Full text of "Bacteriology, general, pathological and intestinal"

BIOLOGY
UBffeRY

BACTERIOLOGY

GENERAL, PATHOLOGICAL

AND

INTESTINAL

ARTHUR ISAAC KENDALL, B.S., PH.D., DR.P.H.

PROFESSOR OF BACTERIOLOGY IN THE NORTHWESTERN UNIVERSITY MEDICAL SCHOOL,

CHICAGO, ILLINOIS

ILLUSTRATED WITH 98 ENGRAVINGS AND 9 PLATES

LEA & FEBIGER

PHILADELPHIA AND NEW YORK
1916

BIOLOGY
RA

Entered according to the Act of Congress, in the year 1916, by

LEA & FEBIGER,
in the Office of the Librarian of Congress. All rights reserved.

THEOBALD SMITH, M.D., LL.D.

PREFACE.

"!N the study of the microscopic forms known as bacteria we have
what might be fitly called the focal points of the various branches of
biological science. Though their investigation may require careful
morphological researches, yet the unmistakable monotony of form
combined with a considerable variation of physiological activity has
compelled the bacteriologist to pay much attention to means by which
such physiological variations may be more or less accurately registered
in order that they may serve as a supplementary basis for classification.
Again, with unicellular organisms the manifestations of cell activity
become the most important phenomena for study. These manifesta-
tions bring together the fields of physiology and chemistry and make
bacteriology in one sense a branch of physiological chemistry." 1

" There is no ulterior interest in the study of bacteria as such, which
is a strong impulse in many other departments of biological science. It
is what bacteria do, rather than what they are, that commands atten-
tion, since our interest centers in the host rather than the parasite." 2

The development of bacteriology has followed very closely the
gradual improvement of the optical parts of the compound microscope,
and to a lesser degree, the perfection of other instruments of precision
on the one hand, and the production of anilin dyes and a great expan-
sion of the fields of organic and physical chemistry on the other hand.
Naturally the greatest advances in bacteriology have been made along
the lines of morphology, staining and diagnosis, because the application
of the microscope, anilin dyes, and the preparation and use of cultural
media to bacterial problems is relatively simple and direct. The final
chapters of bacteriology, in which the problems of immunology are
of paramount interest, will be intimately associated with an unfolding
of the chemistry of cellular activity, as Theobald Smith has so clearly
pointed out in the opening paragraphs of this discussion.

1 Theobald Smith. The Fermentation Tube, Wilder Quarter Century Book, 1893,
p. 187.

2 Theobald Smith. Some Problems in the Life History of Pathogenic Microorgan-
isms, Amer. Med., 19Q4, viii, 711.

vi PREFACE

The chemistry of bacterial activity is not thoroughly studied at
the present time and many of its problems must await the develop-
ment of new methods of chemistry and physics, as well as a refine-
ment of existing methods. Nevertheless, sufficient information exists
to warrant its presentation in concrete form, partly to emphasize its
deficiencies, chiefly to indicate its relation to the biology of the bacteria,
which are potentially "living chemical reagents," as Professor Folin
has so aptly termed them.

In the last analysis, the interest and importance of bacteria centers
around "what they do rather than what they are," and the elucida-
tion of this aspect of bacteriology lies largely within the field of
biochemistry.

The relation of the chemistry of bacterial nutrition to the study
of intestinal bacteriology in health and disease is self-evident; some
of the more general aspects of this subject are briefly set forth in the
chapter relating to intestinal bacteria.

It is with great pleasure that the writer acknowledges his indebted-
ness to his colleagues in the Northwestern Univeristy Medical School
for many valuable suggestions, to Doctors Noguchi and -Amoss, of
the Rockefeller Institute, for the privilege of using the original plates
illustrating the Treponemata and Poliomyelitis, and to Mrs. N. M.
Frain for the line drawings in the text. Finally, the writer would
acknowledge his deep obligation to Miss Bertha J. Schwarz, Secretary
of the Department of Bacteriology, for her invaluable assistance in
the preparation of the manuscript and in reading the proof of the
book. A. I. K.

CHICAGO, 1916.

CONTENTS.

SECTION I.-GENERAL BACTERIOLOGY.

INTRODUCTION. THE DEVELOPMENT AND SCOPE OF
BACTERIOLOGY.

CHAPTER I.
THE MORPHOLOGY OF BACTERIA.

PAGE

Normal and Abnormal Forms Size and Weight Structure and Con-
stituents of the Bacterial Cell Reproduction and Cell Division
Cell Grouping, Classification, and Mutation 17-35

CHAPTER II.

THE PHYSIOLOGY OF BACTERIA AND THE EFFECT OF ENVIRONMENTAL

INFLUENCES.

Rate of Reproduction Motility Germination of Spores Longevity
Effects of Moisture, Oxygen, Temperature, Light and Electricity,
Gravity, Osmotic Pressure Production of Enzymes, "Toxins, Pto-
maines and Pigments Symbiosis, Antibiosis, Commensalism . . 36-55

CHAPTER III.
THE CHEMISTRY OF BACTERIA.

Chemical Constitution of Bacteria and Composition of Morphological
Components of Bacterial Cell Food Relationships of Bacteria,
Bacterial Nutrition 56-67

CHAPTER IV.
BACTERIAL METABOLISM.

The Nature of Bacterial Metabolism Nitrogen Metabolism, Carbon
Metabolism Reactions of Bacterial Metabolism Significance of
Bacterial Metabolism Putrefaction and Fermentation . 68-83

viii CONTENTS

CHAPTER V.
SAPROPHYTISM, PARASITISM AND PATHOGENISM.

PAGE

The Cycle of Parasitism The Cycle of Pathogenism Distribution of
Parasitic and Pathogenic Bacteria in Nature How Parasitic and
Pathogenic Bacteria Reach Man How they Reach the Body,
Portals of Entry, Where They Multiply in the Body, Where and
How They Escape from the Body Balanced Pathogenism and
Epidemiology 84-110

CHAPTER VI.

INFECTION AND IMMUNITY.

Classification of Immunity Infection Theories of Immunity . . . 111-131

CHAPTER VII.
Anaphylaxis, Allergy or Hypersensitiveness 132-141

CHAPTER VIII.
ANTIGENS AND THE TECHNIQUE OF SERUM REACTIONS.

Nature of Antigens and Antibodies Agglutinins and Precipitins
Lysins, Hemolysis and Complement Fixation Aggressins Opson-
ins, Tropins Bacterial Vaccines . .....".. . . 142-174

CHAPTER IX.
BACTERIOLOGICAL TECHNIQUE.

Methods for the Microscopic Study of Bacteria Staining Methods-
Media Cultivation of Bacteria, Study of Bacterial Cultures . . 175-223

CHAPTER X.
BACTERIOLOGICAL EXAMINATION OF MATERIAL FROM PATIENT AND CADAVER.

Autopsy Procedure Blood Cultures Cerebrospinal Fluid Peritoneal,
Pleural and Pericardial Fluids Pus Examination of Urine,
Feces, Sputum, Buccal and Pharyngeal Material Bacteriological
Examination of the Eye, Ear and Nose The Utilization of Animals
for Bacterial Diagnosis and Experimentation ....... 224-240

CHAPTER XI.
PRACTICAL STERILIZATION, ANTISEPSIS AND DISINFECTION.

Laboratory Sterilization Physical Agents, Chemical Solutions, Test-
ing and Standardizing Liquid Disinfectants Gaseous Disinfectants
Disinfection of Sputum, Vomitus, Feces and Urine, Fomites, Skin
and Hands, Instruments . . ... -V 241-254

CONTENTS IX

SECTION II.-PATHOGENIC BACTERIA.

CHAPTER XII.

PAGE

The Pyogenic Cocci 255-268

CHAPTER XIII.
The Streptococcus-Pneumococcus Group 269-291

CHAPTER XIV.
The Meningococcus-Gonococcus Group 292-309

CHAPTER XV.
Micrococcus Melitensis 310-312

CHAPTER XVI.
The Alcaligenes Dysentery Typhoid Paratyphoid Group . . . 313-352

CHAPTER XVII.
The Coli Cloacae Proteus Group 353-362

CHAPTER XVIII.
The Mucosus Capsulatus Group 363-366

CHAPTER XIX.

Glanders, Anthrax, Pyocyaneus, Infectious Abortion, Aciduric Bacteria 367-387

CHAPTER XX.

Diphtheria Group . . . 388-406

CHAPTER XXI.
Hemorrhagic Septicemia Group 407-416

CHAPTER XXII.

HEMOGLOBINOPHILIC BACILLI.
Influenza, Pertussis, Koch- Weeks, Morax-Axenfeld and Ducrey Bacilli . 417-427

CHAPTER XXIII.

TUBERCLE BACILLUS GROUP.
Human, Bovine and Avian ... 428-462

x CONTENTS

CHAPTER XXIV.

PAGE

Leprosy and Acid-fast Bacteria other than the Tubercle Group . 463-471

CHAPTER XXV.

ANAEROBIC BACTERIA.

Tetanus, Botulinus, Aerogenes Capsulatus, Malignant Edema and

Symptomatic Anthrax . . 472-498

CHAPTER XXVI.
Cholera Group 499-513

CHAPTER XXVII.
Treponemata and Spirocheta 514-532

SECTION III.-HIGHER BACTERIA, MOLDS,
YEASTS, FILTERABLE VIRUSES, AND
DISEASES OF UNKNOWN ETIOLOGY.

CHAPTER XXVIII.

Trichomycetes, Actinomycetes, Hyphomycetes and Saccharomycetes . 533-554

CHAPTER XXIX.
Filterable Viruses Diseases of Unknown Etiology 555-578

SECTION IV.-GASTRO-INTESTINAL
BACTERIOLOGY.

CHAPTER XXX.
Gastro-intestinal Bacteriology 579-600

SECTION V.-APPLIED BACTERIOLOGY.

CHAPTER XXXI.
Bacteriology of Milk 601-613

CHAPTER XXXII.

Bacteriology of the Soil, Water and Air '^ . . . . 614-625

SECTION I.

GENERAL BACTERIOLOGY.

INTRODUCTION THE DEVELOPMENT AND SCOPE OF

BACTERIOLOGY.

BACTERIOLOGY is that branch of Natural Science which treats of
the structure, functions and chemistry of bacteria. Bacteria are
intimately related to many fields of human activity, therefore bacterio-
logy is inseparably associated with a number of the arts and sciences.
In those branches of science which treat of the diseases of plants,
of animals and of man, bacteria enter into complex reciprocal relations
with their hosts as parasites or pathogens, relations which are neither
purely bacterial, animal nor vegetal in their limitation. A new science,
Immunology, is rapidly developing which is concerned chiefly with
the elucidation of these relationships between host and parasite.

Bacteria are the smallest in size and simplest in structure of known
visible living organisms. They are rigid unicellular organisms devoid
of chlorophyll or other photodynamic pigment; they possess no
morphologically demonstrable nucleus and reproduce by simple
transverse fission, the resulting individuals being of approximately
equal size.

Bacteria are ubiquitous in their distribution; they are found in all
climates in association with animal and vegetable life. Some thrive
at temperatures but slightly above the freezing point of water; the
majority flourish between 15 and 40 Centigrade; some even develop
in thermal springs at a temperature of 70 Centigrade. Free or atmos-
pheric oxygen is essential for most types of bacteria, but to a few it
is actually a poison.

Bacteria are ordinarily classed as plants, but they exhibit several
prominent characteristics which suggest a relationship with the lowest

animals. The most important of these is the absence of photodynamic
2

18 /. ',' ". ; . : INTRODUCTION

pigment (chlorophyll), which implies an analytical or destructive
function in the economy of Nature.

The great majority of bacteria are saprophytic, living upon dead
organic matter, which they transform into simple compounds suitable
for plant use. These bacteria are Nature's analysts. Some are para-
sitic on living plants and animals; a few are progressively pathogenic
for man and animals. It is this last group, few in numbers, but for-
midable in that their activities are in partial opposition to those of
man and animals, that has given to bacteria all the notoriety which
they possess.

Anton von Leeuwenhoek, a Dutch spectacle maker, appears to have
been the first to see bacteria: in 1675, with lenses of his own grinding,
he examined various putrescent fluids, drops of water, scrapings from
his teeth, and his own diarrheal discharges. He says in his writings,
collected and edited by Robert Hooke, 1 "With great astonishment
I observed everywhere through the material which I was examining,
animalcules of the most minute size, which moved themselves about
very energetically." It is possible to recognize cocci, bacilli and
spirilla in his drawings, and it is almost certain that he actually observed
motility among his organisms. The learned monk, Athanasius Kircher,
observed and described "minute living worms" as early as 1659, but
his optical equipment was inferior to that of von Leeuwenhoek and it
is doubtful if he actually saw bacteria.

Improvements in the microscope opened a new world for investiga-
tion and speculations concerning the doctrine of the Spontaneous
Generation of Life led to numerous experiments of increasing refine-
ment that finally resulted in the brilliant researches of Pasteur, and
Tyndall, who showed by numerous ingenious and carefully executed
experiments that the phenomena in putrescible fluids erroneously
interpreted as spontaneous generation did not take place when proper
precautions in manipulation were observed. About 1835 achromatic
lenses for the microscope reached a state of perfection compatible
with the examination of minute objects and the microscope was almost
immediately applied to the study of various morbid processes, with
remarkable success. Bassi (1837) discovered a fungus which caused
a contagious disease of silk worms known as muscardine; Cagniard
de Latour and Schwann observed and described the yeast plant in
liquids undergoing alcoholic fermentation.

1 Collected Memoirs of Anton v. Leeuwenhoek, Royal Society of London, 1675, 1683.

INTRODUCTION 19

Ehrenberg (1838) began his classification of animalcules and in his
group of Vibrionia described several "species" of organisms, as follows:

1. Bacterium rigid and filamentous organisms.

2. Vibrio flexuous and filamentous organisms.

3. Spirillum rigid spiral filamentous organisms.

4. Spirocheta flexuous spiral filamentous organisms.

This classification, which contains terms widely used in bacterial
nomenclature today, was followed in 1872 by the important contribu-
tions of Cohn upon "Bacteria," the starting-point of modern bacterial
classification.

The diseases of man naturally attracted much attention and in 1839
Schoenlein examined the crusts of that disease of the scalp known as
favus with the microscope and found the mycelia of the fungus now
known in his honor as Achorion schoenleinii.

The extensive studies of Pasteur upon yeasts and the "diseases"
of beer and wine, upon the diseases of the silk worm (pebrine and
flacherie), upon furunculosis and puerperal sepsis, 1 upon anthrax
and anthrax immunization (attenuated viruses) chicken cholera, and
somewhat later, rabies laid broad foundations for the development
of the science of bacteriology.

Among the most important technical discoveries which have con-
tributed to the development of bacteriology are: The improvement
in the achromatic lens (about 1835) and the perfection of the sub-
stage condenser (Abbe) ; the use of cotton for air filters in flasks and
test-tubes by Schroeder and von Dusch (1854), the sterilization of
culture media by heat (Pasteur, Tyndall, Koch and others), the
introduction of anilin dyes as staining reagents by Weigert and Ehrlich
(1877), and finally, the use of solid culture media and the plate method
for pure cultures by Koch in 1881.

Sir Joseph Lister (1867) published an epoch-making contribution
entitled, "On the Antiseptic Principle of the Practice of Surgery,"
in which is clearly set forth the importance of bacteria in surgery
and the principles of surgical asepsis that have revolutionized this
branch of medicine.

About 1878 Koch isolated the anthrax bacillus in pure culture from
the blood of infected animals, grew the organisms for several generations
in the clear aqueous humor of the eye of the ox, and then reinjected
the organisms into experimental animals and reproduced the disease.
For the first time a specific microbe was clearly and convincingly

1 Compt. rend. Acad. d. Sci., 1880, xc, 1033.

20 INTRODUCTION .

shown to be the etiological factor of a bacterial disease. Koch also
found that the anthrax bacillus formed spores.

From this time bacteriology developed with amazing rapidity. In
1882 Koch startled the world with the announcement of the dis-
covery of the tubercle bacillus; and in rapid succession, typhoid,
diphtheria, cholera, tetanus and other well-known pathogenic bacteria
were isolated and studied in pure culture.

In 1882 Metchnikoff published the first of his highly important
contributions to immunity and phagocytosis, and a decade later von
Behring and Kitasato announced the discovery of diphtheria antitoxin.

The last three decades have not only witnessed the rise and develop-
ment of those most brilliant chapters of medicine, infection and im-
munity; but sanitation, agriculture, many industries and other fields
of human activity have benefited largely by the development of
bacteriology.

In medicine the diagnosis of bacterial disease has reached a high
degree of precision, and bacteriological diagnosis is an important
branch of medical science. The most important problem for the future
is to create a system of Bacterial Therapeutics of equal efficiency.

CHAPTER I.

THE MORPHOLOGY OF BACTERIA.

A. MORPHOLOGY NORMAL FORMS:

Coccus, BACILLUS, SPIRILLUM.

B. MORPHOLOGY ATYPICAL AND AB-

NORMAL FORMS.

1. Variation.

2. Degeneration and Involution.

3. Pleiomorphism.

4. Branching.

C. SIZE OF BACTERIA: WEIGHT OF

BACTERIA.

D. STRUCTURE AND CONSTITUENTS OF

THE BACTERIAL CELL.
1. Cell Membrane, Ectoplasm,
Capsule, Zooglea.

2. Cell Substance, Cytoplasm,
Nucleus, Metachromatic and
Polar Granules, Flagella,
Spores, Germination of Spores,
Arthrospores.

E. REPRODUCTION AND CELL DIVISION

IN BACTERIA.

F. CELL GROUPING.

G. CLASSIFICATION OF BACTERIA.

1. Relation of Bacteria to Higher

Plants.

2. Classification.

H. MUTATION. CONSTANCY OF TYPES.

A. NORMAL FORMS: COCCI, BACILLI, SPIRILLA.

THE normal forms of the true bacteria are very simple, and are
included in three fundamental types: the sphere (coccus, plural cocci),
the straight rod (bacillus, plural bacilli), and the curved rod (spirillum,
plural spirilla) . There is in addition a group of organisms intermediate
between the true bacteria and the molds, which is characterized by
a filamentous type of growth. The members comprising this group
of filamentous organisms are commonly known as the higher bacteria
or Chlamydobacteriacese. An organism belonging to one of these
groups always reproduces its kind under normal conditions; that is,
a coccus always reproduces a coccus, a bacillus always reproduces a
bacillus, and a spirillum always reproduces a spirillum.

Cocci. A single coccus is typically spherical, although those
organisms in which division is taking place may be temporarily some-
what elongated in one diameter, thus appearing oval in outline at this
stage of their development. They may even resemble very short
bacilli in extreme instances. The habitual occurrence of cocci in pairs,
frequently with their proximate surfaces flattened, is a noteworthy
morphological characteristic of certain members of this group. They
are referred to as diplococci. The flattening of the proximated sur-
faces may be associated with an elongation of the axes of the organisms
parallel to the plane of apposition, which leads to "coffee bean"

22 THE MORPHOLOGY OF BACTERIA

shaped diplococci, exemplified in the meningococcus and gonococcus,
or to an elongation of the axes perpendicular to the plane of apposition,
in which event the organisms are "lance-shaped" diplococci, as for
example the pneumococcus.

Bacilli. Bacilli are rod-shaped, cylindrical organisms in which a
longer and a shorter dimension may be recognized. They are typi-
cally circular in cross-section. When division is taking place the shorter
bacilli may be temporarily oval or even circular in outline. The dimen-
sions of bacilli vary considerably: some are habitually long, some are
short, some are thick, some are thin. The ends may be convex, less
commonly flat or even concave. A few bacilli are not typically isodia-
metric, but appear in outline as club-shaped, spindle-shaped, or even
more or less conical (cuneate) rods. Less commonly, slightly curved
rods are met with; the curvature takes place along the longer dimension.

6 ooooooo

FIG. 1. The normal types of bacteria. 1-6, cocci; 7-13, bacilli; 14-16, spirilla;
1, micrococcus; 2 and 3, diplococci; 4, tetracoccus; 5, sarcina; 6, streptococcus (the
lower chain includes an arthrospore) ; 7 and 8, bacilli; 9, 10, 12, and 13, bacilli with
various granules; 11, strep tobacillus ; 14, vibrio; 15, spirillum; 16, Spirocheta trepo-

Spirilla. Spiral bacteria, like the bacilli, exhibit a longer and
a shorter dimension; unlike the bacilli, the longer axis is curved in
three planes of space. The curvature may be slight, less than a com-
plete turn, in which event the organism is "comma-shaped" when
viewed under the microscope; it may be a series of open curves, giving
the organism a sinuous outline; or it may be very much curved, so
that the organism resembles a somewhat closely coiled spring in out-
line. As a rule, the curvature is symmetrical and uniform in each
instance.

The cocci, through almost imperceptible morphological gradations,
merge into the bacilli, and the bacilli, through the slightly curved
forms, merge into the spirilla. Even in the spirilla slight differences
in curvature are usually discernible. Thus, a culture of the cholera
vibrio may contain many straight, uncurved organisms in addition

ABNORMAL FORMS 23

to the slightly curved rods which are the characteristic morphologic
forms. There are a few bacteria in which the morphology is still a
subject of controversy. For example, Micrococcus melitensis is called
Bacillus melitensis by some observers. The vast majority of bacteria,
however, are easily referable to their proper morphological type by
simple inspection under the microscope.

B. ABNORMAL FORMS: VARIATION, DEGENERATION AND IN-
VOLUTION, PLEIOMORPHISM AND BRANCHING.

Variation. The composition of the medium in which bacteria are
growing, the age of the culture, and to a limited degree even the tem-
perature of incubation influence somewhat the average size of bacteria.
Given constant conditions, however, bacteria growing in a favorable
environment exhibit constancy of form and size, although a few organ-
isms in every culture are somewhat larger or smaller than their fellows,
appearing as occasional giants or dwarfs. These occasional giant and
dwarf forms represent normal variations in size from the average or
mean.

Degeneration and Involution. Bacteria growing in an unfavorable
environment, brought about by the accumulation of waste products,
by undue changes in reaction resulting in excessive acidity or alka-
linity, by the presence of harmful chemicals, or by specific antago-
nistic substances, may gradually assume atypical shapes, probably the
direct result of these harmful influences. These atypical organisms
may exhibit little or no resemblance to the normal organism, either in
form or size; they may or may not develop into normal organisms when
they are placed again in a favorable environment. If the change is
a morphological one, the atypical organisms are designated involution
forms : thus, plague bacilli grown in nutrient agar containing 3 per cent,
common salt appear as swollen, balloon-like bodies, notably unlike
the typical short rod-shaped bacillus. If, on the contrary, the
organisms permanently lose some morphological or chemical charac-
teristic, they are spoken of as degeneration forms. Thus, anthrax
bacilli heated for several hours at 43 to 44 C. lose their ability to
form mature spores.

Pleiomorphism. By pleiomorphism is meant a permanent or semi-
permanent change in the normal form of the organism. A pleio-
morphic organism would be one which might at one time resemble
a bacillus, again a coccus, or even a spirillum, depending upon the age

24 THE MORPHOLOGY OF BACTERIA

and growth of the organism or the fitness of the culture medium. This
phenomenon is rarely or never met with among the pathogenic bacteria.

Branching. Among the individual organisms comprising a culture
in artificial media of tubercle, diphtheria or glanders bacilli, and to a
lesser extent of other bacilli, a certain number appear as definitely
branched rods: the typical organism in each instance does not exhibit
branching. Branching has also been demonstrated in the spirilla. 1
Bacillus bifidus appears habitually as a rod-shaped organism with
bifurcated ends in artificial media, although it is an unbranched
bacillus in its normal habitat, the intestinal tract of nurslings. Occa-
sionally, bacteria, as the tubercle bacillus, may exhibit branching in
the animal body as well as in cultures, although less commonly.

The cause of this branching is unknown, and at least two theories
have been advanced in explanation of it: each theory has a certain
amount of evidence in its favor. One theory assumes that branching
is the result of unfavorable environmental conditions, and it has been
shown that old broth cultures of diphtheria bacilli contain branched
organisms; young cultures contain few or no branched forms. The
assumption is that old cultures contain harmful products of metab-
olism which cause the diphtheria bacillus to assume branched forms.
The second theory asserts that the appearance of branched forms
among bacteria demonstrates a relationship between them and higher
organisms, which are habitually branched. Bacteria, according to
this theory, exhibit branching as a part of their normal development.

Branching does not necessarily take place under conditions which
would appear to be unfavorable or partially inimical to their growth,
and, on the other hand, it may be observed occasionally when environ-
mental conditions should be favorable for development. It appears
to be reasonable to assume that branching may be a normal develop-
mental process in the life history of the organism, although the phylo-
genetic significance of branching is as yet undetermined.

C. SIZE AND WEIGHT OF BACTERIA.

Size. The unit of measurement for microscopic objects is the
micron (/*), which is 0.001 of a millimeter, or approximately ^m of
an inch, in length. Bacteria are the smallest known living organisms
which have been seen with the microscope. Measured with this unit,
they exhibit considerable differences in size. The average sized pus-

1 Reichenbach, Centralbl. f. Bakteriol., 1901, xxix ,553.

STRUCTURE AND CONSTITUENTS OF BACTERIAL CELL 25

producing coccus is 0.8 micron in diameter; Micrococcus melitensis,
the smallest of the Coccacese, varies in diameter from 0.3 to 0.5 micron.
The largest known bacillus, B. biitschlii 1 is 3 to 6 microns in diameter
and from 40 to 60 microns in length. The smallest known bacillus,
B. influenzse, is but 0.2 by 0.5 micron in diameter; an average sized
bacillus would measure about 2 microns in length and 1 micron in
diameter. Spirillum colossum 2 is from 2.5 to 3.5 microns in diameter.
The cholera vibrio is about 2.5 microns long and 1 micron in diameter.
There are certain living viruses of unknown morphology, so-called
ultramicroscopic or filtrable viruses, which are either somewhat
smaller than any known bacteria or more plastic. Viruses belonging
to this group derive their name from the fact that they retain their
viability even after passage through the pores of standard, unglazed
porcelain filters, which will hold back even the smallest bacteria.

Weight of Bacterial Cell. The weight of a bacterial cell is depend-
ent upon its size and its specific gravity. According to Rubner, 3 the
specific gravity of common bacteria varies between 1.038 and 1.065. 4
B. coli is an average sized cylindrical rod (bacillus), measuring 1
micron in diameter and 2 microns in length. The volume of a cylinder
is the product of the diameter squared, multiplied by 0.7854, multiplied
by the length of the cylinder. The volume of a single colon bacillus
consequently would be (0.001) 2 X 0.7854 X 0.002, or 0.00000000157
c.mm. The weight of a single colon bacillus would be the volume
multiplied by the specific gravity, which is approximately 1.040 or
0.00000000163 mg.; that is to say, sixteen hundred million colon bacilli
would weight approximately one milligram. For purposes of com-
parison it may be stated that a single red blood corpuscle (human)
weighs about 0.00008 mg., about fifty thousand times the weight
of a single colon bacillus.

D. STRUCTURE AND CONSTITUENTS OF THE BACTERIAL CELL.

The typical bacterial cell consists essentially of protoplasmic cell
substance, endoplasm, enclosed by a rigid cell membrane, ectoplasm.

1. Cell Membrane. Ectoplasm. Bacteria appear to possess a
special external boundary layer, cell membrane, or ectoplasm, which is

1 Schaudinn, Arch. f. Protistenk., 1902, i, 306.

2 Errera, Recueil de 1'Instit. botanique (Universite de Bruxelles), 1901, v, 347.

3 Arch. f. Hyg., 1903, xlvi, 41; 1890, xi, 385.

4 Stigell (Cent. f. Bakt., 1908, xlv, 487) finds that the specific gravity of the same
organism varies somewhat with the medium in which it is grown. The specific gravity
of ordinary bacteria varies commonly between 1.120 and 1.35, older cultures being as
a rule of less specific gravity than younger cultures of the same kind.

26 THE MORPHOLOGY OF BACTERIA

rigid and maintains the shape of the organism. Generally speaking,
this cell membrane is intermediate in character between that char-
acteristic of animal and of plant cells respectively, being somewhat
more developed than the former, less highly specialized as a rule than
the latter. Some authorities consider the cell membrane of bacteria
to be merely a concentrated external layer of endoplasm.

The thickness of the cell membrane varies among different varieties
of bacteria, and it appears to be somewhat -thinner in young organisms
of a given variety than in the older individuals of the same kind.
Ordinarily it is not seen, and special stains are required to demonstrate
it clearly. In certain spore-forming bacteria, however, the cell mem-
brane is occasionally seen after the spore has matured within the cell,
as a thin, feebly staining shadow, outlining the original contour of the
organism. Bacteria which plasmolyze easily also show the cell wall
clearly after the cell contents have shrunken away from it.

Capsule. A considerable number of bacteria are surrounded by
mucin-like envelopes, particularly when they are observed in the animal
body or grown in albuminous fluids. This envelope or capsule fre-
quently disappears when the organisms are grown in ordinary media.
This has led to the theory that a capsule represents an hypertrophy
of the ectoplasm. The significance of capsules is still a matter of
controversy. Two principal theories have been advanced to explain
the significance of capsules : according to one theory, bacterial capsules
are purely degenerative phenomena; the more widely accepted theory,
which has much evidence in its favor, maintains that capsule formation
is closely related to the virulence of the organisms. 1 The demonstration
of capsules may be an important factor in the identification of certain
bacteria, for example, the pneumococcus.

Zooglea. A very few bacteria exhibit a slimy intracellular substance
which causes cohesion between considerable numbers of bacterial
cells. This intracellular substance, zooglea, is colored lightly by
ordinary staining methods. It is not found in any of the pathogenic
bacteria.

2. Cell Substance. Cytoplasm. The cytoplasm or endoplasm of
living bacteria (particularly in young cultures) is usually a clear,
colorless, highly refractile, homogeneous appearing substance, although
at times various granules may be seen within it. Vacuoles also are
met with, usually in older bacteria. The cytoplasm usually stains
readily with basic anilin dyes. A few bacteria, notably B. viride and

1 Eisenberg, Centrabl. f. Bakteriol., 1908, xlv, 148.

STRUCTURE AND CONSTITUENTS OF BACTERIAL CELL 27

B. chlorinum, contain a yellowish pigment in the cytoplasm suggesting
chlorophyll, and the so-called purple bacteria similarly possess a
purple colored pigment, bacteriopurpurin.

Nucleus. The occurrence of a demonstrable morphological nucleus
in bacteria is by no means definitely settled: the typical bacterial
cell can not be separated chromoscopically into a nucleus and cyto-
plasm. Those who have thoroughly studied the question by staining
methods, notably Nakanishi, 1 believe that the whole bacterial cell,
as it is ordinarily seen, is potentially a nucleus surrounded by a very
thin film of cytoplasm. Others believe the nucleus substance is dis-
tributed throughout the cell in very finely divided granules : Zettnow 2
is the champion of the latter theory. He believes that the bacterial
cell, as it is viewed following the usual staining processes, is endoplasm
in which the nuclear substance is finely divided and uniformly dis-
tributed. Some observers deny that a nucleus exists at all. Chemical
analyses show beyond doubt that bacteria contain a relatively high
percentage of substances usually regarded as essentially of nuclear
origin. It is quite certain, therefore, that, although there may be no
morphologic nucleus demonstrable by ordinary staining methods,
nuclear material is present in abundance in the organism.

Metachromatic Granules. Certain types of bacteria, notably mem-
bers of the diphtheria and hemorrhagic septicemia groups, exhibit
one or more highly refractile granules in an otherwise homogeneous
endoplasm when they are examined unstained with the higher powers
of the microscope. These granules are few in number in the diphtheria
bacillus group and are distributed somewhat irregularly throughout
the cell, one or more granules usually being greater in diameter than the
cell itself, thus giving the rod a swollen appearance. In the hemor-
rhagic septicemia group these granules are arranged symmetrically,
one at each end of the organism, polar granules. Such granules are
called Ernst-Babes or metachromatic granules. They color differently
from the rest of the cell when they are stained with methylene blue,
appearing as mahogany-red spots in the deep blue endoplasm. They
retain the stain rather tenaciously. Many theories have been advanced
to explain their significance, but nothing definite is known about them,
except that these granules appear to differ widely in chemical composi-
tion. Some are colored brown with iodine, suggesting that they may
be related to glycogen. 3 Some stain black with osmic acid, suggesting

1 Centralbl. f. Bakteriol., 1901, xxx, 97, 145, 193, 225.

2 Ztschr. f. Hyg., 1899, xxx, 1; Festschr. z. 60 Geburtstage vonR. Koch, 1903, p. 383.

3 A. Meyer, Flora, 1899, Ixxxvi, 428.

28 THE MORPHOLOGY OF BACTERIA

that they may be fatty or lipoidal in composition, while others are
probably complex phosphorus-containing compounds. 1 Not all of these
varieties of granules are met with in the same organism. 2 Among
the higher bacteria granules of sulphur or of iron are demonstrable
respectively in the sulphur and the iron bacteria.

Flagella. All minute particles suspended in water or other fluids
of low viscosity are in constant motion. This motion, which is irregular
and tremulous, was first described by Brown: 3 it is variously termed
Brownian movement, pedesis, or molecular movement. Brownian
movement may be rapid or slow, extensive or circumscribed, depending
upon the nature of the particles and the composition and temperature
of the fluid in which they are suspended. This is not true motility,
even though each individual particle moves independently of the other
particles in an irregular orbit, for the particles as a whole do not
permanently change their relative positions. Dead bacteria and many

FIG. 2. Flagella. 1'and 6, peritrichic flagella; 2 and 4, monotrichic flagella; 3 and 5,

lophotrichic flagella.

living bacteria, notably the cocci, exhibit the Brownian movement.
Many bacilli and spirilla, on the contrary, possess the power of
independent motility, that is, they can progressively and permanently
change their relative positions in space. Motile bacteria are provided
with one or more long, delicate, contractile filaments flagella which
are probably the organs of locomotion. These flagella cannot be
demonstrated on living bacteria, except possibly by dark-ground illu-
mination, and ordinary staining reactions usually fail to reveal them.
Special staining methods show them clearly. They appear to arise
from the cell membrane. 4 Their arrangement and number is varied
among bacteria in general, but relatively constant for a particular
variety of bacterium: they are thinner as a rule on younger bacterial
cells, thicker on older organisms. 5 A cholera vibrio has a single

1 Grimme, Centralbl. f. Bakteriol., 1904, xxxvi, 952.

2 For literature see Marx and Woithe, Centralbl. f. Bakteriol., 1900, xxviii, 1, 33, 65,
97; Krompecher, ibid., 1901, xxx, 385, 425; Gauss, ibid., 1902, xxxi, 92.

3 Edinburgh Phil. Jour., 1828, v, 358; 1830, viii, 41.

4 Schaudinn, Arch. f. Protistenk., 1903, i, 421.

6 De Grandi, Centralbl. f. Bakteriol., 1903, xxxiv, 97.

STRUCTURE AND CONSTITUENTS OF BACTERIAL CELL 29

flagellum at one or both ends of the organism; in the typhoid bacillus
they are distributed around the sides of the organism but do not occur
at the ends.

Spores. Endospores. Many bacteria die when their environment
becomes unsuited for further growth. Death may result from the
presence of inimical substances, the absence of essential foods, or the
intervention of unsuitable physical conditions. Death is manifested
by a cessation of chemical interchange between the bacterial cell and
its environment. There is a group of bacteria, however, usually of
saprophytic origin, which is able to survive even prolonged exposure
to unfavorable environmental conditions by passing into a latent stage
during which chemical interchange with the environment is at an
extremely low ebb. This latent stage or hibernation has been known
to last for more than two decades in certain instances, and yet the
organisms have resumed their original luxuriant growth when placed

<*>

FIG. 3. Types of bacterial spores.

under favorable conditions. The bacteria which exhibit this latent
state produce within their substance highly refractile, spherical or oval
bodies called spores. Spores are not found in very young, actively
growing cultures, as a rule. Spore formation is ushered in by a clouding
of the endoplasm of the bacterial cell, which gradually becomes granular.
The granules coalesce, eventually appearing as the mature spore which
is surrounded by a dense membrane, frequently exhibiting a double
contour when stained by dilute carbol fuchsin. 1 The spore membrane
(ectoplasm) is relatively impermeable to heat and disinfectants and
confers the resistance to physical agents which spores exhibit upon
them. But one spore is formed in an individual bacterium, except
under most unusual conditions. It is to be emphasized, consequently,
that spore formation is not a reproductive process. The mature spore
may form in the center of the bacterium, at, or near one end. The
spore may be round or oval, and greater or lesser in diameter than
the parent cell. If the spore is greater in diameter it distends the
cell membrane, producing a spindle-shaped organism if the spore is

1 Meyer, A., Practicum der botanischen Bakterienkunde, Jena, 1903.

30 THE MORPHOLOGY OF BACTERIA

in the center of the rod : if the spore is at one end, a drumstick-shaped
organism results. Usually the size and position of the spore is fairly
constant in a given type of bacteria. Spore formation is most common
among the anaerobes, fairly common among the saprophytic bacteria,
practically absent in the pathogenic bacteria, and practically never
takes place spontaneously in the human or animal body. The spiral
organisms rarely produce spores, and, with the exception of Sarcina
pulmonum, spore formation is practically never observed in the cocci.

It has never been satisfactorily determined whether spore formation
is a regular definite stage in the life history of bacteria which produce
them or whether spores are produced rather under the stress of unfavor-
able evironmental conditions.

Germination of Spores. When bacterial spores are placed in an
environment favorable to the vegetative activity of the cell, they
germinate: the dense membrane which constitutes the ectoplasm of
the spore softens, usually at the pole or the equator, and the vegeta-
tive rod emerges, at first as a small bud, then rapidly assumes the
typical size and shape of the fully mature cell. The development of
the anthrax bacillus from the spore is usually in the line of the longer
axis, polar germination: B. subtilis, on the contrary, usually emerges
at right angles to the larger axis of the spore, equatorial germination.
Many spores are circular in outline, and in such cases the relation of the
developing vegetative cell to the axis of the spore is unknown. Fre-
quently the remnants of the spore membrane remain attached to one
end of the newly formed vegetative cell, appearing as a cap, as it were.
Some spores do not appear to rupture as germination takes place.
the newly forming organism appears to absorb the entire spore and
its ectoplasm, incorporating the entire structure by solution in the
vegetative cell.

Arthrospores. Certain organisms belonging to the coccal group,
more particularly the streptococci, exhibit from time to time cells
which are decidedly larger than their fellows. These cells are more
highly refractile, they usually possess a granular cytoplasm, and fre-
quently stain somewhat irregularly. They have been designated by
Hueppe as arthrospores. These arthrospores appear to have no
unusual resisting powers, and they are in no sense to be regarded as
true spores. It is very probable that they are involution forms.

REPRODUCTION AND CELL DIVISION 31

E. REPRODUCTION AND CELL DIVISION.

Bacteria are structurally the simplest known organisms which main-
tain an independent existence: all their vital functions are exhibited
in a single asexual cell devoid of a morphologically definable nucleus.
The absence of sexual characters and of a morphologic nucleus makes
bacterial reproduction mechanically a simple process, and doubtless
the rapid sequence of generations observed in various bacteria depends
in part upon this simplicity of structure.

Reproduction takes place in the following manner: A bacterial
cell placed in a favorable environment increases in size until it reaches
a maximum which is relatively constant for each variety; then a slight
equatorial constriction occurs, which deepens until a distinct septum
is produced by invagination, which divides the original cell into two
morphologically complete, fully mature individuals of approximately
equal size. It is obvious that this septum consists ordinarily of at
least two layers, since one layer is required to complete each of the
dividing individuals. Successive generations may be produced at
intervals which may be as frequent as every fifteen minutes in the
more rapidly growing types. Septation usually takes place deliberately ;
that is to say, the septum forms relatively slowly. Diptheria bacilli
and possibly related bacteria divide somewhat differently; the parental
cell appears to be under tension when the septum becomes visible,
and the daughter cells spring apart suddenly when septation is com-
pleted. So forcible is this separation that the daughter cells lie at an
angle with each other: Nakanishi 1 has observed that the septum in
this group of organisms frequently forms at a metachromatic granule.
Septation in the Bacillacese and Spirillacese normally takes place
at right angles to the long axis of the organism, and midway between
the ends, thus effecting the separation into two individuals with the
minimal expenditure of material; in the Coccacese, which are usually
isodiametric, no economy of material in septation is apparent, and no
known force determines the initial plane of septation: subsequent
fission may be definitely related to the initial plane. Noguchi 2 has
brought forward striking evidence and photographic illustrations in
favor of the view that the Spirocheta (Treponemata) may reproduce
by longitudinal fission rather than by transverse fission. If this view
be generally adopted, it would contradict the "minimal requirement

1 Centralbl. f. Bakteriol., 1900, xxvii, 641.

2 Jour. Exper. Med., 1912, xv, 201.

32 THE MORPHOLOGY OF BACTERIA

theory," which assumes that transverse fission, the more economical

process both with respect to amount of material and expenditure

of energy, holds universally for bacteria, as has previously been
maintained.

F. CELL GROUPING.

In Bacilli and Spirilla, where septation typically occurs at right
angles to the long axis of the organism, it is obvious that no geometrical
arrangement of cells is possible other than the formation of chains of
rods or of spirals if the individual organisms remain adherent. The
cocci, on the other hand, are spherical and have no longer or shorter
axis, consequently a definite sequence of septation in one, two or
three planes of space can give rise to (1) chains of cocci, if the plane
of septation is always in one plane of space; (2) groups of four cocci,
if septation takes place alternately in two planes of space; or (3) in
packets of cocci, if septation is alternate in three planes of space.
Many cocci do not exhibit a definite sequence of planes of septation.

G. CLASSIFICATION OF BACTERIA.

Relation to Higher Plants. The position of Bacteria in the Plant
Kingdom is indicated in the following table :

Plant Kingdom.

Cryptogamia
Thallophyta

Phanerogamia

Algae

Schizomycetes
(Bacteriacese)

Fungi

Saccharomycetes
Blastomycetes

(yeasts)

i
Lichens
1
Hyphomycetes
(molds)

Eubacteriaceae
Coccacese
Bacillacese
Spirillacese

Chlamydobacteriaceae
Streptothrix
Phragmidothrix
Crenothrix
Cladothrix
Actinomycetes

A complete natural classification of bacteria is impossible at the
present time. The monotony of form observed in this group of
organisms merely suffices to classify them into three great divisions:
Cocci, Bacilli, and Spirilla. Further subdivision into groups which
are potentially families, genera and species is accomplished by arrang-

CLASSIFICATION OF BACTERIA 33

ing them according to their physiological and chemical activities.
Even this artificial procedure is unsatisfactory, for bacteriological
diagnosis is a subject which has developed under the stress of practical
needs, and as bacteria play a part in many fields of activity, it has
inevitably followed that the criteria whereby they are recognized
vary greatly according to the art or science in which they are contem-
plated. Even the same species may be identified by wholly different
characteristics. Notwithstanding the difficulties which surround the
grouping of bacteria, Migula 1 has worked out a system of classification
based upon purely morphological characteristics, which effects a
primary separation of bacteria into smaller subdivisions, which is
moderately satisfactory so far as it goes, and it is the one commonly
adopted.
With certain additions it is as follows:

THE TRUE BACTERIA: EUBACTERIACE.E.

1. Coccacece. Cells in the free state spherical.

(a) Micrococcus. Cells spherical. No definite sequence of planes of septa-

tion.
(6) Diplococcus. Organisms habitually occur in pairs.

chains.

(d) Tetracoccus. Planes of septation alternate, and at right angles in two

planes of space. Form groups of four or tetrads.

(e) Sarcina. Planes of septation alternate, at right angles, in three planes

of space. Form packets.

(/) Planococcus. Motile cocci, provided with flagella.
(g) Planosarcina. Motile sarcina, provided with flagella.

2. Bacillaceoe. Cells elongated and cylindrical; straight.

(a) Bacterium. Non-motile. No flagella.
(6) Bacillus. Cells motile. Peritrichic flagellation.

(c) Pseudomonas. Cells motile. Polar flagellation. Single flagellum or
tufts of flagella at one or both poles of the organism.

3. Spirillacece. Cells elongated and cylindrical; spirally twisted about the long

axis.

(a) Spirasoma. Cells rigid and slightly curved; without flagella.
(6) Microspira. Cells rigid and slightly curved; with one, rarely several,

polar flagella.

(d) Spirocheta. Cells flexuous, closely coiled; flagellation unknown.

THE HIGHER BACTERIA.

4. Chlamydobacteriacece. Cells enclosed in a sheath.

(a) Streptothrix. Cell division always in one plane.

(6) Phragmidothrix. Cell division in three planes of space; very delicate
sheath.

(d) Cladothrix. Cells more or less branched.

5. Beggiatoacece (Thiothrix). Cells contain sulphur granules.

1 System d. Bakterien, Jena, 1907.

34 THE MORPHOLOGY OF BACTERIA

H. MUTATION: CONSTANCY OF TYPES. 1

True mutation or discontinuous variation is rarely observed among
bacteria, although a few instances are on record which have been sub-
jected to satisfactory scrutiny. Mutation must be carefully differen-
tiated from the loss of one or more characteristics of bacteria during
cultivation; the loss or suppression of one or more characteristics is
fairly commonly observed among bacteria. Pigment production, and
proteolytic activity as for example the ability to liquefy gelatin-
are frequently lost to cultures of bacteria during prolonged cultiva-
tion, but these properties may be regained when the organisms are
placed once more in a suitable environment. Similarly, strains of
fermenting bacteria may temporarily, or even permanently, become
unable to decompose certain carbohydrates. Change in virulence,
or loss of virulence is rather commonly noticed among pathogenic
bacteria grown outside the animal body. It is even possible to so
parasitize organisms by prolonged cultivation upon one medium that
they will develop not at all, or slowly at best, on other media. Thus,
a strain of B. proteus has been grown continuously upon agar with
frequent transfers for four years, and the organism will no longer
grow in broth. Similarly, B. bulgaricus is an obligate milk parasite.
Exposure to unfavorable environmental conditions may also suppress
important characters: Pasteur's celebrated experiment of growing
anthrax bacilli at 43 C. for some hours and establishing an asporeless
variety is a familiar example. The suppression of characters as out-
lined above is frequently important as the starting point for new
adjustments between pathogenic bacteria and their hosts.

Turning to the production of disease in man, it is certain that at
least some organisms produce the same reaction today they did years
ago : tuberculosis appears to be the same disease today it was centuries
ago, as is evidenced by the lesions found in Egyptian mummies. Clini-
cally, the observations of Hippocrates would be a fair exposition of the
phenomena seen in tuberculous patients at the present time. Leprosy
also appears to be the same entity now it was during the middle ages,
although the geographical distribution is much more restricted. With
respect to more acute diseases, which require more careful examination
to differentiate them, the evidence is less certain, although typhoid
bacilli do not appear to have changed since they were first isolated

1 Eisenberg, Ueber Mutationen bei Bakterien und anderen Mikroorganismen in
Ergebnisse d. Immunitatsforsch. experimentellen Therapie, Bakteriologie und Hygiene,
Berlin, 1914, pp. 28-142, for summary.

MUTATION: CONSTANCY OF TYPES 35

by Gaffky three decades ago. It appears to be reasonably certain
from what is known of bacteria and the manifestations of disease
they induce that mutation is an infrequent phenomenon: attenuation
and the partial suppresion of characteristics, on the contrary, appear
to be quite common. The available evidence indicates that bacterial
types are stabile under natural conditions : there is no definite evidence
in favor of the view that bacteria change slowly or abruptly either in
their morphology or in the changes they induce in their environment
in the sense that entirely new, unrelated types are developed de now
from preexisting types. This does not preclude the possibility that
such changes have taken place in the past, rather that such changes,
if they have taken place, have not been definitely established.

CHAPTER II.

GENERAL PHYSIOLOGY OF BACTERIA THE EFFECT
OF ENVIRONMENT ON BACTERIA.

A. RATE OF REPRODUCTION.

B. MOTILITY: RATE OF MOTION.

C. SPORULATION: GERMINATION OF

SPORES.

D. LONGEVITY.

E. MOISTURE: DESICCATION.

F. OXYGEN. AEROBIOSIS AND ANAERO-

BIOSIS.

G. TEMPERATURE.

1. General.

2. Cold.

3. Heat.

H. HEAT PRODUCTION.

I. LIGHT AND ELECTRICITY.

J. GRAVITY, OSMOTIC PRESSURE, AGI-
TATION AND CHEMOTAXIS.
K. ENZYMES, TOXINS. PTOMAINS.
L. PIGMENTS.

1. Photodynamic.

2. Phosphorescent.

3. Fluorescent.

4. Chromogenic.

M. SYMBIOSIS, ANTIBIOSIS, COMMENSAL-
ISM.

N. MEDIA COMPOSITION AND REAC-
TION.

O. GROWTH IN ANIMAL BODY.

A. RATE OF REPRODUCTION.

ONE of the striking characteristics of the Bacteriacese is their rapidity
of reproduction. Among the most actively growing types of bacteria,
as, for example, the cholera vibrio, successive generations may appear
at intervals as frequent as every fifteen minutes when the environ-
mental conditions are most favorable : that is to say, ninety-six genera-
tions are theoretically possible in twenty-four hours. If this rate of
reproduction could be maintained for three days, the progeny of a
single organism would occupy a space not less than that of the com-
bined waters of the earth. Fortunately, nature imposes many restraints
which limit the numbers of bacteria. The rapid accumulation of waste
products, the exhaustion of nutrient material, and the enormous
death rate in culture media even after a comparatively few hours'
growth, together with other factors restrict development to such a
degree that the actual number of living descendants of bacteria in
cultures or in nature falls far short of the theoretical number. Many
bacteria develop more slowy than this, however. They may require
hours or even days to arrive at maturity. The tubercle bacillus, for
example, grows comparatively slowy in artificial media (where such
observations are of necessity made), and the frequency of septation,
even in the most rapidly growing bacteria, is greatly affected by
environmental factors.

SPORULATION: GERMINATION OF SPORES 37

Generally speaking, when nutritional conditions are favorable, the
rate of reproduction is influenced by temperature, growth being most
rapid when the temperature is optimum for the organism, less rapid
when the temperature exceeds or falls below this point.

