The eggs of infected ticks are also in-
fected, and the infection may be transmitted to the third
generation of ticks.

Duval and Todd l state that multiplication of S. Duttoni
takes place in vitro in a culture medium made with hens'
eggs and mouse blood. Leishman believes that certain
chromatin bodies present in the eggs and nymphs of the
ticks are the developmental forms of the spirochaetes.

Blood spirochaetes have been found in many animals,
e.g. cattle (S. Theileri), mice (S. muris), fowls (S. galli-
narum), and geese (S. anserina).

Spirochaeta pertenuis. — Castellani 2 found in the
yaws (frambcesia) granulomata a delicate spirochaete
resembling the S. pallida of syphilis closely, but even
more delicate and difficult to stain than the latter organism,
and named the S. pertenuis. It is present also in the
spleen and lymphatic glands in the disease and in inoculated
monkeys. Rabbits can be inoculated in the testicle and
Noguchi has obtained cultures.

Some observers have supposed yaws to be a manifesta-
tion of syphilis, but (1) syphilitic patients can be inoculated
with yaws ; (2) syphilis may supervene on yaws ;

1 Lancet, 1909, vol. i, p. 834.

2 Brit. Med. Journ., 1907, vol. ii, p. 1511.

496 A MANUAL OF BACTERIOLOGY

(3) Neisser and Castellani have shown that monkeys
inoculated with syphilis are not immune to yaws, and vice
versa ; and (4) Castellani 1 has shown that the yaws antigen
and anti-bodies are distinct from the syphilis antigen and
anti-bodies, though the ordinary Wassermann test may
react with yaws.

Spirochaetes are also present in the ulcerating granuloma
of the pudenda of Guiana (Wise) and Australia, in malig-
nant growths, in ulcers, in the mouth (p. 570), and in
Vincent's angina (p. 296).

Staining methods. — Blood-smears may be stained with
the Leishman or Giemsa stain (p. 102).

Trichomonas vaginalis. — This parasite is found in the acid vaginal
mucus in 50 per cent, of those examined. It must not be mistaken
for a spermatozoon. It is a pear-shaped body, measuring 12 to
30 p. in length, and from the blunt end three flagella are given off.

A much smaller species, T. intestinalis, measuring 4 to 15 p., has
been met with in the intestinal canal of man in conditions associated
with diarrhoea.

Syphilis

Various bacterial organisms have been described in
this disease, e.g. by Lustgarten, Eve and Lingard, Van
Niessen, de Lisle and Jullien, etc., and bodies regarded
as protozoa by Siegel, de Korte, and others. In March
1905, Schaudinn 2 noted the constant presence of a spiriform
organism or spirochaeta (S. pallida, or Treponema or
Spironema pallidum) in various lesions in acquired and
congenital syphilis. The T. pallidum varies from 6 to 15 /m
in length, averaging 8-9 /UL (Plate XXII. a and 6). It is
much more attenuated than the majority of spirochaetes,
having a maximum thickness of 0'3 yu, has from three to
twelve, usually from six to eight, twists, forming a close,

1 Journ. of Hygiene, vol. vii, 1907, p. 558.

2 Arbeit, a. d. Kaiser. Gesurtdheitsamte, xx. 1905.

PLATE XXII.

a. Treponema pallidum from condyloma (T. pallidum with
Spirochaeta refringens). Indian-ink method. X 1000.

b. Treponema pallidum. Smear from condyloma. Giemsa.
X 1500.

SYPHILIS 497

regular, and narrow spiral, is actively motile, possessing
a single delicate flagellum at each end, and it may have
an undulating membrane. It stains feebly and with
difficulty. Another spirochaete, the S. refringens, fre-
quently accompanies, and must not be mistaken for, the
T. pallidum in ulcerating lesions ; the former is more
refractile and coarser, has fewer twists and forms a wider
spiral, and stains deeper and more readily than the latter.
The T. pallidum is found generally in all primary and
secondary lesions of syphilis, e.g. the primary sore and
adjacent lymphatic glands, in the papular and roseolar
eruptions, in condylomata and mucous patches. It has
also occasionally been found in the spleen and blood. In
congenital syphilis the T. pallidum is met with in the
bullous eruptions, blood, and organs, and is particularly
abundant in the spleen and liver (Plate XXIII. a).

Tertiary lesions are generally considered to be non-
infective, and the T. pallidum is usually difficult to find
in them. It has, however, been detected in the peripheral
portions of gummata and in syphilitic aortitis, and may
persist in the body for years after the primary lesion.
Noguchi, after a careful search, has detected the spiro-
chaete in the brain in cases of general paralysis (in 48
cases out of 200 examined) and also in the posterior columns
in a case of tabes.

The T. pallidum is now universally regarded as the
specific organism of syphilis, being present not only in
the human lesions but in experimental lesions of inoculated
apes (see below). It must be recognised that spirochaetes
are of frequent occurrence in various non-syphilitic ulcera-
ting and other lesions, e.g. in the mouth and in pyorrhoea,
in yaws and ulcerating granuloma (in yaws they are specific
forms, see p. 495), in ordinary ulcers and in carcinomatous
tumours. Generally the T. pallidum can be distinguished
microscopically from the other species, but care is necessary.

32

498 A MANUAL OF BACTERIOLOGY

When material from a rhesus monkey inoculated with
syphilis is placed in collodion sacs which are introduced
into the peritoneal cavity of another monkey, a great
multiplication of the organism takes place in the
contents of the sacs a month after the operation.1
Noguchi has obtained cultures of the Treponema pallidum
by making use of serum water (serum 1 part, water 3
parts), sterilised for fifteen minutes at 100° C. on three
days, to which fragments of fresh sterile tissue of a rabbit
(kidney, heart-muscle) were added. Rabbits are inoculated
with syphilis in the testicle and the spirochaete- containing
testicular material is employed to inoculate the tubes,
which are then incubated at 35°-37° C. under strictly
anaerobic conditions. Multiplication of the spirochaetes
commences forty- eight hours after inoculation. The
primary cultures are somewhat difficult to obtain, but
once obtained sub-cultivation is easy. Both thick and
thin forms of the Treponema were obtained, which Noguchi
considers may be distinct varieties.

Metchnikoff and Roux (also Griinbaum) found that the
chimpanzee is very susceptible to syphilis, and can readily
be inoculated from manj the T. pallidum being found in
the lesions.

Macacus rhesus is also somewhat susceptible, likewise
the M. cynomolgus and the Chinese bonnet monkey, but
not the mandril. By several passages through a rhesus
monkey the syphilitic virus becomes attenuated, so that
in man it produces merely a local lesion.2 Syphilis may
also be inoculated on the eye or testicle of the rabbit.

Although the central nervous systems of rabbits and
monkeys are refractory to direct inoculation with T.
pallidum, Noguchi has succeeded in inducing some of the
symptoms (convulsions) and lesions of general paralysis

1 Levaditi and Mclntosh, Ann. de Vlnst. Pasteur, xxi, 1907.

2 Metchnikoff, Journ. of Prev. Med., 1906, August.

PLATE XXIII.

a. Treponcma pallidum. Section of liver of fetus (congenital
syphilis.) Levaditi's method. X 1500.

b, Coccidium oviforme. Section of rabbit's liver, x 350.

SYPHILIS 499

in these animals by the following method. Intravenous
inoculations of dead spirochaete cultures were given every
five days over a period of five months, an interval of five
months was then allowed to elapse, and finally the living
spirochaetes were introduced into the brain, subdurally
or intra-cerebrally.

Attempts by Metchnikoff and Roux to prepare an anti-
syphilitic serum by inoculating apes and goats with
syphilitic virus proved unsuccessful (as did earlier experi-
ments with other animals by Hericourt and Richet). The
syphilitic virus as ordinarily introduced into man by sexual
intercourse probably takes some hours to become gene-
ralised, for Metchnikoff found experimentally in apes that
if the seat of inoculation were treated with a calomel
ointment up to eighteen hours after inoculation infection
was prevented.

By triturating cultures of the Treponema in salt-solution,
heating to 60° C. for sixty minutes, and adding O5 per cent,
of carbolic acid, Noguchi has prepared an agent, termed
Luetin, which can be used for a cutaneous reaction for
the diagnosis of syphilis. In syphilitic infection redness,
sometimes becoming pustular, develops at the site of
inoculation.

The syphilitic virus does not pass through aBerkefeld
filter, and hence is not ultra-microscopic. It is readily
destroyed by heat (52° C.) and antiseptics. Treatment
with mercury and with salvarsan (" 606 ") and neo-
salvarsan cause diminution or disappearance of the spiro-
chaetes.

In central nerve lesions salvarsan is more effective
when injected into the central nervous system, but this
procedure is not free from danger. To obviate this, the
salvarsan may be injected intravenously and then some
of the patient's serum is injected into the spinal canal by
lumbar puncture.

500 A MANUAL OF BACTERIOLOGY

Examination for the T. pallidum

1. Examination in fresh preparations. — Scrapings from the deeper
layers of the chancre, etc., may be emulsified in physiological salt
solution and examined microscopically, particularly with dark-
ground illumination (p. 139).

Another useful method is the Indian-ink method. A scraping is
obtained from the lesion as above, and the fluid thus obtained is
placed on a slide and an equal quantity of ink added. The ordinary
commercial Indian inks may be used, Giinther Wagner's being
particularly good (p. 81). The ink must be examined microscopi-
cally to prove the absence of spirillar forms, which sometimes occur
in it. The serum and the ink are then rapidly and thoroughly
mixed and smeared over the slide so that a pale brown colour
results. The material dries in a minute or slightly less, and may
be examined directly with the oil-immersion lens, or the wet pre-
paration may be covered with a cover-glass and examined.

The preparations, which keep for a considerable time, show the
red blood-cells as large clear circular areas in a brownish -black
field, the bacteria and debris as white rods, dots, etc., and spiro-
chaetes, as clear white spirals (Plate XXII. a).

It is particularly important in using this method that in so far
as possible serum alone be used, and that a minimal amount of
mucous material or fibrin be mixed with the ink. The presence of
mucus results in the taking up of a large amount of the colouring
matter of the ink, with the result that a smear of the requisite
colour and thickness cannot be made. If too much serum is used
the albuminous material appears to precipitate the colour from the
fluid and a finely granular appearance is seen microscopically, which
is practically worthless for diagnostic purposes. Again, if too
much ink is used, the surface of the smear is increased in size to
such an extent that the task of examining it thoroughly is greatly
lengthened.

Coles * notes a useful point in the recognition of the treponema,
namely, that if the number of turns of the spiral of the syphilitic
spirochaete be counted, six or seven turns will be found in a length
equal to the diameter of a red blood-cell. The distance from the
top of one spiral to the next is from 1 to 1-2 p. As red blood-cells
measure about 7-5 p in diameter, on an average six or seven turns
will be equal to the diameter of a red blood-cell. The treponema
varies in length from 6 to 15 p., or even more, and consequently
1 Brit. Med. Journ., May 8, 1909.

THE WASSERMANN REACTION 501

contains from six to fourteen and sometimes twenty or more turns.
This measurement of the length of the spiral is usually possible,
and is of the greatest value in identifying the treponema.

2. Stained preparations. — Smears from chancres, etc., may be
stained by the Giemsa method.

The smears are fixed for ten minutes in absolute alcohol. The
preparations are then stained in a dilute solution of the Giemsa
solution for two to twenty-four hours, washed in distilled water,
dried, and mounted. (The dilute Giemsa is prepared by adding
one drop of the Giemsa stain to a cubic centimetre of distilled water,
and rendering alkaline with one drop of 0-01 per cent, potassium
carbonate solution.) The preparations may also be stained in the
undiluted Giemsa stain for half to six hours. Leishman's solution
may also be used or the Giemsa method described under " Malaria."

Sections may be stained by Levaditi's method :

(1) Fix pieces of tissue about 1 mm. thick in 10 per cent, formalin
for twenty -four hours.

(2) Wash in water, and harden in 96 per cent, alcohol for twenty-
four hours.

(3) Wash in distilled water for some minutes (until pieces sink).

(4) Place in 3 per cent, silver nitrate solution at 37° C. for three
to five days in the dark.

(5) Wash in distilled water for some minutes, and then place
in the following solution at room temperature for twenty-four to
forty-eight hours.

Pyrogallic acid ...... 2-4 grm.

Formalin ....... 5 c.c.

Distilled water . . . . . 100 c.c.

(6) Wash in distilled water, dehydrate in absolute alcohol, clear
in xylol, embed in paraffin, cut, and mount.

The spirochaetes are stained black or brown (Plate XXIII. a),
the tissues yellow.

Some have asserted that the spirochaetes seen in the tissues after
staining by this method are artifacts or are composed of filaments
of elastic tissue.1

3. The Wassermann reaction or antigen test. — This is applied in
the diagnosis of syphilitic conditions, and as a confirmatory test
of the syphilitic nature of such conditions as tabes dorsalis and
general paralysis of the insane. The test is based on complement-
fixation (p. 183). In this method an " antigen " consisting of
micro-organisms, or an extract thereof, fixes its homologous immune-

1 See Saling and Miihlens, Centr. f. Bakt. (Orig.), xlii and xliii.

502 A MANUAL OF BACTERIOLOGY

body, and the complex then takes up complement ; this is demon-
strated by the use of a haemolytic system (p. 184).

As a matter of fact, however, the Wassermann reaction, as it
is preferably termed, is apparently not a true antigen reaction, for
substances may be used as antigen which are soluble in alcohol,
and various non-specific bodies may be similarly employed. More-
over, the nature of the substances which act as amboceptor and
fix the complement is uncertain ; some regard them as globulins,
others as lipoids, and while Wassermann considered them to be
specific anti-bodies, others believe them to be derived from a peculiar
degeneration or breaking down of the tissues in syphilis. Again,
the reaction is not confined to syphilis : it may also be obtained
with the syphilitic " antigen " in malaria, trypanosomiasis, yaws,
leprosy, and the early stage of scarlatina.

In the original method a fresh salt-solution extract of the liver
of a syphilitic fetus was used as the " antigen." Levaditi employed
a similar extract of the dried and powdered liver. The test-sub-
stance was the blood -serum inactivated by heating to 56° C. for
half an hour or cerebro -spinal fluid of the patient. The complement
was guinea-pig serum, and the haemolytic system sheep's corpuscles,
and a serum haemolytic for these corpuscles.

Many modifications of this method have since been introduced
both as regards the reagents employed — antigen, complement, and
haemolytic system — and the manner of carrying out the test. These
may now be briefly considered and the manner of carrying out the
test described.

(a) Antigen. The various substances which have been used as
antigen include :

1. A watery or alcoholic extract of syphilitic fetal liver.

2. Alcoholic extract of normal liver or heart-muscle 1(human,

ox, sheep or guinea-pig), with or without previous extrac-
tion with acetone.

3. Alcoholic extract of normal heart-muscle with the addition

of cholesterin.

4. Various artificial mixtures, e.g. lecithin and cholesterin,

sodium glycocholate or taurocholate.

5. Extracts of pure cultures of the Treponema pallidum obtained

by Noguchi's method.

Probably the most widely employed antigen at the present day
is number 3, the so-called " Sachs antigen."

1 Heart muscle is peculiar in that it contains a large amount of lipoid
substances.

THE WASSERMANN REACTION 503

•

This is prepared by extracting 1 grm. of heart-muscle with 10 c.c
of absolute alcohol for three or four days. To 4 parts of this
extract add 5 parts of a 1 per cent, alcoholic solution of
cholesterin.

The solution of antigen must be tested as regards its possible com-
plement fixing properties alone and in the presence of a known
positive syphilitic serum. This is done by taking a series of dilutions,
e.g. from 1 in 2 to 1 in 64, adding to each of these hsemolytic system
and complement, and observing (after incubation) the least dilution
of antigen which ceases to fix. This having been ascertained, this
particular dilution of antigen is now tested with a good known
positive syphilitic serum to see that it does fix complement under
these conditions. If the antigen does not fulfil these requirements
it must be rejected and a fresh one prepared.

(b) Hcemolytic system. — This may be serum haemolytic for ox,
sheep, or human red-blood corpuscles, with the homologous cor-
puscles (see p. 183).

The hcemolytic serum or hcemolytic amboceptor is usually prepared
in the laboratory by injecting a rabbit with washed red-blood
corpuscles (see p. 184), but the horse is occasionally employed.
The haemolytic serum should be of high titre and may conveniently
be stored in quill-tubing or in small ampoules, which after sealing
are heated in a water-bath to 57° C. for an hour on three or four
successive days and kept in a dark cool place. This sterilises and
destroys the complement, leaving only the haemolytic amboceptor.

The blood corpuscles. — If blood is obtained from the slaughter-
house it should be defibrinated at the time of bleeding. If human
blood is used (Noguchi's method), the blood from a prick is allowed
to drip into citrated saline solution. In either case, the corpuscles
are washed twice with 0-85 saline solution and a sufficiency is used
to make a 5 per cent, suspension by volume in the saline solution.
This is preferably used fresh, but will keep for a day or two in the
ice-safe.

(c) Complement. — Fresh guinea-pig serum diluted to 1 in 10 with
0-85 per cent, saline is usually employed, though in some methods
the complement present in the serum to be tested is made use of.
In the method here described the amount of complement present
in the fresh guinea-pig's serum is ascertained, the serum being then
diluted to double the minimum amount required to produce complete
haemolysis.

Every fresh lot of the hsemolytic amboceptor should be tested
both as to its haemolytic activity and also as to the amount of
complement necessary to produce complete haemolysis with varying

504

A MANUAL OF BACTERIOLOGY

TABLE A.

TABLE illustrating the manner of testing the activity of hsemolytic
amboceptor.

5 per

Amboceptor

Guinea-

cent, sus-

dilution
Amount in
each test

pig com-
plement
in 50 per

pension
of
sheep's

Salt
solution.

Amount of
amboceptor
used

Dilution
of ambo-
ceptor

Result.
Amount of
Haemolysis

0-1 c.c.)

cent. sol.

cor-

puscles

1:10

0-1 c.c.

0-5 c.c

1-8 c.c.

0-01 c.c.

250

+ + + +

1:20

0-1 c.c.

0-5 c.c.

1-8 c.c.

0-005 c.c.

500

+ + + +

1 :50

0-1 c.c.

0-5 c.c.

1-8 c.c.

0-002 c.c.

1250

+ +

1 :70

0-1 c.c.

0-5 c.c.

1'8 c.c.

0-00142 c.c.

1750

-f +

1 :100

0 1 c.c.

0-5 c.c.

1-8 c.c.

0-001 c.c.

2500

+

1:200

0-1 c.c.

0-5 c.c.

1-8 c.c.

0-0005 c.c.

5000

+

Control (with-

out ambo-

ceptor)

0-1 c.c.

0-5 c.c.

1-9 c.c.

—

—

0

Three to four times the minimum complete haemolytic dose of
hsemolytic serum is employed in the test, viz. in the above test where
a dilution of the amboceptor of 1 in 500 is the minimum quantity
showing complete haemolysis, a dilution of about 1 in 150 would be
used.

TABLE B.

TABLE illustrating the manner of testing the amount of complement
necessary to produce complete haemolysis in the presence of
varying amounts of amboceptor.

0-5 c.c. of a 5 per

cent, suspension

Results.

of Sheep's

Amboceptor diluted

corpuscles
sensitised with

Undiluted
complement

0-05 c.c. of the

diluted

amboceptor

1:5

1:10

1 :45

0-55 c.c.

0-15 c.c.

+ + + +

+ 4-4-4-

+ + + +

0-55 c.c.

0-12 c.c.

+ + -f +

+ + + +

+ 4-4-4-

0-55 c.c.

0-10 c.c.

+ + + +

4- + + +

+ 4- +

0-55 c.c.

0-08 c.c.

+ + + +

+ + +

+ +

0-55 c.c.

0-05 c.c.

+ +

+

+

(The number of + signs indicates the amount of haemolysis ;
+ + + + = complete, 4- 4- + = nearly complete, + + — partial,
+ = trace.)

THE WASSERMANN REACTION 505

amounts of the hsemolytic amboceptor. In this way the haemolytic
amboceptor is standardised and the manner of carrying out these
two tests is illustrated by the Tables (A and B) on p. 504.

(d) Fluid to be tested. — Either the blood-serum or the cerebro-
spinal fluid is used in the test. If the test is carried out in small
test tubes, then 5 c.c.-lO c.c. of blood must be withdrawn from a
superficial vein with a sterile 10 c.c. syringe. If, however, the test
is carried out in small quill -tubes (which the writer believes is quite
as efficient as with larger tubes), then only 1 c.c. or 2 c.c. of blood
or less are needed (a Wright's capsule-full will suffice), and this may
be obtained from the ear or by binding some small rubber tubing
round the thumb and puncturing the soft tissues at the side near
the nail : when bleeding ceases, the rubber ligature should be
removed and re -applied, and this may be repeated two or three
times. The blood should be collected in a sterile tube and allowed
to coagulate ; this may be hastened if necessary by placing the
tube of blood in the warm incubator for half an hour and centri-
fuging. After the serum is separated, it is pipetted into another
tube, which is then heated in a water- bath to 56° C. for half an
hour immediately before testing in order to destroy its content of
complement. The latter procedure is important as a proportion
of sera from diseases other than syphilis may react positively if
un heated.

In certain nervous diseases, e.g. general paralysis and tabes, it
may be necessary to test the cere bro -spinal fluid, which may react
positively when the serum is negative. The fluid is obtained by
lumbar puncture. An amount rather larger than the serum is
required ; it should be free from blood and cellular elements (which
may be removed by centrifuging if necessary) and it should not be
heated.

THE TEST. — Tubes about 3 in. by -^ in. diameter are used when
the " large quantity " method is carried out and the necessary
quantities of the reagents are measured with 1 c.c. pipettes divided
into hundredths. In the " small quantity " method tubes about
1| in. by i in. internal diameter are used, and the necessary quan-
tities are measured with a Wright's pipette furnished with a rubber
teat and having a volume or unit marked with a grease -pencil
about f in. from the point, and also a four -unit mark higher up.
Dilutions are made with 0-85 per cent, saline solution, and in all
cases the same total volume should be maintained in all the tubes.
The " small quantity " method may be now described ; if the
" large quantity " method is adopted the principle is precisely
similar but larger volumes of the reagents are used. For the

506 A MANUAL OF BACTERIOLOGY

details of the "small quantity" method here given I am indebted
to my friend and colleague, Dr. F. E. Taylor, who has elaborated
it in this form.

For a single serum six small quill tubes are required, five being con-
trols and the sixth the test, and for every additional serum two more
tubes are required. The six tubes (and the additional ones also
when more than one serum is being tested) are arranged in two rows
in a metal rack which is immersed in a water-bath and maintained
throughout the test at a temperature of 38°-40° C. To each tube
in the back row run in 4 volumes of saline solution with the marked
Wright's pipette, into each tube of the front row four volumes of
the antigen suitably diluted as ascertained by the standardisation
of the antigen. Then commencing from the left hand, add to each
of the first two tubes (back and front) one volume of diluted known
normal inactivated serum (these form the negative control mixtures).
To the next two tubes add one volume of diluted known syphilitic in-
activated serum (these form the positive control mixtures). Then
to each of the next two tubes add one volume of the inactivated
serum to be tested, diluted 1 in 2, and repeat this with as many
series of tubes as there may be sera to be tested. Next to every
tube add one volume of suitably diluted complement, and leave in
the warm bath for five minutes. After this add five volumes of the
prepared haemolytic system and leave in the bath for fifteen minutes.
The haemolytic system is prepared by mixing in bulk four volumes
of suitably diluted inactivated haemolytic serum and one volume of
20 per cent, suspension of washed sheep's corpuscles. The pipette
used for the additions of the reagents should be rinsed with saline
solution between each constituent of the test.

At the end of fifteen minutes the tubes are centrifuged and the
pressure or absence of haemolysis noted. If haemolysis has occurred,
the fluid in the tubes form a clear red solution without deposit of
corpuscles, whereas if fixation is complete the corpuscles are
deposited at the bottom of the tube while the fluid above is colourless
and transparent.

All the tubes in the back row should show haemolysis as they
contain no antigen, tube 1 in the front row should also show haemo-
lysis as it contains antigen and a negative serum, tube 2 in the front
row should show no haemolysis as it contains both antigen and
positive serum. In the remaining front row tubes, haemolysis or
fixation will occur according as the sera are negative or positive
respectively.

If it be desired to obtain some idea of the amount of syphilitic
amboceptor present in a positive serum, a quantitative estimation

INFUSORIA 507

may be carried out by putting up a series of tubes containing either
diminishing quantities of the serum or diminishing quantities of
antigen, the other constituents remaining the same in either case.

As human serum generally contains amboceptor hsemolytic for
sheeps' corpuscles, Flemming * devised a method in which the
test serum itself with sheep's corpuscles constitutes the haemolytic
system, and the test is also carried out with Wright's pipettes.
Emery 2 in his method makes use of human corpuscles and of the
complement present in the test serum so that addition of com-
plement is unnecessary (in this case, of course, the test -serum is
not inactivated).

The examination of a very large number of cases of syphilis by
different observers indicates that the test is of very considerable
value and diagnostic significance. In conditions such as tabes
dorsalis and general paralysis of the insane, which on other grounds
are generally regarded as due to syphilis, 52 per cent, give the
reaction. A positive reaction may be said to show a positive, and
probably active, syphilitic infection, but a negative reaction does
not necessarily exclude syphilis. Energetic mercurial treatment
may render the reaction negative.

(4) Forges' reaction. — If syphilitic serum be added to a solution
of lecithin or other lipoid substances, in many cases it gives a white
precipitate. Normal or non-syphilitic serum gives no precipitate.
This has been tried extensively as a substitute for the Wassermann
reaction, but it is not so delicate.

Class III. — Infusoria (Ciliata)

The Infusoria are protozoa the locomotive organs of which
consist of cilia, and in which the nuclear apparatus is differentiated
into a vegetative macronucleus and a generative micronucleus.
The cytoplasm is enclosed within a cuticle, an oral aperture is
present in the form of a slit or pore, and waste matter is extruded
by a pore, constant in position, but, as a rule, visible only when in
use. A contractile vacuole is generally present. Reproduction
usually takes place by fission, which is preceded by division of the
two nuclei, the micronucleus by mitosis, the macronucleus by direct
division.

The Infusoria are not of much pathological importance, but are
common in ponds and ditches, e.g. Paramecium and Vorticella.

1 Lancet, 1909, vol. i, p. 1512.

2 Ibid. 1910, vol. ii, September 3.

508

A MANUAL OF BACTERIOLOGY

Balantidium (Paramecium) coli

This is an intestinal parasite of swine, occasionally met with in

man in conditions associated with chronic diarrhoea and dysentery.
It is somewhat ovoid in shape, the ends being bluntly pointed, is
covered with cilia, measures 65 to 85 /x in
length, and has a superficial resemblance
to the ordinary Paramecium.

According to Saville Kent, the Balan-
tidium coli is to be distinguished from the
ordinary forms of water paramecia by the
following characters : The Bal. coli is some-
what spindle-shaped or ovoid, and bluntly
pointed at each end, one and a half to
twice as long as broad, measuring ^i^ in.
to T J-^ in. in length ; the paramecium is
more cylindrical, four times as long as
broad, measuring yl^ in. to ¥TF in. in
length. The oral aperture in Bal. coli is
near one extremity (Fig. 57) ; in para-
mecium it is situated at about the middle
of the ventral surface. In Bal. coli the
cilia round the oral aperture are as long
again as those over the body generally ; in

paramecium the whole of the cilia are of the same length.

The Bal. coli seems undoubtedly sometimes to be a cause of

dysentery.1 Bal. coli is a common parasite of pigs and may

contract infection from these animals.

FIG. 57. — Balantidium
coli.

Examination of Flagellated and
Ciliated Forms

(1) These may be examined fresh in the fluid in which they are
present, by mounting on a slide, and covering with a cover-glass
one edge of which rests on a bristle to avoid pressure.

(2) Permanent mounts may be made by the Heidenhain method
(p. 485).

(3) Films may be made in the ordinary way, and stained with
weak carbol-fuchsin or Leishman's stain. (The organisms are
apt to be distorted. )

(4) The following method, devised by Rousselet (Journ. Quekett

1 Strong and Musgrave, Johns Hopkins Hosp. Ball., vol. xii, 1901,
p. 31 ; Bureau of Gov. Laboratories, Manila, Bull. 26, 1904.

COCCIDIA 509

Microscop. Club, 2nd series, vol. vi, no. 36, p. 5, March, 1895) for
preserving Rotatoria, may be tried. In those forms which are
non-contractile, kill by adding a drop of J per cent, osmic acid,
wash immediately in water, and preserve in 2£ per cent, formalin.
Contractile forms may be first narcotised by adding a drop or two
of 2 per cent, cocaine solution, then killed with the osmic and
preserved as before.

Class IV. — Sporozoa

The sporozoa are exclusively endoparasitic protozoa, the adult
lacking organs for locomotion and for the capture of food, and
multiply by some method of sporulation, often very complex.
Binary fission is almost unknown in this group. A parasite during
the nutritive or " trophic " phase, when it is absorbing nutri-
ment and growing at the expense of its host, is termed a trophozoite ;
when it is mature and ready for sporulation it is termed a sporozoite
or schizont. The spores are of various kinds, and may develop
outside the body or in a second host.

Order. — Coccidiidea

The Coccidiidea, with a single exception, are intra-cellular during
the trophic stage, and present a dimorphism or alternation of
generations ; the one is endogenous and asporular, determining
the reproduction of the parasite within the host, the other exogenous
and sporular and permitting of infection.

Coccidial Disease of Rabbits

This is a disease caused by a sporozoon, the Coccidium (Eimeria)
oviforme or cuniculi, and often met with in warrens and hutches ;
in some of the former as many as 90 per cent, of the animals may be
affected. The young animals suffer most, and become infected when
they cease to suckle and commence to eat green food, the adult ani-
mal as a rule resisting the disease. The affected animals waste, suffer
from enteritis, and a large proportion die in from one to three weeks,
the condition being known as " wet-snout " among the keepers.
The parasites occur in the intestine, bile-ducts, and liver in large
numbers. Each parasite is ovoid in shape, measuring 36 p. in length
and 22 p, in breadth, is enclosed in a firm translucent cyst, which
encircles a very granular protoplasm. Sometimes this protoplasm

510

A MANUAL OF BACTERIOLOGY

becomes condensed so as to form a spherical mass lying free within
the cyst (Fig. 58, A). In the intestine and bile-ducts the parasites
are attached to the epithelial cells, and in the liver, if the animal
lives beyond the acute stage, set up some remarkable changes. The
affected liver is studded with greyish-white nodules varying in

FIG. 58. — Diagram of Development of Coccidia.1

size from a pin's head to a pea. On making sections and examining
them microscopically, it is found that these nodules consist of
dilated bile-ducts filled with a much hypertrophied and convoluted
mucous membrane, which forms branched projections covered with
cubical epithelium, among which the parasites occur in great numbers
(Plate XXIII. b). A curious fact is that subcutaneous or intra-
venous inoculation, or inoculation into the liver of a healthy rabbit
with the coccidia from another rabbit, fails to induce the disease.

1 This diagram is reproduced by permission from Daniel's Tropical
Medicine and Hygiene, 2nd ed. 1913 (John Bale, Sons, and Danielsson).

COCCIDIA 511

The coccidium has a complicated developmental history, and
infection only seems possible in one of the stages. In order to study
the life -cycle the parasite must be placed under suitable conditions,
and an infusion of rabbits' faeces, kept at the ordinary temperature,
is perhaps as good a cultivating medium as any, the changes being
watched by means of interlamellar films. Reproduction may be
either asexual or sexual, and may be endogenous, within the host,
or exogenous, outside the host. In the asexual cycle, division of
the protoplasm and nucleus of the coccidium takes place and the
cyst comes to contain large numbers of spores (Fig. 58, A). The
cyst-wall then ruptures, the spores are liberated, pass into other
intestinal or hepatic cells and reproduce the coccidium once more
(Fig. 58, A). In the sexual cycle, the protoplasm of some coccidia
remains undivided with a single nucleus and the cyst has a weak
spot, known as the micropyle ; these are the female cells or macro-
gametes (Fig. 58, B). In other coccidia, the protoplasm having
attained maximum growth, divides into a mass of actively motile
thread-like bodies, the male elements or microgametes. The cyst-
wall then ruptures and the microgametes, penetrating the micropyle
of the macrogametes, fertilize them. In the fertilised macrogamete,
which is a zygote known as an " oocyst " and is non-motile, the
micropyle closes and the cyst is discharged with the faeces of the
animal. On damp ground, the nucleus and protoplasm divide
into four spherules. Each spherule becomes elongated, and again
divides into two somewhat crescent-shaped bodies, around each
pair of which a new, somewhat spindle-shaped capsule forms (Fig.
58, D). In this condition the parasite is very resistant, and may
remain alive for six months, undergoing no further change unless
introduced into another animal. If a young rabbit swallows with
its food these crescentic spores, the enclosing capsule is dissolved,
and each crescent becomes a rounded amoeboid mass, and this
again divides up into many crescentic spores. These spores are
apparently motile, and enter the epithelial cells of the intestine,
gall-bladder, and bile-ducts, where a process of growth and
differentiation occurs, and the fully developed parasite is ultimately
reproduced.

Coccidial disease, or, as it is sometimes termed, psorospermosis,
is occasionally met with in animals, as the sheep, and a wasting
disease of young pheasants due to coccidia has been described by
McFadyean.1 Coccidiosis also occurs in grouse and poultry, due
to Eimeria avium ; in the latter causing " scour," which may be
attended with considerable loss.

1 Journ. Comp. Path, and Therapeut., 1895.

512 A MANUAL OF BACTERIOLOGY

In man, coccidial disease has been described (but rarely) in the
liver, gall-bladder, ureter, etc.1

Rixford and Gilchrist 2 described two cases of protozoan infection
of the skin and organs, accompanied by great destruction of tissue
and ending in death. The organisms were spherical, 1 to 21 p
diameter, surrounded by a thick capsule, enclosing granular bioplasm
(C. immitis).

The Ruffer-Plimmer bodies of cancer were at one time believed
to be coccidia (p. 554).

The term " psorospermosis " has been applied to human infection
with coccidium, Sarcosporidia (p. 532), etc.

Examination

(1) The coccidial forms are readily examined in the fresh state-
The only bodies they are likely to be mistaken for are certain ova.

(2) Paraffin sections of rabbit's liver containing coccidia may
be stained much in the same way as tuberculous tissues — viz. warm
carbol-fuchsin ten minutes, decolorise cautiously in 5 per cent,
acid, and counter-stain in methylene-blue. Sections may also be
stained in the Ehrlich-Biondi stain for one to two hours.

Order. — Haemosporidia

The general characters of this group are :

(1) Life at the expense of the red blood-corpuscles, at least
during a portion of the life-cycle.

(2) Endogenous multiplication by spores, by which the life-
cycle is repeated within the host.

(3) Development of a form which becomes free in the plasma,
and which is the commencement of a sexual cycle to be completed
in a second host.

(4) Inoculability, but only from one animal to another of the
same species.

The group includes the malaria parasite and similar parasites in
mammals and birds, the haemogregarines, Drepanidium of the
frog, and perhaps the Piroplasmata.

1 Journ. Comp. Path, and Bact., 1898, June, p. 171.

2 Johns Hopkins Hosp. Reps., vol. i, 1896, p. 209.

MALARIA 513

Malaria

Malaria is caused by parasitic protozoa, placed in the
genus Plasmodium (Hcemamceba), the credit of the discovery
of which must be given to Laveran, who described the
parasite as occurring in four phases, viz. (1) spherical
bodies, (2) flagellated bodies, (3) crescentic bodies, and
(4) segmented or rosette bodies.

The parasites cannot be cultivated beyond one
generation, but inoculation of healthy individuals with
the blood of malarial patients reproduces the disease,
and the same structures or parasites are found in the blood
of these infected persons. Inoculation experiments on all
animals except man have proved negative, and in the
latter the inoculation must be intravenous.

In the various forms of malarial fever the parasites have
the same general characters, though there are distinct
differences between them, by which they can be recog-
nised and the type of fever differentiated. In each there
is an endo-corporeal cycle within the host, through which
the recurrent attacks are developed ; there is also an
extra-corporeal cycle of development outside the body of
the host, whereby the infection of fresh individuals becomes
possible. Each of these cycles needs separate description.

If the blood of a malarial patient is examined an hour
or two before, or at the very commencement of, the febrile
paroxysm, the parasite will be recognised as a pale, ill-
defined mass of protoplasm within the red corpuscles, of
which a variable proportion are infected, the size of the
parasite varying in the different types of fever. When
some hours old a variable number of blackish pigment-
granules of melanin make their appearance. These subse-
quently coalesce into smaller groups, and the latter again
into one or two larger, more or less centrally disposed,
masses. The parasites exhibit more or less amoeboid

33

514 A MANUAL OF BACTERIOLOGY

movement, and the melanin granules are frequently in a
state of tremor. Later on most of the parasites (now
schizonts) become divided into a variable number of
segments, which separate and become spherical, the blood-
corpuscle breaks down, the spherical bodies or spores are
set free, and a certain number of them, again becoming
attached to red corpuscles, develop into the first stage of
the parasite. The melanin granules and some of the
spores are ingested by phagocytes, and after some time the
melanin is deposited in the spleen and liver.

The parasite, termed a plasmodium, or better, an amce-
bula, contains a vesicular nucleus and a nucleolus, and
the melanin granules are present in the surrounding proto-
plasm. When segmentation occurs, each segment contains
a portion of both the nucleolus and the protoplasm. The
maturation of each " brood " of parasites is coincident
with a fresh febrile paroxysm. In the subtertian (per-
nicious) forms of malarial fever there exist in the blood
for some time after the subsidence of the acute paroxysms
well-marked non-motile, crescentic or sausage-shaped
bodies, with rounded ends, the so-called " crescentic
bodies " or " crescents " ; their longer diameter is greater
(^) than that of a red corpuscle, their protoplasm is finely
granular, and contains at about the centre several well-
marked pigment-granules. In the crescentic forms the
extremities of the crescent often appear to be joined by a
delicate membrane (Fig. 64, / and j) ; this is the remains
of the blood- corpuscle in which the parasite has developed.

When a " wet " specimen of malarial blood from a
case of pernicious or sub-tertian malaria is kept under
observation (p. 523), it not infrequently happens that
after a time the so-called flagellated " bodies " make their
appearance. These consist of a central protoplasmic mass
attached to which are from one to six delicate flagella
measuring 20-30 jm in length (Fig. 59, c). The flagella

THE MALARIA PARASITE

515

are actively motile and disturb the corpuscles, but the
body itself does not move much. Frequently one or
more of the flagella break away and swim free, remaining
active for several hours. The flagellated bodies are never
seen in the freshly drawn blood, and Ross has found that
flagellation does not occur if the finger be pricked through
a spot of vaseline, the blood remaining covered with the

FIG. 59. — Development of the malaria parasite in the mosquito.
a, b, and c, the male gametocyte ; d, e, and /, the female
gametocyte ; /, fertilisation of the female gametocyte by a
microgamete. (After Ross and Fielding- Ould.)

film of grease. Careful observation has shown that the
flagellated bodies develop from " crescents " in subtertian
malaria, and from special rounded parasites, difficult to
distinguish from the schizonts, in the benign tertian and
quartan fevers.

Various theories were held in the past as to the nature
of these flagellated bodies. Through the brilliant researches
of Ross, which have been confirmed and extended by
observers in all parts of the world, it is now known that
these cells are sexual elements. The flagellated body
represents the male cell or " male gametocyte," the flagella
(" gametes ") being analogous to the spermatozoa of

516 A MANUAL OF BACTERIOLOGY

higher animals. The female cells or female gametocytes
or gametes are non-flagellated, and are fertilised by the
entrance of one of the flagella of a male gametocyte. This
fertilisation takes place in the stomach (middle intestine) of
certain species of mosquito, and after fertilisation a series of
changes ensues resulting in the formation of spore-like
bodies, which are injected when the insect bites its victim,
and thus the infection of fresh individuals with the malaria
parasite takes place. The first demonstration of the nature
of " flagellated bodies " was given by Opie and MacCallum
on the Halteridium, a parasite of pigeons (p. 528), and this
forms a good example of the value of abstract research
to practical medicine (see p. 528). Ross also followed the
development of the malaria-like Proteosoma of sparrows,
etc., in the mosquito, Culex fatigans. The development
of the malaria parasite of man in the mosquito is as follows,
according to Ross and Fielding- Ould.1 It is not known
what determines whether an amoebula will become a
sporocyte or a gametocyte. When the sexual cells or
" gametocytes " are ingested with the blood by the mos-
quito, they pass into the middle intestine. Within a few
minutes the corpuscles enclosing them break down, the
parasites are set free, and quickly become spherical or
ovoid (Fig. 59, c, e, and/). One or two spherical granules
are often attached to the naked parasites, and may repre-
sent polar bodies (Fig. 59, c and /). Very soon the male
cells become flagellated (Fig. 59, c), and before long the
flagella or " microgametes " break away from the parent
cell, and by their own motility make their way through
the liquor sanguinis. Should one come in contact with
a female cell or " macrogamete," it fuses with the latter,
uniting with the nucleus (Fig. 59, /), fertilisation is com-
pleted, and a " zygote " is formed. As the zygote at this
stage is motile it is known as a " travelling vermicule " or

* Thompson Yates Laboratories Report, vol. iii, pt. ii, p. 183.