B. MOTILITY: RATE OF MOTION.

The rhythmic contractions of the flagella, with which practically
all motile bacteria are provided, drive the organisms through fluid
media in which they may be suspended, some slowly, some rapidly.
Not all bacteria even in the same culture exhibit motility. The char-
acter of the motion may be direct, serpentine, oscillatory, or irregular.
Rarely, the flagella appear to produce local currents in the medium
which immediately surrounds the organism. Various environmental
factors incite or inhibit motility. Chemotactic substances may attract
bacteria, thus in a sense directing their line of movement. Other
substances, as protoplasmic poisons, paralyze bacterial movements.
Oxygen appears to increase the motility of aerobic bacteria, and it
inhibits motility in the anaerobes. Generally speaking, in favorable
media motility increases with the rise in temperature to the optimum.
If this temperature is exceeded, even by a very few degrees, motion
ceases.

The rate at which bacteria progress through a fluid is a variable
one, although with a given organism under favorable conditions it
appears to be fairly constant. It must be remembered that the
apparent rate of motion observed under the microscope is increased
proportionately to the increase in magnification. Lehmann and
Fried 1 have measured the average speed of certain bacteria in fluid
media in millimeters per second. They find that of the -cholera vibrio
to be 0.03, typhoid bacillus 0.018, B. subtilis 0.01, B. megatherium
0.0075. If a man traveled at a rate of speed in proportion to his size
as great as that of the cholera vibrio, he would average more than a
mile a minute.

C. SPORULATION: GERMINATION OF SPORES.

Many saprophytic bacteria form within themselves spores which
appear apparently under the stimulus of the stress of conditions unfa-
vorable for the continued vegetative growth of the organism. Sporula-
tion, in other words, appears to be a specialized mechanism for the

1 Arch. f. Hyg., 1903, xlvi, 314.

38 GENERAL PHYSIOLOGY OF BACTERIA

perpetuation of the organism during periods of environmental unfit-
ness. Whether spore formation is a definite phase in the life-history of
spore-forming bacteria is not definitely settled. Sporulation is rarely
observed when the temperature of the environment falls much below
15 C., although considerable latitude is observed among the spore-
forming bacteria in this respect. Spores are rarely, if ever, produced
within the tissues of the animal body: if the tissues are exposed to
the air, however, particularly postmortem, spore formation may take
place. No bacteria progressively pathogenic for man are known to
form spores.

The unusual resistance of mature spores to desiccation, to exposure
to dry and moist heat, and to disinfectants may be due either to
their low content of water, for spores contain less than half of the
water contained in the normal vegetative cell, to the relatively thick

FIG. 4. Germination of bacterial spores. 1, by absorption of spore membrane;
2, equatorial germination; 3, polar germination.

refractile spore membrane, or to unusual concentrations of fatty and
lipoidal substances. Experiments by Lewith 1 would suggest that the
relative desiccation of the contents of spores as compared with the mois-
ture content of the vegetative organism would be the most plausible
explanation of their resistance to heat without apparent injury. He
found that egg albumen (dried) suspended in 5 per cent, of water coagu-
lated at 145 C.; suspended in 18 per cent, of water, coagulation took
place at 90 C.; with 25 per cent, of water, at 80 C.; and in a consider-
able volume of water (amount not stated) coagulation occurred when
the temperature reached 56 C.

The resistance of spores to physical conditions varies somewhat
according to the organism in which they are formed. Generally
speaking, however, several minutes' exposure to the temperature
of boiling water (100 C.) may fail to kill them. Dry heat is less
effective than moist heat, for an exposure of 160 C. for one and one-
half hours is required to certainly sterilize glassware containing spores.
Ten to fifteen pounds live steam pressure for fifteen minutes is required

1 Arch. f. exp. Path. u. Pharmakol., 1890, xxvi.

MOISTURE AND DESICCATION 39

to effect sterilization of liquids and organic matter in general. Direct
sunlight will kill spores after days of exposure.

Germination of bacterial spores takes place when they are placed
in a suitable nutritive environment in which the temperature, moisture
and oxygen relations are favorable. The vegetative cell breaks through
the spore membrane apparently after the latter has lost its refractility,
and reproduction by fission proceeds anew, and persists until environ-
mental conditions again lead to sporulation.

D. LONGEVITY.

The duration of life in the individual non-spore-forming bacterium
is unknown, but it is greatest apparently when the organism is quiescent
or nearly so. This condition is realized most commonly when bacteria
are exposed to temperatures slightly above freezing in a dark place.
This question has been studied recently under unusual conditions.
A mastodon was discovered in Siberia which had been uncovered by
an unusual recession of the ice. This animal was found to be practi-
cally intact, and cultures made with proper precautions from the
center of the proboscis contained bacteria indistinguishable from
Sarcina lutea and other well known air organisms. 1 If these cultures
are authentic, a most unexpected instance of bacterial longevity has
been unearthed, for this animal has undoubtedly been frozen for
hundreds of years.

Spores have been dried and kept in a cool dry place for more than
two decades, and yet developed with their usual luxuriance when placed
in a favorable environment. Dried anthrax spores thus retain not only
their viability but their virulence unimpaired for years. Practically,
the. average duration of life among bacteria is comparatively brief.

E. MOISTURE AND DESICCATION.

Bacteria normally contain at least 80 per cent, of moisture in their
substance, and they develop typically only in media containing con-
siderable amounts of moisture. Bacteria do not vegetate normally
in desiccated media, but many varieties resist drying for considerable
periods. Advantage is taken of the restriction of bacterial develop-
ment in the absence of suitable amounts of moisture in various pro-
cesses of drying meats and other foodstuffs; desiccated foods will
keep for weeks under the proper conditions. Bacterial spores pro-

1 Russian Academy of Science, 1911-1912.

40 GENERAL PHYSIOLOGY OF BACTERIA

tected from direct sunlight are extremely resistant to drying, but
they develop with characteristic vigor when environmental conditions
become suitable. Even non-sporogenic bacteria may develop after
days or weeks of desiccation. Many pathogenic bacteria are eliminated
from the body enveloped in albuminous material, as in sputum. These
organisms thus protected may resist drying for many days, provided
they are not exposed to direct light. The following table indicates
the relative viability of various bacteria pathogenic for man to air
drying. 1

1. Gonococcus, few hours.

2. Cholera vibrio, few hours to two days.

3. Plague bacillus, one to eight days.

4. Diphtheria bacillus, twenty to thirty days.

5. Streptococcus pyogenes, fourteen to thirty-six days.

6. Pneumococcus, nineteen to fifty-five days.

7. Staphylococcus pyogenes, fifty-five to one hundred days.

8. Typhoid bacillus, up to seventy days.

9. Tubercle bacillus, two to three months.

F. OXYGEN: AEROBIOSIS AND ANAEROBIOSIS.

Oxygen, either in the free state or combined, is essential to the
growth of all known bacteria. The majority of bacteria grow best in
the presence of free (atmospheric) oxygen, although the percentage
of this gas necessary to support bacterial life may be considerably
less than that occurring normally in the air. Some bacteria appear
to be wholly dependent upon free oxygen, and they are called obligate
aerobes. A small group of bacteria, on the contrary, grow only in the
absence of free oxygen, and more than minimal concentrations of this
gas are actually poisonous to them. Those bacteria which grow only
in the absence of free oxygen are called obligate anaerobes. The vast
majority of bacteria are facultative with respect to their oxygen
requirement, growing best in the presence of atmospheric oxygen but
able to develop either in the presence of small amounts of free oxygen,
as in. the tissue of the body and certain parts of the intestinal tract,
or they are able to obtain their oxygen from chemical compounds, as
certain simple sugars, if free oxygen is not available. These organisms
are called facultative anaerobes. The maximum tolerance of bacteria
for oxygen varies very considerably, as the following table indicates:

Oxygen content of the air is taken as 100 per cent.

1 Fischer, Vorlesungen iiber Bakterien 1903, II Aufl., 110.

TEMPERATURE 41

MAXIMUM OXYGEN TOLERANCE.

Atmospheric oxygen,
Per cent.

B. (clostridium) butyricus ........... 1 . 35

B. chauvei ................ 5.00

B. edematis maligni ........... . . 3.25

Purple bacteria (Molisch) ......... about 90.00

Thiosulphate bacteria (Nathansson) ........ 400.00

B. prodigiosus .............. 3000.00

G. TEMPERATURE.

1. General. The extreme temperature limits of bacterial growth
are very slightly above C. to 80 C. inclusive. Some bacteria,
notably those found in the Arctic regions, appear to develop even at
C.; others, chiefly those found in soil, feces, and certain thermal
springs, grow even at 80 C., a degree of heat considerably above that
at which the protoplasm of most animals and plants is coagulated.
The vast majority of bacteria, however, develop best within a range
of temperature from 15 C. as a minimum to 40-43 C. as a maximum.
All bacteria exhibit three cardinal thermic points : a minimum tempera-
ture, below which growth ceases; an optimum temperature, at which
growth is most luxuriant and rapid; and a maximum temperature,
above which growth ceases, and the organisms die. Fischer 1 has
classified bacteria according to their thermic relations as follows:

Minimum. Optimum. Maximum.

1. Psychrophilic bacteria . 15-20 30 Many water bacteria.

2. Mesophilic bacteria . 15-25 37 43 Pathogenic bacteria and

others.

3. Thermophilic bacteria . 25-45 50-55 85 Spore-forming bacteria

from soil, feces, and
thermal springs.

Bacteria which are progressively pathogenic for man and warm-
blooded animals develop within a much narrower range of tempera-
ture than the saprophytic bacteria which are found chiefly in nature,
as the following table, also taken from Fischer, 2 indicates:

Difference between

minimum and

Minimum.

Optimum.

Maximum.

maximum.

B. phosphorescen

B. fluorescens

. 5

20-25

B. subtilis

. 6

Vibrio cholerae

. 10

B. anthracis .

. 12

B. diphtherise

. 18

33-37

Mic. gonorrhese

. 25

B. tuberculosis

. 30

B. thermophilus .

. 40

1 Vorlesungen iiber Bakterien, 1903, II Aufl.

2 Loc. cit., 106.

42 GENERAL PHYSIOLOGY OF BACTERIA

The saprophytic bacteria, as for example B. subtilis, which develop
through a relatively wide range of temperature are also called
Eurythermic bacteria. The pathogenic bacteria, as for example the
tubercle bacillus, which exhibit but little latitude in this respect, are
called Stenothermic bacteria.

2. Cold. All bacteria grow best and most rapidly in an environ-
ment which is maintained at the optimum temperature for the organism.
If this temperature is lowered even a few degrees, the rate of reproduc-
tion is proportionately reduced. As the temperature approaches C.,
there is complete or nearly complete cessation of growth with a corre-
sponding complete or nearly complete restriction of chemical inter-
change between the organism and its environment. The viability,
and in the pathogenic bacteria the virulence, is not seriously impaired
even by exposure to these low temperatures for considerable periods
of time. Practical advantage is taken of this restriction of bacterial
development by cold in the preservation of food by refrigeration or
by cold storage, and also for the preservation of laboratory cultures
of many non-spore-forming bacteria by placing them in the ice-box
at 5-10 C. So resistant are bacteria to low temperatures that they
may be actually frozen solid and kept in this state for days and even
weeks without killing all the individuals of the cultures. Alternate
freezing and thawing is much more disastrous to them than simple
freezing. Thus, typhoid bacilli may be suspended in water and
exposed to a freezing mixture of ice and salt at 18 C. for several
weeks without killing all the organisms, although the majority of them
are killed within a few hours. At the end of a week fully 90 per cent,
are dead; over 95 per cent, succumb by the end of four weeks' continual
freezing; but from four to six months' continuous freezing is required
to kill all of the typhoid bacilli. The survivors appear to be no more
resistant to subsequent freezing than similar organisms which have
not been frozen. It is a noteworthy fact that bacteria suspended in
colloidal substances, as egg albumen, are much more resistant to freez-
ing than similar organisms frozen in water. Alternate freezing and
thawing in colloids is much less disastrous to bacteria, in other words,
than the same freezing in aqueous solutions. It is probable that the
mechanical factor of crystallization which takes place when water is
frozen actually crushes many of the bacteria, thus accounting, in part
at least, for the greater death rate in aqueous solutions than that
observed in colloids. When bacteria are once frozen, further lowering
of the temperature has surprisingly little influence upon the death
rate. Typhoid and colon bacilli will survive freezing, in moderate

HEAT PRODUCTION 43

numbers at least, in liquid air ( 176 C.) or even liquid hydrogen
(252 C.) for several hours, and develop vigorously when they are
again placed in a suitable environment at the optimum temperature.

3. Heat. Bacteria are distinctly injured by exposure to even slight
increases of temperature above that optimum for their growth, although
there are considerable differences met with among different kinds of
organisms in this respect. Generally speaking, the saprophytic bac-
teria exhibit greater latitude than the pathogenic bacteria. If the
maximum temperature of growth be exceeded by even a very few
degrees, the death of the organisms follows rather promptly. The
greater the degree of heat, the shorter the time required to kill them.
Therefore, the thermal death point of bacteria, that temperature at
which specific organisms die, is dependent not only upon the actual
temperature to which they are exposed, but also to the length of time
of exposure. A standard exposure of ten minutes has been proposed,
so that the thermal death point of the bacterium may be defined as
the lowest temperature to which it must be exposed for ten minutes
under constant conditions to ensure the sterility of the culture. The
determination of the thermal death point is influenced by many factors
besides the kind of organism under observation and the temperature.
Older cultures are usually less resistant than younger cultures of the
same kind. The reaction of the medium (acids particularly decrease
thermal resistance), the presence of extraneous substances as mucin
and other non-conductors of heat, all play a part. Certain modifica-
tions in the characteristics of bacteria are observed when they are
exposed for several hours at the maximum temperature of growth
or a degree or ,two above this point. For example, anthrax bacilli,
which habituallyjorm spores, lose this property when they are exposed
to 44 C. for several hours.

Dry Heat, Moist Heat Dry heat is less effective in killing bacteria
than moist heat. This is shown by the high temperature to which
glassware and other apparatus must be exposed in order to kill spores,
a temperature of 160 C. for one and one-half hours being required
to ensure sterility. Moist heat, which is best obtained by dry steam
under pressure, will kill even the most resistant spores in fifteen
minutes at fifteen pounds pressure.

H. HEAT PRODUCTION. .

The energy liberated by bacteria during the decomposition of organic
substances by bacterial growth is partly utilized by them for their

44 GENERAL PHYSIOLOGY OF BACTERIA

anabolic requirements. A larger part, however, is dissipated as heat.
The heat generated in actively growing cultures of bacteria can be
detected with sensitive thermometers, provided losses due to radiation
and evaporation are guarded against. The heat production is not
great as a rule, although in certain fermentations it may rise as high
as 12-15 above the uninoculated controls. The decomposition of
protein and protein derivatives (putrefaction) usually gives rise to less
heat than the decomposition of carbohydrates (fermentation) under
the same conditions.

I. LIGHT AND ELECTRICITY.

The vast majority of plants possess a photodynamic pigment,
chlorophyll. This pigment can synthesize inorganic substances, as
CO2 and water, together with nitrates, into complex organic compounds
through the energy of the sun's rays acting upon it. Plants possessed
of this pigment, therefore, are the synthetic agents of nature. Usually
this pigment is green; it may, however, be brown or red, the latter
pigment being characteristic of certain algaB. A group of the higher
bacteria, the Rhodobacteriaceae, possess a photodynamic pigment,
bacteriopurpurin, which appears to be analogous to chlorophyll of
the green plants. These sulphur bacteria prefer light and move toward
it. 1 The action of sunlight on this bacteriopurpurin enables them to
decompose CO 2 and to utilize the oxygen thus obtained to oxidize
H 2 S.

All other known bacteria have no photodynamic pigment. Light
is not a source of energy to them, and they are distinctly harmed by
it; they grow best in darkness. Direct daylight kills them rapidly,
and even prolonged exposure to diffuse light may be fatal. Bacteria
are more rapidly killed by exposure to the sun's rays in June, July
and August 2 than exposure of the same time in November, December
and January. Expressed differently, many bacteria which are killed
after an exposure of from one to two hours' direct sunlight in summer
require an exposure of from two to three hours in winter to accomplish
the same result.

Of the spectral rays, the red and infra-red rays, aside from the heating
effect, are without noteworthy action on bacteria. The blue, violet,
and ultraviolet rays, on the contrary, are distinctly bactericidal.

1 Yost, Plant Physiology, 223.

2 In the Northern Hemisphere.

LIGHT AND ELECTRICITY 45

These rays are chemodynamic and it is very probable that the death
of bacteria exposed to them in organic media results from the formation
of H 2 O 2 or other germicidal substances from the substrate. Bacteria
are also killed in non-decomposable media when they are exposed to
the ultraviolet rays. It should be remembered that one of the most
important characteristics of ultra spectral emanations is their very
short wave length. Glass is opaque to them where quartz is trans-
parent.

Electricity. It is difficult to differentiate sharply between purely
electrical effects and chemical changes which are induced in media
of various kinds by the action of electric currents. Generally speaking,
strong electrical currents sterilize media in which bacteria are growing,
but it is by no means certain that the electric current per se is the
important factor. Zeit 1 has made a careful, extensive and accurate
study of the action of various kinds of electric currents on bacterial
growth, and his conclusions are as follows:

"LA continuous current of 260 to 320 milliamperes passed through
bouillon cultures kills bacteria of low thermal death points, in ten
minutes by the production of heat 98.5 C. The antiseptics produced
by electrolysis during this time are not sufficient to prevent growth of
even non-spore-bearing bacteria. The effect is a purely physical one.

" 2. A continuous current of 48 milliamperes passed through bouillon
cultures for from two to three hours does not kill even non-resistant
forms of bacteria. The temperature produced by such a current does
not rise above 37 C. and the electrolytic products are antiseptic but
not germicidal.

" 3. A continuous current of 100 milliamperes passed through bouillon
cultures for seventy-five minutes kills all non-resistant forms of bac-
teria even if the temperature is artificially kept below 37 C. The
effect is due to the formation of germicidal electrolytic products in the
culture. Anthrax spores are killed in two hours. Subtilis spores were
still alive after the current was passed for three hours.

"4. A continuous current passed through bouillon cultures of bacteria
produces a strongly acid reaction at the positive pole, due to the libera-
tion of chlorin which combines with oxygen to form hypochlorous acid.
The strongly alkaline reaction of the bouillon culture at the negative
pole is due to the formation of sodium hydroxid and the liberation of
hydrogen in gas bubbles. With a current of 100 milliamperes for two
hours it required 8.82 milligrams of H 2 SO 4 to neutralize 1 c.c. of the

1 Jour. Am. Med. Assn., November, 1901.

46 GENERAL PHYSIOLOGY OF BACTERIA

culture fluid at the negative pole, and all the most resistant forms of
bacteria were destroyed at the positive pole, including anthrax and
subtilis spores. At the negative pole anthrax spores were killed also,
but subtilis spores remained alive for four hours.

" 5. The continuous current alone, by means of DuBois-Reymond's
method of non-polarizing electrodes and exclusion of chemical effects
by ions in Kruger's sense, is neither bactericidal nor antiseptic. The
apparent antiseptic effect on suspensions of bacteria is due to electric
osmose. The continuous electric current has no bactericidal nor
antiseptic properties, but can destroy bacteria only by its physical
effects heat or chemical effects, the production of bactericidal
substances by electrolysis.

" 6. A magnetic field, either within a helix of wire or between the
poles of a powerful electro-magnet, has no antiseptic or bactericidal
effects whatever.

"7. Alternating currents of a three-inch Ruhmkorff coil passed
through bouillon cultures for ten hours favor growth and pigment
production.

"8. High frequency, high potential currents Tesla currents have
neither antiseptic nor bactericidal properties when passed around a
bacterial suspension within a solenoid. When exposed to the brush
discharges, ozone is produced and kills the bacteria.

"9. Bouillon and hydrocele-fluid cultures in test-tubes of non-
resistant forms of bacteria could not be killed by Rontgen rays after
forty-eight hours' exposure at a distance of 20 mm. from the tube.

" 10. Suspensions of bacteria in agar plates and exposed for four
hours to the rays, according to Rieder's plan, were not killed.

"11. Tubercular sputum exposeu to the Rontgen rays for six hours
at a distance of 20 mm. from the tube, caused acute miliary tubercu-
losis of all the guinea-pigs inoculated with it.

" 12. Rontgen rays have no direct bactericidal properties. The
clinical results must be explained by other factors, possibly the pro-
duction of ozone, hypochlorous acid, extensive necrosis of the deeper
layers of the skin, and phagocytosis."

J. GRAVITY, OSMOTIC PRESSURE, AGITATION, CHEMOTAXIS.

1. Gravity. The majority of bacteria suspended in liquids are
not killed even by four hours' exposure to direct pressure of from 2000
to 3000 atmospheres (one atmosphere of pressure is equal to approxi-

GRAVITY, OSMOTIC PRESSURE, AGITATION, CHEMOTAXIS 47

mately 15 pounds to the square inch, or one kilogram per square
centimeter of surface). Bacteria are weakened, however, by these
great pressures, as is evidenced by a diminution in virulence, decreased
pigment production, and the partial or complete inability to multiply.
It is a curious fact that motile bacteria may retain their motility after
an exposure of several hours to 2000 atmospheres from the pressure
liquids, even although their powers of reproduction are quite lost.

Liquids are practically non-compressible, consequently direct pres-
sure does not affect the volume of the liquid in which bacteria are
suspended, nor does this pressure affect the amount of gas dissolved
in the liquid. If, however, bacteria are exposed in liquids to gas
pressure in the place of direct pressure, the germicidal action of the
gas plays the prominent part in the final result. The amount of gas
dissolved in the liquid increases with increase of pressure, consequently
feebly germicidal gases may become powerfully germicidal as the
pressure is increased. Thus, bacteria suspended in water overlaid
by CO2, which is feebly germicidal at ordinary pressures, are rapidly
killed if the pressure is gradually increased; that is, CO 2 under these
conditions becomes strongly bactericidal. According to Certes, 1
600 atmospheres pressure of an inert gas, as nitrogen, will not kill
anthrax bacilli.

Diminished Pressure. Diminished pressure, aside from lowering
the oxygen tension to a point below that necessary for the growth of
aerobic bacteria, does not interfere seriously with bacterial growth.

2. Osmotic Pressure. The boundary layer, ectoplasm, of every
bacterial cell reacts like a semi-permeable or osmotic membrane.
Through this membrane must pass all the elements necessary to the
nutrition of the organism. A normal bacterial cell always tends to
maintain a greater concentration of solutes within its substance than
exists in the surrounding medium; hence the pressure from within
upon the cell membrane is somewhat greater than the pressure from
without upon the cell membrane, and the cell is consequently in a
state of continual turgor. The osmotic pressure exerted by dissolved
substances varies very greatly. Those of high molecular weight, as
albuminoses or peptones, exert little or no osmotic pressure. Crys-
talloids, on the contrary, may exert very considerable pressure. Thus,
a 30 per cent, solution of dextrose exerts a pressure of about 22 atmos-
pheres. A bacterial cell placed in such a solution is under a great
strain. If bacteria which are in a state of equilibrium with reference

1 Compt. rend. Acad. de sc., 1884, 99, 385.

48 GENERAL PHYSIOLOGY OF BACTERIA

to the osmotic pressure of a solution are suddenly introduced into
media containing a greater concentration of solutes, the contents of
the cell diminish somewhat in amount, due to the rapid withdrawal of
water leaving the rigid cell membrane visible. This shrinkage of the
cell contents is spoken of as plasmolysis. 1 This shrinkage of the cell
contents would indicate that the cell membrane is differentially more
rapidly permeable to water than to crystalloids. All bacteria are not
plasmolyzed when they are suddenly introduced into hypertonic
solutions, and some organisms exhibit the phenomenon of plasmo-
lysis to a much greater extent than others. Plasmolysis does not neces-
sarily result in the death of the organism. It appears to be a fact that
older bacteria are frequently more readily plasmolyzed than younger
individuals of the same kind. The observations of Nicolle and Auclaire 2
would indicate that bacteria which retain the Gram stain are less
readily plasmolyzed than Gram-negative bacteria. Whether Gram-
positive bacteria which have become Gram-negative due to prolonged
cultivation in artificial media invariably follow the same rule is not
known.

If bacteria are gradually subjected to solutions of greater or lesser
osmotic pressure, they usually accommodate themselves to these
changes without visible effect. If bacteria are introduced abruptly
into solutions of low osmotic pressure or distilled water, water rapidly
passes through the cell membrane of the bacteria faster than the solutes
within the cell can pass out, thus rapidly increasing the intracellular
pressure until frequently the cell membrane ruptures, permitting the
escape of some of the cell contents. This phenomenon is called
plasmoptysis. 3 Most bacteria do not plasmoptyze readily, and it is
problematical how much importance should be attached to either
plasmolysis or plasmoptysis in practical bacteriology.

3. Agitation. Bacteria grow best in quiet surroundings, although
a slight amount of agitation is usually harmless and may be even
beneficial if it tends to dislodge waste products from the immediate
surroundings of sedimented organisms. Rapid agitation frequently
retards the multiplication of bacteria in fluid cultures, and Meltzer 4
and Horvath 5 have shown that violent shaking gradually kills bacteria;
not, however, by rupturing the cell membrane. The organisms undergo

1 Fischer, loc. cit., p. 23.

2 Ann. de 1'Inst. Pasteur, 1909, xxiii, 547.

3 Fischer, loc. cit., p. 48.

4 Ztschr. f. Biol., 1894, xxx.
6 Pfliiger's Arch., 1887, xvii.

ENZYMES, TOXINS, PTOMAINS 49

a gradual disintegration, and the injurious effects observed are said
by these observers to be not purely mechanical.

4. Chemotaxis. Bacteria respond to various chemical stimuli.
Substances which can be used by them for nutritional purposes, as
various constituents of laboratory media, appear to attract bacteria.
Harmful substances, as acids or alkalis, may act in the reverse manner.
Oxygen is a powerful chemotactic agent for many aerobic bacteria,
while many anaerobes are repelled by it. The mutual chemotactic
relations of bacteria and leukocytes, and the well-defined tendency
of certain invasive bacteria to localize in definite tissues or organs
of the animal body .are interesting fields for speculation. Nothing
conclusive is known about these relations.

K. ENZYMES, TOXINS, PTOMAINS.

Enzymes. The phenomena of chemical interchange between
bacteria and their environment indicate that enzyme activity plays
an important part in bacterial metabolism.

Enzymes may be defined as substances of unknown composition
produced by living cells which incite specific chemical reactions with-
out permanently combining with the products of reactien. A small
amount of enzyme acting under favorable conditions will cause a
relatively extensive transformation of substance without itself being
used up or inactivated. There is, however, a limit to the amount of
transformation which a given amount of enzyme can accomplish,
for the accumulation of reaction products tends to restrict enzyme
action; the removal of reaction products appears to extend enzyme
action somewhat. All bacterial cells appear to produce or to possess
enzymes, probably several, which may be divided somewhat arbi-
trarily into two classes, the extracellular or exo-enzymes, and the
intracellular or endo-enzymes.

Exo-enzymes. Exo-enzymes are those which are excreted from the
organism and appear as soluble, filterable and, frequently, diffusible
enzymes, which may be obtained in an active state from filtrates of
cultures of bacteria. Their diffusion from the bacterial cell and their
filterability suggests that they may be relatively simple in molecular
aggregation. Their function is essentially a "preparatory" one, for
they transform potential nutritional substances, as proteins, carbo-
hydrates or fats, to simpler compounds which are assimilable by the
bacteria. It is very probable that the exe-enzymes work uneconem-

50 GENERAL PHYSIOLOGY OF BACTERIA

ically in the sense that they transform more material than the
organisms require : this phenomenon is exhibited in the extensive lique-
faction of gelatin by proteolytic bacteria, as B. proteus. The organism^
which elaborates such an exo-enzyme probably derives but little energy
from its activity, and, conversely, probably expends comparatively
little energy in the elaboration and secretion of the exo-enzyme.

Endo-enzymes. Comparatively little is known of the endo-enzymes :
it is -generally believed that they are comparatively non-diffusible, at
least in an active state, and that they are non- or but slightly filter-
able. This suggests that they are relatively complex in their mole-
cular aggregation. Their function is probably to act upon the nutrient
substances which the cell has assimilated, partly to liberate energy
from them, and partly to participate in the organization of the cell
constituents. These endo-enzymes work economically in contra-
distinction to the exo-enzymes in the sense that the substrate is appar-
ently changed by them in proportion to the requirements of the cell.
Endo-enzymes may be obtained from bacterial cells when the latter
disintegrate, provided the rupture of the cells is not accomplished by
violent chemical means. Probably the phenomena of autolysis which
many bacteria exhibit when they are placed in an environment free
from food may be due, in part at least, to the autodigestion of the
organisms by their endo-enzymes.

Classification of Enzymes. Enzymes are usually classified according
to the substrate they act upon : thus, proteolytic enzymes, or proteases,
split proteins or protein derivatives into simpler compounds; carbo-
hydrolytic enzymes split starches or polysaccharides into simpler
carbohydrates; fat-splitting ferments, lipases, split fats into glycerin
and fatty acids. The above enzymes are -hydrolytic in character,
that is, they effect cleavage of protein or carbohydrate or fat or of
glucosides by splitting the molecule into simpler molecules which
simultaneously take up hydrogen and oxygen in the proportions to

form water, thus:

CH 2 NH 2 CO-NHCH 2 COOH + H 2 O (+ enzyme) = CH 2 NH 2 COOH + CH 2 NH 2 COOH.
Glycyl-glycine. Glycine. Glycine.

Ci 2 H 22 On + H 2 O ( + lactase) = C 6 Hi 2 O 6 + C 6 Hi 2 O 6 .
Lactose. Dextrose. Galactose.

CH 2 O-CO-CH 3 CH 2 OH

I I

CHO-CO-CHs + 3 H 2 O (+ lipase) = CHOH + 3 CH 3 COOH

Acetic acid.

CH 2 O T CO 7 CH 3 CH 2 OH

Triacetin, Glycerin.

ENZYMES, TOXINS, PTOMAINS 51

The question of specificity of action of bacterial enzymes is not
definitely settled. There is some evidence in favor of the view that
exo-proteolytic enzymes produced by various bacteria act upon a
variety of proteins: thus, the cholera vibrio produces a soluble pro-
teolytic enzyme which will digest casein, coagulated blood serum, egg
albumen, fibrin and gelatin. Other organisms, as the staphylococcus,
produce an exo-enzyme which will hydrolyze casein, coagulated blood
serum and gelatin: its action upon other proteins is not definitely
established. The important question are the products of hydrolysis
of the same protein by proteolytic enzymes from different bacteria
the same is not definitely settled; it is probable, however, that the
products differ. This suggests that the proteolytic enzymes of bacteria
are not mere "catalyzers" which accelerate reactions in relatively
unstable substances that would take place spontaneously but much
more slowly; these enzymes (proteolytic enzymes) may not only incite
reaction, they may guide it, as it were, along lines of cleavage which
would not be followed in the absence of this enzyme. The carbo-
hydrate and the fat-splitting enzymes have much less latitude in
splitting the carbohydrates and fats respectively than the proteolytic
enzymes, for these substances are less complex in structure and com-
position than the proteins and protein derivatives.

Fuhrmann 1 has classified enzymes of bacterial origin into four types
as follows:

A. SCHIZASES (HYDROLYTIC) CLEAVAGE ENZYMES.

1. Proteases, protein-splitting enzymes. Pepsin, Trypsin (Lysins, Coagu-

2. Carbohydrate-splitting enzymes. Amylase, Cellulase, Pectinase, Gelase,

Invertase, Lactase.

3. Glucoside-splitting enzymes. Emulsin (Synaptase).

4. Fat-splitting enzymes. Lipases (esterases).

B. OXIDIZING ENZYMES.

Tyrosinase, Acetic bacteria, Oxydase.

C. REDUCING ENZYMES.

Reductases.

D. FERMENTATION ENZYMES.

Zymase, Urease, Lactic acid enzyme.

The bacteriolysins are of particular importance in bacteriology:
of the bacteriolysins, those which liberate unchanged hemoglobin
from red blood cells (hemolysins) and those which digest hemoglobin
(hemodigestins 2 ) are intermediary in their general properties between
enzymes and toxins, if indeed there is any tangible distinction between

1 Vorlesungen uber Bakterienenzyme, Jena, 1907.

2 Van Loghem, Centralbl. f. Bakteriol., 1912-1913, Ixvii, 410.

52 GENERAL PHYSIOLOGY OF BACTERIA

them. Vaughan 1 has studied both enzymes and toxins extensively,
and has summarized admirably the points of resemblance between
exo-enzymes and exo-toxins as follows:

"1. Both are destroyed by heat. 2

"2. They act in very dilute solution.

"3. When repeatedly injected into animals in non-fatal doses they
cause the body cells to elaborate antibodies which neutralize the toxin
(or the enzyme) both in viw and in vitro.

"4. In the development of their effects a period of incubation is
required.

"5. It has been shown (by Abderhalden) by optical methods that
they have a cleavage effect upon proteins they split complex proteins
into simpler bodies; in other words, they have a proteolytic action.

"6. They are specific in two senses: (a) they are specific according
to the cell which produces them; (6) they are specific in the antibody
elaborated in the animal body after repeated injections of non-fatal
doses."

Bacterial toxins are usually classified as exo- or soluble (extra-
cellular) toxins, and endo- (intracellular) toxins. The former are
soluble and diffuse out from the bacterial cell into the surrounding
medium. Very few bacteria produce exo-toxins: the best known are
those of the diphtheria, tetanus, and botulismus bacilli. To these
specific antitoxins are known. Endo-toxins are non-diffusible and are
locked up in the bacterial cell; they are liberated only when the
cell disintegrates. No specific antitoxin has been produced for an
endo-toxin.

Ptomains. Ptomains are soluble, basic, nitrogen-containing sub-
stances formed from proteins or protein derivatives by the action of
microorganisms. They are non-specific, relatively poor in oxygen
content, and probably simpler in composition than either exo- or endo-
toxins. No antibodies have been produced against them. Some are

poisonous, many are not.

\
\

L. PIGMENTS.

With the exception of bacteriopurpurin, which occurs in the sulphur
bacteria and is supposed to be photodynamic and, therefore, somewhat
analogous to the chlorophyll of the higher plants, the significance of

1 Protein Split Products, Lea & Febiger, Philadelphia and New York, 1913.

2 Although they are somewhat more resistant to heat than the cells which produce
them.

SYMBIOSIS, ANTIBIOSIS AND COMMENSALISM 53

pigment formation, which is a striking cultural characteristic of many
bacteria, is wholly unknown. The pigment they produce does not
protect them 'against strong Light, and achromogenic strains may be
cultivated from the chromogenic varieties without apparent loss in the
cultural or chemical characters of the organisms. It is very probable
that these pigments are chiefly waste products of metabolic origin.

Pigments are produced in darkness and sunlight rapidly destroys
many of them. Oxygen is not necessary for their production, for the
non-colored leukobase is the form in which the pigment is excreted
by bacteria, but oxygen is necessary for the development of color from
this leukobase.

Pigment-producing bacteria may be grouped into four classes :

1. Bacteria producing photodynamic pigment. Certain sulphur
bacteria which produce bacteriopurpurin.

2. Phosphorogenic bacteria which produce a luminous substance
somewhat analogous to that of glow-worms. These organisms are
chiefly marine forms, as B. phosphorescens.

3. Fluorogenic bacteria which produce a pigment soluble in water
and culture media; this usually exhibits complementary colors as it
is viewed by reflected and transverse light respectively.

4. Chromogenic bacteria. The pigment produced is usually insol-
uble in water and soluble in organic solvents. The color varies accord-
ing to the organism producing it. The more common colors are red,
orange, yellow, green, blue, violet, brown, and black pigment. These
colored pigments are usually referred to as lipochromes because of
their solubility in organic solvents and their general relationship to
fats. Many of them give well-defined and constant absorption when
they are viewed spectroscopically in solutions. 1

M. SYMBIOSIS, ANTIBIOSIS AND COMMENSALISM.

The biological relations of bacteria are of the greatest importance
in the economy of nature and in the production of disease. Bacteria
do not grow in pure culture in nature, although they may do so in the
tissues of man or animals, as disease-producing bacteria (pathogenic
bacteria). In nature, where the reduction of dead complex organic
material to mineralized salts is the striking function of bacteria, the
successive steps in the degradation of organic matter are carried on
by different kinds of microbes. The various steps appear to vary

1 Sullivan, Jour. Med. Research, 1905, xiv, 109.

54 GENERAL PHYSIOLOGY OF BACTERIA

somewhat, but the process is on the whole an orderly and definite one.
The association of various kinds of bacteria in this process, where
each succeeding kind profits by the activities of the preceding kind,*
is a symbiotic one; that is, the several types of organisms mutually
profit by their combined activities.

It frequently happens that the products of symbiotic activity may
be greater than the sum of the products of the separate activities of
the organisms. 1 On the contrary, many instances are known in which
one kind of organism by its activity actually crowds out a preexisting
organism, as for example, the lactic acid bacteria which sour milk.
They produce sufficient lactic acid from the fermentation of the lactose
to kill the proteolytic forms. This substitution of one type of organism
by another is known as antibiosis: the latter organism profits wholly
at the expense of the first organism.

It not infrequently happens that one type of bacterium profits by
the activity of another type of organism without benefiting the former
in return. If two types of bacteria are concerned, the process is known
as metabiosis; if the bacterium is living on a host, the relationship is
spoken of as parasitism.

N. MEDIA COMPOSITION AND REACTION.

Most bacteria grow best in a medium containing a large percentage
of moisture in which diffusible proteins or protein derivatives are
present as sources of nitrogen: these substances are better adapted to
the dietary needs of the majority of bacteria than are ammonium
salts or even simple amino acids. A very few bacteria (nitrifying
bacteria) cannot grow in media containing organic nitrogen compounds :
a few strictly pathogenic bacteria appear to require nitrogen as it
exists in the highly complex tissues of man or animals for their growth.
Many bacteria can utilize carbohydrates for their carbon, hydrogen,
and oxygen requirements. Some bacteria appear to be able to utilize
fats for their carbon requirement.

A neutral or feebly alkaline reaction is best adapted to the develop-
ment of the vast majority of bacteria; a few types develop best in a
medium which is distinctly acid the aciduric bacteria. 2 Mineral
acids are germicidal; organic acids may be utilized by bacteria for
foods.

1 Kendall, Jour. Am. Med. Assn.. 1911, Ivi, 1084.

2 Kendall, Jour. Med. Research, 1910, xxii, 153.

GROWTH OF BACTERIA IN THE ANIMAL BODY 55

O. GROWTH OF BACTERIA IN THE ANIMAL BODY.

The vast majority of bacteria do not grow in the tissues of the body,
although a small number of organisms, the parasitic bacteria, live
habitually on the surface of the body or on mucous membranes, usually
without producing noticeable effects. A small, formidable group of
bacteria, the progressively pathogenic bacteria, actually invade the
tissues; they may produce within the host inhibition of function or
anatomical changes incompatible with health.

CHAPTER III.

THE CHEMISTRY OF BACTERIA. THE EFFECT OF
BACTERIA ON THEIR ENVIRONMENT.

A. GENEEAL.

B. CHEMICAL CONSTITUTION OF BAC-

TERIA.

1. Elementary Composition.

2. Chemical Constitution.

3. Chemical Composition.

C. COMPOSITION OF THE MORPHOLOGI-

CAL COMPONENTS OF THE
BACTERIAL CELL.

1. Cell membrane.

2. Capsule.

3. Cytoplasm.

4. Spores.

D. FOOD RELATIONSHIPS OF BACTERIA.

1. General.

2. Sources of Food.

(a) Nitrogen.
(&) Carbon.

(d) Oxygen.

(e) Inorganic Salts.

A. GENERAL CHEMISTRY OF BACTERIA.

THE practical significance of bacteria is summed up in the nature
and extent of the chemical changes which they induce in their environ-
ment, the result of their multiplication and vegetative activity. These
changes are essentially analytical, for the function of bacteria in
nature is to transform dead organic matter from complex unstable
combinations of carbon, hydrogen, nitrogen, and oxygen, which are
worthjess in the economy of nature, to fully mineralized, stable inor-
ganic compounds of these elements, which may be resynthesized by
plants.

A small but formidable group of bacteria, chiefly those pathogenic
for plants, animals and man, act directly upon the living plant or
animal organism, producing changes in them which may be tempor-
arily incompatible with their well-being, and not infrequently lead to
their death and eventually to their mineralization. The pathogenic
bacteria, therefore, are also analytical in their activities and do not
differ essentially in this respect from the saprophytic types.

It is necessary to consider briefly the method of the interchange
of material between the vegetable and animal kingdoms in order to
understand the full significance of bacterial action in the economy of
nature. All animals require preformed organic compounds for their
sustenance. They are unable to build up these compounds of which
their tissues are composed from chemical elements or from simple
inorganic salts. They are, therefore, dependent directly or indirectly
upon the synthetic activities of green plants for their foodstuffs. The
green plants by virtue of the chlorophyll contained within their leaves

GENERAL CHEMISTRY OF BACTERIA 57

and stems possess the power of combining CO 2 , water and nitrogenous
salts under the influence of sunlight directly into the highly complex
proteins and carbohydrates essential for animal food. These products
of the synthetic activity of the plants are utilized by the animal
kingdom for food; directly by the herbivora, indirectly by the carni-
vora. These substances are either broken down within the digestive
tract of the animal body and reconstructed to form the tissues and
supply energy to the animal, or eliminated as excreta. The excreta
of animals are not sufficiently simple in composition, as a rule, to be
used directly by plants, and the tissues of dead animals and plants
are of little value in their complex state for plant foods. Further
cleavage, both of the excreta of animals and the dead bodies of plants
and animals, is necessary to make the elements contained within them
utilizable by plants, and this cleavage is brought about by bacterial
activity. Various saprophytic bacteria act successively upon these
complex organic compounds, changing them, chiefly by hydrolytic
cleavage, into stable, fully mineralized salts, which are directly utiliz-
able in this state by the chlorophyll-bearing plants. There is, there-
fore, a constant rotation of the various elements which enter into the
composition of animal and plant tissues between the plant and animal
kingdoms respectively by means of an anabolic or constructive process
in the one (plants), and a catabolic or destructive process in the other
(animals). The cycle as outlined, however, is not a continuous one,
for there are important gaps in the process of cleavage and in the pro-
cess of synthesis which if left unbridged by the bacteria would eventu-
ally arrest all vital activity both of plants and animals, and all life
would then inevitably cease on this planet. These gaps between the
animal and vegetable kingdoms are filled by the analytical activity
of bacteria.

A small group of bacteria, on the other hand, is also important
from the synthetical point of view. A certain amount of nitrogen is
lost in the animal and vegetable kingdoms by various natural agencies,
and this supply of nitrogen must be made good from sources which
are not directly available either to plants or to animals. Approxi-
mately 80 per cent, of the atmosphere is made up of nitrogen, and a
certain group of bacteria, "the nitrogen-fixation" bacteria so-called,
which are found chiefly on the nodules or roots of leguminous plants,
are able to draw upon this great reservoir of atmospheric nitrogen
and synthesize it into nitrogen-containing compounds which plants
can utilize directly.

58 THE CHEMISTRY OF BACTERIA

Another type of bacterial activity of importance is the oxidation
of ammonia, the final step in the degradation of protein, into nitrites
and nitrates. This is carried on by the nitrifying bacteria of the soil.
Contrary to the generally accepted idea, therefore, the activities of
the majority of bacteria are not in opposition to the activities of man,
animals, and plants; bacteria are indispensable agents in the economy
of nature.

B. CHEMICAL COMPOSITION OF BACTERIA.

1. Elementary Composition. Bacteria normally contain the same
elements in their substance that the higher plants and animals contain,
viz., carbon, nitrogen, hydrogen, oxygen and phosphorus, together
with smaller amounts of sodium, chlorine, sulphur, potassium, calcium,
magnesium, and traces of iron.

2. Chemical Constitution. The elements carbon, hydrogen, nitro-
gen and oxygen, and to a certain extent phosphorus, and perhaps
sulphur are united to form proteins, nucleoproteins, carbohydrates,
and fats. The inorganic substance of bacteria is made up of the other
elements mentioned above in variable proportions. Of these elements,
carbon, hydrogen, nitrogen, oxygen and phosphorus are the most
important. 1

TABLES ILLUSTRATING THE CHEMICAL COMPOSITION OF
BACTERIA.

1. .PERCENTAGE OF THE ELEMENTS IN ASH-FREE " MYCOPROTEiN." 2

C H N

per cent. per cent. per cent.

52.1-52.6 7.3-7.38 14.5-14.9

2. PERCENTAGE COMPOSITION WITH RESPECT TO ORGANIC AND INORGANIC
CONSTITUENTS.

Putrefactive Bacillus Tubercle

bacteria. 3 prodigiosus. 4 bacilli. 5

Water 83.42 85.45 85.00

Protein 13.96 10.33 8.50

Extractive 1.00 0.70 4.00

Ash 0.78 1.75 1.40

Residue 0.84 1.77 1.10

1 Certain acid-fast bacteria can be grown in media containing theoretically but five
elements: carbon, hydrogen, nitrogen, oxygen, and phosphorus. Lowenstein, Centralbl.
f. BakterioL, Original, 1913, Ixviii, 591. Wherry, Centralbl. f. Bakteriol., 1913, Ixx,
115. Kendall, Day and Walker, Jour. Inf. Dis., 1914, xv, 428.

2 Kruse, Allgemein. Microbiol., p. 62.

3 Nencki and Scheffer, Ueber die chemische Zusammensetzung der Faulnisbakterien,
Beitr. z. Biol. d. Spaltpilze. Nencki, Leipzig, 1880, Jour. f. prakt. Chemie, N. F., xix,
u. xx.

4 Kappes, Analyze d. Massen Kulturen einiger Spaltpilze u. d. Soorhefe, Leipzig, Diss.,
1889.

5 Ruppel, Die Proteine, 1900, Heft 4, Beitr. z. exp. Therapie., Ztschr. f. physiol.
Chemie, xxvi.

CHEMICAL COMPOSITION OF BACTERIA 59

COMPOSITION OF BACTERIA. 1

Water.

In per cent, dry residue.
Acetone CHCla

Phosphorus,
per cent.

per cent.

extract.

extract. 3

in fat. 4

Glanders

. . . 76.5

10.5

11.7

8.6

2.5

Chicken cholera .

. . . 79.3

10.8

7.5

6.3

2.4

Cholera

73 4

9.8

8.7

6.8

2.4

Dysentery (Shiga)

. . . 78.2

8.9

12.8

10.6

1.6

Proteus vulgaris

. . . 80.0

10.7

10.9

7.1

1.6

Typhoid . . .

. . . 78.9

8.3

15.4

10.6

1.2

Anthrax 2

. . , 81.7

9.2

6.3

1.5

0.9

Pseudotuberculosis

. . . 78.8

10.4

15.6

10.3

0.8

B. pneumonias

. . . 85.5

10.4

15.4

10.3

0.8

B. coli ....

. . . 73.3

8.3

15.2

11.8

0.8

B. prodigiosus

. . . 78.0

10.5

9.0

6.6

0.5

B. psittacosis .

. . . 78.0

9.5

11.1

7.0

0.5

B. diphtherias

. . . 84.5

7.0

5.2

0.2

B. pyocyaneus

. . . 75.0

9.8

15.8

10.7

0.2

It will be seen that from 75 to 86 per cent, of the bacterial celt is
water. The remainder of the cell consists chiefly of protein, carbo-
hydrate-like bodies, extractives (fats, fatty acids, waxes and lipoids),
and inorganic salts. Of these, the nitrogenous substances vary greatly
in amount, depending upon the composition of the medium in which
the organisms are grown. _ The extractives (fats, waxes, lipoids, and
fatty acids) are most prominent in the tubercle bacillus and the acid-
fast group. Some extractives, however, are found in all bacteria,
they being greater in amount on a medium containing carbohydrate
and protein than on one containing protein alone. The chemical
determination of the extractives is very unsatisfactory, partly because
of the difficulty in breaking up the cell sufficiently to facilitate the
entrance of the solvent.

3. Chemical Composition of Bacteria. The percentages of the ele-
ments and various constituents of bacteria, as indicated in the above
tables, is at best only approximate. Other factors very markedly
influence the composition of the organisms.

Of these, the age of the culture, the temperature at which it is
grown, and the composition of the medium in which the organisms
are grown are the most important. Generally speaking, young cul-
tures appear to contain rather more dry residue than older cultures,
and bacteria grown at 37 C. contain more dry residue than those
grown at 20 C. 5 The inorganic constituents of the broth influence

1 Nicolle and Alilaire, Ann. 1'Inst. Past., 1909, xxiii, 547.

2 Asporeless. 3 From acetone extract. < From CHCls extract.

5 The decrease in dry residue observed in old cultures is partly attributable to auto-
lysis of bacteria ; this is usually observed earlier in cultures maintained at 37 C. than in
corresponding cultures kept at 20 C. Growth is more rapid at this higher temperature,
and recessive changes due partly to the accumulation of waste products are seen earlier.

THE CHEMISTRY OF BACTERIA

the composition of bacteria markedly. Cramer 1 has found that the
percentage composition of the ash of the cholera vibrio varies within
very considerable limits as the organism is grown under different
conditions. The following table indicates in a general way the influ-
ence of these factors:

=4
II-

9.30

22.30

25.90

1.34

2.75

3.73

1.25

2.50

4.12

28.70

34.80

10.90

7.90
16.90

39.80
7.97

2.10
46.70

23.00

11.40

49.20

Ash content of bacteria in dry substance
Ash content of moist mass ....
Ash content of medium in moist mass
Phosphoric acid in bacterial ash .
Phosphoric acid in media ash
Chlorine in bacterial ash ....
Chlorine in media ash

The phosphorus content of the medium in these experiments, as
shown in the above table, was varied almost twenty times, but in the
bacterial organisms it varied scarcely three times. The variation in
chlorine content was somewhat greater.

Even as important an element as nitrogen is subject to rather wide
variations in bacteria, as Cramer 2 and Lyons 3 have shown. The fol-
lowing tables summarize Cramer's and Lyons's results. They were
obtained by growing certain bacteria mentioned specifically below on
a medium consisting fundamentally of 1.5 per cent, agar, to which were
added various substances, as indicated in the tables, respectively
Media A, B, and C. The general procedure was to grow the bacteria
at 37 C. for several days, to wash them off with salt solution, to free
them from adherent media by centrifugalization and washing, to dry
the washed organisms in vacua to constant weight, and to analyze
the dry residue for extractives and ash.

CRAMER.

Nitrogen substance.

Ether-alcohol
extractives.

Ash.

* !

Organism.

Pfeiff er bacillus . .