THE MALARIA PARASITE

517

" ookinet " ; it passes into the outer wall of the mosquito's
stomach, where it becomes encysted (Fig. 60, a, 6). At
this period the zygote is about 7-8 ^ in diameter. If
development proceeds, it acquires a distinct capsule and
begins to grow rapidly, and when mature at the end of a

FIG. 60. — Development of the malaria parasite in the mosquito.
(After Ross and Fielding-Ould.)

week or more, according to the temperature, is 60 /x in
diameter, and projects into the body-cavity of the insect
(Fig. 60, b). Its substance next divides into eight to
twelve portions, or " zygotomeres," then each zygotomere
becomes a spherical body, or " blastophore " (Fig. 60, c),
and each blastophore develops upon its surface a number
of spindle-shaped, radially disposed bodies, or " zygoto-
blasts " (Fig. 60, d). When the zygote reaches maturity

518

A MANUAL OF BACTERIOLOGY

the blastophores disappear, leaving its capsule packed with
large numbers (" thousands ") of free zygotoblasts. The
capsule then ruptures, and the zygotoblasts are poured
into the body-cavity of the mosquito. The " blasts "
measure 12-16 JUL in length, taper at each extremity, and

The
Mosquito Phase

Exogenous
,or Sexual Cycle

FIG. 61. — Diagram of the asexual and sexual cycles of the
malaria parasite.

possess a central nucleus (Fig. 60, e), and they make their
way to all parts of the body of the host, and accumulate
in the salivary or poison glands, whence they are dis-
charged by the middle stylet (hypopharynx) of the pro-
boscis, when the insect " bites," into the circulation of a
fresh vertebrate host. Here, presumably, the blasts be-
come attached to erythrocytes and develop into amcebulse.

THE MALARIA PARASITE 519

The diagram1 (Fig. 61) represents in graphic form the asexual
and sexual cycles of reproduction of the malaria parasite.
So far as is known, malarial infection is conveyed only
through the bite of infected mosquitoes of the sub-family
Anophelince. It has been repeatedly proved that infected
mosquitoes convey infection, and that if mosquitoes be
excluded human beings may live in the most malarious
districts without contracting the disease.

Mosquitoes (Culicidce) are distinguished from other mosquito-
like insects by the fringe of scales on the wings. The common
mosquitoes belong to the sub-family Culicince. The Anophelince,
are usually less abundant (but there is great variation in different
districts), and bite mainly at night ; the females alone are blood-
suckers. Some species breed in natural collections of stagnant,
others in slowly running fresh, water well supplied with lowly forms
of vegetable life. If the head of a mosquito be examined with a
hand-lens, three sets of appendages will be noticed. In the middle
is the stout proboscis containing the stinging and suctorial appa-
ratus ; situated at the base of this are two palpi, one on either side,
and outside these again are two antennae, which are more or less
hairy. In Anophelince, both male and female, the palpi are as long
as the proboscis ; in the female Culex (also in Stegomyia and many
other genera) they are short and stumpy. In Anophelince the scales
on the veins of the wings are usually arranged in alternating light
and dark patches, giving a speckled or dappled appearance, different
as a rule from anything seen in Culex. (Some Culices have a similar
arrangement, and it is wanting in A. maculipennis and A. bifurcatus.)
The front or costal margin of the wing in Anophelince is almost
always marked with dark blotches. Anopheles, as a whole, is a
more slender insect than Culex, and when at rest its body is all in
one line, whereas Culex is angular or hump-backed. The important
species known to carry malaria are Anopheles maculipennis in Europe,
N. Africa, and N. America, A. bifurcatus in Europe, Myzomyia
funesta and Pyretophorus costalis in Central and W. Africa, and
Cellia argyrotarsis in tropical America. Other species, e.g. Myzo-
rhynchus sinensis, Cellia Kochii, and others, are less important
carriers.

(On Mosquitoes, see Theobald, Brit. Museum Monograph, and
Allbutt's System of Med., ed. 2, vol. ii, pt. 2 ; Giles, Handbook of

1 This figure is reproduced by permission from Daniels' Laboratory
Studies in Tropical Medicine (Bale, Sons, & Danielsson, 1908).

520

A MANUAL OF BACTERIOLOGY

the Gnats and Mosquitoes ; Daniels, Laboratory Studies in Tropical
Medicine, ed. 3, 1908.)

There are probably at least three species of malaria
parasite l occurring in the various types of malarial fever
in man, though some authorities (e.g. Laveran) regards
the forms as varieties of a single species, and the following
are the differential characters between them :

(1) Benign quartan fever (Fig. 62). — The quartan parasite

FIG. 62. — The quartan parasite : a, 6, c, d, amoebulse ; e,
sporocyte ; /, free spores ; g, female gametocyte with so-
called polar body ; h, male gametocyte. (After Rees.)

(Plasmodium malarice) completes its asexual life- cycle
in seventy- two hours ; there are two complete days without
an attack, and reckoning the day of the previous attack,
an attack occurs every fourth day, hence the name " quar-
tan." It commences as a small amcebula, which is feebly
motile. It enlarges, becomes pigmented, and motility
ceases, the pigment-granules being numerous and coarse.
The parasite finally occupies nearly the whole of the
corpuscle, which, however, is but little altered (a-d).

Towards the end of the apyrexial period the pigment
collects in the centre, and segmentation takes place with
the formation of a symmetrical rosette (e), and afterwards
of six to twelve spores (/). The quartan parasite does not

1 Hewlett, Trans. XlVth Internal. Congress of Hygiene, vol. ii. 1908,
p. 141,

PLATE XXIV

a. Malaria.

Parasite of benign tertian fever.
X 1500.

Smear of blood.

b. Malaria.

Gametocyte of benign tertian parasite,
of blood. X 1500.

Smear

THE MALARIA PARASITE

521

form crescents, and the flagellated bodies (h), which are
rarely seen, are developed from large pigmented parasites.
(2) Benign, or spring, tertian fever (Fig. 63 ; Plate
XXIV. a). — The benign tertian parasite (Plasmodium vivax)
completes its asexual life- cycle in forty- eight hours, an
attack occurring every other day, or, reckoning the day
of the previous attack, every third day. In the early
stage it resembles the quartan, but shows much more

FIG. 63. — The benign tertian parasite : a, b, c, d, amcebulae ;
e, sporocyte ; /, free spores ; g, female gametocyte with so-
called polar bodies ; h, male gametocyte. (After Rees.)

active amoeboid movement. The pigment-granules are
also finer than in the quartan, and incessantly change
their position. The parasite finally invades the whole
corpuscle, which becomes enlarged and pale. Enlarge-
ment of the corpuscles is a marked feature in the benign
tertian infection (d).

Segmentation takes place, but is unsymmetrical (e),
resulting in the formation of a grape-like cluster of twelve
to twenty spores (/). As in the quartan, no crescentic
bodies are developed, and the gametocytes (g, h) are
similar to, but larger than, the quartan (Plate XXIV. b).

(3) The cestivo-autumnal, malignant, pernicious, or sub-
tertian, fevers (Fig. 64). — This parasite (Laverania malarice)
(or parasites, for it has been divided into three species

522

A MANUAL OF BACTERIOLOGY

by the Italian observers, viz. the pigmented and the
unpigmented quotidian and the malignant tertian, but
this is not generally accepted) is much smaller than the
quartan or benign tertian, and when it reaches the stage
of multiplication it disappears from the peripheral blood
and collects in the internal organs, spleen, liver, cerebral
capillaries, and bone-marrow. It is actively amoeboid,
seems to change its position within the corpuscle, and
the pigment-granules are very fine in the young parasites,

g h j

FIG. 64. — The sub-tertian parasite : a, b, c, amoebulse ; d, sporo-
cyte ; e, free spores ; /, g, h, female gametocyte ; j, k, I,
male gametocyte. (After Rees.)

but early aggregate into large clumps. The fission forms
(d, e) are only met with in the internal organs. Multiple
infection of the corpuscles may also occur. The corpuscles
often suffer severely from the infection, some being shrivelled
and spinous, others dark in colour, " brassy " ; they may
also be altered or destroyed without being actually invaded
by the parasite. It is in this form that the crescentic
bodies appear (/, j). These, however, are not met with
at the very commencement of the attack, but appear in
a week or so, and may not disappear until some weeks
after the termination of the attack. This parasite is met
with in the sub-tertian, or so-called malignant, types of
fever, which are characterised by irregularity of the fever,

PLATE XXV.

a. Malaria. A tertian " rosette." Smear of blood X 1500.

b. Halteridium DanilewsJcyi. Smear of pigeon's blood. X 1500.

DIAGNOSIS OF MALARIA 523

considerable blood destruction, often accompanied by
haemoglobinuria, and cachexia ; coma is another complica-
tion in certain instances, probably caused by massing of
the parasites in the cerebral capillaries.

The cure of malaria by quinine is regarded as being due
to a poisonous action on the parasites analogous to that
exerted on numerous protozoa, amoebae, for example,
being injuriously affected by so little as a 1-50,000 solution
of quinine hydrochlorate.

No toxin can usually be demonstrated in the blood of
those suffering from a malarial attack, but Rosenau and
his co-workers have found that the filtered blood, taken
when the temperature is rising, produces a malaria-like
paroxysm. By withdrawing blood containing parasites,
adding glucose, and incubating at 37° C., the multiplicative
cycle of the malaria parasite, as seen in the blood, is passed
through in the culture tube.

A malaria-like parasite (Plas. Kochii) occurs in apes, in which it
produces fever.

The nature of Blackwater fever, so called from the presence of
haematuria and haemoglobinuria, has given rise to much discussion.
By some it is considered to be a disease sui generis, of unknown
etiology. By others it is regarded as a form of malaria, either of
an intense type, or in which the kidneys are especially involved, or
as due to malarial infection plus quinine. It may be that under
particular conditions, of the nature of which we are at present
ignorant, haemolysins may be set free and cause haemolysis, the blood-
pigment being eliminated by the kidneys.

Clinical Examination

The blood of malarial patients may be examined either in the
unstained or stained condition.

Examination in the unstained condition. — The finger or lobe of
the ear is pricked, and a droplet of blood taken up on a clean cover-
glass, which is then placed upon a slide, so that the droplet of
blood spreads out into a thin layer between the two glasses. The
cover-glass may then be ringed with oil or vaseline to prevent

524 A MANUAL OF BACTERIOLOGY

evaporation. A little practice is required to judge the right quan-
tity of blood. The preparation should be examined with a TV-inch
oil-immersion lens.

Examination in the stained condition. — To prepare stained films
the finger or ear is pricked of the malaria or other blood parasites,
e.g. trypanosomes, and a droplet of blood is taken up on the edge
of the end of a slide '(the spreader), which is then applied to the
surface of a second slide and, holding the spreader at an angle of
45°, it is pushed along the surface of the second slide so that a thin
film of the blood is left behind, and the process is repeated for as
many films as are required. A little practice is required to gauge
the right quantity of blood. Other methods of preparing blood-
films are to deposit a droplet of blood on a cover-glass ; another
cover-glass is applied, and the two are separated so that each is
smeared with a thin film of blood, or a droplet of blood on a slide
may be spread with a cigarette paper, or with the shaft of a needle.
Whatever method is adopted, the film is allowed to dry in the
air, and may then be fixed (not if Leishman's stain is used). In
order to fix, the smears should be immersed in a mixture of equal
parts of absolute alcohol and ether for not less than ten minutes,
preferably for half an hour ; this gives excellent results. In hot
countries a saturated solution of corrosive sublimate may be used.
The methods detailed at p. 97 may also be employed.

Staining is usually carried out with Leishman's stain (No. 12,
p. 102). The blood films, unfixed, are flooded with a few drops
(5-10) of the stain, which is spread by tilting, and in hot weather
the preparation should be covered with a capsule to prevent evapo-
ration. After a half to one minute distilled water is added and
mixed with the stain, in sufficient amount to produce an abundant
precipitate, and the mixture should appear pinkish ; the water
should be about double the amount of stain used, and staining is
continued for five, or in some cases for ten, minutes. The staining
should be continued until the nuclei of the leucocytes are a rich
purple when examined with a low power. The film is then rinsed
with distilled water, a little distilled water is left on the film, which
is watched under the low power until the red corpuscles appear
red ; this takes half a minute or more. The water is now tilted
off the film, and the slide on edge allowed to dry, or it may be blotted
and dried. Fresh films stain better than old ones ; if the films
are old, staining with the diluted stain should be prolonged for
ten or fifteen minutes and differentiation with distilled water
may take five minutes. Jenner's or Giemsa's blood-stain may be
similarly used.

DIAGNOSIS OF MALARIA 525

The writer is indebted to Dr. A. C. Coles of Bournemouth, for
the following method of staining blood-parasites.

In order to obtain good stained films of blood containing para-
sites it is essential to have good slides, well cleaned, a film of blood
spread as uniformly as possible, and to avoid any precipitation of
the stain on the surface of the film.

Slides are best cleaned with whiting or Creta preparata, made
into a paste with water, or with Windowlein, a preparation used
for cleaning windows. Rub the whiting thinly over the surfaces
of the slide, and when dry rub off with a clean cloth.

The impedimenta required for staining the blood film are :

1. Drop bottle of about giij capacity containing distilled water ;

2. Pipette bottle of about Jij to 3iij capacity for the staining

solution ;

3. Bottle of Giemsa's staining solution ;

4. Bottle of Merck's pure methylic alcohol ; both well corked ;

5. A Politzer's bag ; and preferably, though not essential,

6. A curved piece of window glass, 8 inch x 4 inch.

Into the perfectly dry pipette bottle pour some of the Giemsa's
solution, and add about twice as much pure methylic alcohol ;
shake up and keep well stoppered.

Drop from the pipette bottle just enough of the diluted Giemsa's
solution to cover the film. Allow it to act for about ten to twenty
seconds [if longer, especially in a hot climate, the alcohol evaporates
and precipitates the stain].

Then drop on as much distilled water as the slide will hold —
that is, about eight times as much water as stain — allow the stain
and distilled water to mix, and stain for the requisite time.

It is better, however, in order to prevent the precipitation of the
stain, to pour off the diluted stain and water from the film on to
the surface of a piece of slightly curved plate-glass, and immediately
place the slide, film side downward, on this. The duration of
staining varies according to the temperature of the room and the
nature of the film — generally speaking, ten to twenty minutes
give excellent results ; but a good plan is to remove the film,
flood off the stain with distilled water, and examine under low
power. If the nuclei of the leucocytes are of a ruby-red colour,
the staining is successful. If they are blue, the film is insufficiently
stained, and it should be replaced on the staining fluid ; if they
are blackish red, it is too deeply stained for most purposes, and all
that is required is to pour distilled water on the surface, watching
the effect (easily seen by holding the slide over a piece of white
paper), and as soon as the whole film is faintly pink the staining

526 A MANUAL OF BACTERIOLOGY

will be good. This method of staining, generally known as Giemsa's
new method, closely resembles Leishman's, but very much more
distilled water is added.

The exact tint or colour of the objects stained in this way will
depend largely on the reaction of the distilled water used to dilute
the stain. If the water is acid (as most distilled water is) the
red blood-corpuscles are stained a reddish, if alkaline they are
often bluish, in colour.

When the film has been sufficiently stained, do not pour off the
stain and then wash, but flood off the stain with distilled water
and so avoid any deposition of precipitate.

When the film has been quickly washed, it is essential to dry it
as quickly as possible, otherwise decolorisation proceeds. The
films should not be dried with filter or blotting-paper ; it tends to
leave fluff. They are best dried by blowing on the surface with
air from a Politzer's bag.

Films of blood which have been kept for some time, especially
in the tropics, will never stain well. Films should therefore be
stained at once, and they will keep indefinitely in a dry place.
The method of packing stained or unstained films face to face or
wrapped in paper is a barbarous one ; the surfaces soon get scratched
and dirty. The best plan is to pack them back to back in a racked
box, or if this is not at hand, stick a small piece of gummed paper
at the end of the slide on the film side, and when this is thoroughly
dry, but not before, they can be packed together.

It is essential that the films should be absolutely dry before they
are mounted, and if they are mounted in Canada balsam or cedar-oil
they will sooner or later fade and be perfectly useless. The best plan
is to mount them in parolein or liquid paraffin as described by Coles
(Lancet, April 1, 1911), which has lately been advocated by Giemsa.

If the above-named stains are not available staining may also
be done in a half-saturated aqueous solution of methylene-blue or
in Loffler's blue for half an hour, washing in water, and counter-
staining with a very weak eosin solution for a few seconds, washing
and drying. Manson recommends treating the films with a very
weak acetic acid — two or three drops to the ounce of water — to
dissolve out the haemoglobin, and, after washing, staining in the
following solution for half a minute :

Borax ....... 5 parts

Methylene-blue ..... 0-5 part

Water 100 parts

washing, drying, and mounting in xylol balsam.

PLASMODIUM PR^ECOX 527

Haematoxylin (Ehrlich's, or Mayer's haemalum) is preferable for
permanent preparations, and in hot countries, where methylene-
blue rapidly fades. The preparations may be counter-stained with
a weak solution of eosin.

Ross recommends for rapid diagnosis the use of thick blood films,
from which the haemoglobin is first removed with very dilute acetic
acid ; the films are then stained with Leishman's stain, and
examined with a J-inch objective. Practice is required for this
method.

In order to demonstrate the flagellated organisms Manson
recommends the following procedure : Thirty or forty strips of
thick blotting-paper (3 inches by 1J inch), each having an oblong
hole (^ inch by f inch) cut lengthways in the centre, are prepared,
moistened with water, and laid on a sheet of window glass. A
patient is selected in whose blood the crescentic form is plentiful,
and a minute droplet of the blood, about the size of a pin's head,
is expressed from a prick. A clean slide is then breathed on, and
the droplet of blood picked up on it and spread out with a needle
so as to cover an area f inch by £ inch. The slide is immediately
inverted over a blotting-paper cell and pressed down sufficiently
to secure perfect apposition. The rest of the paper cells are simi-
larly covered with blood-charged slides. In from half to three-
quarters of an hour the slides are removed and dried by gentle
warming, and then fixed with absolute alcohol for five minutes.
The alcohol is allowed to evaporate, and the films are treated with
a few drops of 15 per cent, acetic acid to dissolve out the haemo-
globin. The slides are then washed in water and stained with
weak carbol fuchsin (20 per cent.) for six to eight hours, washed
in water, dried, and mounted.

N.B. — Negative results in the examination for the malaria
parasite must be accepted with caution unless repeated. A single
undoubted parasite is sufficient to establish the diagnosis. Quinine
causes the disappearance of the parasite. The parasites in the
sub -tertian fever disappear during the apyrexial intervals (except
the crescents), and are most likely to be found at the commencement
of the attack — i.e. when the temperature is rising. The parasites
of the other forms are larger and more obvious during the apyrexial
intervals.

[For further particulars on Malaria and on the demonstration of
the malaria parasite, see Daniels' Laboratory Studies in Tropical
Medicine, 1908.]

528 A MANUAL OF BACTERIOLOGY

Plasmodium prsecox

Syn. Proteosma Grassii, Hcemamoeba relicta.

This parasite (commonly called " proteosoma ") is met with in
sparrows and other birds, in which it invades the red blood-cor-
puscles, and its structure and development are practically identical
with those of the benign malarial parasites of man. It grows from
a minute granule into an amoeboid plasmodium, which ultimately
segments and forms a rosette. In some specimens of blood flagel-
lated male gametocytes make their appearance, similar to those
of malaria, the flagella break away from the main mass, fertilise
other non-flagellated or female cells, and a series of changes ensues
analogous to those occurring in the malaria parasite (p. 516). The
fertilisation and development of the fertilised cell take place in the
stomach of a mosquito (Culex fatigans), by which the infection is
transmitted to other birds.

Halteridium Danilewskyi

This is an elongated, curved parasite (also known as Hcemo-
proteus or Hcemamoeba Danilewskyi}, found in the red corpuscles of
certain birds (pigeon, crow, etc.), and embracing the nucleus (Plate
XXV. b). By some it is included among the malaria-like parasites
(Plasmodium). At an early stage it much resembles the proteosoma,
but as it grows it becomes elongated, pigment -granules appear,
and are either scattered throughout the protoplasm or collect in
two groups, one at each extremity. Finally, the parasite occupies
nearly the whole of the corpuscle, dislocating its nucleus. The
fully grown parasites may be differentiated into two forms, one
of which remains almost completely unstained when treated with
methylene-blue, the other staining deeply with this dye (Opie).
When the blood is withdrawn, the corpuscles disintegrate and
liberate the contained parasites, which assume a circular outline,
and a certain number become flagellated. It is only the non-staining
form which becomes flagellated. These two varieties of the parasite
are the male and female cells respectively, and the fertilisation of
the female cell by a free flagellum has been actually observed by
MacCallum.1 It can hardly be doubted that the development of
the fertilised cells takes place in some insect, but the definitive
host has not yet been discovered with certainty.

The presence of these parasites induces rise of temperature,

1 Journ. Exper. Mecl, vol. iii, 1898, pp. 79, 103, 117.

THE PIROPLASMATA 529

deposition of melanin, and changes in, and enlargement of, the
spleen and liver, analogous to those occurring in malaria in man.
The Halteridium parasite, according to Schaudinn, is a stage in
the life-cycle of a trypanosome (see p. 494).

Somewhat similar parasites are frequent in the blood of the lower
vertebrates (see Plate XXVI. &).

The Piroplasmata x

Syn. Pyrosoma, Bdbesia.

The Piroplasmata form a somewhat anomalous group, but are
usually included in the Haemosporidia of the Sporozoa. They
differ from the Plasmodia in the following respects : absence of
pigment, non-fragmenting of the nucleolus, division into two or
four only, and frequency of extra-corpuscular forms. They cause
many diseases in animals, are conveyed by ticks, but are unknown
in man. (A piroplasma was described as the causative organism
of Rocky Mountain spotted fever by Wilson and Chowning, but
the observations appear to be erroneous, see p. 546). The body
of a piroplasma is typically pear-shaped (Plate XXVI. a), but
rounded and rod forms occur. Two nuclear masses are present,
one larger than the other.

The developmental cycle in the ticks has not been worked out,
but Koch has observed peculiar rayed forms with P. bigeminum,
and Christophers 2 various developmental forms with P. canis.
Miyajima states that a piroplasma of Japanese cattle (apparently
P. parvum) in blood broth develops into typical trypanosome forms.3

Piroplasma bigeminum. — This is the parasite of the well-known
Texas fever of cattle, a disease which is characterised by fever,
emaciation, anaemia, haemoglobinuria, and enlargement of the liver
and spleen.

The disease causes considerable loss among cattle, and is met
with in various parts of the world, America, Australia, South Africa,
Malaya, the Philippines, the Roman Campagna, Greece, Roumania,
and North Ireland.

In the acute type of the disease a small proportion (1-5 per cent.)
of the red corpuscles in the peripheral circulation contain pairs of

1 See Hewlett, Trans. XlVtk Internal. Cong, of Hygiene, Berlin,
vol. ii, 1908, p. 146 ; Minchin in Allbutt's System of Med., ed 2, vol. ii,
pt. 2, p. 86.

2 Brit. Med. Journ.. 1907, vol. i, p. 76.

3 Philippine Journ. of Science, vol. ii, 1908, p. 37.

34

530 A MANUAL OF BACTERIOLOGY

pyriform bodies 2-4 /j. in length and 1-5-2 p. in largest diameter.
One end of each body is rounded, and the body gradually tapers to
a point at the other end, and the pair lie close together, their tapering
ends directed towards each other. A dark spherical body is present
at the rounded end of the parasite.

Some of the young parasites exhibit amoeboid movements when
the blood is examined on a warm stage. In the internal organs
the parasites are more numerous ; in the kidney and liver 10-25
per cent, of the corpuscles contain them, in the heart-muscle
50 per cent. In the mild type 5-50 per cent, of the corpuscles in
the circulating blood may be infected at one time or another, and
the parasite appears in some cases as a coccus-like body at the
periphery of the corpuscle. This appears to become enlarged and
spindle-shaped, then to taper in the middle, divide, and so give rise,
to the pyriform bodies. Occasionally minute free coccoid bodies
are seen in the plasma, and at times two to five minute (0-5 p)
coccoid cells are present in the red cells. After death the pyriform
bodies seem to become spherical or angular.

Sexually differentiated gametes are not known with certainty
but flagellated forms have been described.

The disease is transmitted through the bites of ticks (Rhipi-
cephalus annulatus, R. australis). The female tick, after biting an
infected ox and sucking its blood, falls off and lays its eggs ; the
eggs hatch in two to six weeks' time, and the daughter ticks transmit
the disease to other animals through their bites.1 The disease may
be to some extent controlled by prophylactic measures designed
to destroy the ticks, and to prevent infection thereby.

A partial immunity is enjoyed after an attack of the disease,
but by repeated attacks the immunity may be rendered absolute.
By inoculation with the blood of an affected animal in which the
fever has subsided, a transient illness in the inoculated animal
is produced together with partial immunity, and by a second
or third inoculation the immunity may be much increased. The
mortality from such a procedure amounts to 3-5 per cent.2

P. parvum causes Rhodesian red-water of cattle. It is not
directly inoculable, and is conveyed by the tick R. appendiculatus.

P. equi causes biliary fever in horses.

P. canis causes epidemic jaundice in dogs (Plate XXVI. a).
It is conveyed by the ticks Hcemaphysalis leachi in South Africa,

1 See Smith and Kilborne, Texas or Southern Cattle Fever, United
States Dep. Agricult. Bull. No. 1, 1893.

2 See Tidswell, Report on Protective Inoculation against Tick Fever,
New South Wales, Dep. Pub. Health, vol. i, 1898 ; vol. ii, 1900.

PLATE XXVI.

a. Piroplasma canis. Film of blood, x 1500.

&. Hwmocystidium (Hcemoproteus) najce. Pigmented parasite of
Cobra (Naja hajce).

MICROSPORIDIA 531

E. sanguineus in India, and Dermacentor reticulatus in Europe.1
(On Ticks, see Nuttall, Journ. Eoy. Inst. of Public Health, vol. xvi,
1908, p. 385.)

H aemogregarina

The Hsemogregarines (which must be distinguished from the
Gregarines) are unpigmented parasites, not amoeboid, typically
having an elongated body or vermicule, occurring in the blood,
mostly in cold-blooded vertebrates, but several species have of
late been found in mammals (dog, jerboa, palm squirrel), though
not in man. In the dog, the parasite (Leucocytozoon canis) occurs
as an elongated, curved or doubled-up body in the polymorphonu-
clear leucocytes. It is encapsuled and contains a single granular
nucleus. Encystment with sporulation occurs in the bone -marrow,
and a sexual development is stated to occur in a tick.

Hcemogregarina (Drepanidium, Lankesterella) ranarum inhabits
frogs (Eana esculenta), and possesses both an intra- and an extra-
corpuscular phase. In the former the parasite occurs as an elon-
gated gregarine-like body within the red corpuscles, which increases
in size until its length is 10-15 p ; it then divides into numerous
small or a few large gymnospores. In the first case the spores
may number fifty, are 3-5 p in length, occur in May or June, and
are exclusively within the erythrocytes ; in the latter case the
spores measure 5-8 \i in length, are five to fifteen in number, and
develop within cells in the blood-forming organs. The extra-
corpuscular phase, commencing within the corpuscles, ends in an
elongated organism possessing a vermicular movement, and free
in the plasma. Similar parasites are frequent in the lower verte-
brates, e.g. snakes.

Order. — Myxosporidia

In this group the trophozoite is amoeboid, and the species are
almost exclusively parasites of fish, in the young stage being intra-
cellular (" fish psorosperms ").

Order. — Microsporidia

The Microsporidia are cell parasites of invertebrates, especially
arthropods, and the trophozoite is more or less amoeboid.

1 See Nultall and Grab am -Smith, Journ. of Hygiene, vols, iv to viji,
1904-8.

532 A MANUAL OF BACTERIOLOGY

Nosema bombycis causes pebrine, a disease of silkworms, which
is of considerable importance commercially, for the silk industry
in France was once threatened with extinction owing to its ravages.
The infected worms do not grow normally, cease to eat, and die, or
may form abnormal pupse. Within the body of the affected worms
a large number of roundish, highly refractile corpuscles are found.
Pasteur ascertained that the disease was propagated by healthy
worms eating with their food the excreta of infected ones. The
moths were thus infected, and laid infected eggs. By allowing each
moth to lay its eggs separately, and subsequently examining the
body of the moth microscopically, he was able to separate the healthy
from the diseased, and the eggs of the former were kept, while those
of the latter were destroyed. According to Pfeiffer,1 when the
worms eat the excreta containing the corpuscles mentioned above,
these lose their capsule and form large amoeboid masses which
penetrate the muscles and blood-corpuscles. The amoeboid masses
then become encapsuled and are yellow and granular. Later on
the bright roundish corpuscles form within them.

The Isle of Wight bee disease is caused by Nosema apis, which
is mainly confined to the alimentary tract.

Another disease of silkworms is known as flacherie, but is due
to a bacterium, Micrococcus bombycis. It is contagious, and can
be transmitted by inoculation.

Order. — Sarcosporidia

The parasites belonging to this order are not thoroughly worked
out. They complete their life-history in the substance of striated
muscular fibres : such are the well-known Miescher's corpuscles.
Few instances of this class of parasite are recorded in man, but it
occurs in the monkey 2 and also in the ox. T. Smith 3 describes
the characters and development of a species found in mice.

A parasite, Rhinosporidium kinealyi, nearly allied to the fore-
going, causes a polypoid condition in the nose in the tropics. If
a section be made of the mass, cysts (pansporoblasts) will be seen
in the deeper layers containing many refractile rounded nucleated
bodies, the spores. Neither the life-history nor the mode of trans-
mission of the parasite is known.

1 Zeitschr.f. Hyg., vol. iii, 1888, p. 3.

2 De Korte, Journ. of Hygiene, vol. v, 1905, p. 451

3 Journ. Exper. Med., vol. vi, No. 1, 1901, p. 1.

CHAPTER XIX

SCARLET FEVER— HYDROPHOBIA— INFANTILE PARA-
LYSIS — TYPHUS FEVER — YELLOW FEVER -
DENGUE— PHLEBOTOMUS FEVER— VACCINIA AND
VARIOLA— MALIGNANT DISEASE

Scarlet Fever

VARIOUS organisms have been described in scarlet fever —
a bacillus by Eddington, a streptococcus by Frankel and
Freudenberg, protozoa by Mallory and others. The
disease may be milk-borne, and in the historic Hendon
outbreak a streptococcus was claimed by Klein to be
the specific infective agent, but the researches of Crookshank
and others seem to disprove this.

In 1885 an epidemic of scarlet fever occurred in Mary-
lebone, and was traced to infection conveyed by milk
supplied from a farm at Hendon. The infection could
not be traced to any human source, and it was therefore
concluded that the cows themselves were affected with
scarlet fever, and infected the milk. A vesicular eruption
was found on the udders and teats of the cows, and this
was regarded as the local manifestation of bovine scar-
latina. From the vesicles and crusts Klein isolated a
streptococcus which, although closely resembling the
Streptococcus pyogenes (as then known), differed slightly
from it ; on inoculation into calves it produced death,
with lesions of the kidney resembling those of the human
disease. Klein also isolated the same streptococcus in

533

534 A MANUAL OF BACTERIOLOGY

five out of eleven cases of the disease in man. The con-
clusions which Klein and Power came to were, therefore,
that scarlet fever is communicable to, and may exist in
cows, the milk thereby becoming infected and conveying
the disease to man, and that a streptococcus is the specific
infective agent.

The Hendon outbreak was reinvestigated by Axe and
Crookshank.1 Axe found that, so far from there being
no source of human infection, cases of scarlet fever had
occurred near the dairy within a short time of the out-
break, and the eruptive disease of the cow was shown by
Crookshank to be cowpox, while the so-called streptococcus
of scarlet fever he regarded as a variety of the S. pyogenes.
The existence of bovine scarlet fever is entirely discredited
by the veterinary profession, both here and on the Continent.

In 1909 a milk-borne epidemic occurred in certain
districts in London and Surrey, and was traced to milk
derived from one farm. The outbreak was investigated
and reported on by Hamer and Jones, who again traced
it to infection of the cows. Hunting 2 reviews the evidence
and shows how little there is to support this conclusion,
as there is no doubt that the family of one of the employees
on the farm were suffering from scarlatina.

Scarlatina seems to be inoculable in the chimpanzee
and some of the lower apes. It is now regarded as being
caused by a filter-passer.

Gordon 3 reinvestigated the bacteriology of scarlatina with
special reference to the Streptococcus scarlatince or conglomeratus
of Klein. He found that this organism differs distinctly in its
cultural characters from other varieties of streptococci, and that
it occurs constantly in the mucous secretion on the surface of the
tonsils and fauces and in the nasal, but not in the aural, discharge

1 On the Hendon outbreak, see Trans. Path. Soc. Lond., 1888 (Refs.).

2 Journ. Roy. Sanitary Inst., vol. xxxii, 1911, p. 62.

3 (a) Rep. Med. Off. LOG. Gov. Board for 1898-99, p. 480 ; (b) ibid.
for 1899-1900, p. 385.

HYDROPHOBIA 535

in scarlatina. It is also present in a somewhat modified form in
the blood and tissues post mortem. It was not found in four
non-scarlatinal throats examined. Gordon concluded, therefore,
that the S. scarlatina or conglomeratus is the "specialised and
essential agent " of scarlatina. It is pathogenic to mice.

Cumpston ! investigated the biological characters of 101 strep-
tococci isolated from scarlet fever, applying Gordon's tests (p. 233).
The majority corresponded with the S. longus type.

Baginsky and Sommerfeld, Class and Jaques also isolated strep-
tococcoid organisms in scarlatina, but they possessed no very
distinctive cultural characters.

It seems very doubtful if streptococci are the etiological agents
in scarlet fever ; they are probably secondary infective agents.
It is remarkable how frequently diphtheria complicates scarlatina.

Mallory detected small bodies, 2-7 p in diameter, staining deli-
cately but sharply with met hylene -blue, and occurring in and
between the epithelial cells of the epidermis and in the lymph-
vessels and spaces of the corium. He regards these as protozoa,
but others consider them to be degenerated leucocytes (see p. 537).

The blood in the early stages of scarlatina gives the Wassermann
reaction (p. 502).

%

Hydrophobia 2

Hydrophobia attacking man is invariably contracted
through the bite of an animal affected with the disease,
In the lower animals the disease is termed rabies, and is
most frequent in the dog, but the cat, wolf, and deer are
also subject to it, and other animals can be infected by
inoculation. The disease may assume two forms — the
raging and the paralytic. The latter is not met with in
man, unless certain rare forms of acute ascending paralysis
(e.g. Landry's) be manifestations of it. In the dog either
may occur, but in rodents the paralytic form is almost
always the one assumed. In man the incubation period
is very variable ; it is never less than about twenty days,

1 Journ. of Hyg., vol. vii, 1907, p. 599.

2 See Marie, La Rage, 1901 ; Scientific Memoirs Gov. of India, Nos.
30 and 44; Luzzani, Ann. de VInst. Pasteur, xxvii, 1913, p. 1039
(Bibliog.).

536 A MANUAL OF BACTERIOLOGY

and possibly may be as long as two years, or even more ;
the average seems to be about ten weeks. In the rabbit,
after inoculation from the dog, the incubation period is
about two to three weeks.

The virus resides in the central nervous system, as was
shown by Pasteur. Inoculation with emulsions prepared
from the medulla and with the saliva conveys the disease,
but the filtered emulsions are usually inactive, and the
other tissues and fluids of the body, excepting the lacrimals
and suprarenals, are non-infective.

Remlinger * has found that after very complete tritura-
tion the virus may pass through a porcelain filter.

No micro-organism has been demonstrated with certainty
in rabies. Negri has described the constant presence of
structures — the Negri bodies — particularly in the grey
matter of the hippocampus major, which he regards as
protozoa. They are of varying size, apparently encap-
suled, taking a homogeneous purplish colour in smears
stained with eosin and methylene-blue, the smallest
spherical and structureless, larger ones with a central
granule or nucleus, the largest, round, ovoid or elongated,
containing several (as many as eight) granules (Fig. 65).
They occur abundantly in animals suffering from chronic
rabies, but in the acute type are scanty, though still to
be found ; in " fixed virus " (p. 538) they are very small.
So constantly are the Negri bodies present in rabies, and
absent in non-rabic conditions, that their presence or
absence forms a rapid and simple means of diagnosis.2

Inasmuch as the rabies virus is filterable, the view
taken by Prowazek of the nature of the Negri bodies is
that they represent the tissue reaction to invasion by the
parasite, the parasite being an extremely minute one

1 Bull, de I'InsL Pasteur, iv, 1904, p. 342.

2 See Williams and Lowden, Journ. Infect. Diseases, vol. iii, 1906,
p. 452.

HYDROPHOBIA

537

and contained within the body and belonging to a group
of the Protozoa termed the Chlamydozoa. In the same
category he would place the trachoma bodies, the Mallory
bodies of scarlatina and the Councilman bodies of variola.
Noguchi believes that the Negri bodies or derivatives
from them can be cultivated in his medium used for the
Trep. pallidum (p. 497).

Babes states that the virus is destroyed at a tempera-
ture of 60° C., but the medulla and other infective material

FIG. 65. — Smear from hippocampus major of rabid dog :
n, nucleus of nerve-cell; b. b, the Negri bodies (eosin
and methylene-blue). (After Williams and Lowden.)

retain their virulence for months in glycerin. He has
described certain lesions present in the medulla in cases
of rabies, the so-called rabic tubercles. These consist of
an invasion of the peri-ganglionic spaces by an accumulation
of round-cells, with degeneration of the cells of the bulbar
nuclei.

Van Gehuchten has described as pathognomonic of
rabies certain lesions in the sympathetic and cerebro-
spinal ganglia, especially those of the pneumo-gastric.
These ganglia consist normally of a supporting tissue

538 A MANUAL OF BACTERIOLOGY

holding in its meshes large ganglionic cells with distinct
well-staining nuclei, each being enclosed in a capsule
lined with endothelium. The changes in rabies consist
in atrophy of the ganglionic cells, which become shrunken
and no longer fill the enclosing capsule, and their nuclei
at the same time become ill-defined and stain badly.
A number of new-formed cells also appear within the
ganglionic capsules. Ravenel and McCarthy studied
twenty- eight cases of rabies in various animals, and consider
that these capsular and cellular changes in the ganglia,
taken in conjunction with the clinical manifestations,
afford a rapid and trustworthy means of diagnosis of
rabies, but that the absence of these changes does not
necessarily imply that rabies is not present. They also
consider that the rabic tubercle of Babes is present suffi-
ciently often to furnish valuable assistance in cases where
the central nervous system only is obtainable.1

Pasteur showed that the virus can be attenuated by
desiccating the infective nerve matter, and in this way
was able to prepare a vaccine which protects animals from
otherwise fatal doses of the virus. Advancing a step
further, he used his vaccines to treat individuals who had
been bitten by rabid animals, but in whom the symptoms
had not yet developed, and so inaugurated the present
system of anti-rabic inoculation as carried out at the
Pasteur and other institutes.

To prepare the anti-rabic vaccines, a rabbit is inocu-
lated subdurally with an emulsion made from the medulla
of a rabid dog. When the animal dies, a second rabbit is
similarly inoculated from the first, and the passage through
rabbits is continued until a " fixed " virus is obtained,
with which the first symptoms appear on the seventh or
eighth day, and which kills with certainty in about ten

1 See Journ. Compar. Pathol. and Therapeut., vol. xiv, pt. i, 1901,
p. 37.

HYDKOPHOBIA 539

days. This having been attained, two or three rabbits
are inoculated subdurally every day, so that there is a
daily supply of animals dead of the disease. The spinal
cord is removed with aseptic precautions, cut into con-
venient segments, and suspended in bell jars containing
a layer of caustic potash at the bottom, which serves to
desiccate them. The jars are dated, and preserved in
glass cases in a dark room, kept at a constant temperature
of about 23° C. In Paris the vaccine fluids are prepared by
triturating portions of the dried cords in sterile broth,
so as to form an emulsion — 1 cm. of cord in 5 c.c. of sterile
broth, of which 1 c.c. (i.e. 2 mm. of cord) forms a single
dose. At the commencement of treatment the cords
which have been dried for fourteen days are used, at the
end of treatment those which have been dried for only
three days ; the latter are much more virulent, and would
communicate the disease but for the previous treatment.
The rabbits employed should all be of the same weight
(2J kilogrammes in Paris) ; if the rabbits are small, a
slightly shorter period of desiccation of the cords would
be necessary. The treatment varies in duration according
to the severity of the case, which is gauged by the number
and situation of the bites and by the species of animal.
Bites on exposed parts are regarded as much more serious
than those through clothing, and on the face, where
efficient treatment is difficult, than on the hands, and
wolf-bites than dog-bites.

The doses are injected subcutaneously in the flank,
and do not produce much constitutional disturbance.
At first there is a feeling of lassitude, and considerable
muscular tenderness at the seat of inoculation, which
later on passes off. At Lille, where there are only a few
cases under treatment at a time, the cords, after drying
for the requisite period, are placed in pure sterile glycerin.
In this they retain their virulence unimpaired for about

540

A MANUAL OF BACTERIOLOGY

a month. This method does away with the necessity for
the daily inoculation of rabbits, a rabbit being inoculated
occasionally as required. The system of dosage employed
at the various anti-rabic stations differs somewhat ; the
following is that employed at Lille, 2 mm. of cord being
emulsified in 5 c.c. of sterile broth, or physiological salt
solution :

ORDINARY TREATMENT.

ORDINARY TREATMENT.

Day of Days of

Day of Days of

treat- desiccation

treat- desiccation

ment. of cord.

ment. of cord.

1 (two injections)

14 and 13

13 . . .

3

2

12 and 11

14 (two injections)

9 and 8

3

10 and 9

15

7 and 6

4

8 and 7

16 .

5

5 .

6

17 .

4

6 .

5

18 .

3

7

4

8 . ! !

3

9 (two injections)

9 and 8

FOR SEVERE BITES, in Addition.

10

7 and 6

19 (two injections) . 7 and 6

11 .

5

20 ., . 5 and 4

12 .

4

21 . . . .3

At Buda-Pesth a dilution method has been employed ;
instead of drying the cords, an emulsion is made with
the fresh cord, and this emulsion is considerably diluted
for the earlier doses, dilutions of 1 in 10,000 to 1 in 6000,
corresponding to cords dried for from fourteen to eight
days. Semple * has found that a carbolised emulsion
of the cord may be employed as the inoculating agent.
An 8 per cent, emulsion of the cord in physiological salt
solution with 1 per cent, carbolic acid is kept at 37° C.
for twenty-four hours. At the end of this time an equal
volume of physiological salt solution is added and the
emulsion bottled aseptically. This vaccine will keep for
months.