66.6

70.0

53.7

17.7

14.63

24.0

12.56

9.10

9.13

Bacillus H-28 4 . .

73.1

79.6

59.0

16.9

17.83

18.4

11.42

7.79

9.20

Pneumonia bacillus .

71.7

79.8

63.6

10.3

11.40

22:7

13.94

10.36

7.88

Rhinoscleroma bacillus

68.4

76.2

62.1

11.1

9.06

20.0

13.45

9.33

9.44

1 Quoted by Kruse, Allgemeine Mikrobiologie, p. 88.

2 Arch. f. Hyg., 1893, 151. 3 Ibid., 1897, xxiii, 30.

4 From water.

COMPOSITION OF THE BACTERIAL CELL

LYONS.

Medium.

Nitrogen-
containing
substance.

Ether
extractives.

Alcohol
extractives.

Ash.

Organism.

Pfeiffer bacillus .

62.75

58.88

45.88

1.68

3.502.67

12.17

17.30

29.60

7.16

2.79

3.09

Bacillus No. 28 1 . .

71.8159.12

46.25

3.32

3.842.84

11.39

15.19

22.78

6.51

3.66

4.18

"Thread bacillus." .

61.06

44.31

33.25

1.74

2.24

1.87

18.40

21.80

27.50

8.09

4.50

3.02

Medium A agar, 1.5 per cent.
Medium B agar, 1.5 per cent.
Medium C agar, 1.5 per cent.

peptone, 1 per cent,
peptone, 5 per cent,
peptone 1 per cent.; dextrose, 5 per cent.

It will be seen that the nitrogen content of the bacteria grown in
a medium containing nitrogen plus carbohydrate is almost 25 per cent,
less than the nitrogen content in the same organisms grown in the same
nitrogen medium but with no carbohydrate. The nitrogen content is
greatest in the carbohydrate-free medium, the extractives are greater
in the carbohydrate-containing medium. This decrease in the nitrogen
content in pathological bacteria grown in sugar media may be of
considerable importance, particularly in the preparation of vaccines
and other antigens. Nothing is known definitely of the distribution
of nitrogen in bacteria, but this reduction of 25 per cent, in the
nitrogen content may well influence somewhat the immunizing value
of vaccines.

C. COMPOSITION OF THE MORPHOLOGICAL COMPONENTS OF
THE BACTERIAL CELL.

1. Cell Membrane. Typical cells of higher plants contain cellulose,
and bacteria were formerly differentiated sharply from the plant
kingdom because cellulose could not be found in them. Later observa-
tions would suggest that cellulose or substances chemically closely
related to it are demonstrable in certain bacteria. Dreyfuss 2 appears
to have identified cellulose in bacteria from pus and in B. subtilis;
Hammerschlag 3 claims to have isolated cellulose from tubercle bacilli,
Dzierzgowski and Rekowski 4 appear to have found cellulose in diph-
theria bacilli; more recently Tamura 5 has demonstrated a hemi-cellulose

1 From water.

2 Ztschr. f. phys. Chemie, 1893, xviii, 375.

3 Sitzber. Akad. Wiss., Wien, xiii, 12.

4 Arch. Soc. Biol., St. Petersburg, 1892.

6 Ztschr. f. phys. Chem., 1914, Ixxxix, 289.

62 THE CHEMISTRY OF BACTERIA

in the same organism. So that the ability of at least certain bacteria
to elaborate cellulose can hardly be doubted.

Emmerling 1 identified chitin in Bacterium xylinum, and Irvanoff 2
gives the following percentage composition of the cell membranes of
B. pyocyaneus, B. megatherium and B. anthracis: C, 46 per cent.;
H, 6.7-7 per cent. ; N, 8.4-8.8 per cent. ; which is empirically very similar
to chitin. Chitin is chemically a polymer of glucoseamine, CH 2 OH.-
(CHOH) 3 .CHNH 2 .CHO, which in turn is an amino hexose very similar
to dextrose, except that it has an amino group adjacent to the aldehyde
gro'up. Chitins are typically animal in origin, and are rarely, if ever,
found in typical plants, hence the distribution between cellulose and
chitin in bacteria is important as suggesting relationships to the
vegetable or animal kingdoms.

Many bacteria stain brown with iodin, and the assumption is that
the cell membrane of such organisms, or the cell substance contains
substances similar to glycogen. According to Arthur Meyer, 3 many
bacteria color blue with very small amounts of iodin; brown or
red-brown with an excess of iodin; indicating that there is a very
small amount of starch and a relatively large amount of glycogen
or amylodextrin in the substance. Similar observations have been
made by Heinze 4 and Levene, 5 who have isolated a substance from
tubercle bacilli which reacts chemically like glycogen.

2. Capsule. The capsules of the capsule-forming bacteria contain
considerable amounts of a mucinous substance apparently a glyco-
protein. Cultures of bacteria which do not ordinarily exhibit capsules
occasionally produce spontaneously viscid, mucinous substances in
artificial media; thus, strains of rabbit septicemia bacilli and glanders
bacilli may become viscid after repeated transfers. 6 Broth cultures
of tubercle bacilli may similarly become mucinous. 7 Rettger's observa-
tions 8 make it very probable that these viscid substances are true
mucins.

3. Cytoplasm. The cytoplasm of bacteria consists chiefly of the
bacterial protein, which appears to be specific in character for any

1 Berichte d. chem. Gesell., 1899, 541.

2 Hofmeister's Beitrage, 1902, i, 524.

3 Flora, 1899.

4 Centralbl. f. Bakteriol., 2te Abt., 1903, xii; 1904, xiv.
6 Jour. Med. Research, 1901, vi, 135.

6 Theobald Smith, Transactions of First Annual Meeting of National Association
for the Study and Prevention of Tuberculosis.

7 Weleminsky, Berl. klin. Wchnschr., 1912, xlix, 1320; Kendall, Walker and Day,
Jour. Infec. Dis., 1914, No. 11.

8 Jour. Med. Research, 1903, x, 101.

COMPOSITION OF THE BACTERIAL CELL 63

given organism, together with enzymes and at least minimal quantities
of all the products of its metabolism.

Regarding the nature of the protein substance in bacteria, but little
is known, although 50-80 per cent, of the dried substance of the
bacterial cell consists of protein and protein derivatives. Conspicuous
among these protein derivatives are the nuclein constituents, nucleins,
nucleoproteins, and nucleic acids; they occur constantly in bacteria
and apparently the greater part of the protein of the bacterial cell
consists of these nuclear constituents. Nucleins and nucleoproteins
have been isolated from many bacteria: from B. subtilis by Van de
Velde; 1 from the plague bacillus by Lustig and Galeotti; 2 from the
typhoid bacillus by Paladino-Blandini; 3 from the tubercle bacillus
by Von Ruck 4 and Ruppel; 5 from the diphtheria bacillus by Aronson; 6
and Carapelle 7 has identified a glyco-nucleo-protein in B. prodigiosus.

Numerous observations indicate that nuclein bases (xanthin bases)
are found in bacterial cells; thus, Lustig and Galeotti 8 identified
xanthin in plague bacilli. Nashimura 9 obtained xanthin bases in the
dried residue of a water bacillus in the following amounts: xanthin 0.07
per cent.; guanin, 0.14 per cent.; adenin, 0.08 per cent. No hypox-
anthin was found.

The amino-acids of bacterial protein have not been thoroughly
studied. The variable nitrogen content even of the same organism
as it is grown in different media and under different conditions would
suggest that quantitative determinations of nitrogenous substances
would be somewhat unsatisfactory. Qualitatively, so far as available
data show, many amino-acids found in protein of higher animals and
plants have been isolated or identified in bacterial cells. These amino-
acids appear to differ in amount in different organisms, and several
have not been isolated at all up to the present time. Vaughan, Wheeler,
and Leach 10 conclude that the bacterial substance contains carbo-
hydrates, nuclein bodies and polymers of mono- and diamino-acids.
They are glyco-nucleo-proteins. Kruse 11 and Vaughan 12 have arrived at

1 Ztschr. f. phys. Chem., viii.

2 Deutsch. med. Wchnschr., 1897, 225.

5 Baumgarten's Jahresberichte, 1901, 228, ref.

4 Prophylactic Immunization against Tuberculosis, Report No. 1, Asheville", 1912, 3.

6 Ztschr. f. phys. Chem., 1898, xxvi.

6 Arch. f. Kinderheilkunde, vol. xxx.

7 Centralbl. f. Bakteriol., 1907, xliv, 440.

8 Loc. cit.

9 Arch. f. Hyg., xviii, 325.

10 Tr. Assn. Am. Phys., 1902, p. 243.

11 Allgemeine Microbiologie, p. 65.

12 Protein Split Products, p. 437.

64 THE CHEMISTRY OF BACTERIA

the same conclusion. The analysis of one hundred grams of dried
tubercle bacilli by Ruppel 1 indicates the importance of the nucleins in
bacterial proteins.

Grams.

Nucleic acid (tuberculinic acid) 8.5

Nucleoprotamin 25.5

Nucleoproteid 23.0

Albuminoids (keratin, etc.) 8.3

Fat and wax 26 . 5

Ash 9.2

Carbohydrates. Glycogen or some similar carbohydrate, which is
readily detected by the mahogany color it gives with iodine, is found
in many bacteria, as has been stated previously, but it is extremely
difficult to decide definitely whether it is limited exclusively to the cell
membrane or scattered somewhat diffusely through the cytoplasm
as well.

Fats and Fatty Derivatives. Fats, fatty acids, lipoids and waxes,
which may be demonstrated by staining bacteria with Sudan III,
Scharlach R, and osmic acid, occur in variable amounts in the tubercle
bacillus and other acid-fast bacilli. The amount of these extractives
may be very great in the acid-fast group, varying from 26 to 40 per cent,
of the total dry residue. Considerable discussion has centred around
the distribution of these substances, many authorities claiming that
the fats and waxes are contained in the cell wall of the organism, while
others maintain that these substances are scattered throughout the
cell substance as well. In the acid-fast bacilli it is probable that these
fats are both intra- and extracellular, for analyses show that a certain
amount of them can be extracted from intact bacilli, while still more
can be extracted when the organisms are broken up. The following
table from Kresling 2 illustrates the distribution of the fatty substance
of the tubercle bacillus:

I. CONTENTS OF THE DRIED TUBERCLE BACILLI IN THE
PREPARATION OF TUBERCULIN.

Per cent.

Moisture (dried at 100 -l 10 C.) 3.9375

Moisture (dried in desiccator) 3 . 08

Ash 2.55

Nitrogen - 8.575

Nitrogen-containing substances (albumin) reckoned by multiply-
ing the amount of N by the factor 6.25 (the N of lecithin and
other substances soluble in chloroform, benzol, ether, and

alcohol were not reckoned) 53 . 59

Fatty substances in medium after the first four determinations 38.95
Other N-free substances, reckoned as the difference . . . . . 9725

1 Loc. cit.

2 Centralbl. f. Bakteriol., 1901, xxx, 909,

FOOD RELATIONSHIPS OF BACTERIA 65

II. FATTY SUBSTANCE OBTAINED BY EXTRACTION WITH

CHLOROFORM, POSSESSES THE FOLLOWING

CHARACTERISTICS :

Melting point 46 C.

Acid number ' 23.08

Reichert-Meissl number 2.007

Hehner number 74 . 236

Saponification number 60.70

Ether number 36 . 62

Iodine number (according to Hubl) 9 . 92

III. THE FATTY SUBSTANCE OBTAINED BY EXTRACTION WITH
CHLOROFORM CONTAINS:

Per cent.

Free fatty acids 14.38

Neutral fats and esters of fatty acids 77.25

Alcohols separated from the fatty acid esters (with melting point

43.5-44 C.) 39.10

Lecithin 0.16

Cholesterin Not determined

Substances directly soluble in water 0.73

Substances soluble in water which are formed by the complete

saponification of the fatty substances 25 . 764

Inorganic Constituents. The most conspicuous inorganic element
found in the ash of bacteria is phosphorus, and the content of phos-
phorus, recovered as phosphoric acid, frequently reaches as high as
half the total ash weight. It is probable that a considerable part of
this phosphorus is combined with nucleic acid to form nucleo-protein.

4. Spores. The chemical composition of spores is not well deter-
mined, but the generally accepted theory is that they contain relatively
less water and consequently a greater proportion of proteins and ash.
Reinke 1 has suggested that the sporoplasm is an anhydride of the
cytoplasm of the vegetative cell. Sporulation implies that relatively
considerable amounts of water must be taken up by the spore sub-
stance in order to regain the proportion of this substance found in the
parent organism.

D. FOOD RELATIONSHIPS OF BACTERIA.

1. General. Food is any substance which a living organism may
utilize, either by making it a part of its living material or as a source
of energy. Food which is suitable for utilization by any organism must
contain all the elements necessary for the building up and maintenance
of that organism. Analyses of bacterial cells, which have been given
in preceding tables, show them to be made up of the same elements
as those of the higher plants and animals; viz., carbon, hydrogen,
oxygen, nitrogen, and phosphorus, together with smaller amounts

1 Quoted by Kruse, Allgem. MikrobioL, p. 57,

66 THE CHEMISTRY OF BACTERIA

of sodium, potassium, sulphur, calcium, and magnesjum. Foods to
be fully suitable for bacterial needs, therefore, should contain these
elements. It should be stated, however, that the food requirements
of bacteria vary within wide limits, but the above statements are
generally applicable.

2. Sources of Food. (a) Nitrogen. The nature of the compounds
in which nitrogen must be presented to bacteria as food varies greatly
among the different groups. The nodule bacteria found in the nodules
on the roots of many leguminous plants actually utilize atmospheric
nitrogen: nitrifying bacteria found chiefly in the soil derive their
nitrogen requirement chiefly from mineral salts which are oxidized
through their activities to nitrites and eventually to nitrates. From
this very simple source of nitrogen these bacteria are able to synthesize
the complex nitrogen-containing proteins of their bodies.

The majority of bacteria, including not only the saprophytic organ-
isms but most of those pathogenic for man, animals, and plants as
well, thrive in media in which nitrogen is presented to them as peptones,
albumoses, or even certain amino-acids; in other words, upon the pro-
ducts of protein digestion. The more strictly pathogenic organisms,
as the gonococcus, may require nitrogen in the form of highly specific
tissue proteins. Generally speaking, animal protein or its derivatives
is more easily utilized by bacteria than protein of vegetable origin.

(6) Carbon. The simplest carbon compound which occurs naturally,
CO 2 , cannot be used by bacteria, except certain nitrifying bacteria,
as a source of energy, for it is already fully oxidized. The carbon
of proteins and their derivatives, of carbohydrates, and of fats, on the
contrary, is readily utilizable by most bacteria. As a rule, hydro-
carbons of the aliphatic series are not attacked by the microorganisms,
but compounds containing oxygen as well as carbon and hydrogen are
better adapted for microbial food. Organic acids, as acetic acid,
aspartic, tartaric, and many oxy acids are utilizable by some bacteria.
The simpler alcohols can be used, but by very few bacteria. The
complex alcohols, like glycerin and mannite, on the other hand, are
available food materials for many.

The best nitrogen-free food compounds for microorganisms are the
carbohydrates, particularly those containing six and twelve carbon
atoms, the hexoses and bioses respectively. Carbohydrates containing
four, five, or any number of carbon atoms not a multiple of three are
usually not readily attacked by bacteria. Starches and cellulose are
not generally utilizable, although certain types of organisms, notably

FOOD RELATIONSHIPS OF BACTERIA 67

those found in the intestinal tracts of herbivora, appear to decompose
them very readily.

(c) Hydrogen. Hydrogen is readily obtained by microorganisms
from organic compounds containing available carbon, nitrogen, and
hydrogen, but not apparently from water.

(d) Oxygen. Oxygen is indispensable to the life of all living organ-
isms as a source of energy and for structural purposes. A few bacteria,
the obligately aerobic bacteria, can live only in the presence of free
oxygen; another small group, the obligately anaerobic bacteria, live
either in the absence of free oxygen or at best in the presence of minimal
amounts of it; more than minimal amounts of free oxygen act as
specific poisons to them. The majority of bacteria are facultative with
respect to their oxygen requirements; that is, they can either live
in the presence of free oxygen or derive their oxygen needs from
organic compounds, usually the carbohydrates or proteins.

(e) Inorganic Salts. Inorganic salts are used by bacteria almost
wholly for structural purposes. The requirement for mineral com-
pounds is very little, for these substances do not on the average make
up more than 7 to 10 per cent, of the solid matter of the bacterial cell.
The essential elements and the percentage of them found in the ash
of certain bacteria have been referred to previously, and it was stated
that the amount of inorganic salts found in the bodies of the bacteria
bore a rather direct relationship to the salt concentration of the media.
Of the inorganic elements, phosphorus is the most important, for it
makes up nearly 50 per cent, of the ash. Phosphorous in contra-
distinction to any other inorganic salt is absolutely indispensable to
bacterial growth. It is combined organically in nucleo-proteins,
glyconucleo-proteins, and nucleic acids, which form the greater part
of the protein of the bacterial cell.

CHAPTER IV.

BACTERIAL METABOLISM.

I. GENERAL.

II. THE NATURE OF BACTERIAL MET-
ABOLISM.

III. NITROGEN METABOLISM.

IV. CARBON METABOLISM.

V. QUALITATIVE CATABOLIC REAC-
TIONS OF BACTERIA.

A. In Media Containing Only

Utilizable Nitrogenous
Substances.

B. In Media Containing Both

Utilizable Nitrogenous
Substances and Utilizable
Carbohydrates .

VI. THE QUALITATIVE INFLUENCE OF
UTILIZABLE CARBOHYDRATES
UPON THE ELABORATION OF
PROTEOLYTIC ENZYMES.
VII. QUANTITATIVE MEASURE OF BAC-
TERIAL METABOLISM.
VIII. THE SIGNIFICANCE OF BACTERIAL
METABOLISM.

IX. FERMENTATION AND PUTREFAC-
TION.

I. GENERAL BACTERIAL METABOLISM.

Two distinct phases may be recognized in the life-history of a
bacterial cell; an anabolic or constructive phase, during which the
cell becomes morphologically complete; and a catabolic, vegetative,
or fuel phase, in which the mature organism reacts chemically upon
its environment to provide the energy (fuel) necessary for the main-
tenance of the cell. Chronologically, the anabolic phase precedes the
catabolic phase; that is to say, the bacterial cell must be morpho-
logically complete before it can bring about its characteristic energy
transformations; practically the two phases overlap somewhat.

The actual amount of material required for the anabolic phase of the
bacterial cell is very small, for the actual weight of the average
bacterium is but 0.000,000,0016 of a milligram, approximately (see
page 25). The structural phase is practically ended, aside from the
replacement of comparatively slight losses of substance incidental
to the elaboration of soluble enzymes or to additional requirements
for the formation of structural elements, such as capsules, when the
organism is morphologically complete. The waste incidental to the
utilization of material for purely anabolic needs is likewise very small
in amount, and the total environmental change attributable to the
purely constructive phase of bacterial metabolism is slight andordin-
arily disregarded. 1

Kendall, Jour. Med. Res., 1911, N. S. f xx, 140.

GENERAL BACTERIAL METABOLISM 69

The amount of material required for the catabolic (vegetative or
fuel) phase of the bacterial cell, on the contrary, is relatively large.
The energy requirement of cellular organisms varies rather with the
area of their surface than according to their actual volume; conse-
quently, very minute organisms, as bacteria, in which the surface is
relatively very great in comparison with their size, would require much
more material for energy purposes than for structural purposes. For
example, the total surface area of a million average-sized cocci (each
1 micron in diameter) would be approximately 3.1416 sq. mm.; the
weight of these organisms, assuming the specific gravity to be 1.030
(which is reasonably accurate), would be about 0.00054 mg. The
combined surface of all the cocci in an actively growing broth culture
of such organisms would be very considerable. It must be remembered,
however, that these figures do not carry any specific basis for the
measurement of bacterial activity in terms of chemical or physical
phenomena; they merely express in a very general manner the physical
basis for the apparent disproportion observed between the size of
bacteria and the amount of change they induce in their environment.

The energy phase commences theoretically when the cell is morpho-
logically complete, and it is a continuous process which ends only with
the death of the cell. It may be reduced to a minimum when the cell
enters upon a latent state of existence, as in spore formation; it is
greatest when the organism is growing in a favorable medium at the
optimum temperature, and it is restricted proportionately when
environmental conditions become unfavorable.

The life-history of a culture in which innumerable bacteria are
growing can not be sharply divided into the anabolic and catabolic
phases. During the first few hours after inoculation, however, the
anabolic aspect predominates; later the catabolic aspect predominates.
Thus, colon bacilli inoculated into dextrose broth fermentation tubes
do not produce gas in visible amounts during the first few hours of
incubation, although the medium gradually becomes turbid, due to the
rapid multiplication of bacteria. Somewhat later gas formation is
observed, and it then proceeds with considerable rapidity. The
production of gas is indicative of a period of great vegetative activity
in which large numbers of mature colon bacilli utilize the dextrose for
their energy requirements. Still later the production of gas ceases,
the activities of the organisms diminish, and the culture finally dies
out as waste products accumulate in sufficient amounts.

Those bacteria habitually pathogenic for man induce less striking
physical and chemical changes in their environment, as a rule, than

70 BACTERIAL METABOLISM

do the saprophytic types, as Theobald Smith 1 showed long ago. Thus,
typhoid bacilli are relatively inert culturally; they form no gas in
sugar media, no indol, and do not liquefy gelatin; on the contrary,
B. coli and even more strikingly B. proteus are characterized by strik-
ing cultural changes; B. coli produces deep-seated changes in protein,
resulting in the production of indol; it produces gas from sugar media,
but it does not liquefy gelatin. B. proteus behaves much like B. coli
in sugar media, but liquefies gelatin as well. These marked changes
in the composition of the medium, namely, the production of indol
from protein, the production of gas from sugar, and the liquefaction
of gelatin, are all phenomena associated with the vegetative or fuel
phase of bacteria.

H. THE NATURE OF BACTERIAL METABOLISM.

Chemically considered, the anabolic phase of bacterial activity is
one characterized by the synthesis of relatively simple substances,
chiefly nitrogen-containing, into the complex specific bacterial proto-
plasm through a series of synthetic reactions among which reductions
and condensations appear to be the more prominent. It is very
probable that many of these condensation reactions are hydrogenic
in nature; that is, two simpler molecules are united into one molecule
of greater complexity through the removal of hydrogen and oxygen
from them in the proportions to form water.

As simple illustrations: the formation of lactose from a molecule
each of dextrose and galactose,

C 6 Hi 2 O 6 + C 6 Hi 2 O 6 = Ci2H 22 Oii + H 2 O
Dextrose. Galactose. Lactose.

the formation of a polypeptid, glycyl-glycin, from two molecules of
glycocoll, 2

NH2.CH 2 .COOH + H.NH.CH 2 .COOH = NH 2 .CH 2 .CO.NH.CH 2 .COOH + H 2 O
Glycocoll. Glycocoll. Glycyl-glycin.

and the formation of the glyceride of a fatty acid from glycerin and
acetic acid may be cited,

CH 2 .OH + HOOC.CHa = CH 2 .O.O.CH 3

CH.OH + HOOC.CHs = CH.O.O.CHs + 3 H 2 O
I I

CH 2 .OH + HOOC.CHs CH 2 O.O.CH 3
Glycerin. Acetic acid. Triacetin.

1 Fermentation Tube, Wilder Quarter Century Book, 1893, p. 219. (See also Kendall,
Day and Walker, Jour. Am. Chem. Assn., 1913, xxxv, 1201-1249, for analytical data.)
Fischer, Ber. d. deutsch. chem. Gesell., 1906, xxxix, 530.

NITROGEN METABOLISM 71

The catabolic phase is essentially analytic; it is characterized
chemically by a series of reactions in which the cleavage of more
complex compounds to simpler ones with their simultaneous or sub-
sequent oxidation, involving the liberation of energy, is a noteworthy
feature. The catabolic phase is chiefly a series of oxidations of carbon
and hydrogen. (For illustrative catabolic reactions see infra, pp. 73, 76.)

m. NITROGEN METABOLISM.

Bacteria, like all known living things, contain nitrogen in their
substance, and nitrogen in some form is absolutely indispensable for
the building up of their structure. Nitrogen, in other words, is an
absolutely essential element in the constructive phase of the bacterial
cell. The form in which nitrogen must be presented to bacteria in
order to be utilizable by them varies with the kind of organism. The
nitrogen-fixing bacteria found on the roots of leguminous plants can
utilize the nitrogen of the atmosphere; some nitrifying bacteria can
utilize the nitrogen of ammonium salts. (These two groups of organ-
isms appear to be the only ones which can oxidize nitrogen.) Many
bacteria can obtain their nitrogen from amino-acids. The majority 1
of bacteria pathogenic for man and the higher animals are somewhat
more exacting in this respect and require more highly organized
nitrogen, as peptones and proteoses, while a small group of obligately
human pathogenic bacteria, as the gonococcus, grows only in media
containing nitrogen as it exists in the highly specialized protein of
human origin, at least during their first growth outside the human
body on artificial media.

he vegetative phase of bacterial metabolism is essentially a series
of oxidations of carbon and hydrogen; nitrogen can not be oxidized by
the great majority of bacteria, and consequently it appears to yield
little or no energy to them. When nitrogen-containing compounds
as amino-acids, peptones, albumoses, or proteins are utilized for the
energy requirements of these organisms, the nitrogen (amino nitrogen)
is usually eliminated from the ^mino-acid complex incidental to the
oxidation of the carbon and hydrogen; the nitrogen thus eliminated
appears in soluble form in the culture medium as ammonia. This
process is true deaminization. . Nitrates and even nitrites may be
sources of energy to many bacteria, usually, however, because of their
valuable oxygen content. To summarize, bacteria must have available
nitrogen for their structural needs,, but nitrogen, except for the nitrogen-
fixing and nitrifying bacteria, is not as a rule a source of energy to
them, because the great majority of bacteria can not oxidize it.

72 BACTERIAL METABOLISM

IV. CARBON METABOLISM.

Carbon is an important structural element for bacteria, and it is
equally indispensable as a source of energy, for the oxidation of carbon
is an important feature of the catabolic activity of the majority of
microorganisms. The reduced form in which this element is present
in amino-acids and other protein derivatives appears to be particularly
adapted for structural purposes; for fuel purposes it is less available,
possibly because of the necessity of introducing free oxygen into the
carbon complex to provide the requisite energy for the vegetative
activities of bacteria, as well as the additional amount of work required
to eliminate the nitrogen of the amino-acid molecule fdeaminization).
It is generally stated that bacteria with relatively few exceptions
fail to grow with their customary vigor in sugar-free media from which
free (atmospheric) oxygen is excluded; the relative absence of available
oxygen in such compounds would explain this phenomenon, in part
at least.

The carbohydrate molecule, which contains no nitrogen and in which
the carbon is already partially oxidized, can be utilized for fuel purposes
by most bacteria with less expenditure of energy for its preparation
than can be the case with most amino-acids, peptones, or proteins;
for this reason it is very probable that utilizable carbohydrate is
act6d upon by many bacteria in preference to protein carbon. In
this sense utilizable carbohydrate protects or shields protein or protein
derivatives from bacterial attack for their fuel requirements; it does
not protect protein from bacterial breakdown to supply their structural
requirements, however.

The net result of this selective protective action of carbohydrates
for protein is important because the amount of material required to
provkL* energy for the bacterial cell far exceeds the amount of material
required to build up the bacterial cell. The chemical transformations
incidental to the anabolic phase of bacterial metabolism are insignificant
in amount and ordinarily not noticeable; on the contrary, the chemical
transformations associated with the catabolic phase of bacterial
metabolism are relatively very considerable in amount; and the
nature and extent of those chemical reactions which are associated
with the transformation of material for energy are important not
only for the identification of bacteria, they collectively comprise the
important specific function of bacteria.

QUALITATIVE CATABOLIC REACTIONS OF BACTERIA 73

V. QUALITATIVE CATABOLIC REACTIONS OF BACTERIA.

The chemical changes observed in cultures of ordinary bacteria are
chiefly those associated with the breakdown of organic substances for
energy they are reactions of the catabolic phase of bacterial meta-
bolism. It should be again emphasized that the energy reactions the
catabolic reactions are those which are most profoundly influenced
by the composition of the nutritive substrate upon which the organisms
are grown.

A. Reactions of Bacteria in Media Containing Only Nitrogenous
Substances (Proteins or Protein Derivatives) Which are Utilized
for the Energy Requirements of Bacteria. Proteins are composed
of amino-acids, of which some seventeen are recognized. Bacteria
which decompose protein appear to act upon these amino-acids in the
last analysis, and several types of reaction are recognized at the present
time. Each kind of organism utilizes protein or protein derivatives
somewhat differently and characteristically, but in general one or more
of the following types of reactions are involved either successively or
simultaneously in the catabolism of these substances. The reactions
follow i 1

1 . R.CH 2 .CHNH 2 .COOH + H 2 = R.CH 2 .CH 2 .COQH 4- NH* Re-
ductive deaminization of amino-acid to fatty acid with the same
number of carbon atoms.

2. R.CH 2 .CHNH 2 .COOH + H 2 O = R.CH 2 .CHOH.COOH + NH 3v
Hydrolytic deaminization of amino-acid to oxy-acid with the same
number of carbon atoms.

3. R.CH 2 .CHNH 2 .COOH + O 2 = R.CH 2 .CO.COOH + NH 3 . Oxi-
dative deaminization of amino-acid to keto-acid with same number of
carbon atoms.

4. R.CH 2 .CHNH 2 .COOH-R.CH 2 .CH 2 .HN 2 + CO 2 . Carooxylic
decomposition of amino-acid to amine with one less carbon atom.

5. R.CH 2 .CH 2 .COOH -* R.CH 2 .CH 3 . + CO 2 . Carboxylic decom-
position of fatty acid.

6. R.CH 2 .CH 2 .COOH + 3O = CH 2 .COOH + CO 2 + H 2 O. Carbo-
xylic decomposition with the formation of a fatty acid with one less
carbon atom.

A few illustrations will indicate the nature of these changes in amino-
acids with the production of certain substances of clinical interest :
1. Formation of indol from tryptophan. Indol is a substance pro-

* See Kruse, Allgem. Mikrobiol., 505-536, for literature.

BACTERIAL METABOLISM

duced in the intestinal tract from tryptophan (an amino-acid found
in protein), chiefly by B. coli and B. proteus. The reactions through
which tryptophan is changed to indol by these organisms are as
follows. 1

\ CH 2 .CHNH 2 .COOH

+ H 2 =

Tryptophan.

CH 2 .CH 2 .COOH

V NH

(deaininization)
Indol propionic acid.

+ NHs Indol propionic acid
+ 3O =

\ _ CHz.COOH

+ CO 2

CHs

H 2 O->

Indol acetic acid.

C0 2

Skatol + 3 O

Skatol.

CO 2 + H 2 O

Indol.

Indol contains little or no energy for most bacteria, and it is left
as such in the culture medium or the intestinal tract. Indol is fre-
quently absorbed from the intestinal tract, but it has little or no
energy for the human body it is oxidized in the liver to indoxyl

and is excreted and appears
in the urine as indican

o
/

O - S - ONa

II
Q

B. coli, B. proteus, and other organisms which "form indol" utilize
the alanin radical of the tryptophan molecule (alpha amino propionic
acid) for energy, first eliminating the nitrogen (deaininization), then
oxidizing the carbon. The indol radical which is left is not a source
of energy; it can not be oxidized by these organisms, consequently
it' remains as such in culture media or is absorbed from the intestinal
tract.

Nencki, Sitzungsber. Wien. Akad., 1898, II Abt., xcviii, 412.

QUALITATIVE CATABOLIC REACTIONS OF BACTERIA 75
2. Production of phenolic bodies from tyrosine.

+ NH 3

Paraoxyphenyl propionic acid

+ 3 O =
CH 2 .CHNH 2 .COOH CH 2 .CH 2 .COOH

Tyrosine. Paraoxyphenylpropionic acid

(deaminization)

OH OH OH

C0 2 + H 2 -> + C0 2 + C0 2

Paracresol

CH 2 .COOH CH 3 + 3 O =

Paraoxyphenyl Paracresol. Phenol,

acetic acid.

Phenol is not oxidizable by bacteria, hence it remains as such
unchanged in the culture media. Phenol (or cresol) may be absorbed
from the intestinal tract, but it appears eventually in the urine as an
ethereal sulphate, precisely as indol appears in the urine as indican.
Indol and phenolic bodies are not found in cultures containing utiliz-
able carbohydrate the bacteria which produce indol and phenols
from tryptophane and tyrosine, respectively, can obtain their requisite
energy far more directly and economically from the sugar than from
the nitrogen-containing amino acid. Doubtless the same general
principle applies to the formation of these aromatic substances in the
intestinal tract.

3. Formation of amines from amino-acids by bacterial action.

(a) Cadaverin from lysine. 1

CH 2 .CH 2 .CH 2 .CH 2 .CH.COOH CH 2 .CH 2 .CH 2 .CH 2 .CH 2

I I - I I- +C0 2

NH 2 NH 2 NH 2 NH 2

Lysine. Cadaverin.

(6) Putrescin from ornithin. 2

CH 2 .CH 2 .CH 2 .CH.COOH CH 2 .CH 2 .CH 2 .CH 2

| | -'/ "! I +00,

NH 2 NH 2 NH 2 NH 2

Ornithin. Putrescin.

1 Ladenburg, Ztschr. f. phys. Chem., 1886, xix, 780.

2 Ellinger, Ztschr. f. phys. Chem., 1902, xxix, 334; Ber. d. deut. chem. Gesell., 1889,
xxxi, 3183; ibid., 1900, xxxii, 3542.

76 BACTERIAL METABOLISM

H C NH V H C NH V

\r*t \r | fi

II //^- II //^*

C N * = C N * + CO 2

I I

CH 2 CH 2

I I

CHNH 2 CH 2 NH 2

I
COOH

Histidine. Betaimidazoleethylamine.

According to Vaughan, 1 betaimidazoleethylamine is possibly the
active poisonous principle of the protein molecule. Recent investiga-
tions would suggest that its liberation in the intestinal tract as the
result of bacterial decomposition of protein there and its absorption
into the body may be associated with symptoms of considerable
severity. The substance is not formed as a product of bacterial
metabolism in media containing utilizable carbohydrates.

B. Reactions of Bacteria in Media Containing Both Utilizable
Nitrogenous Substances (Protein and Their Derivatives) and Carbo-
hydrates. Carbohydrates contain no nitrogen; consequently pure
carbohydrate solutions are not complete foods for bacteria, they
are important chiefly as sources of energy to them. Generally speak-
ing, carbohydrates containing two, four, five, seven or eight carbon
atoms are not readily fermentable by bacteria. Those containing
six carbon atoms, particularly dextrose, are most readily utilizable,
those with three carbon atoms, generally speaking, somewhat less
so. Bioses, containing twelve carbon atoms, and starches appear to
be hydrolyzed to sugars containing six carbon atoms before they are
finally oxidized.

The final utilization of sugars for energy by bacteria varies accord-
ing to the type of organism; the following qualitative reactions are
illustrative of some of the general types of decomposition usually met
with. It must be remembered that the exact quantitative utilization
of carbohydrates by bacteria and the nature and composition of many
of the intermediary products formed from them are still uncertain.

1. C 12 H 22 On + H 2 O = C 6 H 12 O 6 + C 6 H 12 O 6 .

Lactose. Dextrose. Galactose.

Hydrolytic cleavage of a biose to two molecules of hexose sugar.

2. C 6 Hi 2 O 6 = 3CH 3 COOH. Pure acetic acid fermentation.

3. C 6 H 12 O 6 = 2CH 3 CHOHCOOH. Pure lactic acid fermentation.

1 Protein Split Products.

QUALITATIVE CATABOLIC REACTIONS OF BACTERIA 77

4. C 6 H 12 O 6 = CH 3 .CH 2 .CH 2 .COOH + 2C0 2 + 2H 2 . Pure butyric
acid fermentation.

5. C 6 Hi 2 6 = 2CH 3 CH)H + 2C0 2 . Pure alcoholic fermentation.

6. 2C 6 H 12 O G + H 2 = 2CH 3 .CH#)H.COOH + CH 3 .COOH +
C 2 H 5 OH' + 2CO 2 + 2 + 2H 2 . The type of fermentation produced
by B. coli in dextrose broth. 1

The sugars containing six carbon atoms appear to be somewhat
more utilizable than their corresponding alcohols: thus, the Shiga
bacillus (B. dysenteriae) can not ferment mannite; it can, however,,
readily ferment dextrose. This would suggest that the aldehyde
group CHO is somewhat more readily attacked than the alcohol
group CH 2 OH , for mannite has no aldehyde group and dextrose
has an aldehyde group. The alcohols in general appear to be less
readily acted upon by bacteria than are the corresponding aldehydes or
even organic acids, provided the latter are not too greatly dissociable.

The products of fermentation of higher alcohols, as mannite, by
bacteria are somewhat different from those of the corresponding sugars
(aldoses). The chief points of difference, according to our present
knowledge, consist principally in the production of more alcohol
when the 'higher alcohols are utilized than when the corresponding
aldoses are concerned. This has been worked out satisfactorily for
certain bacteria, notably the colon and the typhoid bacilli, by Harden. 2
It is not definitely known for many other organisms. The gas-forming
bacteria, as a rule, produce more gas and more alcohol from the alcohols
of the Ce series than from their corresponding aldoses. This gas
formation appears to result from the decomposition of formic acid
by the activity of a specific enzyme, formiase, according to the equa-
tion HCOOH = CO 2 + H 2 O. 3 Thus, B. coli and related gas-forming
bacteria, according to this theory, produce the ferment, formiase,
while B. typhosus, which also produces formic acid from the decom-
position of dextrose, does not possess this ferment and consequently,
forms no gas in sugar solutions. Formic acid is, therefore, somewhat
prominently represented among the decomposition products of carbo-
hydrates by the typhoid bacillus, while formic acid is either not
present or present in small amounts in corresponding cultures of colon
bacilli. 4

The qualitative changes produced in fats and lipoidal substances
by bacteria are not well known.

1 Kruse, Allgemeine Mikrobiologie, p. 294.

2 Jour. Hygiene, 1905.

3 Franzen and Stuppuhn, Ztschr. f. physiol. Chem., 1912, Ixxvii, 129.
Clark, Science, November 7, 1913.

78 BACTERIAL METABOLISM

VI. QUALITATIVE INFLUENCE OF UTILIZABLE CARBOHYDRATES
UPON THE ELABORATION OF PROTEOLYTIC ENZYMES.

Certain bacteria, as for example B. proteus, characteristically pro-
duce proteolytic enzymes which rapidly dissolve gelatin by hydrolytic
cleavage. These enzymes are exo-enzymes; that is, they may be
obtained sterile and free from bacteria simply by passing gelatin
liquefied by their action through sterile unglazed procelain filters.
Although the bacteria which elaborated the enzymes are removed by
this filtration, the sterile filtrate still contains the active enzyme which
will liquefy considerable amounts of gelatin. The function of these
enzymes is to prepare the gelatin for assimilation by the proteus bacil-
lus: the gelatin is broken down by enzyme action to gelatin peptone
or even to polypeptids. The proteus bacillus does not produce soluble
gelatin-splitting enzymes in gelatin containing utilizable carbohy-
drate, although sugar-free gelatin contains them in considerable
amounts. These gelatinases, however, will liquefy sugar-gelatin quite
as readily as sugar-free gelatin, indicating that the enzyme itself
is not inactivated by the sugar, at least in the amount usually
employed, 1 per cent. The same phenomenon is observed in cul-
tures of the cholera vibrio and many other bacteria which liquefy
sugar-free gelatin. Extensive investigations by Auerbach, 1 and by
Kendall, Day and Walker 2 have shown that the gelatinase, which,
as has been noted, is produced only in sugar-free gelatin, although
it liquefies sterile sugar gelatin, prepares protein for utilization by
these bacteria for purely catabolic purposes; if the organisms have
access to utilizable carbohydrate the enzyme is not produced by
them, because they utilize the sugar, not the protein, under these
conditions as the source of their energy. These observations indicate
how fundamentally the metabolism of bacteria is influenced by the
nature and composition of the substrate upon which they are grown.

VH. QUANTITATIVE MEASURE OF BACTERIAL METABOLISM.

It is possible to measure the nitrogen metabolism of bacteria under
varying conditions with a very considerable degree of accuracy in
spite of the minute amounts of products involved. Such measure-
ments are not only indicative of the nature and degree of the decom-

1 Arch. f. Hyg., 1897, xxxi, 311.

2 Jour. Am. Chem. Assn., 1914, xxxvi, 1962.

QUANTITATIVE MEASURE OF BACTERIAL METABOLISM 79

position of purely nitrogenous substances by bacteria; they furnish
quantitative evidence of the extent of the utilization of carbohydrates
by bacteria in preference to nitrogenous substances for fuel (catabolic)
purposes; that is to say, such measurements evaluate the nitrogen
metabolism of bacteria in purely protein solutions, and their nitrogen
metabolism in media containing both protein and utilizable carbo-
hydrate.

Such determinations have been made for a large series of bacteria
by Kendall and Farmer, 1 and Kendall, Day and Walker. 2 The gen-
eral method followed is to measure the amount of ammonia (deamin-
ization) which appears in fluid cultures of bacteria under various
conditions of growth. The following table shows, respectively, the
change in reaction (to neutral red as an indicator in terms of T acid
or alkali per 100 c.c. media) and the increase in ammonia (milli-
grams per 100 c.c. media), as certain bacteria are grown for ten days
in plain and dextrose broth respectively. The broths are identical
in initial composition and reaction, except that the "dextrose broth"
contains in addition to the ingredients of the "plain broth" 1 per cent,
of chemically pure dextrose. All other conditions are exactly parallel.
The results are averages of several strains of the same organism in
various lots of media. It will be seen that B. alcaligenes, for example,
which ferments no sugars, produces an alkaline reaction (indicated
as " " in the table) both in plain and dextrose broth: the amounts
of ammonia in both media are nearly the same.

All the organisms which ferment dextrose produce less ammonia
in the dextrose medium than in the corresponding sugar-free medium,
although the numbers of living bacteria were found to be greater in
the former than the latter. The small amount of ammonia in the
dextrose broth appears to be largely the nitrogenous waste incidental
to the utilization of protein for structural purposes: the relatively
large amount of ammonia observed in the corresponding sugar-free
broths is the combined "structural waste" and the " deaminization"
incidental to the utilization of protein for their energy requirement.
The progressively pathogenic bacteria, as the diphtheria, typhoid
and dysentery bacilli, produce much Ies3 ammonia in sugar-free
media than do the same organisms in various lots of media. 3 (Kendall,
Day and Walker.)

1 Jour. Biol. Chem., 1912, xii, 13, 215, 219, 465; xiii, 63. Methods ^iven here.

2 Jour. Am. Chem. Soc., 1913, xxxv, 1201-1249.

3 Ibid.

80 BACTERIAL METABOLISM

Sugar-free broth. Sugar broth.

Ten-day observations. Reaction. 1 NH.s Reaction. 1 NH.s

B. alcaligenes -1.25 +3.50 -1.15 +5.30

B. dysenterise (Shiga) .... -0.30 +4.20 +2.80 0.00

B. dysenteries (Flexner) ... -0.25 +3.10 +2.45 0.00

B. typhosus -0.45 +5.40 +3.30 +0.60

B. diphtheria ...... 0.50 +3.10 +2.80 +1.05

B. of hemorrhagic septicemia . 0.20 +4.70 +2.25 +0.35

B. paratyphosus alpha and beta 0.10 +7.50 +3.90 +1.20

B. icteroides -0.10 +4.20 +3.80 +2.10

B. of hog cholera avirulent . . -1.25 +16.45 +3.70 +1.05

B. of hog cholera virulent . . -0.75 +8.40 +2.65 +1.05

B. of fowl cholera .... - .00 +13.65 +3.35 +0.70

B. of Morgan - .33 +29.50 +3.90 +29. 66 2

B. coli - .00 +24.40 +4.90 +0.35

B. cloacae - .20 +39.20 -0.30 +36. 40 2

B. proteus . ...,..- .98 +58.40 +3.55 +1.40

Sp. cholerse - .45 +62.80 +2.00 +0.70

Sp. of Finkler and Prior . . . -1.00 +27.30 +1.50 +0.70

Sp. of Metchnikoff . . . . -4.30 +41.30 +2.70 +0.70

B. pyocyaneus -1.85 +30.30 -1.33 +41.50

Streptococcus +0.70 +1.40 +5.00 +0.70

Staphylococcus -0.75 +38.70 +3.75 +0.70

Mic. tetragenus +1.00 +2.10 +3.00 +0.70

Mic. melitensis 0.10 +6.30 +3.50 +0.70

Vm. SIGNIFICANCE OF BACTERIAL METABOLISM, WITH SPECIAL

REFERENCE TO THE SPARING ACTION OF UTILIZABLE

CARBOHYDRATE FOR PROTEIN.

Considerable emphasis has been placed upon the sparing action
of utilizable carbohydrate for protein in the preceding pages. It now
remains to summarize the salient features of this aspect of bacteriology
and to indicate briefly by means of a few illustrations precisely how
a comprehension of the principles underlying bacterial metabolism
may be made use of in controlling, or at least influencing the action
of these microorganisms upon their environment. The examples
selected are chosen rather with a view of indicating the extreme range
of the subject than for completeness along any limited line of inves-
tigation.

]. The Composition of Bacteria. Experiments quoted previously
(page 60) show very clearly that the percentage of composition of the
bacterial cell varies according to the medium in which it is grown.
Particularly striking is the difference in nitrogen content when the
same bacterium is grown in media of the same nitrogenous composition
and reaction with and without the addition of utilizable carbohydrate.

1 Neutral red, = alkaline reaction, + = acid reaction.

2 These organisms can utilize 1 per cent, of dextrose without forming enough acid
to inhibit their growth; after the dextrose is used up they attack the protein for their
fuel needs hence the ammonia production in a medium containing utilizable sugar.
During the initial period when sugar is present, the ammonia value is very little, and
the reaction is acid.

THE SIGNIFICANCE OF BACTERIAL METABOLISM 81

2. The Recognition of Bacteria. The recognition of many kinds
of bacteria, as for example members of the intestinal group, depends
upon the reactions these organisms induce in various sugars. Thus,
B. alcaligenes ferments no sugars; B. dysenterise ferments dextrose
with the production of acid; B. proteus ferments dextrose and sac-
charose with the evolution of gas and the production of acid; B.
coli ferments dextrose and lactose with the evolution of gas and the
production of acid; B. coli coagulates milk, while B. proteus charac-
teristically peptonizes it. All of these reactions are explained per-
fectly upon the theory that utilizable carbohydrate protects protein
from bacterial breakdown. Thus, B. alcaligenes does not utilize
any carbohydrate; as is well known, it is carnivorous. B. dysenterise
can utilize dextrose, and consequently it produces acid in a medium
containing both protein derivatives and this sugar: similarly, B.
proteus and B. coli ferment dextrose and in addition a specific biose.
B. proteus, however, does not ferment lactose, hence it attacks the
protein of milk; while B. coli, which does ferment lactose, produces
an acid coagulation in milk: the acid resulting from the fermenta-
tion of the milk sugar (lactose) protects the proteins of the milk.
In each instance the organisms attack the utilizable carbohydrate
whenever it is present, in preference to the protein for their energy
requirements. If bacteria did -not habitually utilize carbohydrate
in preference to protein for their fuel needs, these fermentation reac 1 -
tions would be of no value whatsoever as diagnostic tests for these
various microorganisms.

3. Certain bacteria, notably B. proteus, produce active, soluble
(extracellular) enzymes when grown in sugar-free gelatin, that
bring about an energetic liquefaction of this medium, which becomes
alkaline in reaction. If the organisms are grown in dextrose gelatin
no liquefaction takes place; the bacilli produce CO 2 and H 2 as well
as acid in dextrose gelatin, using the sugar in preference to the protein
for their energy needs. The liquefied gelatin containing the soluble
gelatinase may be sterilized by passage through a Berkefeld filter,
thus removing all bacteria. The filtrate will liquefy sterile plain or
sterile dextrose gelatin, thus proving that the soluble enzyme, which
prepares gelatin for assimilation by proteus bacilli (and which is
only produced in a carbohydrate-free medium), acts specifically on
the protein irrespective of other substances which may be present.
In this instance the presence of utilizable sugar in cultures of living
proteus bacilli protects the protein (gelatin in the instance cited)

BACTERIAL METABOLISM

from bacterial attack, and inasmuch as proteus bacilli prepare gelatin
for assimilation through the action of a proteolytic ferment, the
ferment is not elaborated by them under these conditions. A pre-
cisely similar restriction of the development of gelatin-liquefying
ferments by utilizable sugars occurs in cultures of cholera vibrios and
other bacteria which habitually liquefy this medium. In each instance
the same explanation holds true.

4. Diphtheria bacilli do not produce their characteristic powerful
extracellular toxin in the presence of utilizable carbohydrate dex-
trose as Theobald Smith 1 showed several years ago. The toxin is
only formed in sugar-free media. In this case again the dextrose
shields the protein of culture media from attack by the diphtheria
bacillus, and consequently prevents the formation of toxin which
is apparently a true excretion produced incidental to the utilization
of protein for energy by these organisms. Similarly, tetanus and
Shiga bacilli fail to produce toxin in the presence of utilizable carbo-
hydrates.

5. Colon and proteus bacilli produce considerable amounts of
indol in sugar-free media, but no indol in the same media to which
utilizable sugar has been added. Here again the carbohydrate is
attacked by these organisms in preference to the protein. The fol-
lowing table summarizes briefly the salient features of the above
discussion :

Sugar- free media. 2 Dextrose media. 2

Nitrogen substance, Nitrogen substance,

Chemical composition of bacteria. per cent. per cent.

1. Pfeiffer bacillus 70.0 53.7

Pneumo bacillus 79 . 8 63 . 6

Rhinoscleroma bacillus . 76 . 2 62 . 1

2. Diphtheria bacillus

3. B. tetani ....

B. dysenterise (Shiga)

4. B. proteus

Sp. cholerse .

5. B. coli, B. proteus:

Odor ....
Reaction .
Products .

Sugar-free media.
Powerful extracellular toxin

of which on the average

0.005 c.c. kills guinea-pigs.
Powerful extracellular toxin

produced.
Toxin present.
Soluble, extracellular gela-

tinase formed.
Soluble, extracellular gela-

tinase formed.

Foul.

Strongly alkaline.
H2S, indol, phenols, am-
monia, etc.

Sugar media.