Undoubtedly the Pasteur inoculations will protect
animals from rabies, the duration of immunity after

1 Sc. Mem. Gov. of India, No. 44.

ANTI-KABIC INOCULATION 541

vaccination in the dog being at least three years. In man
the efficacy of the treatment can only be judged by
statistics. The mortality after bites by supposed rabid
animals is variously stated, the most favourable being
about 16 per cent. (Leblanc). At the Pasteur Institute,
Paris, among 2730 cases treated in which the animal
which inflicted the bites was proved to be rabid by inocu-
lation experiments, nineteen deaths occurred — a mortality
of O7 per cent. In 1910, 401 cases were treated, with
no death ; in 1911, 341 cases, with one death ; in 1912,
395 cases, with no death, being mortalities of O'OO, 0'29,
and 0*00 per cent, respectively.

The failure of the treatment may be due to two causes :
(1) delay in its commencement, and (2) a short incubation
period. The principle of the treatment probably depends
upon the long incubation period of the disease, owing to
which it is possible to forestall the disease, and to immunise
the body by the inoculations before its onset. If, unfor-
tunately, the infective material should be very virulent,
and the incubation period thereby reduced to the lower
limit, it may be impossible to do this before the onset
of the disease, and the same is the case if the commence-
ment of the treatment be delayed. Pasteur's system of
inoculation is useless when the disease has declared itself.

By vaccinating animals by the Pasteur method by a
long series of injections, and with the most virulent material,
the blood-serum acquires " anti- " properties, and this
" anti-rabic " serum is said to be of service in the treatment
of the declared disease.

Variations from typical rabies have been described both in
animals and in man under such names as " chronic rabies," " abor-
tive rabies," etc. Harvey, Carter, and Acton * describe a spon-
taneous disease in dogs due to a general infection with B. pyocyaneus.
which closely simulates rabies. By subdural inoculation the disease

1 Veterinary Record, July 22, 1911, p. 57.

542 A MANUAL OF BACTERIOLOGY

is reproduced in rabbits, with paresis of the hind legs and death in
from sixteen to twenty-one days. The Negri bodies are absent,
the course of the disease differs somewhat from rabies, and the
B. pyocyaneus can be isolated from the brain and blood.

Diagnosis of Rabies

In a case of suspected rabies in a dog the animal should not be
killed immediately, but should be kept under observation until
it dies, or for three or four weeks, and then killed.

1. Moderately thin smears on slides are made from (a) the cortex
in the region of the fissure of Eolando (the crucial sulcus in the
dog), (b) the hippocampus major, (c) the cerebellum. They are
dried in the air, fixed for five minutes in methyl alcohol, and then
stained in weak Giemsa (1 drop stain, 1 c.c. distilled water ; with
1 drop of 1 per cent, potassium carbonate solution to every 10 c.c.
of the dilute stain) for three hours. The stained films are then
washed in running tap-water for one to three minutes, dried with
filter-paper, and examined for the Negri bodies.

Or the moist films may be fixed in methyl alcohol, and without
drying stained for one minute in a mixture of 10 c.c. distilled water,
3 drops of a saturated alcoholic solution of basic fuchsin, and 2 c.c.
of Loffler's methylene-blue. Eosin-methylene-blue mixtures may
also be used.

The cytoplasm of the bodies stains orange, pink, red, or magenta,
the central nuclei are granular, and appear bluish or purplish.

Luzzani considers that the Negri bodies can generally be well
seen in teased up fresh material unstained. It is stated that structures
resembling the Negri bodies may be present in the brain after death
from snake-bite. s <

2. If the Negri bodies cannot be detected, inoculation should
be performed. The brain should be removed as soon as possible,
and if it cannot be manipulated immediately, should be placed
in sterile glycerin. From the middle of the floor of the fourth
ventricle a small piece about the size of a pea is removed ; this is
triturated and thoroughly emulsified in a sterile watch-glass by
means of a sterile glass rod with a bulbous end, a little sterile broth
being used to make the emulsion, and sufficient being added to
measure about 10 c.c. The hair on the head of a good-sized rabbit
is cut close, the animal is anaesthetised with ether, the skin on the
scalp reflected and a trephine hole made through the skull. The
centre of the trephine hole should be in the middle line, and on

INFANTILE PARALYSIS 543

the line drawn between the posterior corners of the eyes, the
diameter of the trephine being about ^ inch. A little of the
emulsion is drawn up in a small syringe, having a fine needle, and
two or three drops are injected beneath the dura mater. The
operation is carried out with antiseptic precautions, the wound
closed, and a little wool and collodion dressing applied.

If the material injected be from a rabid animal, the first symptoms
will be noticed in from ten to fourteen days. The inoculated animal
loses control over its hind legs and throws them about peculiarly
when running. This increases, and in another day or so the
animal is apt to fall when running, and in another day or two the
hinder extremities become paralytic, and the animal is unable to
move, and dies shortly. The onset of symptoms is hardly ever
delayed beyond twenty-one days.

Van Gehuchteri's method. — The ganglion is placed in absolute
alcohol for twelve hours, the alcohol being changed once ; it is then
embedded, and sections are cut. These are stained for five minutes
in Nissl's methylene-blue and mounted. Or the material may be
fixed in 10 per cent, formalin before staining. The capsular changes
are best shown by staining with haematoxylin and eosin.

Babes' method. — A piece of the medulla or cord is hardened in
alcohol and stained with anilin red, and sections are prepared.

Infantile Paralysis 1

Infantile paralysis or acute anterior poliomyelitis occurs
sporadically and also in epidemics.

Various organisms have been described in this disease,
but recent researches, particularly by Levaditi, Land-
steiner, and Flexner, have proved that the virus is a
filter-passer.

Injection of emulsions of the affected cord into the
brain, spinal cord, peritoneal cavity, and blood-stream
of monkeys reproduces the disease with the same clinical
and pathological features as in man. The disease can
be carried on from monkey to monkey by inoculation,

1 See Levaditi, Journ. Roy. Inst. of Public Health, vol. xix, 1911, pp. 1
and 65 (Bibliog.) : Flexner and others, Journ. Amer. Med. Assoc.,
1910-1911.

544 A MANUAL OF BACTERIOLOGY

but does not seem to be transmissible to other animals.
The salivary and some of the lymphatic glands contain
the virus.

Flexner has observed a case of spontaneous infection
in the monkey, and found that the naso-pharyngeal
mucosawas infective, so that this is probably the channel of
infection in man. Flies belonging to the genus Stomoxys
are stated to be capable of transmitting infection. Human
cerebro-spinal fluid was not found infective in some
instances, but monkey cerebro-spinal fluid is infective
(infectivity in this case may depend on the stage of the
disease).

Human ascitic fluid inoculated with the filtered fluid
from emulsions of cord became turbid, but no organism
could be detected microscopically, and the culture can
be carried on from tube to tube (Flexner and Noguchi).
Monkeys which have recovered from an attack are refrac-
tory to inoculation. A certain degree of active immunity
may be established by subcutaneous injection of the virus.
The serum of immunised and recovered animals possesses
considerable neutralising power for the virus. Attempts
are now being made to prepare a curative serum.

Some cases of the acute ascending paralysis of Landry
may be forms of this disease (see also p. 535).

Buzzard, from a case of the latter disease, isolated a
coccus which induced a rapidly spreading palsy on sub-
dural inoculation into rabbits.

Typhus Fever l

Many organisms have been described in this disease.
Nicolle, in Tunis, has found that typhus fever of man
is communicable to the chimpanzee by inoculation and
from the anthropoid to the Chinese bonnet monkey.

1 See Hewlett, Practitioner, July 1911, p. 112 (Refs.).

TYPHUS FEVER 545

Nicolle and Conseil have found it possible directly to infect
the Macacus sinicus and rhesus monkeys from human
cases.

Nicolle ascertained that the blood is virulent from the
commencement of infection and continues so until the
day after the temperature becomes normal. The dog and
rat are quite refractory. The disease appears to be trans-
mitted by the body-louse (P. vestimenti), not by the flea,
as suggested by Matthew Hay.

The blood from a mild case does not produce immunity
on injection, nor does a mild attack itself induce any
appreciable immunity. On the other hand a severe
infection induces considerable immunity. Nicolle and
Jseggy have not detected any microbe in affected persons
or animals. As the polymorphonuclear leucocytes suffer
considerably during the attack, undergoing fragmentation
of the nucleus and necrosis, it is suggested that the micro-
organism may be intra-leucocytic.

Other researches have been carried out in America on
the typhus of Mexico, known locally as " Tabardillo."
Anderson and Goldberger first showed that the Macacus
rhesus monkey could be directly infected with Mexican
typhus. Ricketts and Wilder have confirmed this, and
find that typhus blood is not infective if passed through a
Berkefeld filter, indicating that the micro-organism is of
appreciable size. They also find that the disease is con-
veyed by the body-louse, and, moreover, that the infection
is hereditary in the louse, the second generation of lice
derived from infected lice apparently being still infective.
Neither bugs nor fleas conveyed the disease.

In the blood of typhus patients Ricketts and Wilder
detected a small bacillus, measuring 2 /UL in length by O6 yu,
in breadth, tending to stain at the poles and belonging to
the group of the hsemorrhagic septicsemic bacteria. It is
cot numerous, and is found from the seventh to the twelfth

35

546 A MANUAL OF BACTERIOLOGY

day of the disease. It is also found in infected lice, but
could not be cultivated. A similar micro-organism was
also observed in Mexican typhus blood by Gavino and
Girard, and by Campbell, and the latter also finds that
the blood is not infective if passed through a Chamberland
F filter.

Ricketts and Wilder also discuss the relationship between
typhus fever and Rocky Mountain spotted fever.1 Some
years ago Wilson and Chowning made observations on a
typhus-like fever occurring in limited tracts of country
near the Rocky Mountains and ascribed it to a Piroplasma.
Subsequent research, however, failed to confirm this,
though the disease appears to be conveyed by a tick, and
not by fleas, lice, etc. There are clinical differences
between typhus and Rocky Mountain spotted fever ; more-
over, the guinea-pig is susceptible to the spotted fever
but not to typhus, and a monkey immunised to typhus is
susceptible to spotted fever. Ricketts believes that the
spotted fever is due to a bacillus which can be found
in the ovary of the tick and is agglutinated by the serum
in dilutions of 1-500.

Cathoire has made observations on complement fixation
in typhus. Using as an antigen an alcoholic extract of
typhus spleen, marked complement fixation was obtained
with the serum of typhus cases.

Yellow Fever

As far back as 1889 Sternberg described a bacillus —
" Bacillus X " — in yellow fever, a facultative anaerobic
organism, very pathogenic to rabbits. In 1897 Sanarelli 2
described his Bacillus ictero'ides, which later investigation

1 The name is an unfortunate one, for this disease is quite distinct
from " spotted fever " — epidemic cere bro -spinal meningitis.

2 Ann. de VInst. Pasteur, xi, 1897, pp. 443, 673, and 753.

YELLOW FEVER 547

has proved to be an organism belonging to the Gartner
group (see p. 371).

Reed and Carroll 1 critically examined the B. ictero'ides
and its relation to yellow fever. Their conclusions were
that the Bacillus X belongs to the colon group, the B.
ictero'ides to the Gartner group, that the B. ictero'ides and
hog- cholera bacillus produce the same lesions in animals
and mutually protect against each other, that the B.
ictero'ides causes in swine all the symptoms and lesions of
hog cholera, and that the blood of hog cholera agglutinates
the B. ictero'ides in a much more marked degree than does
the blood of yellow fever.

Reed, Carroll, and Agramonte,2 having thus shown the
etiological position of the B. ictero'ides to be untenable,
directed their attention to the transference of yellow fever
through the agency of mosquitoes. Finlay, of Havanah,
suggested many years ago that yellow fever might be
propagated through the intermediary of a mosquito —
Stegomyia calopus (fasciata) — and with this species these
investigators worked. They allowed mosquitoes to bite
yellow-fever patients at various stages of the disease, and
the infected mosquitoes were subsequently allowed to bite
eleven individuals, two of whom contracted yellow fever.
It is true this is not a very convincing experiment, but
it is to be noted that during the period of fifty-seven days
among a population of 1400 non- immune Americans there
were only three cases of yellow fever, and that two of these
had been bitten by contaminated mosquitoes within five
days of the commencement of their attacks. The matter
was put to the further test of experiment in the following
manner.3 Under the same observers a camp was estab-
lished with several tents each occupied by one to three

1 Journ. Exper. Med., vol. v, pt. iii, p. 215.

2 Philad. Med. Journ., October 27, 1900, p. 790.

3 Journ. Amer. Med. Assoc., February 16, 1901, p. 431.

548 A MANUAL OF BACTERIOLOGY

non-immune individuals, and precautions were taken to
prevent the introduction of yellow fever from outside.
Five individuals were bitten by infected mosquitoes, and
four out of the five contracted yellow fever, no other
occupants of the camp being attacked by the disease.
Subsequently several non-immune individuals were exposed
to yellow fever infection from soiled linen, yellow-fever
discharges, etc., in a mosquito-proof hut from which
mosquitoes were excluded, with entirely negative results.
These experiments prove, therefore, that yellow fever is
conveyed by mosquitoes only, and further work by
Americans and Cubans, and by French and Brazilian
Commissions, has entirely confirmed these researches
and conclusions. It has been found that to convey infec-
tion, it is necessary for the mosquitoes to bite the patient
during the first three or four days of the illness, but they
do not become infective until about the twelfth day after
feeding, and then retain their infectivity indefinitely.
All these facts point to a protozoon as being the causative
organism, but none has been found with certainty.

The Americans have shown that the blood-serum after
filtration through a porcelain filter is still infective ; the
organism, therefore, is probably ultra-microscopic, at least
at one stage. Seidelin 1 describes extremely small rounded
bodies with a minute chromatin point and feebly staining
protoplasm, without pigment, in the blood corpuscles.
Somewhat similar, but larger, bodies may also be present
in the organs and free in the plasma. Macfie and Johnston 2
state that they have found elements similar to those
described by Seidelin in the red corpuscles in practically
every case of yellow fever examined.

1 Journ. Pathol. and Bacterial, vol. xv, 1911, p. 282.

2 Proc. Roy. Soc. Med., vii, No. 3, 1914 (Med. Sec.), p. 49.

DENGUE AND PHLEBOTOMUS FEVER 549

Dengue

No organism, bacterium or protozoon, has been demon-
strated in this disease. The intra-venous inoculation of
filtered dengue blood into healthy individuals is followed
by an attack ; the organism is therefore probably ultra-
microscopic. The disease can be transmitted by a mos-
quito, Culex fatigens, and this is probably the common
mode of infection.1

Phlebotomus Fever

A fever of short duration (three days) occurs in South
Austria, the malady being somewhat like dengue. It is
known locally as " pappataci," and an apparently identical
disease has been described by Birt 2 in Malta under the
name of " phlebotomus fever." Investigation has shown
that this disease is conveyed by the bite of a dipterous fly,
the sand-fly (Phlebotomus pappatasii). " Canary fever,"
" Shanghai fever," " Chitral fever," and the seven days
continued and " sand-fly " fevers of India are probably
of the same nature. The virus in phlebotomus fever passes
through a Berkefeld filter.

Further research must decide whether these and dengue
are distinct diseases or whether they are all manifestations
of dengue.

Variola and Vaccinia

The specific contagia of these two diseases appear
to be filter-passers.

Variola is inoculable on man the calf and the monkey,
vaccinia on the rabbit in addition.

1 Ashburn and Craig, Philippine Journ. of Science, vol. ii, 1907, p. 93.

2 Journ. Roy. Army Hed. Corps, August 1910.

550 A MANUAL OF BACTEKIOLOGY

A large number of observations have been made with
vaccine lymph, but no distinctive bacterium has been
obtained except by Klein and Copeman. Usually the
ordinary pyogenic organisms and many saprophytic forms
can alone be isolated. Klein observed the presence of a
bacillus in vaccinia, which was subsequently more fully
studied by Copeman.1 It was found in vaccine vesicles
at an early stage, but at maturation could no longer be
detected. It is a very fine bacillus, and these observers
were unable to cultivate it. Subsequently Copeman
found a similar organism in variola, and succeeding in
cultivating the bacillus from both sources in eggs, and from
such egg-cultures was able to inoculate calves. Klein,2 by
storing variola crusts in 50 per cent, glycerin and so getting
rid of the saprophytic forms, has cultivated an organism
which he terms the Bacillus albus variolce. Morphologically
it closely resembles the bacillus observed in vaccine lymph ;
it forms small white, opaque, coherent colonies on agar,
but grows very feebly on gelatin. Involution forms occur,
and it seems to belong to the group of diphtheria and xerosis
bacilli. On inoculation into calves some approach to,
but not typical, vaccinia was produced. Moreover, the
inoculated calves were not immune to subsequent vaccina-
tion. Copeman 3 inoculated glycerinated vaccine lymph
in which the extraneous organisms had died out into
collodion capsules filled with beef broth and inserted them
in the peritoneal cavity of rabbits, and observed zooglcea
masses made up of bodies resembling spores which he
regards as the resting stage of the specific microbe.

De Korte finds that the vesicles, both in variola and in
vaccinia, are sterile before maturation, and regards the bac-
terial forms that have been isolated as secondary infections.

1 Milroy Lectures on Vaccination, 1898.

2 Rep. Med. Off. Loc. Gov. Board for 1896-97, p. 267.

3 Brit. Med. Journ., 1901, vol. i, p. 450.

VARIOLA AND VARICELLA 551

The failure to isolate a bacterial form has induced many
observers to seek for a parasitic protozoon in variola and
vaccinia. L. Pfeiffer in 1887 observed roundish or ovoid
bodies in the lymph in both diseases, which he regarded
as sporozoa. Guarnieri found small bodies, about half
the size of the nucleus, in the epithelial cells of the skin
in the prepustular stage of variola (Cytoryctes variola).
Small shining amoeboid bodies were also noticed in the
epithelial cells of the corneae of guinea-pigs inoculated
with vaccine lymph. L. Pfeiffer confirmed Guarnieri's
work, and also described these amcebiform parasites in
the blood in variola and vaccinia, and of vaccinated calves.
J. Clarke, and RutTer and Plimmer in this country described
somewhat similar appearances. Ruffer and Plimmer
describe the supposed protozoon as a small round body,
about 3 JUL in diameter, lying within a clear vacuole in
the protoplasm of the epithelial cell.

Councilman, Magarth, Brinkerkoff, Tyzzer, and Calkins l
in America have found the Guarnieri body in variola and
vaccinia in man and animals, and regard it as a protozoon
and the causal agent of these diseases.

Ogata found bodies which he regards as parasitic pro-
tozoa and the causative agent of the disease in variolous
and vaccine lymph. Reed likewise observed small granular
amoeboid bodies having a diameter of about one-third
that of a red blood- corpuscle, similar apparently to those
described by L. Pfeiffer, in the blood of vaccinated children
and monkeys, but also observed them — and this is impor-
tant— occasionally in the blood of normal children and
monkeys.

Funck, Roger and Weil, and Calmette 2 have also
observed various bodies and retractile granules in lymph.

1 Journ. Med. Research, vol. xi, 1904, p. 173 -^Philippine Journ. of
Science, vol. i, 1906, p. 239. $& ; 1*\

2 Ann. de Vlnst. Pasteur, xv,jl901. No. 3, p. 161,

552 A MANUAL OF BACTERIOLOGY

The monkey and rabbit are both susceptible to vaccinia ;
in the latter animal the pustules are mature on the third
day and immunity is acquired by the sixth day.

Ferroni and Massari state that appearances similar to
those described by Guarnieri can be obtained in cornese
inflamed by croton oil or Indian ink, and therefore believe
that the so-called parasites are derived from the nuclei
or from emigrated leucocytes. Salmon considers that the
so-called parasites in vaccinia and variola are more or less
condensed balls of chromatin of extra- epithelial origin
derived from the migratory polynuclear leucocytes.
According to von Prowazek these cell inclusions (the
Guarnieri bodies, etc.) in this and other conditions (e.q.

'I \ ij

scarlatina) are not parasites, but consist of plastin and
nuclease, and are derived from the cells in which they occur.

De Korte 1 has observed in the variolous and vaccine
vesicles before maturation large amoeboid bodies (10 /u),
which he believes to be protozoa (Sporidium vaccinate).
In vaccine lymph refractile motile granules occur in
abundance, believed by De Korte to be spores.

Fornet 2 by treating variola or vaccine lymph with ether
finds a stage when all the bacteria are killed but the specific
virus is uninjured. By inoculating this etherised lymph
into nutrient broth and keeping at 37° C., the broth culture
inoculated in man produces typical vesicles even after
two months incubation, and moreover the culture can
be carried on from tube to tube. In the broth, minute
rounded bodies can be detected which may be the specific
micro-organism.

The relationship of vaccinia to variola has been a very
vexed question. With few exceptions (Ceely, Hime,
Simpson, Klein, King, Copeman) attempts to inoculate

1 Trans. Path. Soc. Lond., vol. Ivi, 1905, p. 172.

2 Trans. XVIIih Internal. Cong. Med. Lond., 1913, Sect, iv, pt. ii,
p. 119.

VARIOLA AND VARICELLA 553

variola on the calf have failed. In the successful cases
the lymph obtained from the calf has, on inoculation upon
children, produced typical vaccinia without any untoward
results. The positive results obtained by the inoculation
of variolous material being so few, a doubt arises whether
in these cases there may not have been some fallacy, such
as accidental contamination with vaccinia. Simpson,
however, performed his experiments within the precincts
of a smallpox hospital and away from possible vaccine
infection, and Copeman * found that variola may be readily
inoculated upon monkeys, and after several passages
through these animals is easily inoculable upon the calf.
He suggests, therefore, that vaccinia in the calf was origin-
ally due to infection with inoculated smallpox, so prevalent
at the time of Jenner's discovery. A somewhat parallel
instance of the attenuation of a virus by passage through
another animal is recorded by Sticker and Marx in the
case of birdpox, which produces an extensive smallpox-
like eruption in fowls and pigeons. In fowls and in pigeons
the virus retains its pathogenic properties for each bird
unaltered for any number of inoculations, but the pigeon
strain, after a few inoculations into fowls, completely loses
its virulence for the pigeon. There seems little doubt,
therefore, that vaccinia is modified variola, and the rationale
of vaccination rests upon a scientific basis.

The preparation of vaccine lymph is fully described by Blaxall.2
Calves are vaccinated with lymph under aseptic precautions, and
five days later the contents of the vesicles are scraped off, the pulp
is triturated in a machine, and is then placed in six times its weight
of sterilised 50 per cent, pure glycerin in distilled water, and stored
for about a month in test-tubes, until agar cultivations show that
extraneous bacteria have died out, when it is issued for use. It
remains very active for fifty to sixty days, after which it begins to
deteriorate.

1 Brit. Med. Journ., 1901, vol. i, p. 1134, and 1901, vol. ii, p. 1736.

2 Rep. Med. Off. Loc. Gov. Board for 1898-99, p. 35.

554 A MANUAL OF BACTERIOLOGY

Green * rapidly prepares vaccine lymph by killing off the extra-
neous organisms with chloroform vapour.

Blaxall 2 has more recently used oil of cloves as a sterilising agent
in the preparation of calf lymph.

Malignant Disease

The analogies between carcinoma and sarcoma and many infec-
tive diseases have led investigators to search for micro-organisms
in these conditions.

Bacteria have been repeatedly looked for, but Shattock was
unable to isolate any bacterial form from malignant disease. Doyen
isolated a micrococcus (M. neoformans, p. 232), but, though fre-
quently present, it is not causative.

A great impetus was given to the study of parasites in malignant
disease by the publication of a paper by Russell. He observed,
by certain methods of staining, small corpuscles within the epithelial
cells. They were spherical in shape, 4 to 10 /z in diameter, occurring
singly or in groups, were apparently homogeneous, and surrounded
by a capsule. Russell regarded these structures as belonging to
the " sprouting fungi " (Blastomycetes), and they have since been
known by the name of " fuchsin bodies " or " Russell's corpuscles."

Subsequently structures were observed within the epithelial
cells of carcinoma which were regarded by many investigators as
parasitic protozoa.3 These structures are round or ovoid, 2 ^ to
10 fj, in diameter, with a very distinct outline, as though encapsuled,
and clear refractile contents in which is a smaller body of variable
size analogous to a nucleus (Fig. 66, a). Occasionally the refractile
contents present a radial striation or a granulation.

These bodies are usually single, but may number as many as
eight or ten, and sometimes they invade the epithelial nucleus.
The Ruffer's or Plimmer's body, however, is a structure probably
analogous to the archoplastic vesicle of the cells of reproductive
tissue (Fig. 66, 6). Save for the presence of these structures,
there is no proof that protozoa are present in, or are the cause of,
carcinoma.

Another hypothesis of the nature of malignant disease is that
it is due to a blastomycetic infection (see p. 462).

Washbourn and others have observed infective venereal tumours

1 Rep. Med. Off. Loc. Gov. Board for 1900-01, p. 639.

2 Ibid. 191 1-12, p. 361.

3 See Ruffer and Walker, Journ. Path, and Bact., vol. i, 1893, p. 395.

MALIGNANT DISEASE 555

in dogs. These have been stated to be sarcomata, but are probably
granulomata.

Malignant disease occurs in all classes of vertebrates, and is
generally inoculable on an animal of the same species as that from
which it is derived, but not on other animals. The carcinoma of
mice has been the subject of much investigation of late. In the
writer's opinion, the trend of recent research is to show that malig-
nant disease is not due to a micro-parasite, but is derived from the

FIG. 66. — a, Buffer's or Plimmer's body in a cancer-cell ;
b, the archoplastic vesicle in spermatid of mouse. (After
Farmer, Moore, and Walker.)

irresponsible division of cells of the normal or of embryonic tissues.1
If there be a parasite, in all probability it is intra-cellular, like the
organism of plant cancer (Bacterium tumefaciens) described by
Erwin Smith.2

The molluscum bodies have likewise been regarded as parasitic
(coccidial) in nature, but with them also inoculation and cultivation
experiments have failed. The virus is stated to be a filter-passer,
as is also the case with bird molluscum.

Certain malignant-like tumours of birds are also filter -passers,
e.g. chicken sarcoma.

1 For further information consult Pathology, General and Special,
ed. 3, R. T. Hewlett (Churchill, 1912).

2 Trans. XVIIth Internal. Cong. Med. Land., 1913, Sect, iii, pt. ii,
p. 281.

CHAPTER XX

SOME DISEASES NOT PREVIOUSLY REFERRED TO, WITH
A DISCUSSION OF THEIR CAUSATION — MICRO-
ORGANISMS OF SKIN AND MUCOUS -MEMBRANES

APPENDICITIS. — The following Table * shows the usual kinds and
relative frequency of the infections in appendicitis :

Micro-organism.

Acute
appendicitis.

Chronic
appendicitis.

Bacillus coli in pure culture

70 per cent.

90 per cent.

,, with staphylococci

15

6

„ „ streptococci .

7

Very rare.

Staphylococci alone

4

1 per cent.

Streptococci ,,

Very rare.

Very rare.

Other organisms or combinations

4 per cent.

3 per cent.

It is not improbable that in a still greater percentage of cases
a mixture of organisms is present at first, the Bacillus coli subse-
quently crowding out the other forms. The Bacillus proteus,
B. pyocyaneus, and B. Welchii also occasionally occur.

Castellani 2 describes a bacillus, pathogenic to guinea-pigs,
isolated from a case of gangrenous appendicitis. Morphologically
it resembled the Shiga-Kruse dysentery bacillus, and was non-
motile, produced acid and gas in glucose and maltose and curdled
milk, but did not ferment mannite, lactose, and sucrose.

BEBI-BEBI. — Various observers have attempted to cultivate a
micro-organism in this disease. Cocci have been described by
Pekelharing and Winkler, Hunter, Okata and Kokubo, a sporing
bacillus by Rost, and Hamilton Wright suggests that the disease
is due to an intoxication, the result of a gastro- duodenal infection

1 Battle and Corner, Diseases of the Vermiform Appendix, 1904.

2 Brit. Med. Journ., 1907, vol. i, p. 1513.

556

CONJUNCTIVITIS 557

with a large Gram-positive bacillus (unisolated). Daniels suggested
that the epidemiology of the disease is best explained on the hypo-
thesis of a protozoan infection conveyed by lice. The writer and
De Korte * also suggest a protozoan infection, the organism perhaps
being eliminated in the urine.

Other views are that beri-beri may be a peripheral neuritis due
to arsenical poisoning, or that it is caused by the absence of certain
nutritive elements from polished rice. The evidence in favour of
the latter view seems to be accumulating, and it has been found
that essential nutritive constituents (vitamines ?) are present in the
husk of rice which is removed in polishing.

BRONCHITIS. — Ritchie 2 concludes that acute bronchitis is an
infective disease, but is not due to any one specific organism, the
most important causal bacteria being the S. pneumonice and strep-
tococci. In every case of acute bronchitis numerous pathogenic
bacteria are present in the bronchi, which are usually sterile in
health. The commonest organisms are B. pneumonice, B. influenzce,
and M. catarrhalis. Spirochaetes are present in some forms of
tropical bronchitis ; in others Castellani has described oidium-
like and yeast-like organisms.

CHANCRE, SOFT. — An extremely small bacillus, first described
by Ducrey,3 has been found in the ulcers and buboes. It has not
been inoculated successfully on animals, but can be inoculated
from a chancre, experimentally, from man to man. The bacillus
does not stain by Gram's method, and can be cultivated on blood
agar, on which it forms shining greyish colonies 1 mm. in diameter,
or in guinea-pig blood.4

CONJUNCTIVITIS. — Conjunctivitis is of several varieties:

(a) Acute contagious conjunctivitis, due to the Koch-Weeks
bacillus. This is a slender, non-motile organism, 1-1-5 ^ in length,
occurring singly or in pairs, both free and within the pus-cells.
It is decolorised by Gram's method, and is difficult to cultivate,
growing best on a serum-agar mixture, on which it forms small,
punctiform transparent colonies. It is hardly pathogenic to
animals, but in man sets up a typical acute conjunctivitis.

(6) Chronic catarrhal conjunctivitis, due to the Morax-Axenfeld
diplo-bacillus. This organism is 2 /j, long by 1 /M broad, is not
stained by Gram's method, and can be cultivated on blood-serum
which is liquefied, or serum agar.

1 Journ. Trop. Med., October 1, 1907, p. 315.

2 Journ. Path, and Bact., vol. vii, No. 1, p. 1.

3 Comp. Rend. Congres Internal, de Dermatologie (Paris, 1889), p. 229.

4 Himmel, Ann. de Vlnst. Pasteur, xv, 1901, p. 928.

558 A MANUAL OF BACTERIOLOGY

(c) Gonorrhceal conjunctivitis.

(d) Diphtheritic conjunctivitis.

(e) Conjunctivitis of streptococcic origin.

(/) Conjunctivitis of pneumococcic origin. — Usually in children,
and accompanied with coryza and scanty muco-purulent discharge.

(g) Micrococci (aureus and albus) and B. coli may also occasionally
cause conjunctivitis.

DIARRHOEA (SUMMER) OF INFANTS. — Booker,1 in an elaborate
paper, came to the following conclusions : " No single micro-
organism is found to be the specific exciter of the summer diarrhoea
of infants, but the affection is generally to be attributed to the
activity of a number of varieties of bacteria, some of which belong
to well-known species, and are of ordinary occurrence and wide
distribution, the most important being a streptococcus and the
Proteus vulgaris."

Lesage obtained a bacillus from the " green diarrhoea " of infants
which he believed to be the cause of this complaint. It is a small,
motile, non -liquefy ing bacillus, producing on gelatin a whitish
expanded growth with crenated margins, and giving rise to a green
fluorescence in the medium. The B. pyocyaneus may be an
occasional cause.

In cases with blood and mucus in the stools, the B. dysenteric^
(Shiga-Kruse type) has been found to be present in America and
in this country. In London, Morgan has isolated in a number of
cases a bacillus which in its fermentation reactions is nearly allied
to the hog-cholera bacillus (see p. 372). Lewis 2 found that non-
liquefying and non-lactose-fermenting bacilli are more frequent in
the faeces of children suffering from diarrhoea than in normal children,
and believes that Morgan's bacillus has a causal relationship in many
cases. Alexander 3 also found Morgan's bacillus more frequent in
diarrhoea cases than in normal children.

Ralph Vincent ascribes the disease (which he terms " zymotic
enteritis ") to the ordinary organisms of putrefaction gaining access
to milk and multiplying and causing alterations therein.

The stinking motions of the diarrhoea of children have been
ascribed to the action of organisms belonging to the Proteus group,
particularly B. proteus (P. vulgaris, see p. 621), which occurs in
putrefying matter, sewage, and in the intestine. (This organism
may also cause abscesses and cystitis, and a form of meat poisoning

1 Johns Hopkins Hosp. Reps., vol. vi, 1897, p. 159 (Bibliog.).

2 Rep. Med. Loc. Gov. Board for 1911-12, p. 265, and ibid, for 1912-13,
p, 375.

3 Ibid. 1911-12, p. 288.

DYSENTERY 559

has been ascribed to its action.) Filtrates of cultures were found
by S. Martin to produce a fall of temperature, collapse, and diarrhoea
in rabbits.

CANINE DISTEMPER. — According to Galli- Valeric,1 this is caused
by a bacillus (B. caniculce) intermediate in character between the
coli-typhoid and hsemorrhagic septicaemic groups of organisms.
Torrey and Rahe 2 confirm Ferry and M'Gowan's observations on
a bacillus (B. bronchisepticus) present in distemper. It does not
ferment any sugars and litmus milk becomes markedly alkaline.

Evidence has also been brought forward that distemper is due
to a filter passer (Carre). Probably the term " distemper " may
include several different diseases.

DYSENTERY. — Dysentery must be regarded as a term applied
to a series of clinical symptoms associated with colitis which is
due to different specific agents. There are at least two forms of
the disease, one, the so-called tropical or endemic dysentery, met
with especially in the East, and characterised by chronicity, a ten-
dency to relapses, amenability to treatment with ipecacuanha, and
the occurrence of the single liver abscess as a sequela ; the other,
epidemic dysentery, met with in all parts of the world, particularly
in times of war and famine, not amenable to ipecacuanha, and not
followed by liver abscess. There are also probably other forms
occurring in small outbreaks or sporadically. Tropical dysentery
is due to the Amoeba coli, which is found abundantly in the stools,
especially in the acute stage, and also in the liver abscesses (see
p. 484).

In the epidemic dysentery of Japan and other parts of the world
a bacillus, or group of bacilli, has been isolated by Shiga, Flexner,
Strong, Kruse, and others. This is the B. dysenteries described
at p. 376.

Coli-form bacilli have been isolated from cases of dysentery.
Calmette in Tonkin isolated the B. pyocyaneus, and this organism
seems to have been the cause of a small outbreak in New York
State investigated by Lartigau.3 In Japan, Ogata isolated a fine
Gram-staining, liquefying bacillus which does not seem to have
been met with by subsequent observers. Spirochaetes have been
found in large numbers in a form of dysentery occurring in Bordeaux.

Vedder and Duval,4 as a result of the study of a number of cases

1 Centr. f. Bakt. (Ref.), xli, 1908, p. 563. See also M'Gowan, Journ.
Palhol. and Bacterial., vol. xv, 1911, p. 372 (Bibliog.) and xvi, p. 257.

2 Journ. Med. Research, xxvii, 1912, p. 291 (Bibliog.).

3 Journ. Exper. Med., vol. iii, No. 6, p. 595.

4 Ibid. vol. vi, 1902, No. 2, p. 181.

560 A MANUAL OF BACTERIOLOGY

of acute dysentery in the United States, conclude that the disease,
whether sporadic, " institutional," or epidemic, is due to the B.
dy sentence of Shiga.

The B. dysenterice (Shiga type) has been isolated by Eyre,
McWeeney, and others from cases of ulcerative colitis or asylums
dysentery in the British Isles (see pp. 376-379).

The Balantidium coli (p. 507) and certain parasitic worms may
also induce a dysenteric condition.

SKIN DISEASES : Acne. — In the acne pustules, the M. pyogenes
var. aureus, with or without var. albus, is almost invariably present,
and a staphylococcic vaccine generally acts extremely well. In
the comedoes a Gram-positive, Hofmann-like bacillus (B. acnes)
is present in considerable numbers, and may be the cause of the
comedo. This organism was cultivated by Fleming on a neutral
agar to which glycerin and oleic acid are added. Siidmersen and
Thompson * cultivate it on an acid ( + 40) serum-agar. The organism
is anaerobic, at least at first, and will grow in glucose-agar stabs.
In culture the organism is diphtheroid. A vaccine prepared with
it is of service in the comedo stage.

Eczema is produced by the action of the pyogenic cocci (M.
pyogenes, var. aureus and albus). Virulent cultures of these organ-
isms, with or freed from their toxins, seem, however, to produce
an impetigo rather than eczema. But the filtered cultures, i.e.
toxins, are harmful to the skin, and when applied to it for one or
two days by means of moist warm pads a typical papular or vesi-
cular eczema ensues. Probably in the human subject in addition
to the micro-organisms some peculiarity in the soil is necessary
for the disease to develop.2 In so-called seborrhceic eczema, a
non-liquefying micrococcus which forms butyric acid has been
isolated.

Impetigo. — The large vesiculo-bullous eruption of impetigo con-
tagiosa is caused by the Streptococcus pyogenes ; the small pustule
in the neighbourhood of hair-follicles, impetigo of Bockhart, is
caused by the M. pyogenes var. aureus. The B. diphtheria may
also cause an impetigo (p. 273).

Pemphigus. — A diplococcus has been isolated in acute pemphigus
by Demme, and in the chronic form by Dahnhardt. Bulloch and
Russell Wells, in this country, seem to have isolated an identical
organism, and the following description of it is taken from their
papers. Cocci 0-8 to 1-5 p in diameter, mostly arranged as diplo-
cocci, and staining by Gram's method. On surface agar the organ-

1 Journ. of Pathol. and Bacterial., vol. xiv, 1910, p. 224.

2 Whitfield, Practitioner, February 1904, p. 202.

MEASLES 561

ism forms a thick, white, shining growth. In stab agar the growth
has a " nail-shaped " appearance. The colonies on agar are at
first round, but later, in seven days, they throw out lateral pro-
jections and assume a rosette appearance. On gelatin the growth
is slow and slight, with some, but not marked, liquefaction. On
blood-serum the growth resembles that on agar. On potato a
whitish, semi-transparent film forms. Milk is curdled. In brotlj
it causes a general turbidity, with a whitish sediment, and some-
times a pellicle, which soon sinks. Guinea-pigs and mice inoculated
or vaccinated with the organism died in four to eight days, fine
haemorrhage, occurring in the lungs, and the cocci being obtained
from the blood. No bullae appeared on the skin. The B. pyo-
cyaneus may cause dermatitis and bullous eruptions (see p. 239).

The pyogenic cocci or their toxins may produce various bullous
eruptions, e.g. pemphigus neonatorum and contagiosus and hydroa
gestationis.1

Herpes zoster. — Pfeffer observed bodies in the cells of the vesicles
which he believed to be protozoa. Gilchrist, however, regards
these merely as altered nuclei.

FOOT AND MOUTH DISEASE. — Various organisms have been
described in this disease, but a German commission comprising
Loffler and Abel 2 stated that they were unable to prove its etio-
logical significance. Lofiler and Frosch have determined that the
organism must be a very minute one, as it passes through the
smallest- pored porcelain filter.

MASTOID DISEASE. — See " Otitis Media."

MEASLES. — Doehle and Behla described small flagellated bodies
which they believed to be protozoa in this disease. Canon and
Pielicke found small bacilli in the blood, which Tchaikovsky con-
firmed. They are motile, do not stain by Gram's method, and
can be cultivated on agar and serum, on which they form delicate
colonies. Czajkowski has found a similar organism. Lesage 3
cultivated a small micrococcus from the nasal mucus and blood,
which produced a fatal haemorrhagic septicaemia in animals. The
influenza bacillus is present in many cases. The organism is
probably a filter-passer.

MENINGITIS may be caused by S. pneumonice (60 per cent, of
acute cases), D. intracellularis, Still's diplococcus, B. tuberculosis,
gonococcus, and micrococci and streptococci.

MUMPS (EPIDEMIC PAROTITIS). — Mecray and Walsh isolated from

1 Brit. Med. Journ., 1902, vol. i, p. 73.

2 Centr. f. Bakt., xxiii, 1898, March.

3 Compt. Rend. Soc. Biol, 1900, p. 203.

36

562 A MANUAL OF BACTERIOLOGY

the parotid and blood in some cases of mumps a coccus resembling
that described by Laveran and Catrin. It occurs chiefly as a
diplococcus, but also in large groups. The colonies form circular,
white, shining points, with slow growth and gradual liquefaction.
On potato a white growth occurs ; on blood-serum a plentiful cream-
coloured growth ; and in litmus milk production of acid with
coagulation.

NOMA AND CANCBUM OBIS. — Grawitz in 1890 observed bacilli
in the affected tissues in this disease, others fusiform bacilli with
or without other organisms ; Comba considered that there was
probably no specific organism ; Durante found the M. pyogenes,
Var. aureus, with B. proteus, and Ravenna the same micrococcus
with the typhoid bacillus. Diphtheroid bacilli have also been
isolated. Weaver and Tunnicliff l in a case of cancrum oris observed
the presence of fusiform bacilli and spirilla. Hellesen 2 isolated a
diplococcus from a case of noma. The organism is not unlike the
pneumococcus, but possesses no capsule, is Gram-positive, gives
a general turbidity in broth with acidity, forms no gas from glucose,
curdles milk with acid production, and forms punctate, whitish-
grey, translucent colonies on surface agar. On inoculation into
animals a specific necrosis was produced.

Bishop and Ryan, in two out of three cases, isolated an organism
which culturally and morphologically resembled the diphtheria
bacillus, but which only produced some local inflammation on
inoculation into guinea-pigs. In the third case the M. pyogenes,
var. aureus, and the Streptococcus pyogenes were isolated. Guizzetti,
and Freymuth and Petruschky have isolated the Klebs-LofHer
bacillus in noma.

OpPLEB-BoAS BACILLUS. — Met with in the stomach, particularly
in cases of carcinoma, and its detection is suggestive of this con-
dition. The bacilli occur in masses, are long and filiform and non-
motile, and frequently join one another at an angle. They measure
usually 6-8 p. in length, but vary between 3 and 10 p.. The organism
has been cultivated, and is facultative anaerobe, non-sporing and
Gram-positive. It curdles milk and forms lactic acid from various
sugars.

OTITIS MEDIA. — The Streptococcus pneumonice is perhaps the
commonest organism met with ; next in frequency comes the
Streptococcus pyogenes, and then the pyogenic cocci. In scarlatinal
otitis media, Blaxall found the S. pyogenes to be always present,
and generally accompanied by other organisms, pyogenic cocci,

1 Journ. Infectious Diseases, vol. iv, 1907, p. 8 (Bibliog.).

2 See Lancet, 1908, vol. i, p. 955.

PELLAGRA 563

etc. In thirty-seven cases of mastoid disease Blake found the
following organisms, and remarks that as a rule the same were
found in the middle ear :

Streptococcus ....... 12

Staphylococcus ....... 5

Diplococcus (? pneumonice) ..... 6

Streptococcus and diplococcus .... 5

Streptococcus and Bacillus fetidus ( ? colon bacillus) 3

Streptococcus and Bacillus pyocyaneus ... 1

Streptococcus and diplococcus .... 1

Streptococcus, micrococcus, and diplococcus . . 2

In two of the cases no organisms could be isolated.

OZJENA (ATROPHIC RHINITIS). — Lowenberg described in this
disease encapsuled bacilli somewhat resembling the pneumo-bacillus
morphologically. Some Italian observers found bacilli apparently
identical with the diphtheria bacillus. Abel * described a bacillus
somewhat resembling the pneumo-bacillus. It is this organism
which produces the atrophy of the mucous membrane, but the
fetor is due to the decomposition of the secretions produced by
other organisms.

Perez 2 isolated an organism in ozsena (Cocco-bacillus fetidus
ozcence) which has the following characters : it is a short bacillus
with rounded ends, non-motile, does not stain by Gram's method,
does not liquefy gelatin, does not ferment lactose nor curdle milk,
but forms indole and ferments urea. Its cultures are foul-smelling,
and it is pathogenic for guinea-pigs, mice, rabbits, and pigeons.

PELLAGRA. — Many hypotheses have been propounded to account
for the causation of this disease, in which no micro-organism has
been detected with certainty. It formerly was supposed to be due
to the consumption of maize, which contains toxic substances.
Lombroso suggested that spoilt maize is the cause, toxic substances
being produced by Penicillium glaucum. Of parasitic theories,
Ceni and others suggest infection with Aspergillus fumigatus or
A. flavescens. Tizzoni attributes it to the pleomorphic, Strepto-
bacillus pellagrce (which may be a pleomorphic form of an
actinomycotic organism). Sambon on epidemiological data
believes that a protozoan parasite is the agent and is trans-
mitted by small biting flies of the genus Simulium. The sun's rays
have also been supposed to cause the affection. 3

1 Zeitschr. f. Hyg., xxi, p. 89.

2 Ann. de Vlnst. Pasteur, xiii, 1899, p. 937, and xv, 1901, p. 409.

3 See First Progress Rep. of the Thompson-McFadden Pellagra
Commission.

564

A MANUAL OF BACTERIOLOGY

T PERITONITIS. — Treves gives the following Table of the micro-
organisms found in peritonitis :

Frankel

Tavel and Tanz

Found alone

Found alone

Found in
association

Bacillus coli communis

11

15

16

Streptococcus
Staphylococcus
Pneumococcus

7
1
1

3
2

0

15
6

2

20

20

39

Dudgeon 1 believes the B. coli is frequently a secondary agent and
not the primary infection. He finds that the M. pyogenes, var.
albus, is very commonly present from the first, and may exert a
protective action by determining the occurrence of phagocytosis.

PSILOSIS OR SPRUE. — Carnegie Brown 2 considers this disease to
be due to an abnormal fermentation in the intestine brought about
by some organism, bacterial or protozoan, which has not yet been
isolated.

PUERPERAL FEVER. — This condition may be either a localised
infection with intoxication (sapraemia), or a localised infection with
general infection (puerperal septicaemia) ; in both the primary
seat of infection may be perinseal or vaginal lacerations, or the
contents of the uterus or the placental site. The infecting organisms
may be S. pyogenes, pure (20 per cent.), or with other organisms
(30 per cent.), occasionally the S. pneumonia?, B. coli, M. pyogenes,
var. albus, M. pyogenes, var. aureus, M. gonorrhoea?, B. Welchii,
and diphtheroid bacilli. These are rarely alone, but generally occur
with one or other of the organisms named. The B. diphtheria? may
exceptionally be met with.3

PURPURA. — Hsemorrhagic septicaemia may be caused by a number
of capsulated bacilli allied to the B. pneumonia? of Friedlander *
(see pp. 258, 404), as well as by streptococci and pyogenic cocci.
Paratyphoid infection may be accompanied with purpura.

1 Bacteriology of Peritonitis (Constable, 1905).

2 Sprue and its Treatment (Bale, Sons, & Danielsson, 1908).

3 See Foulerton, Practitioner, March, 1905, p. 387.

* See Howard, Journ. Exp. Med., vol. iv, 1899, p. 149 (Bibliog.).

ACUTE RHEUMATISM 565

PYORRHCEA ALVEOLARIS (Rigg's disease). — Goadby1 has found
the following organisms to be probably causative in this disease :
M. citreus granulatus, M. pyogenes, var. aureus, streptococci, M.
catarrhalis, and diphtheroid bacilli, and has used vaccine treatment
with success. Eyre and Payne 2 have found similar organisms,

RAT-BITE DISEASE. — A disease occasionally met with in England
but commoner in Japan, and consequent on the bite of a rat. It
is characterised by weekly bouts of severe fever lasting two or three
days. No organism has been detected.3

RHEUMATISM (ACUTE). — The opinion has gained ground of late
years that acute rheumatism is an infective disease. A number of
observers have isolated streptococci and micrococci in this disease,
and Singer regards the disease as merely an attenuated form of
pyaemia. Menzer considers that rheumatic fever is not due to any
one organism, but is a particular reaction in predisposed persons
to various microbes, especially streptococci. In 1897 Achalme
isolated an anaerobic anthrax-like bacillus from several cases.
This bacillus agrees in all its characters with the B. Welchii (enteri-
tidis sporogenes), and is believed by the writer 4 to be identical with
the latter ; it is probably a terminal infection or a contamination.
Poynton and Paine 5 in 1899 obtained from eight successive cases
a diplococcus (D. rheumaticus) which in broth develops into a
streptococcus. Injected intravenously into rabbits the diplococcus
frequently produces enlargement and inflammation of the joints
with effusion, and occasionally valvulitis and endocarditis. In
man the organism was demonstrated in the vegetations, pericardium,
tonsils, and rheumatic nodules, and has been isolated from the
blood, pericardial fluid, cardiac vegetations, and tonsils.

Andrewes and Horder found that two strains of the D. rheumaticus
corresponded with the S. fcecalis (p. 234).

Beattie 6 also obtained a streptococcus from the synovial mem-
brane of cases of acute rheumatism, which regularly produced
arthritis, and occasionally endocarditis, in rabbits. Goadby has
observed similar effects with a streptococcus obtained from the
mouth.

The manner in which typical acute rheumatism generally reacts

Proc. Eoy. Soc. Med., February 1910 (Odontological Section).
Ibid. December 1909.

See Hewlett and Rodman, Practitioner, July 1913, p. 86.
Trans. Path. Soc. Lond., vol. lii, pt. ii, 1901, p. 115.
Lancet, 1900, vol. ii, p. 861 et seq. ; Trans. Path. Soc. Lond., vol. Iv,
1904, p. 126.

6 Journ. Pathol. and Bacteriol., vol. xiv, 1910, p. 432.

566 A MANUAL OF BACTERIOLOGY

to salicylates suggests a protozoan organism, if an organism be the
cause.

RHEUMATOID ARTHRITIS (ARTHRITIS DEFORMANS). — This disease,
which is probably not a single one, may sometimes be caused by
an intestinal, urinary, pyorrhceic, or other toxaemia. Blaxall l
found in the synovial fluid, and occasionally in the blood, a minute
bacillus measuring 2 p. in length. It possessed marked polar
staining, was decolorised by Gram's method, and could only be
stained by prolonged (3-5 days) immersion in anilin methylene blue.
The organism can be cultivated on agar, on serum, and in broth.
In a clear broth, after three days, minute shining, yellowish particles
appear and increase in amount, giving rise on shaking the flask to
an appearance of " gold dust." Inoculation experiments on
animals failed.

Poynton and Paine 2 isolated a diplococcus (? a form of their
D. rheumaticus) from an osteo-arthritic joint, which produced
arthritis, with osteo-arthritic changes, when injected intravenously
into rabbits.

Crowe3 has found a micrococcus of peculiar type in the urine
in many cases. It may be isolated on the neutral-red egg medium
(p. 235), and a vaccine prepared with it seems to be of service in
treatment. The organism is allied to the M. epidermidis and has
been named by Crowe M . deformans.

RHINOSCLEROMA. — A bacillus has been described in this disease.
It is a short rod, with rounded ends, encapsuled, and frequently
linked in pairs. The organism is non-motile, does not stain by
Gram's method, and forms on gelatin a whitish growth without
liquefaction like that of Friedlander's pneumo -bacillus. Milk is not
coagulated. The organism is slightly pathogenic. It is doubtful
if it is the causal agent.

RINDERPEST. — Simpson, Koch and Eddington described bacilli
in this disease, but Nicolle and Adil-Bey have found that the virus
passes through a procelain filter, and the organism therefore is
probably ultra-microscopic.

TRACHOMA. — Various organisms have been observed in this
disease, e.g. a diplococcus by Sattler, gonococcal-like organisms
by Lindner and others (it is even suggested that the organism may
be an "involuted" gonococcus), the Koch-Weeks bacillus, the
Morax-Axenfeld diplobacillus and the pneumococcus. Minute
cell-inclusions, which may be demonstrated by the Giemsa method,

1 Lancet, 1896, vol. i, p. 1120 (Bibliog.).

2 Brit. Med. Journ., 1902, vol. i, p. 79.

3 Lancet, i, 1913, p. 1377, and ii, 1913, p. 1460.

UNDULANT FEVER 567

are present in the epithelial cells, regarded by Halberstaeder and
Prowazek as Chlamydozoa *• (p. 537). The disease is inoculable
on apes and the virus is stated to be a filter-passer. The causative
organism cannot yet be said to be known.

UNDULANT FEVEK.2 — Synonyms : Rock, Mediterranean or Malta
fever. A disease met with especially on the Mediterranean littoral,
but also in South Africa, India, China, the Philippines, and the
subtropical countries of America, and clinically often simulating
typhoid fever.

A minute micrococcus (M. melitensis}, first described by Bruce,
is the cause of the disease.

Microscopically, the organism from cultures occurs as a coccus,
single, in pairs, or in short chains ; it is easily stained by the ordinary
anilin dyes, but is Gram-negative. In hanging-drop cultures it
shows decided movement, which may be only an active Brownian
movement, but is perhaps a true motility inasmuch as Gordon has
described the presence of flagella (other observers have failed to
find them). The organism may be isolated from the spleen of a
cadaver.

On agar it grows as minute transparent colonies, which first
appear when inoculated from the spleen in 90 to 125 hours. In
thirty-six hours more the colonies become amber-coloured, and
later still in four to five days, they become opaque, of a slightly
orange colour, and round with granular margins. On gelatin a
whitish growth slowly forms without liquefaction, and in broth a
diffused cloudiness forms, with a white deposit and without film-
formation. Litmus milk becomes alkaline without curdling. Alkali
is also produced in glucose media, but galactose, maltose, and
saccharose are unchanged (see Table, p. 248). The distribution of
the M . melitensis in the body corresponds closely with that of the
B. typliosus ; thus it is abundant in the spleen, relatively scanty
in the blood, and is excreted in the urine.

The M. melitensis maintains its vitality outside the body in
the dry state in dust or on clothing for two to three months, in tap-
or sea-water for a month. The thermal death-point is about
55° C.

Inoculated into animals no result usually ensues ; in the monkey,
however, a febrile condition is produced, with enlarged spleen,
sometimes terminating in death, the course of the temperature
resembling that of the disease in man. By intra-cerebral inoculation

1 Berl. Idin. Woch. No. 24, 1909.

2 See Reports of the Mediterranean Fever Commission (Royal Society),
pts. i-vii, Harrison & Sons, 1904-1907.

568 A MANUAL OF BACTERIOLOGY

Durham found that the organism becomes pathogenic for the rabbit
and guinea-pig, otherwise it is without effect. For the diagnosis
of the disease the agglutination reaction is most valuable. It may
be carried out by the microscopic method, a forty-eight-hours'
broth culture being employed, the details of the process being the
•same as described at p. 191. Dilutions of 1 in 30, 1 in 50, and
1 in 100 should be prepared, as well as controls with normal serum,
for old laboratory strains sometimes agglutinate with normal serum
in dilution of 1 in 20 or 30 (see p. 192. Neglect of this precaution
ed Bentley to ascribe kala-azar to a Malta fever infection). The
organism being minute, it is necessary to use the yL-inch oil-immer-
sion, the £-inch with a high eyepiece and draw-tube extended, or
better, a J-inch dry objective. Bassett-Smith *• for agglutination
tests prefers the sedimentation method, for which an emulsion of
a forty-eight-hour old agar culture in physiological salt solution
should be employed. Three dilutions of the serum are made,
1 in 40, 1 in 100, and 1 in 400, and the tubes are placed in the blood-
heat incubator for two hours and the results noted. The tubes
should then be allowed to stand at laboratory temperature and the
results recorded after a further period of twelve hours. In some
two thousand observations, only once was a positive agglutination
obtained with a control serum. Complement-fixation tests may
also be employed and are satisfactory. Absence of agglutination
does not necessarily negative a diagnosis of undulant fever : in cases
of long duration it may be absent. Isolation of the organism from
the blood is another method that may be used, but similarly may
fail in long-standing cases.

The disease may be conveyed to monkeys by contact, by inhala-
tion of infected dust, and by feeding. Mosquitoes and other insects
do not seem to convey it.

The investigations of the Mediterranean Fever Commission have
shown that the main source of infection of man is by goat's milk.
Goats may be infected (and are largely so in endemic districts, e.g.
Malta and South Africa) without showing any symptoms, and
excrete the organism in large numbers in their milk. Since
goat's milk has been boiled the incidence of the disease in Malta
has fallen from 663 cases in 1905 to 7 cases in 1907 in the Army,
and in the Navy there were no cases in 1907 (Bruce).

Toxin, vaccine, and serum therapy. — The M. melitensis forms no
extra-cellular toxin, but Macfadyen obtained an endotoxin by
disintegration. Attempts to prepare an anti-serum have not been
successful. A vaccine prepared with cultures killed by heat (see

1 Journ. of Hyg., xii, 1912, p. 497.

SKIN AND CONJUNCTIVA 569

p. 219) has been used in the chronic form of the disease by Bassett-
Smith l and others with some amount of success (dose 100 to 500
millions)

An organism, the M. paramelitensis, has been found by Negre
and Raynaud in certain cases of undulant fever. In such cases,
the blood may not agglutinate the M. melitensis but does agglu-
tinate the M. 'paramelitensis. A case of this kind is recorded by
Bassett-Smith.2 As regards treatment, yeast or yeast-products
have been found of service in the neuritis of the disease. Vaccines
(100 to 500 millions) should be given every five to seven days : they
are contra-indicated when the pyrexia is continuous or remittent.

Micro-Organisms of the Skin and Mucous
Membranes

Skin. — In the normal clean skin micro-organisms are scattered
here and there in cracks of the horny layer and in crevices around
hairs and glands, but such skin is not swarming with microbes.
The S. pyogenes and M. pyogenes, var. aureus, albus, and citreus,
and the M. epidermidis (albus) of Welch, are the commonest (see
p. 229). Equally common on the skin and scalp is the scurf micro-
coccus isolated by Gordon (see Table, p. 230). Sarcinae, bacilli,
and moulds occur also. On the skin of the groin, scrotum, and
vulva the smegma bacillus occurs. From sweating feet various
organisms have been isolated, which on culture evolve a disagreeable
odour, among which is the Bacterium fetidum of Thin.

Conjunctive. — Some observers have stated that the conjunctiva
is generally sterile. A certain number of organisms are, however,
usually present, though they are not numerous, and if artificially
inoculated the excess is rapidly eliminated. The B. xerosis can
often be isolated.

Randolph 3 states that the normal conjunctiva always contains
organisms, the commonest species being the Micrococcus epidermidis
(albus) of Welch.

Lawson 4 found the normal conjunctiva to be sterile in 20 per cent,
of cases and pyogenic cocci to be rare, and, when present, non-
virulent.

1 Journ. of Hygiene, vol. vii, 1907, p. 115.

2 Journ. Trop. Hed. and Hygiene, February 15, 1913.

3 Archives of OphthalmoL, vol. xxvi, 1897, p. 379.

4 Trans. Jenner Inst. Prev. Med., vol. ii, p. 56 ; also Griffith,
Thompson Yates Lab. Rep., vol. iv. pt. i, 1901, p. 99.

570 A MANUAL OF BACTERIOLOGY

Nose. — In the anterior nares crusts and vibrissae micro-organisms
are present in great abundance, but, contrary to the usual opinion,
StClair Thomson and the writer l showed that the mucous mem-
brane of the interior of the nose is comparatively sterile, and when
organisms are present they are very scanty compared with the
number of organisms inspired.2 Moreover, organisms artificially
deposited were found to be rapidly disposed of. After two hours,
for example, prodigiosus inoculated on to the inferior turbinate
could not be detected by cultivation. Wurtz and Lermoyez
asserted that the nasal mucus is germicidal, but StClair Thomson
and the writer 3 were unable to confirm this, though it may have
an inhibitory action.

Air-passages. — Below the larynx under normal conditions the
air-passages are free from micro-organisms. Expired air is also
free from organisms, and the air from the naso-pharynx after passing
through the nasal cavities is deprived of the majority of its
organisms.4

Mouth. — Micro-organisms of all kinds are present in the buccal
cavity in the greatest abundance — leptothrix, bacilli, pyogenic
cocci, sarcinae, and spirilla are almost always to be found. The
Streptococcus pyogenes, M. pyogenes, var. aureus, and Streptococcus
pneumonice are frequently present. Certain organisms have their
normal habitat in the mouth, are difficult to cultivate, and are of
considerable importance in the production of dental caries.5 Well-
defined micrococci and streptococci also occur in the saliva (M.
salivarius, p 231, and 8. salivarius, p. 234). The normal saliva is
germicidal to some extent. (See also p. 460.)

Stomach and intestine. — Although a vast number of organisms
gain access to the stomach, a large number are destroyed by the
acid gastric juice. At the same time a considerable proportion
are able to survive — sarcinae, and lactic and butyric acid bacilli.
In normal nurslings the mouth and stomach contain few bacteria —
a few cocci, and some bacilli of the B. coli and B. lactis aerogenes
groups. The small intestine contains remarkably few organisms
of the same types. In the large intestine bacteria are extremely
numerous, particularly Gram-positive ones. These are mostly

1 Medico-CMrurg. Trans., vol. Ixxviii, 1895 (Bibliog.).

2 Other observers, however, have not altogether confirmed this.
See Iglauer, Laryngoscope, 1901, November, p. 363.

3 " The Fate of Micro-organisms in Inspired Air," Lancet, 1896
January 11.

4 Ibid.

5 See Goadby, Mycology of the Mouth.

STOMACH AND INTESTINE 571

slender, slightly curved bacilli of moderate size, the B. bifidus of
Tissier, which often has a bifid extremity, also a somewhat similar
organism, B. acidophilus of Moro, but capable of developing in an
acid medium, a few B. Welchii, and a diplococcus. The Gram-
negative forms are B. coli, B. lactis aerogenes, and cocci. In bottle-
fed children the same organisms occur, but the preponderating
organisms are Gram-negative of the B. coli type, with many cocci
and streptococci. In childhood and adolescence organisms of the
bifidus type become less numerous but putrefactive anaerobes
become more so, particularly B. Welchii and B. putrificus (coli) of
Bienstock ; the latter is a long, slender, Gram-positive bacillus with
large terminal spores. During adult life the putrefactive anaerobes
tend to become still more numerous, and the putrefactive decom-
positions they produce are regarded by Metchnikofi as standing in
causal relation to old age. In the healthy adult the stomach,
duodenum and jejunum contain relatively few organisms, from
the lower ileum to the rectum the intestinal contents are crowded
with bacteria, and the greatest number of anaerobic organisms occur
here and putrefactive changes are most in evidence.1 Kendall2
has described the presence of a bacillus (B. infantilis) in large
numbers in a condition of infantilism, associated, according to
Herter, with chronic intestinal infection. The organism is a Gram-
positive, motile, sporing bacillus belonging to the subtilis group.
It is aerobic and facultatively anaerobic, grows readily on the
ordinary culture media, and ferments dextrose and saccharose with
the production of acid only, but lactose is hardly attacked. In a
dog and a monkey diarrhoea was produced by feeding with it.

Urinary and genital organs. — The meatus urinarius and distal
portion of the urethra contain a few organisms, which increase in
number in inflammatory conditions, and Gram-negative cocci may
be found (see p. 248). The deeper portion of the urethra, however,
is free from organisms, and the bladder is sterile. The genital
tract in the female up to the middle zone of the cervix contains
organisms, but the uterus and Fallopian tubes are normally sterile.
The B. vagince of Doderlein, a large Gram-positive bacillus capable
of growing in an acid medium, is frequently present in considerable
numbers in the vagina.

1 See Herter, Bacterial Infections of the Digestive Tract, 1907.

2 Journ. Biolog. Chemistry, vol. v, p. 419.

CHAPTER XXI

THE BACTERIOLOGY OF WATER, AIR, AND SOIL, AND
THEIR BACTERIOLOGICAL EXAMINATION— SEWAGE
—BACTERIOLOGY OF MILK AND FOODS

Some of the Commoner Organisms found in the Air, Water
and Soil.

Bacterial Content of Waters and the Factors
influencing it. Filtration, etc.

THE bacterial flora of natural waters is a very varied one.
The organisms met with in surface waters, such as streams,
ponds, and shallow wells, are derived from the air and
soil through which the water has passed, and if not con-
taminated from human or animal sources, from the air of
towns, from sewage or manure, consist mainly of non-
pathogenic bacilli, the majority of which are chromogenic
and non-liquefying, and develop best on culture media
at a temperature of 18° to 22° C. or thereabouts, not at
blood heat ; also of some sarcinse and a few micrococci ;
B. coli and B. Welchii are usually absent. When, however,
the water passes through cultivated lands, or receives
sewage, the number of organisms is enormously increased ;
a large proportion of them liquefies gelatin and develops
at blood heat, and B. coli and B. Welchii appear more or
less numerously. Whereas water from shallow wells has
a bacterial content nearly as great as the surrounding
surface water, that from deep wells, especially in the chalk,
is remarkably free from organisms. The following Table

572

BACTERIAL CONTENT OF WATERS 573

illustrates the number of organisms that may be met with
in water from different sources :

Source Number of organisms

per cubic centimetre.

Freshly fallen snow . . . 34-38

Ice (very variable) 30-1700

Rain water (Paris) . . . 4-5
Rhone, above Lyons . . 75
Rhone, below Lyons . . 800
Rhine, at Miihlheim . . . average about 20,000
Thames, at Hampton (Frank-
land) (variable) 2000-90,000

Deep well in the chalk (Kent

Company) .... 3-19
Surface well .... 1200
Spring water, Reigate (Frank-
land) 8

Lake of Lucerne . . . 8-50

Loch Katrine (Frankland) . 74
Filtered water supplied to London

(Houston) .... average rarely exceeds 100

Sewage (Frankland) . . 26,000,000

The number of bacteria in a natural water varies con-
siderably with its source, at different seasons, and under
different climatic conditions. The Table x on p. 575
illustrates the seasonal variation in certain raw London
waters.

The following factors modify the number of organisms
present in the water :

(1) Storage of unfiltered water. — A large storage capacity
permits of the water being admitted when the source
(river, etc.) is in its best condition, so that foul water, in
flood time or drought, may be avoided. Moreover, storage
alone usually markedly diminishes the number of organisms,
partly by subsidence, partly by lack of aeration, and partly
probably owing to the struggle for existence going on
among them (see also p. 361).

1 Houston, Seventh Ann. Rep. Hetropol. Water Board, 1913.

574 A MANUAL OF BACTERIOLOGY

(2) Thickness of fine sand in the filter-beds. — Efficient
sand nitration removes quite 99 per cent, of the organisms
originally present. The fine sand only has to be taken
into account in estimating the removal of organisms and
efficiency of a filter bacteriologically. It probably should
form a layer not less than 3 ft. to 3 ft. 6 in. in thickness.
Moreover, a filter-bed is not efficient at first, but becomes
so when the surface film forms, composed of sedimented
particulate matter, and of a zoogloeal mass of bacteria and
algae.

(3) The rate of filtration. — The removal of organisms is
less perfect when the rate of filtration is increased ; this
should not exceed about 1*5 gallons per square foot per
hour.

(4) The renewal of the filter-beds. — New, or recently
cleaned, filter-beds allow a greater number of organisms
to pass through. The beds must be cleaned from time
to time by raking up and clearing away the surface layer
of sand, for as time goes on the rate of filtration becomes
slower and slower, though the bacterial efficiency of the
filter-beds does not appear to be reduced by prolonged
use. The normal bacterial efficiency seems to be
rapidly regained after cleaning — within two or three
days.

Besides storage and filtration, sedimentation in the
presence of fine particles, either naturally present or
artificially added, may also effect a marked removal of
micro-organisms from water. Thus, by the addition of
alum, an old method of clarifying turbid water, a large
number of the organisms present are carried down in the
precipitate.

The Clark process of softening water may also reduce
the number of organisms present, but is very uncertain
(Moor and Hewlett). By the Porter-Clark rapid process,
however, in which the precipitate of calcium carbonate is

iS 2

m

af

a|

as

oo
ioio

oooioo

s 5 s

111 a

•2^3

5 S

1 2

^5

o • <

11:

10 t~

I a

*

, _ ,

SiHOl
i-HrH

^

ne
es.

-

31

576 A MANUAL OF BACTERIOLOGY

removed by filtration through canvas bags, very con-
siderable purification is effected.1

Houston has introduced an " excess lime " method.
Enough lime is added to the water to render it decidedly
alkaline and germicidal for the colon bacillus in five to
twenty-four hours (for raw Thames water, about 1 of lime
in 5000 of water). At the end of this period a sufficiency
of pure stored water is added so as to precipitate the
excess of lime. With Thames water, 3 parts of raw water
with 1 part of stored water would be the approximate
quantities.

The Tables on pp. 577 and 578 illustrate the influence
of storage and of sand filtration on the bacterial content
of a water.

The Bacteriological Examination of Water2

The bacteriological analysis of water affords valuable
indications as to the purity or otherwise of a water, and,
if properly carried out, will indicate a pollution so small
in amount as to be incapable of detection by chemical
methods.

The specimen of water should be collected in clean bottles
of about 100-200 c.c. capacity, sterilised preferably by
heat. If, however, the bottles be thoroughly cleaned and
rinsed out with a little strong sulphuric acid, and then
thoroughly rinsed several times with the water to be
examined before taking the specimen, no error will be
introduced. The stopper of the bottle should be tied
down with a thin layer of cotton-wool enclosed between

1 Nankivell, Journ. of Hyg., xi, 1911, p. 246 ; Hewlett and Nankivell,
Rep. Med. Off. Loc. Gov. for 1911-12, p. 350.

2 See Savage, Bacteriological Examination of Water Supplies (Lewis,
1906) ; Thresh, Examination of Water and Water Supplies (Churchill,
Ed. 2, 1913) ; Houston, Gordon and others in Reps. Med. Off. Loc. Gov.
Board, 1899-1904 ; Houston, Reports to the Metropolitan Water Board
and Studies in Water Supply (Macmillan & Co., 1913).

+*

1 1 1

1 1 1

rH 00 O

in TH esj

000
<N <N <N

CO <M CO

ift O O

S •§

fao

CO rjl CO

* I

i i I

t/3 U J

37

578

A MANUAL OF BACTERIOLOGY

snaxvAV
KOOKOT: riiv
sivxoi dKVHO

.op o o n-
O «o ob i> o\

Cl

OS
CO

cb

0

co

I

S

TH

CO

OS

co

<s s s s

5n

CO
O4

CD

04

2

00

CN

0
CO

CO
CN

r-

CO
04

•(sajdures iQSf)
saaijBM. paAuop
-samtjqi,

.04 00 t^- CM

dJ^ o t^- OS

OS

00

,.

H

cT

(M

op

CM

co

10

CO

CO

(M

^

.h- p 10 co

•<« 04 04 04

CO

04
O4

O4

0

s

-

-

04

-

P

•(saiduiBS COG)

.CO 00 r-l n

C5

>0

0

^

00

OS
0

Cl

cb

g

lO

P

CJeo cb io N

oo

Cl

0

.04 10 0 00

CO

CM CN"'(N ,_ CM -

CO

00
04

10
04

•(saiduiBS 006)

PUB

jlJBMq^nog

ds Is?

0

00
rH

Cl

l~

CO

CO
CM

Cl
Cl

CM

0

cb

04

oo

.0 <f p 0

<{ie eb •* vb

?

«

04
04

0

CN

00

b

IO

9

04

*

10
04

xasaippiK

. r» ci o co

CO

cb

S

9

Cl

Cl

i

oo

6

5r

z

OS

10

I-

CO

CO

-is s s :

CO

T-

%

p^

-

«

b

-

-

g

p'tmo

^£ S ^ S

CO
'S.

3

CO

cb

ITS

CO

CM

TH

ia

CO*

2,

S

*«*

. P r* O *"«

-
cb

S

H

2

Cl

BQ

OS

t>

t>

o

E

b

s

•

JIJBJ uo:>duiax

=S g s g

I

co

l-H

co
db

£

0

o1

00
0

M

co

en

10

<eb ib ob lb

CD

-

-

0

04

£

04

04

CN

eb

uopuoi -g;

.*»»?"* 9

rh cb 00 "* ^
O4 (M CO *n

i

>0

3

«

0

CM

0

0

OS

3

I-

op

Cl
Cl

oo
ob
co

10

b
eo

CN
10

CN

5

ri

'

(saiduiBS £681)
•jaAi?[ Max

^ co co m

a

CM

,_

to

CM

t^

0

-n

pr

09
04

^5 IO O rH IO

0

oo

05

00

CO

CM

0

rH

fc

04

r,

£

r,

-

00

CO

!

«

eb

I i

CO
(M
*§

|

O

>

<1

r-l
rH

«"S s

o "£ *>-, ti

$ 3 1 £

|

1
i

I September

1 October .

I November

1

1

ft

CO

o

I

I February

I

EXAMINATION OF WATER 579

two pieces of muslin, and the bottle should be not quite
filled. In taking the specimen the following details should
be attended to :

(1) If taken from a tap, the water should be allowed
to flow for at least five minutes before the specimen is
collected.

(2) The water from a cistern is not a representative
sample of the water-supply ; to be so the specimen should
be taken direct from the mains.

(3) If taken from a stream or pond, the bottle should be
held about a foot below the surface and away from the
edge before the stopper is removed.

(4) If taken from a well the conditions should be noted,
e.g. whether the well has been recently disturbed or not,
whether the pumps have been in operation, etc., for such
may markedly influence the number of bacteria found.

The specimen should then be examined with as little
delay as possible, for if allowed to stand for any time a
large increase in the number of bacteria may take place.
Frankland, for example, found that in distilled water,
even at the ordinary temperature, organisms multiply
enormously :

Number of organisms
Hours in 1 c.c.

0 1,073

6 6,028

24 7,262

48 48,100

In water of good quality the organisms are found to
multiply much more rapidly during the first few days,
after which time they become less and less numerous ;
but in impure water multiplication is slower, and the
number more persistent, while in very impure water the
number may diminish. It is essential, therefore, if reliable
results are to be obtained, for the specimen to be examined

580 A MANUAL OF BACTERIOLOGY

at once (within three hours). If this cannot be done the
specimen should be packed in ice ; the cold will then
inhibit multiplication to any extent. Special double-
chambered metal boxes are made for this purpose : the
bottle containing the sample (not less than 60 c.c. ; the
writer prefers to have not less than 200 c.c.) is placed in
the inner chamber, the outer chamber (which surrounds
the inner) being filled with a mixture of ice and sawdust,
and the whole is packed in a wooden box with felt lining.
According to Remlingler,1 the addition of 10 per cent, of
common salt to the sample preserves the original bacterial
content of the water unaltered up to ninety-six hours after
taking the sample, without icing. Besides the sample
packed in ice, a " Winchester quart " of the water may
also be collected for examination for the spores of the
B. Welchii (enteritidis sporogenes).

The routine bacteriological examination of the specimen
may be carried out according to the scheme (here somewhat
modified) drawn up by a committee of the Royal Institute
of Public Health.2

PROCEDURES. — The following procedures should be
carried out :

(a) Enumeration of the organisms which will develop
aerobically in gelatin at 20° C.

(b) Enumeration of the organisms which will develop
aerobically in agar at 37° C. (Enumeration is carried out
by counting the number of colonies which develop in the
plates [see p. 79].)

(c) Search for Bacillus coli, and identification and
enumeration of this organism if present.

(d) Search for, and enumeration of, streptococci.

As a routine measure it is not necessary to search for

1 Comp. Rend. Soc. BioL, Ixx, p. 64.

2 Journ. State Med., vol. xii, 1904, p. 471.

EXAMINATION OF WATER 581

the Bacillus Welchii (enteritidis sporogenes), but in special
instances it may be desirable to do so.

The bottle must be well shaken to mix the sample.
Before removing the stopper, it and the neck of the bottle
should be swabbed with absolute alcohol, which is then
carefully ignited and allowed to burn away.

MEDIA, TIME OF INCUBATION, ETC. — For the gelatin
count ordinary nutrient gelatin is employed, the period of
incubation being seventy-two hours. In hot weather it
may be necessary to use 15-20 per cent, gelatin (unless
an incubator which can be cooled is available), but the
development of the colonies is slower. For the agar count
ordinary nutrient agar is used, the period of incubation
being forty to forty-eight hours.

The media should preferably be recently prepared and
be standardised to a reaction of +10.

In addition to the actual numbers of organisms which
develop in the gelatin and in the agar, a comparison of the
ratio of the number of organisms developing in gelatin at
20° C. to those developing in agar at 37° C. also gives
useful indications. With a pure water this ratio is gene-
rally considerably higher than 10 to 1 ; with a polluted
water this ratio is approached, and frequently becomes
10 to 2, 10 to 3, or even less. The actual number of
organisms growing at blood-heat is of considerable value
apart from any question of ratio.

In certain instances it is true that this ratio may be
unreliable. Thus with surface waters, especially in the
tropics (as pointed out by Horrocks) varieties of the
B. fluorescens liquefaciens and non-liquefaciens and B.
liquefaciens may be abundant and grow well at blood-heat.

Distilled water gelatin and agar have also been recom-
mended, but since the organisms of polluted water develop
better in the ordinary nutrient media, the latter are
preferable for routine use.

582 A MANUAL OF BACTERIOLOGY

AMOUNTS TO BE PLATED, SIZE OF DISHES, etc. Gelatin.—
For an ordinary water amounts of Ol, 0*2 and O3 c.c.
may be plated in Petri dishes of about 10 cm. diameter,
preferably done in duplicate.

Agar. — Two plates may be made with 0*1 and O2-
O3 c.c., and are preferably duplicated.

The desired volume of water should be run into the sterile Petri
dish by means of a sterile 1 c.c. pipette graduated in hundredths,
The tubes of gelatin should be melted in a water-bath at a low
temperature (40° C.). A tube is taken from the water-bath, wiped
to prevent the adherent water running down into the Petri dish,
its mouth is singed in the Bunsen flame to sterilise it, and the
contents are then quickly poured into the dish and mixed with the
water by tilting the dish several times.

The agar tubes must first be boiled, then cooled to about 45° C.,
and similarly treated, or surface plates may be made.

If waters are constantly being examined, it saves trouble to have
the gelatin and agar in small flasks, 30-60 c.c. of the former and
20-40 c.c. of the latter ; a flask of each will then be used for an
examination.

In dealing with an unknown water, and in all cases of
doubt, additional plates should be prepared with a dilution
of the water (made with sterilised tap-water) of ten or
hundred fold, according to circumstances.

The amount of the medium in a plate should be 10 c.c.

The counting is done with the naked eye, preferably in
daylight, any doubtful colony being determined with the
aid of a lens or low power objective. The number of
liquefying colonies in the gelatin plates should also be
noted. The plates should be inspected daily, in order
that the count may be made earlier should liquefaction
render this necessary.

In examining an ordinary drinking-water there is no need ever
to dilute. As 1000 or 1500 colonies can be counted in a plate, and
if the number on a plate should be, owing to crowding, uncountable,
ipso facto this would be sufficient to condemn without an actual
count. Dilution is necessary when dealing with river or other

EXAMINATION OF WATER 583

water known to be polluted, and of which an estimate of the number
of organisms present is desired. In order to count the colonies if
very numerous, ink lines may be drawn across the bottom of the
Petri dishes so as to divide them into sectors. Ruled paper discs
(Pakes's discs) upon which the dishes are placed can also be obtained.
The colonies in the sectors are then much more easily counted :
or if the colonies be very numerous and evenly distributed, the
number in two or three of the sectors may be counted, and the
total number on the plate estimated by calculation.

SEARCH FOR BACILLUS COLI, ETC. — Various media may
be employed for the detection, isolation, and enumeration
of B. coli. The writer generally employs as a preliminary,
glucose bile-salt peptone-water, but many other media
may be employed, e.g. formate or neutral-red broth, or
if the organism is abundant, neutral-red bile-salt agar.

As a routine, 50 c.c. should be the minimal quantity
examined for the presence of the Bacillus coli, quantities
from a minimum of 0*1 c.c. to a maximum of 50 c.c. being
added to the tubes of culture media.

It is preferable to add the water directly to the tubes
of culture medium, even with the larger amounts, and
not to concentrate the bacteria by any method. The
culture media may be diluted with at least an equal volume
of the water without interfering with their cultural pro-
perties, and large tubes or small flasks are used for the
larger amounts.

In the case of glucose or lactose bile-salt peptone- water,
the medium may for the larger amounts be prepared of
double strength. The glucose or lactose bile-salt peptone
water should be incubated at 42° C. for not less than
forty-eight hours.

For composition of glucose formate broth, glucose and lactose
bile-salt media, and neutral-red broth, see p. 590, et seq. While a
lactose medium has the advantage of excluding a number of forms
which, though fermenting glucose, do not ferment lactose, and are
therefore not typical B coli, Houston has found that a glucose

584 A MANUAL OF BACTERIOLOGY

medium is more delicate than a lactose one. For general purposes,
quantities of from 0-1 to 25-0 c.c. may be added to tubes of the
medium selected. For the examination of an ordinary drinking-
water, the writer usually employs five tubes with 1 c.c. of the
water in each, two tubes (double strength) with 10 c.c. in each, and
one tube (double strength) with 25 c.c. For the larger amounts
large test-tubes and boiling tubes must be employed.

If the medium shows changes (acid + gas) suggestive of the
presence of B. coli, it is only presumptive evidence of the presence
of this organism. Occasionally other organisms produce a similar
change, e.g. B. lactis aerogenes, B. cloacae. Hence the necessity for
the isolation and identification of the organism as recommended
in the next section.

ISOLATION OF BACILLUS COLI, IF PRESENT. — If indica-
tions of the presence of the Bacillus coli be obtained in
the preliminary cultivations (acid + gas), the organism
must be isolated and identified. If several tubes show
acid + gas, one or two of the tubes with the smallest
quantities of the water should be used for this purpose.

This may be done by making surface cultures on plates
(sloping tubes generally suffice) of either (a) litmus lactose
agar, reaction + 10 ; (6) litmus lactose bile-salt agar ;
(c) Conradi and Drigalski agar, which the writer generally
employs ; or (d) ordinary nutrient gelatin. Agar media,
incubated at 37° C., have the advantage of saving time.
(For composition of media, see p. 590, et seq.)

IDENTIFICATION OF, AND TESTS FOR, THE BACILLUS
COLI. — Having obtained coli-like colonies on the plates
made from the preliminary cultivations of the water,
various tests must be used for identification. The organ-
ism should conform in morphology, motility and staining
reactions with the characters of the typical B. coli as given
at pp. 379-387, and must be subjected to various cultural
tests, e.g. the " flaginac " reactions of Houston (p. 384).
The writer generally employs these, with the addition of
the fermentation reactions given by dulcitol, mannitol,
and adonit litmus peptone water, and gelatin for absence

EXAMINATION OF WATER 585

of liquefaction. If atypical Bacilli coli (see pp. 388 and
389) are met with, the fact should be noted, but their
significance is not yet fully determined.

STREPTOCOCCI. — It is a distinct advantage to search
for streptococci. They may be looked for by making
hanging- drop preparations of the fluid media employed for
the preliminary cultivation of the B. coli (glucose or lactose
bile-salt peptone water, etc.) The presence or absence of
streptococci in these tubes gives also a quantitative value
to the examination, just as in the case of B. coli, and the
result obtained should be stated. The streptococci can
be readily isolated on Conradi-agar plates.

According to Houston (loc. cit.), faeces contain at least 100,000
streptococci per gramme. The type of streptococcus generally present
is one forming short chains, producing a uniform turbidity in
broth, acid and clot in litmus milk within five days at 37° C., and
non-pathogenic for mice. (See Table, p. 235.)

BACILLUS WELCHII. — As already stated, it is not essential
as a routine procedure to search for the Bacillus Welchii
(enteritidis sporogenes), though in certain instances it may
be of advantage to do so. A negative result in such
cases is probably of more value than a positive one.