No toxin produced; several
cubic centimeters medium
fails to kill guinea-pigs.

No toxin produced.

No toxin present.
No gelatinase formed.

No gelatinase formed.

None.

Strongly acid.

H 2 , CO 2 , lactic acid.

1 Tr. Assn. Am. Phys., 1896.

2 Nitrogenous constituents and reaction precisely the same in both sugar-free and
sugar-containing media. The only difference is that the dextrose medium contains
1 per cent, of dextrose in addition. The organisms studied have, therefore, a choice
between protein and sugar for catabolic purposes.

FERMENTATION AND PUTREFACTION 83

DC. FEEMENTATION AND PUTREFACTION.

The terms "fermentation" and "putrefaction" have been confused
and even used synonymously- in bacteriological, chemical and even
legal nomenclature, but they represent essentially distinct and generic
types of bacterial activity. They indicate, or should indicate
respectively, microbic decomposition of two quite distinct types of
organic compounds, the carbohydrates and closely related nitrogen-
free compounds, on the one hand (fermentation), and nitrogenous
organic substances on the other hand, putrefaction. There are
substances intermediate in character between carbohydrates and
proteins, or fats and nitrogen-containing compounds in which it
would be difficult to predict a priori which term would be correct
glucose amine is such a substance. Glucose amine is an amino-
aldose, containing both nitrogen and carbohydrate groupings. Such
instances, however, are uncommon and do not militate against the
correctness of the general theory that fermentation and putrefaction
are distinct processes. 1

Fischer 2 has defined fermentation in the broad sense it should be
used in bacteriology, essentially in the following terms: "Fermenta-
tion is the biochemical decomposition of nitrogen-free compounds,
chiefly carbohydrates, by the action of microorganisms." Similarly,
putrefaction is defined as "The biochemical decomposition of nitro-
genous organic compounds by the action of microorganisms."

Fermentation and putrefaction are probably enzyme phenomena.

Transposing the sparing action of utilizable carbohydrate for
protein, which has been repeatedly emphasized in the preceding
pages, it may be stated that in the catabolic phase of bacterial metab-
olism "fermentation takes precedence over putrefaction," 3 meaning
by that that bacteria which can utilize carbohydrate derive their
energy requirements from the utilizable carbohydrate when they are
growing in media containing both carbohydrate and protein. The
results of this sparing action of utilizable carbohydrate for protein
have been indicated in the preceding pages, sections V-VIII, inclusive.

1 Kendall, Jour. Med. Research, 1911, N. S., xx, 140-144.

2 Vorlesungen iiber Bakterien, 1903, II Aufl., 206.

3 Kendall, Jour. Med. Research, 1911, N. S., xx, 140-144.

CHAPTER V.

SAPROPHYTISM, PARASITISM, AND PATHOGENISM.

I. DEFINITIONS AND LIMITS.
II. THE CYCLE OF PARASITISM.

III. THE CYCLE OF PATHOGENISM.

IV. DISTRIBUTION OF PARASITIC AND

PATHOGENIC BACTERIA IN NA-
TURE.

V. How PARASITIC AND PATHOGENIC
BACTERIA REACH MAN.

A. The Occurrence of Parasitic

Bacteria upon the Bodies of
Healthy Men and Animals.

B. How Pathogenic Bacteria

Reach the Body.

1. Air-borne Infection.

(a) Dust.
(6) Droplet.

2. Soil-borne Infection.

3. Water-borne Infection.

4. Food-borne Infection.

5. Animal Carriers.

(a) Direct Contact.
(6) Indirect Transfer.

Transfer.

(d) Intermediary Host.

6. Human Carriers.

7. Contact Infection.

8. Germinal and Prenatal

Infection.

C. Portal of Entry: Atria of In-

vasion.

1. Skin and Adnexa: Ear,

Eye. SubcutaneousTis-
sue, Tonsils, Salivary
Glands, Nasal Cavity,
Lungs.

2. Mucous Membranes:

Mouth, Stomach, In-
testines.

3. Geni to-urinary System:

Vagina, Uterus, Ure-
thra, Urinary Bladder
and Ureter, Kidneys.

D. Where Bacteria Multiply in

the Body.

E. Where and How Bacteria

Escape from the Body.
VI. BALANCED PATHOGENISM; EPIDEMI-
OLOGY.

I. DEFINITIONS AND LIMITS.

THE most conspicuous and important function of bacteria in the
economy of Nature is to maintain a continuity between the Animal
and Vegetable Kingdoms by restoring in utilizable form to the Plant
World the elements contained in the complex organic compounds
which comprise the dead bodies of plants, animals and their products.
Bacteria dissipate much of the energy accumulated in these dead
bodies and oxidize the elements contained in them to inorganic, fully
mineralized salts. These salts are resynthesized by the chlorophyll-
bearing plants through the energy of sunlight to carbohydrates,
proteins and fats, and in these complex combinations the elements
are again available for animal food.

The bacteria which live upon this dead organic matter, and whose
function it is to effect its degradation and ultimate mineralization,
are called saprophytic bacteria. They are specifically the most

DEFINITIONS AND LIMITS 85

numerous, chemically the most active, and economically the most
important members of the phylum Bacteriacese. They are rarely
pathogenic, that is, they rarely initiate disease in man or the lower
animals. Whenever they are found associated with morbid processes
their presence is usually to be explained on the ground that they are
secondary invaders.

A smaller group of bacteria are parasitic, that is, they exist upon
the bodies of living plants, animals or men. Many of them are rarely
met with in Nature far removed from their respective hosts. Their
activities are not usually in opposition to those of their host and their
presence is therefore unnoticed. They may become invasive, how-
ever, whenever the natural barriers, which ordinarily suffice to keep
them out, are impaired.

From the parasitic bacteria there has been gradually evolved a
small but formidable group of organisms, the pathogenic bacteria,
whose activities are in partial opposition to those of their host. The
pathogenic bacteria, like the parasitic bacteria, require a living host,
but they differ from the parasitic forms in that they actually invade
their hosts and induce progressive disease from host to host.

There are no sharply definable limits between these three groups
of bacteria, the saprophytic, parasitic, and pathogenic; the latter
appear to have arisen from the former by a process of evolution.
Certain general modifications in the general types of chemical activity
manifested by these groups are discernible, however, which are partly
the result and partly the cause of their change in environment as
they have passed from a saprophytic to a parasitic existence. Promi-
nent among these modifications and activities is a gradual decrease
in the intensity with which the parasitic and pathogenic bacteria act
upon their environment.

The essential function of the saprophytic bacteria in Nature is to
effect a rapid, deep-seated degradation of organic matter to simple
compounds; these organisms decompose a relatively large amount
of substance in a relatively short time. They are chemically active
and many of them form highly resistant spores which enable them
to survive prolonged periods of environmental vicissitude. The
habitually parasitic bacteria, on the other hand, which exist upon
the bodies of living animals, and the progressively pathogenic bacteria
which develop within the tissues of animals are not subjected to
extremes of temperature and food supply; they rarely or never form
spores. The chemical activity of these organisms is usually much

86 SAPROPHYTISM, PARASITISM, AND PATHOGENISM

less pronounced than that of the saprophytic bacteria. 1 Indeed,
intense chemical activity would be incompatible with their continued
parasitic existence, for the damage to their host would be insupport-
able. The parasitic and pathogenic bacteria do not, for example,
produce widespread liquefaction of the tissues, even when large
numbers of them are actually growing in the body of the host. The
growth of invasive organisms in the animal body is characterized by
subtle changes in the composition of the tissues of the host and the
development of these reciprocal reactions between host and parasite,
which collectively are included in the newly developed science of
Immunology.

It would appear, therefore, that in their evolution toward parasit-
ism, those bacteria which could thrive without producing deep-seated
and rapid degradation of proteins, that is to say, whose metabolism
approached more closely the intracellular metabolism of their host,
would be the more adaptable to a parasitic existence, and this is in
accord with what is known of the chemistry of these organisms.
Their metabolism approaches rather closely that of their host.

H. THE CYCLE OF PARASITISM.

The cycle of parasitism for bacteria whose life cycle is such that
but a limited excursion outside their host is possible for them and
this appears to be the case for the majority of organisms parasitic on
man consists of three separate and well-defined stages, as Theobald
Smith 2 has so clearly pointed out. They must first reach an appro-
priate host; secondly multiply at least temporarily thereon, and
thirdly escape to other suitable hosts. Each phase of this parasitic
existence must be exactly fulfilled, otherwise the cycle is broken and
that particular strain dies out. It is not surprising, therefore, that
the bacteria habitually parasitic for man are found variously upon
the surface of the body in the upper respiratory tract, the gastro-
intestinal tract, or upon the mucous surfaces which are in direct
communication with the exterior. Escape from the body of the host
to other hosts is readily accomplished from these positions. 3

Under special conditions, parasitic bacteria may actually invade
the body of the host and become, therefore, temporarily pathogenic.

1 Theobald Smith, Am. Med., October 22, 1904, viii; Kendall, Boston Med. and
Surg. Jour., 1913, clxix, 749.

2 Theobald Smith, loc. cit.

3 Theobald Smith, loc. cit.

THE CYCLE OF PATHOGEN ISM 87

Such an invasion is usually subsequent to a preexisting disease or to
local weakening of the tissues which under normal conditions suffice
to exclude these organisms. The disease produced by parasitic organ-
isms is usually non-specific in character and sporadic in distribution,
and ordinarily it does not attain epidemic proportions. The bacteria
which have penetrated into the tissues of the host are locked up there,
as it were, and their descendants cannot escape to other hosts, at least,
in numbers sufficient to perpetuate the invasive strain, for these
organisms have not perfected their pathogenic cycle. Parasitic
organisms, in other words, are " opportunists," as Theobald Smith
has admirably called them, rarely initiating disease, but usually able
to penetrate the body as secondary or terminal invaders. The colon
bacillus, for example, is an habitual parasite in the gastro-intestinal
tract of man and many animals. Under certain conditions it may
become invasive, causing cystitis, appendicitis, peritonitis, or other
inflammatory lesions, but it does not ordinarily become progressively
pathogenic for successive hosts, producing epidemics of cystitis, ap-
pendicitis or peritonitis. The staphylococcus is a common inhabitant
of the skin of healthy man. When the continuity of the epidermis is
destroyed, the organism may become invasive, causing furuncles,
osteomyelitis, or endocarditis. The pneumococcus is found in the
respiratory tract of many normal men, particularly in large cities,
where it exists as an "opportunist," ordinarily producing no harmful
effects, but frequently becoming invasive and producing a variety of
lesions when the general resistance of the host is lowered. 1 These
parasitic bacteria have not perfected their mechanism of entry into
the tissues of the host, and of escape from the tissues to the exterior,
consequently those strains which accidentally become invasive are
locked up in the body and, as a rule, either are overwhelmed by
thjeir host or perish with it. They are imperfectly pathogenic, in other
words.

m. THE CYCLE OF PATHOGENISM.

Habitually pathogenic bacteria those organisms which produce
progressive, specific disease from host to host actually invade the
living bodies of animals or man. This invasion may be direct* in which
event the microorganisms actually enter the tissues or body fluids

1 Recent studies by Cole and his associates indicate that the ordinary "mou^"
pneumococcus differs serologically from the strains found in the saliva of pneumonl^
cases. It is not improbable that similar serological differences may be demonstrated
in the group of the streptococci.

88 SAPROPHYTISM, PARASITISM, AND PATHOGENISM

and multiply there, or it may be indirect, in which instance their soluble
toxins alone are absorbed by the host. The cycle of pathogenism,
therefore, is more complex than the cycle of parasitism ; it necessitates
lodgement of the invading microbe on the body of the host, the location
and penetration of the necessary portal of entry (which involves an
initial skirmish between the organism and the non-specific natural
defences of the host), growth within the tissues of the host in the
presence of opposition there, escape from the tissues to the surfaces
of the host or to some channel in communication with the exterior
and, finally, the transmission of the organism, directly or indirectly,
to other suitable hosts. If the organism cannot force an entrance to
the tissues of the host, that is, if the natural defences of the host
suffice to keep out the prospective invader, the latter usually perishes
and no infection takes place; if the organism does penetrate the tissues
of the body, the invasion and growth of the microorganism leads to
disturbances of structure, function or composition of the host, which
are abnormal and inimical to his well-being. The production of disease,
therefore, depends ordinarily upon the ability of the microorganism
to multiply in the tissues or the body fluids of the host; bacteria which
cannot force an entrance into the tissues of the host, multiply there
and escape to the exterior and eventually to other susceptible hosts
do not produce progressive disease.

The nature and extent of the disease produced depends upon several
factors: (1) the kind of microorganism; (2) the number of micro-
organisms; (3) their ability to locate and force an entrance to the
tissues of the body (their virulence, in other words) ; (4) the location
and extent of their multiplication in the tissues of the host; (5) the
response of the tissues of the host to this invasion, and (6) the nature
and extent of the secondary, specific defense of the host in response to
the invasion.

The contagiousness of a disease depends upon the ability of the
invading organisms to escape from their host in sufficient numbers to
infect new hosts and to survive environmental vicissitudes until new
hosts are reached. A few examples will indicate the principal vari-
ants of the pathogenic cycle commonly met with among progressively
pathogenic bacteria.

The tubercle bacillus ordinarily gains entrance to the host through
the air passages. The organisms pass through the alveoli of the lungs,
set up infection there, and gradually are shut off from communication
with the exterior through the formation of the tubercle. After a

DISTRIBUTION OF PARASITIC AND PATHOGENIC BACTERIA 89

longer or shorter time, these tubercles eventually break down, typically
into the air passages and discharge there large numbers of tubercle
bacilli. These are coughed up by the patient and are eliminated from
the body, usually in enormous numbers, by droplets and in the sputum.
Pulmonary tuberculosis is typically a chronic, focal disease. The
perpetuation of the tubercle bacillus is assured through their elimina-
tion from the diseased body in enormous numbers through long periods
of time, their ability to resist desiccation, and the relative directness
with which they reach other hosts.

The typhoid bacillus gains entrance to the body through the mouth
and the intestinal tract. The organisms penetrate the intestinal
mucosa, develop in the internal organs, particularly the spleen, and
after a rather definite excursion in the tissues of the* body, enter into
the intestinal tract again, either through ulcers or the gall-bladder or,
occasionally, they appear in the urine. They are eliminated from the
body in great numbers, either with the feces, or less commonly, the
urine, and they gain access immediately to other subjects through
direct contact or more or less indirectly through water or food, in
sufficient numbers to set up infection in at least some of them.

The gonococcus is transmitted directly by contact. Occasionally
the infection may be somewhat less direct, involving the conjunctiva.

The plague bacillus may be transmitted from host to host,. either
directly in the case of pneumonic plague, where great numbers of
plague bacilli are coughed up from the lungs of one patient and trans-
mitted through inhalation to other patients, or somewhat more
indirectly, as is the case in bubonic plague. Bubonic plague appears to
be a true septicemia; the plague bacilli circulate, at least temporarily,
in the blood, and they are removed from the blood of one patient and
transmitted to another patient (either man or rat) through the agency
of the flea, which acts potentially as an hypodermic syringe, as it were,
in this instance. Plague bacilli are locked up in the tissues of the host
and were it not for the agency of a suctorial insect, as the flea, bubonic
plague would almost certainly disappear, because the organisms have
not perfected for themselves any mechanism of escape from one V)st
to the other.

IV. DISTRIBUTION OF PARASITIC AND PATHOGENIC BACTERIA

IN NATURE.

It has been shown in previous sections that comparatively few, if
indeed any, of those bacteria habitually parasitic or pathogenic for

90 SAPROPHYTISM, PARASITISM, AND PATHOGENISM

man are found in Nature far removed from rather intimate association
with their hosts. This is in accordance with the fact that few, if any,
of these organisms are provided with spores which would enable them
to survive exposure to long periods of conditions unfavorable to their
growth. It is true, however, that some, at least, of these organisms,
as for example, the typhoid bacillus, can survive for longer or shorter
periods of time in the soil, particularly if it be frozen, or in water, for
days or even weeks. There is little evidence that these bacteria multiply
extensively outside the body; on the contrary, they tend to die off
rather rapidly. In any event, their existence depends upon their
reaching a suitable host again within a comparatively brief period.

There- are a few spore-forming bacteria which occasionally infect
man when associated conditions are favorable for them. Of these
the bacillus of lockjaw, B. tetani; of botulism, B. botulinus; the
gas bacillus, B. aerogenes capsulatus; and the anthrax bacillus are
well-known. These organisms are not habitual parasites, however;
they are "saprophytic opportunists." That is, they could in all
probability exist if man were eliminated from their environment.

V. HOW PARASITIC AND PATHOGENIC BACTERIA REACH MAN.

A. The Occurrence of Parasitic Bacteria upon the Bodies of Healthy
Men and Animals. The continual exposure of the skin of man to his
environment makes it almost inevitable that microbes shall collect
there. It is quite probable, however, that the large number of micro-
organisms which reach the skin are not only non-pathogenic, they are not
even habitually parasitic. Most of them are found there only trans-
iently. Certain organisms, however, occur among these adventitious
microbes, which appear to be habitual parasites, and many of these
bacteria, under certain conditions, produce disease. Of these, Staphy-
lococcus aureus and albus and Streptococcus pyogenes are almost
invariably present not only on the skin, but on the exposed mucous
membranes, particularly those of the nose and throat. The influenza
bacillus, diphtheria bacillus, the pneumococcus, and even the tubercle
bacillus, meningococcus and other organisms may also be occasionally
found, particularly in the nose and throat of healthy men. The
occurrence of these organisms is readily explained; the secretions of
the nose and throat, as well as that of the skin are excellent culture
media for these organisms, which collect at these sites and grow upon
the various secretions and desquamated cells.

PARASITIC AND PATHOGENIC BACTERIA 91

The majority of these organisms, however, particularly the coccal
forms, as the staphylococcus, streptococcus and pneumococcus are to
be regarded as "opportunists"; they do not of themselves initiate
disease, as a rule. They are to be regarded rather, as Theobald Smith
has called them, "organisms of the diseased state/' because of their
invasion of the bojdy secondary to other, intercurrent diseases. Even
the tubercle bacillus and the diphtheria bacillus, particularly the
latter, have been found in the mouths of men who apparently have
had neither tuberculosis nor diphtheria, yet these organisms appear
to be virulent when tested in the usual manner and presumably might
be able to incite disease whenever conditions favor their entrance to
the tissues of the body. Theoretically at least, people who harbor
these organisms are potential sources of danger to others. Even the.
internal organs of healthy individuals may contain parasitic bacteria
without harm, although these organisms naturally are not present in
large numbers. Tubercle bacilli have been found occasionally in
lymph glands in normal man and in cattle. Intestinal bacteria also
occur not infrequently in the apparently healthy tissues of the body.
In rare instances, B. coli may be present in the urinary bladder without
causing noteworthy symptoms.

B. How Pathogenic Bacteria Reach the Body. The manner in
which bacteria of the "opportunist" type reach the body has been
considered above. It is now necessary to consider the manner in which
bacteria which cause progressive disease from man to man reach
the body.

1. Air-borne Infection. Bacteria which cause progressive disease,
particularly of the respiratory tract, are discharged from the diseased
body principally through the mouth and nose and find lodgment in the
environment of the patient through the medium of the air, from whence
they settle upon various substances, as food, clothing, and walls and
floors of rooms. These bacteria probably do not proliferate to any
extent outside of the body, but they resist drying and may remain
fully virulent for considerable periods of time and potentially able to
infect a certain proportion of those individuals who may be exposed
to them.

These air-borne infections are transmitted in at least two rather
distinct ways: (a) by dust, and (6) by droplet infection.

(a) Organisms which are transmissible through dust must first
of all be able to survive considerable periods of drying. The larger
particles of dust to which bacteria may become attached soon settle

92 SAPROPHYTISM, PARASITISM, AND PATHOGENISM

from the air, but smaller particles may remain suspended for some
time, depending on the velocity of air currents and the nature, size
and shape of the particles. Dusting and sweeping in rooms naturally
stir up particles which have settled from the air, and even larger par-
ticles may be resuspended in this way. Tuberculosis has frequently
been suspected to have been transmitted through the inhalation of
infected dust particles, that is, particles of dust which have dried
tubercle bacilli adhering to them. Careful investigation has shown that
houses in which careless consumptive patients have lived have been
responsible for the transmission of tuberculosis. The ward-room of
a battle ship is known to have become infected with tubercle bacilli
early in its career and at least two successive details of officers con-
tracted tuberculosis in this place. Guinea-pigs exposed on the floor
of these so-called tuberculous rooms are quite frequently successfully
infected with the tubercle bacillus.

The extent to which dust dissemination is a factor in transmitting
disease, however, is not at all definitely known. It must be emphasized
that the transmission of disease through dust is not necessarily a
very direct one, because the inciting organisms may pass a very con-
siderable period of time in dust before they reach a favorable host.
In this sense, transmission of disease by dust is a relatively latent one.

(6) Droplet Infection. Fliigge 1 and his pupils were the first to
demonstrate that minute droplets of spray may be eliminated from the
mouth during talking, sneezing and coughing. These droplets are
frequently carried through the air for some distance, even as much as
ten meters in a quiet room. Usually the more minute particles remain
suspended in the air for some time. The possibility of droplet infection
has been definitely proven in the following manner: Agar plates
containing sodium carbonate are placed at various heights and dis-
tances from the experimenter, who places in his mouth a solution of
phenolphthalein and then talks in a natural manner, expelling droplets
containing phenolphthalein during his speech. This dye is transmitted
with the droplets until they reach the agar plates, where bright red
spots are produced which are very readily observed. In like manner,
cultures of B. prodigiosus placed in the mouth will infect agar plates
at similar distances.

The transmission of disease by droplet infection may be, and fre-
quently is, a very direct one. Bacteria which are air-borne or borne
by droplets may remain alive for several weeks in indirect sunlight,

1 Ztschr. f. Hyg., 1897, xxv, 179.

PARASITIC AND PATHOGENIC BACTERIA . 93

but all of them are readily killed if they are exposed to direct sunlight.
The virus of whooping-cough, mumps, measles, influenza, cerebro-
spinal meningitis, pneumonic plague, tuberculosis, the exanthemata,
the diphtheria bacillus, and possibly the pneumococcus may be spread
in this manner. Air-borne infections probably rarely take place
in the open air where the sunlight is strong. This does not apply
to droplet infections where one individual coughs, talks or sneezes
directly into the face of another. Air-borne infections, particularly
droplet infections, are potentially common where overcrowding occurs,
as in tenements, public gatherings, railway trains, schools, and factories.

2. Soil-borne Infections. Those bacteria which are occasionally
pathogenic for man and produce sporadic disease in man, and whose
habitat is the soil, are for the most part spore-forming organisms. They
commonly enter the body through wounds. Of these the bacillus of
tetanus, malignant edema, symptomatic anthrax, of anthrax, and the
gas bacillus are the best known but, with the exception of the latter,
they are not habitually human parasites. Of those bacteria which
are habitually pathogenic for man, typhoid, cholera, paratyphoid
and probably dysentery may be soil-borne, but ordinarily infection
with these organisms does not take place through the soil.

3. Water-borne Infection. The viruses of excrementitious diseases
typhoid, paratyphoid, dysentery, and cholera are not infrequently
transmitted from man to man through contaminated water. Feces
containing these organisms get into water supplies, reach man again,
incite disease in man, again escape in the feces and reenter water
courses, thus being recirculated. The cycle may be somewhat more
complex, as for example, when typhoid dejecta are thrown upon the
ground and are eventually washed directly into water supplies and
thus reach man again.

4. Food-borne Infection. A considerable number of pathogenic
bacteria may reach man through food, although food which is infected
is usually rendered so through the handling of it by man. Milk is
probably the most common food thus to be infected and it is par-
ticularly dangerous for two reasons. In the first place, its opacity
makes it difficult to distinguish foreign substances which may be in it;
and again, it contains all the elements which are necessary for the
food of man and incidentally for the majority of bacteria. Scarlet
fever, diphtheria, tuberculosis both human and bovine, Malta fever,
epidemic sore throat or tonsillitis, typhoid, dysentery, foot-and-mouth
disease, many diarrheas of children, milk sickness, and the organisms

94 SAPROPHYTISM, PARASITISM, AND PATHOGENISM

of cholera infantum, and, rarely, Asiatic cholera as well, all may be
transmitted from milk.

Shell-fish, particularly oysters, have been known to transmit enteric
diseases. This has been due, in the past, largely to their exposure
in the estuaries of rivers where sewage flowed freely over them.
Typhoid bacilli enter the mantle cavity of the shell-fish, remain alive
there and enter the digestive tract in a viable state when the shell-
fish are consumed in an uncooked condition.

Meats, particularly from beef and swine, have been known to transmit
paratyphoid fever, botulismus (sausage poisoning) and meat-poisoning
as well. There is, in addition, a group of cases with somewhat insidious
symptoms, which are probably due to the consumption of food, par-
ticularly meat, which has been decomposed by saprophytic bacteria.

5. Animal Carriers. The microbic diseases which are transmissible
to man from animals and from man to man by animals are varied
in character. They comprise protozoan and bacterial infections and
the so-called " filterable viruses." Of these diseases, comparatively
few are common to man and animals. Microorganisms may be trans-
mitted to man by animal carriers in at least four distinct ways:

(a) By direct contact.

(b) By indirect transfer.

(d) By intermediary hosts.

(a) Direct Contact. The transfer of glanders from the horse to
man, of anthrax from cattle and sheep to man and of hydrophobia
from dogs to man represents direct transfer of the virus from the sick
animal to the well man. Other diseases are thus transmitted, but the
examples given suffice for illustration.

(b) Indirect Transfer. Insects are common carriers in the indirect
transmission of the virus of disease from man to man. Flies are known
to have carried typhoid bacilli from typhoid dejecta to milk or other
food, which in turn has been consumed by man, resulting in infection.
The same insect, doubtless, when conditions are favorable, can and
does carry other enteric bacteria paratyphoid, dysentery and even
cholera organisms. It is very probable that other insects also par-
ticipate in the indirect transmission of bacteria. Acute conjunctivitis,
particularly that form which is prevalent in Egypt, is supposed to
be spread in this manner.

(c) Mechanical Transfer. Suctorial insects are known to transmit
the viruses of certain diseases which circulate in the blood stream of

PARASITIC AND PATHOGENIC BACTERIA 95

animals to man, incidental to feeding. Thus, the flea transmits the
plague bacillus from rat to man, from man to man, and possibly from
man to the rat. The louse similarly spreads the virus of typhus from
man to man. In the instances cited the insect is probably not a true
intermediary host, for the virus does not necessarily multiply in the
insect, nor does the virus undergo any essential transformation, so
far as is known, in the insect. Nevertheless, the transmission of the
viruses of these diseases bubonic plague and septicemia, for example,
depends upon the agency of suctorial insects for their passage from
host to host. Other insects also transmit disease, but the evidence
in a majority of instances is somewhat less definite than the cases
cited.

(d) Intermediary Hosts. Certain insects, notably mosquitoes,
transmit disease from man to man only after the virus has passed
an extracorporeal cycle in the extrinsic host the mosquito in this
instance. Thus, Anopheles transmits malaria from man to man and
Stegomyia fasciata, or as it is now called, Aedes calopus, transmits in
similar manner, the virus of yellow fever. Transmission in these cases
is through the female insect and a definite interval (latent period)
must elapse between the time of biting the patient and the time when
the mosquito becomes infective to the non-immune host.

6. Human Carrriers. Individuals who are apparently healthy
occasionally harbor within their bodies (in free communication with
the exterior, however, either through the respiratory tract, the gastro-
intestinal tract, the urinary tract or the skin) bacteria which are
capable of inciting disease in others. Such individuals are known as
bacillus carriers; frequently they eliminate these pathogenic bacteria
in large numbers.

The bacillus carrier may or may not give a history indicating
recovery from an infection of the specific organism which he "carries."
Bacillus carriers may be temporary carriers, in which event they harbor
the pathogenic bacteria for but a few weeks; or they may become
habitual carriers, in which case the organism may be excreted for
considerable periods of time, even years. The excretion may be
constant or intermittent.

The typhoid bacillus is a common organism to be thus carried. It
appears to localize eventually in the gall-bladder or the bile ducts, less
commonly in the urinary bladder, and it may appear occasionally
in large numbers in the feces or urine of the carrier. Women are more
commonly found to be typhoid carriers than men. Similarly, para-

96 SAPROPHYTISM, PARASITISM, AND PATHOGENISM

typhoid, dysentery and cholera organisms may be excreted in the feces
through long periods of time, rarely or never, however, in the urine.
Slowly progressing focal diseases, as pulmonary tuberculosis are,
in a sense, spread by carriers, for the patient may survive for years,
excreting daily large numbers of tubercle bacilli. The line of demarca-
tion, in other words, between the human bacillus carrier and the
patient in whom a focal disease is chronic for long periods of time is
not sharply circumscribed.

7. Contact Infection. The direct transmission of bacteria from man
to man is well exemplified in the venereal diseases, gonorrhea and
syphilis, which are usually transmitted by direct contact. Diseases
of the respiratory tract, as tuberculosis, diphtheria and whooping-
cough, may be transmitted directly from patient to patient by kissing,
swapping chewing gum, by eating utensils, etc. Soiled fingers may
transmit the typhoid bacillus from a typhoid patient to other indi-
viduals. All of the excrementitious diseases may be spread in a
similar manner, under certain conditions.

8. Germinal and Prenatal Infection. True germinal infection implies
that a disease-producing microorganism is carried by the ovum or
spermatozoa and incorporated in the embryo prior to its development.
This method of transmission is not definitely worked out, although it
has been claimed that syphilis may be thus transmitted by the male
to the ovum in utero, the mother remaining uninfected by the disease.

In prenatal infections the organisms must pass the placental
barrier. This implies that the fetus becomes infected directly from the
maternal blood stream, or by continuity of growth of the organisms
through the placenta. The placental form of infection is not conceded
by all observers, but it is reasonably certain that congenital syphilis
may be contracted thus. Smallpox, measles, dysentery, various
pyogenic infections, and, rarely, pneumonia are occasionally said to be
prenatally transmitted to the fetus. With respect to tuberculosis,
there is difference of opinion. A very few cases are on record in which
prenatal infection seems almost certainly to have taken place, for the
newborn infant exhibited lesions which were so far advanced that no
other explanation than prenatal infection suffices to explain them.

C. Portal of Entry; Atria of Invasion. The bacteria which cause
infection in the human body may be provisionally divided into two
great groups: those of exogenous origin, which are not habitual
parasites of man; and those of endogenous origin, which are habitual
parasites of man.

PARASITIC AND PATHOGENIC BACTERIA 97

The girat majority of specific microbic diseases (in contradistinction
to non-specific inflammations) are incited by bacteria of exogenous
origin. These organisms must enter the host directly through their
respective appropriate atria to produce characteristic disease. For
example, the typhoid bacillus only causes typhoid fever when the
organism is swallowed and enters the body through the intestinal
tract. Infection of a skin- wound with typhoid bacilli will not result
in typhoid fever. Similarly, cholera vibrios do not produce the disease
cholera unless they enter the body through the gastro-intestinal tract,
although if cholera vibrios are introduced through the skin in experi-
mental animals they tend to migrate toward the intestinal tract, thus
suggesting a special affinity for the intestinal tissues. Pathogenic
bacteria of exogenous origin produce in general, progressive specific
disease from man to man. Bacteria of endogenous origin, on the
other hand those which occur habitually as " opportunists" on the
surface of the body or on mucous membranes opening to the exterior
ordinarily exist as harmless parasites. They may, however, and
occasionally do, become invasive, inciting local or generalized inflam-
matory reactions as a rule, rather than well-defined clinical syndromes
which are frequently so characteristic of infections with exogenous
pathogenic bacteria. The bacteria of the "opportunist" type do not
ordinarily gain entrance to the tissues of the body through sharply-
circumscribed atria and the disease they produce is usually not epidemic
in character.

1. Skin and Adnexa The intact skin is a natural barrier which
protects the underlying tissues of the body from bacterial invasion.
Its free exposure to the environment suggests that a great variety of
organisms find lodgment upon it from time to time; a majority of
these organisms are harmless, and probably transient saprophytes
which come and go irregularly. The moisture and excretions, how-
ever, appear to favor the limited development of a few types of bac-
teria, mainly those of the coccal group, which occur with sufficient
regularity to be regarded provisionally as habitually parasitic bacteria.
Of these, the pyogenic cocci are usually the most numerous; they
exist as "opportunists" on the surface of the skin or penetrate into
hair follicles and the ducts of the cutaneous glands, ordinarily, however,
without becoming invasive so long as the continuity of the skin is
maintained. Abrasions and cuts furnish a portal of entry to the sub-
cutaneous tissues, in which these parasitic bacteria frequently set up
inflammatory reactions. Friction may actually force them through

98 SAPROPHYTISM, PARASITISM, AND PATHOGEN ISM

the intact skin. The plague bacillus and certain types of staphylococci
are said to pass through the skin occasionally in this manner.

Streptococci and staphylococci are the more common habitual
bacterial parasites found on the skin. Staphylococcus epidermidis
albus (Welch), a variant of Staphylococcus pyogenes albus, is a particu-
larly common factor in the causation of the troublesome, but relatively
benign stitch abscesses which frequently develop where sutures are
introduced through the skin.

The damaged skin is the usual portal of entry for spore-forming
bacteria as well as the cocci mentioned above. Spores of the bacilli
of tetanus, anthrax, symptomatic anthrax, malignant edema and the
"gas bacillus," (B. aerogenes capsulatus, Welch) may pass to the
underlying tissues through abrasions of the skin and cause either
localized infections or widely distributed lesions. Even so insignifi-
cant an abrasion as an insect bite may furnish the necessary atrium
for infection. The umbilicus of the newborn furnishes a portal of
entry for certain bacteria; particularly severe is the infection of the
stump of the umbilicus with B. tetani, causing that very fatal "tetanus
neonatorum" which has been so common in the tropics in the past.
"Contused wounds and compound fractures are particularly dangerous;
the inflamed tissues furnish anaerobic conditions particularly favoring
the growth of anaerobic bacteria, as the tetanus and gas bacilli.
Clean-cut wounds are usually less liable to infection with anaerobic
bacteria. The free flow of blood with its bactericidal properties
washes out many bacteria, inhibits the growth of residual microbes,
and by virtue of the clot which soon seals the wound prevents the
entrance of other organisms.

The sebaceous secretions, particularly of the axilla and external
genitalia, are good culture-media for certain acid-fast bacteria, par-
ticularly B. smegmatis. The cerumen of the external ear is frequently
infected with Micrococcus cereus flavus, and the puncture of the
tympanic membrane may lead to direct infection of the middle ear
from the outside, with this or other organisms. Infection of the
middle ear may also take place directly through the Eustachian tube.
The blood and lymph may also deposit bacteria in the middle ear.

The conjunctiva, by virtue of its very exposed position, must
receive bacteria upon it very frequently. Its polished surface and
the mechanical cleansing by the flow of tears (which do not possess
germicidal properties) usually suffice to remove adventitious bacteria
and to prevent bacterial development under ordinary conditions. The

PARASITIC AND PATHOGENIC BACTERIA 99

conjunctival sac, which receives the washings from the conjunctiva,
is probably the recipient of many bacteria; of these B. xerosis occurs
with sufficient regularity in the conjunctival sac to be regarded as a
normal inhabitant. The pneumococcus is also found there. These
organisms are "opportunists," occasionally causing severe acute
conjunctivitis, although usually they are benign. Certain bacteria
affect the conjunctiva fairly readily. Among these organisms, the
gonococcus is particularly troublesome, causing a most severe inflam-
mation. Ophthalmia neonatorum, a gonorrheal infection of the con-
junctivse of the newborn of infected mothers, has been in the past
a most common cause of blindness. It has been claimed that the
meningococcus may occasionally pass from the eye through the tear
duct to the nasal cavity, and from there to the meninges.

Subcutaneous Tissue.- Many bacteria, particularly exogenous
pathogenic bacteria, do not develop in the subcutaneous tissues, as
for example, the majority of those organisms which induce specific pro-
gressive disease from man to man such as typhoid and cholera organ-
isms. On the other hand, many of those bacteria which are habitually
parasitic on the skin may produce infections of the subcutaneous
tissues which vary in severity from mild inflammations to severe
cellulitis. The staphylococci and streptococci are among the more
important of this type.

Tonsils. The crypts of the tonsils afford mechanical protection
to bacteria which gain access to them and the secretions and tissue
undoubtedly provide the necessary nutritive elements, consequently
it is not surprising to find many types of bacteria in them. Staphy-
lococci are almost invariably present and streptococci, particularly
non-hemolytic varieties, are very common. The tonsils, which are
in very direct communication with the lymphatic system, are impor-
tant atria of invasion, particularly for streptococci, and many cases
of low-grade infections of the body appear to have originated from
the passage of bacteria through the tonsils to the tissues of the body.
The extent to which the normal tonsils destroy bacteria their value
in the non-specific initial defense of the body against bacterial invasion
in other words is not clearly established. Generally speaking, how-
ever, the tonsils appear to bear the brunt of attack in certain diseases
and they are of undoubted importance in shielding the body from
invasion through the lymphatic tract by directly holding back these
bacteria. The promiscuous removal of tonsils, particularly in the
young, has no justification from available knowledge. The removal
of diseased tonsils is quite a different matter.

100 SAPROPHYTISM, PARASITISM, AND PATHOGENISM

Salivary Glands. The salivary glands of the mouth are sometimes
invaded by bacteria.

Nasal Cavity. Large numbers of bacteria, indeed practically all
known bacteria may at one time or another gain access to the nose
through the inhalation of air containing dust, by droplet infection,
from the tear ducts, and in other ways. The air which is inhaled is
freed from bacteria before it enters the trachea, largely during its
tortuous passage over the turbinates; the moist surface of the nasal
mucosa effectively arrests the progress of bacteria, which adhere to
it. The constant secretion of mucus encloses many of these organ-
isms, which are removed mechanically with the mucus. There is no
evidence that the nasal secretions are germicidal. The permanent
nasal flora is very limited, however. The pseudodiphtheria bacillus
is very frequently found there and pneumococci, streptococci and
staphylococci are relatively common. The true diphtheria bacillus
is found in the nasal cavity of about 1 per cent, of healthy individuals.

Lungs. The expired air in quiet, normal breathing is sterile:
also, the inhaled air is practically sterile before it reaches the bronchi,
for the moist tortuous passages of the nasal cavity mechanically
retain bacteria; the same mechanism prevents the expulsion of bac-
teria during exhalation, unless the breath is expelled forcibly either
through the nose or mouth. Bacteria leave the nose or mouth in
expired air only when the expiration is forcible enough to eject finely
divided droplets from the mouth or nose respectively.

The lungs are protected from bacterial invasion not only by the
tortuous nasal air passages, but by the ciliated epithelium which
covers the surface of the mucosa of the bronchi and bronchioles.
The rhythmic contractions of these cilia carry upward and outward
those bacteria which may have penetrated so deeply into the respira-
tory passages. Inhibition of the activity of these cilia by cold or other
environmental conditions may be a potent factor in the establishment
of infection in the respiratory tract. Occasionally bacteria succeed
in reaching the terminal bronchioles and alveoli of the lungs : they
are normally removed by the phagocytic activity of leukocytes (micro-
phages) or of certain fixed tissue cells (macrophages) . In spite of
these barriers, however, the lungs occasionally become infected. The
pneumococcus and tubercle bacillus are the most common primary
invaders of the lungs. Streptococci are more frequently secondary
invaders, although many primary lobular pneumonias are caused by
this organism.

PARASITIC AND PATHOGENIC BACTERIA 101

2. Mucous Membranes. The moist surface of mucous membranes
makes them excellent culture media for many bacteria which can
grow at the temperature of the body. The physiological secretions
which bathe these membranes, with the exception of the stomach,
are usually without germicidal properties; at best, their antiseptic
properties are w r eak. The removal of bacteria from such surfaces
is probably for the most part mechanical. The secretion of mucus,
which has been shown to enclose bacteria, may be an important factor
in their elimination.

Mouth. The mouth is a most important portal of entry for the
great majority of bacteria, both pathogenic and non-pathogenic,
which are associated with man. All of the intestinal bacteria, harmful
or benign, many of the bacteria which are associated with morbid
processes of the respiratory tract, and several which induce specific
lesions of the brain and spinal cord enter through this atrium. A
great majority of viruses which infect the respiratory tract and the
cerebrospirial axis also leave the body through the mouth or nose.

The normal flora of the mouth is quite varied, 1 including not only
bacteria which are ordinarily regarded as harmless, but also organisms
which occasionally or frequently incite disease. Thus, from 20 to 40
per cent, of healthy individuals living in large cities harbor typical
and apparently virulent pneumococci in their mouths; 2 about 2 per
cent, of school children harbor typical diphtheria bacilli in their
mouths. 3 Rarely, tubercle bacilli have been detected in the mouths
of apparently normal individuals.

It is worthy of note that an occasional abscess in the cervical region
may contain spiral organisms; frequently a careful examination will
reveal a sinus connecting the abscess with the mouth, perhaps origi-
nating at the base of a carious tooth. Dental caries is usually regarded
as a bacteriological process. The removal of bacteria from the teeth
and gums can not be satisfactorily accomplished by antiseptic mouth
washes and the saliva possesses no germicidal properties. Bacteria
are removed from the teeth mechanically by friction and are trans-
ported from the mouth to the stomach during the processes of mas-
tication and deglutition. ' The oral flora is most numerous before

1 For full literature and descriptions see Miller, Die Mikroorganismen der Mundhohle,
Leipzig, 1892, and Goadby, Mycology of the Mouth, 1903.

2 Recent observations by Cole and his associates indicate that the ordinary "mouth"
pneumococci differ in their serological reactions from pneumococci isolated directly
from pneumonia lesions.

3 Moss, Guthrie and Gelien have found a much larger proportion of diphtheria bacillus
carriers during a period when diphtheria was epidemic.

102 , ' SAPRaP-HYTISM, PARASITISM, AND PATHOGENISM

eating and almost absent immediately after eating a hearty meal.
Tubercle bacilli are swallowed thus and many of them eventually
appear in the feces.

Stomach. The acidity of the stomach during gastric digestion,
by virtue of the free hydrochloric acid of the gastric juice, is a potent
factor in the destruction of bacteria which reach the stomach both
from the mouth and the respiratory tract. Mineral acids are much
more powerful germicides than organic acids. The normal stomach,
therefore, is quite free from inflammations or irritations attributable
to the activity of bacteria. Many bacteria, however, run the gauntlet
of the stomach successfully, especially when the stomach is empty
(when the concentration of hydrochloric acid is very low) and pass
into the intestinal tract, where the conditions are much more favorable
for their growth. The passage of bacteria through the stomach prob-
ably takes place either very early in gastric digestion, when the
hydrochloric acid is not at its "digestive concentration" (about 0.2
per cent.), or after gastric digestion has ceased. When water or
other fluids are drunk, which do not call forth gastric juice, bacteria
doubtless pass through the stomach unharmed, and it is probable
that organisms included mechanically within food particles may
occasionally escape the action of the gastric acidity.

Certain aciduric bacteria 1 and even yeasts which are tolerant of
acid may be found occasionally in the normal stomach, but rarely
or never pathogenic bacteria. Abnormally, particularly when the
hydrochloric acid is deficient, many bacteria are found in the stomach
contents. Obstruction of the pylorus tends to increase the number
of bacteria in the stomach by promoting stasis of food. This con-
dition is particularly common in carcinoma of the pylorus. The
Oppler-Boas bacillus, sometimes called B. geniculatus, one of the
aciduric bacteria, is so frequently found in this pathological condition
it was at one time supposed to be an accessory factor; it is now known
to have no relationship to gastric carcinoma. B. geniculatus is also
found very commonly in cases of achlorhydria. Sarcina ventriculi
is also found in similar conditions.

The gastric acidity will destroy the toxins of B. diphtherise and
B. tetani; the toxin of B. botulinus is not inactivated by the gastric
juice. The toxins of the paratyphoid group of bacteria also appear
to be resistant to gastric digestion.

1 Kendall, Jour. Med. Research, 1910, N. S., xviii, 153.

PARASITIC AND PATHOGENIC BACTERIA 103

Intestines. The abundant intestinal contents, which vary some-
what in composition and reaction at different levels, provide conditions
which make the intestinal tract a very efficient combined incubator
and culture medium. Many kinds of bacteria may theoretically find
conditions well adapted to their rapid development there and it is
not surprising to find that bacterial proliferation is greater both in
nature and extent in the intestinal tract than in any other known
medium. It has been conservatively estimated that the average
daily fecal excretion of bacteria in a healthy adult on a normal diet
is expressed by the truly enormous number, 33 x 10 12 . About 47 per
cent, of the nitrogen of the feces is contained in the bodies of these
bacteria which, when dried, weigh nearly 0.5 gram.

The upper level of the intestinal tract, particularly the duodenum,
is relatively free from bacteria during interdigestive periods. The
duodenal bacterial population increases rapidly when food enters
this section of the alimentary canal and decreases when the food
passes to lower levels. The numbers of bacteria increase very greatly
where stasis of food becomes more marked and in the cecum and
large intestines generally there are continually present enormous
numbers of bacteria. 1

The types of bacteria found in the intestinal tract are influenced
markedly by the nature of the food of the host and by the ability of
the organisms themselves to change their metabolism to meet varia-
tions in the composition of this food. Those bacteria which can best
meet alternations in diet of the host are the ones which naturally
persist. The bacteria contained in the food itself may also play a
prominent part in determining the nature of the organisms which
are found in the intestinal tract. The colon bacillus is particularly
labile in meeting dietary alternations in the intestines and this organism
constitutes about 80 per cent, of the bacteria which can be isolated
from the feces of the adult.

At birth the intestinal tract is sterile and the embryonal feces, the
meconium, which is passed during the first eighteen hours after birth,
is sterile. Following this period of sterility there is a period lasting
about three days on the average, in which various adventitious organ-
isms are met with in the dejecta. The normal nursling flora begins
to appear by the end of the third day, following the ingestion of
breast milk. The dominant organism of this nursling flora is ordi-
narially an obligate anaerobe, Bacillus bifidus, which is one of the

1 Kendall, Jour. Med. Research, 1911, xxv, 126-130.

104 SAPROPHYTISM, PARASITISM, AND PATHOGENISM

best known examples of obligately fermentative organisms. It does
not thrive on a purely protein diet but requires carbohydrate, which
is normally supplied by the breast milk. Breast milk, it will be remem-
bered, contains on the average about 7 per cent, of lactose, 3 per cent,
fat and but 1.5 per cent, protein. The proportion of carbohydrates
to protein in the diet decreases as the infant becomes older and the
diet becomes more liberal, and this decrease in the percentage of
carbohydrate is associated with a diminution in the number of the
obligately fermentative bacteria, particularly of Bacillus bifidus, and
their gradual replacement by organisms which can thrive well on a
diet containing variable proportions of carbohydrate and protein. 1

Bacillus coli is a most labile organism with respect to its ability
to develop in the carbohydrate and protein constituents of the intes-
tinal contents at the ileocecal region and lower levels; this organism
is represented to the extent of fully 80 per cent, in the feces of healthy
men. Smaller numbers of other bacteria, as Micrococcus ovalis,
Bacillus acidophilus, B. proteus, B. mesentericus, B. aerogenes cap-
sulatus and many other varieties are found transiently or semi-per-
manently in the intestinal contents. Exogenic bacteria occasionally
invade the tissues of the body through the intestinal mucosa. Thus
typhoid, paratyphoid and dysentery bacilli and cholera vibrios may
produce severe infections. The tubercle bacillus may pass through
the apparently intact intestinal wall without leaving any evidence
of its passage. It is supposed that this organism penetrates the
intact mucosa and enters lymphatic channels suspended in fats and
eventually proliferates in deeper tissues.

3. Genito-Urinary System. Vagina. The vagina has an acid reac-
tion and it harbors very few bacteria, but immediately afterchild-
birth the reaction may become temporarily alkaline. The bacillus
of Doderlein, however, occurs so commonly, that it may be provision-
ally regarded as a normal inhabitant and a few strains of aciduric
cocci are not infrequently detected in cultures from the fundus of the
vagina. The Gonococcus and Treponema pallidum are the more
common pathogenic organisms whose portal of entry is the vagina.

Uterus. The normal uterus is sterile and the acid reaction of the
vagina and the closure of the cervix uteri tends to maintain sterility
under normal conditions. During menstruation and childbirth the
mechanical defenses of the uterus are impaired. The organ itself
appears to possess no specialized powers of resistance to infection.

1 A more detailed discussion on intestinal bacteria and their significance will be found
in Chapter xxx.

PARASITIC AND PATHOGENIC BACTERIA 105

Urethra. The urethra in health is practically free from bacteria.
The flow of urine mechanically frees it from bacteria. The external
orifice of the urethra, however, frequently contains an acid-fast organ-
ism, Bacillus smegmatis, which can be differentiated from the tubercle
bacilli only by animal inoculation, and, very frequently, Bacillus coli.
The gonococcus and Treponema pallidum may invade the tissues
through the urethra.

Urinary Bladder and Ureter. The slightly alkaline reaction of the
urine affords a good culture medium for many bacteria and infection
of the bladder by B. coli, B. proteus, B. typhosus and other micro-
organisms is by no means uncommon. It is probable that infection
occurs much more frequently through the blood or lymph than through
an ascending infection from the urethra. B. proteus appears to grow
with great luxuriance in the urinary bladder and a typical cystitis
may be readily incited in dogs by injecting virulent cultures of the
organism directly into the bladder. Occasionally a descending infec-
tion from an inflamed kidney may result in cystitis: whether a true
ascending infection through the ureter to the kidney takes place is
not definitely proven.

Kidneys. The kidneys are normally free from bacteria, but infec-
tion of one or both kidneys through the blood stream is a well-estab-
lished phenomenon. A variety of organisms may thus infect the
kidney. The cocci of suppuration frequently incite acute nephritis
and tubercle bacilli induce chronic infection. Theoretically, any
invasive organism which enters the blood stream may localize in the
kidney and establish metastatic foci there. The organ is susceptible
to specific bacterial toxins as well as to the bacteria themselves.

I). Where Bacteria Multiply in the Body. Practically no organ
or part of the body, except such structures as the nails, are free from
invasion with one or another kind of organism. The obvious com-
plexity of the subject makes it difficult or even impossible to present
in concrete form, a statement which shall indicate specifically the
types of organisms which incite infection in association with the
particular organs or tissues where they become localized. It is impor-
tant in this connection, however, to remember that a great majority
of progressively pathogenic bacteria exhibit rather marked affinities
for special tissues, and that they invade the tissues through definite
atria. The organisms which are habitually parasitic, on the contrary
the "opportunists" as Theobald Smith has so clearly pointed out,
are less exacting in this respect, as a rule, and they may invade the

106 SAPROPHYTISM, PARASITISM, AND PATHOGENISM

tissues whenever the natural barriers skin, mucous membranes, and
so on weaken and become vulnerable.

The following table indicates the more common and important
bacteria, parasitic or pathogenic, which may invade the tissues, and
the organs where they tend to localize and develop.

SKIN:

Staphylococcus and streptococcus groups.

Acid-fast group: tubercle bacilli, lepra bacilli, smegma bacilli.

Anaerobic group: tetanus, gas bacillus.

Anthrax.

"Bottle" bacillus (spore of Melassez).
NOSE, THROAT AND ADNEXA:

Staphylococcus group.

Streptococcus and pneumococcus group.

Diphtheria and pseudodiphtheria group.

Influenza and pertussis group.

Pneumobacillus, rhinoscleroma and ozena group.

Bacillus fusiformis and spirillum group.

Meningococcus and catarrhalis group.

Acid-fast group 'Chiefly tubercle bacilli and leprosy.

Blastomycetes and hyphomycetes.

Virus of poliomyelitis and unknown viruses, mumps, etc.

(Organisms of dental caries and pyorrhea not included above.)
EYE AND EAR:

Streptococcus and pneumococcus group.

Staphylococcus group.

Diphtheria and pseudodiphtheria group.

Influenza group.

Koch- Weeks and Morax-Axenfeld group.

Gonococcus.

Proteus group.

Pyocyaneus group.
LUNGS :

Streptococcus and pneumococcus group.

Pneumobacillus group.

Acid-fast group: tubercle bacillus.

Influenza and pertussis group.

Plague bacillus, anthrax bacillus and B. psittacosis.

Colon and typhoid group.

Actinomyces and hyphomycetes.
PELVIC ORGANS:

Streptococcus and Staphylococcus group.

Gonococcus and Treponema pallidum.

Tubercle bacillus and smegma bacillus.

Micrococcus melitensis.
SEROUS FLUIDS:

1. Cerebrospinal fluid:

(a) Fluid usually clear: tubercle bacillus and Treponema pallidum. Virus
of poliomyelitis.

(&) Fluid turbid: Pneumococcus, streptococcus, meningococcus, B. influ-
enza, B. typhosus, B. coli.

2. Pleural and pericardial fluids:

(a) Fluid usually clear: tubercle bacillus.

(&) Fluid turbid as a rule: pneumococcus, streptococcus, B. influenzce,
Pneumobacillus group, Bacillus typhosus, Staphylococcus.

3. Peritoneal fluid:

Streptococcus group.
Coli and typhoid group.
Tubercle bacillus (?).

BALANCED PATHOGENISM; EPIDEMIOLOGY 107

BLOOD:

Streptococcus and pneumococcus group.

Staphylococcus group.

Typhoid, paratyphoid and dysentery group.

B. coli.

Recurrent fever and treponemata.

B. pestis.

Certain filterable viruses: Yellow fever, poliomyelitis (?).

Tubercle bacillus (occasionally).
INTESTINAL CONTENTS, FECES:

B. bifidus and B. acidophilus group (chiefly in infants).

B. coli, B. lactis aerogenes, proteus and cloacae group.

Alcaligenes, paratyphoid, typhoid and dysentery group.

Streptococcus and Micrococcus ovalis groups.

Mucosus capsulatus group.

Spore-forming group: Aerobic B. mesentericus, B. subtilis, B. anthracis.

Anaerobic B. aerogenes capsulatus, B. tetani, B. botulinus.

Acid-fast group: tubercle bacilli, bovine and human; grass bacilli.

Spiral group: Vibrio cholerse, Sp. of Finkler and Prior.

E. Where and How Bacteria Escape from the Body. It appears
from foregoing considerations that those microorganisms which are
progressively pathogenic for man habitually invade the tissues through
atria characteristic for each microbe. Their escape from the tissues
through appropriate channels in direct communication with the out-
side is equally important. Bacteria of the "opportunist" type fre-
quently perish within the tissues because they lack a perfected
mechanism of escape to the outside. Progressively pathogenic bac-
teria leave the body through two principal avenues the mouth and
nose, and the feces. Less commonly, certain types may pass to the
outside in the urine. The skin is not a very important factor in the
elimination of pathogenic bacteria. The paths of pathogenic bacteria
from the tissues to the outside are varied, but very constant for each
special organism and the discussion of this phase of their activity is
reserved for the Section on Specific Organisms.

VI. BALANCED PATHOGENISM; EPIDEMIOLOGY.

It has been helpful, for clearness and discussion, to distinguish
rather sharply Setween parasitic and pathogenic bacteria and in a
majority of specific instances such a differentiation can be readily
established. There is no hard and fast line of demarcation, however,
between organisms of the "opportunist" type and those progressively
pathogenic, for it is undoubtedly true that some "opportunists" may
exhibit epidemic tendencies for limited periods if a combination of
conditions arise which favor the distribution of the organisms and
either increase the invasive powers of the microbe or decrease the
resistance of the host. The limited spread of such bacteria is far more

108 SAPROPHYTISM, PARASITISM, AND PATPIOGENISM

frequently attributable to unusually direct transfer of organisms by
a common vehicle through a series of susceptible hosts than to the
escape of the microbes from one host to another. Thus, milk-borne
epidemics of septic sore throat may be extensive and involve many
patients, but secondary transfer from man to man is relatively uncom-
mon. These bacteria have not, as a rule, perfected their mechanism
of escape from the tissues of one host to those of another. The
epidemics are usually of brief duration and it is probable that the
surviving microbes return to their original parasitic state.

Of far greater importance is a probable tendency of many progres-
sively pathogenic bacteria to act more and more on the defensive;
to gradually disembarrass themselves, on the one hand, of the offen-
sive weapons which originally conferred upon their possessors the
ability to invade their host, and, on the other hand, to perfect what-
ever defensive weapons they may have possessed the rudiments of. 1
Such a change, as Theobald Smith has pointed out, would be difficult
to detect, because an elimination of the more aggressive type and its
gradual replacement by a strain in which the defensive elements were
more prominently represented would require years for its accomplish-
ment. Such a change in the activities of the microorganisms would
probably be accompanied by reciprocal activities of the host, so that
eventually a strain of microorganisms would be evolved which had
reached a state of relative equilibrium with the host. Unusually
virulent strains of microbes would tend to perish with their hosts,
and unusually susceptible hosts would tend to perish with their
invaders. A mutual adjustment of virulence and resistance between
the surviving hosts and microbes would lead eventually to one of
three conditions:

1. Gradual extinction of the microorganism;

2. The gradual assumption^ of a parasitic or " opportunist " exist-
ence, or

3. A more perfect pathogenism in which the mechanism of invasion,
multiplication within the tissues and escape to other hosts is accom-
plished without acute damage to the host.

It might well happen that the introduction of such "balanced"
strains into new fields would lead to temporary disaster, as for
example, the highly fatal epidemic of measles when this virus first
gained a foothold in the South Sea Islands.

1 Theobald Smith (Some Problems in the Life History of Pathogenic Microorganisms,
Am. Med., 1904, viii, 711) clearly stated and discussed this hypothesis over a decade
ago, and it is surprising how little cognizance has been taken of it.

BALANCED PATHOGENISM; EPIDEMIOLOGY 109

Theobald Smith 1 has mentioned the diphtheria bacillus as an
organism which possibly exhibits a tendency toward a parasitic exis-
tence. The toxin of the diphtheria bacillus is not a poison specific
for man; many animals, as the horse and guinea-pig, are very sus-
ceptible to it. Yet the diphtheria bacillus is almost obligately a human
pathogen. The ever-increasing occurrence of avirulent, non-toxin
producing strains which are otherwise perfectly typical, and the
frequent occurrence of individuals whose serum contains small
amounts of natural antitoxin might be interpreted as an indication
that strains of this organism are becoming gradually accustomed to
a purely parasitic existence in the upper respiratory tract of man on
the one hand, and that man has acquired some specific resistance to
the microbe on the other hand.

The tubercle bacillus (typus humanus) is an excellent example of
an exquisitely balanced pathogenic microorganism. Its metabolism is
not markedly different from that of the host and the typical disease
excited by it is focal, chronic, and slow-going. Years may elapse before
the host finally succumbs. The development of the organisms within
the tissues of the host does not appear to lead to the formation of
substances which arouse the latent offensive and defensive mechanism
of the host to acute antagonism. During this long period the tubercle
bacilli establish communication with the outside and, in a majority
of cases, countless myriads of bacilli escape from the host before
death removes him as a source of infection. Occasionally tubercle
bacilli become widely disseminated in the body, causing rapidly fatal,
generalized miliary tuberculosis. These organisms perish with their
host.

It is well known that the virulence of bacteria, many kinds at least,
can be increased decidedly by passage from animal to animal by
providing an artificial portal of entry and of exit from animal to
animal. This is accomplished by injecting the organisms into a
first animal and reinjecting them, at brief intervals, into other animals.
In such instances there is a direct continuity of growth from animal
to animal, greater than is met with in naturally occurring infections.
It is worthy of note that bacteria of the "opportunist" type are, gen-
erally speaking, more successfully exalted in virulence under these
conditions than the progressively pathogenic forms.

There is yet another feature of Pathogenism which is worthy of
note. From time to time almost any bacterial disease, for example,

1 Theobald Smith, loc. cit.

110 SAPROPHYTISM, PARASITISM, AND PATHOGENISM

typhoid, plague or influenza, may leap suddenly to epidemic propor-
tions, spread rapidly and then subside again, to be succeeded by
sporadic cases which gradually diminish in numbers and in severity.
The bacteria causing these outbreaks appear to acquire somehow and
somewhere, an unusual degree of invasiveness and they spread rapidly,
especially in thickly settled areas, and as rapidly lose their unusual
activities and subside to what appears to be their usual level of viru-
lence. It is very probable that those strains of pathogenic bacteria
in general, which suddenly acquire unusual virulence are short-lived,
partly because their hosts perish before the microbes can escape to
new hosts. Not infrequently, these or similar epidemics are preceded
by mild, atypical disease, which may not be specifically recognized,
and during this initial period the bacteria may be quite widely dis-
tributed. 1

1 Kendall, Boston Med. and Surg. Jour., 1915, clxxii, 851.

CHAPTER VI.
IMMUNITY AND INFECTION.

2. Passive Immunity.

(a) Antibody Im-
munity.
(6) Chemotherapy.

3. Mixed, Active and

Passive Immunity.
II. INFECTION PRIMARY AND SEC-
munity. . ONDARY.

1. Active Immunity. A. Defenses of the Host, Non-

(a) Natural Ac- specific and Specific.

GENERAL PHENOMENA OF IMMUNITY.
I. CLASSIFICATION OF IMMUNITY.

A. Natural or Inherited Im-

munity.

1. Racial.

2. Individual.

B. Acquired or Induced Im-

quired Im-
munity.
(6) Artificial Ac-

Ill. THEORIES OF IMMUNITY.

A. The Humoral Side-chain or
Ehrlich Theory of Im-

quired Im- munity.

munity.

B. The MetchnikorT or Phago-
cytic Theory of Im-
munity.

GENERAL PHENOMENA OF IMMUNITY.

IT has long been recognized that man and animals exhibit refrac-
toriness to infection with specific bacteria or other microorganisms
which cause serious epizootics in closely related animals. Man is,
as a rule, quite free from the epizootic diseases of animals domesticated
by him, and the domestic animals are usually not infected with the
organisms which incite progressive disease in man. Thus, man does
not contract chicken cholera and domestic animals do not become
infected with the typhoid bacillus. Furthermore, closely related
animal species may exhibit striking differences in susceptibility to
the same disease; for example, field mice are readily infected with
the glanders bacillus, but house mice are quite resistant to infection
with this organism, and ordinary sheep readily succumb to anthrax
although Algerian sheep are practically immune to infection with the
anthrax bacillus.

This inherent or congenital resistance or refractoriness to infection
with a specific microorganism, when general among the individuals of
a species or group of animals or of man is termed natural or inherited
immunity. It is not necessarily absolute; lowering the natural resis-
tance of the individual may render him susceptible to infection. Thus

112 IMMUNITY AND INFECTION

hunger, experimental (phloridzin) diabetes, fatigue produced by pro-
longed exercise in treadmills, and excessive chilling by the removal
of hair have been shown to decrease resistance in experimental animals.

It is also a matter of common observation that the uniform exposure
of a number of theoretically susceptible individuals of the same species
to a virus does not lead to uniform infection; a certain small number
usually perish, a larger proportion become mildly or severely ill and
recover. The greatest number are not especially affected, as a rule.
Those individuals who escape infection in one epidemic may succumb
to infection during a subsequent epidemic of the same disease. This
phenomenon of individual variation in susceptibility is well exempli-
fied in water- and milk-borne epidemics of typhoid fever where typhoid
bacilli are widely distributed in a water or milk supply. A small
number become infected, but the greater number escape the disease.
The incidence of scarlet fever, of diphtheria or of other infectious
diseases among the members of the same family frequently illustrates
this same phenomenon. This resistance to infection exhibited by
certain individuals of a susceptible species is termed inherited
immunity.

Susceptible individuals who survive a naturally acquired or arti-
ficially induced infection as smallpox, measles, typhoid fever or
vaccinia are frequently resistant or refractory to subsequent infec-
tion with the same virus. They have developed a resistance to specific
infection, they have acquired immunity, in other words. This type
of immunity, which results from actual infection, is termed active,
acquired immunity. It is the outcome of a successful struggle between
the host and the invading microbe during which the former, through
cellular activity, produces or increases antibodies specifically inimical
to the latter. The immunity which is thus laboriously produced
is frequently fairly persistent. It is more commonly observed fol-
lowing invasion by exogenous, progressively pathogenic bacteria than
infection with endogenous microorganisms of the "opportunist" type.
Indeed, infection with the latter not infrequently results in increased
susceptibility to subsequent infection with the same species of microbe.
Thus, recovery from one attack of typhoid fever usually confers last-
ing immunity upon the individual; one attack of lobar pneumonia,
on the other hand, appears to predispose the individual to subsequent
infection with the pneumococcus.

The injection of specific immune substances or antibodies into
susceptible individuals may confer upon them transient or temporary

CLASSIFICATION OF IMMUNITY 113

immunity to the specific infection; the host is a passive recipient of
antibodies in such instances. These alien antibodies, however, soon
diminish in potency or disappear, leaving the susceptibility of the indi-
vidual to infection at its original level. Immunity induced by the
injection of specific antibodies is termed passive acquired immunity.
The transitory immunity to diphtheria or tetanus following the
injection of diphtheria or tetanus antitoxin is an example of passive
acquired immunity.

Immunity may be localized or general in the same individual, and
different individuals frequently exhibit varying degrees of resistance
or susceptibility to the same virus.

I. CLASSIFICATION OF IMMUNITY.

Both immunity and susceptibility are relative; there is probably
neither absolute immunity nor complete susceptibility to any infec-
tion. There is furthermore, no hard and sharp line of demarcation
between the various types of immunity; nevertheless, it is convenient
to assemble the prominent manifestations of immunity into several
types or classes.

A. Natural or Inherited Immunity. The inherited power of resist-
ing specific infection manifested by a large proportion of the individuals
comprising a family, genus or species is termed inherited or natural
immunity. It may be:

1. Racial. Observed in specific families, genera or species of the
animal kingdom, or

2. Individual. Observed in individuals of the same species. Indi-
vidual natural immunity may also be sexual observed in males or
females of the same species.

B. Acquired or Induced Immunity. The resistance or non-sus-
ceptibility to infection following naturally acquired or artificially
induced specific diseases, or resistance passively brought about by the
introduction of specific protective substances is termed acquired or
induced immunity.

1. Active Immunity. (a) Natural. Following naturally acquired
disease, as for example, immunity following recovery from smallpox
or typhoid fever.

(b) Artificial. Brought about by the introduction of attenuated
or killed viruses, vaccines or toxic products of bacteria into a sus-
ceptible host. The toxic products of bacteria may be either those
8

114 IMMUNITY AND INFECTION

excreted during life, or products arising from their disintegration.
Immunity to smallpox following vaccination and immunity to typhoid
fever following the injection of killed cultures of typhoid bacilli are
familiar examples of this type of immunity.

There is usually a period of increased susceptibility to infection
immediately after the introduction of the virus or its products, in
artificially acquired immunity. This period of susceptibility is fol-
lowed by an increase in resistance to the virus. If the process of
immunization is repeated several times, the initial level of resistance
to infection may be raised very materially. Thus, prophylactic vac-
cination with killed typhoid bacilli (anti-typhoid vaccination) increases
the resistance of the recipient of the vaccine to typhoid infection to
such a degree that his chances of acquiring the disease are greatly
lessened. It is also probable that in the event of infection of the
protected individual with the typhoid bacillus, both the duration and
severity of the attack will be diminished.

2. Passive Immunity. (a) Antibody Immunity. Introduction into
the host of specific products of immunity (antibodies) as diphtheria
antitoxin.

(6) Chemotherapy. The use of chemicals for preventing or modifying
infection.

Passive immunity is induced by the injection of antibodies into
the host, which have been developed in another animal. The recipient
of these antibodies is protected only so long as they remain in the
body. The immunity, however, is effective almost immediately after
injection; there is no latent period.

3. Mixed, Active and Passive Immunity. Mixed artificially acquired
immunity is induced by the simultaneous injection of specific anti-
bodies and the weakened or attenuated virus; resistance to infection
is usually increased at once (passive immunity), while at the same
time the host begins to react to the virus and to produce antibodies
thereto (artificially acquired immunity).

The factors which predispose the host to or protect him from inva-
sion by microorganisms are usually varied and complex. Relatively
simple explanations of the mechanism involved suffice to account for
the phenomenon in specific instances, however. For example, frogs
and hens are not naturally susceptible to infection with the anthrax
bacillus, whose optimum temperature of growth is 37 C., yet infection
could take place if the body temperature of either animal were brought
to this level, as Pasteur showed nearly two decades ago. A change

INFECTION 115

in environment may predispose to infection; the carnivora in their
native state are quite resistant to infection with the tubercle bacillus,
whereas in captivity they may succumb readily. Similarly, man
placed in bad hygienic surroundings appears to be distinctly more
vulnerable to many infectious diseases than he is when his environ-
ment is more sanitary. Unhygienic conditions, however, are rela-
tively complex in their reactions on man, for the attendant evils of
overcrowding, underfeeding and increased exposure to infection
undoubtedly play a part.

Heredity also appears to be an important factor in determining the
average severity of infection in certain types of endemic disease.
Measles is a common and usually fairly mild disease of childhood
among civilized people. Among aboriginal populations, as those of
the South Sea Islands where the inhabitants had not been exposed
to measles previous to the advent of Europeans, the introduction
of the virus has resulted in a veritable plague during which large
numbers of the people died. This phenomenon of hereditary acquired
tolerance for specific endemic disease may conceivably be even more
specific; for example, strains of a given organism might produce
mild disease in areas where it has been endemic for generations and
yet be rapidly fatal for alien populations who may have in turn become
partly tolerant for other strains of the same organism. If such prove
to be the case, unrestricted emigration may lead to temporary dis-
turbances in the balance between specific microorganisms on the one
hand and hosts on the other a feature which Theobald Smith called
attention to many years ago. 1

Racial differences in susceptibility are occasionally met with even
in the same species. Negroes and Indians are more susceptible to
infection with tubercle bacillus than the Caucasian race. The Jews
appear to be somewhat more resistant to infection with the tubercle
bacillus than the other branches of the Caucasian race.

H. INFECTION.

Pathogenic bacteria which reach the host do not necessarily incite
disease; they may be, and undoubtedly are, frequently overcome by
the body without inducing symptoms. This initial resistance to infec-
tion involves an initial struggle between host and microorganism which
brings into play non-specific lines of defense of the macroorganism

1 Theobald Smith, Tr. Assn. Am. Phys., 1893.

116 IMMUNITY AND INFECTION

consisting collectively of the skin, mucous membranes of the respira-
tory and gastro-intestinal tracts and other intact barriers discussed
in the preceding chapter. If this initial line of defense holds, the host
overcomes the prospective invader and the latter frequently perishes.
Repeated microbic assaults may be successful if the first fails. On
the other hand, if the microbe prevails and penetrates the initial line
of defense, invasion of the tissues of the host occurs and the micro-
organism encounters a second line of defense which is made up of
two rather distinct factors cellular and humoral. The cellular
defense of the host resides in the leukocytes which circulate in the
the body fluids and in certain fixed tissue cells in the lungs, lymph-
spaces and glands, the Kupfer cells of the liver, as well as large cells
which appear in serous cavities. These cells engulf and destroy
certain types of invading microorganisms. The humoral defense
resides in the natural, non-specific power of the blood and lymph to
destroy limited numbers of microorganisms or to so interfere with
their nutrition or other functions as to prevent their development
within the body. The humoral defense is frequently effective against
bacteria which do not succumb to the cellular defense of the body,
and vice versa.

It is recognized that certain environmental factors predispose to
infection. Thus, extreme climates, excessive humidity, or exposure
to unhygienic conditions, bad air, poor or insufficient food, lack of
exercise or fatigue may react upon the individual in ways not defi-
nitely understood and reduce his resistance to microbic invasion on
the one hand, and his ability to rally his specific, anti-microbic
mechanism on the other hand. Intracurrent disease frequently
weakens the initial lines of defense, permitting bacteria of the " oppor-
tunist" type to become invasive. Thus, furunculosis frequently is
a complication of diabetes, pneumonia not uncommonly terminates
a case of tuberculosis. Renal and cardiac disease may weaken the
normal barriers of the body, permitting a variety of infections with
endogenous bacteria.

It is a well-attested fact that certain occupations or professions
cause or promote pathological conditions which predispose to infec-
tion. Prominent among these is participation in arts or industries
which involve exposure to poisonous or irritating dust or fumes. The
incidence of tuberculosis among those frequently exposed to organic
or inorganic dust is a striking example of the relation of occupation
to infection.

THEORIES OF IMMUNITY 117

When an invading microorganism has reached a suitable atrium of
the body, overcome the initial defense of the host at that point, and
has successfully resisted the normal humoral or cellular opposition
of the host, a new phase of the struggle becomes prominent, during
which the host gradually develops a specific attack upon the invader,
bringing into action latent forces which constitute the third and last
defense of the body. The invader also may change its weapons to
some degree to meet the antimicrobic activity of the host and the
result of the struggle may be complete recovery from infection, chronic
disease, the bacillus carrier state, or death of the host.

The initial and secondary defensive powers of the host, therefore,
are both cellular and humoral in character. The intact skin and
mucous membranes of the gastro-intestinal, respiratory and genito-
urinary tracts are important initial non-specific lines of defense. The
phagocytic activity of leukocytes and certain fixed tissue cells, and
the natural, normal bactericidal substances of the blood and lymph,
which bathe the initial line of defense, are important adjuvants in
maintaining the integrity of these initial barriers to infection. In
certain infections the humoral factors are the more important, while
in others the cellular mechanism is conspicuous.

The defensive mechanism against the same bacterium may be
different in one or another animal. For example, dogs and rats are
relatively immune to infection with the anthrax bacillus. The immu-
nity observed in the dog appears to be due to phagocytic activity of
leukocytes which engulf and destroy anthrax bacilli which may have
gained entrance to the body. 1 The rat, on the contrary, enjoys immu-
nity not because its leukocytes engulf and destroy anthrax bacilli;
the blood of the rat possesses soluble, non-specific bactericidal sub-
stances which destroy anthrax bacilli. Frequently both the cellular
and humoral elements are engaged either simultaneously or succes-
sively as the struggle between host and invading organism proceeds.

m. THEORIES OF IMMUNITY.

Two distinct explanations have been advanced to account for the
mechanism of immunity as it is observed during the course of disease :
the cellular or phagocytic theory championed by Metchnikoff and
his followers, and the humoral theory developed by Ehrlich.

Both of these theories, the cellular and the humoral, have in com-

1 Hektoen, Jour. Am. Med. Assn., 1906, xlvi, 1407.

118 IMMUNITY AND INFECTION

mon, tacitly at least, two important features:, the specificity of the
protective substances (antibodies) formed as the result of infection,
and the principle that no new mechanism is evolved de novo to meet
the conditions existing during an infection; rather, there is an increase
in activity along definite lines in the preexistent, latent or reserve
mechanism of defense.

Neither theory affords a satisfactory explanation of all the features
of immunity following infection and it is very probable that cellular

911-

FIG. 5. Side-chains, first order (antitoxins and antif erments) . 1, side-chain attached
to cell; c,.haptophore group; 2, side-chain to which is attached a toxin molecule; 3,
a cast off side-chain of the first order: antitoxin or antif erment ; 4, a toxin or enzyme
molecule; a, toxophore group; 6, haptophore group; 5, a toxoid: the toxophore group
is destroyed, leaving the haptophore group (6) intact; 6, a toxin molecule attached to
a cast-off side-chain (antitoxin), illustrating the neutralization of toxin by antitoxin in
the blood stream.

activity and the production of specific antibodies is more important
in certain types of infection, while phagocytic activities are more
intimately concerned in other types.

A. The Humoral, Side-chain or Ehrlich Theory of Immunity.
According to Ehrlich's conception, every cell of the body has two
functions: a physiological function, which constitutes a special type
yt/bf activity of the cell secretory for a glandular cell, contractile for
Vv a muscle cell, or conductive for a nerve cell and a nutritional func-
tion, which is concerned with the removal of the necessary food sub-
stances from the general supply circulating in the blood or lymph

THEORIES OF IMMUNITY

119

channels, and the appropriation and eventual utilization of these
specific food materials by the cell. These nutritional substances
undoubtedly serve two purposes: Structural, to replace cellular
waste, and Fuel, to supply cellular energy.

The nutritional requirements of the individual cell are varied as
their activities are varied, and Ehrlich conceives that each cell
possesses a number of chemical affinities or receptors, for convenience
of discussion designated as "side-chains" or "haptines," which are

FIG. 6. Side-chains, second order (agglutinins and precipitins) . 1, side-chain attached
to cell; c, haptophore group; b, zymophore group (agglutinophore or precipitinophore
group); 2, side-chain to which is attached a bacterial cell; a, haptophore group of
bacterial cell; 3, a cast-off side-chain of the second order, agglutinin or precipitin; 4,
a side-chain attached to a bacterial cell (agglutination); 5, a bacterial cell; a, hapto-
phore group; 6, an agglutinoid; the zymophore group is destroyed, leaving the hapto-
phore group intact.

the means of attaching to the cell by chemical union, the essential
nutritive substances preparatory to their assimilation. When the
particular food attached to the cell by chemical affinity anchored
by the side-chain, to use Ehrlich's terminology has been assimilated,
more of the same kind of food is removed from the blood stream and
attached to the cell, in accordance with its normal physiological
requirement. The cell, acting through its side-chain, does not exhibit

120 IMMUNITY AND INFECTION

discrimination between nutritive substances and irritating or harmful
substances which may accidentally possess the same combining
affinity for the cell. Consequently, when poisonous substances pos-
sessing chemical affinities similar to those of the normal food sub-
stances circulate in the blood stream, they may become attached to
the cell in place of the normal physiological nutrients. The anchoring
of these poisonous substances, unlike the attachment of normal
nutrient substances, is followed by damage to the cell, or, in extreme
cases, by the death of the cell. 1 If the cell is not actually killed by the
presence of the toxic substance acting upon it through the side-chains,
it is irritated, as it were, and the toxic substance imposes a twofold
burden upon the cell loss of the side-chains to which it is attached
and which are essential to maintain the nutrition of the cell, and
greater or lesser damage to the function of the cell, due to toxic inhibi-
tion of its normal activities. A cell cannot disembarrass itself of the
poison, nor can it assimilate it. It can, however, throw off the side-
chain with the poison still firmly united to it chemically; the extruded
poison cannot enter into chemical combination with other cells pos-
sessing the same chemical affinity, for it is already attached to a side-
chain. Its combining power is saturated.

Side-chains are a necessity to the cell, however; without them
the cell would starve. Consequently the cell regenerates new side-
chains of precisely the same kind to replace those thrown off after
being bound to non-assimilable substances. If enough of the soluble
poison or toxin circulates in the blood stream, this process of union
of toxin to the cell by its side-chains and its expulsion from the cell
with the side-chains attached to it is so frequently repeated that the
cell regenerates side-chains in excess of the normal requirements, in
accordance with the Weigert theory of overproduction. This casting
off of supernumerary side-chains is important. Were they not cast
off the cell would be vulnerable to toxin in direct proportion to the
extra number of side-chains, which would furnish extra bonds for its
attachment. As the cast-off side-chains circulate in the blood stream,
however, they are an element of protection to the cell, for they retain
their original combining power for the toxin and unite with it and
neutralize it as it circulates in the blood stream; that is, before it
can reach the cell itself. It will be seen, therefore, that the same

1 If the toxic material circulates in the blood stream but does not become attached
to the body cells, it is harmless to the host, according to this theory, and the host is
naturally immune.

THEORIES OF IMMUNITY

121

mechanism of the living body which is susceptible of being poisoned
becomes the protective agent if it circulates in the blood stream. It
is obvious that the cast-off side-chains constitute antitoxin. The
body as a whole is qualitatively the same after as before these side-
chains are formed in excess of the normal cellular needs; the difference
is a quantitative one. An animal is naturally immune, according to

FIG. 7. Side-chains, third order (bacteriolysins, hemolysins and cytolysins). 1,
side-chain attached to cell; c, haptophore group; b, complementophile group; 2, side-
chain to which is attached a bacterial cell (6) and complement (5) ; 3, a cast-off side- chain
of the third order; amboceptor; 4, a cast-off side-chain to which are attached a bacterial
cell (6) and complement (5) illustrating lysis; 5, complement.

this theory, if the cells of the body do not unite with toxin, that is,
if they do not contain side-chains which fit the toxin " as a key fits a
lock," to use Emil Fischer's analogy. Toxin may circulate in the
blood stream of such animals, but it does not unite with the cells.

Side-chains of the First Order. From the standpoint of the side-
chain theory, the toxin molecules consist of two groups-^a combining

122 IMMUNITY AND INFECTION

or haptophore group, and a poisoning or toxophore group. The former
is relatively thermostabile, the latter thermolabile. If toxin is heated
to 70 C. for a few minutes, or allowed to stand for several weeks, it
will be found that the poisonous property of the toxin has disappeared,
or has been materially reduced. It still retains its original powder of
uniting with and neutralizing antitoxin, however. The thermolabile
toxophore group has been destroyed or weakened by the heating
process, or on standing. The thermostabile group the haptophore
group has not been impaired. Toxin which has lost part or all of
its original poisoning properties, but which still unites with antitoxin
is called toxoid.

The soluble toxins of the diphtheria and tetanus bacilli are not
simple substances; they contain at least two physiologically separate
poisons. Thus, the toxin of the diphtheria bacillus contains in addi-
tion to the poison which produces acute symptoms, a second poison
which acts slowly and appears to be responsible for postdiphtheritic
paralyses and emaciation. This second poison has less affinity for
antitoxin than the acute poison, and it is called a toxone. Similarly,
the tetanus toxin appears to consist of at least two distinct poisons
tetanospasmin, which has an especial affinity for nerve cells and
which elicits the acute symptoms of tetanus, and tetanolysin, which
causes hemolysis of erythrocytes. The injection of soluble or exo-
toxins produced by bacteria leads to the formation of soluble specific
antibodies which are called antitoxins. Antitoxins are supernumerary
side-chains which have been produced in excess of the physiological
needs of the cell, in response to the stimulus of a specific toxin, and
cast off into the blood stream.

It has been shown that repeated injections of solutions containing
active enzymes as, for example, rennin into animals, is followed
by the appearance in the blood stream of specific antibodies which
will prevent the activity of the homologous enzyme. These anti-
bodies, or anti-enzymes, as they are called, exhibit the specificity and
other characteristics which distinguish antitoxins.

Antitoxins and anti-enzymes are called side-chains of the first
order by Ehrlich. They possess the property of combining with and
neutralizing their respective toxins or enzymes.

Side-chains of the Second Order. If substances of greater com-
plexity than those just described are needed for the nutrition of the
cell, some preliminary treatment, probably in the nature of digestion,
may be required to prepare these substances for assimilation after

THEORIES OF IMMUNITY 123

they are bound to the cell. A side-chain of the first order, which
possesses simply a combining group, does not provide the requisite
power of digestion, according to Ehrlich, and to effect this digestion
side-chains of somewhat more complex structure are required. Side-
chains of this more complex type, side-chains of the second order,
possess not only a combining group for the foodstuff, but a digestive
group as well. This digestive or zymophore group, as it is called, acts
upon foodstuffs after they are anchored to the cell by the combining
or haptophore group. The complete side-chain of the second order,
therefore, is composed essentially of a combining or haptophore group,
and a zymophore group as well. The haptophore group of the second
order side-chain is relatively stabile, but the zymophore group is
labile and readily becomes inactive without, however, impairing the
original combining ability of the side-chain. Side-chains of the second
order are as vulnerable to pathological substances possessing the
requisite chemical affinity as side-chains of the first order, and repeated
irritation of a cell by such pathological substances leads eventually
to an overproduction of side-chains of the second order and an
elimination of the supernumerary side-chains in excess of the physio-
logical need of the cell into the blood stream. Side-chains of the
second order which are thus cast off from the cell in response to the
stimulation of bacterial or other alien protein are of importance
immunologically. If the serum of an animal containing such side-
chains is brought into contact with a suspension of the homologous
bacterium, the organisms are sooner or later clumped together or
agglutinated. If, on the contrary, the serum is brought into contact
with a clear solution of the homologous protein, a precipitate forms.
These reactions are highly specific and those side-chains which cause
agglutination of the specific bacterium pr precipitation with the
homologous protein solution are called respectively, agglutinins and
precipitins.

The relative instability of a zymophore group of a side-chain of the
second order may be inferred from the following experiment:

A serum obtained by injecting a horse with repeated graduated
doses of typhoid bacilli will clump or agglutinate the specific organism
in high dilution. If the serum is heated to 60 or 70 C. for a few
minutes, or if it has been kept for a long time, it will no longer clump
the bacilli, or, at least, it will clump them imperfectly. If such a serum
is allowed to stand in contact with typhoid bacilli for an hour or two
then removed by centrifugalization, it will be found that the bacilli

124 IMMUNITY AND INFECTION

will no longer agglutinate with a fresh, highly potent agglutinating
serum. The bacteria are saturated with the combining group of the
serum whose agglutinophore group had been inactivated by heating.
This experiment shows that the combining group is relatively stabile,
and that it is active even though the zymophore group is inactive.
A side-chain of the second order which has lost its ability to cause
agglutination with a specific organism, but which still retains its
combining power, is called an agglutinoid. It bears a striking resem-
blance to a toxoid in that the active or ergophore group is destroyed,
but the combining group remains intact.

Sera containing specific precipitins readily lose their ability to form
precipitates with the homologous protein. The precipitins have
changed to precipitinoids, due to a functional loss of their precipitino-
phore group.

The part played by side chains of the second order, agglutinins and
precipitins, in immunity is not well understood. Their relation to
immunity is less clear than the relation of antitoxin to immunity.

Side-chains of the Third Order. Nutritive substances of large mole-
cular aggregation may require considerable modification to fit them
for cellular assimilation. Such substances are removed from the
blood stream and bound to the cell by side-chains of the third order.
They are then acted upon by an enzyme (complement) which is
also present in the blood stream. It will be seen that both the nutri-
tive element and a digestive enzyme circulate in the blood, but that
no reaction occurs between them until they are both united by a side-
chain of the third order, which must therefore consist essentially of
two combining groups. One of these, the cytophilic group or hapto-
phore, unites specifically with the nutritive element. The other com-
bining or haptophore group, the complementophilic group, unites
with the enzyme or complement which is present in the blood stream.

Side-chains of the third order are called amboceptors because they
possess two combining groups. An excessive irritation of a cell by a
substance capable of uniting with the cytophilic group of a side-chain
of the third order will lead to overproduction and elimination of these
side-chains precisely as toxins lead to an overproduction of side-chains
of the first order (antitoxin formation). The side-chains of the third
order, furthermore, exhibit specificity for the substance which led to
their overproduction, just as antitoxins exhibit specificity for their
homologous toxins.

It has been shown that the zymophoric group of a side-chain of the

THEORIES OF IMMUNITY 125

second order is permanently a part of the structure. The comple-
ment, which is analogous to the zymophore group of the second order,
is not attached to a side-chain of the third order until the cytophilic
group of the latter has combined with its antigen. The zymophore
group of the second order side-chain is readily destroyed and it
cannot be replaced. The zymophoric group of the third order side-
chain is not an integral part of the structure, and it can be introduced
under appropriate conditions.

Third order side-chains or amboceptors are cytolysins. Those
specific for bacteria are called bacteriolysins; those specific for blood
are called hemolysins; and those specific for the cells of various tissues
or organs are called cytolysins.

The activity of the lysins, according to the Ehrlich theory, depends
on the union of non-specific complement and a specific antigen by the
specific amboceptor. A union of antigen and amboceptor may take
place in the absence of complement, but a union of antigen and com-
plement cannot take place in the absence of amboceptor. The
amboceptor, like other haptophore groups, is relatively thermostabile.
The non-specific complement (found in fresh blood serum from any
animal) is thermolabile and readily destroyed.

Thus far it has been assumed that the cells of the body defend
themselves against toxins, alien protein or alien cells by the formation
of specific antibodies or side-chains. Welch 1 has made the important
suggestion, which has experimental evidence in its favor, that bac-
teria may also produce side-chains which are specific for certain
cells of the host. A struggle between host and microbe, therefore,
would not be one-sided; a dual attempt at immunization is going
on during a bacterial invasion, in which the microbe attempts to
protect itself against the specific weapons of the host as the host
attempts to protect itself against the weapons of the invading micro-
organism. Thus, bacteria grown in media containing agglutinating
sera gradually lose their agglutinability, but this acquired loss of
agglutinating power is not exhibited by descendants of the inagglu-
tinable strain 'grown for some time in media not containing agglutinins.

The side-chain theory, originally formulated to explain antitoxin
immunity, but enlarged in its scope to include the phenomena of
agglutination, precipitation and cytolysis, has been subjected to much
adverse criticism. It was assumed that toxin and antitoxin, for
example, united in simple proportions as a strong acid and a strong

1 Huxley Lecture, 1902.

126 IMMUNITY AND INFECTION

base unite; the chemical analogy of toxin-antitoxin union to form
an inert mixture comparable to a salt was further accentuated by the
effect of moderate degrees of heat in hastening the reaction between
the two. A very thorough investigation of the quantitative neutraliza-
tion of toxin by antitoxin revealed the error of this supposition and
Ehrlich was led to assume a very complex structure for the toxin mole-
cule, in which there existed several fractions possessing individually,
different affinity for antitoxin.

Madsen and Arrhenius 1 studied the toxin-antitoxin union from
the standpoint of physical chemistry and found that the slightly dis-
sociated reactive substances united in conformity with the law of
mass action of Guldberg and Waage. Their conclusion was that
toxin and antitoxin react like a weak acid and weak base, and that it
is a reversible reaction, so that a mixture of toxin and antitoxin always
contains free toxin, free antitoxin and toxin-antitoxin, the relative
amounts being calculable according to the law of mass action. The
observations of Theobald Smith 2 and of many other observers that
neutral mixtures of toxin and antitoxin would induce active immunity
in experimental animals are in harmony with this-view. Biltz 3 has
advanced an hypothesis, based upon the assumption that toxin and
antitoxin are colloids, which in essence assumes that the toxin-anti-
toxin reaction is a phenomenon of adsorption, quite unlike the reaction
of a weak acid and a weak base.

The humoral theory of immunity fails to attribute to phagocytic
cells any prominent part in immunity. No theory has been advanced,
up to the present time, which explains all the phenomena of humoral
immunity; whatever the final solution may be, the side-chain theory
as developed and defended by Ehrlich must, and always will be, a
worthy monument to a great man.

B. The Cellular or Phagocytic Theory of Immunity. The cellular
theory of immunity, formulated and championed by Metchnikoff,
had its inception in observations of the nutritive activities of amebse,
which could be watched under the microscope. It was observed
that these simple, transparent protozoa, when about to feed, ap-
proached and flowed around a minute organism, as a bacterial cell.
Shortly after engulfment the contour of the ingested bacterium lying
within the substance of the ameba became less and less distinct and

1 See Arrhenius, Immunochemie, Leipzig, 1907, for full details.

2 Jour. Exp. Med., 1909, xl, 241, Active Immunity Produced by So-called Balanced
or Neutral Mixture of Diphtheria Toxin and Antitoxin.

3 Ztschr. f. physiol. Chem., 1904, 615.

THEORIES OF IMMUNITY

127

finally disappeared entirely. His attention was soon directed to a
small, transparent crustacean, daphnia, within whose body cavity
could be distinguished minute wandering cells which exhibited ameboid
movements. The physiological significance of these ameboid cells
which are potentially leukocytes was not clear until it was found
that they engulfed and digested certain yeast spores that occasionally
gained entrance to the body cavity of the crustacean. If the yeast
spores were not too numerous the wandering cells flowed around and
eventually destroyed them; if, on the contrary, the number of yeast
spores was too great, the wandering cells could not remove the entire

FIG. 8. Phagocytosis of gonococcus.

number and the residual spores germinated and killed the host. It
was evident that the phagocytic activity of the ameboid cells played
a prominent part in protecting daphnia from an infection with the
yeast.

Next Metchnikoff injected anthrax bacilli into the lymphatic sac
of frogs and found again that wandering cells leukocytes engulfed
and destroyed the bacteria, thus preventing infection and death of
the frog. This line of observation was followed through an extensive
series of lower animals, mammals, and finally in man, where the
engulfment of the meningococci, gonococci, pneumococci, and staphy-
lococci by polymorphonuclear leukocytes during the course of acute
infections with these organisms afforded a striking demonstration

128 IMMUNITY AND INFECTION

of the phagocytic activity of leukocytes which circulate normally in
the blood and lymph streams. These and many other observations
and experiments led to the formulation of the phagocytic theory of
immunity. Natural immunity, according to this theory, is leukocytic
immunity that is, the natural barriers of the body, reenforced by
the activity of leukocytes in the blood and lymph streams which
bathe the intact skin, mucous membranes, etc., suffice to protect
the body against invasion by moderate numbers of bacteria or other
microorganisms. Infection of the body, according to this view, is
attributable to a failure of the leukocytic defense, or to too large
numbers of invading organisms, or both factors combined.

Metchnikoff classified phagocytic cells of the body into two groups :

1. .Macrocytes or Macrophages. Large mononuclear cells and certain
fixed tissue cells, particularly of the spleen, liver, lungs, and lymph
nodes. Macrophages are active in the removal of necrotic tissue,
injured blood cells, and similar abnormal cellular elements of the
body, and in chronic bacterial' infections, notably in tuberculosis,
leprosy, and actinomycosis. They contain a digestive enzyme
macrocytase which dissolves or digests these abnormal cells.

2. Microcytes or Microphages. Chiefly polymorphonuclear leuko-
cytes which occur in the blood stream. They engulf bacteria and
similar cells. Microcytes contain a digestive enzyme microcytase
which dissolves or digests bacteria.

The substance which Ehrlich regards as complement is normally
present in the leukocytes as macro- and microcytase, according to
Metchnikoff. These cytases are liberated into the blood stream when
the leukocytes are destroyed (phagoly sis) .

The phenomenon of phagocytosis may be divided into three separate
and distinct phases: the method of approach of the phagocytic cell
to its prey (chemotaxis), the engulf ment, and finally the digestion or
destruction of the latter.

The Method of Approach. It was a matter of observation by Metch-
nikoff and his followers that phagocytosis was more marked in mild
bacterial infections and during recovery than in severe infections and
the early acute stages of the disease. The importance of chemotaxis
as the attractive force of leukocytes to bacteria, however, was not
clearly realized until Massart and Bordet 1 showed by ingenious
experiments that non- virulent bacteria apparently secrete substances

1 Ann. Inst. Past., 1891, v, 417.

THEORIES OF IMMUNITY 129

which draw phagocytic cells to "them. 1 Virulent organisms of the
same strain not only do not appear to attract leukocytes, but they
appear to repel them. Bordet explained the increase of virulence of
bacteria through passage in experimental animals on the ground that
the less virulent individuals were engulfed and killed; the more viru-
lent members survived and produced a thoroughly virulent strain.
Yaillard and Vincent 2 and Vaillard and Rouget 3 showed that bacterial
toxins may repel or paralyze leukocytic activity; if tetanus spores
are bathed with tetanus toxin before injection into the animal body,
the leukocytes do not collect at the point of injection, the spores ger-
minate and the animal dies of tetanus. If, however, the spores are
washed free from tetanus toxin and then injected, leukocytes appear
at the site of inoculation, engulf the spores, and either destroy them
or prevent their germination.

The mechanism of chemotaxis has been a subject of much discus-
sion. Evidence is accumulating which would suggest that chemo-
tactic stimuli of bacterial origin which reach leukocytes enter the
phagocytic cell in greater concentration on that side which is nearer
the source of the chemotactic substance, lowering the surface tension
at that point. A flow of protoplasm in this direction, in obedience
to the lowered resistance, will result in the protrusion of a pseudo-
podium, which will continue to advance until the surface tension is
equalized. 4 This generally occurs when the leukocyte has flowed
around or engulfed the organism.

Engulfment. The earlier view associated the protrusion of pseudo-
podia and the subsequent engulfment of bacteria or other cell as an
auto voluntary act of the leukocyte. The inclusion of inert particles,
as dust or other minutely comminuted granules, would appear to
discredit this hypothesis. The engulfment of living or inert bacteria
or other minute bodies is, as Wells aptly expresses it, 5 " but an exten-
sion of the phenomena of chemotaxis. When the substance toward
which the leukocyte is drawn is small enough, the leukocyte simply
continues its motion until it has flowed entirely about the particle."

Digestion. The ultimate solution of engulfed substances other
than purely inert particles is by intracellular enzymes contained within

1 Inert particles, as coal dust, are engulfed by phagocytic cells; it is difficult to explain
this phenomenon on the basis of chemotaxis.

2 La semaine medicale, 1891, xi, No. 5.

3 Ann. Inst. Past., 1892, vi, No. 6.

4 See Well's Chemical Pathology, 1914, 2d ed., pp. 230-251 (Saunders & Co.), for
an excellent resume of the literature.

6 Well's Chemical Pathology, 1914, 2d ed., p. 238 (Saunders & Co.).
9

130 IMMUNITY AND INFECTION

the phagocytic cells. These enzymes are of two kinds: macrocytase,
present in the macrophages, and microcytase, found in the micro-
phages. 1 Van de Velde, 2 Buchner, 3 Hahn, 4 and Bordet 5 have demon-
strated such endo-enzymes. The solution of bacteria engulfed in leuko-
cytes can be shown by appropriate staining methods; the organisms
gradually lose their ability to take up stain and eventually disappear.

At this stage of the development of the phagocytic theory of immu-
nity, the important part played by the blood serum in preparing bac-
teria for phagocytosis was prominently set forth in the investigations
of Wright and Douglas, 6 although foreshadowed by the excellent and
comprehensive observations of Denys and LeClef 7 and Neufeld and
Rimpau. 8 Wright and Douglas showed that leukocytes, freed care-
fully from adherent serum by washing with salt solution, would not
engulf bacteria, or, at least, but slowly. The addition of serum from
a normal or immunized animal caused active phagocytosis to take
place. The substances in the blood serum which prepare bacteria
for engulfment by leukocytes were called "opsonins" by Wright and
Douglas: the immune opsonins which are specifically increased in
immunized animals are almost certainly identical with the substances
called bacterial tropins by Neufeld and Rimpau. That the opsonic
substances of the serum act primarily upon the bacteria rather than
upon the leukocytes was clearly shown by the observations of Hektoen
and Reudiger. 9 Streptococci suspended in plasma, blood serum or
defibrinated blood were engulfed by leukocytes. Leukocytes, washed
free from serum or plasma, were without phagocytic action upon the
same bacteria. If the streptococci, however, were allowed to stand
in contact with serum, plasma, or defibrinated blood for a short time
at 37 (a much longer exposure at to 4 C. was necessary), then
washed free from adherent serum or plasma, and exposed to washed
leukocytes, active phagocytosis took place.

The present tendency is to ascribe to phagocytosis an important
part both in the destruction of many kinds of invading bacteria and
in the removal of alien or abnormal cells as well. The importance

1 For a detailed discussion of leukocytic enzymes, see Opie, Jour. Exp. Med., 1905,
viii, 410.

2 La Cellule, x, 2; Cent. f. Bakt., 1898, xxxiii, 692.

3 Miinchen. med. Wchnschr., 1894, 718.

4 Arch. f. Hyg., 1895, xxviii, 312.
8 Ann. Inst. Past., 1895, ix, 398.

6 See Wright, Studies in Immunization, 1909, Constable.

7 La Cellule, 1895, xi.

8 Deutsch. med. Wchnschr., 1904, 1458.

9 Jour. Inf. Dis., January, 1905, ii, No. 1.

THEORIES OF IMMUNITY 131

of substances contained within the plasma or blood serum, which
prepare bacteria for phagocytosis to use Wright's terminology has
modified somewhat the original conception of phagocytosis as proposed
by Metchnikoff.

The phagocytic theory and the humoral theory of immunity would
appear to be in direct opposition. Metchnikoff maintained that the
fundamental basis of immunity resides in the phagocytic activity of
macro- and microphages. He believed that the humoral immune
bodies are derived either from leukocytes or the organs in which
they are formed the bone marrow and lymphatic system. The
champions of the humoral theory, on the other hand, would attribute
the healing principle to soluble substances contained in the body
fluids. The leukocytes and other phagocytic cells, according to the
extremists who advocate this theory, would be rega/ded as scavengers
merely, whose function it is to remove the debris dead bacteria or
disabled bacteria after they are overwhelmed by the activity of the
soluble natural and immune antibodies.

A final decision of the importance of cellular and humoral factors
in immunity cannot be made at the present time. It is not unlikely
that both theories will be modified somewhat as additional evidence
accumulates.

CHAPTER VII.
ANAPHYLAXIS, ALLERGY OR HYPERSENSITIVENESS. 1

PROTEIN fed to man or animals is reduced to simple compounds,
chiefly amino-acids, by the action of gastro-intestinal enzymes before
it is absorbed from the alimentary canal. These gastro-intestinal
enzymes act rapidly under normal conditions, and without an appre-
ciable latent period. One noteworthy result of digestion is a complete
denaturization of all ingested protein before it enters the tissues of
the host; absorption of unaltered or partially-digested protein is
prevented or reduced to a minimum.