For the isolation of B. Welchii, 500 c.c. of the water may be
filtered through a Pasteur-Chamberland filter, the deposit is sus-
pended in 5 to 6 c.c. of sterile water, and 1 c.c. of the suspension
added to each of five to six tubes of sterile milk, which are then
heated to 80° C. for ten minutes in a water-bath, and incubated
anaerobically at 37°C. for forty-eight hours (filter-brushing method).
A better method *• is to employ large boiling tubes or small Erlen-
meyer flasks, each containing 25 to 50 c.c. of sterile milk. To each
tube a quantity of water equal to that of the milk is added, the
tubes are then heated in a water-bath to 80° C. for fifteen to twenty
minutes, some sterilised oil or melted vaseline is poured on the
surface to exclude air, the tubes are cooled in water to 37° C. or
thereabouts, and incubated for forty-eight hours at 37° C. Not
less than 200 c.c. of the water should be used. The typical change

1 R. T. Hewlett, Trans. Path. Soc. Lond., vol. Iv, 1904, p. 123

586 A MANUAL OF BACTERIOLOGY

in the milk (see p. 428) indicates the probable presence of the
organism. To make sure that the change is due to the B. Welchii
and not to the C. butyricum, 1 c.c. of the whey per 100 grm. of
body-weight should kill a guinea-pig in forty hours when injected
subcutaneously.

The virulence of a peptone-water culture has been suggested as an
index of contamination, but in the writer's hands has not given
reliable results. If sufficient peptone and salt be added to a
measured volume of the water to form a 1 per cent, solution of the
former and a \ per cent, solution of the latter, the mixture incubated
at 37° C. for twenty -four hours and injected intraperitoneally into
a guinea-pig, a bad water is stated to kill, whereas a good one does
not. The amount to be injected is 2 c.c. and death should ensue
within forty -eight hours.

INTERPRETATION OF RESULTS. — The interpretation of
the results of the bacterioscopic examination of water is a
difficult matter, for which experience is necessary. Just
as in chemical analysis, it is not possible to lay down an
absolute standard, a knowledge of the source and sur-
rounding conditions being of the greatest importance in
forming an opinion. The ultimate aim is, of course, the
detection of sewage or fsecal pollution ; the bacterioscopic
analysis does not give any information as to the suitability
of the water for household, trade, or factory purposes.

Number of colonies on the gelatin plates. — The number of
colonies represents approximately the number of organisms
in the original sample capable of development aerobically
at 20° C. in gelatin. This number in a good water rarely
exceeds 100 or 150 ; in pure waters, particularly those
coming from deep chalk-wells, there may be only a few—
5 to 10 per c.c. (the results are always expressed in numbers
per cubic centimetre of the original water). In waters of
poorer quality the number may approach 500 per c.c.
Anything over this casts suspicion on the water, and
1000 per c.c. or more should probably condemn the sample,
always supposing, of course, that multiplication in vitro
can be excluded by the proper storage of the sample

EXAMINATION OF WATER 587

bottle in ice. As a rule in water of good quality liquefying
organisms are scanty, while in a polluted water they are
numerous.

Number of colonies on the agar plates. — As mentioned
before (see p. 581), it is the ratio of the number of organisms
developing on the agar plates to the number of those
developing on the gelatin plates that is of importance.

Number of B. coli. — The detection and enumeration of
B. coli are regarded by all as perhaps the most important
part of water examination. The number of B. coli is esti-
mated from the amounts of water that have been added
to the tubes of media, which, however, assumes that the
organism is regularly distributed throughout the sample,
and this must so far as possible be ensured by thorough
mixing. The results generally come out fairly concor-
dantly, though irregularities exceptionally occur which
can only be obviated by making duplicate sets of cultures.

It is better to state the result as " B. coli present in

c.c. of water " rather than to say that so many B. coli are
present, though as a matter of fact the latter statement
is approximately correct. Adopting the writer's method
for B. coli (p. 584), if none of the tubes contains B. coli,
we say that " B. coli is absent from 50 c.c. " ; if the 25 c.c.
tube contains B. coli, but not the remainder, " B. coli is
present in 25 c.c. but not in less," and so on.

If nothing is known about the water, the following
standards may be adopted :

(a) Waters of good quality. — B. coli absent in 50 c.c.
of the water.

(b) Waters of medium quality. — B. coli present in 50 c.c.
but absent in 25 c.c.

(c) Waters of poor quality. — B. coli present in 50 c.c. and
25 c.c., but absent in 10 c.c.

(d) Waters of suspicious quality. — B. coli present in
50 c.c., 25 c.c., and 10 c.c., but absent in 1 c.c.

588 A MANUAL OF BACTERIOLOGY

(e) Waters unfit for drinking. — B. coli present in 1 c.c.
or less.

Waters which show no B. coli in 50 c.c. are of a high degree of
purity, and therefore the proved absence of this organism in this
amount, and still better in larger quantities, is of great value.

B. coli should be absent from at least 50 c.c. of spring or deep
well water, possibly from greater amounts.

In upland surface waters the presence of B. coli in 40, 10, or even
2 or 1 c.c. means contamination, but not necessarily a contamination
which it is essential to prevent. It may be from contamination
with the excreta of animals grazing on the gathering areas, and is
by no means necessarily from sewage or other material containing
specific organisms of infection. If B. coli are present in numbers
greater than, say, 500 per litre (or even in that amount), such a
water is suspicious, as it is rare to get so many B. coli in a water
from the kind of animal contamination indicated, and further
investigation is desirable. In filtered samples the number of
B. coli is, as a rule, considerably reduced.

In surface wells B. coli in large numbers indicate surface or other
'contamination, generally very undesirable if not actually dangerous.

It must clearly be understood that the presence of the B. coli
in water is used as an index of pollution, just as the organic ammonia
is in a chemical analysis. This organism is not necessarily harmful
in itself ; it is what it indicates, viz. pollution, probably with human
excremental matters, which may contain the organisms of specific
disease, e.g. typhoid, dysentery, and cholera. As a routine, the
typhoid bacillus is never looked for, and the statement sometimes
seen in the report on the bacteriological examination of a sample
of water that " no typhoid bacilli have been detected " is of little
value. It is on the general results of the examination, as detailed in
preceding pages, that a conclusion is arrived at respecting the purity
or otherwise of a water.

Bacillus Welchii. — This organism being abundantly
present in fseces and sewage, its presence in water has been
suggested as an indication of pollution. Its spores, how-
ever, are very resistant, and it might, therefore, gain
access to the water in ways other than by direct pollution —
e.g. in dust — and for this reason the committee did not
recommend the search for this organism as a routine

EXAMINATION OF WATER 589

procedure. On the other hand, Thresh l lays a good deal
of stress on it, and the following are standards suggested
by him, based on an examination for, and detection of,
B. coli and B. Welchii :

1. Water showing the absence of organisms capable of fermenting
glucose, and of the B. Welchii. These we regard as being free from
any evidence of pollution.

2. Waters showing the absence of organisms capable of fermenting
glucose, but containing the B. Welchii, or its near ally. In the few
cases of this kind which have come under our observation we have
inferred the absence of sewage pollution, but the possible presence
of water derived from fertile soil. This inference has been verified
on more than one occasion.

3. Waters containing organisms capable of fermenting glucose,
but not lactose, but free from the spores of the B. Welchii. These
are regarded as unpolluted.

4. Waters differing from No. 3 only in containing spores of the
B. Welchii. These we regard as free from sewage pollution, but
as probably containing soil washings.

5. Waters containing lactose fermenters, none of which belongs
to the Bacillus coli group, and free from the spores of the B. Welchii.
These we do not regard as being sewage-polluted, but as containing
surface water or subsoil washings.

6. Waters resembling No. 5, but containing the spores of the
B. Welchii. These waters are usually from a source requiring careful
watching, manurial matter probably being used on the collecting
area.

7. Waters containing organisms of the colon group other than
the B. coli, but no spores of the B. Welchii. These we do not regard
as dangerously polluted, but as probably coming from a source
such as that referred to under No. 6.

8. Waters containing organisms of the colon group other than
the B. coli, and also spores of the B. Welchii. Pollution indicated,
but possibly from a source not close at hand. The necessity for
frequent examination is essential, especially after heavy rains, as
such waters usually sooner or later show more serious signs of
pollution.

9. Waters containing the true B. coli, but no spores of the B.
Welchii. Such waters are occasionally met with. No opinion can
be expressed without an intimate knowledge of the source. We

1 Public Health, 1904.

590 A MANUAL OF BACTERIOLOGY

have had such water from a source absolutely free from the possi-
bilities of contamination, but usually subsequent examination
has revealed the presence of the spores of the B. Welchii. The
proximity of manured soil is strongly indicated.

10. Waters containing the true B. coli and spores of the B. Welchii.
These we regard as being decidedly contaminated with faecal matter
of recent origin.

Streptococci. — Streptococci are abundant in faeces and
sewage, but are extremely rare, if ever present, in unpolluted
natural waters ; hence the value of their detection. Strep-
tococci as a class are delicate organisms, and it was supposed
that their presence indicates recent pollution.1 Horrocks,
on the other hand, believes that they maintain their
vitality longer even than B. coli, and this is rather the
opinion at present. We need further data before we can
exactly estimate the value of streptococci as indicators of
pollution. There can be no question, however, that the
detection of many streptococci, together with B. coli,
indicates serious pollution.

There can be no doubt of the value of the bacteriological examina-
tion of water, but it cannot entirely supplant chemical analysis,
which on account of its rapidity and the valuable data it yields
will probably always remain an integral part of the examination
of potable waters. If the water be pure and uncontaminated, the
bacteriological examination will occupy three days ; but if con-
tamination be present, though it may be presumed in the same time,
ten days or a fortnight may be required to convert this presumption
into a certainty, owing to the length of time necessary for deter-
mining the characters of the organisms present.

Media Employed for the Isolation of B. Coli

(1) Carbolised gelatin. — Ordinary nutrient gelatin with the addi-
tion of 0-05 per cent, of phenol. (Hardly used now.)

(2) Bile-salt peptone water (MacConkey and Hill). — The com-
position of this medium is as follows : Sodium taurocholate 0-5 grm.,
glucose or lactose 1-0 grm., peptone 2-0 grm., water 100 c.c. The

1 Houston, Rep. Med. Off. Loc. Gov. Board for 1898-99.

ISOLATION OF BACILLUS COLI 591

constituents are dissolved by heating ; the mixture is filtered, and
after filtration sufficient neutral litmus solution is added to give a
distinct colour. The medium is then distributed into Durham's
fermentation-tubes and sterilised by steaming for twenty minutes
on three successive days. The medium may be put up in various
sized tubes, a measured volume in each — e.g. 10 c.c., 20 c.c., 25 c.c.,
etc., according to the quantity of water which is to be added. For
the larger quantities the medium may be made double the above
strength. The inoculated tubes are incubated at 42° C. for forty-
eight hours. The B. coli reddens and ferments both the glucose
and lactose media, so that gas collects in the fermentation tube.

(3) Neutral-red broth (Hunter, Makgill, Savage). — The dye known
as neutral-red (Griibler's) is reduced by the action of the B. coli,
the colour changing to a canary yellow, accompanied by a green
fluorescence. The B. enteritidis (Gartner) also reduces neutral-
red, but the B. typhosus does not do so, nor do streptococci,
B. pyocyaneus, and Vibrio cholerce. Some anaerobes also possess
a reducing action. Glucose agar or broth (0-5 per cent, of glucose)
is employed, and to every 10 c.c. of the medium 0-1 c.c. of a 0-5
per cent, aqueous solution of neutral-red is added. Savage recom-
mends the following procedure : 10 c.c. of the water are added to
a 10 c.c. tube of neutral-red broth ; also to 40 c.c. of the water
contained in a bottle or flask a 10 c.c. tube of the broth of quadruple
strength is added. Both are incubated at 37° C., and examined
daily up to eight days. If reduction occurs, B. coli is almost
certainly present in the water ; if reduction does not occur its
presence is highly improbable.

(4) Glucose formate broth (Pakes). — To ordinary meat infusion
1 per cent, peptone, 0-5 per cent, sodium chloride, 2 per cent, glucose,
and 0-4 per cent, sodium formate are added. When these have
been dissolved by heating, the medium is neutralised (indicator,
litmus), and after neutralisation 2 c.c. of normal caustic soda solu-
tion per litre are added ; the broth is then steamed for twenty
minutes, filtered, and distributed into test-tubes, 10 c.c. in each,
which are steamed for twenty minutes on each of three successive
days. These tubes are inoculated with the water, and incubated
anaerobically at 42° C. for twenty-four to seventy-two hours. Tubes
showing any growth at the end of twenty-four, forty-eight, or
seventy-two hours are removed and examined microscopically and
by plating.

(5) Bile-salt lactose agar (MacConkey). — This medium is prepared
by adding to 1000 c.c. of tap-water in a flask 2 per cent, of peptone,
0-5 per cent, of sodium taurocholate, and 1-5 per cent, of agar.

592 A MANUAL OF BACTERIOLOGY

The mixture is autoclaved at 105° to 110° C. for 1^ hours, cleared
with a small addition of white of egg, and filtered. To the nitrate
1 per cent, of lactose is added. The medium is then distributed
into test-tubes, 10 c.c. in each, and sterilised by fifteen minutes'
steaming on three successive days. Plates are made and incubated
at 42° C. for forty-eight hours. The colonies of organisms which
ferment lactose with the formation of acid are surrounded with a
cloudiness or haze owing to the precipitation of the taurocholate.
Neutral-red or krystal violet may be added (proportions, see Nos. 3
and 6).

(6) Conradi-Drigalski agar. Mixture A. — To 1 litre of acid
beef broth (p. 54) add :

Witte's peptone . . . . .10 grm.

Nutrose 10 „

Sodium chloride . . . . 5 „

Steam for one hour, and add 25 grm. of powdered agaf. Steam
for three hours, bring to a reaction of + 10, and filter through
" papier Chardin."

Mixture B. — Boil for a few minutes 100 c.c. of Kubel-Tiemann's
litmus solution, add 15 grm. of pure powdered lactose, and boil
again for a few minutes.

Add B to A, and to this mixture add 2 c.c. of a hot 10 per cent,
solution of anhydrous sodium carbonate and 10 c.c. of a 0-1 per cent,
solution of krystal violet. The medium is then tubed, 10 c.c. being
placed in each test-tube, and sterilised.

In using the medium it should be employed as surface plates.
The required number of tubes are melted in a water-bath, and
their contents poured out into sterile Petrie dishes and allowed to
set. These sterile plates are then placed in the warm incubator
for an hour or so with the lids slightly tilted at one edge, so that
the surface of the medium may dry somewhat. The matter to
be plated is sufficiently diluted, and from a few drops to 0-5 c.c.
are run on to the surface and spread by means of a glass rod bent
into a flattened hook, and sterilised by boiling. On this medium
in forty-eight hours B. coli forms large red colonies, B. typhosus
and B. dy sentence small blue colonies, and streptococci small delicate
red colonies. Other organisms are to a large extent inhibited from
developing.

(7) 8.D.8. rebipelagar (Houston). — " Rebipelagar " has been
much used by Houston * for the isolation of B. coli. It has the
following composition : Agar 20 grm., taurocholate of soda 5 grm.,

1 First Rep. on Research Work, Met. Water Board, 1908.

SPECIFIC ORGANISMS IN WATER 593

lactose 10 grm., neutral-red 4 c.c. of a 1 per cent, solution, peptone
20 grm., water 1 litre. The S.D.S. rebipelagar has the following
composition : Agar 20 grm., taurocholate of soda 5 grm., lactose
2-5 grm., neutral-red 4 c.c. of a 1 per cent, solution, peptone 20 grm.,
saccharose 2-5 grin., dulcitol 2-5 grm., salicin 2-5 grm.

The Isolation of Specific Organisms
from Water

The principal disease-producing organisms conveyed by water
are the B. typhosus, B. paratyphosus, B. dysenterice, and Vibrio
cholerce.

THE ISOLATION OF B. TYPHOSUS, B. PARATYPHOSUS, AND B.

DYSENTERIC FROM WATER. — There is great difficulty in isolating
the B. typhosus from water that has been very copiously contami-
nated with specifically polluted sewage, there is, therefore, far
greater difficulty when the specific pollution has been small in
amount. The earlier records of the isolation of the B. typhosus
must be accepted with much scepticism, as the methods of identi-
fication were formerly incomplete and unsatisfactory. It is neces-
sary to bear in mind that usually, when drinking-water has suffered
sewage-pollution, the amount of the pollution is relatively very
minute when compared with the great bulk of the water-supply.
Moreover, allowing ten days as the average incubation period of
typhoid fever, another week before the disease comes under notice,
and another week before the fact that an epidemic is in progress
is recognised, at least a month will have elapsed between the date
of infection of the water-supply (supposing this to have occurred
on one occasion only, as may be the case) and the taking of the
samples for examination, a period during which all the typhoid
bacilli may have died out. The contamination of water may,
however, be of an intermittent nature.

Numerous methods * have been devised for the isolation of the
typhoid bacillus from an infected water. With rare exceptions,
it is impossible to detect the organism by direct plating ; it is too
scanty and too mixed with other organisms to admit of this, and
therefore concentration of the bacterial content of the water must
be attempted. The following are some of the methods which have
been suggested for this purpose ; they serve equally well for B.
paratyphosus and B. dysenterice.

1 See H. S. Willson, Journal of Hygiene, vol. v, 1905, p. 429 ;
McWeeney, Brit. Med. Journ., 1909, vol. ii, p. 866.

38

594 A MANUAL OF BACTERIOLOGY

1. Filtration through a porcelain filter. — By passing one to two
litres of the water through a sterile Pasteur-Chamberland filter,
the whole of the organisms present may theoretically be collected
in a few c.c.s. Practically, however, a large proportion of the
organisms are lost in the process : perhaps they get carried into
and remain in the superficial layers of the filter-candle, and for this
reason, though sometimes employed, this method has been largely
given up.

2. Concentration. — W. J. Wilson 1 has devised the following
method : The water is placed in one or two Winchester quart
bottles, and 10 c.c. of nutrient broth are added for every litre. The
bottles are placed in a water- bath maintained at 37°-40° C., and
are connected by rubber corks and tubing with a condenser (at a
lower level) through which cold water continuously passes, and
the tube of the condenser is connected to a large bottle (at a still
lower level). This bottle is kept partially exhausted by means of
a filter-pump. The water evaporates and is thus concentrated, the
evaporated water being condensed and collected in the exhausted
bottle. It requires twenty-one to twenty-two hours to evaporate
a litre of water. The water remaining in the bottles, now concen-
trated to a few c.c.s., is then plated on Conradi-Drigalski or mala-
chite-green agar.

3. Chemical precipitation. — These methods depend on the forma-
tion in the water of a fine, inert precipitate, which entangles and
carries down with it a large proportion of the bacteria present. Thus
in the Vallet-Schiider 2 method, to 2 litres of the water are added
20 c.c. of a 7-75 per cent, solution of sodium hyposulphite and
20 c.c. of a 10 per cent, solution of lead nitrate. The precipitate
is allowed to settle or is centrifuged off, is dissolved in a small
volume of a saturated solution of the hyposulphite, from which
plates are made in suitable media. Ficker 3 uses ferrous sulphate
after making the water faintly alkaline with caustic soda ; the
ferrous hydrate formed carries down the micro-organisms (this
must be a risky procedure, as the typhoid bacillus is very sensitive
to caustic alkalies). Iron oxychloride may also be used as the
precipitant. H. S. Willson (loc. cit.) employs alum. A stock
solution of alum is prepared, containing 10 grm. per 100 c.c., and
of this sufficient is added to the water to obtain 0-5 grm. to the
litre. After the precipitate of aluminium hydrate has formed,
the vessel is well shaken to mix its contents, and the mixture is

1 Brit. Med. Journ., 1907, vol. i, p. 1176.

2 Zeitschr.f. Hyg., xlii, No. 2, p. 317.

3 Hyg. Rundschau, xiv, No. 1, 1904. p. 7.

ISOLATION OF BACILLUS TYPHOSUS 595

centrifuged for fifteen minutes at 2000 revolutions per minute.
The clear, supernatant fluid is then syphoned or poured carefully
off from the precipitate, and the mass of precipitate in the conical
extremity of the tube stirred up with the little fluid (0-5 to 1 c.c.)
remaining. The suspension is then plated out on Conradi-Drigalski,
malachite -green or brilliant -green, agar. This seems to be a very
promising method.

4. Serum agglutination. — An anti-typhoid serum — the serum of
an animal which has been inoculated several times with the typhoid
bacillus, having the power of agglutinating typhoid bacilli — if
added to a water would presumably agglutinate any typhoid bacilli
into masses which will sediment or may be centrifuged off. The
method has been used by Schepilewsky,1 who adds 10 to 20 c.c. of
the water to flasks containing 50 c.c. of nutrient broth, to which
after three or four days incubation at 37° C. an addition of the
typhoid serum is made, and after standing for some hours and
centrifuging, the deposit is plated out.

5. Method of enrichment. — The principle of this method is to
devise a medium which will allow of the multiplication of the
typhoid bacillus and at the same time prevent, or at least retard,
the growth of B. coli and allied forms. Almost all the methods
which have been introduced for this purpose fail, inasmuch as
though they inhibit the growth of a great many organisms, they
do not inhibit the growth of the B. coli, or, if they do, inhibit the
B. typhosus to a still greater degree. Roth 2 found that caffeine
in broth would retard B. coli, but allow B. typhosus to multiply.
The method has been further elaborated by Hoffmann and Ticker,3
who convert the water itself into a nutrient medium by the addition
of 1 per cent, of nutrose, 0-5 per cent, caffeine, and 0*001 per cent,
of krystal violet. The mixture is incubated at 37° C. for not more
than twelve to thirteen hours, at the end of which time the typhoid
bacilli should have multiplied to such an extent as to permit of
direct isolation by plating, the B. coli being inhibited. Many
observers have shown, however, that while caffeine may materially
help, it cannot be entirely relied on to eliminate B. coli and allied
forms.

6. Process of Cambier. — Cambier 4 has devised a process based on
the idea that an actively motile organism will find its way through
the pores of a porcelain filter more quickly than feebly or non-

1 Centr. f. Bakt., Orig., xxiii, No. 5, 1903.

2 Hyg. Rundschau, xiii, 1903, p. 489.

3 Ibid, xiv, 1904, p. 1.

4 Rev. dHyg., 1902, p. 64.

596 A MANUAL OF BACTERIOLOGY

motile forms. His procedure is to make use of a special alkaline
peptone medium, which is placed in a glass jar. In this is immersed
a Pasteur-Chamberland filter-candle half filled with the same
solution, to which is added a little of the fluid to be examined, and
the whole is incubated at 37° C. Sooner or later growth appears
in the fluid outside the candle, and Cambier states that if typhoid
bacilli be present they will make their appearance before B. coli.
In hands other than those of Cambier, however, the method has
not proved successful.

7. Fuchsin agar (Endo). — One litre of 3 per cent, nutrient agar
is made alkaline with 10 c.c. of 10 per cent. NaOH solution after
neutralisation. Pure lactose 10 grm. and saturated alcoholic
fuchsin solution 5 c.c. are added, and after mixing, 25 c.c. of fresh
10 per cent, solution of sodium sulphite are added. The medium
when cold should be colourless. The medium is used as surface
plates, and on it typhoid and paratyphoid colonies are colourless,
coli colonies are red.

8. Malachite-green media. — Loffler has found that malachite
green (No. 120 Hoechst) in the proportion of about 1 in 5000 in
media inhibits the growth of B. coli while still permitting the
growth of B. typhosus. The dye may be added either to liquid or
to solid media. The medium recommended by Loffler l is com-
posed of 3 per cent, agar made with meat infusion, with 1 per cent,
nutrose, and containing in every 100 c.c. 2-2-5 c.c. of a 1 per cent,
solution of malachite green. On this medium the B. typhosus
grows in twenty-four hours as delicate, slightly crinkled colonies,
surrounded by a colourless zone (due to alkali formed by the bacilli).
Thus it is possible to detect one colony of B. typhosus among 300 to
600 colonies of other bacteria. As a medium for " enriching "-
i.e. for specially advancing the growth of the B. typhosus — Loffler
recommends a 15 per cent, gelatin, prepared with beef -juice and
peptone, and containing per 100 c.c. 3 c.c. of doubly normal phos-
phoric acid and 2 c.c. of 2 per cent, malachite-green solution. With
the suspected matter, firstly, one series of malachite -gelatin plates
is prepared and incubated at 25° C. for twenty to twenty-four
hours ; secondly, a tube of malachite gelatin is inoculated and
incubated at 37° C. for twelve to twenty-four hours ; from this a
second tube is inoculated and incubated at 37° C., and then plated
out on malachite gelatin and incubated at 25° C. The colonies of
B. typhosus are well marked after twenty to twenty -four hours, as
large as a pin's head, transparent, highly refractile, light grey and
granular. Their shape is circular or oval, and they show charac-

1 Deutsch. med. Woch., 1906, No. 8.

ISOLATION OF BACILLUS TYPHOSUS 597

teristic offshoots resembling a bone-corpuscle or the body of an
acarus. By using this 15 per cent, gelatin, which can be incubated
at 25° C., there is the double advantage of speedy growth and
formation of very characteristic colonies.

Houston recommends S.D.S. rebipelagar (p. 592) with the addi-
tion of malachite -green to the extent of 1 in 5000 (0-2 grm. to the
litre). On this medium B. typhosus forms colourless colonies ;
most other bacteria do not grow, or appear as blue-black colonies.

9. Werbitzlci's China green agar. — For this 3 per cent, nutrient
agar (reaction +13) is used, and to every 100 c.c. of the agar
1-4-1-5 c.c. of a 0-2 per cent, aqueous solution of china green
(Griibler's) are added.

10. Brilliant green agar. — Conradi devised an agar containing
brilliant green and picric acid, and this has been modified by Fawcus *
as follows : To 900 c.c. of tap-water are added sodium taurocholate,
5 grm. ; powdered agar, 30 grm. ; peptone, 20 grm. ; and sodium
chloride, 5 grm. Dissolve the constituents by steaming for three
hours, filter through wool, and bring to a reaction of + 15 (by
means of lactic acid or NaOH, as the case may be). In 100 c.c. of
distilled water dissolve 10 grm. lactose and add this to the former
filter, distribute in flasks (100 c.c. in each), and sterilise. At time
of using, melt and add to each 100 c.c., 2 c.c. of a 1-1000 aqueous
solution of brilliant green and 2 c.c. of a 1-100 aqueous picric acid
(extra-pure, Griibler's). Typhoid forms round, transparent refrac-
tile colonies of a light pale green colour by transmitted light, B. colt
dark green colonies with an opaque spot at the centre.

CONCLUSION. — The writer would suggest for the isolation
of B. typhosus from water : (1) Concentration of the
organism by precipitation with alum (Willson's method)
or iron oxychloride, followed by plating of the precipitate
on Conradi-Drigalski agar, or, better, on malachite green
agar (Loffler's or Houston's, No. 8 above), or brilliant-
green agar (No. 10 above) ; (2) enrichment by Loffler's
method and subsequent plating. In all cases the organism
isolated must be examined as to its morphological, cultural,
and biological characters, and should have its agglutination
and Pfeiffer reactions tested with a high-grade typhoid
serum. Two organisms which are likely to be mistaken

1 Journ. Roy. Army Med. Corps, February 1906, p. 147.

598 A MANUAL OF BACTERIOLOGY

for the B. typhosus, unless all tests are applied to them,
are the B. (fcecalis) alkaligenes and B. (aquatilis) sulcatus.
Both occur in the dejecta and in polluted water, and are
very like the B. typhosus in morphology, motility, staining,
and cultural reactions, but neither agglutinates with
typhoid serum. The B. alkaligenes sometimes produces
a brownish growth on potato, it renders litmus milk
alkaline and produces alkali, but no gas, in glucose, lactose,
dulcitol, mannitol, saccharose, and salicin. The B. sulcatus
hardly grows at 37° C. and is almost a strict ae'robe, little
growth occurring in the depth of a stab. Some varieties
of typical and of atypical B. coli agglutinate with typhoid
serum, so that a positive agglutination reaction does not
necessarily prove that an organism is B. typhosus.

THE ISOLATION or THE CHOLERA BACILLUS FROM WATER. — The
detection of Koch's comma bacillus (Vibrio cholerce) in water, as
in the case of the typhoid bacillus, is a matter of some difficulty,
as this organism is rapidly overgrown by the ordinary water bacteria.
In the examination of suspected water supplies, the best method
to employ for the detection of this organism is to take advantage
of the fact, first noted by Dunham, that the cholera vibrio
multiplies with great rapidity in alkaline saline peptone solution.
The suspected water is examined as follows : To 300-500 c.c. of
the water are added 1 per cent, each of pure peptone and of common
salt ; the mixture is made faintly alkaline with sodium carbonate,
distributed in a dozen small Erlenmeyer flasks having a layer not
more than an inch deep in each, the flasks are loosely capped with
caps of filter-paper, and incubated at 37° C. At intervals of ten,
fifteen and twenty hours respectively, hanging- drop and cover-glass
preparations are made from the top of the liquid, an which there
is often a surface film, and care must be taken not to disturb this ;
these are then examined microscopically for vibrios and spirilla.
At the same time agar (3 per cent.), or, better, blood alkali agar
(p. 446) plates are prepared and incubated at blood-heat. Any
colonies that appear which resemble the cholera spirillum are
examined microscopically ; if the organisms are comma-shaped,
they are at once subcultured into peptone water and other media.
The original peptone water cultures are tested for the indole reaction
with pure hydrochloric acid, withdrawing some of the contents of

STERILISATION OF WATER 599

the flasks with a sterile pipette. Any likely vibrios isolated
must have its cultural and biological reactions investigated and be
tested for the agglutination and Pfeiffer reactions with a high-grade
cholera serum.

On the survival of the typhoid and cholera organisms in water,
see pp. 360 and 437 respectively.

Ice and ice-creams may be examined by methods similar to those
used for water, the material being first melted at a low temperature.
Some of the fluid should also be centrifuged and the deposit
examined microscopically for gross contamination.

The infection in typhoid fever and cholera, and perhaps
also in bacillary dysentery, is perhaps more frequently
water-borne than conveyed in any other way. It might
be supposed that the acid gastric juice would prevent this,
and it may do so in many instances. Experiments by
Macfadyen l showed that, whereas in fasting animals,
to which suspensions in water of the cholera vibrio were
administered, living vibrios pass into the intestine, when
the vehicle is milk none could be detected in the intestines.
The inference is that when there is no food there is no
gastric juice secreted and the organisms are able to pass
into the intestine, but when food is present the gastric
juice is secreted and the organisms are destroyed.

STERILISATION OF WATER. — This may be done on the small scale
by heat, by the use of germicidal agents, or by filtration through a
filter (see p. 601). Heat may be applied by simple boiling, or by
the use of apparatus in which the water is heated to 65°-90° C.,
and the outgoing hot water is cooled by the ingoing cold water,
which itself is thus warmed, thereby effecting economy in fuel
(Griffiths' and other sterilisers). The chemical germicides that
have been employed are (1) sodium bisulphate, 15 grains to the
pint ; (2) Potassium permanganate, sufficient to tinge the water
deeply for at least half an hour ; (3) chlorine gas or iodine tablets,2
in both cases the taste of the agent being destroyed by the addition
of sodium sulphite ; (4) copper and copper sulphate. Sufficient
metal is dissolved from bright copper in twenty-four hours to destroy

1 Journ. of Anat. and PhysioL, vol. xxi.

2 Nesfield, Journ. Prev. Med., vol. xiii, 1905, p. 623.

600 A MANUAL OF BACTERIOLOGY

typhoid and cholera. Copper sulphate 1 in 100,000 or less is
similarly germicidal, and in still smaller quantities (1 in 1,000,000)
destroys algae, and has been used for the purification of reservoirs
overgrown with algae. On the large (also small) scale, chlorine
derived from hypochlorites is one of the simplest and most efficient
agents. Moor and Hewlett * showed that 0-25 part of chlorine
(equivalent to about 0-75 part of good chloride of lime) per million
parts of chalk water is sufficient to kill B. coli in half an hour.
The taste disappears quickly in bright sunlight and on standing,
or may be removed by an addition of sodium sulphite. If the
water is organically polluted, more chlorine must be used.

Ozone produced by high-tension electric discharge is also employed
on the large scale for the sterilisation of water-supplies, e.g. at
Chartres (see also p. 637).

EXAMINATION OF SHELL-FISH. — Shell-fish may come from sewage-
polluted layings (see p. 362). The following method may be
employed for their examination (after Houston) :

The outside of the shells are cleansed by thorough scrubbing and
rinsing in tap-water, and a final rinse in sterile water. The fish
after cleansing are laid on a sterile towel. The operator then
cleanses his hands and opens the shells aseptically with a sterile
oyster-knife, care being taken to avoid loss of their contained liquor.
The liquor as each fish is opened is poured into a sterile litre cylinder,
and the fish is cut up with sterile scissors and added to the liquor
in the cylinder. Ten fish should be treated, the volume of fish -f-
liquor noted, and sterile water is then added to make up to 1 litre ;
100 c.c. liquid therefore corresponds to one fish. In addition,
four dilutions of the liquid are prepared — 1 in 10, 1 in 100, 1 in
1000, and 1 in 10,000. With the liquid and dilutions gelatin and
agar plate cultivations are prepared for the enumerations of the
organisms present. Cultures are also made in litmus lactose bile-
salt peptone water and in milk for the enumeration and isolation
of B. coli and B. Welchii respectively, taking 100 c.c., 10 c.c., and
1 c.c. of the liquid, and 1 c.c. of each of the four dilutions ; in this
way the contents of the fish, ranging from one fish to TTnrl___ of
a fish, are examined. The process and principles involved corre-
spond to those described for water. Houston has suggested for
oysters as a lenient standard less than 1000, and as a stringent
standard less than 100, B. coli per oyster. Even ten B. coli per
fish should be viewed with suspicion, for Hewlett and others have
shown that oysters from pure layings contain no B. coli.

Watercress, etc., may be examined in a similar manner, 100 grm.
1 Rep. Med. Off. Loc. Gov. Board for 1909-10, p. 559.

FILTERS 601

being weighed out and transferred bit by bit with sterilised forceps
and scissors to a flask containing 900 c.c. of sterile water. The
flask is shaken vigorously, and the washings examined in a manner
similar to that employed for shell-fish.

FILTERS. — Reference has already been made to the
removal of organisms in water by sand nitration. With
regard to niters for domestic use, few of those in the market
are capable of doing more than removing particles of
suspended matter, while they allow from 5 to 50 per cent.,
or even more, of the bacteria present in the water to be
filtered, to pass through. Such filters are, of course,
useless for the prevention of disease — in fact, rather favour
it, by engendering a false sense of security ; and when in
use for some time without cleaning, the water after filtra-
tion may be worse, bacteriologically and chemically, than
before filtration.

Woodhead and Wood x found that the only filters which
were capable of completely removing organisms were the
Pasteur-Chamberland, Berkefeld, and Porcelaine d'Amiant.
The Berkefeld, while more rapid in action than the other
two, after being in use for a few days may allow some
organisms to appear in the filtrate. This, perhaps, is due
rather to a growth of organisms through the pores of the
filter-candle than to a direct passage. Lunt 2 found that
while the ordinary water bacteria, such as the B. fluorescens
liquefaciens, appeared in the filtrate from a Berkefeld filter
within a few days of the infection of the sample, the typhoid
bacillus and the comma bacillus similarly introduced had
not passed through the filter four or five weeks after
infection.

Horrocks,3 however, does not confirm this, and has
found that when sterile water is inoculated with typhoid

1 Brit. Med. Journ., 1894, vol. ii, p. 1053 et seq.

2 Trans. Brit. Inst. of Prev. Med., vol. i, 1897.

3 Brit. Med. Journ., 1901, vol. i, p. 1471.

602 A MANUAL OF BACTERIOLOGY

bacilli and run daily through a Berkefeld filter, the bacilli
appear in the nitrate in one or two weeks, whereas this is
not the case with the Pasteur-Chamberland. The writer
has made some similar experiments, which partially, but
not entirely, support Horrocks's conclusions. Much
evidently depends upon the chemical composition of the
water.

Messrs. Doulton have constructed a porcelain filter which
seems to be perfectly efficient, like the Pasteur-Chamber-
land. All porcelain niters should be cleaned weekly by
well scrubbing with a nail-brush and boiling in water
containing some sodium carbonate.

The Bacteriological Examination of Water-
Filters

The large majority of water-filters at present in use are incapable
of preventing organisms from being washed through into the
filtrate. In order to ascertain whether this is the case with any
particular filter, it should be sterilised in the steam steriliser, and
water containing organisms of known species (B. prodigiosus, B.
violaceus, and M. agilis are very suitable) should be passed through
it for twenty-four hours. This water and the filter should during
this period of the examination be maintained, if conveniently
possible, at a temperature below 5° C. This will almost invariably
prevent any growth or multiplication of the organisms. Samples
should be taken immediately after the filtration has begun, and at
intervals during the day, and again at the end of twenty -four hours.
If they are all sterile, the filter is capable of preventing organisms
from being directly washed through. In the case of filters of very
great density or depth of filtering medium, it may be necessary to
prolong the period of examination beyond the first day ; but most
ordinary filters which permit organisms to be washed through do
so within the first few hours.

Protozoa and Algae in Water

The examination of water for the minute forms of life other
than bacteria, and their enumeration, can be carried out by the

BACTERIOLOGY OF AIR 603

Sedgwick-Rafter method.1 A 6-inch glass funnel is plugged at the
bottom of the stem with a perforated rubber cork, over the upper
end of which a disc of fine silk bolting cloth, cut by a wad-cutter,
is laid. Sharp, clean, dry quartz sand is then poured into the
stem of the funnel to the depth of half an inch above the plug.
The sand should be of such a size that the grains will pass through
a sieve of 60 meshes to the inch, but not through one of 120 meshes.
The sand is washed into place and well moistened with a little
distilled water free from organisms.

The water to be examined is thoroughly shaken and 500 c.c.
are poured into the funnel ; it runs through the sand, which detains
any organisms it may contain. After the water has all passed
through, the rubber plug is carefully removed and the sand washed
down into a test-tube with 5 c.c. of distilled water. The contents
of the test-tube are agitated and the tube is allowed to rest until
the sand has deposited. Immediately this is the case the super-
natant fluid is decanted into a second test-tube, carrying with it
the organisms. One cubic centimetre of this is withdrawn by a
pipette from midway between the top and bottom and transferred
to the counting plate. This consists of an ordinary glass slide on
which a rectangular brass cell (20 x 50 mm.) is cemented, so
enclosing exactly 1000 square mm. The brass cell is 1 mm. thick,
so that the cell contains exactly 1 c.c. The preparation is covered
with a cover-glass and examined with a low power.2

The Bacteriology of Air

Just as in water, the bacteria in the air vary considerably
at different times and seasons, under different conditions,
and in various localities. The species met with are mostly
saprophytes, consisting largely of chromogenic forms. A
number of moulds occur (as spores), and, in fact, ordinarily
are in large excess, together with yeasts and torula?.

It is not easy for micro-organisms to become diffused
through the atmosphere ; they are incapable of a volun-
tary rising, and cannot be torn from a fluid or moist solid

1 Calkin, Twenty-third Ann. Rep. State Board of Health, Massa-
chusetts, 1891.

2 On the microscopy of water, see Whipple, Microscopy of Drinking
Water,

604 A MANUAL OF BACTERIOLOGY

medium by a strong current of air. The medium on
which they are growing must dry up completely and crumble
into fine dust before they can be distributed through the
agency of air-currents (but see p. 365).

The number of organisms in the air varies with the
season, with rain, with altitude, with movement, etc. At
Montsouris, Miquel found in one cubic metre of air 49
organisms in winter, 85 in spring, 105 in summer, and
142 in autumn. After heavy rain the air is largely freed
from organisms. Frankland found at Norwich Cathedral
at an altitude of 300 feet 7 organisms in two gallons, while
on the ground 18 were found ; at the Golden Gallery at
St. Paul's two gallons of air contained 11 organisms ; in
St. Paul's churchyard the number was 70. On high
mountains organisms are nearly absent from the air, and
the same is the case at sea at a distance from land exceeding
about 100 miles. Organisms are much fewer in the air
of the country than in that of towns. At the entrance-hall,
Natural History Museum, South Kensington, Frankland
found in the morning 30 organisms ; in the afternoon,
when many visitors were present, the number had risen
to 292, showing the influence of movement. By keeping a
volume of air absolutely still, enclosed in a box the walls
of which were smeared with glycerin, Tyndall was able
to free it completely from particles and organisms. The
writer found from 43 to 150 organisms per 10 litres of air
in some of the principal streets of London during the
daytime.

Gordon,1 by exposing dishes of neutral-red broth to the
air, or by aspirating air through neutral-red broth (p. 591)
has been able to detect the presence of the S. salivarius,
M. epidermidis, and scurf micrococcus (p. 230) in air
subjected to human contamination. By these tests and
by the use of B. prodigiosus as an indicator he concludes
1 Reps. Med. Off. Loc. Gov. Board for 1902-1904.

EXAMINATION OF AIR 605

that particles of saliva are disseminated as far as 40 feet
in the act of loud speaking, indicating the possibility of
the wide distribution of such pathogenic organisms as
the tubercle, plague, and influenza bacilli and the pneumo-
coccus by speaking, and still more so by coughing.

The number of dust particles in the air may be enormous.
In London Macfadyen and Lunt observed as extremes from
20,000 to nearly 600,000 per c.c. The ratio of micro-
organisms to dust particles is therefore a very small one.

Bacteriological Examination of Air

A number of methods have been devised for the estimation of
the number of micro-organisms in the air, of which the following
are the principal ones :

(1) Plate method. — Melted sterile nutrient gelatin is poured into
a sterilised Petri dish, and allowed to set. The plate is then exposed
to the air, by removing the lid, for a given time — one, five, ten,
or fifteen minutes, etc. — the lid is replaced, and the plate incubated
at 22° C. for some days. The number of colonies of moulds, bacteria,
yeasts, etc., is counted, and, having estimated the area of the
gelatin plate,1 the result is expressed as the number of organisms
falling per square foot per minute. The results obtained by this
method are roughly comparative, but no estimate can be formed
from it of the number of organisms contained in a given volume of
the air.

(2) Hesse's method. — This is a quantitative method for estimating
the number of organisms contained in a given volume of air. The
apparatus consists of a glass tube 30 in. long by 1^ to 2 in. in diameter.
One end of this tube is plugged with a rubber cork through which
a glass tube passes, the other end is covered with a piece of sheet
rubber perforated with a hole £ to | in. in diameter ; over this is
placed another sheet of rubber, unperforated. The small tube
being plugged with cotton-wool, the whole is sterilised for an hour
in the steam steriliser. Just before use 40 to 50 c.c. of melted
sterile nutrient gelatin are poured into the tube, and its walls
coated with the medium. The tube is then strapped horizontally
on to a tripod stand, and the small tube connected by means of a

1 The area of a circular dish is calculated by multiplying the square
of the diameter by 0'785.

606 A MANUAL OF BACTERIOLOGY

piece of rubber tubing to an aspirator consisting of two flasks
arranged so as to form a reversible syphon. A litre of water is
poured into the flask connected with the tube, and the outer sheet
of rubber having been removed from the end of the tube, the water
is syphoned over to the second flask, placed at a lower level, and
an equal volume of air is thus aspirated through the tube. The
second flask is then connected with the tube, and the position
of the flasks being reversed the water is again syphoned over and
a second litre of air passes through the tube, and this process is
repeated until 5, 10, 15, or 20 litres of air have been drawn through
the tube. The rate of flow is controlled by a screw-clamp on the
rubber connecting-tube ; it should not exceed half a litre per minute.
With this rate of flow all the organisms are deposited on the gelatin-

A B c

FIG. 67. — Frankland's tube for air analysis.

coated tube. The aspiration being completed the rubber tube is
disconnected and the sheet of rubber replaced over the end of
the tube, which is then incubated, and the colonies are counted
when they have developed.

(3) Petri's method. — Petri aspirates the air through a glass tube
containing sterilised sand, kept in place by fine wire -gauze wads.
When the sample has been taken the sand is distributed in Petri
dishes, and melted sterile gelatin is poured over it and allowed to
solidify, plate cultures being thus prepared. The objection to
this method is the presence of the opaque particles of sand in the
culture medium.

(4) Frankland's method. — The air to be examined is aspirated
through a tube 5 in. in length and £ in. in diameter (Fig. 67). One
end of the tube is open, the other (c) is plugged with cotton-wool.
At a distance of 1 in. from the open end the tube is slightly con-
stricted to support a plug of glass wool (A). At a distance of 2^ in.
from this plug the tube is again constricted to support a second
plug (B), consisting of glass-wool and finely powdered cane-sugar,
supported in front and behind by plugs of glass-wool. Several
such tubes having been prepared, they are placed in a tin box and
sterilised at 130° C. for three hours, and can then be easily trans-
ported without risk of contamination. When required for use,
a tube is quickly removed from the box, being handled by the

EXAMINATION OF AIR

607

plugged end, which is connected by stout rubber tubing to aspi-
rating flasks such as are used in Hesse's apparatus. The tube is
clamped horizontally to a retort stand, and by attaching the second
flask to a small hand exhaust-pump, the water can be syphoned
over from the first flask, a corresponding volume of air passing
through the tube. When the desired volume of air has been
aspirated through the tube, it is disconnected j-«rt^

and placed in another sterile tin box. As many
tubes as desired can be employed to control
one another or to examine the air in different
localities and under different conditions. All
the samples having been taken, the tubes are
manipulated on returning to the laboratory.
The tubes, as before, being handled by the
ends only, a file-mark is made across the centre
of each tube, which is then broken in half and
the plugs of glass-wool and sugar are shaken,
or pushed by means of a sterile wire, into a
sterile flask of about 250 c.c. capacity. Into
this 10 or 15 c.c. of liquefied sterile nutrient
gelatin are then introduced ; the sugar dis-
solves, the glass-wool becomes disintegrated,
and a roll-culture is made on the walls of the
flask, which is incubated at 22° C., and the
colonies are counted when they have deve-
loped.

(5) Sedgwick and Tucker's method. — One of
the best and most convenient methods for the
bacteriological examination of air. A glass

tube of special form is employed (Fig. 68) ; pIG> 68. Sedgwick

this consists of an expanded portion (A) about and Tucker's tube
15 cm. long and 4-5 cm. in diameter ; one end of for air analysis,
this is contracted so as to form a neck 2-5 cm. in
diameter and in length ; to the other end is fused a glass tube (B c)
15 cm. long and 0-5 cm. in diameter. The neck of the tube is plugged
with cotton-wool, and two cotton-wool — or, better, glass-wool — plugs
are inserted in the narrow tube, one at its open end, the other (c)
about 6 to 8 cm. from the wide part. The whole is then sterilised.
When cool, the narrow part of the tube, from its origin at the wide
part down to the first plug (c), is filled with powdered cane-sugar
(No. 50, B.P. gauge) which has been carefully dried and sterilised
at 120°-130° C. The tube is again sterilised at 120°-130° for two
or three hours, the greatest care being taken not to melt the sugar.

608 A MANUAL OF BACTERIOLOGY

After sterilisation the tube is ready for use. The wool plug is
removed from the mouth and a measured volume of air is aspirated
through the layer of powdered sugar by means of a small hand
air-pump, the volume of air being measured by the displacement
of water in a flask. Having taken the sample (5 to 20 litres), the
wool plug is replaced in the neck. The powdered sugar is then
shaken down into the wide part of the tube (A), and 15 c.c. of melted
sterile nutrient gelatin are poured in. The powdered sugar readily
dissolves in the melted gelatin, and when solution is complete a
roll-culture is made in the tube, just as in Esmarch's method (p. 83).
The tube is then placed in an incubator at 20° C., and the colonies
are allowed to develop.

In both Frankland's and Sedgwick and Tucker's methods the
sugar, after powdering and sifting and before introducing into the
tubes, should be thoroughly dried by keeping in the warm incubator
for several days ith occasional stirring. Unless this be done, the
sugar is apt to cake and discolour during sterilisation.

Soil

The upper layers of soil contain large numbers of organisms,
chiefly bacilli. The species are very varied ; among pathogenic
ones may be named the bacillus of tetanus and of malignant oedema.
The B. mycoides is very abundant, and the varieties of Proteus,
the hay and potato bacilli, are common, while the nitrifying forms
are of course present, but do not develop on ordinary media.

Below five or six feet aerobic organisms become scanty, but the
anaerobic and thermophilic ones are still met with. The number
of organisms present in soil is variable, from 200,000 to 45,000,000
in ordinary earth, while in dirty and busy streets there may be as
many as 1,000,000,000 per grm. According to Houston, unculti-
vated sandy soil averages 100,000, garden soil 1,500,000, and sewage
polluted 115,000,000 per grm.

Houston * found that in virgin soils the B. coli, B. Welchii, and
streptococci are practically absent, but that in soils polluted with
animal excrement by manuring or otherwise the spores of B. Welchii
are present in great abundance, also B. coli and streptococci if the
pollution be of recent date.

The length of time pathogenic bacteria retain their vitality in
buried corpses has been the subject of experiment by Losener,2

1 Rep. Med. Off. Loc. Gov. Board for 1889-1900.

2 Centr.f. Bakt. (VQ Abt.), xx, 1896, p. 454.

EXAMINATION OF SOIL 609

who injected cultures into the bodies of pigs, which were then
wrapped in linen, placed in wooden coffins, and buried. The
conclusions he arrived at were that, provided the soil has good
filtering properties, there is practically no chance of the dissemina-
tion of a virus.

Klein,1 experimenting with the bacilli of diphtheria, cholera,
plague, typhoid fever, etc., also found that the vitality and infective
power of these organisms passed away in a comparatively short
time, in most cases within a month.

On the survival of the typhoid and cholera organisms in soil
see also pp. 363 and 437 respectively.

Examination of Soil

The bacteria in the soil may be examined by adding traces of the
soil to sterile nutrient broth, thoroughly crushing and soaking it,
and then making plate or roll cultures, aerobic and anaerobic.

To make anything like an accurate quantitative examination is
almost impossible. Weighed amounts of the soil, after thorough
pulverisation in an agate mortar, may be introduced into sterile
test-tubes and thoroughly exhausted by repeated washing with
sterile water or broth, plate cultivations being made with the

Various forms of boring apparatus have been devised for with-
drawing soil from different depths.

Sewage 2

Sewage is exceptionally rich in organisms, but the numbers present
are variable. Jordan in Massachusetts found an average of 708,000
per cubic centimetre. Laws and Andrewes found from 905,000
to 11,216,000, the latter being the highest number obtained. The
number of organisms naturally varies at different seasons and
with the amount of dilution. The organisms present are very
varied, but moulds, yeasts, and sarcinse only occasionally occur.
A few micrococci are met with and streptococci are present in
considerable numbers, at least 1000 per c.c., but bacilli, especially
liquefying forms, largely predominate. The commonest species

1 Rep. Med. Off. Loc. Gov. Board for 1898-99, p. 344.

2 See various Reports to the London County Council by Clowes,
Houston, Laws and Andrewes ; Klein, Houston, Reps. Med. Off. Loc.
Gov. Board for 1897-1904 ; Rep. of the Sewage Commission.

39

610 A MANUAL OF BACTERIOLOGY

are the B. fluorescens liquefaciens and varieties, several varieties of
Proteus, the B. filamentosus, varieties of the B. mesentericus, B.
mycoides, B. subtilis, B. cloacce, and the colon bacillus. The latter
numbers from 20,000 to 2,000,000 per c.c., and the other bacilli
mentioned number 200,000 to 2,500,000 per c.c. Many anaerobic
sporing bacilli are also found, especially the B. Welchii, the spores
of which number from 30 to 2000 per c.c., averaging 500-600.
Foreign bacteria introduced into sewage are probably soon sup-
pressed by the predominant species of the sewage.

The air of well- ventilated sewers differs but little from that of
the external air, and the organisms in it contrast with those of
sewage by the abundance of moulds. Specific organisms may,
however, gain access to it (p. 365).

The powerful liquefying and solvent actions of the bacteria
present in sewage have suggested a means of dealing with sewage
so as to make use of these properties, and many bacterial systems
of sewage disposal have been devised. The principle most widely
adopted is to run the sewage into large covered reservoirs (septic
tanks), where it remains at rest for twenty-four to forty-eight hours.
Here it is under practically anaerobic conditions, and anaerobic
bacteria exert their action on the solids, partly dissolving them,
partly disintegrating them, with the formation of a sludge which
has to be cleared out from time to time. From the septic tanks
the sewage passes on to beds composed of broken brick, coke, or
some similar material, through which it slowly percolates, and
here it is subjected to the action of aerobic organisms, which com-
plete the decomposition to such an extent that the effluent does
not affect fish life nor putrefy, so that it may be run into a stream
without causing a nuisance. Four sets of these aerobic bacterial
beds are usually provided, each set being worked in turn for six
hours and resting for eighteen hours during the twenty-four hours.
The effluent from such bacterial beds may contain as many bacteria
as, or more than, the sewage itself. Pathogenic organisms may
be present in it, for Houston found that the B. pyocyaneus added
to the beds soon appeared in the effluent.

On the survival of the typhoid and cholera organisms in sewage
see pp. 363 and 437 respectively.

Examination of Sewage and Sewage Effluents

To ensure a fair average sample, the sewage or effluent should
be collected in small portions at intervals. The portions are mixed,

EXAMINATION OF SEWAGE

611

strained through muslin, and dilutions of 1 in 10, 1 in 100, 1 in 1000,
and 1 in 10,000 made with sterile tap-water. These are then
examined according to the following scheme :

Tests.

Procedure.

Amount of sewage in c.c.

1. Total number of

Gelatin and agar plate

0-001, 0-0001, 0-00001

bacteria

cultivations

2. Number of spores

Gelatin plate cultures

1-0, 0-1, 0-01

of aerobes

with material pre-

viously heated to 80°

C. for ten minutes.

3. Number of spores

Agar plate cultures

1-0, 0-1, 0-01

of anaerobes

with material pre-

viously heated to 80°

C. for ten minutes

and incubated anae-

robically

4. Number of organ-

Surface gelatin plates

0-001, 0-0001, 0-00001

isms liquefying

gelatin

5. Spores of B. Wel-

Milk cultures heated

0-1, 0-01, 0-001

chii (enter itidis

to 80° C. for ten

sporogenes)

minutes and incu-

bated anaerobically

6. Number of B. coli

Surface-plates of Con-

radi - Drigalski, or

0-001, 0-0001, 0-00001

bile-salt media, etc.,

as described for

water (p. 591)

7. Number of strep-

Surface-pl&tes of Con-

0-01, 0-001, 0-0001

tococci

radi - Drigalski me-

dium (p. 592)

EFFLUENTS ONLY.

8. Incubate some of the effluent in beakers at 22° C. and 37° C. for

some days. A good effluent should yield little or no unpleasant
odour (an unpleasant odour indicates the presence of decom-
posable organic matter, and such an effluent might give rise to
a nuisance).

9. Place a gold-fish or two in a bowl of the effluent. The fish will

live in, and be unaffected by, a satisfactory effluent. (This may
be done only by a licensee under the Vivisection Act.)

612 A MANUAL OF BACTERIOLOGY

Milk1

Milk is an admirable nutrient soil for the development
and multiplication of micro-organisms, and, though sterile
in the udder,2 as delivered to the consumer may contain
an appalling number of bacteria. In milk as ordinarily
supplied there are from one to five million bacteria per c.c.,
and it frequently contains ten to fifteen millions, with an
average of about three to four millions. Hewlett and
Barton found an average bacterial content of about
1,500,000 in London milk as delivered at the railway termini
(the range was from a minimum of 20,000 to a maximum
of 8,390,000), but this does not represent the condition of
the milk as delivered to the consumer, for the bacteria
present rapidly multiply in warm weather. Eyre 3 in
the middle of summer found the following rate of multi-
plication :

Microbes per c.c.

Initial content . . . 56,000

After 12 hours . . . 526,000

After 24 hours . . . 20,366,000

After 30 hours . . . clotted

A similar specimen in the middle of winter gave the
following results :

Microbes per c.c.

Initial content . . . 20,000

After 12 hours . . . 24,000

After 24 hours . . . 43,000

After 30 hours . . . 280,000

1 See Houston, Rep. to the London County Council, No. 933, 1905 ;
MacConkey, Journ. of Hygiene, vol. v, 1905, p. 333 ; Hewlett and
Barton, ibid. vol. vii, 1907, p. 22 ; Savage, Rep. Med. Off. Loc. Gov.
Board for 1909-10, p. 474, and Milk and the Public Health (Macmillan,
1912) ; Swithenbank and Newman, Bacteriology of Milk.

2 The " fore " milk may contain organisms which have lodged in the
milk-ducts, and it is extremely difficult to obtain completely sterile milk.

3 Journal of State Medicine, vol. xii, 1904, p. 728.

BACTERIAL CONTENT OF MILK 613

In New York, Park estimated the average bacterial
content of milk as supplied to the consumer at 1,000,000
per c.c. in winter and 5,000,000 per c.c. during the hot
months. Eyre (loc. cit.) states that, as the result of his
observations, the numbers are in London about 3,000,000
to 5,000,000 in December, January, and February, and
20,000,000 to 30,000,000 in June to September, smaller
numbers than these always being associated with the
presence of boric acid or formaldehyde. Even in so-called
sterilised milks bacteria are rarely completely absent.

Cream is even richer in bacteria than milk, and averages
about 8,000,000, and may contain as many as 30,000,000
organisms per c.c. Although all the ordinary species may
be met with, milk has a bacterial flora largely its own,
comprising many forms producing lactic and butyric acid
fermentations. Organisms also occur having more or
less specific effects, and giving rise to bitter milk, viscid
milk, etc. The lactic ferments are mostly non-sporing,
the butyric chiefly sporing, species. The commonest of
the lactic ferments are Streptococcus lacticus (non-gas-
forming) and B. acidi lactici (gas- forming), which has
some similarity to the colon bacillus (see Table, p. 381).
Another common lactic organism is the Oldium lactis, a
mycelial form, the colonies of which appear as little fluffy
tufts. In addition to the organisms named, pathogenic
species may be met with — viz. the tubercle, diphtheria,
typhoid, paratyphoid, Gartner, dysentery, and comma
bacilli, the M . melitensis, M. pyogenes, and the Streptococcus
pyogenes (lactic-acid-forming streptococci are also common).
The B. coli and B. Welchii are generally present in milk,
and the B. lactis aerogenes is sometimes found (p. 389).
Scarlatina (see " Scarlatina ") and foot-and-mouth disease
may likewise be conveyed by milk, and the diarrhoea of
infants is largely due to the use of milk swarming with
microbes, some of which in themselves may be harmful,

614 A MANUAL OF BACTERIOLOGY

and which also by the products they form tend to set up
gastro-enteritis. The percentage of samples infected with
tubercle bacilli varies much : Barton and Hewlett found
only one out of 26 samples taken at London railway
termini. The supply of the large dairy firms is also
comparatively free from tuberculous infection, as con-
siderable precautions are taken to exclude tuberculous
animals. For the quarter ending March 31, 1911, of
760 samples examined for the London County Council,
106, or 13-9 per cent., were found to be tuberculous, and
since 1907 of 5698 samples, 640, or 11-2 per cent., proved
tuberculous (see also p. 321). A poisonous body, tyro-
toxicon (p. 38) has been isolated from milk and milk
products. Sources of contamination and infection are
derived from the insanitary conditions of many farms and
dairies and the dirty methods of those handling the milk.
In order to render milk wholesome for infants and free
from infective organisms under the present conditions of
supply, two methods may be adopted — sterilisation and
pasteurisation. To ensure sterilisation it is necessary to
heat the milk to boiling-point for six hours, or to expose
it for a shorter period to steam under pressure. Such
treatment, . however, markedly alters the flavour of the
milk, and is said to diminish its nutritive value. If the
milk be heated to a temperature not exceeding 70° C.,
the flavour and nutritive qualities are far less altered, while
the pathogenic species are all destroyed. This method is
termed " pasteurisation," and consists in heating the milk
to about 60°-68° C. for twenty to thirty minutes. Pas-
teurisation destroys 92-99 per cent, of the total organisms
present. The objections to pasteurised milk are that the
natural enzymes present in fresh milk are destroyed and
such heated milk is stated to induce scurvy rickets,1 the

1 Dr. Lane-Claypon denies this, and considers that the enzymes in milk
are derived from the bacteria in it (Rep. to theLoc. Gov. Board, 1913).

BACTERIOLOGY OF MILK

615

lactic-acid-forming organisms are killed, and if the treated
milk be kept, the residuum of resistant putrefactive, etc.,
bacteria multiply enormously, without obvious change in
the milk and " returned " milk can be utilised again and
again. Pasteurised milk should be rapidly cooled and be
consumed within twenty-four hours of treatment. Behring
has advocated the addition of formaldehyde to all milk
used for the feeding of children. Another method for
sterilising milk is the Budde process,1 in which the milk,
after the addition of hydrogen peroxide, is heated for
three hours to 52°-53° C. All non-sporing organisms are
destroyed, and the added hydrogen peroxide is decomposed
into H90 and 0.

All milk should be distributed in closed bottles, and
pasteurised milk should be consumed within thirty-six
hours of treatment.

The thermal death -point of pathogenic organisms in milk is
as follows : 2

Organism.

Temperature.

Period of Exposure.

B. tuberculosis

60° C.

20 min.

B. typliosus

60° C.

2 min.

B. cliphtherice .

60° C.

1 min.

Spir. cholerce .

60° C.

1 min.

B. dysenteries .

60° C.

10 min.

M.fnelitensis .

60° C.

20 min.

The thermal death-point of tubercle bacillus, especially in milk
has been the subject of some controversy (see also p. 309). De Man
found that an exposure of fifteen minutes at 65° C. was necessary
l£> destroy the infective properties of tuberculous milk. Bang, of
Copenhagen, considers that pasteurisation cannot always be relied
upon, and recommends that milk should be heated to 85° C. The
writer found that the vitality of the ordinary non-virulent laboratory
cultures was destroyed by a temperature of 60° C. acting for ten

1 Hewlett, Lancet, 1906, vol. i, January 27.

2 Rosenau, Hygienic Lab., Washington, Bull. 42, 1908.

616 A MANUAL OF BACTERIOLOGY

minutes, and that the infective properties of tuberculous sputum,
tested on guinea-pigs, were destroyed by a temperature of 65° C.
acting for fifteen minutes in five out of six instances. Woodhead s
experiments (First Royal Commission on Tuberculosis) gave irregular
results which seem to be explained by Theobald Smith's careful
work.1 This showed that tuberculous milk was rendered non-
infective by heating to 60° C. for ten to fifteen minutes, provided
there was no formation of a surface scum ; the latter seems to protect
the bacilli. Russell and Hastings 2 confirmed Smith's experiments,
and assert that it is sufficient to heat milk to 60° C. (140° F.) in
a closed receptacle for a period of not less than twenty minutes in
order to destroy the tubercle bacillus. The surface scum forms
on milk only when it is heated in contact with air ; all pasteurisers,
therefore, should be closed vessels. The writer has devised a
simple form of domestic pasteuriser, which is made by Messrs.
Allen and Hanbury.

The occurrence of so-called leucocytes and pus-cells in
milk must be considered. A certain number of cells
resembling polymorphonuclear leucocytes are always
present in milk, more numerous during the first week of
lactation and then accompanied by colostrum corpuscles.
An excess of these cells may indicate some local inflamma-
tory affection of the udder, or, if streptococci and blood
are present in addition, suppuration, but not necessarily,
for Russell and Hoffman, and Revis have shown that a
very large cell count (500,000-1,000,000, or even 10,000,000,
per c.c.) may often be obtained from quite healthy cows.
The nature of these cells has been the subject of an extended
investigation by Hewlett, Villar, and Revis.3 Their con-
clusion is that the majority of these cells are not leucocytes,
but are germinal cells of the secreting epithelium of the
udder. Blood may also be present transitorily in health
(Revis). The presence of squamous epithelial cells indi-
cates desquamation from the teat or udder or from the
hand of the milker — i.e. want of cleanliness.

1 Journ. Exper. Med., vol. iv, 1899, p. 217.

2 17 ih Ann. Rep. Wisconsin Agricult. Exp. Station.

3 Journ. of Hygiene, vols. ix, x, xi, and xiii.

SOUR MILK 617

There is no doubt that micro-organisms are far more
abundant in milk as supplied to the consumer than should
be. This arises from the ignorance and carelessness of
those charged with the duty of providing and distributing
this important article of diet. The udder and teats of
the cow and the hands of the milker (who should wear a
special dress) should be wiped before milking, and all
vessels should be clean and steamed or scalded before use.
The milk should be cooled at once, some more efficiently
closed vessel than the present form of milk churn adopted,
and the milk not stored, but forwarded without delay
by the railway companies in special refrigerator vans.
Distribution in bottles would be a great improvement.

The following might be suggested as a bacteriological
standard for milk : * (a) Number of organisms not to
exceed 1,000,000 per c.c. ; (b) absence of excess of leu-
cocytes or of pus- cells ; (c) B. coli, B. Welchii, and strep-
tococci should not be present in 1 c.c. or less; (d) the
sediment after centrifuging should be less than 100
parts per million ; (e) the milk as delivered should not
have a temperature above 10° C. ; (/) absence of pathogenic
organisms.

Sour milk. — Sour milk is used as an article of diet in many parts
of the world, e.g. Bulgaria. In these sour milks a particular micro-
organism or a variety of it, the B. bulgaricus or " bacillus of Massol,"
is generally present in association with lactic streptococci. It is
a large, pleomorphic, Gram-positive bacillus, non-motile, non-
sporing, growing best at about 40° C., but only in milk or in culture
media made with milk or whey. It has been much employed for
the preparation of a soured milk which is of considerable service
in the treatment of certain disorders.2

1 See " Rep. of a Committee on Milk Supply," Philad. Med. Journ.,
October 1900, p. 758 ; Park, Journ. of Hygiene, vol. i, 1901, p. 391 ;
Houston, loc. cit.

2 See Hewlett and others, Brit. Med. Journ., 1910, vol. ii (Bibliog.).

618 A MANUAL OF BACTERIOLOGY

Examination of Milk

Number of organisms per c.c. — This is carried out by diluting the
milk to 1 in 1000 — 1 in 1,000,000 with sterile water, or preferably
nutrient broth, as a better mixture is obtained. Plates are then
made either in gelatin or in distilled water agar (1| grm. powdered
agar, distilled water 1 litre, Eastes), or preferably in both media.

B. coli, B. Welchii, and streptococci. — These are searched for
quantitatively by the methods detailed for " Water " (pp. 576-586).
Amounts of milk in decreasing decimal order from 100 c.c. to
0-000001 c.c. should be examined. The B. coli must be differentiated
from B. lactis aerogenes and B. acidi lactici (see pp. 389, 381).

Pathogenic organisms. — The detection of these, with the exception
of the tubercle bacillus, is difficult and uncertain. In all cases
the milk should be centrifuged and the deposit examined.

1. For the detection of the tubercle bacillus * staining methods
are almost useless (except in cases of advanced tuberculosis of the
udder or when the milk of a single cow is examined) and inoculation
must be performed. At least 250 c.c. of the milk should be centri-
fuged at 2000 to 2500 revolutions per minute for an hour. As
many organisms become entangled in the cream, it is advisable to
stop the machine after half an hour, stir in the cream, and again
centrifuge. The fluid is poured or pipetted off carefully, so as
not to disturb the sediment, leaving about 3 c.c. in the tube. The
sediment and the remaining fluid are then well mixed and about
1 c.c. is inoculated subcutaneously and intraperitoneally into
two guinea-pigs respectively (see also p. 329). For staining, a
process of solution of the milk may be employed, 20 c.c. of the milk
being mixed with 1 c.c. of a 50 per cent, potash solution, and heated
in a water-bath until the solution turns brownish ; 20 c.c. of acetic
acid are then added. The mixture is shaken, heated in a water-
bath for three minutes, and centrifuged for ten minutes. The
fluid is poured off, 30 c.c. of hot water are added to the sediment,
and the mixture is again centrifuged. Films are then prepared
from the sediment, and stained for the tubercle bacillus (see also
p. 325), the films being always treated with alcohol as well as with
acid.

Non-pathogenic acid-fast bacilli occur in milk (p. 340).
2. The diphtheria bacillus is searched for by making serum
cultures from, and inoculating guinea-pigs with, the sediment.

1 See Delepine, Rep. Med. Off. Loc. Gov. Board for 1908-09, p. 134.

EXAMINATION OF MILK 619

If a diphtheroid organism is detected it must be isolated and
examined by culture tests and animal inoculation.

In milk and cheese a bacillus is frequently met with closely resembling
the diphtheria bacillus in its morphological and cultural characters ,
it is, however, quite non-pathogenic.1

3. The typhoid, paratyphoid, Gartner, and dysentery bacilli
and cholera vibrio may be searched for by the methods given for
" Water."

(4) The M. pyogenes and the Streptococcus pyogenes may be
searched for by means of plate cultures on glycerin agar.

(5) Examination of sediment. — Houston and Savage (loc. cit.)
have devised methods for the quantitative estimation of the sedi-
ment by centrifuging in special graduated tubes. For the micro-
scopical examination of the sediment the milk is centrifuged
for twenty minutes at 1500 revolutions per minute, and the upper
fluid is pipetted or syphoned off. Some of the sediment should be
examined with the f in. and £ in. objectives for the presence of
" dirt," e.g. hairs, straw, etc. Three smear preparations are then
made, each with four drops of the sediment, which are spread evenly
over three-fourths of the slide. The slides are air-dried, and may
be treated with a mixture of absolute alcohol and ether for ten
minutes. One slide is stained with Loffler's blue, another by
Gram's method for streptococci, and a third by the tubercle method.
The Loffler's blue specimen gives a general idea of the number of
bacteria present, and of the presence of cells.

From what has been said above (p. 616), considerable caution
must be exercised in stating the presence of pus-cells. Streptococci
present are not necessarily pathogenic, as non-pathogenic lactic-
acid-forming streptococci are common. For counting the number
of cells present, Revis 2 employs a centrifuge tube of 10 c.c. capacity,
the lower third of which is contracted to 0-8 cm. in diameter, and
contains 1 c.c. The procedure is as follows :

In the tube are placed 5 c.c. of the well-mixed milk, diluted to
the 10 c.c. mark with 0-8 per cent, salt solution. After inserting
a rubber stopper the contents are well mixed. The tube is then
centrifuged at about 2000 revolutions per minute for two minutes,
the cream is broken up by violently shaking the upper part of the
tube, and the rotation continued for four minutes longer. A glass
rod, fitting roughly the narrow neck of the tube, is inserted, and
the major part of the milk poured off, and the upper part of the
tube well rinsed with water to remove cream, etc. ; the contents

1 See Scientific Bull. No. 2, Health Dept., City of New York, 1895, p. 10.

2 Journ. of Hygiene, vol. x, 1910, p. 58.

620 A MANUAL OF BACTERIOLOGY

of the narrow end down to within \ in. of the deposit are sucked out
with a fine glass pipette, the upper part of the tube is wiped clean
and the tube is then filled to the 10 c.c. mark with salt solution.
The tube, having been violently shaken till all the deposit is dis-
tributed through the liquid, is then rotated for four minutes, and
the liquid down to within | in. of the deposit again removed. In
the case of small deposits, two to three drops of saturated aqueous
solution of methylene-blue are added, and the deposit is stirred
up by blowing through a fine glass capillary pipette (which is
afterwards used for filling the counting chamber). After fifteen
minutes, water is added to the 1 c.c. mark, and counting done in
the usual way with a Thoma-Zeiss blood counter. Counting should
not be restricted to the ruled spaces, but the field should be so ar-
ranged that a definite number of squares is included, and fields are
counted all over the chamber. At least two different preparations
should be made of the same deposit for counting.

FOOD PoisoNiNG.1 — Apart from the presence of the ordinary
poisons, food may be poisonous on eating — (a) naturally, e.g. certain
fish, (6) from the results of the activity of micro-organisms with
the formation of toxic products, the ordinary " ptomine poisoning "
(see p. 38), in which case the poison is pre-formed and is ingested,
(c) from infection with certain organisms, particularly B. enteritidis,
which generally induce gastro-enteritis. In the last named, symp-
toms do not usually ensue until a lapse of twelve to forty-eight
hours after the consumption of the food. Mayer and Mandel
describe an outbreak following the consumption of broiled fish,
in which B. proteus was isolated from the stools and was agglutinated
by the patients' serum.

Meat is not likely to convey any infective disease with the excep-
tion of tuberculosis and anthrax. It may be examined by cultures
and plate cultivations, and by inoculation and feeding experiments.
Tinned meats, etc., frequently contain sporing organisms of the
B. subtilis and mesentericus groups. They may be examined by
aerobic and anaerobic cultures, and by feeding mice. Poisonous
ptomines are occasionally present. The B. enteritidis occurs in
meat, and causes a form of poisoning (see p. 371 ).2 In certain
intoxications due to bad meat, known as " botulism," Van Ermengen
isolated the B. botulinus (see p. 427).

Bread. — Troitzki states that new bread contains no micro-
organisms, but Waldo and Walsh found that such organisms as
the comma bacillus are not destroyed by passing through the ordeal

1 See Savage, Hep. to the LOG. Gov. Board, No. 77, 1913.

2 See Savage, Eep. Med. Off. Loc. Gov. Board for 1909-10, p. 446.

oes not grow at
turbidity in broth,

23 13 -g

§.2 .-g

5£l°~

C 3 Co +-

•SgsS*

ca-S^S

o S a %

K-SJ o

- >>•*> ^4:

I!!1!
ii?!l

S ^2 tC ^ 'c8

o a

K

5I

• sT

It

,

iJ I!

'

O

OQ

W 02

il

s

.

m

it ii

-UIBJf)

+ + +

•uoijmuioj
oiodg

+ + +

Organism
and its size.

s 1 1!

•3 a5 a

•§

"2 *u

622 A MANUAL OF BACTERIOLOGY

of the baker's oven. Cut bread forms a good nidus for the develop-
ment of pathogenic organisms.

The Bacillus prodigiosus may grow upon various food-stuffs, and
give rise to suspicion of foul play. L. Parkes x describes cases of
diarrhoea which he suggests were caused by this organism.

Butter contains from two to forty-seven millions of micro-
organisms per gramme. Tubercle bacilli have been found in butter,
and the comma bacillus artificially introduced survives for over a
month. " Acid-fast " non-pathogenic forms also occur (p. 340).

For the isolation of the tubercle bacillus from butter and cheese
the only certain method is by inoculation. Butter may be melted
and allowed to stand in the incubator at 37° C. for some days, and
the sediment inoculated. As this involves the multiplication of
septic organisms, it is preferable to centrifuge the melted butter,
keeping it melted during the process, and to inoculate the sediment
immediately.

Clothing, etc. — Attempts have been made to examine clothing,
bedding, flock, etc., by bacteriological methods for filth contamina-
tion, but without much success.

1 Brit. Med. Journ., 1905, vol. ii, 1330.

CHAPTER XXII

HEAT — STEAM DISINFECTION — CHEMICAL DISINFECT-
ANTS— THEORY OF DISINFECTION— METHODS OF
DETERMINING DISINFECTANT POWER

Disinfection x

NATURAL agencies restrain the multiplication of disease
organisms, but enough survive to determine the persistence
of infective diseases, and to call for measures by which
communities attempt to cope with them. These measures
are broadly isolation, prophylactic inoculation, general
improvement in sanitation and nutrition, and disinfection.
In the present chapter the methods by which the fourth
means of protection may be applied are considered. Dis-
infection implies the removal or the destruction of infective
properties, but, for practical purposes, should be understood
to mean the killing of the infective organisms to which
those properties are due. For this purpose, the two
agencies ordinarily used are heat and chemical action,
though, in addition, other methods can occasionally be
employed for destroying or excluding micro-organisms.
Such are light, desiccation, and filtration.

HEAT. — Fire is the simplest and most efficient agent
for destroying infective matter. Burning should always
be employed where possible, as for rags, old clothing or
bedding, native huts, etc.

For surfaces which would not be unduly injured, such
as stables, pens, yards, etc., a torch-fire generated by means

1 See Hewlett, " Milroy Lectures," Lancet, 1909, vol. i.
623

624 A MANUAL OF BACTERIOLOGY

of the cyclone burner described by Forbush and Fernald
has been favourably reported on by Stiles. The apparatus
consists of a portable tank, from which paraffin gas oil
is driven by a pump through a hose (such as is used for the
delivery of oil) to which is attached a pole, consisting of
an iron pipe 12 ft. long, which is protected by a covering
of wood, and to the end of which is attached a cyclone
nozzle. The fine spray from the nozzle is ignited, and the
resulting fierce flame passed over the surfaces to be dis-
infected. The thorough wetting with water of all such
surfaces would practically abolish danger from fire, and
by proper adjustment of the power of the flame, and
experience on the part of the operator, the method is an
efficient one.

Dry heat may also be used, and forms the basis of some
disinfectors (Ransome's), but is not nearly such an efficient
means as moist heat. The objections to dry heat are,
that to ensure the destruction of bacteria and spores the
temperature must be high and the heating prolonged.
Koch and Wolfhiigel found that two hours at 150° C. did
not always ensure sterilisation, and Gaflky and Loffler
state that the spores of some organisms are killed only
by exposure to hot air at 140° C. for three hours. Moreover,
dry heat has little power of penetration, and it requires
many hours for the centre of a mass of bedding, or the
like, to attain the temperature requisite for sterilisation,
while some articles and fabrics are distinctly injured by
the prolonged heating. The highest temperature which
can be safely adopted for a dry-heat disinfector is about
120° C., and then if large masses have to be treated the
heating has to be continued for from eight to ten hours.
A rise of 5° C. above this temperature is sufficient to
damage many woollen goods, which enhances the objections
to a dry-heat disinfector, as it is difficult to keep the
temperature of a large chamber constant.

STEAM DISINFECTION 625

For the reasons given above, disinfection by dry heat
is often impracticable ; on the other hand, moist heat is
more effective, is found to work well in practice, and is
now generally adopted. In the household, for articles
which cannot be burnt, brisk boiling for an hour or so will
suffice.

Steam disinfection. — For public disinfectors, steam under
pressure — i.e. at a pressure greater than that of the atmo-
sphere— is employed. Steam under pressure has not such
a deleterious action on articles, with the exception of
leather, as dry heat, while its penetrating powers are far
greater. By " saturated steam " is meant steam at the
temperature at which it can condense, and the tempera-
ture of the condensation point rises as the pressure increases.
By " superheated steam " is meant steam at a temperature
higher than that at which it can condense ; therefore
superheated steam has to be cooled down into the state of
saturated steam before condensation ensues. If super-
heated steam is used for disinfection, it loses heat by
conduction, and the rise in temperature of the articles
treated approximately corresponds to the fall in tem-
perature of the steam. With saturated steam, on the
other hand, immediately it is cooled an enormous amount
of latent heat is set free by the change in state from the
gaseous to the liquid condition, therefore saturated steam
is a far more efficient disinfectant than superheated steam.
These considerations should always influence the choice
of a steam disinfecting apparatus for efficient working.

The Equifex disinfector is worked with saturated steam
at 10 Ib. pressure (239° F.). The chamber consists of a
cylinder of mild steel, made without steam jacket, so as
to avoid risk of superheating. The cylinder is lagged with
non-conducting composition and wood, to reduce loss of
heat by radiation, and, as usually supplied, is furnished
with separate doors for infected and disinfected articles

40

626 A MANUAL OF BACTERIOLOGY

respectively. An arrangement can be supplied to prevent
both doors being opened simultaneously. The Washington-
Lyons apparatus, or its modifications, is an elongated
cylindrical boiler with double walls, forming a jacket, and
a door at each end. The chamber is of sufficient size to
admit bedding, and is built into the partition wall between
two rooms, so that each door opens into a different room.
Into one of the rooms the infected articles are conveyed,
and are placed in the disinfector as lightly packed as
possible ; when disinfected they are removed by the
opposite door into the other room, thereby avoiding all
chance of reinfection. Steam at a pressure of about 20 Ib.
is admitted into the jacket and then passes to the inner
chamber, the object of the jacket being to warm the
chamber, and so prevent condensation. For the same
purpose hot air is sometimes injected beforehand to warm
the chamber and articles, and after the steam disinfection,
can again be injected for drying. The length of time re-
quired for disinfection does not exceed a half to one hour.
In Thresh's disinfector the steam is generated from a
saline solution (calcium chloride), which has a boiling-
point0 (105 C.) higher than that of water.

The thermal death-point of a number of organisms in pure
culture has been determined by many investigators. Eyre suggests
the following as " standard conditions " for determining thermal
death-points :

1. Length of " time exposure " to be ten minutes.

2. Emulsion to be prepared from " optimum cultivation."

3. The vehicle in which culture is suspended to be sterile salt
solution or sterile distilled water.

4. Strength of emulsion to correspond to about 1 milligramme of
culture per cubic centimetre.

5. Bulk of emulsion to be not less than 3 c.c.

6. Emulsion to be contained in test-tube of 1-5 cm. diameter with
walls 1 mm. thick.

7. Emulsion to be exposed to moist heat in a water-bath regulated
by a delicate and accurate thermo -regulator.

CHEMICAL DISINFECTANTS 627

8. Broth cultivations and agar plates both to be used in deter-
mining the death of the bacteria, and the period of observation of
these cultures to be extended, when necessary, to seven or fourteen
days. The experiments to be repeated at least once.

9. Thermal death-point to be first roughly determined to within
5° C.

10. Thermal death -point to be finally determined to within
1° C., and to be defined as that temperature which causes the
death of all micro-organisms exposed to it, within the ten minutes
in these standard conditions.

LIGHT is not used directly for disinfection, but indirectly
in nature and in our homes may not be an unimportant
factor. It has previously been referred to at p. 23. Sun-
light, and artificial light rich in violet and ultra-violet radia-
tions, such as that emitted by a quartz mercury vapour
lamp, are efficient germicides. The latter has been
tested by Barnard and the writer with excellent results,
but, unfortunately, the germicidal rays have practically no
power of penetration and are stopped even by thin glass.

DESICCATION, although one of Nature's methods of
disinfection, is not made use of to any extent by man
except as an inhibitory agent for the preservation of many
articles of food. Shattock and Dudgeon found that many
bacteria, e.g. B. coli. and B. typhosus, rapidly succumb to
complete desiccation, but B. pyocyaneus maintained its
vitality for two years under these conditions.

FILTRATION is a method of disinfection by exclusion,
and in the form of sand filtration and filtration through
porous porcelain, as in the Berkefeld and Pasteur-Chamber'
land filters, is made use of for the sterilisation of water
and other fluids.

CHEMICAL DISINFECTANTS. — A large number of chemical
substances variously known as germicides, antiseptics,
disinfectants, deodorants, etc., have the power of inter-
fering with, or masking the results of, the vital activities
of micro-organisms. Germicides are substances which

628 A MANUAL OF BACTERIOLOGY

kill bacteria or germs ; antiseptics, by inhibiting bacterial
development, prevent sepsis or putrefaction ; and by
" disinfectant " is meant a substance which prevents the
action of, or destroys, infective matters, while deodorants
destroy or absorb foul-smelling gases the result of putre-
factive and similar processes. All germicides are disinfec-
tant and antiseptic, but many antiseptics, though pre-
venting or inhibiting the development of bacteria, are not
necessarily germicidal.

Many deodorants act largely mechanically, and although
often not germicidal, and hence not ideal disinfectants,
are of some value in preventing the deleterious and depres-
sing effects of the emanations from decomposing organic
matter. Such are charcoal, ashes, dry mould, and peat
(peat has also a germicidal action). Other deodorants,
such as quicklime and chloride of lime, act chemically.

The germicides and antiseptics may be considered together,
for although many antiseptics are not germicidal, all the
germicides in small amounts act as antiseptics. The prin-
cipal germicides and antiseptics are the halogen elements,
the mineral acids, a large number of metallic salts, phenol
and many coal-tar derivatives, and various organic bodies
and essential oils.