The importance of a denaturization of protein before it enters the
tissues becomes apparent when a comparison is made between the
effects of parenteral injections of the end-product^ of prnfpin
on the one hand, and of the unaltered protein jtsplf nn tVi
Repeated parenteral injections of amino-acids in moderate amounts
appear to be without serious or noteworthy effects upon experimental
animals. A single parenteral injection of an unaltered protein is also
without visible effect, as a rule. A second parenteral injection of the
same protein, after an interval of ten to fourteen days, frequently is
followed by a rather definite train of symptoms, severe in character
and wholly unlike the negative response to a corresponding treatment
with amino-acids or normal end-products of gastro-intestinal digestion.

Sensitization. The first parenteral injection of a protein 2 which
is foreign to the body, or in some instances, natural for the body but
alien for the blood, is without visible effect upon the animal, but leads
to its sensitization to the specific protein. The sensitizing agent is
variously referred to as a sensitizer, sensibilisinogen, or anaphylac-
togen, and may be effective in very small doses. Rosenau and Ander-
son 3 were able to sensitize guinea-pigs with one-millionth of a cubic
centimeter of horse serum; Wells 4 has sensitized the same animal

1 For an excellent resume of the literature of anaphylaxis complete to 1912, see
Hektoen, Jour. Am. Med. Assn., 1912, Iviii, 1081.^

2 Proteins deficient in tryptophane or tyrosin are said not to sensitize.

3 Bull. 29 and 36, Hygienic Laboratory, Washington, D. C., 1906, 1907.

4 Wells' Clinical Pathology, 1914, 2d ed., 180.

REINJECTION OF THE HOMOLOGOUS PROTEIN 133

with one twenty-millionth of a gram of crystallized egg albumen.
Usually 0.001 to 0.1 c.c. of serum is an effective sensitizing dose.

A latent period intervenes between the initial injection of the
animal with sensitizing protein and sensitization on the average
this is about ten to fourteen days. Gay and Southard 1 showed, how-
ever, that the time necessary to effect sensitization depends somewhat
upon the size of the sensitizing dose, larger amounts requiring longer
periods than smaller amounts. White and A very 2 have found that a
relation exists between the minimum sensitizing and the maximum
intoxicating dose, larger amounts of protein being required on rein-
jection to elicit a reaction when the sensitizing dose is very small, and
vice versa.

Reinjection of the Homologous Protein. Repeated injections of the
homologous protein spaced at intervals less than ten days do not,
as a rule, cause symptoms of acute anaphylaxis after a third or a
fourth injection, however, there appears at the site of the first injec-
tion a swelling, usually indurated and more or less edematous, which
may lead to extensive necrosis and sloughing. These local reactions,
the so-called Arthus 3 phenomenon, are closely related phylogenetically
to the anaphylactic symptoms described below.

If the second parenteral injection is made after sensitization is
established usually after ten to fourteen days symptoms follow
almost immediately, which vary somewhat according to dosage and
the site of inoculation. A very large dose frequently results in rapid
death, the Theobald Smith phenomenon. 4 Very broadly speaking,
it requires 200 to 2000 times as much protein to cause acute anaphy-
laxis as to effect sensitization.

Intravenous or intracerebral injections of moderate doses are fol-
lowed very soon by a period of excitement (in dogs, followed by a
period of depression), 5 the animal is restless and moves about in a
bewildered manner and shows signs of respiratory embarrassment.
It coughs (a normal guinea-pig rarely or never coughs) and scratches
the corners of its mouth. This state is followed by dyspnea, with
involvement of the diaphragm and bronchial musculature leading to

1 Jour. Med. Res., 1908, xviii,.407.

2 Jour. Inf. Dis., 1913, xiii, 103.

3 Compt. rend. Soc. Biol., 1903, Iv, 20; 1906, Ix, 1143.

4 Theobald Smith, Jour. Med. Res., 1905, xiii, 341; Otto, Leuthold-Gedenkschrift,
1096, i, 153.

5 Guinea-pigs in general react most strikingly to anaphylactic stimuli; man is less
sensitive. Rabbits, sheep, goats, horses, and birds, in the order mentioned, are less
susceptible than man. Cold-blooded animals appear to be refractory.

134 ANAPHYLAXIS, ALLERGY OR HYPERSENSITIVENESS

bronchial spasm and later to paralysis of respiration, 1 lowered blood-
pressure, frequently cyanosis, an4 jdeath. Smaller intravenous injec-
tions are followed by the saijfce sympt$|Kis of excitement and respiratory
involvement, but to a lessfc degree, j Frequent micturition and fluid,
often bloody stools togetheV^ItK^reat prostration and dyspnea are
usually observed. The animal cannot stand and may die after several
hours, or eventually recover.

Intraperitoneal injections elicit similar symptoms. Subcutaneous
injections rarely cause acute death; as a rule the animal has a febrile
reaction and repeated injections may be followed by the Arthus pheno-
menon. If the animal survives an anaphylactic reaction it is fre-
quently observed to be more refractory or even temporarily immune
to subsequent injections of the same protein. This refractory state
is called anti-anaphylaxis by Besredka and Steinhardt. 2 This period
of refractoriness is of variable duration.

The postmortem appearance of guinea-pigs which have died from
the effects of acute anaphylaxis is usually striking and characteristic.
The lungs remain fully distended when the thorax is opened, the cut
surface is rather dry, and death appears to have resulted from
asphyxiation due to a tonic spasm of the bronchial musculature. 3

Severe but non-fatal anaphylactic reactions are accompanied by a
lowering of the body temperature, lowered arterial pressure, leucopenia,
frequently with a temporary partial or complete loss of coagulability
of the blood, 4 followed by a secondary febrile rise of temperature and
a leukocytosis in which polymorphonuclear leukocytes and frequently
eosinophiles 5 are increased. Animals killed during the early acute
symptoms show but little distention of the lungs the lesions may
resemble those of an acute toxic gastro-enteritis. Ecchymoses and
ulcers may be found occasionally in the stomach and intestines,
together with parenchymatous degeneration of the liver and particu-
larly the kidneys, which may lead eventually to fatty degeneration of
these organs.

The symptoms of anaphylaxis may be masked or even prevented by
the administration of certain drugs immediately before the reinjec-
tion of these atropin, chloral hydrate and similar narcotics are con-
sidered particularly efficient.

1 Auer and Lewis, Jour. Am. Med. Assn., 1909, liii, 6; Biedl and Kraus, Wien. klin.
Wchnschr., 1910, 844.

2 Ann. Inst. Past., 1907, xxi, 117, 384.

3 Auer and Lewis, loc. cit.

4 Biedl and Kraus, Wien. klin. Wchnschr., 1909, 363; Friedberger and Grober, Zeit.
f. Immunitatsforsch., 1911, ix, 216.

B Moschowitz, New York Med. Jour., 1911, Ixxxxiii, 15.

THE NATURE OF THE POISON, ANAPHYLATOXIN 135

THE NATURE OF THE POISON, ANAPHYLATOXIN.

The anaphylactic reaction, like other serological reactions, appears
to depend upon the elaboration of a specific antibody in the sen-
sitized animal. The specificity of the reaction is very striking in the
physiological sense the serum of one animal fails to sensitize for the
serum of an unrelated animal. Egg protein of one species also fails
to sensitize an animal against the egg protein of another species.
Osborne and Wells, 1 using vegetable proteins which can be obtained
in a state of relative purity, have shown that sensitization, in the
last analysis, depends chiefly upon the chemical composition of the
sensitizer. Thus, one vegetable protein fails to sensitize against a
second, unlike protein, even though they be derived from the same
seed.

The specificity of the reaction is striking it takes place only in
response to a second injection of the_jipmplogQus protein, but the
symptomatology is essentially the same, irrespective of the sensitizer.
The promptness with which the reaction appears after the reinjection
suggests at once that the poison, is radically different from a true
bacterial toxin, which invariably requires a definite latent, period
before symptoms can be detected,.] In this respect the anaphylatoxin
resembles somewhat an alkaloidal poison. Up to the present time no
antitoxins have been prepared. The action of the poison is peripheral
rather than central, according to Auer and Lewis. 2 Schultz 3 and
others have shown that it acts powerfully upon smooth muscle fibers;
Biedl and Kraus 4 and others have shown that an injection of peptone
into dogs elicits symptoms and pathological changes indistinguishable
from those of anaphylaxis. They were inclined to regard the anaphy-
latoxin as similar to, or possibly identical with peptone. Animals
immune to anaphylactic reactions react slightly or not at all to peptone
injections.

Passive anaphylaxis may be induced in a non-sensitized animal
by an injection of the serum of a sensitized animal. Usually a few
hours elapse before the recipient of the specific antibody is reactive,
however. The experiments of Pearce and Eisenbrey, 5 of Weil, 6 Dale, 7

1 Jour. Inf. Dis., 1913, xii, 341.

2 Loc. cit.

3 Hygienic Laboratory Bulletin, 1912, No. 80.

4 Wien. klin. Wchnschr., 1901, No. 11.

5 Journ. Inf. Dis., 1910, vii, 565.

6 Jour. Med. Res., 1913, xxvii, 497; 1914, xxx, 87, 299.

7 Jour. Pharm. and Exp. Therap., 1913, iv. 167.

136 ANAPHYLAXIS, ALLERGY OR HYPERSENSITIVENESS

Schultz 1 and others indicate that the reaction occurs within the cells
of the body rather than in the blood stream. The urine of anaphy-
lactic animals is toxic and 2 c.c. is frequently sufficient to kill guinea
pigs with anaphylactic symptoms, according to Pfeiffer. 2
<( Anaphylaxis may be defined as a congenital or acquired condition
of hypersensitiveness of man or animals tojthe parenteral introduction
of proteins, which is incited byjme or more injections of Jbacterial,
plant, animal or huma_n protein^ Active acquired hypersensitiveness
can be transmitted to non-sensitized individuals by the injjectionpf
the serum of an anaphylacticized individual, inducing in the recipient
of the serum a condition of passive anaphylaxis.) Anaphylaxis, there-
fore, belongs to the group of immuhological reactions.

Theories. Vaughan 3 has shown that all proteins may be split into
two fractions if they are heated with alcoholic potassium hydroxide;
one portion, insoluble in alcohol, when injected into animals gives
symptoms indistinguishable from those of anaphylaxis, irrespective
of the protein. The alcohol-soluble fraction is not toxic. The alcohol-
insoluble fraction obtained from various animal, vegetable, and
bacterial proteins always reacts the same, not only symptomatically,
but quantitatively as well. His theory is that the protein molecule
consists of two parts: an archon or nucleus, which is poisonous and
elicits the symptoms of anaphylaxis when it is injected parenterally
into animals, and common to all proteins; and additional groups
which are non-poisonous, but confer upon a protein by their number
and arrangement, its specificity. When a protein is injected paren-
terally into an animal, the cells of the animal elaborate an enzyme
which will specifically disintegrate it. Among the products of disin-
tegration is the poisonous nucleus or archon in a more or less free
state. The liberation of this substance causes acute poisoning of the
host. This substance, for which no antibody or antitoxin has been
prepared so far, is the "endotoxin" of bacteria.

Many of the phenomena of anaphylaxis are readily explained in
the light of Vaughan's work. The latent period or pre-anaphylactic
state which intervenes between the injection of a protein and the
appearance of sensitization is the time required to mature the specific
enzyme. The specificity of the enzyme (called forth by the stimulus
of alien protein in the tissues) is determined by the arrangement and

1 Loc. cit.

2 Zeit. f. Immunitatsforsch., 1911, x, 550.

3 Protein Split Products, 1913, for full discussion.

THE NATURE OF THE POISON, ANAPHYLATOXIN 137

number of groups arrayed around the poison group of the protein;
the similarity or identity of the symptoms of anaphylaxis irrespective
of the protein depends upon the liberation of the poison nucleus
(common to all proteins) in a relatively free state. The induction
of passive sensitization depends upon the injection of this specific
enzyme, which is present in the serum of a sensitized animal, into a
non-sensitized animal. 1

Vaughan regards the formation of a specific proteolytic enzyme in
response to the injection of alien protein into the tissues as a protec-
tive mechanism to rid the body of foreign substance; the theoretical
importance of this conception as a purposeful reaction is clearly
shown in bacterial infections. The incubation period of many bac-
terial infections is about two weeks, during which clinical symptoms
are not pronounced. This is interpreted as the time required by the
cells of a host to mature a specific enzyme capable of disintegrating
the alien protein (bacterial cells). The symptomatology of bacterial
disease is caused largely by the liberation of the poisonous nucleus
of the bacterial protein in special tissues or organs. Natural immunity
to bacterial disease, according to this theory, is due to the inability
of the organism to grow in the tissues of the host; active immunity
is conferred on the host by the presence of a persistent enzyme which
will disintegrate the specific organism whenever it is reintroduced
into the body.

Chemically, the poison nucleus or endotoxin is stated by Vaughan
to resemble beta-imidazoleethylamine, described previously. 2 The
specificity of the anaphylactic reaction depends upon the cleavage of
the protein molecule by a specific proteolytic enzyme with the libera-
tion of a non-specific poisonous product of protein degradation.
Abderhalden 3 and his associates have demonstrated proteolytic
enzymes in the blood stream.

Friedberger 4 has shown that a poison may be obtained by incubating
the inactivated serum of a sensitized animal with an excess of com-
plement and homologous (sensitizing) protein, which, when injected
into guinea-pigs, elicits the symptoms of anaphylaxis. It is not
true toxin, for no antibody is produced in response to repeated, sub-
lethal injections; it appears to differ from Vaughan's poison in that

1 The importance of the degradation of protein in the alimentary tract can be appre-
ciated in the light of what has been stated about anaphylaxis.

2 See page 76.

3 Zeit. f. physiol. Chem., 1912, Ixxxii, 109; Abwehrfermente des tierischen Organismus,
Berlin, 1913.

4 Zeit. f. Immunitatsfbrsch., iv, 636; vii, 94; Ueber Anaphylaxie, Ibid., 1911, ix, 394
(in collaboration with Goldschmidt, Schmanowsky, Schiiltze, and Nathan).

138 ANAPHYLAXIS, ALLERGY OR HYPERSENSITIVENESS

it is destroyed or inactivated at a temperature above 65 C. The
poison does not form if complement is not present in solution with
the inactivated serum and antigen, which would suggest a resemblance
to other cytolytic reactions in which the specific amboceptor is acti-
vated by complement.

The essential distinction between the theory of Vaughan and that
of Friedberger would appear to rest upon the nature of the poisonous
substance liberated; Vaughan would maintain the specificity of the
enzyme and the identity of the poisonous substances formed from
various proteins. Friedberger's theory, which was developed several
years after Vaughan's first work, would emphasize the distinction
between an enzyme and the specific amboceptor, which requires com-
plement for its activation. Keysser and Wassermann 1 and more
recently, Jobling and Petersen 2 have found that serum shaken with
kaolin, chloroform, and other agents will absorb substances from
serum, leaving the remainder toxic for guinea-pigs; the reaction
induced by the injection ef small amounts of altered serum resembles
closely that of anaphylaxis. Jobling and Petersen believe that the
toxic substance originates not from bacteria necessarily, but from
serum itself. Under normal conditions, anti-enzymes prevent the
normal serums from causing auto-autolysis; kaolin, bacteria, etc.,
added to the serum, absorb and thus remove the anti-enzymes, thus
permitting the serum to digest itself. In other words, the poisonous
substance may originate in the serum rather than in the bacteria or
other alien protein. These facts do not necessarily detract from
Vaughan's theory, but until more is known of the entire subject, a
final discussion of the mechanism of anaphylaxis must be postponed.

ANAPHYLAXIS IN MAN.

Natural Hypersensitiveness. It has long been known that the
inhalation of organic substances as the pollen of various plants, or
emanations from horses or guinea-pigs, of peptone or other similar
material may excite acute coryza and that train of symptoms popu-
larly recognized as " hay-fever" or "pollen fever" in some, but by no
means all, individuals. If the specific pollen or dust is rubbed on the
nasal mucosa of these sensitized individuals a violent reaction will
take place. Other individuals develop a severe urticaria if they eat
certain proteins: the flesh of arthropods, particularly crabs and
lobsters, vegetables, eggs, milk are known to excite symptoms in indi-

1 Ztschr. f. Hyg., 1911, Ixviii, p. 535.

2 Jour. Exp. Med., June, 1914, xix, p. 480.

ANAPHYLAXIS IN MAN 139

viduals who exhibit an "idiosyncrasy" to one or another of these
substances. This idiosyncrasy to foreign protein may be either
congenital or post-natal; the protein is supposed to have passed
unchanged through the intestinal tract in the latter case. The pheno-
mena in these instances are explained on the basis of sensitization
with specific protein; a mild anaphylactic reaction occurs when
the specific dust reaches the nasal mucous membrane or the specific
protein enters the digestive tract.

The tendency at the present time is to regard certain clinical and
pathological symptoms of bacterial infections particularly fever
and the production of specific pathological lesions as manifestations
of anaphylaxis as outlined by Vaughan. 1 The body is sensitized to
the alien protein, be it organic dust, protein of the food, or invasive
bacteria; the anaphylactic reaction takes place when the homologous
protein is brought into contact with the sensitized individuals through
the proper channels. It will be remembered that the incubation period
in many bacterial infections was explained as the time elapsing between
the arrival of the alien protein (bacterial cells) in the tissues of the
host and the maturing of a specific proteolytic enzyme that would
effect their disintegration. The symptomatology of bacterial infec-
tions, according to Vaughan, is largely due to the liberation of the
anaphylatoxin incidental to the lysis of the residual organisms.

Artificial or Acquired Hypersensitiveness. The phenomena
grouped for convenience as acquired hypersensitiveness are met with
chiefly in connection with the administration of the sera of animals
immunized for therapeutic purposes. Three types of anaphylactic
reaction may be recognized :

1. Sudden Death. A very few cases are on record in which the
administration of antitoxin for therapeutic purposes, either for
immunization or curatively, has been followed within a few minutes
or hours by death. Already, in 1896, Gottstein 2 had collected 12
which followed the injection of diphtheria antitoxin, 8 of whom were
diphtheritic, 4 healthy individuals. About 1 in every 50,000 appears
to be the proportion of deaths due to an injection of therapeutic sera.
The symptoms are essentially those observed in sensitized experi-
mental animals which die shortly after the injection of the homologous
protein. Behring, Kitasato and other observers had noticed many
years ago, when antitoxin was first prepared on a large scale, that
animals immunized with large amounts of tetanus or diphtheria toxin

1 Loc. cit.

2 Therap. Monatschr., 1896, Heft 5.

140 ANAPHYLAXIS, ALLERGY OR HYPERSENSITIVENESS

occasionally succumbed to a subsequent small dose of the homologous
toxin, although the blood serum of these animals contained much
specific antitoxin.

2. Serum Sickness or Serum Disease. Attention was first directed
to serum sickness by von Pirquet and Schick, 1 who noticed that there
occasionally developed in individuals who had received an injection
of antitoxic sera, usually after seven to fourteen days, fever and a
rash which might be urticarial, scarlatinal, or, in the more severe
cases, morbilliform; enlargement of lymph glands, particularly those
near the site of inoculation; and joint pains, more frequently of the
metacarpal joints. A slight edema, frequently of the angioneurotic
type, was also occasionally observed. The fever is usually slight
and there are signs of respiratory embarrassment, not as a rule
marked, but occasionally severe. These reactions, sudden death
and serum sickness, are more common in asthmatics, and in those
individuals presenting the pathological syndrome known as status
lymphaticus.

According to Moschowitz, 2 these individuals, particularly the
asthmatics, present an eosinophilia. The exact cause of sudden death
following the administration of diphtheria antitoxin is not definitely
known, but it has been assumed that respiratory involvement is a
potent factor. The appearance of serum disease seven to fourteen
days after the administration of antitoxin is supposed to depend upon
the fact that some of the alien protein (serum) remains in the body
during the period of pre-anaphylaxis (period of sensitization), and
that this residual protein is broken down by the mature specific enzyme
or enzymes with the liberation of a poisonous substance which causes
the anaphylactic shock.

3. Arthus Phenomenon. During the course of immunization against
rabies by the Pasteur method it is frequently noticed that after three
or four injections a subsequent injection causes symptoms of inflam-
mation at the site of the first injection, and that this phenomenon is
repeated, usually, but not always, with diminishing intensity at the
site of earlier injections as the treatment progresses. This inflam-
matory reaction at the site of injection is not due to bacterial infection
ordinarily, but is rather an expression of anaphylaxis. It is comparable
to the Arthus phenomenon produced in rabbits by successive injec-
tions of serum referred to above. Also in re vaccination (vaccinia)
a so-called accelerated reaction may occur the second time the indivi-

1 Die Serumkrankheit, Leipzig, 1905.

2 New York Med. Jour., 1911, Ixxxxiii, 15.

ANAPHYLAXIS IN MAN 141

dual is vaccinated. This accelerated reaction again is a mild edition
of the Arthus phenomenon.

4. Prophylaxis. At first sight it might appear that the administra-
tion of diphtheria and tetanus antitoxin for therapeutic purposes would
be a dangerous procedure. If there is reason to suspect that the
patient would react to the injection of antitoxin it is advisable to
inject 0.1 or 0.2 c.c. subcutaneously and wait half an hour. If no
symptoms develop, the full dose may be given without danger; it
is generally believed that even if mild symptoms do follow the initial
injection, the full dose may be given with safety after half an hour;
the first injection appears to abort what otherwise might be a reaction
dangerous to the patient

The present method of concentrating diphtheria antitoxin by frac-
tional precipitation of the globulin 1 appears to reduce very materially
the incidence of serum sickness. According to German investigators,
antitoxin which has stood for one or two months has lost to a very
considerable extent the substance or substances which cause the
symptoms of serum sickness.

Practical and Theoretical Considerations. A. Advantage is taken
of the sensitization of individuals by bacterial protein during certain
bacterial infections, particularly those with the tubercle bacillus,
B. mallei, and in syphilis, for diagnostic purposes. It has been shown
almost beyond doubt that individuals suffering from these diseases are
sensitized to the bacterial protein, and it is possible to make a fairly defi-
nite clinical diagnosis by introducing extracts of the specific organisms
into the skin and inducing there an anaphylactic reaction which, if the
dose is small, is local in character, but which may be general and severe
if the dose is increased in amount. The von Pirquet, Calmet, Moro, and
Koch methods of utilizing tuberculin for diagnostic purposes are directly
dependent upon this reaction of hypersensitiveness. The diagnostic
use of mallein and luetin depend upon the same phenomenon.

B. Advantage is also taken of the specificity of the anaphylactic
reaction for the recognition of proteins. Wells and Osborn 2 and
many others have sensitized guinea-pigs with proteins and then
injected into these sensitized animals proteins which are to be iden-
tified either specifically or phylogenetically. The nature and extent
of the anaphylactic reaction in these animals furnishes the most deli-
cate test (except possibly the precipitin test) which is available for
such investigations.

1 Banzhof, Johns Hopkins Hosp. Bull., 1911, xxii, 241.

2 Loc. cit.

CHAPTER VIII.

ANTIGENS AND THE TECHNIC OF SERUM
REACTIONS.

NATURE OF ANTIGENS AND ANTIBODIES.
AGGLUTININS AND PRECIPITINS.
LYSINS.

Hemolysis and" the Complement Fixa-
tion Reaction.

AGGRESSINS.

OPSONINS, TROPINS. BACTERIAL VAC-
'CINES.

NATURE OF ANTIGENS AND ANTIBODIES.

THOSE substances which cause specific antibody formation when
they are introduced into the tissues or the body fluids of the host are
called antigens. Their chemistry is as yet unknown, but available
evidence would indicate that they are protein in nature and highly
organized chemically. Degradation products of proteins, as albu-
moses and peptones and carbohydrates and fats, are not ordinarily
antigenic, that is, they do not lead to antibody formation when they
are introduced into the animal body. 1 The antigenic properties of
lipoids are still a subject of controversy: lipoids appear to play a
prominent part in certain types of immunological reactions, but their
ability to stimulate specific antibody formation cannot be regarded
as proven at the present time. 2

The function of antibodies as specific offensive weapons of the
host against alien organisms or their products has long been recog-
nized in bacteriology, and most important laboratory diagnostic
methods have been elaborated through a study of the reactions
between specific antigens and their respective antibodies. Antibodies
are soluble and are found in various concentrations in blood serum
derived from immunized animals. Many attempts have been made
to determine changes in the chemical composition or physical proper-
ties of immune sera from those of normal serum. Atkinson, 3 Gibson,

1 The injection of carbohydrates and fats may, however, lead to specific enzyme
formation. See Rohmann, Antigene Wirkung der Kohlenhydrate, Deutsch. med.
Wchnschr., 1914, xl, 204.

2 See Pick, Kolle, and Wassermann, Handbuch der pathogenen Mikroorganismen,
2d ed., Bd. I, for discussion of the chemistry of antigens.

3 Jour. Exp. Med., 1899, iii, 649.

AGGLUTININS. AGGLUTINOIDS AND PROAGGLUTINOIDS 143

and Banzhaf 1 and others have found that the sera of horses immunized
to diphtheria toxin show a marked increase in globulin content, with
a decrease in albumin content. Beljaeff 2 could find no appreciable
change in the refractive index, specific gravity, freezing point or
reaction of the serum of an immune animal above that of a normal
animal.

The chemical nature of antibodies, aside from their apparently
close relation to globulin, has not been determined. There is evidence
that antitoxin molecules may be larger than toxin molecules, how-
ever. Martin and Cherry 3 found that toxins could be forced through
dense porcelain filters impregnated with gelatin, which would restrain
antitoxin, and Arrhenius and Madsen 4 determined that the toxin mole-
cule diffused several times as rapidly as the antitoxin molecule, from
which observation they assumed that the antitoxin molecule was
larger than the toxin molecule.

AGGLUTININS. AGGLUTINOIDS AND PROAGGLUTINOIDS.

Gruber and Durham 5 appear to have been the first to clearly demon-
strate specific clumping in broth cultures of typhoid and cholera
organisms when their respective sera were added to them. Somewhat
later Widal, 6 and independently Griinbaum, 7 utilized the principle
of the specific agglomeration of bacteria by their immune sera for
the diagnosis of typhoid fever. They found that relatively early in'
the disease, sera of typhoid patients clumped typhoid bacilli from
broth cultures. Pfaundler 8 observed that typhoid bacilli grown in
broth containing low concentrations of specific sera grew out into long,
tangled filaments, the " thread" reaction. Originally this phenome-
non was regarded as highly specific, but it has largely given way to the
macroscopic or microscopic agglutination test.

Agglutination in the bacterial sense may be defined as a clumping
or agglomeration of bacteria from a uniform suspension in a fluid
medium, brought about by the addition of specific antibodies
agglutinins. It takes place in two stages if motile bacteria are con-
cerned. First there is loss of motility "immobilization" and

1 Jour. Exp. Med., 1910, xii, 411.

2 Cent. f. Bakt., Orig., 1903, xxxiii, 293, 396.

3 Proc. Royal Soc., 1898, Ixiii.

4 Festskrift Statens Serum Institute, 1902.
6 Munchen. med. Wchnschr., 1896, No. 13.

6 La Semaine Medicale, 1896, No. 13.

7 Brit. Med. Jour., 1897, May 1, and Munchen. med. Wchnschr., 1897, 330.

8 Cent. f. Bakt., 1898, xxiii, 9, 71, 131.

144 ANTIGENS AND THE TECHNIC OF SERUM REACTIONS

eventually clumping. Smith and Reagh 1 working with a non-motile
hog cholera bacillus have demonstrated both flagella and somatic
agglutinins, the former paralyzing the activity of the flagella, the
latter agglomerating the organisms themselves. Non-motile bacteria
usually agglutinate somewhat more slowly than motile organisms.
Small amounts of neutral salts are necessary for the clumping of
bacteria, 2 although a union of the specific organism and its agglutinin
will take place even if salts are absent. The specific substance (or
substances) of the bacterial cell which reacts with the specific antibody
of the serum (agglutinin) is known as agglutinogen. Closely related
bacteria, as typhoid and paratyphoid bacilli, may possess a certain
amount of agglutinogen in common, but, as a rule, the specific
organisms are clumped in immune sera at much, greater dilution than
related organisms are clumped. Also, the specific organisms will
remove the agglutinin completely from immune sera, while closely
related bacteria only remove that portion of the agglutinating sub-
stance which is common to both organisms, leaving behind the
specific agglutinin which will then agglutinate the specific organism,
but not its closely related fellow; that is to say, closely related
bacteria will react with the common or group agglutinin, but fail to
absorb the specific agglutinin.

Experience has shown that the sera of normal adults frequently
contain agglutinin which will clump various bacteria and the potency
of these "normal" or natural agglutinins may even be sufficient to
clump moderate numbers of typhoid bacilli in dilutions as great as
1 to 30. The sera of normal nurslings contain only minimal amounts
of normal agglutinins as a rule, and the conclusion has been drawn
that normal agglutinin may be either:

(a) Group agglutinin, derived from mild infection with closely
related organisms, or

(6) True immune agglutinins resulting from mild or unrecognized
infection with the specific organism.

No definite distinction has been noted between natural and immune
agglutinins; the latter are usually present in sera, however, in much
greater concentration than the former.

The site of formation of agglutinins in the body is not definitely
known, although lymphoid tissues appear to be intimately concerned,
especially bone-marrow and the spleen. Pryzgode 3 states that

1 Jour. Med. Res., August, 1903, x, No. 1.

2 Bordet, Collected Studies in Immunity, 1909 (translation by Gay).

3 Wien. klin. Wchnschr., 1913, xxvi, 84.

AGGLUTININS AGGLUTINOIDS AND PROAGGLUTINOIDS 145

cultures of spleen tissue in vitro will form specific agglutinins for
typhoid bacilli if the virus is brought into contact with the tissue
cells. As a general rule the concentration of a specific agglutinin is
greater in the blood stream than in the tissues of the body.

Preparation of Specific Agglutinating Sera. Specific agglutinating
sera for experimental purposes are best obtained from rabbits, whose
serum normally contains no agglutinin. Several, usually three to
five intravenous injections of 1, 2, 3 and 5 loopfuls respectively of
killed cultures of typhoid bacilli at eight-day intervals, produce pow-
erful agglutinating sera. The animal is bled about two weeks after
the last injection. For large amounts of agglutinating sera horses or
asses must be used.

Properties of Agglutinins. Agglutinins are of unknown chemical
composition, but they may be separated from solution by those pre-
cipitants w r hich throw down globulins, and they may be removed
from solution by absorption in animal charcoal. Toward heat they
are moderately resistant, usually remaining active after an exposure
of twenty minutes to 55 C., a degree of heat sufficient to inactivate
complement. Agglutinins, therefore, appear to be quite distinct
from bacteriolysins. The temperature at which agglutinins are de-
stroyed depends upon their specificity, agglutinins for plague bacilli
being more sensitive than typhoid agglutinins. The reaction of the
medium also affects .their stability. Alkalis, even in dilute solution,
rapidly destroy agglutinins; acids are somewhat less harmful. Nat-
urally the duration of exposure to these various agents exercises an
important influence upon their resistance. Agglutinins do not appear
to pass through parchment membranes, but it is stated that agglu-
tinogen will slowly diffuse under similar conditions. This would sug-
gest that the agglutinin molecule is larger than the agglutinogen
molecule. Preserved in a dry state, in a cool place away from light,
agglutinins preserve their properties unimpaired for days. In
solution and upon standing agglutinins rapidly lose their property
of clumping bacteria, but they still retain their original ability to unite
firmly with bacteria. Ehrlich designates agglutinins which have
lost their ability to cause clumping but still retain their combining
power for agglutinogen, agglutinoids. In his terminology they are
side-chains of the second order which have lost their agglutinophore
(ergophore) group. Agglutinins acting in neutral salt-free media
also fail to cause clumping of bacteria, but in this case the addition
of a small amount of NaCl or even some weak acid very soon brings
10

146 ANTIGENS AND THE.TECHNIC OF SERUM REACTIONS

about a typical reaction. 1 This and similar observations have
attracted attention to the similarity between the precipitation of
bacteria to which agglutinin is anchored by neutral salts, and the
precipitation of finely suspended 'clay by the addition of neutral salts;
the inference has been drawn that the phenomenon of agglutination

one of physico-chemistry.

Specificity of Agglutination Reactions: Group Agglutinins. The
composition of the agglutinogen that constituent of the bacterium
which stimulates agglutinin formation is unknown, but it appears
to be complex and probably not a single chemical compound. Closely
related bacteria may possess in common a small amount of agglu-
tinogen a least common multiple, as it were which stimulates the
production of "group agglutinin" that reacts with related bacteria
more or less in proportion to their content of the common antigen or
agglutinogen. . The specific agglutinin produced by the entire agglu-
tinogen content of an organism is more potent and fails to react with
related bacteria. Thus, the serum of an animal immunized against
B. typhosus may agglutinate that organism in a dilution of 1 to 3000;
B. paratyphosus will be agglutinated in a dilution of 1 to 300 by the
same serum, and B. coli would agglutinate only in a dilution of 1 to
50. The group agglutinin in this example would be effective for B.
paratyphosus in a dilution of 1 to 300, but in greater dilutions it would
be ineffective. For B. coli in the instance cited, the group agglutinin
is ineffective in dilutions above 1 to 50.

The common or group agglutinin for B. paratyphosus in this typhoid
serum could be quantitatively removed by leaving it in contact with
a large number of paratyphoid bacilli for a few hours, then centrifu-
galizing to remove the organisms. The residual serum would contain
only agglutinin specific for B. typhosus. If B. typhosus were added
to the serum, all the agglutinin both "group" and specific would
be removed.

As a general rule, group agglutinins constitute a minor fraction of
the total agglutinin and in practice the degree of dilution of the serum
used in specific cases is ample to exclude error. It occasionally happens
that sera of low dilution, especially those rich in agglutinoids, fail to
clump the specific organism; as the serum is diluted more and more
the phenomena of clumping become more and more marked; finally
a degree of dilution is reached beyond which the serum again becomes
ineffective. The initial negative agglutination in concentrated serum

1 Bordet, Ann. Inst. Past., 1899, xiii, 225.

AGGLUTININS AGGLUTINOIDS AND PROAGGLUTINOIDS 147

is known as a " proagglutinoid" reaction; it is attributed by Ehrlich
to the presence of " agglutinoids" in the serum side-chains of the
second order which have lost their agglutinophore group, but still
retain their combining group (haptophore group). These " agglu-
tinoids," which are deteriorated agglutinins, have a greater affinity
for the agglutinogen of the bacteria than have the unchanged agglu-
tinins, and consequently prevent the latter from becoming attached
to the organisms. If the serum is diluted a point is reached where
the agglutinoids are numerically too few to interfere with the action
of the agglutinins, which usually far outnumber the agglutinoids. As
the serum is more and more diluted a point is eventually reached
where the content of agglutinin is insufficient to react with the bacteria.
If, however, bacteria are cultured in this dilute serum, they frequently
develop into long, thread-like, interwoven filaments, the so-called
"thread-reaction" of Pfaundler. It is obvious that the maximum
dilution at which a serum will agglutinate bacteria depends somewhat
upon the number of organisms; there is, in other words, a relation
between the amount of agglutinin in the serum and agglutinogen in
the bacteria.

Non-agglutinable Bacteria. Occasionally strains of bacteria, as
B. typhosus, freshly isolated from the body, may not agglutinate
with the specific serum. This resistance to agglutination is supposed
to result from some unknown change in the agglutinogen of the bac-
terium during its development in the body. A similar loss of agglu-
tinability may be experimentally brought about by growing the
bacteria in gradually increasing concentrations of specific agglutinat-
ing serum outside the body. This inagglutinability is usually lost
after a few days' development on artificial media; the organisms will
then clump in a characteristic manner in a serum that originally was
ineffective.

The Reaction of Agglutination. The practical value of the reaction
of agglutination depends upon the visible clumping or agglomeration
of a suspension of bacteria in a fluid medium containing some neutral
salt, when a relatively small amount of immune serum specific for
the organism is added to it. The reaction may be expressed thus:

Organism (Agglutinogen) + Specific Serum (Specific Agglutinin)
= Agglutination.

If a specific organism is added to an appropriate dilution of unknown
serum with proper precautions, and characteristic clumping takes
place, or if a known specific serum is added with suitable precautions

148 ANTIGENS AND THE TECHNIC OF SERUM REACTIONS

to a suspension of an unknown organism, and characteristic clumping
takes place, a specific diagnosis of the serum or of the organism can
be arrived at. In the first instance, a diagnosis of disease may be
made; in the second instance the identity of an organism may be
established. The laboratory diagnosis of typhoid and paratyphoid
fever, of the various types of bacillary dysentery and of other bacterial
infections is frequently made by testing the serum of the patient
for agglutination with a known culture of the organism. 1 The labora-
tory identification of specific bacteria, conversely, is frequently estab-
lished or corroborated through their agglutination with known specific
agglutinating sera.

The reaction of agglutination may be made either microscopically
or macroscopically.

1. Microscopic Method. A drop of serum from a patient, diluted to
the proper degree, is mixed with an equal amount of a broth culture
of the desired organism on a clear cover-glass, 2 and then suspended
over the cavity of a hollow ground slide, ringed with vaseline to pre-
vent evaporation, and examined under the microscope. Motile bac-
teria, as for instance B. typhosus, soon lose their motility (immo-
bilization) and gradually collect in small groups which tend to coalesce
into larger and larger clumps, leaving the field between them practically
free from organisms. The bacteria are not necessarily killed by agglu-
tination. The reaction ordinarily is complete within two hours.
Killed cultures of bacteria may be used in place of living cultures
but the reaction is usually less clear-cut.

The advantage of the microscopic method lies chiefly in the small
amount of serum required to perform the test. One of its chief disad-
vantages lies in the relative inaccuracy of the dilution of the serum.
(See chapter on B. typhosus for full discussion of technic.)

2. Macroscopic Method. Various dilutions of serum, accurately
measured by volumetric pipettes, are brought into small, sterile test-
tubes, together with suspensions or broth cultures of the bacteria.
Agglutination is manifested by the gradual accumulation of a floccu-
lent sediment of bacteria, leaving the supernatant liquid perfectly
clear. Control tubes without serum remain uniformly clouded.

The part played by agglutinins in immunity is unknown; the

1 The technic and precautions to be observed are discussed individually in the chap-
ters upon specific pathogenic bacteria.

2 For a majority of bacteria, eighteen-hour cultures in 0.1 per cent, dextrose broth
are particularly advantageous. Cultures grown in plain broth are usually much less
actively motile and agglutinate less readily.

PRECIPITINSPRECIPITINOIDS 149

concentrations of agglutinins in immune sera, as measured by present-
day methods, throws no light upon the degree of immunity or the
prognosis. Very severe typhoid infections, for example, may show
little agglutinin in their sera, and mild cases may exhibit sera com-
paratively rich in agglutinin content. Their chief value at the present
time lies in their relation to the diagnosis of disease.

PRECIPITINS. PRECIPITINOIDS.

In the preceding section it was shown that the sera of animals
immunized with various bacteria contained substances agglutinins
which agglutinated the specific organisms. Kraus 1 showed that
these immune sera would cause a precipitate when they were added
to clear filtrates of the specific organisms. During the process of
immunization, therefore, specific antibodies, termed precipitins, are
formed, which react with the specific soluble antigen, precipitinogen,
in germ-free filtrate of broth cultures of the specific organisms, to
form a precipitate. Later investigations have shown that any soluble
protein, as egg-albumen, injected into experimental animals may
stimulate the production of specific precipitins which will cause a
precipitation in clear solutions of the homologous protein. These
reactions have a marked specificity: The sera of animals immunized
against casein of cows' milk, for example, will cause precipitation in
clear solutions of this protein, but will fail to cause precipitation in
solutions of casein from human milk. The sera of closely-related ani-
mals may contain small amounts of "group" precipitins, and biological
relationships have been established, based upon the community of
these antibodies. Thus, the sera of certain anthropoid apes 2 are
said to be precipitated by the sera of animals immunized to the serum
of man; sera from the lower- monkeys fail to react with the human
serum. From these observations the inference has been drawn that
these anthropoid apes are more closely related to man than are the
lower monkeys. 3

Precipitins closely resemble agglutinins in their method of formation,
their resistance to physical agents and their reactions. Like the
agglutinins, they possess both a thermostabile haptophore or combining
group and a thermolabile ergophore group. The precipitinophore

1 Wien. klin. Wchnschr., 1897, 736.

2 Griinbaum, Lancet, January, 1902.

3 See Nuttall, Jour. Hyg., 1901, i, No. 3; Proc. Royal Acad., November, 1901, Ixix;
Proceedings Cambridge Philosophical Society, January, 1902; Brit. Med. Jour., April,
1902, i, for full details.

150 ANTIGENS AND THE TECHNIC OF SERUM REACTIONS

group is very labile and readily becomes non-functionating, but the
combining group is relatively stabile. A precipitin which has lost
its ergophore group is called a precipitinoid.

The precipitate formed by a specific serum acting upon a clear
solution of the antigen (precipitinogen) probably is derived from the
serum, because very dilute solutions of the immunizing protein will
throw down a relatively bulky precipitate, far too great in amount
to come from the antigen in the dilution used. 1

Precipitins have been extensively studied in their relation to cer-
tain aspects of Forensic Medicine, but they have little practical value
in the laboratory diagnosis of bacterial disease. They are found in
sera under the same conditions as agglutinins, but the technic for their
demonstration is more involved than that for agglutinins. Their
relation to immunity is unknown, but probably similar to that of
agglutinins.

LYSINS.

Mention has been made (see preceding section) of the bactericidal
power of fresh blood serum of a normal animal and man. This impor-
tant discovery, that normal sera contain substances that will destroy
moderate numbers of bacteria, was made by Nuttall, 2 who also
observed that there was a limit to this destructive activity and that
this property was lost upon standing, or rapidly destroyed by an
exposure of the serum to 55 C. for half an hour. Buchner 3 corrobo-
rated and extended these observations and designated the . unknown
stabile component "alexin." Pfeiffer 4 then showed that the destruc-
tive action of normal sera could be increased many fold above its
original level for a specific organism if that organism were repeatedly
injected into an animal in sublethal, but gradually increased doses.
The serum of such an animal would still destroy only moderate num-
bers of heterologous bacteria, but relatively great numbers of the
homologous bacteria. This observation opened the way for the highly
important study of active acquired immunity against bacteria.
Pfeiffer observed that heating immune sera to 50 to 56 C. for half
an hour destroyed their bactericidal properties, precisely as Nuttall
had found that natural, non-specific bactericidal properties were
destroyed under similar conditions. Bordet 5 then discovered that the

Welsh and Chapman, Ztschr. f. Immunitasforsch., 1911, ix, 517.

Ztschr. f. Hyg., 1888, iv, 353.

Cent. f. Bakt., 1889, v, 817; vi, 1, 561.

Ztschr. f. Hyg., 1894, xviii, 1 ; 1895, xix, 75-100.

Ann. Inst. Past., 1895.

LYSINS 151

addition of a small amount of unheated blood serum from a non-
immune animal would "reactivate" the heated inactive immune serum
and restore its bactericidal power to its original level. These experi-
ments collectively demonstrated clearly that :

1. Normal sera had an inherent but limited destructive action upon
a variety of bacteria.

2. That this destructive or bactericidal action could be greatly
increased for specific organisms through repeated injections of sub-
lethal doses of them. 1

3. That both normal and immune sera lost their bactericidal prop-
erties by heating them to 55 C. for half an hour.

4. That immune sera would regain their specific bactericidal power
if a small amount of fresh normal blood serum of a non-immune
animal were added to them. 2

Bordet 3 showed similarly that the red blood cells of an alien animal
were also destroyed to a limited degree by the serum of a normal
'animal, but that the destruction could be greatly increased for specific
erythrocytes if they were repeatedly injected into an experimental
animal. The blood serum becomes specifically hemolytic. Here
again Bordet 4 found that heating an immune serum to 55 C. for
thirty minutes destroyed its activity, but that a small amount of
fresh serum from a non-immune animal (whose serum per se would
not dissolve the homologous cells) would reactivate the serum. Thus,
both specific bacteriolytic sera and specific hemolytic sera must con-
tain two distinct components a thermostabile component resisting
an exposure to 55 C. for half an hour and contained only in the
immune serum, and a thermolabile component destroyed or inacti-
vated at 55 C., which is present both in active immune bacteriolytic
and hemolytic sera, and also in normal sera. To the thermolabile
substance present in unheated normal and immune sera, Bordet gave
the name "alexin;" to the thermostabile specific substance in immune
sera he gave the name "substance sensibilitrice." He regarded the
"substance sensibilitrice" as a sensitizer or mordant which made
bacteria or blood cells vulnerable to the ferment-like or digestive
action of the "alexin."

Ehrlich and Morgenroth 5 studied the phenomena of hemolysis in

1 Presumably leaving the original non-specific bactericidal power at its initial level
for all except the specific organism, and possibly for closely related forms.

2 Moxter, Cent. f. Bakt., 1899, xxvi, 344.

3 Loc. cit.

4 Ann. Inst. Past., 1898, xii, No. 10.

6 Berl. klin. Wchnschr., 1899, No. 1 and 22. See also Collected Studies on Immunity,
Ehrlich, translated by Bolduan, 1910.

152 ANTIGENS AND THE TECHNIC OF SERUM REACTIONS

great detail and demonstrated by very careful and ingenious experi-
ments that the phenomena observed by Bordet were fundamentally
correct. They showed :

1. That inactivated specific hemolytic serum (heated to 55 C.)
was absorbed by the homologous red blood cells, and that these " sen-
sitized " cells, separated from the serum after a few hours and washed
carefully, were readily hemolyzed when resuspended in salt solution
to which was added a small amount of fresh, unheated, normal guinea-
pig serum.

2. The supernatant residual fluid from which the red blood cells
had removed all the immune body was incapable of causing hemolysis
of the homologous red blood cells when fresh normal serum was added
to it. The erythrocytes, in other words, quantitatively removed the
thermostabile "substance sensibilitrice" from solution.

3. If normal sera were allowed to remain in contact with the same
red cells for an equal length of time, and these red cells were then
removed by centrifugalization and resuspended in salt solution con-
taining normal fresh serum, no hemolysis took place, leading to the
conclusion that the thermolabile substance (alexin of Bordet) is not
removed from solution by erythrocytes. Apparently alexin is not
bound or anchored directly to the red cells.

4. Finally, it was shown that inactivated immune serum, red blood
cells and fresh normal serum could be maintained at C. without
apparent hemolysis. At 37 C. the same solution soon exhibited com-
plete hemolysis. Thus, at the lower temperature, the normal serum
failed to cause hemolysis. If the mixture maintained at C. were
centrnugalized, however, after some hours, and the red blood cells
washed thoroughly and resuspended in salt solution, hemolysis
promptly occurred when a small amount of normal serum was added
to the suspension, thus showing clearly that the inactivated immune
serum was bound or anchored by the red blood cells at C., even
though activation did not take place.

Ehrlich substituted the term " amboceptor" for Bordet's term
"substance sensibilitrice" and complement for the term "alexin,"
and conceived that the immune body amboceptor consisted essen-
tially of two combining or haptophore groups one the cytophilic
group, possessing a specific combining power for the specific cell
(bacterium or erythrocyte), the other, complementophilic group,
combining with the non-specific complement. According to this
theory the union of complement to specific cell takes place through the

LYSINS 153

amboceptor; Bordet maintains that neither the specific cell (antigen)
of itself nor the substance sensibilitrice (amboceptor) of itself unites
with alexin (complement). When both are simultaneously present,
however, alexin is absorbed. In other words, amboceptors as such do
not exist, according to this view, and consequently complement can-
not be bound to the specific cell by a complementophile (haptophore)
group.

Multiplicity of Amboceptors and Complement. The researches of
Nuttall and Buchner and of Moxter 1 have shown that fresh normal
serum possesses definite but limited bactericidal powers, apparently
not specific (for a variety of bacteria may be destroyed) which are
destroyed by an exposure of thirty minutes to 55 C. Furthermore,
the "inactivated" serum appears to regain its original bactericidal
value for various organisms when it is mixed with a relatively small
amount of normal serum. In other words, normal serum and specific
immune serum (unheated) alike appear to depend upon thermostabile
amboceptor and thermolabile complement for their bacteriolytic and
hemolytic activities. They differ in the highly specific potency of
the immune serum for its homologous cell. Ehrlich and Morgenroth 2
believe that the normal or natural cytolytic activities of sera depend
upon a multiplicity of specific amboceptors, each for its specific red
blood cell or other cell, and Pfeiffer 3 has made similar observations
for the normal bactericidal powers of blood. Ehrlich and Morgen-
roth have attempted to demonstrate a multiplicity of complements in
normal sera also; heated normal sera injected into normal animals
are claimed by the Ehrlich school to give rise to anticomplementoids,
the supposition being that the heat has destroyed the ergophore
group of complement but not its combining group, giving rise to a
"complementoid," precisely as a toxin which has lost its toxophore
group becomes a toxoid. There appears to be no theoretical limit
to the anti- and anti-antibodies which may thus be produced by
various increasingly complicated investigations. Bordet and Gay 4
deny the multiplicity of complement.

Fixation of Complement. Bordet and Gengou, 5 in a series of experi-
ments, brought forth experimental evidence of the unity of com-
plement and, incidentally, developed a method of investigation now

1 Loc. cit. 2 L OC> eft.

3 Harben Lecture, Jour. Royal Inst. Public Health, 1909, xvii, 385.

4 Collected Studies in Immunity by Bordet and his associates (translated by Gay,
1909).

8 Ann. Inst. Past., 1901, xv.

154 ANTIGENS AND THE TECHNIC OF SERUM REACTIONS

extensively utilized to demonstrate the presence of various specific
immune antibodies. If a specific immune body (as for example,
the serum of an animal immunized to typhoid bacilli) .is heated to
55 C. for half an hour, then added to a suspension of typhoid bacilli
together with normal unheated serum, a union between the bacilli,
the specific antibody of the serum (amboceptor, substance sensibili-
trice) and the complement (alexin) will take place. If the proportions
of the three reactive bodies are correct, all the complement or alexin
will be bound, provided the mixture is incubated a few hours at 37
C. If, now, red blood cells and inactivated immune serum specific
for the red blood cells are added to the mixture of bacteria, immune
body and complement, no hemolysis should be noticed, because the
complement is quantitatively anchored to the bacteria-immune serum
complex. If, on the other hand, the inactivated immune serum added
to the suspension of typhoid bacilli be not typhoid immune serum,
complement will not be bound to the bacteria, for the specific ambo-
ceptor or substance sensibilitrice will not be present. The complement
or alexin, therefore, is not anchored to the bacteria, and it is free to
act when the red blood cells and their specific inactivated serum are
added to the mixture of bacteria and serum. Under this condition
hemolysis occurs, because the red blood cells, inactive immune body
and complement unite. The production of hemolysis being visible,
it acts as an indicator in such instances. Wassermann and his asso-
ciates have utilized this method of "fixation of complement" for the
serologic diagnosis of syphilis, and gradually a relatively large number
of diagnoses of clinical importance have been developed along the
same lines.