Theory of chemical disinfection. — The theory of chemical
disinfection is not yet fully understood. It is probable,
as suggested by Paul and Kronig, that the degree of ionisa-
tion of a solution may have an important bearing on its
disinfecting efficiency.

Paul and Kronig l made a number of experiments on
the M. pyogenes, and spores of anthrax, with a view of
determining the effects of various acids, bases, oxidising
agents, and metallic salts on bacteria. The salts of
mercury, gold and silver exert a marked germicidal action,
strongest in the case of mercury, while the platinum salts

1 Zeitschr.f. physikal. Chem.: 1896, xxi, p. 414.

THEORY OF DISINFECTION 629

are almost inactive. The efficiency of mercuric chloride
is markedly lessened by the presence of sodium chloride
or other chlorides. Of the oxidising agents, nitric, chromic,
chloric, and permanganic acids act in the order stated ;
chlorine has the most powerful action of the halogens.
Phenol acts better in a 5 per cent, solution than in higher
concentrations, and the efficiency is increased by the
addition of sodium chloride, but diminished by the presence
of alcohol, and under the most favourable conditions it is
not such a powerful germicide as mercuric chloride. Mer-
curic chloride dissolved in absolute alcohol has little or no
efficiency, and the addition of sodium chloride reduces its
activity. Organisms in masses are less readily acted upon
by antiseptics than when they are isolated.

The efficiency of a germicidal salt in solution seems to
vary with its dissociation. It is believed that the molecules
of a salt in solution are more or less dissociated into con-
stituent electrified atoms or " ions," and the greater the
dissociation the more active will the substance be as a
germicide. Taking mercuric chloride, bromide and cyanide,
it is found that the ionisation of the chloride is greater
than that of the bromide, and this is more ionised than
the cyanide, and the following results show that the
germicidal power of the three is in this order : 1

Number of
colonies which developed.

After After

Solution. 20 minutes' 85 minutes'

treatment. treatment.

1 mole HgCl2 in 64 litres . 7 0

1 „ HgBr2 „ „ . 34 0

1 „ Hg(CN)2 in 16 litres 8 33

Since the amount of this dissociation may be greatly
influenced by the presence of other substances, much

1 Findlay, Physical Chemistry, 1905.

630 A MANUAL OF BACTERIOLOGY

caution should be exercised in adding salts, etc., to increase
solubility or prevent precipitation, as the addition may
seriously impair germicidal or antiseptic power (see p. 635).
The disinfection process is a gradual one. In the early
stages of disinfection large numbers of organisms are killed,
but the rate of killing becomes slower and slower as time
elapses. Madsen and Nyman and Miss Chick 1 have found
that if the results be plotted, ordinates representing the
numbers of surviving bacteria, and abscissa the corre-
sponding times, the points lie on a logarithmic curve. The
curve so obtained, in fact, appears to be similar in form
to that of a " unimolecular reaction," and may be expressed

1 n

by the formula - - log - = K, where % and n2 are
t2-t± &^2

the numbers of bacteria surviving after times t± and t2
respectively, and K is a constant. In the case of disinfec-
tion of anthrax spores with phenol, Miss Chick found the
mean value of K to be 0'44. In the case of B. paratyphosus,
however, the course of the disinfection is different unless
the culture is very young, and Miss Chick concluded that
the older individuals are less resistant than the younger.
The progress of heat disinfection apparently follows the
same course. Miss Chick asserts that the act of disinfec-
tion is a unimolecular reaction, but it is difficult to accept
this view. Disinfectants in emulsion tend to be more
efficient than when in solution.

Factors modifying disinfectant action.2 — The efficiency of
a disinfectant liquid partly depends on its concentration.
The rate of penetration into bacterial cells decreases as
the concentration increases above a certain limit. Most
disinfectants yield, therefore, a greater amount of disin-
fectant energy per gramme-hour in dilute than in strong

1 Journ. of Hygiene, vol. viii, 1908, p. 92 (Summary and Bihliog.).

2 This section is largely taken from Applied Bacteriology, Moor and
Hewlett, 1907.

FACTORS MODIFYING DISINFECTION 631

solutions. In oil, glycerin, or alcohol, disinfectants lose
some or most of their activity. Spores in anhydrous
glycerin, oil, or vaseline, are not killed at a temperature
lower than 170° C. acting for half an hour.1 Of fats,
lanolin alone seems compatible with disinfectant efficiency.
Some disinfectants form an emulsion on the addition of
water, and their efficiency for a given amount of active
material may vary within wide limits according to the
manner in which they are emulsified. The temperature
at which the organism is exposed to the disinfectant has
a considerable influence on the extent or rate of disinfec-
tion. Up to the optimum temperature at which the
organism to be disinfected grows on the medium in which
it is exposed the activity of a disinfectant may fall off
as the temperature rises, owing to the increased vigour
which the organism derives from the improvement in
its conditions in respect of temperature. A relatively
small difference of temperature — two or three degrees —
may make an appreciable difference in the activity of the
disinfectant, and in the examination of disinfectants the
failure to remember this fact has led to serious error.
Above the optimum a rise of temperature increases the
activity of the disinfectant, sometimes to an enormous
extent. The same is sometimes the case even at tem-
peratures below the optimum, when the organism is in
unfavourable conditions for growth. A mixture of dis-
infectants in many cases has a more powerful effect than
can be produced by either separately (Chamberland). The
resistance of bacteria to disinfection by chemical agencies
is extremely variable and is also selective. Bacteria of
one class may be many times more sensitive to one
disinfectant than to another when both substances
exert an equal effect on bacteria of another class. The
presence of organic matter may profoundly modify the
1 Bullock Journ. of Hygiene, xiii, 1913, p. 168.

632 A MANUAL OF BACTERIOLOGY

action of chemical disinfectants, particularly those acting
by oxidation, considerably reducing their efficiency.

Requirements for an efficient disinfectant. — The conditions
which should be satisfied by an efficient disinfectant for
general use are simple, but not easy to obtain. Because
a disinfectant effect depends on the strength of the solution,
the substance should have an approximately definite
efficiency for particular organisms in given conditions,
and for the same reason it should be permanently homo-
geneous. In practice disinfectants must be used with
water or in an aqueous solvent ; it should, therefore, yield
a stable solution or uniform emulsion in all proportions.
Because bacteria as presented for practical disinfection
usually have some organic coating, it should be stable in
the presence of organic matter ; and as this coating is
often of a greasy character, it should, especially if intended
for use on dirty or greasy surfaces, have high solvent power
for grease. For use when heat can also be applied, whereby
its activity is enhanced, unless it breaks up, it should be
stable at all reasonable temperatures. These conditions
may be considered to be indispensable. It is further
desirable that it should have a sufficiently high specific
efficiency to allow of its being used in a readily diffusible dilu-
tion ; that it should yield a cheap solution or emulsion, not
act on metals, and be neither caustic nor toxic. Some dis-
fectant substances may now be considered more in detail.

Acids. — All acids have disinfectant action, and their
relative values are interesting in the respect that for them
a general law has been fairly well established by Von
Lingelsheim, and confirmed by Boer — namely, that the effi-
ciency varies with the degree of acidity. Solutions of acids
not of equal percentage concentration, but of equal acidity,
have approximately the same disinfectant efficiency what-
ever may be the acid, and whether it be inorganic or organic.

The acids have no great practical application in dis-

ACIDS AND ALKALIES 633

infection. That which has been most commonly used is
sulphurous acid, applied either direct from burning of
sulphur (in which case it will also contain S03 if there is
sufficient moisture to hold the sulphur dioxide in solution)
or by the use of the liquefied gas. It produces a slow
superficial disinfection of a weak and uncertain character
even under laboratory conditions. Such experiments
avoid, however, to a far greater extent than is possible in
practice the difficulty of diffusion, and the unequal diffusion
of sulphurous acid in air and its small power of penetration
make it less efficient in practice. To obtain even the
poor efficiency which is its maximum possible it is neces-
sary for the air to be damp and the room most carefully
sealed, and in these conditions it is often more injurious
to the objects under treatment than to the bacteria against
which it is directed. One of the most efficient methods of
applying sulphurous acid disinfection is by means of the
Clayton apparatus. The gas is generated by burning
sulphur in a current of air at a high temperature, and
contains, in addition to S02, traces of higher oxides of
sulphur. It is also a very efficient vermin-killer, destroying
rats, cockroaches, bugs, fleas, flies, etc.

Alkalies and soaps. — The degree of alkalinity of a solu-
tion affects, but does not by itself altogether determine,
its germicidal power, which is also dependent on the nature
of its metal. The hydrates of thallium, lithium, barium,
calcium, potassium, sodium, and ammonium have widely
different efficiencies, roughly in the order named. For
practical purposes only those of potassium, sodium, and
calcium need be considered.1 They exhibit notably the
characteristic of all disinfectants in that they work much
more vigorously in hot than in cold solution. It is to the
hydrates or alkaline carbonates of potassium and sodium

1 See Forrest and Hewlett, Journ. Roy. Army Med. Corps, February
1904.

634 A MANUAL OF BACTERIOLOGY

that the soaps owe such power as they possess against
naked organisms. The relative efficiency of soaps in
practical disinfection may be understated by the results
of comparative experiment on laboratory cultures because
the resistance of the microbe itself to disinfection by
chemical substances, and, indeed, by other agencies, may
be small compared with the resistance offered by the
envelope of grease or greasy dirt, derived from perspiration,
pus, fat, and the oily grime which pervades cities and is
everywhere caused by handling. A disinfectant of greater
efficiency than soap on a laboratory culture may, therefore,
be of much less efficiency on an infection in actual practice.
Soaps are incompatible with most disinfectant substances,
but not with all. Biniodide of mercury can be prepared
with soap, and for surgical purposes is a disinfectant of
high value. The " carbolic soaps " of commerce are, for
the most part, worthless.

Caustic lime, used generally as a 20 per cent, milk, has
considerable disinfectant power, and has been applied to
the disinfection of fseces. For this purpose care has to
be taken to break up any lumps of excreta, and whenever
practicable a heat process, of which the efficiency and
rapidity may be greatly increased by an alkaline disinfec-
tant, is much to be preferred. Lime is inefficient against
»the more resistant organisms, and lime-washing cannot
be considered a sufficient precaution against them or
against infections, such as those of scarlet fever and small-
pox, of which the exciting organism is unknown.

Halogens. — The disinfectant values of dry chlorine,
iodine, and bromine are low. Both in a dry and a damp
state chlorine is inconvenient, and the others are costly ;
and the use of halogens is therefore practically confined
to solutions, notably " chloride of lime " (a mixture of
calcium hypochlorite, hydrate, and chloride) and hypo-
chlorite of soda (chloros). These have a powerful effect on

HALOGENS 635

laboratory cultures, but in practice need to be used in
excess proportionate to the amount of organic matter
which may be present. Thus, for instance, a 1 per cent,
solution of hypochlorite of soda mixed with an equal
volume of urine loses the whole of its available chlorine
almost immediately, and becomes inert as a germicide.
Where the amount of organic matter is small, and the
objects are not likely to be injured, the hypochlorites are
among the best of known disinfectants, provided they
are used fresh. The slow addition of hydrochloric acid,
yielding nascent chlorine, increases the activity of a hypo-
chlorite considerably. A solution of iodine is now used
for skin disinfection in surgical practice. Iodine trichloride
is a powerful disinfectant, of which the use has been
suggested, among other purposes, for the sterilisation of
water. Nessfield has suggested the use of chlorine for
sterilising water on the large scale, and iodine for the
same purpose on the small scale (p. 599). Chloride of
lime or other hypochlorite may be used for sterilising water
on the large scale (p. 600).

Other inorganic substances. — Solutions of salts of mercury
exercise a powerful disinfectant action in proportion to
the amount of dissolved metal which they contain. The
most commonly used is the perchloride (corrosive sub-
limate). Apart from its extremely poisonous character,
it has the disadvantage of forming with albuminoid sub-
stances both insoluble and soluble compounds of little or
no germicidal value, sulphuretted hydrogen converts it
into the insoluble and inert sulphide, and it acts on some
metals. The addition of acids or salts (e.g. hydrochloric
or tartaric acid or sodium or ammonium chloride) prevents
or largely reduces the formation of insoluble compounds ;
but it does not prevent the reactions resulting in soluble
substances, it may reduce the germicidal power, and the
action of perchloride in the presence of albuminoids is

636 A MANUAL OF BACTERIOLOGY

therefore very variable. The reduction in germicidal
power by addition of sodium chloride is well seen from the
following results (Finlay, loc. cit.) :

Number of colonies

16 litres of solution contained developing after treat-

ment for 6 minutes.

1 mole HgCl2 8

1 „ HgCl2 + 1 mole NaCl . . 32

1 „ HgCl2 + 2 moles NaCl . . 124

1 „ HgCl2 + 4 „ Nad . . 382

1 „ HgCl2 + 10 „ NaCl . . 1087

Extremely high values were at one time given for the
germicidal efficiency of corrosive sublimate. This is now
known to have been due to its powerful inhibitory action,
traces of the substance carried over into the subcultures
preventing growth (see p. 643).

The Local Government Board recommended the fol-
lowing solution of corrosive sublimate for disinfecting
purposes :

Corrosive sublimate J oz.

Hydrochloric acid . . . 1 oz. fl.

Anilin blue . . . . . 5 gr.

Water 3 gals.

This forms a solution of 1-900 nearly ; it would be pre-
ferable to use 1 oz. of corrosive sublimate.

The biniodide is also a powerful disinfectant when
dissolved in potassium iodide. It is not affected by
albuminoids nearly as much as is per chloride, and may
be incorporated with soap.

Soluble silver salts are powerful disinfectants, weaker
than mercuric chloride, but far less sensitive to albumi-
noids ; in blood-serum, for instance, silver nitrate is
several times as powerful as corrosive sublimate. They
are incompatible with chlorides, except in certain organic

FORMALDEHYDE 637

combinations, from which silver chloride is only partially
precipitated. Silver salts are poisonous, though less so
than those of mercury.

Iron and zinc salts have been credited with useful
disinfectant action ; but, in fact, their value is very small,
and no practical account need be taken of them. A very
strong antiseptic power has been attributed to copper
salts, which, according to some experiments, exercise a
sufficient disinfectant action on sporeless organisms, such
as the B. typhosus, to enable drinking water to be sterilised
from such infections by the small quantity of copper which
it dissolves (p. 599).

There is some ground for connecting the disinfectant
action of metallic salts with a reducing action on some
forms of protoplasm, as pointed out by Loew.

The permanganates have considerable germicidal power
when in strongly acid or alkaline solution, but the readiness
with which they are affected by organic substances makes
them unsuitable for practical use. Peroxides and ozone
are open to the same objection, and have less disinfectant
power. Hydrogen peroxide is used in the Budde process
for sterilising milk (p. 615), and ozone has been practically
applied in the sterilisation of water-supplies (p. 600).

Organic substances. — The methane and the aromatic
series furnish the disinfectants which are most important
in practice.

Alcohol itself possesses some disinfectant power for
sporeless organisms, but only when absolute or in very
strong solution.

Formaldehyde is by far the most important of the
methane group. It can be applied either as a solution
(formalin) or as gas. The gas can be produced by the
incomplete combustion or oxidation of methyl alcohol,
by the evaporation, with or without pressure, or spraying
of formalin, either alone or mixed with calcium chloride

638 A MANUAL OF BACTERIOLOGY

or glycerine, by the depolymerisation by heat of the solid
polymer paraformaldehyde, or by mixing this substance
with potassium permanganate. Many forms of apparatus
have been designed for the production of formaldehyde gas
for disinfection. In any form the gas seems to give little
more than superficial disinfection, and to require precau-
tions to ensure diffusion throughout the atmosphere of a
room. The conditions desirable for disinfection by for-
maldehyde gas are saturation of the air with moisture,
maintenance of a good room temperature, sealing of the
room, the use of at least 60 grm. of formaldehyde per
1000 cubic feet (preferably more, up to 120 grm.), and in
the case of large rooms mixture of the gas with the air of
the room, either mechanically or by the provision of a
multiplicity of inlets for the gas into the atmosphere. By
the use of a vacuum formaldehyde can be evaporated in
a closed chamber at temperatures indifferent to many
substances which will not stand steam at 100°, and con-
siderable penetration can be obtained (Defries process).
As a spray formalin can be used in any ordinary apparatus.
Formalin seems to have a very slow germicidal action, for
tested by the Rideal- Walker method, its carbolic co-
efficient is only about 0-7 for the B. typhosus. Yet 2
per cent, formalin kills anthrax spores in two or three
days and gaseous formaldehyde is similarly active.

Of the aromatic series, the number of substances and
preparations is extraordinarily large. The standardisation
of methods of examination will, it is to be hoped, eliminate
the less efficient.

The best known is phenol (carbolic acid). Its saturated
solution contains about 9 per cent. It is only slightly
affected by albuminoids, and generally is stable in the
presence of organic matter at ordinary temperatures. Its
compounds, when it forms any, have themselves some
disinfectant action. With acids this action is usually

PHENOL 639

greater than that of pure phenol, with alkalies less. Light
tends to decompose it, but the efficiency is not affected.
It is poisonous and caustic. For practical uses its chief
value is as a standard, as its disinfectant value is com-
paratively low, and for spore-bearing organisms it is
practically useless. Like the cresols, its efficiency is
greatly increased by the addition up to saturation of
common salt or hydrochloric acid. The following results
well demonstrate the increased germicidal power of phenol
by additions of sodium chloride (Findlay, loc. cit.) :

Anthrax spores treated.
Number of colonies develop-

Solution ing after treatment (days).

0137

3 per cent, phenol .... 6300 1390 1260 950
3 + 1 per cent. NaCl . 5720 1450 1320 360

3 + 8 per cent, NaCl . 1940 150 50 0

Probably the addition of salt alters the distribution of the
phenol between the water and the cells, the salt increasing
the concentration of the phenol in the bacterial cells.

" Crude carbolic acid " consists mainly of cresols and
higher phenols in proportions largely dependent on the
source of the tar from which they are prepared ; phenol
is nearly absent from it By themselves the cresols are
extremely insoluble in water ; in oil or alcohol they have
little or no disinfectant value. Cresols are much reduced in
efficiency by albuminoids. In saturated salt solution the dis-
infectant value of crude carbolic acid is greatly increased.

Ordinarily neutral tar oils with no appreciable disin-
fectant value are left in, or mixed with, tar distillate, and
the saponified product produces an emulsion with water.
Innumerable products of this type are made. Their
efficiency varies not only with their active ingredients, but
also with the character of the emulsions which they form,
from about the same as that of phenol to about three times

640 A MANUAL OF BACTERIOLOGY

as much. Commercially they are known as soluble carbolic
acid, soluble creosote, etc. Creolin is a type of numerous
preparations of the same character. They are all poisonous
and sensitive to albuminoids. If naphthalene is present
in excess it is deposited in cold weather on standing. Lysol
is mainly a solution of the cresols in fat or linseed oil,
saponified, with addition of alcohol. It gives a clear
solution with water, having slightly less efficiency on naked
bacteria than cresol, much superior solvency for grease,
and equal sensitiveness to albuminoids. A number of
proprietary disinfectants of high germicidal power are
now to be obtained. Such are cyllin, McDougalPs M.O.H.
fluid, izal, kerol, etc. The active agents appear to be
oxidised hydrocarbons without phenol and cresol, in
emulsion in glue, soaps, oils, etc., and they are compara-
tively non-toxic. The active principle of cyllin is an
oxidised hydrocarbon, having a di-phenyl nucleus in place
of the single phenyl present in carbolic acid ; it is insoluble
in water, hence for the purpose of even distribution in
water it is emulsified with a neutral hydrocarbon oil.
The finished product contains 50 per cent, of the active
principle, and is free from carbolic acid and its homologues.
The active principle of kerol consists of oxidised hydro-
carbons with a di-phenyl nucleus and contains no phenol
or cresol. The germicidal efficiency, expressed as the
carbolic- acid co- efficient (p. 645), of a number of substances
is given in the Table on page 641.

Some of the anilin dyes, especially purified methyl violet
or pyoctanin, have been claimed to be powerfully antiseptic
in solutions of 1-500 to 1-1000.

Chloroform is a powerful antiseptic, but at least 1 per cent,
must be present to act as a germicide ; it is costly, and
not much used as a practical disinfectant, but in bacterio-
logical and physiological chemistry is a useful antiseptic
for preserving solutions which putrefy easily.

CARBOLIC ACID CO-EFFICIENTS

641

lodoform is valuable for dusting wounds, though its
penetrating odour is objectionable, and has led to the
introduction of many substitutes. Its value as an anti-

Carbolic Acid Co-efficients obtained by the Rideal- Walker
Method1 (p. 644)

Disinfectant.

Observer.

Date
of
experi-
ment.

Organism.

Carbolic acid
co-efficient
(carbolic acid
= 1).

Absolute alcohol

Fowler

8-05

B. typhosus

0-03

Boric acid

Walker

10-04

M

0 (?)

Chinosol

Fowler

11-03

M

0-15

Chloros .

j>

1-04

,,

21-0

„ (with 50 per

cent, urine) .

Walker

7-06

,,

8-0

Copper sulphate

ff

6-04

M

0-04

Cyllin* .

Fowler

11-06

,,

14-0

,, (with 50 per

cent, urine) .

n

5-06

M

11-0

Cyllin .

Klein

5-05

M. pyogenes

9-3

;>

Simpson and

6-06

B. pestis

34-0

Hewlett

Formalin

Fowler

3-05

B. typhosus

0-7

Hydrochloric acid .

Walker

2-05

n

11-0

Izal*

Fowler

3-06

tf

11-0

Kerol* .

M

9-06

[I

12-0

„ (with 50 per

cent, urine) .

5>

8-06

M

8-5

Little's phenyle

M

5-04

2-0

Lysol

tj

2-06

j}

2-5

Mercuric chloride

>»

8-05

}t

1000-0

»

Walker

8-05

400-0

Potass, permanganate

Fowler

8-05

?>

42-0

(with 3

per cent, organic

matter)

Walker

1-07

M

1-0

Zinc chloride .

»

1-06

»>

0-15

* The germicidal efficiency of these substances has been increased
since the date of the experiments recorded, and they now have a
carbolic-acid co-efficient of from 16 to 20-22.

1 Fowler, Journ. Roy. Army Med. Corps, July 1907.

41

642 A MANUAL OF BACTERIOLOGY

septic has been greatly discussed ; micro-organisms will
develop in nutrient media containing a considerable
proportion, but probably when in contact with living
cells a decomposition is effected, free iodine being liberated,
hence its value.

The essential oils, peppermint, mustard, doves, thymol,
and menthol, are powerfully antiseptic.

Disinfectant powders at best exert but a superficial
action. They act chiefly as deodorants, but may be useful
in preventing the breeding of flies in garbage, etc.

It is useless to add a small quantity of disinfectant to a large
volume of fluid or solid ; the disinfectant must be added in sufficient
amount so that the mixture contains the minimum percentage
which has been found by experiment to be efficient. For this
reason the attempt to disinfect sewers, sewage, streets, etc., by
relatively small quantities of disinfectants is useless, and the money
so wasted would be far better employed in providing more water
for flushing purposes.

In medical practice, while antiseptics can be applied locally with
success and, to some extent, for disinfecting the alimentary tract,1
no substance has yet been discovered which can be administered
with safety to such a degree as to saturate the body, and so exert
a general germicidal action in bacterial infective diseases. Sal-
varsan, perhaps, to some extent possesses this power and has
been used with success in certain general infections, e.g. anthrax.
Protozoa are attacked selectively by many substances, e.g. the
malaria parasite by quinine, spirochaetes by salvarsan, trypano-
somes by atoxyl, trypan red, etc., Piroplasma canis by methylene-
blue, etc.

In surgical practice no unbiased observer can doubt the efficacy
of antiseptic treatment, but many so-called " antiseptic operations "
are marred by faults of omission and commission which render
them far from being perfectly antiseptic. There has been some
controversy between the advocates of " antise:ptic " and of " aseptic"
surgery. Undoubtedly antiseptics do diminish the vitality and
therefore the reparative power of the tissues and aseptic methods
should so far as possible replace antiseptic ones. The skin of the

1 See F. E. Taylor, "Intestinal Disinfection in Alimentary Toxaemia,"
Medical Prets, January 14, 1914.

DETERMINATION OF GERMICIDAL POWER 643

patient and the hands of the operator having been disinfected as
far as possible, no antiseptic should be permitted to come into
contact with the wound, which may be irrigated with warm sterile
physiological salt solution. A dry wound is an important element to
success, and a dry, sterile, unirritating dressing should be employed.
Instruments, sponges, etc., may be kept in sterile salt solution after
the preliminary disinfection — by heat (not sponges) or chemicals.
But the aseptic system requires more care to ensure success than
the antiseptic one, and unless the assistants can be Crusted and
the details rigorously carried out, the latter seems preferable.

The Determination of the Germicidal Power

For determining germicidal power on sporing organisms anthrax
spores are generally used, on non-sporing organisms cultures of
the B. typhosus are usually employed.

(1) Thread method. — Sterilised silk threads are impregnated with
sporing and non-sporing organisms, lightly dried, and then exposed
to the action of the antiseptic solution of a known strength for a
given time. After treatment the threads are thoroughly washed
with distilled water to remove the antiseptic, and sown on the
surface of agar or other suitable culture medium. If no growth
occurs the organisms are assumed to have been destroyed. As a
matter of fact, however, it is extremely difficult to get rid of the
last traces of the antiseptic, which may inhibit growth although the
organisms may yet be alive, a fallacy which caused an exaggerated
value to be assigned to many substances — for example, corrosive
sublimate. The thread method may still be employed, but after
treatment the threads should be sown in broth, or, better still, if
pathogenic organisms be the subject of experiment, inoculated into
a susceptible animal. The writer finds that in disinfection experi-
ments with anthrax spores, surface agar is a much better medium
than broth.

In experiments with corrosive sublimate, by whatever method, the
last traces of this substance must be converted into the inert sulphide
by treatment with hydrogen or ammonium sulphide.

(2) Garnet method. — Small garnets the size of a pea are sterilised,
soaked in a suspension or a broth culture of the organism, removed
and dried. The garnets with the organisms attached are then
soaked in solutions of the disinfectant of known strengths for
various periods of time ; they are then removed from the solution
well washed with sterile water, and finally placed in tubes of broth.

644 A MANUAL OF BACTERIOLOGY

(3) Rideal-Walker or drop-method. — Moor first suggested that the
germicidal efficiency of a disinfectant might be compared with that
of a standard solution of carbolic acid, which has a definite com-
position, is stable, and can be accurately standardised, and Rideal
and Walker devised an ingenious and simple method for carrying
this out. A special test-tube rack is very convenient (Fig. 69), in
.which the lower tier has five holes which hold three or four tubes
containing the solutions of decreasing strengths of the disinfectant
tojbe tested, and two tubes or one tube containing standard carbolic

FIG. 69. — Test-tube rack with test-tubes arranged for the
Rideal-Walker method of testing disinfectants.

acid solution of known strength for comparison. The upper tier
has thirty holes in two rows spaced into six sets of five holes
each. These hold tubes of sterile nutrient broth which are num-
bered from 1 to 30. The test is usually made with a broth culture
of B. typhosus, but other organisms may be employed. The process
is as follows : The five tubes in the lower tier each contain 3 c.c.
of the disinfectant and carbolic solutions. Into each in succession,
at intervals of half a minute, three drops of the typhoid broth
culture are added with a pipette. Half a minute after the last tube
has been inseminated, a loopful is taken from the first tube and
inseminated into the first broth tube, and this process is repeated
at half-minute intervals until all the broth tubes have been inocu-
lated. The inoculated broth tubes are then incubated at 37° C.
for three days, and the occurrence or not of growth is taken as
indicating the killing or non-killing of the organism respectively.
Obviously the first set of five broth tubes inoculated are subcultures

RIDEAL-WALKER METHOD

645

in which the organism has been acted upon by the disinfectant and
carbolic solutions for two and a half minutes, the second set for
five minutes, and so on. The results (taken from an actual test)
may be charted as follows :

B. typhosus, 24:-hour broth culture at 37° C.
Room -temperature 60° F.

Disinfectant

Dilution.

Time culture exposed to action
of disinfectant (in minutes).

Sub-cultures.

Period of
incubation.

Tempera-
ture.

2J

5

7*

10

12i

15

X
X

1-1400
1-1500

+
+

3 days

37° C.

+

*

*

*

*

X

1-1600

+

+

+

*

*

*

X

1-1700

+

+

+

+

*

*

Carbolic

1-100

+

+

+

*

*

*

= growth in the sub-cultures.

= no growth in the sub-cultures.

From this it will be seen that the disinfectant X in a solution of
1 in 1600 kills in the same time (7| minutes) as carbolic 1 in 100.
This result is expressed as a coefficient obtained by dividing the
strength of disinfectant by the strength of carbolic which kills each
in the same time ; in the present instance the co-efficient is YQ00° =
16-0, and this figure is known as the " carbolic acid coefficient."

If nothing is known about the strength of the disinfectant, some
preliminary experiments should be performed with dilutions at
wide intervals as regards strength (e.g. 1-100, 1-500, 1-1000,
1-1500, 1-2000, etc.), and when the limit has thus been approxi-
mately ascertained, the test is performed as above.

Precautions to be taken in carrying out the test. — (1) The culture
should be a broth one about twenty to twenty-four hours old, and
should be free from clumps ; this may be attained by filtration
through paper. Instead of adding drops of the culture to the
solutions, the addition of 0-1 c.c. of culture for every cubic centi-
metre of solution has recently been suggested. The writer regards
this amount as being too large, and would suggest that O'l c.c. of
culture is sufficient.

(2) The carbolic acid (the crystals of which should have a melting-
point of not less than 40-5° C.) should be kept in the form of a
5 per cent, aqueous solution standardised by the bromine method.

646 A MANUAL OF BACTERIOLOGY

Failing this, the solutions may be made with the acidum carbolicum
liquefactum of the Pharmacopoeia, which contains 100 parts of
phenol in 110, but is not absolutely constant in composition.

(3) All measu es, pipettes, and test-tubes used for making dilutions
should be sterile.

(4) The dilutions of the disinfectant and carbolic should be made
with sterile distilled water.

(5) The broth used for culturing and sub-culturing should have
the following composition :

Lemco . . . . . .20 grin.

Peptone . . . . .20 grm.

Salt 10 grm.

Water 1000 c.c.

The medium should be standardised to a reaction of + 10 (Eyre's
scale).

(6) The loop usecj for sulculturing should have an internal
diameter of 3 mm., and be made with platinum wire of 27-28
B.W.G.

(7) Growths in the subcultures should be obtained in those taken
at not less than two and preferably at three of the time intervals
(2£, 5, and 7£ minutes) from both the disinfectant and the carbolic
solutions which correspond.

(8) The temperature at which the determination is made should
be noted, and the strength of carbolic varied accordingly (1-100
for 56°-62° F., 1-110 for 62°-67° F., and 1-120 for 67°-73° F.
for B. typhosus), or the determination may be made at a standard
temperature (e.g. 20° C.) by warming (or cooling) the disinfectant
and carbolic tubes in a water-bath.

(9) When the organism does not form a uniform culture in broth,
a suspension of an agar or other culture must be made in water
and filtered. Sub-culturing in some cases (e.g. with B. pestis and
B. anthracis) must be made on agar or other suitable culture medium.

The method is an admirable one for determining the relative
efficiencies of disinfectants on naked organisms in the absence of
organic matter. But in practice disinfection is almost always
carried out in the presence of organic matter, and various suggestions
have been made with a view of introducing this factor into the
test, for the presence of organic matter may reduce the carbolic-
acid coefficient of many disinfectants (see pp. 632-642, and Table,
p. 641). Among the substances suggested are urine, faeces, 2
per cent, suspension of dried and sterilised faeces (Martin and Chick),
and milk. Kenwood and Hewlett found that the presence of urine

RIDEAL-WALKER METHOD 647

or faeces reduced the carbolic acid coefficient of some proprietary
disinfectants to a greater relative extent than that of carbolic.

The method is also sometimes somewhat erratic in practice, and
a number of determinations may be needed before the strengths of
disinfectant and carbolic which coincide are found. Occasionally
also two strains of B. typhosus may differ widely as regards the
germicidal action of the disinfectant on them, while they are prac-
tically identical as regards the germicidal action of the carbolic.

Woodhead and Ponder have proposed a modification of the
method. In this, B. coli is used as the test-organism and bile-salt
peptone water as the culture medium, a platinum spoon being used
for culturing, and more cultures at shorter intervals up to half
an hour are made.

4. Volatile disinfectants may be tested by moistening the wool
plug of an agar tube, inoculating the agar, and capping with a
rubber cap, and observing whether any growth occurs.

5. Volatile disinfectants may also be tested by exposing silk
threads, pieces of paper or fabrics, splinters of wood, etc., impreg-
nated with organisms, some free, others done up in packets of cotton-
wool, in a room or chamber of known cubic capacity, to the action
of the gas, a known amount of which is present in the chamber.
After exposure for a given time, the threads are sown in broth
tubes, and the tubes incubated.

On the Rideal- Walker method, etc., see Rideal and Walker,
Journ. Sanitary Inst., vol. xxiv, 1903, p. 424 ; Kenwood and
Hewlett, ibid. vol. xxvii, 1906, p. 1 ; Firth and Macfadyen, ibid.
p. 17 ; Kenwood, Public Health, 1908 ; Fowler, Journ. Roy. Army
Med. Corps, July 1907 ; Partridge, Bacteriological Examination
of Disinfectants ; Woodhead and Ponder, Lancet, 1909, vol. ii.

FRENCH WEIGHTS AND MEASURES AND THEIR
ENGLISH EQUIVALENTS

\fji (micron)
1 millimetre

25 millimetres
1 centimetre
2-5 centimetres
5 centimetres
1 gramme
4 grammes

28 grammes
1 kilogramme
0-5 kilogramme
1 cubic centimetre
3£ cubic centimetres

28 cubic centimetres
568 cubic centimetres
1 litre

0-001 millimetre (^5Q00 inch, nearly)
0-04 (Jg) inch.
1 inch.
0-39 inch.

1 inch.

2 inches.

15£ (15-432) grains.

1 drachm (apothecaries'), nearly.

1 ounce (avoirdupois), nearly.

2-2 pounds (avoirdupois).

1 pound (avoirdupois), nearly.

16 minims, nearly (16'23 minims).

1 fluid drachm, nearly.

1 fluid ounce, nearly.

1 pint (f litre).

If pints, or 35 fluid ounces, nearly.

SOLUBILITIES

AMOUNT OF SUBSTANCE CONTAINED IN 10 c.c. OF A
SATURATED SOLUTION

Alcoholic solution of methylene-blue .
Aqueous solution of methylene-blue
Alcoholic solution of gentian violet
Aqueous solution of gentian violet
Alcoholic solution of fuchsin
Aqueous solution of fuchsin
Aqueous solution of corrosive sublimate

0-068 grm.
0-646 grm.
0-442 grm.
0-175 grm.
0-292 grm.
0-066 grm.
0-507 grm.

648

JHD,HODGEND, O.S

INDEX

ABERRATION, 139
Abiogenesis, 4
Abscesses, amoebic, 484

— multiple, 226

— typhoidal, 355

Absorption of complement, 178,

183

Achalme's bacillus, 427, 565
Achorion Schoenleinii, 479
Acid Alcohol in Gram's method,

104

Acid-fast organisms, 299
- in milk, etc., 340
Acne, 219, 228, 229
Actinomyces, cultivation, 453

— varieties, 456
Actinomycosis, 451

— clinical examination, 456

— human, 452

— in cattle, 451

— spread of, 454

— staining of, 452
Adsorption, 167
Aerobic organisms, 21
Agar, 58. See Culture Media
Agglutination, 185,190
Aggressins, 179

Air, bacteriology of, 603

— examination of, 605

— of sewers, 610

Air- passages, organisms of, 570
Air-pump, 48
Alcohol, absolute, 86

— formation of, 35, 385, 471

— for fixing, 85

— and ether for fixing, 97

— as an antiseptic, 365, 637

— methylated, 86

Alessi's experiments, 365
Alexins, 174, 181, 200, 204
Algae, 8, 9

— destruction of, 600

— in water, 602

Alum for purifying water, 574

— method for typhoid, 594
Amboceptor, 174, 175
Amoeba buccalis, 482
Amoeba coli, 482
Amoebae, intestinal, 482
Amoebic dysentery, 482

— diagnosis of, 485
Ammonia not pyogenic, 225

- production of, 30, 125
Anaerobic cultures, 71

- stab, 71

— in nitrogen, 72

— Buchner's tubes, 72

— in vacuo, 72

— in hydrogen, 73

— in formate broth, 76

— in sulphindigotate broth, 76

- Dean's method, 76

- Frankel's method, 74

— Hamilton's method, 72

— writer's method, 75

- plate, 82

Anaerobic organisms, 21, 419
Analysis of yeasts, 466
Anaphylaxis, 168
Angina, Vincent's 269
Anilin dyes as disinfectants, 640

— stains, 99

— water, 99

Animals, dissection of, 123

— inoculation of, 122
Anophelinae, 519

649

650

INDEX

Anthrax, 251

- bacillus of, 252, 200

— diagnosis of, 262

— occurrence of, 258-261

— serum for, 261

- spread of, 258

— symptomatic, 455

— vaccine, 262
Anti- bodies, 149
Anti-endotoxic sera, 42,177
Anti-ferments, 194
Antigen, 150

— test (syphilis), 501
Antiseptic action, conditions

modifying, 630

— power, determination of, 643

— treatment, 642
Antiseptics, 627-643
Anti-sera, 173
Anti-serum, anthrax, 261

— cholera, 444

— colon, 387

— dysentery, 378

— gonococcic, 246

— hydrophobia, 541

— meningococcic, 244

- plague, 400

— pneumonia, 411

— polyvalent, 175

— streptococcus, 237

— tubercle, 322

- typhoid,366
Antitoxic constituent, 167

— treatment, 160
Antitoxin, cholera, 444

— diphtheria, 278

— tetanus, 424
Antitoxins, 150

— in normal blood, 153, 274
Anti-venin, 164
Appendicitis, 556
Archebiosis, 5

Area of dish, 605
Arthritis, 246, 410, 565, 566

— deformans, 566
Arthus phenomenon, 171
Ascitic fluid culture medium, 61
Ascococcus, 17
Ascomycetes, 470

Ascospores of peniciUium, 472

— of yeast, 461, 465

— of yeast, staining, 468
Aseptic treatment, 642
Asiatic cholera, 433
Aspergillus glaucus, 472

— fumigatus, 473

— niger, 472
Atrepsy, 206
Autoclave, 47
Azotobacter, 33

BABESIA, 528

Bacilli, capsulated, 258

Bacilli carriers, cholera, 438

- diphtheria, 273, 286

— dysentery, 378

- typhoid, 359
Bacillus, definition of, 17

- acidi lactici, 381, 613

— acidophilus, 571

— acnes, 560

— aerogenes capsidatus, 240, 427

— aertryck, 373, 374

— albus variola, 550

— alcaligenes, 6, 368, 598

— anthracis, 252

— anthracis similis, 256

— anthracoides, 256

— aquatilis sulcatus, 598

— bifidus, 571

— bottle, 479

- botulinus, 427

— bronchisepticus, 559

— buccalis, 460

— bulgaricus, 617

— butyricus, 35, 432

— cadaveris sporogenes, 431

— caniculce, 559

— capsulatus, 381
hominis, 258

— cavicida, 388

— chauvcei, 429, 431
Bacillus cloacce, 389, 584, 610

- coli, 379

communis, 379

communior, 384

immobilis, 258

INDEX

651

Bacillus coryzce, 297

— diphtheria, 267
columbarum, 298

- diphtheroid, 273, 287, 298,

563, 565, 619

— dysenteria, 352, 376

- enteritidis, 351, 371, 381, 390,

613, 620
— sporogenes, 427, 431

— facalis alkaligenes, 6, 368, 598

— fllamentosus, 610, 621

— fetidus, 569

— fluorescens liquefaciens, 30, 37,

203, 239, 363, 581, 610, 621

— fluorescens non - liquefaciens,

581, 621

— fluorescens stercoralis, 363

— fusiformis, 296, 19

— glanders, 343

— grass, 340

— icteroides, 373, 546

— infantilis, 571

— influenza, 415

— lactis aerogenes, 389, 570, 584,

613

— leprce, 333

— mallei, 343

— megaterium, 621

— mesentericus, 30, 610, 621

— mist, 340

— mucosus capsulatus, 258

— murisepticus, 405

- mycoides, 30, 608, 610, 621
• — neapolitanus, 388

- of Achalme, 427, 565
— • of black quarter, 431

— of chicken cholera, 404

— of Danysz, 373

— of Ducrey, 557

— of Friedlander, 4075 412

— of gastro-enteritis, 371

— of Hofmann, 287

— of hog cholera, 373

— of Johne, 331

— of Koch and Weeks, 557
• — of Laser, 388

- of Lustgarten, 339, 496

— of malignant oedema, 426

— of Morax and Axenfeld, 557

Bacillus of mouse septicaemia, 405

— of ozsena, 563

— of rabbit septicaemia, 404

— of rheumatoid arthritis, 566

— of rhinoscleroma, 566

— of Massol, 617

— of swine fever, 373

— of swine plague, 373, 405

- of symptomatic anthrax, 431

- of syphilis, 496

- of xerosis, 297

- Oppler-Boas, 562

— paracoli, 374

- paradysenterioe, 379

- paratyphosus, 351, 371, 374

— perfringens, 427

— pertussis, 417

— pestis, 392

- pneumonia, 258, 407, 412

- prodigiosus, 36, 250, 621, 622

- proteus, 24, 30, 240, 558, 608,

610, 621

— pseudo-anthracis, 256

— pseudo-diphtheria, 287

— pseudo-dysenteria, 376

— pseudo-tuberculosis, 332, 395

— psittacosis, 371, 373

— putrificus coli, 30, 420, 571

— pyocyaneus, 238, 37, 182, 240,

541, 558, 559, 561, 627

— pyogenes fetidus, 387

— segmentosus, 297

— smegmatis, 338, 329

— subtilis, 15, 491, 610, 621

— suicholera, 373

- suipestifer, 351, 373, 374

— sulcatus, 598

— tetani, 420

— timothy grass, 340

— tuberculosis, 301

— typhosus, 353

— typhimurium, 371, 373

— vagina, 571

— violaceus, 37, 621

— Welchii, 427, 240, 431, 572,

581, 585, 588, 608, 609, 613,
617, 618

— X., 546

— xerosis, 297

652

INDEX

Bacteria, action on artificial
sugars, 22

— classification of, 15

— conditions of life of, 19

— effect of electricity on, 24

- effect of light on, 23

— effect of pressure on, 23

- influence of chemical agents,

on, 21

— influence of oxygen on, 20

— influence of radium on, 24

— influence of temperature on,

20

— nutrition of, 19

— selective action of, 22

— structure of, 9-14

- study of, 66, 118 et seq.