The Determination of Specific Antibodies by the Method of Complement
Fixation. Principle Invoked. When an antigen (bacteria, erythro-
cytes, tissue cells, protein, or other substance which stimulates specific
antibody formation) is mixed intimately with its specific inactivated
immune serum and fresh normal complement a firm union of the
three components takes place. 1 Jhe result of this union is an injury
or destruction of the antigen, ^f the antigen be bacterial cells or
tissue cells there is usually no visible change in the gross appearance
of the mixture, and cultural or chemical investigation must be relied
upon to demonstrate the lytic process. Erythrocytes, on the other
hand, undergo changes in the presence of inactivated specific immune
serum and complement which result in the liberation of hemoglobin,

1 Bordet and Gengou, Ann. Inst. Past., 1901, xv, 290.

LYSIN& 155

which colors the solution deep red. This change is clearly visible
and requires no additional procedure for its demonstration; the
liberation of hemoglobin is in itself an indicator of the reaction which
has taken place.

The relation between antigen, immune serum, and complement
is quantitative; consequently, if the respective amounts of the three
components are correctly proportioned, no free unattached comple-
ment will be present in a mixture of them after an appropriate incuba-
tion at body temperature is practiced to allow of their union. Usually
an hour at 37 C. suffices for this union to take place quantitatively.

These very important observations of Bordet and Gengou have
led to the development of a technic for the diagnosis of infection, and
the identification of antigens by the method of complement fixation.

The underlying principles of the reaction of complement fixation
are three:

(a) The union of specific inactivated immune serum and homologous
antigen.

(b) The quantitative activation of the antigen inactivated specific
immune serum complex by non-specific complement; and

(c) The visible hemolysis that results from the activation of an
erythrocyte inactivated specific immune serum complex by non-
specific complement.

The general plan of procedure is to incubate an antigen (as bacterial
cells) and inactivated serum and complement in proper proportions
for an hour, to permit the three components to unite. A mixture of
erythrocytes and specific inactivated hemolytic serum is now added.
If the reactive substances are properly proportioned and the inac-
tivated serum first added is specific for the antigen (bacteria), no
hemolysis will occur when the hemolytic system is added, because
all the complement present is bound by the bacteria-immune serum
complex. On the contrary, if the inactivated serum is not specific
for the bacterial antigen, no union between the two will take place,
complement will not be bound, and it is free in the mixture. It will
activate the erythrocyte-inactivated immune serum complex, and
hemolysis will occur.

It will be seen that the hemolytic system is added as an indicator;
an absence of hemolysis shows a union of bacterial antigen, inactive
specific bacterial immune serum and complement. Hemolysis shows
that the union has not been formed, the complement was free in the
mixture and it united with the hemolytic system, causing hemolysis

156 ANTIGENS AND THE TECHNIC OF SERUM REACTIONS

in the erythrocyte antigen through the specific amboceptor or
hemolysin.

The method of complement fixation may be employed to examine
sera for specific antibodies, using a known antigen, or to test suspected
antigens with sera containing specific antibodies. The most practical
application of the method in medicine is the serum diagnosis of syphilis,
glanders, and other bacterial infections. t

The Technic of Complement Fixation. The technic of complement
fixation is simple in principle, but it requires the most scrupulous
attention to details. All glassware must be neutral in reaction,
chemically clean, and bacteriologically sterile. Physiological salt
solution (0.85 to 0.90 per cent. C.P. NaCl in neutral distilled water)
used for washing red blood cells and for dilutions should be sterile
and stored in clean containers.

The Wassermann Serum Diagnosis of Syphilis. Five elements enter
into the Wassermann test for syphilis: the antigen, suspected syphilitic
serum, complement, and a hemolytic system consisting of red blood
cells and specific immune hemolytic serum (hemolysin).

Preparation and Standardization of Antigen. The antigen originally
employed by Wassermann and his collaborators was an aqueous
extract of syphilitic tissue which was prepared by suspending one part
by weight of finely comminuted liver of a syphilitic fetus 1 in five parts
of physiological salt solution containing 0.5 per cent, phenol as a
preservative. After several days' violent agitation in the dark it is
strained through several layers of cheesecloth to remove coarser par-
ticles and stored in amber bottles in the refrigerator. Sedimentation
takes place until a brownish, slightly opalescent fluid remains, which
is the luetic antigen.

Later work 2 showed that alcoholic extracts of luetic liver were more
stable than watery extracts. The specific reacting component, accord-
ing to Forges and Meier, is lipoidal in nature, and in this sense it is
not biologically specific; The fixation of complement appears to
depend upon a substance in the antigen, lipoidal in nature, which
effects a union of antigen, immune body and complement. Citron
has proposed the term "lues reagine" for this substance. Alcoholic
extracts of syphilitic liver are prepared by shaking finely comminuted
liver with ten times the weight of absolute alcohol for a few days,

1 The tissue is examined for the specific organism; if Treponemata are abundant it
is converted into antigen, otherwise it is discarded.

2 Especially by Forges and Meier, Berl. klin. Wchnschr., 1908, No. 15.

LYSINS 157

then digesting the mixture at 37 C. for a week. The extract is
filtered through filter paper and placed in the refrigerator.

Alcoholic extracts of normal organs, prepared in the same manner
as luetic livers, have been found to be quite as good as alcoholic
extracts of syphilitic livers for the diagnosis of syphilis. In practice
heart-muscle of normal guinea-pigs, freed from all fat, is used.

Noguchi's Acetone-Insoluble Lipoidal Antigen. 1 Noguchi and others
have shown that .alcoholic extracts of organs may, and frequently
do, contain sufficient amounts of neutral fats, or their hydrolytic
cleavage products, to make the antigen hemolytic or anticomplemen-
tary. These substances are for the most part soluble in acetone, while
the antigenic fraction is insoluble in acetone. One part of fat-free
heart muscle or liver from a guinea-pig is cut into very fine pieces,
mixed with ten parts of absolute alcohol, and extracted in the incu-
bator at 37 C. for a week or ten days, being thoroughly shaken every
day. The soluble substances are freed from the fragments of tissue
by filtration through fat-free filter paper, and rapidly evaporated to
dry ness. 2 Sufficient ether is then added to take up the brownish
residue, and it is then allowed to stand until a clear, slightly colored,
ethereal solution is obtained, free from suspended material. The
ethereal solution is concentrated by evaporation to a point where
separation of a sediment begins, then it is poured into several volumes
(usually ten) of pure acetone. A voluminous precipitate forms at
once, and settles out as a tenacious gummy mass. This is retained,
under acetone, as the antigen. The acetone-soluble solution is dis-
carded. The antigen thus prepared consists largely of lecithins and
related substances. It keeps well and appears to be very sensitive
and reliable. From 0.2 to 0.3 gram are dissolved in a mixture of 1
c.c. of ether (free from alcohol and having a neutral reaction) and
10 c.c. of neutral absolute methyl alcohol. This solution is kept
in an amber bottle in the refrigerator as a stock antigen. One cubic
centimeter of this stock antigen is added to 19 c.c. of physiological
salt solution; this is the antigen used for the test.

Before making a test it is necessary to standardize the antigen.
It is essential to know the anticomplementary titer, that is, that
maximum amount of antigen which will inhibit hemolysis in the
presence of syphilitic serum, but which will not inhibit hemolysis

1 Noguchi, Serum Diagnosis of Syphilis.

2 Best by exposing the nitrate in a broad shallow dish to an air current from an
electric fan.

158 ANTIGENS AND THE TECHNIC OF SERUM REACTIONS

when non-syphilitic serum is used. In addition, the following deter-
minations are sometimes desirable.

The hemolytic titer, that amount of antigen which will of itself
cause lysis of red blood cells, and the antigenic titer, the amount of
complement it will absorb or "fix" in the presence of a definite amount
of specific syphilitic serum.

The anticomplementary titration is made by mixing graded amounts
of antigen and a constant amount of complement (0.1 c.c. of a 10 per
cent, solution 1 ) with constant amounts (0.1 c.c.) of known syphilitic
serum and normal serum, both inactivated.

The various mixtures are incubated in a water-bath at 37 C. for
an hour, then 0.2 c.c. of red blood cell suspension and inactivated
hemolytic serum are added and again incubated in the water-bath at
37 C. The maximum amount of antigen which will give complete
inhibition of hemolysis with syphilitic serum and no inhibition of
hemolysis in the non-syphilitic serum is regarded as the unit.

EXAMPLE OF AN ANTICOMPLEMENTARY TITRATION OF ANTIGEN.

Normal

Comple-

Hemolytic

Tube.

Antigen.

serum
Inactive,

ment, 10
per cent.

Red blood
cells, c.c.

serum
inactivated

Result.

c.c.

units.

0.2

0.1

0.10

1.0

1.5

Complete hemolysis.

0.4

0.1

0.10

1.0

1.5

" "

0.6

0.1

0.10

1.0

1.5

" "

0.8

0.1

0.10

1.0

1.5

CO 3

" "

1.0

0.1

0.10

5 ^

1.0

1.5

Partial inhibition.

1.5

0.1

0.10

J g

1.0

1.5

Marked inhibition.

2.0

0.1

0.10

1.0

1.5

Complete inhibition.

8 2

0.1

0.10

os y>

1.0

1.5

-2

Complete hemolysis.

9 3

0.10

1.0

1.5

Complete hemolysis.

Tube 5, containing 1.0 c.c. antigen, shows beginning inhibition of
hemolysis. This is regarded as the anticomplementary titer of the
antigen.

As a general rule, the hemolytic titer is higher than the anti-
complementary titer. The test is readily made, if desired, by using
the same amounts of antigen mixed with 1 c.c. of red blood cell
suspension and sufficient salt solution to bring the volume to 4 c.c.

It is customary to use one-fourth the anticomplementary titer as
the standard amount of antigen to be used in the actual test. In the

1 Prepared by adding fresh normal guinea-pig serum to physiological salt solution
in the proportion of one part serum to nine parts salt.

2 Serum control. 3 Hemolytic control.

LYSINS 159

example cited, 1.0 c.c. of the antigen was found to be anticomple-
mentary, consequently 0.25 to 0.3 c.c. would be the proper amount
of antigen to employ in the test.

Complement. Fresh guinea-pig serum is the usual source of com-
plement for fixation reactions. The animal should be healthy and
not previously injected with protein of any nature. The serum of
pregnant pigs is not trustworthy. Blood may be obtained directly
from the heart of the living animal by aspiration through a hypo-
dermic needle, from a severed carotid artery, or, more expeditiously
by cutting the throat of the animal, avoiding the esophagus, and
collecting the blood in sterile Petri dishes. The freshly drawn blood
is allowed to stand for a few hours at a low temperature and the serum
is pipetted off. Complement must be kept cold (below 16 C.) and in
the dark. It must be used fresh, for it deteriorates rapidly. In a
frozen condition, however, it will remain active for two or three
weeks. Both the "activating" and combining properties of normal
fresh guinea-pig serum are sufficiently constant for the reaction of
complement fixation.

Hemolytic System. (a) Hemolytic Serum (Hemolysiri). Hemolytic
serum is obtained from rabbits which have been injected with 2 c.c.,
4 c.c., and finally 6 c.c. of a 50 per cent, solution of washed sheep
red blood cells 1 at intervals of two or three days. The injections
may be made intraperitoneally or intravenously, the latter being
preferable. Not less than nine days after the injection the animal
is bled to death from the carotid artery under anesthesia, the blood
being received in sterile test-tubes, which are placed in an inclined
position in the ice-box. The serum is removed, centrifugalized if not
wholly free from blood corpuscles, and placed in small amber bottles
with aseptic precautions. These are heated to 56 C. for half an hour
to effect inactivation (to destroy complement).

(6) Red Blood Cells. Erythrocytes of the sheep are used. The
blood of a sheep is collected either in small sterile flasks containing
one volume of 0.85 per cent, salt solution and 0.5 per cent, sodium
citrate, or in sterile centrifuge tubes. If the former is used, nine
volumes of blood are allowed to flow into the flask and immediately
mixed intimately with the citrate solution, which prevents clotting.
This method is applicable if the blood cannot be centrifuged imme-

1 Fresh red blood cells of the sheep are freed from serum by repeated washings with
physiological salt solution usually five washings "suffice. The corpuscles are then
suspended in a volume of salt solution twice that of the corpuscles themselves.

160 ANTIGENS AND THE TECHNIC OF SERUM REACTIONS

diately. If centrifuge tubes are used, an amount of blood not more
than one-third the capacity of the tube (about 5 c.c.) is collected and
twice the volume of sterile salt solution is added to it. The corpuscles
are sedimented, the supernatant solution is pipetted off, fresh salt
solution is poured in, and the corpuscles resuspended by careful
stirring with a clean glass rod. This process is repeated five times,
each time discarding the washings. The last time the volume occu-
pied by the erythrocytes is read off on the graduations of the tube
and they are suspended in a volume of salt solution twenty times that
occupied by the erythrocytes. This makes a 5 per cent, suspension.
Erythrocytes are obtained by centrifugalization from the citrated
blood in precisely the same manner. This suspension of red blood
cells, kept in a cool, dark place, may be used for two days, but not
longer. Beyond that time the cells deteriorate and hemolyze with
abnormal readiness, thus vitiating the value of the test.

(c) Standardization of Hemolytic System. It is very important to
know with exactness the amount of hemolytic serum (inactivated,
of course) which will effect complete hemolysis of 1 c.c. of a 5 per
cent, suspension of sheep erythrocytes in the presence of a constant
amount of complement. The determination of this factor gives the
hemolytic titer of the hemolytic serum. It is readily determined
by adding to a series of tubes, 0.1 c.c. of fresh guinea-pig serum (com-
plement), 1 c.c. of erythrocyte suspension, and varying amounts of
the inactivated hemolytic serum. The smallest amount of hemolysin
which will effect hemolysis under the conditions stated is the hemolytic
titer orunit. Thus, the following tubes incubated at 37 C. for one
hour showed :

Result.
Complete hemolysis.

Partial hemolysis.
No hemolysis.

0.0025 of this serum is one unit; the hemolytic titer is 0.0025 c.c., in other words.
It is customary to use two units in the actual test, consequently 0.005 c.c. would be
the amount used.

Tube.

Complement.

c.c.

II 1

c.c.

12 2

0.0 c.c.

5 per cent, suspension
sheep erythrocytes.

Inactivated
hemolytic serum.

1 C.C.

0.10 c.c.

1 c.c.

0.075 c.c.

1 c.c.

0.050 c.c.

1 c.c.

0.025 c.c.

1 c.c.

0.010 c.c.

1 c.c.

0.0075 c.c.

1 c.c.

0.0050 c.c.

1 c.c.

0.0025 c.c.

1 c.c.

0.0010 c.c.

1 c.c.

0.00075 c.c.

1 c.c.

0. c.c.

1 c.c.

0. c.c.

Complement control.

2 Erythrocyte control.

LYSINS 161

It must be emphasized that precision of measurement is an absolute
requirement for success; the activating power of complement for
hemolysin does not follow the law of multiple proportions it is rather
an inverse ratio, as Noguchi 1 has pointed out. Relatively less com-
plement is required to induce complete hemolysis in a system contain-
ing four units than is required for a system containing but a single
hemolytic unit.

The serum to be examined for specific antibodies by the method
of complement-fixation must be sterile and free from hemoglobin.
The products of bacterial growths in serum may be anticomplementary
and the presence of hemoglobin in serum also tends to inhibit hemolysis.
Blood, therefore, should be withdrawn with aseptic precautions from
the median basilic vein of the patient into sterile test-tubes, and either
centrifugalized at once and the serum removed from the clot, or
placed in an inclined position in a cool place until the serum has
separated. The serum must be clear 2 and free from erythrocytes or
hemoglobin. 3 It is inactivated at 54 to 55 C. for half an hour in a
water-bath. 4

The Technic of the Test. It is essential that the hemolytic system
erythrocytes, hemolysin, complement be standardized daily.
Varying amounts of hemolysin are added to constant amounts of
erythrocyte suspension and complement, as outlined above. A
known positive syphilitic serum and a known negative syphilitic
serum, together with suitable controls, must be examined along with
the unknown serum to be tested.

The following reagents are required :

1. Sterile physiological salt solution.

2. Fresh guinea-pig serum (complement) add 0.1 c.c. to each tube.

3. Five per cent, suspension of washed sheep erythrocytes in normal
salt solution use 1 c.c. to each tube.

4. Hemolysin (amboceptor) use twice the hemolytic unit (the unit
must be determined daily).

5. Known syphilitic serum inactivate and use 0.2 c.c.

6. Known normal (non-syphilitic) serum, inactivated use 0.2 c.c.

7. The serum to be tested inactivate, use 0.2 c.c.

1 Serum Diagnosis of Syphilis.

2 Blood is best obtained early in the morning, before the patient has eaten; blood
obtained at the height of digestion frequently contains fats which make the serum
turbid.

3 Small amounts of blood, yielding a few drops of serum, may be obtained from the
finger-tip or the lobe of the ear. Massage must not be practised, for there is danger of
damaging erythrocytes with the liberation of hemoglobin.

4 Noguchi, Serum Diagnosis of Syphilis, states that inactivation at 54 C. should be
practised the higher temperature weakens the reactive substance somewhat.

162 ANTIGENS AND THE TECHNIC OF SERUM REACTIONS

Unknown serum.

Known positive
syphilitic serum.

Known normal
non-syphilitic serum.

Controls.

Tube 1.
Serum, 0.2 c.c.
Complement, 0.1 c.c.
Salt solution, 2.7 c.c.

Tube 3.
Serum, 0.2 c.c.
Complement, 0.1 c.c.
Salt solution, 2.7 c.c.

Tube 5.
Serum, 0.2 c.c.
Complement, 0.1 c.c.
Salt solution, 2.7 c.c.

Tube 7.

Complement, 0.1 c.c.
Salt solution, 2.9 c.c.

Tube 2.
Serum, 0.2 c.c.
Complement, 0.1 c.c.
Antigen, 1 1 c.c.
Salt solution, 1.7 c.c.

Tube 4.
Serum, 0.2 c.c.
Complement, 0.1 c.c.
Antigen, 1 c.c.
Salt solution, 1.7 c.c.

Tube 6.
Serum, 0.2 c.c.
Complement, 0.1 c.c.
Antigen, 1 c.c.
Salt solution, 1.7 c.c.

Tube 8.

Complement, 0.1 c.c.
Antigen, 1 c.c.
Salt Solution, 1.9 c.c.

After mixing the tubes are placed in a water-bath maintained at
37 C. for one hour, to permit of the fixation of complement; 1 c.c.
of a 5 per cent, suspension of erythrocytes and two units of hemolysin
are then added to each tube, mixed and reincubated for one hour,
then read. Tubes 1, 3, 5, 7, 6 and 8 should show complete hemolysis.
Tube 4 should show complete inhibition of hemolysis (positive reac-
tion). If such be the case all the reagents are properly adjusted, and
Tube 2, containing the unknown serum, is read. If hemolysis is
absent the reaction is positive; if hemolysis is complete the reaction
is negative. 2

The Method of Noguchi. 3 A rigorous standardization of reagents
is a prerequisite for accuracy in the serum diagnosis of syphilis, and
Noguchi has pointed out that a variable inherent inaccuracy exists
in the Wassermann method. He has shown that human sera may
contain variable amounts of hemolysin specific for sheep erythrocytes.
Human sera, however, contain no hemolysin for human erythrocytes.
The Noguchi modification, therefore, substitutes human red blood
cells (obtained from placenta or at autopsies) for sheep red blood
cells. Rabbits are immunized to carefully washed human erythro-
cytes and the hemolytic unit of the rabbit serum is determined in the
usual manner. The following reagents are required to perform the
Noguchi test:

1. Complement Fresh guinea-pig serum in 40 per cent, dilution
(one part clear fresh serum to 2.5 parts sterile salt solution).

2. Hemolytic Serum Rabbit serum, immunized against human
erythrocytes, is titrated against human erythrocytes to determine the
hemolytic unit. Two units are used in the test.

1 Twice the antigen titer, determined by titration, diluted with salt solution; thus, if
the antigenic titer of the acetone insoluble extract is 0.2 c.c., and the anticomplementary
titer is found to be 1.75 c.c., 0.4 c.c. of the extract are diluted with 0.6 c.c. salt solution
and used in the diluted state. In practice, enough extract should be diluted to last one
day.

2 For a discussion of results, see section on Treponema pallidum.

3 Noguchi, Serum Diagnosis of Syphilis.

PLATE I

Wassermann Reaction. (Simon.) *

A, positive; B, partial; C, negative reaction.

Note undissolved blood corpuscles in A, partial hemolysis in B, and complete hemolysis in C.

LYSINS

163

3. Human Erythrocytes Red blood cells are obtained from a
normal individual, washed thoroughly with salt solution, and made
up as a 1 per cent, suspension in salt solution. 1 c.c. of the suspension
is used in the test.

4. Antigen The acetone-insoluble antigen is used.

5. Patient's Serum Obtained fresh, from 2 to 5 c.c. of blood. It
is used unheated.

6. Known syphilitic serum.

7. Known normal (non-syphilitic) serum.
The test is performed as follows:

Unknown serum.

1
Known positive serum.

Known negative serum.

Controls.

Tube 1.
Serum, 1 drop.
Complement, 1 0.1 c.c.
Erythrocytes, 1.0 c.c.

Tube 3.
j Serum, 1 drop
Complement, 0.1 c.c.
Erythrocytes, 1.0 c.c.

Tube 5.
Serum, 1 drop.
Complement, 0.1 c.c.
Erythrocytes, 1.0 c.c.'

Tube 7.
Complement, 0.1 c.c.

Erythrocytes, 1.0 c.c.

Tube 2.
Serum, 1 drop.
Complement, 0.1 c.c.
Antigen, 2 units.
Erythrocytes, 1.0 c.c.

Tube 4.
Serum, 1 drop.
Complement, 0.1 c.c.
i Antigen, 2 units.
Erythrocytes, 1.0 c.c.

Tube 6.
Serum, 1 drop.
Complement, 0.1 c.c.
Antigen, 2 units.
Erythrocytes, 1.0 c.c.

Mix and incubate one hour in water-bath at 37 C. Remove and add 2 units hemolysin to each
tube and incubate in water-bath for one hour. Tubes 1, 3, 5, 6 and 7 should show complete hemo-
lysis. Tube 4 should show no hemolysis (positive control). If such be the case the reagents are
correctly adjusted and a reading of Tube 2 will be positive (no hemolysis) or negative (hemolysis).

A further simplification of the method has been made by Noguchi.
The hemolysin and antigen respectively may be absorbed on squares
of filter paper, dried, and standardized. In this state they retain
their potency for several weeks. In practice the squares of paper are
added directly to the tubes, thus saving much time.

Complement-fixation in Bacterial Infections. Preparation of Antigen
from Bacteria. Experience has clearly shown that bacterial antigens
should be polyvalent prepared by mixing in equal amounts, several
strains of the same organism. The antigen may be prepared in one
of several ways.

The simplest method is to wash off bacteria from agar slants,, at
the period of maximum growth, with salt solution and shake thoroughly
to make a uniform suspension. A small amount of phenol (0.5 per
cent.) and 3 per cent, glycerin are then added and the whole sterilized
at 56 to 60 C. for one hour. Relatively more of the proteins of the
bacterial cell may be obtained in solution if the bacterial emulsion
is shaken in a shaking machine with sterile, sharp quartz-sand for
twenty-four hours: filtration through coarse Berkefeld filters removes

1 Forty per cent, solution of fresh guinea-pig serum in salt solution.

164 ANTIGENS AND THE TECHNIC OF SERUM REACTIONS

the sand and broken bacterial cells, and the filtrate is preserved with
0.5 per cent, phenol. Besredka prepares a bacterial antigen from dried
bacterial cells, which are obtained by drying bacteria scraped from
agar slants or other solid media over sulphuric acid or calcium chloride.
The dried organisms are ground in agate mortars with crystals of
NaCl to an impalpable powder, which is then gradually rubbed up
in successive portions of water until a physiological salt solution is
obtained (corresponding to 8.5 grams NaCl in a liter of distilled water).

It has been found that much of the antigenic substance of bacteria
is precipitated by an excess of alcohol; a considerable excess of alcohol
is added to a suspension of bacteria, or to an emulsion of the cell
substance prepared according to Besredka's process, outlined above.
The precipitate from the alcoholic solution is separated by filtration,
dried, and ground to an impalpable powder with NaCl crystals.
The powder is gradually brought into solution by the addition of
water in successive amounts until isotonicity is reached. An attempt
is made to create a definite concentration of antigen by starting with
a known quantity of dried bacteria and a corresponding amount of
NaCl crystals. Thus, 1 gram of dried bacterial substance, ground
in a mortar with 0.85 gram NaCl crystals and gradually brought to
a volume of 100 c.c. with distilled water, would yield, theoretically,
an antigen of 1 per cent, strength. Bacterial antigens must be kept
cold and in a dark place, preferably in sealed amber bottles. Deter-
ioration gradually occurs and all bacterial antigens suspended or
dissolved in liquids are relatively unstable.

Standardization of Bacterial Antigens. The standardization of bac-
terial antigen differs in no respect from that of a syphilitic antigen.
The anticomplementary titer and the antigenic titer are determined,
the latter by titration with a specific immune serum.

The Diagnosis of Glanders by the Method of Complement-fixation
The antigen is prepared from glycerin-agar cultures 1 of several strains
of B. mallei incubated at 37 C. for forty-eight hours. The organisms
are autolyzed in distilled water for several hours at a relatively high
temperature (70 to 80 C.), then freed from suspended particles by
filtration through coarse Berkefeld filters. The filtrate is stored in
amber bottles in the ice-box after the addition of 0.5 per cent, phenol.

The anticomplementary titer is determined from a series of tubes
containing constant amounts of complement and graduated amounts

1 Reaction 1.5 per cent, acid to phenolphthalein.

AGGRESSINS 165

of antigen (1 to 20 dilution in salt solution). 1 The total volume of
complement and antigen is brought to 3 c.c. by the addition of salt
solution. After one hour's incubation in the water bath at 37 C., 1
c.c. of sheep erythrocyte suspension and 1.5 units sheep erythrocyte
hemolysin are added and reincubated. That dilution of antigen which
shows the slightest inhibition of hemolysis is taken as the anti-
complementary titer of the antigen. Not more than one-half this
amount, and preferably one-fourth of the anticomplementary titer,
is used in the test.

The actual determination is made in the same manner as for the
Wassermann test. 2 It is well to include a known positive and known
negative glanders serum of the same animal species as the unknown,
together with suitable controls of the hemolytic system. The length
of incubation is determined by the time it takes to effect complete
hemolysis in the known negative and the hemolytic controls. Fre-
quently ten or more hours will elapse before this occurs.

AGGRESSINS.

Progressively pathogenic bacteria appear to differ from parasitic
bacteria or "opportunists" in that they are able to force an entrance
to the underlying tissues of the host through natural, non-specific
barriers which ordinarily suffice to restrain the more parasitic types
of microbes. Bail 3 has advanced an hypothesis, based upon experi-
mental evidence, which attributes the invasiveness of pathogenic
bacteria and their ability to develop in the tissues of the host to
"aggressins." These aggressins, according to Bail, are present and
may be demonstrated in exudates resulting from bacterial infection,
but they are not, as a rule, found in artificial cultures of the same
organism. To demonstrate the action of aggressins, Bail removed
bacteria from exudates by centrifugalization and injected the clear
supernatant fluids, together with a sublethal dose of the homologous
bacterium, into experimental animals. Rapidly fatal infections
developed. The aggressin-containing exudates were not inactivated
by prolonged exposure to 50 C., and it was shown, furthermore, that

1 Usually a range of antigens from 2 c.c. to 0.05 c.c. will be found sufficient.

2 For full discussion of results, see Mohler and Eichhorn, Bureau of Animal Industry
Bulletin 136, April 7, 1911.

3 See Der Problem der bakteriellen Infektion, Bail, in Bibliothek medizinischer Mono-
graphien, xi; see also Miiller in Oppenheimer's Handbuch der Biochemie, 1909, ii, 1,
681.

166 ANTIGENS AND THE TECHNIC OF SERUM REACTIONS

their injection into susceptible animals stimulated the formation
of "antiaggressin," which greatly increased the resistance of the
animal to subsequent infection. The sera of animals immunized
with aggressin-containing fluids conferred a limited degree of immunity
to specific infections in non-immune animals (passive immunity).
It has been claimed by Doerr 1 and others that the aggressins are of
the nature of bacterial endotoxins and that the immunizing properties
of aggressin fluids are due to their content of specific substances
derived from the autolysis of bacterial cells.

The aggressin theory must, for the present, be regarded as not
definitely proved.

OPSONINS. TROPINS. BACTERIAL VACCINES.

A most important contribution to the literature of immunity is
the work of Denys and his associates, 2 who showed that the sera of
rabbits immunized to Streptococcus pyogenes possessed two properties
not exhibited by the serum of a normal animal, namely, the property
of restricting the development of the organism, and the property of
stimulating phagocytosis. Their very comprehensive studies demon-
strated that the leukocytes of normal animals, suspended in the serum
of immunized animals, phagocytized streptococci energetically, but
the leukocytes of immunized animals suspended in normal serum
failed to exhibit phagocytic activity. Their conclusion was that the
immunity of rabbits to the streptococcus resides in the serum. These
observations not only added materially to the restricted field in which
they were cast they brought sharply into focus the interrelation of
the humoral and cellular aspects of immunity.

Wright and Douglas, 3 using a modification of the technic of Leish-
man, 4 were able to study phagocytosis in vitro: by an ingenious series
of experiments they showed that normal serum contains substances
opsonins which prepare bacteria for phagocytosis, as described
in a preceding section (Cellular Immunity). The technic of meas-
uring the potency of opsonins in the sera of normal and infected
individuals, as practised by Wright and his associates, consisted

1 Wien. klin. Woch., 1906, No. 25.

2 Denys and Le Clef, La Cellule, 1895, xii; Bull, de 1'Acad. roy. de Belgique, 1895;
Denys and Marchand, Ibid., 1896; Van de Velde, Ann. Inst. Past., 1886, x; Marchand
Arch, de Med. exp., 1898; Denys, Cent. f. Bakt., 1898, xxiv, 685.

3 See Studies in Immunization, Constable, 1909, for complete biography.
Brit. Med. Jour., 1902, i, 73.

OPSONINS TROPINS BACTERIAL VACCINES 167

essentially in mixing intimately equal volumes of bacterial emulsion,
serum, and leukocytes; after incubation at body temperature the
mixture was spread evenly upon microscopic slides, stained, and
examined with the microscope. The average number of bacteria per
polymorphonuclear leukocyte was determined by direct count. A
comparison, under parallel conditions, of the phagocytic activity of
leukocytes for a specific organism in the serum of a normal individual
and that of an individual infected with the specific organism, accord-
ing to the technic outlined below, was called by Wright the opsonic
index.

Procedure. 1. Leukocyte Suspension. About 0.5 c.c. of blood,
drawn from the lobe of the ear or the tip of the finger, is collected in
a centrifuge tube containing 10 c.c. of sterile physiological salt solu-

FIG. 9. Phagocytosis of streptococci.

tion in which has been dissolved 1 per cent, of sodium citrate; this
mixture is centrifuged at moderate speed until a sharp separation
of blood cells and clear supernatant fluid is obtained. The super-
natant fluid is carefully poured off and the top layer of blood cells,
which contains practically all the leukocytes, is removed to a fresh
centrifuge tube containing 10 c.c. of physiological salt solution.

A second centrifugalization is made, and again the supernatant
fluid, containing the last traces of blood serum, is discarded. The
sediment, rich in leukocytes, is used as the leukocyte suspension in
the test.

2. Suspension of Bacteria. Bacteria from a culture on solid media
are suspended in sterile salt solution and agitated until a fine opales-
cent emulsion is obtained. This is most conveniently accomplished

168 ANTIGENS AND THE TECHNIC OF SERUM REACTIONS

in a shaking machine, but repeated shaking in a stoppered test-tube
containing glass beads will usually suffice. The coarser clumps of
bacteria are removed by filtration through a coarse filter paper. The
density of the bacterial suspension should be such that not more than
ten bacteria per leukocyte will be taken up as the average.

3. Serum. (a) Blood from three or four normal individuals is
collected in capillary tubes; after the serum has separated a "pool"
or mixture is made, composed of equal volumes of each serum. Experi-
ence has shown that "pooled" serum furnishes a more reliable normal
opsonic index than that obtained from a single individual.

(6) Serum from the Patient. This is prepared in the manner
described above.

The Test. A capillary pipette of 1 to 1.5 mm. bore is made by
drawing out a piece of glass tubing previously softened in the flame.
If the tubing is heated in the center until it softens, then, after a few
seconds, drawn slowly and steadily out, the desired size and shape is
readily obtained. A close-fitting rubber bulb attached to the larger
end is a convenience.

A mark about 1 to 1.5 cm. from the capillary end is made with a
wax pencil, and a volume each of the leukocytes, pooled serum, and
bacterial suspension are drawn into the pipette. It is convenient to
separate each ingredient by a small air bubble, to insure uniformity
of volume. Mixing is accomplished by carefully expelling and drawing
back the respective elements into the pipette. Finally, the mixture
is drawn well up into the pipette, the end is sealed in the flame of a
Bunsen burner, and the charged pipette is placed in the incubator
at 37 C. This is the normal or control.

A precisely similar preparation is made, using the serum of the
patient in place of the pooled serum.

Incubation is maintained for fifteen minutes.

The ends of the pipettes are now broken off, and the contents of
each pipette mixed as before. A large drop of each respective mixture
is spread upon clean glass slides, using the same technic as that for
preparing a blood smear, and air-dried. The preparations are stained
with Loffler's methylene blue, Wright's stain, or other stain suitable
for the organism used.

The number of bacteria in fifty, one hundred, or two hundred leuko-
cytes are determined by direct count, and the average number of
bacteria per leukocyte of the normal serum compared with the average
number of bacteria per leukocyte in the pathological serum:

OPSONINS TROPINS BACTERIAL VACCINES 169

EXAMPLE.

Bacteria in Bacteria per

100 leukocytes. leukocyte.

Staphylococcus suspension + pooled serum and

leukocytes 750 7.5

Staphylococcus suspension + patient's serum and

leukocytes 250 2.5

2 5

Opsonic index, patient's serum = '-- or 0.33 per cent.

7 . 5

Numerous observers have been unable to obtain uniform results
with the technic of Wright for opsonic index determination, and this
is not surprising when the many variable factors entering into the
method are reviewed. Attempts have been made to eliminate or
limit the variable factors: Simon proposed a dilution method in
which the pooled and patient's serum are diluted 1 to 10, 1 to 100,
etc., before incubation with the bacteria and leukocytes. That dilu-
tion of serum at which phagocytosis practically ceases in the normal
and patient's serum respectively is taken as a basis for comparison.
Inasmuch as the opsonic index is rarely determined as a guide for
treatment of bacterial disease with bacterial vaccines at the present
time, however, a discussion of these modifications, which are too
involved for practical use, is left for more pretentious volumes.

The Nature of Opsonins. There appears no doubt that the hypo-
thetical substance or substances called opsonin by Wright exist in
normal sera, and it is equally certain that they may be diminished
during infection. Furthermore, opsonin may be increased either in
amount or in potency by careful immunization. The relation of
opsonins to other antibodies, normal or specific, is a subject of con-
troversy at present. The researches of Neufeld and Rimpau, 1 Hek-
toen 2 and others indicate that the normal opsonins those of normal
sera are thermolabile, but those developed during immunization
to a specific organism bacteriotropins are relatively thermostabfte.

It has been suggested that opsonins or bacteriotropins are not to
be distinguished from other immune bodies as normal and specific
amboceptors or agglutinins. The rapidity with which the opsonic
index may be increased or diminished within a few hours following
injections of bacteria, however, would suggest a possible distinction
between these antibodies and the slowly developing specific bacteri-
cidal and agglutinating antibodies.

Vaccine Therapy. The value of vaccines and of autogenous vac-
cination in bacterial prophylaxis and bacterial immunization as set

1 Deutsch. med. Wchnschr., 1904, 1458.

2 Jour. Inf. Dis., 1906, iii, 434; 1909, vi, 78; 1913, xii, 1.

170 ANTIGENS AND THE TECHNIC OF SERUM REACTIONS

forth by Wright marks a distinct epoch in bacterial therapeutics in
spite of the practical failure of his opsonic index determination as
a theoretical guide to immunization and treatment. He has used
bacterial vaccines both for prophylaxis to prevent infection with
specific bacteria and therapeuitically to arrest infection.

Prophylactic Vaccination. The object of prophylactic vaccination is
to increase the resistance of the recipient to specific infection. This
is accomplished by reinforcing the natural initial defenses of the
body with specific antibodies, generated in the host in response to
the injection of the specific microorganism as a vaccine. In prophy-
lactic vaccination the host has ample time to work over the vaccine,
and by prolonging the treatment through repeated graduated doses
the maximum degree of immunity may be expected. To attain the
maximum immunizing effect the bacteria of the vaccine should be as
near their normal state as possible, that is, they should be endowed
with all the antigenic properties they possess in the natural disease
produced by them in the host.

Following the brilliant work of Jenner with cowpox vaccine and the
epoch-making observations of Pasteur, observers are fairly agreed-
that the best results from prophylactic vaccination are obtainable
only by the use of an attenuated living virus. The action of such a
living virus is, as Theobald Smith 1 has aptly expressed it, " a multitude
of feeble blows, each of which produces an immunological response."
The dangers attending the use of attenuated viruses, however, ordi-
narily preclude their employment, due to inability to control the
virulence of attenuated cultures. The possibility of creating carriers
cannot be overlooked. For this reason killed cultures are almost
invariably selected.

It is, of course, impossible to utilize an autogenous vaccine, but
for purposes of immunization a polyvalent vaccine is indicated. The
action of a dead virus is limited practically to a single immunological
response, hence the need of repeated inoculations.

Therapeutic Vaccination. In chronic, long-drawn out focal or local
infections, the invading microbes are either holding their own or
gaining the ascendency and the object of bacterial vaccination is to
turn the tables on the invaders. The products of immunization must
be used at once, arid the organisms comprising the vaccines for this
purpose cannot ordinarily be as resistant as their originals in the host.
The underlying principle of therapeutic vaccination, according to

i Jour. Am. Med. Assn., 1913, Ix, 1591.

OPSONINS TROPINS BACTERIAL VACCINES 171

Wright, 1 is to exploit the normal tissues of the body in the interest
of the infected tissue. For this purpose, microbes similar to those
causing the infection (autogenous organisms) are inoculated into
some other part of the body. This inoculation is not, to use Wright's
phraseology, a mere replica of the original infection; there are two
important points of difference: (1) the microbes of the vaccine are
killed, so that their multiplication within the host is impossible; (2)
the dose of vaccine must be so regulated that the tissues of the host
at the site of inoculation and elsewhere must inevitably win. Victory
of the host is brought about through the elaboration of specific anti-
bodies generated in the healthy tissues on a scale more than adequate
to bring about a destruction of the organisms introduced into the
healthy tissue. The surplus of the specific antibodies will find its
way, through blood and lymph channels, to the focus of infection,
and will reinforce the partially depleted defensive forces which have
ineffectually opposed the invading organisms.

It should be borne in mind that vaccine therapy cannot be reason-
ably applied unless an exact bacteriological diagnosis has been made.
The immunizing effects of vaccines are definitely limited by the
ability of the normal tissues of the patient to produce antibodies;
to inject too frequently or in too large doses may not only be barren
of results it may result in a decrease rather than an increase of
resistance to infection.

It is essential for the best results of vaccination that the focus of
infection be so situated anatomically that the newly formed antibodies
be drawn to the infected area by the production of local hyperemia.
Infections of long standing naturally respond to treatment more
slowly than newly acquired infections.

Preparation of Vaccines. Much discussion has arisen concerning
the use of autogenous vaccines as compared with stock or polyvalent
vaccines. So little is actually known of what vaccines may accomplish
in the body that it is impossible to answer this question definitely.
It is desirable, however, to retain in the vaccine all possible anti-
genie properties which were possessed by the organism in the body.
It is a well-known fact that certain kinds of organisms rapidly lose
their ability to produce disease when they are grown for any length
of time outside the body. Others retain their virulence for some
time. This would appear to indicate that stock vaccines of the former
would be unsatisfactory, while stock vaccines of the latter might be

1 Proc. Roy. Soc. of Med., London, 1910, iii.

172 ANTIGENS AND THE TECHNIC OF SERUM REACTIONS

more successful. It is a safe general rule to state that an autogenous
vaccine is desirable.

The preparation of vaccine is carried out as follows:

1. Obtain pure cultures of the organisms from the lesion or what-
ever material is available. The details of culture vary with the type
of organism that is expected.

2. Inoculation of the pure culture, or cultures in the event of mul-
tiple infection, in suitable media to furnish the desired amount of
growth.

3. Removal of the growth, with sterile precautions, to a sterile
container, such as a test-tube containing sterile glass beads. This
is accomplished by washing the growth from the medium into sterile
saline solution : 5 to 1 c.c. of salt solution are required for an ordinary
agar slant culture. When enough growth is accumulated it is trans-
ferred to the sterile test-tube, being careful that no organisms con-
taminate the upper part, else they may escape sterilization.

4. Sterilization: Heat the bacterial suspension in a water-bath.
Usually one hour at 60 to 65 C. suffices. Care must be taken that
the level of water in the water-bath is well above that of the level
of the suspension in the test-tube.

5. Test sterility of the suspension. Inoculate suitable media and
observe the absence of growth. In skin infections it is sometimes
desirable to exclude the presence of the tetanus bacillus.

6. Shake the suspension vigorously to distribute the organisms
uniformly in it.

7. Standardize: Determine the number of bacteria in a cubic
centimeter. This is very simply accomplished by thoroughly mixing
equal volumes of freshly drawn blood and bacterial emulsion in a
pipette, spreading the mixture on a microscope slide, drying and
staining it with Wright's or Jenner's stain. Determine by actual
counting in a number of fields the proportion of bacteria to red cells.
Knowing the number of red blood cells in a cubic centimeter of blood
(5,000,000,000) and the proportion of bacteria to red blood cells, it is a
simple matter to determine the number of bacteria in the suspension.

A more accurate procedure is to draw up one volume of vaccine in
the erythrocyte pipette of a hemocytometer, dilute to the 101 mark
with a dilute solution of fuchsin or other suitable stain, mix and
transfer to the counting chamber. An enumeration of the bacteria
is made in precisely the same manner that a blood count is made.

OPSONINS TROPINS BACTERIAL VACCINES 173

8. Dilute the suspension to the required degree with phenol, so
that the finished vaccine shall contain 0.25 to 0.5 per cent, of it. This
is the finished vaccine.

9. Redetermine sterility if necessary.

Sensitized Vaccines. Killed bacteria which have been immersed
in a specific serum sensitized vaccines are said to be less liable
to produce general and local reactions. The immunity developed
in response to the injection of these sensitized vaccines is said to
appear more rapidly, and doses thirtyfold those of unsensitized vac-
cines may be injected without serious effect.

The Injection. The skin at the site of injection is cleaned with
soap and water and then with alcohol; or better, after carefully dry-
ing it is painted with tincture of iodin. The required amount of
vaccine is injected subcutaneously through this area, from a sterile
syringe.

The Dosage and Frequency of Injection. It is advisable to begin
with small doses of vaccine, quantities which past experience has
shown to do no harm so far as can be determined by clinical evidence,
and to increase the size of the dose gradually, the injections usually
being given at intervals of about a week. If no change results from
the treatment, larger doses may be tried. If the symptoms become
aggravated the doses should be diminished and given at less frequent
intervals. Generally speaking, in the more acute cases smaller doses
should be selected to begin with, larger doses being reserved for the
more chronic cases. The amounts of vaccine to be injected vary widely
according to different investigators. Generally speaking, the following
figures are fairly representative :

Minimum. 1 Maximum. 1 Average. 1

Staphylococcus ...... 5.0 1000 25

Streptococcus 2.5 100 25

Pneumococcus 2.5 100 25

Goriococcus 2.5 300 30

Coli 5.0 1000 100

Pyocyaneus ........ 5.0 1000 100

Indications for the Use of Bacterial Vaccine. Generally speaking,
bacterial vaccines are contraindicated in acute disease, but may be
employed in practically any localized infection, or an infection which
has become chronic. 2

1 Figures represent millions of organisms.

2 An excellent discussion of the present status of vaccine therapy is that of Theobald
Smith, An Attempt to Interpret the Present-day Use of Vaccines, Jour. Am. Med. Assn.,
1913, Ix, 1591.

174 ANTIGENS AND THE TECHNIC OF SERUM REACTIONS

Results. Opinions differ widely as to the value of vaccines. Accord-
ing to the theory of bacterial vaccination, subacute and chronic infec-
tions which are localized should give the best results, and such indeed
appears to be the case. For example, a streptococcus septicemia
abates and leaves a joint involvement or a heart valve vegetation.
Vaccine therapy has a better chance of producing results during this
secondary stage than during the earlier acute septicemic stage. Gon-
orrheal arthritis, pneumonias which resolve by lysis, pus sinuses, and
localized colon infections are suitable for treatment. In acute inflam-
mations of the mucous membranes of the intestines, bladder, throat,
etc., the results have been either negative or unsatisfactory.

So far as specific organisms are concerned, staphylococcus vaccines
give the most constant and satisfactory results. Furuncles, severe
carbuncles, some cases of acne, and even low-grade staphylococcus
septicemias yield rather readily to vaccine therapy with this organism.
Streptococcic and pneumococcic infections are much more resistant,
generally speaking, to vaccine treatment than are staphylococcus
infections.

CHAPTER IX.

THE MICROSCOPIC AND CULTURAL STUDY OF
BACTERIA.

2. Capsules.

3. Polar Bodies.

4. Flagella.

F. Differential Stains for Bac-
teria.

1. Gram.

2. Ziehl-Neelsen.

3. Gabbett.

4. Polychrome. Stains.

B. Preparation of Stains. 5. Smith Sputum Stain.

C. Technio of Staining Bac- III. STAINING BACTERIA IN TISSUES.

METHODS FOR THE MICROSCOPIC STUDY

OF BACTERIA.
I. LIVING BACTERIA.

A. Hanging Drop.

B. Hanging Block.

C. Dark Ground Illumination.

D. Intra Vitam Staining.
II. STAINING OF BACTERIA.

A. Chemistry of Stains.

teria.

D. Intensive Stains for Bac-
teria.

IV. METHODS AND MEDIA FOR THE

CULTIVATION OF BACTERIA.
V. CULTIVATION OF BACTERIA.

E. Stains for Special Struc- 1 Inoculation of Cultures.

tures of the Bac-
terial Cell.

1. Spores.

Isolation of Pure Cultures.
Incubation of Cultures.
VI. STUDY OF BACTERIAL CULTURES.

METHODS FOR THE MICROSCOPIC STUDY OF BACTERIA.

BACTERIA may be examined directly under the higher powers of the
microscope for their morphology, motility, arrangement, method of
reproduction, and their behavior in specific sera, or they may be
stained with various anilin dyes and chemicals to bring out details
of structure or composition, and their relation to various tissues in
pathological processes.

Glass slides and cover-glasses are conveniently used for this purpose.
Microscopic slides should be made from clear, colorless glass. Cover-
glasses should be made of thin glass. The available working distance
of oil-immersion lenses is somewhat less than 1.5 mm., consequently
cover-glasses should not measure more than 1 mm. in thickness as
a maximum limit. Number 1 cover-glasses are suitable for bacterio-
logical work.

Glass slides and cover-glasses are best cleaned in a mixture of
potassium bichromate, 1 part; water, 4 parts; sulphuric acid, 6 parts.
The bichromate is dissolved in the water with the aid of heat and
cooled; the acid is added slowly with constant stirring. Immersion
in this mixture for twenty-four hours removes dirt and grease from

176 MICROSCOPIC AND CULTURAL STUDY OF BACTERIA

both slides and cover-glasses. The cleaned glassware is removed from
the cleansing bath and washed with running water until neutral to
litmus paper. It is stored either in slightly ammoniacal alcohol, or
dried with a soft cloth, previously freed from grease by boiling in a
5 per cent, sodium carbonate solution.

I. Examination of Living Bacteria. A. Hanging Drop. The motil-
ity, shape, and size of bacteria may be studied in a "hanging-drop"
preparation. A drop of fluid from a bacterial culture in liquid media

FIG. 10. Hollow-ground slide for hanging drop.

is transferred to the center of a thin cover-glass. If the growth is
upon solid media a drop of physiological salt solution 1 is placed upon
the center of the cover-glass as before, and a very small amount of
the culture is removed with a platinum needle and emulsified in it.
Next, the rim of the concavity in a " hollow-ground slide" is ringed
with vaselin and the cover-glass is inverted over it in such a manner
that the drop is suspended in the hollow, but touches neither the
sides not the bottom. The vaselin seals the preparation, causing it
to adhere to the slide, and also prevents evaporation. The prepara-
tion is now ready for microscopic examination. The one-sixth or one-
eighth-inch objective should be used, with the diaphragm partly closed
to reduce the intensity of illumination. It is desirable to focus first
upon the edge of the drop; the edge is sharply defined and readily
located. Bacteria are usually more numerous at the edge than in the
center of the drop.