- thermophilic, 20, 608

— variation of, 6, 16

— vitality of buried, 609
Bacterial poisons, 146

— products, 38
Bacteriological diagnoses. See

EXAMINATIONS

— microscope, 132
Bacteriolysis, 174
Bacteriotropines, 210
Bacterium, definition of, 17

— species of, See Bacillus

— termo, 30, 621

— tumefaciens, 555
Bacteroids, 32
Balantidium coli, 507, 560
Basidia, 470
Basidiomycetes, 470

Bee disease, 532
Beer, 466
Bell- jars, 49
Beri-Beri, 556
Berkefeld filter, 49, 601
Bird-pox, 206, 553
Bismarck brown, 101, 294
Black leg, 431
Black quarter, 431
Blackwater fever, 523
Blastomycetes, 462

— examination, 464
Blastomycetic dermatitis, 463
Bleeding animals, 125

Blood films, 96, 523
Blood, centrifuging,126

— germicidal action of, 199
— serum, 60

- to obtain, 60, 125
Blood-agar, 62, 446

- parasites, staining, 524
Blue pus, 238

Boils, 228

Borax- methylene blue, 526
Bordet- Durham reaction, 188, 192
Bordet-Gengou phenomenon, 183
Boric acid, 641
Bottle baciUus, 479
Botulismus, 427
Bread, 620

Brilliant-green agar, 597
Bromine, 634
Bronchitis, 557, 248, 413
Broncho - pneumonia, 406, 373,

413, 416, 417

Broth, 55. See CULTURE MEDIA
Brownian movememt, 11, 130
Bubonic plague, 391. See Plague
Buchner's method, 72

— tube, 72
Budde process, 615
Butter, 622

— acid-fast bacilli in, 340

CAFFEINE mixture, 595
Cahen's test, 435
Canary fever, 549
Cancer, 554, 232, 462, 486, 511
Cancrum oris, 562
Caps, india-rubber, 52
Capsulated bacilli, 258
Capsule of bacteria, 10

— staining, 112
Carbol-fuchsin, 100

— gelatin, 590

— methylene blue, 99

— thionin blue, 100
Carbolic acid, 638

— crude, 639

— coefficient, 645
Carbuncle, 223
Carmine picro-, 101
Carriers, bacilli, 359

INDEX

653

Cellulitis, 235
Centrifuge, 48

Cerebro- spinal meningitis, 241
Chancre, soft, 557
Cheese, diphtheroid bacillus in,
619

— tubercle bacillus in, 622

— spirillum, 449
Chemotaxis, 202
Chicken cholera, 404
China-green agar, 597
Chitral fever, 549
Chlamydospores, 471
Chlamydozoa, 537, 567
Chloride of lime, 634
Chlorine, 634
Chloroform, 640
Chloros, 634

Cholera anti-serum, 444

— Asiatic, 433

— chicken, 404

— hog, 372

— infantum, 558

— red reaction, 27, 435

— spirillum, 433

— diagnosis of, 446
indole reaction, 435

- in butter, 622

- in milk, 437, 613, 619

— in oysters, 437

- in soil, 437

- in water, 437, 598

— isolation from water, 598

- pathogenesis, 437

- phosphorescence, 440

— toxins, 442

— vaccine, 444
Ciliata, 507
Cirrhosis, hepatic, 387
Cladothrix dichotoma, 460
Classification of bacteria, 15
Clearing, 108

Clinical diagonses. See EXAMINA-
TIONS
Clostridium butyricum, 35, 432

— Chauvcei, 429, 431
Clothing, etc., 622

Clove oil as an antiseptic, 642

— as a clearing agent, 108

Coccidial disease in man, 511
Coccidium oviforme, 509
Cold, effect on bacteria, 20
Coley's fluid, 250
Colitis, 379, 560
Collodion sacs, 121
Colon bacillus, 379
— differentiation from typhoid,
384

- isolation of, 380

- isolation from water, 590, et

seq.

- pathogenicity of, 386

- varieties of, 384, 388
Comma bacillus of cholera, 433
Complement, 174, 175

- deviation, 178, 183

- fixation, 183
Complementoid, 175
Condenser, sub-stage, 139
Conidia, 469

Conjugation in Hyphomycetes,

469

Conjunctive, organisms of, 569
Conjunctivitis, 557
Conradi-Drigalski agar, 592
Contagion, 144
Copper, germicidal action of, 599

— sulphate, germicidal action of,
599

Correction collar, 141
Corrosive sublimate as disinfec-
tant, 635, 629

— action on rubber, 52
— for fixing, 87

Cover-glass specimens, 94

— of blood, 97, 523

— staining, 106
Cream, 613
Creolin, 640
Cresol, 639
Crithidia, 487
Croup, 265
Cryptococcus, 474
Culicidse, 518
Cultures, anaerobic, 71

— hanging-drop, 129

— Indian ink, 81

— plate, 76, 82

654

INDEX

Cultures, preserving, 116

— roll, 82

— shake, 83

— single ceU, 81, 465

— vitality of, 121

CULTURE MEDIA

Agar-agar, 58

- blood, 62, 417, 446, 489
alkali, 446

— brain, 303

— brilliant green, 597

— china-green, 597

— Conradi-Drigalski, 592

— distilled water, 618

— fuchsin, 596

— glucose, 59

— glycerin, 59

— haemoglobin, 63

— litmus, 59

— malachite green, 596

— maltose, 477

— mannite, 33

— nasgar, 242

- potato blood, 441

- rebipelagar, 592

— serum, 62

— wood-ashes, 32
Alkali albumin, 63
Ascitic fluid, 61, 291
Beer- wort, 57

Bile (for typhoid), 370
Bile-salt, 590, 591
Blood-serum, 60

— fluid, 61

— Loffler's, 61
Broth, acid beef, 54, 64,

— ascitic fluid, 61, 291

— egg, 56

— formate, 76, 591

— glucose, 56

— glycerine beef, 56

— Lemco, 56, 646

— peptone beef, 55

— sulphindigotate, 76

— veal, 56
Dieudonne's, 446
Dorset's egg medium, 303
Eggs, 63

Endo's, 596

CULTURE MEDIA (cont.) —
Gelatin, 57

— beer-wort, 58

— carbol, 590

— glucose, 58
Hiss's, 291
Hydrocele fluid, 61
Litmus, 59
Malachite green, 596
Milk, 59

Neutral red, 591
Nitric and nitrous, 31
Pasteur's fluid, 63
Peptone water, 57

— Dunham's, 57
Petruschky's, 385
Potato, 60

— glycerin, 303
Proskauer-Capaldi, 385
Standard, 64
Uschinsky's fluid, 63
Whey, litmus, 385

Cutaneous reaction, tuberculosis,

330
- syphilis, 499

— typhoid, 370
Cultures, roll, 82

— single- cell, 81, 465
Cystitis, 250, 355, 387
Cytases, 203
Cytoryctes variolce, 551
Cytotoxins, 185

DANYSZ bacillus, 373

— effect, 165

— rat vims, 373
Dark-ground illumination, 139
Deneke's spirillum, 449
Dengue, 549

Deodorants, 628

Dermatitis, blastomycetic, 463

— bullous, 239, 561
Desiccation as a disinfector, 627

— influence of, 20, 627
Deviation of complement, 178

— test, 183
Dhobie itch, 479

Diagnosis, bacteriological or
clinical. See EXAMINATIONS

INDEX

655

Diarrhoea of infants, 558, 373,

379, 613

Dilution method, 77
Diphtheria, 265

- etiology of, 266
Diphtheria, associated organisms,

271

— antitoxin, 278

standardisation of, 280

unit of, 284

value of, 285

— diagnosis of, 269, 292

— value of, 270

- and milk, 594

— bacillus, 267

• acid formation, 269

— fermentation reactions, 292

— in noma, 562

— in ozsena, 563

— in pyorrhoea, 565
isolation of, 266

- pathogenic action, 272
persistence of, 270

- pseudo, 287

- thread forms, 267, 295
• toxins, 276

- varieties, 267, 272

— membrane, 273

— in lower animals, 275

— 'of calves, 298

— of pigeons, 298
Diphtheritic roup, 298

— neuritis, 273

- paralysis, 273, 275, 286
Diphtheroid bacilli, 273, 287, 292,

297, 563, 565, 619
Diplococcus crassus, 244

— flavus, 244

— intracellidaris meningitidis,

241

— mucosus, 244

— pneumonice, 407, 234

- rheumaticus, 234, 565, 566

- Still's, 244

Disease, production of, 146
Diseases, causative organisms of,
556

— of beer, 466
Disinfectant powders, 642

Disinfectants, 627

Disinfecting solution of Local

Government Board, 636
Disinfection, 623
Disinfectors, 624-626
Dissection of animals, 123
Distemper, 559
Dorset's egg medium, 303
Dourine, 490
Drepanidium, 531
Dunham's solution, 57
Durham's tubes, 83
Dust in the air, 605
Dysentery, 559
— amoebic, 482

— bacterial, 376, 239

— infusorial, 508

- para, 379

— pseudo, 376

— bacillus, 376, 352

ECZEMA, 560
Effluents, sewage, 610
Egg cultures, 63

— media, 56, 303

Ehrlich's side-chain theory, 152

Ehrlich-Biondi stain, 101

Eimeria, 509

El Tor vibrios, 441

Electricity, effect of, on bacteria,

24
Embedding, gum, 88

- paraffin, 90
Empyema, 355, 410
Endo's fuchsin agar, 596
Endocarditis, infective, 228, 231,

235, 246, 410, 565
Endotoxins, 39, 40, 147
Endotoxic sera, 42, 177

— vaccines, 222
Enrichment methods, 595
Entamceba, 482
Enteritidis group, 351, 371
Enteritis, 371

— fowl, 404, 447, 511

— zymotic, 558
Enumeration of organisms, 80,

220
Enzymes, 36

656

INDEX

Eosin, 100

Epizootic, lymphangitis, 348, 474
Eppinger's streptothrix, 450
Erysipelas, 236
Erythrasma, 479
Esmarch's roll cultures, 82
Ether and alcohol for fixing, 97
Euglena, 487
Evaporation, 49

EXAMINATIONS, BACTERIOLOGICAL,
AND CLINICAL DIAGNOSES —

Actinomycosis, 456

Agglutination, 190

Air, 605

Algae in water, 602

Amoeba coli, 485

Anthrax, 262

Blastomycetes, pathogenic,
464

Butter, 622

Cheese, 622

Cholera, 446

— in water, 598
Ciliated forms, 508
Coccidial disease, 512
Colon baciUus, 387
Complement fixation, 183
Diphtheria, 292

— in milk, 618
Disinfectants, 643
Dysentery, 485
Filters, 602
Flagellated forms, 508
Glanders, 349
Gonorrhoea, 247
Haemolysis test, 182
Hydrophobia, 542
Hyphomycetes, 474
Ice and ice creams, 599
Influenza, 417
Leprosy, 337
Malaria, 523
Malignant O3dema, 430
Milk, 618

Moulds, 474
Opsonic index, 214
Pfeiffer's reaction, 177
Phagocytosis, 210
Plague, 403

EXAMINATIONS, BACTERIOLOGI-
CAL, AND CLINICAL DIAG-
NOSES (cont.) —
Pneumonia, 414
Porges' reaction, 507
Protozoa, 508

— in water, 602
Rabies, 542
Relapsing fever, 496
Ringworm, 479
Sarcina ventriculi, 249
Septic diseases, 239
Sewage, 610
Shell-fish, 600

Smegma bacillus, 329, 339
Soil, 609

Suppuration, 239
Syphilis, 499
Tetanus, 425
Thrush, 474

Treponema pallidum, 499
Trichophytons, 479
Trypanosomes, 475
Tuberculosis, 323

— (milk), 618

Typhoid bacillus in water,
593

— fever, 369
Vincent's angina, 296
Wassermann reaction, 501
Water, 576
Watercress, 600
Welch's bacillus, 430
Xerosis, 297

Yeasts, 468

— pathogenic, 464
Exhaustion theory, 206
Eye-pieces, 135, 140

FARCIN DES BCEUFS, 456
Farcy, 341, 474
Favus, 479
Fermentation, 34, 465

— acetic acid, 36

— alcoholic, 35

— bottom, 466

— butyric acid, 35

— lactic acid, 35

— top, 466

INDEX

657

Fermentation tube, 83
Ferments, 34, 36

- anti-, 194
Films, 94

- blood, 96, 523

— interlamellar, 131
Filters, 49, 601
Filtration, 49, 574, 601

— as a disinfector, 627

- sand, 574

'; Finger and toe " disease, 486
Finkler-Prior spirillum, 448
Fixation of complement test, 183
Fixing specimens, 96

— tissues, 86

— by alcohol and ether, 97

— by corrosive sublimate, 87
Flacherie, 532

Flagella, 11

Flagella staining, 114

Flaginac reaction, 384, 584

Flasks, yeast, 75

Fleas, 402

Flies and disease, 364, 389

— preventing access of, 125

- tsetse, 490

Fluid media, growths in, 68
Food poisoning, 38, 362, 371, 620
Foot and mouth disease, 561
Forceps, 51
Formalin for disinfecting, 637

— fixing tissues, 87

- preserving cultures, 116

- preserving specimens, 116
Formate broth, 76, 591
Fowl enteritis, 401, 447, 511
Frambo3sia, 495
Frankel's pneumococcus, 407

— tube, 74

Frankland's method for air ana-
lysis, 606

Freezing microtome, 89
Friedlander's capsule stain, 112

— pneumo-bacillus, 412
Frozen sections, 88, 108
Fuchsin agar, 596

— bodies, 554

- carbol, 100
Fumigation, 634, 637

Fungi, 8, 9, 469, 470
Fungi imperfecti, 461, 470
Fungus disease, 457
Fusiform bacillus, 19, 296, 562

GAMETES, 481, 515
Gangrene, hospital, 223

- spreading, 235, 425, 427
Gartner group, 351, 371
Gartner's bacillus, 371
Gas production, 37

determination of, 83

Gastric juice, prevention of infec-
tion by, 439, 599

Gelatin, 57. See CULTURE MEDIA

— liquefaction of, 68

General paralysis of insane, 274,

497, 507

Genital organs, organisms of, 571
Gentian- violet, anilin, 99
Germicidal action of blood, etc.,

199

Germicides, 627
Giant ceUs, 300, 311, 312, 334,

346, 451

Giemsa stain, 102, 500, 524
Glanders, 341
Glanders-like disease, 349
Globulin, cell, 199
Globulin of anti-bodies, 167
Glossina, 490
Golding's bottle, 79
Gonorrhoea, 244

— diagnosis of, 247

— lesions in, 246
Gram's method, 102, 99

— Claudius's modification, 105

— Giinther's modification, 104

- thionin, 106

— Weigert, 105, 111
Gram-negative cocci, 248
Granules, metachromatic, 10
Granuloma, ulcerating, 495
Grass bacillus, 340

Grease for stoppers, 48
Gregarines, 530
Griffith's steriliser, 599
Grinding machine, 42
Grouse disease, 405, 511

42

658

INDEX

Guarnieri bodies, 551, 552
Gum for freezing, 88

EDEMAMCEBA, 512, 527, 528
Haematoxylin, 101

— iron, 485
Haemoflagellates, 487
Haemoglobin agar, 63
Hsemogregarines, 530
Haemolysins, 181
Haemolysis, 180

— test, 182
Hsemolytic serum, 181, 184

— system, 184
Haemoproteus, 528
Haemosporidia, 512
Halogens, 634
Halteridium, 528, 515
Hanging- drop cultivations, 129

— anaerobic, 131
Haptines, 155
Haptophore group, 153
Heat as a disinfector, 623
Heat, dry, 624

— moist, 625
Heidenhain's iron-haematoxylin,

485

Hermann's tubercle stain, 327
Herpes zoster, 561
Herpetomonas, 487
Hesse's method for air analysis,

605

Hiss's medium, 291
Hodgkin's Disease, 332
Hofmann bacillus, 287
Hog-cholera, 372
Hot-air steriliser, 45
Humanus longus tubercle bacillus,

319

Hydroa gestationis, 561
Hydrocele-fluid culture medium,

61

Hydrochloric acid, 635, 641
Hydrogen peroxide, 637
Hydrophobia, 535
Hypersensitation, 169
Hyphae, 469
Hyphomycetes, 469

Hyphomycetes, examination, 474

— pathogenic, 472

ICE, organisms in, 573, 599

— creams, 599

Identification of organisms, 119
Illumination, 133

— dark ground, 139
Immersion lenses, theory of, 137
Immune body, 174
Immunity, 195

— acquired, 196, 205

— atreptic, 206

— natural, 196

— active, 205

— passive, 305

— humoral, 202

— phagocytic, 202

— transmission of, 206
Impetigo, 228, 273, 560
Impression specimens, 98
Incubator, 66

Index, opsonic, 210, 214

— determination of, 214
Indian ink method, 81

— for syphilis, 499
Indole, 25

— influence of culture medium,
25

Infantile paralysis, 543
Infection, 144

— modes of, 148
Infective process, 147
Influenza, 415

— cold, 248, 297, 417
Infusoria, 507

Inoculating tubes, method of, 69
Inoculation, intra- venous, 123

— of animals, 122
Insects and disease, 389
Interlamellar films, 131
Intestine, organisms of, 570
Intoxication, 145
Intracellular substances, 39, 40,

140

Intra- venous inoculation, 123
Invertase, 36
Investigation of micro-organisms,

118

INDEX

659

Iodine, 635

— Gram's, 103

— trichloride, 635
lodoform, 641
Irrigation, 127

Isolation of micro-organisms, 77,

119
Izal, 640

JENNER'S blood stain, 102, 524
Johne's disease, 331

KALA-AZAR, 491

Kerol, 640

Klebs-Loffler bacillus, 266

Koch's " comma " bacillus, 433

— postulates, 147
Koch- Weeks bacillus, 557
Kraus's test, 435

LANDRY'S paralysis, 535, 544
Lankesteretta, 531
Laverania malarice, 521
Leguminosse, fixation of nitrogen,

by, 32

Leishman- Donovan body, 491
Leishman stain, 102, 524
Leishmaniosis, 491
Lenses, microscopical, 135, 139

— immersion, 137
Leprosy, 333

— diagnosis of, 337
Leprosy-like disease of rats, 337
Leptothrix buccalis, 460
Leucocytes, migration of, 202

— in milk, 616, 619
Leucocytozoa, 494, 530
Leucocytozoon canis, 530
Leuconostoc, 17
Levaditi's stain, 501
Life-history, studying, 119, 127
Life without bacteria, 2
Light as a disinfector, 627

— effect of, on bacteria, 23
Lime as a disinfectant, 634

— and water purification, 574, 575
Litmus media, 59

Local Government Board disin-
fecting solution, 636

Loffler's methylene blue, 99

— serum, 61
Loop, standard, 646
Luetin, 499

Lustgarten's bacillus, 339, 496
Lymphadenitis, ovine, 332
Lymphangitis, 235

— epizootic, 348, 474
Lysins, 150, 174, 181, 185
Lysol, 640

MACROGAMETE, 481
Macrophages, 203
Madura disease, 457
Hadurella, 459
Mai de caderas, 490
Malachite green media, 596
Malaria, 512

— diagnosis of, 523
- parasites, 519-523

• mosquito phase, 516

species, 519

Malignant disease, 554, 232, 462,
486, 511

— oedema, 425

clinical examination, 430

— pustule, 258
Mallein, 348

— in diagnosis, 349
Malta fever, 567
Marasmus, 239
Mastigophora, 487
Mastoid disease, 561
McConkey stain, 112

— media, 590, 591
McDougall's fluid, 640
McLeod's anaerobic method, 82
Measles, 561

Measurements, microscopical, 142
Measures and weights, 648
Meat, 371, 620

Media, culture, 54. See CULTURE

MEDIA

Medical antiseptics, 642
Mediterranean fever, 567
Meiostagmin reaction, 193
Membranous rhinitis, 273
Meningitis, 410, 561

— cerebro-spinal, 241

660

INDEX

Meningitis, posterior basic, 244
Mercaptan, 25, 37
Mercuric chloride, 635

— iodide, 636
Mercury pyogenic, 225

- vapour lamp, 134, 627
Merismopedia, definition of, 16
Metachromatic granules, 10
Metchnikoff s spirillum, 447
Methylated spirit, 86
Methylene blue, Loffler's, 99

— borax, 526

— carbol, 99

Micrococci, Gram- negative, 248
Micrococcus, definition of, 16

— agilis, 621

— bombycis, 532

— candicans, 621

- catarrhalis, 248, 417

— cereus albus, 231
flavus, 36, 231

— cinereus, 244

— deformans, 566

— epidermidis albus, 230, 604

— flavescens, 230

— gonorrhoea, 244, 566

— lanceolatus, 407

— Melitensis,561, 188, 248

— meningitidis, 241

— neoformans, 232, 554

— paramelitensis, 569

- Pasteuri, 407

- pyogenes aureus, 227

albus, 229

citreus, 229

tenuis, 407

— salivarius, 231, 604

— scurf, 231, 604

— tetragenus, 249

— urecR, 30, 36

— zymogenes, 231
Microgamete, 481
Micrometer, 142
Micro- millimetre, 143
Micron, 143, 648
Microphages, 203
Microscope, bacteriological, 132
Microsporidia. 531
Microsporon, Audouini, 475

Microsporon furfur, 479

— minutissimum, 479
Microtomes, 89, 92
Miescher's corpuscles, 532
Milk, 612

- diphtheria-like bacilli in,

276, 619

- examination of, 618

- leucocytes in, 616, 619

— organisms in, 613

- Pasteurisation of, 614

- pathogenic organisms in, 613

— examination for pathogenic

organisms, 618

- sour, 617

- standard for, 617

— sterilisation of, 614

— curdling of, 35, 69, 383, 613

— culture media, 59

- and tuberculosis, 310, 316, 321,

614

Moeller's spore stain, 114
Molluscum bodies, 555
Morax-Axenfeld bacillus, 557
Mosquitoes, 518

— and malaria, 515-519

— and yellow fever, 547
Motility of organisms, 11, 130
Moulds, 469

Mounting, 102
Mounting sections, 108
Mouse plague, 371

— septicaemia, 405

Mouth, organisms of, 460, 570
Movement, Brownian, 11, 130
Much's tubercle stain, 327
Mucor mucedo, 470

— rouxii, 471

Mucous membranes, organisms

of, 569
Mumps, 561
Mussel poisoning, 38
Mustard oil, 642
Mycelium, 469
Mycetoma, 457
Mycetozoa, 486
Mycoderma, 461
Mycoses, 472
Mycosis tonsillaris, 460

INDEX

661

Mytilotoxin, 38
Myxomycetes, 486
Myxosporidia, 531

NAGANA, 489

Nasal mucus germicidal action

of, 570
Nasgar, 242
Nastin, 335
Necrosis, 37
Needles, 50, 51
Negri bodies, 536
Neisser's stain, 294
Neuritis, diphtheritic, 273
Nitragin, 33
Nitrification, 28
— stages in, 30
— solutions for, 31
Nitrifying organisms, isolation of,

31

Nitrobacter, 30

Nitrogen, fixation of, 32

Nocardia, 459

Noctiluca, 487

Noguchi's method for cultivating

spirochaetes, 497
Noma, 562
Nomenclature, 19
Normal solutions, 64
Nosema, 531, 532
Nose, organisms, of, 570
Nose-piece, 142
Nucleins, 201

OBJECTIVES, 135, 139
Objects, measurement of, 142
(Edema, malignant, 425
Oidium albicans, 474

— lactis, 613
Oil-immersion, lenses, 137

Oils, essential, as antiseptics, 642
Old age, 571
Ookinet, 516

Ophthalmia, 246, 557, 566
Ophthalmitis, 230
Ophthalmo-reaction in glanders
349

— in tuberculosis, 330

Ophthalmo-reaction in tvphoid

370

Oppler-Boas bacillus, 562
Opsonic index, 210, 219

- determination of, 214
Opsonins, 210
Orange-rubin, 101
Organisms and disease, 118 143

147
Organisms, cultivation of, 66

— enumeration of, 77, 220

- identification of, 119

- influence of a mixture of, 21

- isolation of, 44, 77, 119

- of air, water, and soil, 621

— of air- passages, 570

— of conjunctive, 569

— of genital tract, 571

- of mouth, 570

— of nose, 570

— of skin, 569

- of stomach and intestine, 570

— of urinary tract, 571

— ultra-microscopic, 141

— variation of, 6

Osmic acid fixation, 98, 508
Osteomyelitis, 227, 228, 235 353
Otitis, 238, 410, 562
Oven, hot-air, 45
Ozsena, 563
Ozone, 600, 637

FAKES' discs, 583
Pappataci, 549
Pappenheim's solution, 340
Para-colon bacillus, 374
Para- dysentery, bacillus, 379
Para-typhoid fever, 374
Paraffin, embedding in, 90

— sections, 92

Paraffin sections, mounting, 109
Paralysis, diphtheritic, 273, 275
286

— general, 274, 497, 507

— infantile, 543
Landry's, 535, 544

Paramecium coli, 507, 560
Parasites, 145
Parotitis, 561

662

INDEX

Parthenogenesis, 481
Pasteurisation of milk, 614
Pasteur's fluid, 63
Peat, germicidal action of, 363,

437

Pebrine, 560
PeUagra, 563
Pemphigus, 560
Penicillium glaucum, 471
Peppermint oil, 642
Peptone water, 57
Pericarditis, 246, 410
Peritonitis, 410, 564
Permanganates, 599, 637, 641
Pertussis, 417
Petri dishes, 78
Petri's method for air analysis,

606

Petruschky's litmus whey, 385
Pfeiffer's reaction, 174, 177
Phagocytes, 203
Phagocytosis, 202

— estimation of, 210
Phenol, 638
Phlebitis, 226
Phlebotomus fever, 549
Phlogosin, 229

Phosphorescence, 37, 440, 487
Phycomycetes, 470
Physiological salt solution, 95
Picro- carmine, 101

Pictou cattle disease, 387

Piedra, 479

Pigment, formation of, 36

Pink torula, 621

Pinta, 479

Pipettes, 51, 53, 214

Piroplasmata, 528

Pitfield's flagella stain, 115

Pityriasis, 479

Plague antiserum, 400

— baciUus of, 392

— diagnosis, 403

— epidemiology, 400
- pathogenesis, 396

— vaccines, 398
Plasmodiophora brassicce, 486
Plasmodium, 512

— Kochii, 523

Plasmodium malarice, 519

— prcecox, 527

— vivax, 520
Plasticine, 52, 82, 215
Plate bottles, 79

— cultures, 76

agar, 80

anaerobic, 82

gelatin, 78

silica jelly, 31

Platinum needles, 50
Plant's method, 457
Pleomorphism, 16
Pleuropneumonia, 141, 407
Plimmer bodies, 554
Pneumobacillus of Friedlander,

412

Pneumococcus, Frankel's, 407
Pneumono-mycosis, 473
Pneumonia, 406, 371, 373, 391, 413

— diagnosis, 414
- septic, 235

Poisons, bacterial, 37, 146

— tolerance to, 197
Poliomyelitis, 543
Porcelain filters, 49, 601
Porges' reaction, 507
Post-mortems, 123
Postulates, Koch's, 147
Potassium permanganate, 599,

637, 641

Potato, 60. See CULTURE MEDIA
Powders, disinfectant, 642
Precipitins, 194
Pressure, effect of, on bacteria,

23

Products of bacteria, 24
Proskauer-Capaldi media, 385
Proteins, bacterial, pyogenic,

225

— germicidal, 199

— toxic, 39
Proteosoma, 527

Proteus capsulatus hominis, 258

— mirabilis, 621

— vulgaris, 621. See B. proteus

— Zenkeri, 621

— in putrefaction, 24, 30
Protophyta, 8

INDEX

663

Protozoa, 480

— action of drugs on, 642

— in water, 602

Pseudo- diphtheria bacillus, 287
Pseudo- diphtheria, relation to B.

diphtheria, 289
Pseudo-tuberculosis, 331
Pseudo mo nas, 19, 30
Psilosis, 564
Psittacosis, 371, 373
Psorospermosis, 511
Pto mines, 38
Puerperal fever, 564
Pugh's stain, 294
Pump, exhaust, 48
Purpura, 564
Pus, blue, 238

— in milk, 616
Putrefaction, 24

Pysemia, 223, 225, 226, 228, 235
Pyle-phlebitis, 226
Pyoctanin, 640
Pyocyanase, 239, 262
Pyocyaneus infection, 238, 541
Pyocyanin, 238
Pyogenic organisms, 223, 250
Pyorrhoea, 565
Pyrogallic acid, 73
Pyrosoma, 528

QUARTAN fever, 519
Quarter evil, 431
Quinine and malaria, 522

— and tetanus, 422

RABBIT septicaemia, 405
Rabies, 535

— diagnosis, 542

Radium, effect of, on bacteria,

24

Rag-sorter's disease, 258
Rat-bite disease, 565

— virus, 373

Rats and plague, 401
Rauschbrand, 431
Ray fungus, 452

Reaction, Bordet- Durham, 188,
192 i

— cholera-red, 27, 435

Reaction, indole, 25

— meiostagmin, 193

- Pfeiffer's, 174, 177

- Porges', 507

- Voges-Proskauer, 389

- Wassermann, 501
Rebipelagar, 592
Receptors, 154

— chemo-, 195, 206
Relapses, theory of, 368
Relapsing fever, 494
Resolving power, 140
Retention theory, 207
Rheumatism, 565
Rheumatoid arthritis, 566
Rhinitis, membranous, 273

— atrophic, 563
Rhinoscleroma, 566
Hhinosporidium kinealyi, 537
Rinderpest, 566
Ringworm, 475

— cultivation, 476

— examination, 479
Rocking microtome, 92
Roll cultures, 82
Romanowski stain, 102
Roup, diphtheritic, 298
Rubin, 101

Ruffer bodies, 554
Russell's corpuscles, 554

SACCHABIMETEB, 84
Saccharomyces, 461, 465
Saccharomyces anomalus, 467

- cerevisice, 465, 467

— ellipsoideus, 467
— litogenes, 462

— pastoriamis, 467

Saliva, germicidal action of, 570
Salt solution, physiological, 95
Salvarsan, 262, 493, 499, 642
Sand-fly fever, 549
Saprsemia, 225
Saprophytes, 21
Sarcina, definition of, 17

— lutea, 36, 621

— ventriculi, 249
Sarcoma, 462
Sarcosporidia, 532

664

INDEX

Sarkodina, 481
Saturation test, 193
Scarlet fever, 533
Schizomycetes, 8
Schizophycese, 8
Sclerotium, 469
Scour of poultry, 540
Sections, frozen, 88

- paraffin, 90

- fixing to slide, 93

— staining, 108

— to mount, 110

Sedgwick and Tucker's method

for air analysis, 607
Sedimentation test, 192
Septic diseases, 223
Septic tank process, 610
Septicaemia, 149, 223

- a, 149

Sera, anti-microbic, 173
— • antitoxic, 150

- polyvalent, 175, 270
Serum, culture medium of, 60-62

— germicidal action of, 199
Serum disease, 168
Seven- days' fever, 549
Sewage, 609

Sewers, air of, 610

Shake culture, 83

Shell-fish, examination of, 600

— pathogenic organisms in, 362
Side-chain theory, 152

Silica jelly, 31
Silkworms, disease of, 531
Silver salts, 636

- pyogenic, 225
Simulium, 563
Skatole, 28
Skatole-carboxylic acid, 27, 268,

288
Skin diseases, 479, 560

— organisms of, 569
Sleeping-sickness, 488
Slides, cleaning, 95, 114, 524

— hollow-ground, 129
Smallpox, 549
Smear preparations, 94
Smegma bacillus, 338

— staining, 339

Sodium bisulphate, 599
Soil, 608

Soil, nitrification in, 28
Solubilities, 648
Solutions, normal, 64
Sour milk, 617
Species of bacteria, 7, 18
Specimens, preserving patholo-
gical, 116
Spengler's tubercle stain, 326

— views on tuberculosis, 319
SpiriUa, 17, 433. See also Vibrio
Spirillum, definition of, 17

— choleras Asiaticce, 433

- varieties, 439-442

— of cholera, isolation from

water, 598

- of Tinkler and Prior, 448

— Metchnikovi, 447

- Obermeieri, 494

- rubrum, 36, 449

— tyrogenum, 449
Spirochaeta, 18, 493

- Duttoni, 494

— Obermeieri, 494

- pallida, 496

— pertenuis, 495

— recurrentis, 494

— refringens, 496

— Vincenti, 296

— in bronchitis, 557

— in cancer, 495

— in dysentery, 559,

— in ulcerating granuloma, 495

— in ulcers, 449

— in yaws, 495
Spirochaetosis, 493
Spironema pallidum, 496
Spleen, germicidal substance

from, 200

— in immunity, 204
Sporangium, 469
Spore formation, 14
Spore staining, 113, 468

Spores, resistance to heat, 20, 624
Sporidium vaccinale, 552
Sporotrichosis, 473
Sporozoa, 508
Spotted fever, 241

INDEX

665

Spotted fever, of rocky mountains,

546

Sprue, 564

Stage, microscopical, 132
Staining methods, 98. See under
respective names

— cover-glass specimens, 106

— capsules, 112

- flagelloa, 114

- Gram, 102

— sections, 110

- spores, 113, 468

Stains, 98. See under respective

names

Standard loop, 646
Standardisation of antitoxin, 280,
425

— of media, 64
Staphylococcus, 18

— species of. See Micrococcus
Steam as a disinfector, 625

- steriliser, 46
Stegomyia, 547
Sterilisation, 45, 624

— discontinuous 5

— of cotton- wool, 52

— of glass vessels, 53

- of milk, 614
Steriliser, hot air, 45
Steriliser, Griffith's, 599

— writer's, for milk, 616

- steam, 46, 625
Still's diplococcus, 244
Stimulins, 209
Stomach, organisms of, 570
Strangles, 237
Streptococcus, definition of, 16

— diagnostic table, 235

— anginosus, 234

— brevis, 233

— conglomerates , 233, 237, 534

— equinus, 235

— erysipelatis, 236

— foecalis, 234

— longus, 233

— mzdius, 233

— pyogenes, 232

— anti-serum, 237
in milk, 613, 616, 619

Streptococcus, rheunwticus, 234,
565, 566

— vdlivarius, 234, 570, 604

— scarlatince, 233, 237, 534

— viridans, 236 .
Streptothrix infections, 450
Streptothrix, acid-fast, 299, 450

— actinomyces, 453

— Eppingeri, 450, 459

- Freeri, 459

- leproides, 335

— madurw, 458

- Nocardii, 452, 459
Streptotricheae, 450
Sub-stage condenser, 139
Sub-tertian fever, 521
Sugars, resolution of, 22
Sulphurous acid, 633
Supersensitisation, 169
Suppuration, 223

— clinical examination, 239

— conditions modifying, 226

— due to chemical agents, 225

— influence of dose, 227

embolism, 226

injury, 227

Surgical antiseptics, 642
Surra, 490

Swine erysipelas, 405

- fever, 372

- plague, 373
Symbiosis, 21
Symptomatic anthrax, 431
Syphilis, 496

Syringes, 122

TABES, 505, 507

Temperature influence on bac-
teria, 20
Test-tubes, 50
Tetanus, 419

— animals susceptible to, 421

— clinical examination, 425

- bacillus, 420

— — associated organisms, 421
and quinine, 422

- antitoxin, 424
— toxins, 422
Tertian fever, 520

43

666

INDEX

Texas fever, 529

Theobald Smith phenomenon, 171

Thermal death-point of organisms
in milk, 615

Thermal death-point, determina-
tion of, 626

Thermophilic bacteria, 20, 608

Thionin, carbol, 100

Thrush, 474

Ticks, 495, 530

Tinea, 475

Tinfoil, 52

Tissue-fibrinogen, 201

Tissues, preparation of, 86

Tolerance to poisons, 195

Torula, pink, 621

Torulae, 461

Toxins, 39, 153, 162

Toxoids, 156, 157, 281

Toxones, 281, 282

Toxone effect, 165, 282

Toxophile group, 154

Toxophore group, 153

Trachoma, 566

Treatment, antiseptic, 642

- antitoxic, 160
Treponema pallidum, 496
Trichomonas, 487

— species of, 496
Trychophytons, 475
Trypanoplasma, 487
Trypanosoma, 487

- species of, 488^91
Tsetse flies, 490

- fly disease, 489
Tube, microscope, 135
Tubercle anti-sera, 322

— structure of, 300

— bacillus, 301

agglutination, 324

Tubercle bacillus, avain variety,

312, 321

cultivation of, 303

distribution in tissues, 311

mammalian variety, 312,

313

staining peculiarities, 302

— thermal death-point, 309,

615

Tubercle bacillus, toxins of, 308
in the blood, 311

- in butter, 622

- in milk, 316, 321, 614,

618
Tuberculin, Behring's, 307

— bacillary emulsion, 307

— cutaneous reaction, 336

- new, 306

- old, 304

— ophthalmo reaction, 330

- R, 306

- reaction in actinomycosis, 455

— in leprosy, 334

— treatment, 307

— veterinary, 331
Tuberculosis, 299

— anatomy of, 300

— avain, 312

— bovine, 312, 313

— complement fixation in, 323

— diagnosis of, 323 et seq.

— disinfection in, 322

— immunity in, 322

— in the horse, 313

— mammalian, 312

— piscian, 312

— precipitin reaction in, 324

— pseudo-, 331

— Royal Commission on, 314,

319, 320

— Splengler's views on, 319

- spread of, 320
Tuberculous food, 320

— sputum, staining, 324

— tissues, staining, 327

- urine, 329
Tulase, 307

Turpentine, pyogenic. 225
Twort's stain, 485
Typhoid bacillus, 352

— carriers, 359

— in the blood, 354

— in milk, 613

— in water, 360, 593

- serum, 366

— survival of, 359, 360

— isolation from stools, 370

— variation of, 368

INDEX

667

Typhoid vaccine, 367
Typhoid fever, 352

— agglutination reaction, 356,

370

— and oysters, 362

— and sewer gas, 365

- diagnosis of, 369

— in animals, 355
Typhus fever, 554
Tyrotoxicon, 39, 614

ULCERATING granuloma, 495
Ulcerative endocarditis. See

Endocarditis, infective
Ulcers, 449

Ultra- microscopic organisms, 141
Ultra-violet light, 7, 23, 254, 627
Undulant fever, 188
Units, antitoxin, 284
Unna's method, 111
Urea, fermentation of, 30, 36
Urinary organs, organisms of, 571
Urine, colon bacillus in, 387

— smegma bacillus in, 329, 339

— tubercle bacillus in, 329

— typhoid bacillus in, 355
Uschinsky's fluid, 63

VACCINES, dosage of, 221

— endotoxic, 222

- prophylactic, 221

- sensitised, 222

— standardisation, of, 220

- therapeutic, 219

— (also under individual or-

ganisms)

Anthrax, 262

Cholera, 444

Plague. 398

Typhoid, 367

Vaccinia, 553
Vaccinia, 549
Vaginal organisms, 571
Van Ermengem's flagella stain

114

Variola, 549
Vibrio, definition of, 17

- cholera, 433

— Berolinensis. 440

Vibrio Danubicus, 440

- Deneke, 449

- El Tor, 441

- Elwers, 440

- Finkler, 448

- Ivanhoff,448

- Massowah, 440, 441, 443

- Metchnikovi, 447

- Sanarelli, 463
Vibrios of mouth, 449
Vincent's angina, 296, 19
Virulence, to increase, 121
Visibility, limit of, 141
Vitality of cultures, 121
Voges-Proskauer reaction, 389
Volvox, 487

Vorticella, 507

WASSERMANN reaction, 501

- in leprosy, 334, 502

— in malaria, 502

— in scarlatina, 502, 535

— in trypanosomiasis, 502

- in yaws, 495, 502
Water, bacteriology of, 572

- number of organisms in, 573

- effect of sand filtration, 574

— effect of sedimentation, 574

- effect of storage, 573

— bacteriological analysis of,

576

- pathogenic organisms in, 593

- colon bacillus in, 572, 584, 587,

590

— comma bacillus in, 437, 598

- sterilisation of, 599

— sterilisers, 599

- typhoid bacillus in, 360, 593
Watercress, examination of, 600
Weigert's law, 155

- methods, 105, 111
Weights and measures, 648
Whooping-cough, 417

Widal reaction, 187, 190, 356,

370

Wooden tongue, 451
Wool-sorter's disease, 258
Wright's capsule, 214

XEROSIS bacillus, 297

668

YAWS, 495
Yeasts, 460, 9

— analysis of, 466
— isolation, 465

— of fermentation, 465
- pathogenic, 462

Yellow fever, 546, 373

INDEX

ZIEHL - NEELSEN solution,

100

Zinc chloride, 641
Zoogloea, 10
Zygospore, 469
Zygote, 516
Zymolysis, 34

JOSEPH D, HOOGEN D,

BALL ANT YNE AND COMPANY

TAVISTOCK STREET COVENT GARDEN
LONDON

UNIVERSITY OF CALIFORNIA

MEDICAL CENTER LIBRARY

THIS BOOK IS DUE ON THE LAST DATE
STAMPED BELOW

Books not returned on time are subject to a fine of 50c
per volume after the third day overdue, increasing to
$1.00 per volume after the sixth day. Books not in de-
mand may be renewed if application is made before expi-
ration of loan period.

2m-10,'40(9371s)

//-