B. Hanging Block. It is desirable occasionally to follow the
development of bacteria through several generations, to study the
germination of spores, or to examine special structures within the
bodies of individual organisms. The hanging-drop method is
unsuited for this purpose, which presupposes immobilization of the
organism. Hill 2 has invented an ingenious modification of the hang-
ing-drop method, the hanging block, which fulfils this requirement.
His directions for preparing it are:

"Pour melted nutrient agar into a Petri dish to the depth of about
one-eighth or one-quarter inch. Cool this agar and cut from it a block

1 Physiological salt solution is prepared by dissolving 8.5 grams NaCl in distilled
water 1000 c.c.

2 Jour. Med. Research, March, 1902, vii, 202.

METHODS FOR THE MICROSCOPIC STUDY OF BACTERIA 177

about one-quarter inch to one-third inch square and of the thickness
of the agar layer in the dish. This block has a smooth upper and
under surface. Place it, under side down, on a slide and protect it
from dust. Prepare an emulsion, in sterile water, of the organism to
be examined if it has been grown on a solid medium, or use a broth
culture; spread the emulsion or broth upon the upper surface of the
block as if making an ordinary cover-slip preparation. Place the
slide and block in a 37 C. incubator for five to ten minutes to dry
slightly. Then lay a clean sterile cover-slip on the inoculated surface
of the block -in close contact with it, carefully avoiding air-bubbles.
Remove the slide from the lower surface of the block and invert

FIG. 11. Warm stage, electrically heated, for the cultivation of bacteria.

the cover-slip so that the agar block is uppermost. With a platinum
loop, run a drop or two of melted agar along each side of the agar
block, to fill the angles between the sides of the block and the cover-
slip. This seal hardens at once, preventing slipping of the block.
Place the preparation in the incubator again for five or ten minutes,
to dry the agar-agar seal. Invert this preparation over a moist 4.
chamber and seal the cover-slip in place with white wax or paraffin.
Vaselin softens too readily at 37 C., allowing shifting of the cover-
slip. The preparation may then be examined at leisure. 1

1 A light, detachable, electrically heated warm-stage incubator, manufactured by
the Chicago Surgical and Electrical Company according to specifications furnished
by the writer is very satisfactory for this purpose. Bacteria may be maintained con-
stantly at any desired temperature between that of the room and 45 C. for several
days, and observed continuously without difficulty. If the warm-stage incubator is
attached to a graduated mechanical stage, many individual bacteria may be observed
in the same preparation by recording their respective positions as indicated on the
graduated rectilinear stage verniers.
12

178 MICROSCOPIC AND CULTURAL STUDY OF BACTERIA

C. Dark Field Illumination and Ultramicroscopic Examination. For
the study of very minute particles in suspension, the ultramicroscope
of Siedentoff and Zsigmondy 1 has been used, but the dark-ground
illumination apparatus of Reichert, 2 a much simpler device, readily
adjusted to any microscope, has largely supplanted it for bacterial
examinations. With the Reichert apparatus the flagella of bacteria
and other structures of low-refractive index may be observed. Tre-
ponema pallidum in fresh smears from lesions is readily seen with the
dark ground illuminating apparatus.

D. Intra Vitam Staining. Nakanishi 3 has applied the method of
infra vifam staining to the study of spores and granules in living bac-
terial cells. The method consists essentially in emulsifying a small
amount of bacterial growth in normal salt solution containing suffi-
cient aqueous methylene blue to impart a distinct blue color to the
solution. The preparation is viewed as in the hanging-drop slide.
The organisms absorb sufficient dye to impart to them a faint color,
and granules within their bodies frequently stain with moderate inten-
sity. The development of spores from pre-sporogenic granules may be
studied by this method.

II. Staining of Bacteria. A. Chemistry of Stains. The stains of
value for coloring bacteria are almost exclusively anilin dyes which
contain one or, more commonly, several benzene rings. Their color-
ing properties have been shown to depend upon two distinct radicals;
double-bonded atoms as C = C, C = O, C = N, N = N, known as
chromophoric groups, and auxochromic groups, which impart to or
intensify the color. Of the chromophoric groups, NH 2 and OH are
the more important. The latter form salts which may be either basic
or acid in character. Bacteria usually stain best with basic dyes, as
do nuclei of higher plant and animal cells.

The chemistry of the staining process itself is a matter of discussion.
It was formerly held that the cell protoplasm united chemically with
the stain as an acid unites with a basic salt, but later investigations,
particularly those of Michaelis, 4 are not in harmony with this view.
It is probable that the physical state of the cell membrane as well
as the composition of the cytoplasm play a part in the staining process.

B. Preparation of Stains. Stains prepared by Griibler or Merck
are commonly used for the staining of bacteria. They are conveniently

1 Zeit. f. wissenschaftl. Mikroskopie, 1909, xxvi, 391.

2 Miinchen. med. Wchnschr., 1906, 2351; Hyg. Rund., 1907, No. 18; Cent. f. Bakt.
Orig., 1909, li, 14.

3 Miinchen. med. Wchnschr., 1900, No. 6; Cent. f. Bakt., 1901, xxx, 97, 145, 193, 225.

4 Einfiihrung in die Farbstoffchemie, 1902, Berlin.

METHODS FOR THE MICROSCOPIC STUDY OF BACTERIA 179

kept in stock as saturated aqueous or alcoholic (96 per cent.) solutions.
The solubility of stains in water and in alcohol respectively varies,
but, as a rule, the solubility in alcohol is greater than that in water.
Saturated solutions of anilin dyes are unsuited for the staining of
microorganisms, but they are more stable than diluted solutions
provided they are kept in tightly-stoppered bottles away from the
light. Dilutions of saturated solutions are prepared as they are
needed for current use.

C. Technic of Staining Bacteria. 1. Preparation of a film of bac-
teria for staining: A drop of a culture of bacteria from a fluid medium
as broth is removed with a platinum loop and spread upon a clean
cover-glass or glass slide. Bacteria from a solid medium are emul-
sified in a small drop of water on the slide. 1

2. The film of bacteria is allowed to dry in the air; evaporation
may be hastened in the incubator at 37 C.

3. The air-dried film is next fixed by passing it once slowly through
the flame of a Bunsen burner, film side upward; about one-half second's
exposure to the flame suffices; a longer exposure destroys or changes
the staining properties of the organisms.

4. Staining: A 5 per cent, aqueous solution of methylene blue,
fuchsin, or gentian violet, prepared by adding 5 c.c. of filtered satu-
rated stock solution to 95 c.c. of distilled water, is used. The slide
or cover-glass is flooded with the desired stain, and after one to five
minutes, depending upon the intensity of the stain used, the excess
is poured off and the preparation is washed thoroughly with water.
The residual moisture is removed with filter paper or by air-drying,
and a small drop of Canada balsam (dissolved in xylol) is placed in
the center of the stained area. The film is finally enclosed between a
slide and a cover-glass.

D. Intensive Stains for Bacteria. Simple aqueous or alcoholic solu-
tions of anilin dyes are frequently inefficient for staining bacteria and
resort is made to intensified stains. One of the most useful of the
intensified stains is Loffler's alkaline methylene blue, prepared in the
following manner:

1 to 10,000 aqueous solution of potassium hydroxide 2 70 c.c.

Saturated alcoholic solution methylene blue 30 c.c.

1 It is essential that the emulsion shall be but faintly opalescent when viewed by
reflected light; a distinct clouding indicates that too many organisms have been added,
in which event the preparation will be found to be unsatisfactory.

2 Conveniently prepared by dissolving 1 gram of KOH in 100 c.c. distilled water
and adding 1 c.c. of this solution to 99 c.c. of distilled water.

180 MICROSCOPIC AND CULTURAL STUDY OF BACTERIA

Fixed films of bacteria are stained from one to five minutes with
this stain, or the films are flooded with the stain and heated until
steam rises (not bbiled) for one to three minutes. It is difficult to
overstain with Loffler's methylene blue unless evaporation takes
place to such a degree that the stain dries on the slide. The stain is
washed off with water, dried, and mounted.

E. Stains for Special Structures of the Bacterial Cell. 1. Spores.
(a) Flood fixed film of bacteria with carbol-fuchsin 1 and steam (not
boil) for five minutes.

(6) Wash thoroughly in running water.

(d) Wash thoroughly in running water.

(e) Flood with saturated aqueous solution methylene blue (or
Loffler's alkaline methylene blue) and allow to stain one minute.

(/) Wash in water, dry, and mount.
Spores stain red, vegetative cells blue. v

Holler's Spore Stain: 2 (a) Suspend the fixed film of bacteria in
chloroform for two minutes.
(6) Wash with water.

(d ) Wash thoroughly in running water.

(e) Flood with carbol-fuchsin and steam for five minutes.
(/) Wash thoroughly in water. . \* i

(g) Decolorize with 1 per cent, sulphuric acid until excess stain is
removed.

(h) Wash thoroughly in water.

(i) Flood with Loffler's alkaline methylene blue and allow to stain
one minute.

(j) Wash in water, dry, and mount.

Spores stain red, vegetative cells blue.

2. Capsule Stains. Welch Method* (a) Fixed films of bacteria are
flooded with glacial acetic acid for a few seconds.

(6) The acid is poured off and the preparation is washed two or
three times with anilin oil gentian violet, then flooded with the stain,
which is allowed to act for three to five minutes.

(d) Mount in salt solution and examine.
Capsules faint purple, bacterial body deep purple.

1 Saturated alcoholic solution of basic fuchsin, 10 c.c.; 5 per cent, aqueous phenol
solution, 90 c.c.

2 Cent. f. Bakt., 1891, x, 273. 3 Johns Hopkins Hosp. Bull., 1892, 128.

METHODS FOR THE MICROSCOPIC STUDY OF BACTERIA 181

Hiss's Method. 1 (a) Place a drop of sterile blood serum upon a
slide and emulsify bacteria in it.

(b) Dry in the air and fix by heat.

(d) Remove excess of stain by washing in a 20 per cent, solution
of copper sulphate.

(e) Dry with filter paper. Mount and examine.

Capsule faint pink or purple; body of organism deep red or purple.
Rosenow Method. 3 (a) Prepare the smear on perfectly clean cover-
glass.

(b) When smear is nearly dry, cover with 10 per cent, aqueous
solution of tannic acid for twenty seconds.

(d) Flood with anilin-oil gentian violet and steam gently for thirty
to sixty seconds.

(e) Wash thoroughly in water. ^

(/) Cover with Gram-iodin solution, one minute.
(g) Decolorize with 96 per cent, alcohol.

(h) Stain one or two minutes with a saturated (60 per cent.) alco-
holic solution of Griibler's eosin.

(i) Wash in water; dry and mount in balsam.
Capsules pink; bacteria blue.
3. Polar Bodies. -Neisser Stain. 4
Preparation of Stain.

Solution A Methylene blue 1 gram

Ninety-six per cent, alcohol ........ 20 c.c.

Glacial acetic acid 50 c.c.

Distilled water 950 c.c.

Solution B Bismarck brown 1 gram

Distilled water 500 c.c.

(a) The air-dried film, fixed by heat, is flooded with solution A for
three to five seconds.

(b) Wash with water.

(d) Wash with water, dry, and mount.
Polar bodies stain blue; bacterial cells brown.

1 Jour. Exp. Med., 1905, vi, 338.

2 Saturated alcoholic solution of the dye, 5 c.c.; distilled water, 95 c.c.

3 Jour. Infect. Dis., 1911, ix, 1.

4 Ztschr. f. Hyg., 1897, xxiv, 443.

182 MICROSCOPIC AND CULTURAL STUDY OP BACTERIA

4. Flagella. Preparation of Film. 1 (a) Add enough of an eighteen
to twenty-four-hour agar culture to a test-tube containing 5 c.c. of
sterile salt solution to produce a faint turbidity in the upper half of
the solution.

(6) Incubate at 37 C. for thirty to sixty minutes.

(c) Place two or three loopfuls of the suspension upon a perfectly
clean cover-glass and allow to dry spontaneously in the air or in the
incubator.

Do not attempt to spread the films with the platinum loop; agita-
tion breaks off flagella.

Staining Flagella. Pittsfield's Flagella Stain. 2

Preparation of Stain.

(a) Mordant:

Tannic acid, 10 per cent, aqueous solution 10 c.c.

Mercuric chloride, saturated aqueous solution 5 c.c.

Alum, saturated aqueous solution 5 c.c.

Carbol fuchsin 5 c.c.

(b) The Stain:

Alum, saturated aqueous solution 10 c.c.

Carbol fuchsin 5 c.c. or,

Gentian violet 2 c.c.

Flood the dried and fixed film with the mordant and steam gently
for one minute. Wash in running water, air-dry and flood with the
stain. Heat gently two minutes, wash thoroughly in water, air-dry
and mount.

F. Differential Stains for Bacteria. 1. Gram Stain. 3 A most impor-
tant differential method of staining bacteria is that, of Gram. Bacteria
may be divided into two groups: those which retain the initial stain
Gram-positive organisms and those which fail to retain the initial
stain but color with the counter stain Gram-negative bacteria.

It was believed formerly that the organisms which retained the
initial stain the Gram-positive bacteria contained within their
protoplasm, a substance of unknown composition which united
chemically with gentian violet (or other pararoseanilin dye) and
iodin to form a compound relatively insoluble in alcohol. Gram-
negative bacteria did not contain the hypothetical substance, which,
in association with the dye and iodin, was insoluble in alcohol. Treat-
ment of the latter group with alcohol, therefore, would remove the

1 Kendall, Jour. Applied Microscopy, 1901, v, 1836.

2 Medical News, September 7, 1895.

3 Gram, Fortschr. d. Med., 1894, ii,

METHODS FOR THE MICROSCOPIC STUDY OF BACTERIA 183

gentian violet, leaving them unstained. In the unstained condition
the organisms were colored with the second or counter stain. Subse-
quent investigation has largely discredited this view. It has been
shown by Kruse 1 that the cytoplasm of Gram-positive bacteria is
more resistant to autolysis, to the action of trypsin, and to solution in
dilute KOH than that of Gram-negative organisms, probably because
the cytoplasm of the former is less permeable to these various reagents
than is that of the latter. Eisenberg, 2 and Guerbet, Mayer and Schaef-
fer 3 have advanced the hypothesis that Gram-positiveness is due to
the lipoidal content of the cell membrane (ectoplasm) and specifically
to unsaturated fatty acids and phosphatids. The addition of iodin,
according to this theory, through the formation of alcohol-insoluble
combinations with the lipoids in the ectoplasm, renders the cell wall
impermeable to alcohol and thus prevents removal of the dye which
has already penetrated into the cell contents.
Preparation of Stain:

Solution A Saturated aqueous solution of anilin 4 90 c.c.

Saturated alcoholic solution of gentian violet . . 10 c.c. or,

Five per cent, aqueous solution of carbolic acid . . 90 c.c.

Saturated alcoholic solution of gentian violet . . 10 c.c.

The above solutions are unstable, but retain their tinctorial value
for two or three days if they are kept stoppered.

Solution B 5 Distilled water 300 c.c.

Potassium iodide 2 grams

Iodin crystals 1 gram

Solution C Bismarck brown, saturated aqueous solution ... 10 c.c.
Distilled water 90 c.c.

Procedure. (a) Prepare and fix film of bacteria in the usual manner.
(6) Flood with anilin-oil gentian violet (or carbolic gentian violet)
and stain five minutes.

(d) Decolorize with 96 per cent, alcohol until no more stain can be
removed.

(e) Wash thoroughly in water.

(/) Counterstain with Bismarck 1 brown 6 13r two minutes.
(g) Wash in water, dry, and mount.

1 Miinchen. med. Wchnschr., 1910, p. 685.

2 Cent. f. Bakt., 1909, xlix, 465; 1910, li, 115; liii, 481, 551; Ivi, 183.

3 Compt. rend., Soc. biol., Ixviii, 353.

4 Three c.c. of anilin oil are shaken for several minutes in 100 c.c. of distilled water.
The solution is filtered through filter paper to remove the undissolved anilin.

5 This iodin solution is variously known as Gram's iodin solution or Lugol's solution.

6 Dilute aqueous fuchsin, 1 to 10, may be used in place of Bismarck brown.

184 MICROSCOPIC AND CULTURAL STUDY OF BACTERIA

Bacteria which retain the initial stain Gram-positive bacteria
are colored dark purple or blue; those which fail to retain the initial
stain Gram-negative bacteria are brown, or bright pink if fuchsin
is used as a counterstain.

2. Stains for Acid-fast Bacteria. Ziehl-Neelsen Method. 1 (a) Stain
dried and fixed smear with carbol fuchsin, as described on page 180.

(6) Wash thoroughly with water.

(d) Wash with water.

(e) Counterstain lightly with Loffler's alkaline methylene blue.
(/) Wash, dry, and mount.

Acid-fast bacilli and spores red; all others blue.

3. Frdnkel-Gabbet Method. 3 (a) Stain with carbol fuchsin as in
the Ziehl-Neelsen method and wash in water.

(b) Decolorize and counterstain simultaneously with the following
solution :

Methylene blue 2 grams

Water . 75 c.c.

Sulphuric acid 25 c.c.

The counterstain is allowed to act for one minute.

4. Polychrome Stains. Polychrome stains are of special value for
the examination of exudates, body fluids or tissues in which the his-
tological relations of bacteria are to be investigated. These stains,
or modifications of them, are also useful in the study of treponemata,
spirochetes, and protozoa.

Wright's Stain. 4 Preparation. "To a 0.5 per cent, aqueous
solution of sodium bicarbonate add methylene blue (B.X., or ' medi-
cinally pure') in the proportion of 1 gm. of the dye to each 100 c.c.
of the solution. Heat the mixture in a steam sterilizer at 100 C. for
one full hour, counting the time after the sterilizer has become thor-
oughly heated. The mixture is to be contained in a flask, or flasks,
of such size and shape that it forms a layer not more than 6 cm. deep.
After heating the mixture is allowed to cool, placing the flask in cold
water if desired, and is then filtered to remove the precipitate which

1 Ziehl, Deutsch. med. Wchnschr., 1882, 451; Neelsen, Fort. d. med., 1885, 200.

2 Ninety per cent, alcohol containing 3 per cent, by volume of hydrochloric acid.

3 Frankel, Berl. klin. Wchnschr., 1884; Gabbet, Lancet, 1887.

4 Mallory and Wright, Pathological Technic, 1915, 6th ed., p. 382.

METHODS FOR THE MICROSCOPIC STUDY OF BACTERIA 185

has formed in it. It should when cold have a deep purple red color
when viewed in a thin layer by transmitted yellowish artificial light.
It does not show this color while it is warm.

"To each 100 c.c. of the filtered mixture add 500 c.c. of a 0.1 per
cent, aqueous solution of 'yellowish, water-soluble' eosin and mix
thoroughly. Collect on a filter the abundant precipitate which imme-
diately appears. When the precipitate is dry, dissolve it in methylic
alcohol (Merck's 'reagent') in the proportion of 0.1 gm. to 60 c.c.
of the alcohol. In order to facilitate solution the precipitate is to be
rubbed up with alcohol in a porcelain dish or mortar with a spatula
or pestle.

"This alcoholic solution of the precipitate is the staining fluid. It
should be kept in a well-stoppered bottle because of the volatility of
the alcohol. If it becomes too concentrated by evaporation and thus
stains too deeply, or forms a precipitate on the blood smear, the
addition of a suitable quantity of methylic alcohol will quickly correct
such faults. It does not undergo any other spontaneous change than
that of concentration by evaporation.

"A most important fault met with in the working of some samples
of this fluid is that it fails to stain the red blood corpuscles a yellow or
orange color, but stains them a blue color which cannot readily be
removed by washing with water. This fault is due to a defect in the
specimen of eosin employed. It can be eliminated by using a proper
'yellowish, water-soluble' eosin."

Method of Staining. (a) Unheated air-dried films 1 are covered
with the stain, which is allowed to act for one minute.

(6) Add an equal volume of distilled water to the stain and allow to
stand for three minutes.

(d) Dry rapidly with filter paper and mount in balsam. 2
Giemsa Method. 3 Preparation of Stain:

Azur II (eosin) 3.0 grams

Azur II 0.8 grams

Glycerin, C. P 250 c.c.

Neutral absolute methyl alcohol 250 c.c.

The dyes are dissolved in the glycerin at 60 C.; the alcohol, warmed
to 40 C., is then added, thoroughly mixed by shaking, and allowed
to cool slowly to room temperature, then filtered. Immediately before

1 Films more than a few hours old do not stain as well as fresh ones.

2 The balsam must be neutral in reaction.

3 Giemsa, Cent. f. Bakt., 1904, xxxvii, 308.

186 MICROSCOPIC AND CULTURAL STUDY OF BACTERIA

use, 10 c.c. of distilled water are slightly alkalinized by the addition
of two drops of a 10 per cent, solution of sodium carbonate, and
exactly ten drops of the stain are then added.

Staining with Giemsa Solutions. (a) Films are fixed by immersion
in neutral absolute methyl alcohol for one minute, air-dried, and
covered with the diluted stain, which is allowed to act for fifteen to
twenty minutes when ordinary exudates and bacteria are used; for
one to three hours if Treponemata or Negri bodies are sought for.

(b) Wash in water, dry and mount.

5. W. H. Smith's Solution Stain. (a) Stain the fixed smear with
anilin oil gentian violet for one minute.

(b) Wash with water.

(d) Decolorize with 95 per cent, alcohol.

(e) Wash with ether for a few seconds.

(/) Flood with absolute alcohol for five seconds.

(g) Stain with saturated aqueous solution eosin for one to two
minutes.

(k) Wash with absolute alcohol for a few seconds.

(i) Clear with xylol.

( j) Mount in balsam.

III. Staining Bacteria in Tissues. Paraffin sections are preferable,
partly because very thin sections may be cut; chiefly because celloidin
stains somewhat with the stains ordinarily used.

The Gram-Weigert Stain for Bacteria in Tissues. 1 (a) Stain paraffin
sections with anilin oil methyl violet for five to twenty minutes.

(6) Wash in water to remove excess of stain.

(d) Wash in water to remove excess of iodin.

(e) Decolorize with several changes of absolute alcohol until no
more color comes out.

(/) Clear section in xylol.
(g) Mount in neutral xylol balsam.

Mallory and Wright Modification for Celloidin Sections. 2 (a) Stain
sections with lithium carmine for two to five minutes.

(b) Remove excess of stain with acid alcohol.

(d) Dehydrate in 95 per cent, alcohol.

1 Mallory and Wright, Pathological Technic, 6th ed., 1915, p. 432.
* Ibid.

METHODS FOR THE MICROSCOPIC STUDY OF BACTERIA 187

(e) Expose to ether vapor to fix section to slide.
(/) Stain with anilin oil methyl violet for five to twenty minutes.
(g) Remove excess stain with normal salt solution.
(h) Gram-iodin solution for one minute.
(i) Remove excess iodin with water.

( j) Remove moisture as thoroughly as possible with filter paper.
(k) Dehydrate in several changes of anilin oil.
(/) Clear with several changes of xylol.
(m) Mount in neutral xylol balsam.

Staining Tubercle Bacilli in Tissues. (a) Paraffin sections are
covered with carbol-fuchsin and steamed gently for five minutes.

(b) The excess stain is removed with water.

(d) Remove excess of stain and acid with water.

(e) Dehydrate with absolute alcohol.
(/) Clear section in xylol.

(g) Mount in xylol balsam.

Staining Actinomyces in Tissues Mallory Method. 1 (a) Stain
paraffin sections with saturated aqueous eosin for ten minutes.
(6) Remove excess stain with water.

(d) Remove excess stain with normal salt solution.

(e) Remove excess water with filter paper.
(/) Clear in anilin oil.

(g) Remove anilin oil with several changes of xylol.

(h) Mount in neutral xylol balsam.

The clubs stain pink, the filaments blue.

IV. Methods and Media for the Cultivation of Bacteria. One of
the most important procedures in bacteriology is the preparation
of nutritive media in which the morphology, chemistry, and cultural
characteristics of the organism may be studied; furthermore, it is
possible by cultural methods to separate one type of bacterium in
pure culture from associated organisms, and to study its reactions
apart from all contaminating microorganisms. The technic of isolat-
ing and cultivating bacteria is exacting at every step of the process,
from the preparation of glassware to the selection of suitable nutritive
media, and their preparation requires not only scrupulous cleanliness;
it necessitates a most rigorous maintenance of sterility.

1 Loc. cit., p. 433.

188 MICROSCOPIC AND CULTURAL STUDY OF BACTERIA

Bacterial cultivation is usually carried out in glass vessels test-
tubes, flasks, fermentation tubes, and Petri dishes because glass is
transparent and permits an unobstructed view of the reactions taking
place within. It is obvious that glassware employed in bacterial
laboratories must be chemically and bacteriologically clean.

Preparation of Glassware. The method to be employed in the clean-
ing of glassware depends somewhat on the purpose for which it is

FIG. 12. Petri dish.

used. New glassware frequently contains alkali, which is readily
neutralized by diluted acid, hydrochloric or sulphuric. Glassware
that has contained cultures of bacteria is first sterilized in the auto-
clave to remove all danger of infection, then immersed in a strong
solution of soap-powder and soap-suds maintained at a boiling tem-
perature for half an hour. The adherent media is removed with a
brush or swab; a final thorough rinsing in clear water removes all

FIG. 13. Fermentation tubes various types.

traces of soap. Very dirty glassware or glassware in which chemical
determinations are to be made should be cleaned in chromic acid
solution, which is prepared by adding a saturated aqueous solution
of potassium bichromate to a 1 to 3 dilution of sulphuric acid. Twenty-
four hours' exposure to chromic acid removes all traces of organic
matter, as a rule. Following the acid bath the glassware is thoroughly
rinsed in clear water and dried.

METHODS FOR THE MICROSCOPIC STUDY OF BACTERIA 189

The cleaned glassware test-tubes, flasks, or fermentation tubes
is then stoppered with non-absorbent cotton cotton batting which
has a long staple or fiber. The cotton plugs must be carefully fitted
neither too loose, which would permit of the passage of adventitious
microorganisms, nor too tight, for obvious reasons. The cotton plugs
are conveniently prepared from a layer of cotton batting about two
inches square (for a test-tube of ordinary diameter, about 15 mm.),
which is laid squarely over the orifice. The center of the square is
gently pushed down into the neck of the tube for a distance of about
three-fourths to one inch; sufficient cotton protrudes from the tube
to be conveniently grasped by the fingers and removed. It is fre-
quently advisable to cover the cotton plugs with two or three layers
of filter paper, which prevents an accumulation of dust on the cotton.
Wide-mouthed containers are sealed with several layers of unglazed
paper fastened in place with a piece of twine. Flasks are frequently
not plugged with cotton; the neck is simply covered by an inverted
beaker of appropriate size.

Glassware should always be sterilized before media is placed in
it; this is readily accomplished by dry heat. A hot-air sterilizer is
used, in which a temperature of 180 C. is maintained for one hour.
A higher temperature must be avoided, to prevent charring of cotton
plugs. The heat must be increased gradually and diminished gradually,
to prevent cracking of the glass. By this process not only is the
utensil rendered sterile, the plugs of cotton retain their shape when
withdrawn, as well.

A majority of the bacteria pathogenic for man and many parasitic
and saprophytic forms as well require relatively complex organic
compounds containing carbon, hydrogen, nitrogen, and oxygen,
together with other elements for their nutrition. These foodstuffs
provide both the structural and fuel requirements of the organism,
as explained in the chapter on Bacterial Metabolism. Experience has
shown that a medium containing meat infusion, peptone, and salt is
a satisfactory one for many bacteria. This medium may be enriched
by the addition of various ingredients to meet the requirements of the
more fastidious organisms.

Meat infusion is prepared from finely comminuted lean meat 1 freed
from fat. 500 grams of meat are intimately mixed with 1000 c.c. water
and allowed to infuse over night in the refrigerator. It is then strained

1 Beef hearts make a very satisfactory meat infusion and their cost is much less than
the better cuts of meat.

190 MICROSCOPIC AND CULTURAL STUDY OF BACTERIA

through several layers of cheese-cloth, the volume recorded, then
heated to boiling. The coagulum which forms is removed by filtration
through filter paper and the clear, amber-colored fluid, after restoring
the loss due to evaporation, is run into flasks and sterilized in an
autoclave at 15 pounds' pressure for fifteen minutes. This plain meat
infusion contains but little protein; it is relatively rich, however, in
soluble meat extractives, soluble salts and muscle-sugar dextrose.
It is not suitable in itself as a complete nutritive medium for most
bacteria, but it forms the basis of many of the commonly used nutri-
tive media. Meat extract (Liebig's or other kinds) is frequently
substituted for meat infusion. Three grams of meat extract are dis-
solved in a small volume of water, filtered through a cold wet filter
paper to remove fat, and made up to a volume of a liter. The solution
contains some meat extractives, including a relatively large propor-
tion of xanthin bases and a very considerable amount of salts, par-
ticularly sodium chloride. Little or no muscle-sugar is present. It
is distinctly inferior to meat infusion, however, as a basis for cultural
media, especially for the more delicate pathogenic organisms.

The Reaction of Media. Bacteria are relatively sensitive to com-
paratively slight changes in the reaction of their nutritional environ-
ment, and it is essential to create a suitable initial degree of acidity
or alkalinity in media to favor their growth. A reaction neutral to
phenolphthalein slightly alkaline to litmus is suitable for most of
the bacteria pathogenic for man human tissues and blood are slightly
alkaline to litmus. A reaction of 1 per cent, acid (+1.0), using phenol-
phthalein as an indicator, has been recommended by the Laboratory
Section of the American Public Health Association for the routine
bacterial examination of water, ice, sewage, milk, cream, and ice-cream.
A reaction of 1 per cent, signifies that 1 c.c. of normal NaOH would
be required to neutralize the acid in 100 c.c. of the medium. Ten
c.c. of Y NaOH would be required to exactly neutralize one liter of
medium having an acidity of 1 per cent.

The reaction may be determined accurately in the following manner :
to 45 c.c. of distilled water, contained in a porcelain evaporating dish
of 100 c.c. capacity, are delivered exactly 5 c.c. of the medium from a
graduated pipette. The solution is brought to the boiling-point over
the free flame to expel CO 2 and 1 c.c. of a solution of phenolphthalein 1

1 Made by dissolving 0.5 gram phenolphthalein in 100 c.c. 50 per cent, alcohol. This
indicator is colorless in acid solution pink in an alkaline solution. CO2 interferes with
its accuracy as an indicator. It is especially sensitive to organic acids which occur in
ordinary media, hence its value in media titrations.

METHODS FOR THE MICROSCOPIC STUDY OF BACTERIA J91

is added as an indicator. The solution usually remains colorless,
because ordinary media are acid in reaction; Jg NaOH is added slowly
from a burette until a faint pink color appears and persists after one
minute's boiling. From the amount of f alkali required to neutralize
5 c.c. of medium, the reaction of the entire amount is readily com-
puted. Thus:

5 c.c. media are neutralized by 3 c.c. ^ NaOH.
100 c.c. media would be neutralized by 3 c.c. normal (y)NaOH.
1000 c.c. media would be neutralized by 30 c.c. normal (j)NaOH.

To reduce the reaction of a liter of medium whose initial reaction
is + 3.0 to + 1.0, 20 c.c. of normal NaOH would be required. It is
necessary to heat the medium after adding the alkali, in order to
promote the reaction between the acids of the medium and the neu-
tralizing solution and a redetermination of the reaction should be
made to make certain that the desired change in acidity has taken
place. Frequently a second addition of alkali is necessary to create
the desired final reaction.

A satisfactory reaction for cultural media designed for most patho-
genic bacteria may be created by adding ^ NaOH solution a few
drops at a time, to the entire volume, using filter paper dipped in
phenolphthalein solution, and dried, as an indicator. When the paper
shows a faint pink color the addition of alkali is discontinued. The
reaction is practically neutral under these conditions.

The Clarification of Media. It is desirable, in the preparation of
culture media, to remove all insoluble substances. This is accom-
plished by filtration methods, with or without preliminary treatment,
to flocculate the substances in suspension. The addition of non-heat-
coagulable proteins, as gelatin, frequently requires clarification with
a coagulable protein, as egg-albumen, to remove the finely divided
suspended matter.

Clarifying with Eggs. For each liter of medium to be clarified, two
eggs thoroughly whipped in a small amount of water are added. The
temperature of the medium should not exceed 50 C. The eggs are
thoroughly stirred in and the entire mixture is slowly heated to 100
C., either in a double boiler or in the Arnold sterilizer. A firm coagulum
forms during the heating process, which enmeshes the suspended par-
ticles it is desired to remove. The medium should never be disturbed
during the coagulating process. The clear underlying medium is
drawn off and filtered through cotton.

192 MICROSCOPIC AND CULTURAL STUDY OF BACTERIA

Filtration through Cotton. A large glass funnel is lined with a double
layer of absorbent cotton; the layers are placed at right angles,
thus laying the fibers of cotton at right angles. The cotton is moistened
with a small amount of hot water if agar or gelatin is to be filtered.
The medium is then carefully poured into the funnel, care being taken
that the cotton is not displaced by the force of the inflowing fluid. The
first portion of filtrate may not be clear and it is somewhat diluted

FIG. 14. Hot-air sterilizer. Lautenschlager form. (Park.)

with the water originally used to wet the cotton hence it should be
returned and refiltered. Agar and gelatin filter slowly, which may
lead to congelation, therefore the top of the funnel should be covered
to prevent undue loss of heat. Funnels surrounded by a hot water
jacket are sometimes used in the filtration of these media. Media
that are fluid when cold may be often advantageously clarified by
filtration through a good grade of heavy filter paper, with or without
a preliminary clarification with eggs, as occasion demands.

METHODS FOR THE MICROSCOPIC STUDY OF BACTERIA 193

The Distribution of Media. The clarified media are either stored in
flasks or transferred to smaller containers for immediate use. Then
they are sterilized. Most media are used in test-tubes. Test-tubes
are filled from a reservoir, usually a large funnel, the smaller end of
which is provided with a short length of rubber tubing, into which a
short glass tube, constricted somewhat at the outer end, is introduced.
The flow is controlled by a pinch cock, which constricts the rubber
tubing midway between the funnel and the delivery tube. The cotton
plug is removed from a test-tube and the delivery tube is introduced
into the open end of it to a depth of about two inches. The pinch

FIG. 15. Arnold steam sterilizer.
(Abbott.)

STERILIZING CHAMBER

U A 4 A

FIG. 16. Arnold steam sterilizer.
Ordinary type. (Park.)

cock is opened somewhat and the desired volume is allowed to flow in.
The pinch cock is then released to stop the flow, the delivery tube
removed, care being taken that no media touches that part of the
test-tube where the cotton fits, so that it will not adhere to the sides
of the tube, and the cotton plug is replaced. Usually about 8 to 10
c.c. of media are added to a tube.

Sterilization of Media. Media which do not contain coagulable
proteins, gelatin or carbohydrates are sterilized for fifteen minutes in
an autoclave at a live steam pressure of fifteen pounds (121.3 C.).
Media containing gelatin or carbohydrates are sterilized at a lower
temperature by discontinuous sterilization half an hour on three

194 MICROSCOPIC AND CULTURAL STUDY OF BACTERIA

successive days, in flowing steam in an Arnold sterilizer. After each
sterilization the medium is kept at room temperature to permit of the
germination of spores. Lower temperatures are occasionally employed,
particularly for the sterilization of blood serum or other native pro-
teins an exposure of to 70 C. for an hour on each of six successive
days usually suffices. Bacteria may be removed from fluid media
and from various sera and solutions containing thermolabile toxins
or similar products by filtration through sterile porous filters made of
unglazed porcelain or diatomaceous earth Pasteur or Berkefeld
filters. These filters are made with varying degrees of porosity,

FIG. 17. Arnold steam sterilizer. Boston Board of Health type. (Park.)

regulated largely by the thickness of their walls to accommodate vary-
ing needs. Usually the fluid is forced through the walls of the filter
into the center, which is hollow, by suction. The clear, bacteria-free
filtrate passes into a sterile container attached to the filter. The filters
and their necessary accessory parts are sterilized in the autoclave for
fifteen minutes at fifteen pounds live-steam pressure. Turbid fluids
should be passed through several layers of filter paper prior to filtra-
tion, to remove the grosser particles which otherwise would soon
clog the filter. A time limit, usually not exceeding two hours as a
maximum, should be set, beyond which filtration should be stopped

METHODS FOR THE MICROSCOPIC STUDY OF BACTERIA 195

bacteria may be forced through filters and contaminate the filtrate
if the process is carried much beyond this interval.

New, unused filters should be cleaned by running several liters of
clean water through them and they should invariably be tested before
use to guard against " pin-holes."

After filtration the filter is sterilized to kill whatever bacteria have
contaminated it. Then the surface is thoroughly scrubbed with a
brush and 1 per cent, alkaline permanganate solution (potassium per-
manganate 10 grams, water 1000 c.c.) is run through to remove organic
matter. Five per cent, oxalic acid is then passed through to remove

FIG. 91 FIG. 92 FIG. 93 FIG. 94

FIGS. 91 to 94. Types of unglazed porcelain filters. (Park.)

the permanganate solution and the acid finally removed by repeated
washings with water. If the filter becomes so clogged with organic
matter that it can no longer deliver a reasonable amount of filtrate,
the filter is placed in a muffle-furnace, gradually heated to about
250 C., and as gradually cooled. It is then cleaned as before with
permanganate solution, to remove the last traces of organic matter.

Storage of Media. If media are not to be used at once it is necessary
to protect them from evaporation and contamination. Flasks of media
are preserved best by tying paper caps over the cotton plugs if the
period of storage does not exceed a few days, or by pouring melted
paraffin over the plugs if longer periods of storage are contemplated.

196 MICROSCOPIC AND CULTURAL STUDY OF BACTERIA

It is necessary to burn the surface of the plug to destroy surface con-
tamination, then to push the plug into the neck of the flask for a dis-
tance of 1 cm. to make room for the paraffin. Flasks hermetically
sealed in this manner may remain visibly unchanged for weeks or even
months. It is good practice to place a lead foil cap over the paraffin

FIG. 22. Autoclave. (Park.)

plug and lead foil caps are better than paper caps as coverings for cot-
ton plugs. Media in storage should be maintained at a temperature
not exceeding 45 C., in a dry ice-box.

The Preparation of Nutrient Bouillon (Broth). M eat Infusion Broth
To 1000 c,c. of meat infusion (see page 189 for preparation), in a tared,

METHODS FOR THE MICROSCOPIC STUDY OF BACTERIA 197

agate-ware boiler, add 5 grams of common salt (NaCl) and heat to
boiling. Dust 10 grams of Witte peptone over the surface and stir
until it is thoroughly dissolved. Restore the loss by evaporation and
adjust the reaction to the desired degree of acidity. Boil for five
minutes, verify the reaction and filter through filter paper until the
filtrate is perfectly clear. Sterilize in the autoclave.

Meat Extract Broth. 1 To 1000 c.c. of meat extract (see page 190 for
preparation) in a tared agate-ware boiler, add 10 grams of Witte pep-
tone, dusting the peptone on the surface. Heat to boiling, restore
loss by evaporation and adjust the reaction. Continue the boiling for
five minutes, verify the reaction and cool to room temperature. 2 Filter
cold through filter paper until perfectly clear and sterilize.

Nutrient Sugar-free Broth. Meat infusion contains small amounts
of muscle-sugar dextrose usually about 0.1 per cent. This sugar
is present in nutrient meat infusion broth prepared as outlined above.
It is frequently desirable to prepare meat infusion broth free from all
sugars. The dextrose is readily removed by fermentation with Bacillus
coli, adding a broth culture of this organism to the meat infusion before
it is heated and maintaining the infusion at 37 C., for eighteen to
twenty-four hours. The sugar which is attacked by Bacillus coli in
preference to the protein constituents of the medium 3 is quantitatively
removed. The organism must be killed as soon as the sugar is
exhausted, otherwise the protein constituents will be attacked. The
end of the fermentation may be judged with a fair degree of certainty
if one removes some of the infusion seeded with Bacillus coli to a fer-
mentation tube, kept at the same temperature, 37 C.; when gas is
no longer evolved the sugar is exhausted. Sugar-free broth contains
lactic acid, one of the products of fermentation of dextrose by Bacillus
coli. After the sugar is removed the medium is sterilized in the usual
manner, or made directly into sugar-free nutrient meat infusion broth
as outlined above.

Nutrient Sugar Broth. One per cent, of dextrose, lactose, saccharose,
mannite, or other carbohydrate is added to nutrient sugar-free broth
immediately before filtering. Media containing sugars are best steri-
lized in the Arnold sterilizer on three successive days; the high tem-
perature of the autoclave tends to decompose carbohydrates.

1 It is unnecessary to add salt to meat extract.

2 A precipitate containing phosphates, soluble in the hot medium, settles out upon
cooling. It must be removed before the medium is used.

8 See chapter on Bacterial Metabolism.

198 MICROSCOPIC AND CULTURAL STUDY OF BACTERIA

Calcium Carbonate Nutrient Sugar Media. Bacteria grown in
sugar media frequently form acid products from the fermentation of
the sugars the amount of acid products may be sufficient to inhibit
the development of the organisms even after one or two days'
growth. The addition of insoluble carbonates as calcium carbonate
neutralizes the acid as it is formed and thus maintains automatically
a favorable reaction for prolonged development. Bolduan 1 has shown
that pieces of marble about 0.5 centimeters square in 100 c.c. of broth
not only restrain the development of free acid the marble appears to
create a somewhat more favorable medium, especially for the pneumo-
coccus and streptococcus as well. The bits of marble should be-
sterilized in the hot-air sterilizer before they are introduced into the
broth.

Nutrient Glycerin Broth. To 1 liter of sugar-free broth add 3 to
5 per cent, pure, redistilled glycerin immediately before filtering.
Sterilize in the autoclave fifteen minutes at fifteen pounds pressure.
Glycerin broth is extensively used for the cultivation of the tubercle
bacillus 2 and it is frequently employed in the culture of bacteria which
are susceptible to desiccation the glycerin conserves the moisture
and retards evaporation.

The various sugar-broths may be prepared with meat extract as
a basis; pathogenic bacteria develop less luxuriantly as a rule in
extract media than in meat infusion media, however.

Dunham's Solution. Five grams of common salt and 10 grams of
Witte peptone are added to one liter of water and heated to boiling
until the peptone is completely dissolved. Pass through filter paper
until perfectly clear, tube, using 10- c.c. to each test tube, and sterilize
in the autoclave. The reaction does not require adjustment.

This medium is frequently used to test the ability of bacteria to
form indol. Indol is formed in the absence of utilizable sugars by
Bacillus coli; members of the cholera group and other bacteria form
tryptophan by the splitting off of alanin :

CH 2 .CHNH 2 .COOH

\ /\/

\/ NH
Tryptophan

Alanin. The alanin is decom-
posed by the bacteria.

1 New York Medical Journal, May 13, 1905.

2 The reaction of glycerin broth designed for the cultivation of tubercle bacilli should
be +1.0 acid. The organism does not develop well in media neutral to phenolphthalein.

METHODS FOR THE MICROSCOPIC STUDY OF BACTERIA 199

Samples of Witte peptone occasionally do not contain tryptophan,
consequently each lot of peptone should be tested. When an especially
favorable sample is found it should be reserved for this purpose. Plain
neutral sugar-free broth is a better medium than Dunham's solution
for the indol test, and it should be employed for this purpose whenever
possible.

Nitrate Broth. Add 10 grams of Witte peptone to one liter of water
and dissolve by boiling. Then add 0.2 gram chemically pure potas-
sium nitrate free from nitrites and filter. Sterilize in the autoclave.
The reaction does not require adjustment.

Nutrient Gelatin Media. Ten grams Witte peptone and 5 grams
NaCl are added to 1 liter of sugar-free meat infusion 1 and dissolved by
boiling. When the ingredients are in solution, 100 2 grams of "Gold
Label" gelatin are added, a few leaves at a time, and stirred until
dissolved. The reaction is then adjusted to the desired degree and
verified after an additional five minutes' heating. The medium is
cooled to 50 C., and clarified with eggs, using two eggs for each liter.
Filter through a double layer of absorbent cotton in a large glass funnel
until clear, and sterilize. When sterilization is accomplished, cool
quickly and store in the ice-box.

Nutrient Agar. (a) Dissolve 12 grams of powdered or shredded agar
in one liter of meat infusion by the aid of heat and add 5 grams NaCl
and 10 grams Witte peptone. Maintain a boiling temperature for at
least thirty minutes, or until the ingredients are completely dissolved;
restore the loss by evaporation, adjust the reaction, and filter through
a double layer of absorbent cotton in a large glass funnel. Pass through
filter until clear. It is frequently necessary to clarify agar with eggs.
After the reaction is adjusted, cool to 50 C. add two eggs beaten up
in water and mix thoroughly. Heat slowly to the boiling-point, boil
ten minutes, and filter through absorbent cotton; sterilize in the
autoclave.

(b) Prepare " double-strength" meat infusion; 1000 grams of finely
comminuted lean meat are suspended in one liter of water; infuse in
the ice-box for twenty-four hours; heat to boiling and filter through
filter paper. Prepare nutrient meat infusion broth with this strong
infusion as a basis and adjust the reaction to twice the desired acidity
thus, if +1.0 is to be the final reaction, make the infusion broth

1 Meat extract may be used in place of meat infusion, but the medium is not as satis-
factory for pathogenic bacteria.

2 Use 120 grams gelatin during warm weather.

200 MICROSCOPIC AND CULTURAL STUDY OF BACTERIA

+2.0. Dissolve 24 grams agar in one liter of water, boiling steadily
until complete solution is attained. Add to the meat infusion broth,
boil for ten minutes and clarify with eggs in the usual manner. Filter
and sterilize.

Meat Extract Agar. Meat extract agar is made by substituting
meat extract solution for meat infusion; otherwise the process is the
same. The medium must be clarified with eggs.

Glycerin Agar. Five per cent, of glycerin is added to meat infusion
agar immediately before filtration. The reaction for cultivation of the
tubercle bacillus should be +1.0 acid to phenolphthalein. Tubercle
bacilli do not thrive in media neutral to phenolphthalein.

Loffler's Blood Serum. Add one part by volume of 1 per cent,
nutrient dextrose broth 1 to three parts of clear, hemoglobin-free beef
or sheep serum, and distribute in test-tubes. The tubes are placed
in a Koch's serum inspissator or in specially designed racks in an auto-
clave in an inclined position to produce a slanted surface, and slowly
heated to 80 C. This temperature is maintained until the medium
is firmly coagulated. The temperature is then raised to 95 or 100
C., and maintained for an hour on each of three successive days, or to
115 in the autoclave, and maintained for one hour. The medium is
opaque and white and the surface is smooth and should be free from
a metallic lustre when viewed by reflected light. The lustre indicates
an accumulation of salts, which are inimical to the growth of many
bacteria.

Coagulated Serum. 2 Clear blood serum from the dog, sheep, cow. or
other animal, preferably sterilized by filtration through Berkefeld
filters, and with or without the addition of glycerin, is placed in test
tubes and slanted and coagulated in a serum inspissator at a tempera-
ture of 75 to 80 C. An exposure of one hour to this temperature on
each of six successive days is necessary to insure sterility. The medium
should be translucent, free from bubbles, and firm.

Hiss Serum Water Media. Hiss 3 has recommended a serum water
medium for the cultivation of pneumococci and similar organisms.
It is prepared in the following manner :

Sheep or beef serum, 4 clear and free from hemoglobin, is added to
water in the proportion of one volume of serum to three of water.

1 If the liquefaction of blood serum by bacteria is to be tested, sugar-free broth must
be used in place of dextrose-broth.

2 Theobald Smith, Tr. Am. Phys., 1898, xiii, 417.

3 Jour. Exp. Med., 1905, vii, 223.

4 It is advisable to sterilize the serum by passage through an unglazed porcelain filter.

METHODS FOR THE MICROSCOPIC STUDY OF BACTERIA 201

Ten per cent, aqueous solutions of various sugars are prepared and
sterilized, and a sufficient amount of the desired sugar to make a
final concentration of 1 per cent, is added to the sterile serum solution.
Sufficient sterile 5 per cent, litmus solution is added for an indicator.
Fermentation of the carbohydrate is shown by the development of an
acid reaction, and frequently by a well-defined coagulation of the
medium as well.

Endo Medium for the Isolation of Typhoid, Paratyphoid, and Dysentery
Bacilli. I. Preparation of Agar. (a) Prepare plain, sugar-free nutri-
ent agar as described on page 197, using 15 grams of agar per liter.

(6) Adjust the reaction to a point just alkaline to litmus.

II. Preparation of Indicator. (a) Prepare a 10 per cent, solution
of basic fuchsin in 96 per cent, alcohol. This solution is fairly stabile
if kept away from the light.

(b) Prepare a 10 per cent, aqueous solution of chemically pure
anhydrous sodium sulphite (1 gram in 10 c.c. water). This solution
does not keep.

' (c) Add 1 c.c. of "II, a" to 10 c.c. of "II, b" and heat in the Arnold
sterilizer for twenty minutes. The color of the fuchsin is nearly
discharged if the solutions are of proper strength. This solution must
be prepared each day it does not keep.

III. Preparation and Use of Endomedium. (a) Add 1 gram of
C. P. lactose (free from dextrose) to 100 c.c. of agar and place in the
autoclave until melted and the lactose is thoroughly dissolved.

(b) Add a sufficient volume of "II, c" (about 1 c.c) to impart a
faint pink color to the medium.

(c) Pour into sterile Petri dishes and allow to harden in a dark
place with the covers partly removed. When cool the medium should
be colorless when viewed from above and a very faint pink when viewed
from the edge. The medium must be kept in a dark place because
the color is restored by the action of daylight.

Those bacteria which ferment lactose as Bacillus coli form lactic
acid which restores the color of the medium in the immediate neigh-
borhood of the colony; the colony therefore is colored red. Some
aldehydes also restore the color, but it is not very probable that alde-
hyde production is commonly observed among the lactose-fermenting
organisms. Non-lactose fermenting bacteria grow as colorless colonies.

If the plates are to be incubated two or three days it may be
advisable to increase the agar to 2.5 per cent, to limit the diffusion of

202 MICROSCOPIC AND CULTURAL STUDY OF BACTERIA

color from the acid colonies. For rapid isolations the medium with
the normal percentage of agar is preferable. 1

The Technique of Inoculation of Endomedia is described on page 231.

Lactose Litmus Agar. I. Prepare 1 per cent, lactose nutrient agar
by adding 10 grams of C. P. lactose (free from dextrose) to one liter
of plain nutrient agar. Adjust the reaction to a point slightly alkaline
to litmus. Tube and sterilize in the Arnold sterilizer.

II. Prepare an aqueous solution of litmus either a 5 per cent,
solution of purified litmus (Merck) or a 1 per cent, solution of azo-
litmin (Kahlbaum) and sterilize.

To Use Lactose Litmus Agar. Add about 1 c.c. of sterile litmus
solution to a sterile Petri dish and pour over it the melted lactose
agar, previously inoculated with the desired material. For water and
milk, add 1 c.c. of water or milk (diluted to the proper degree) to the
Petri dish before adding the lactose agar. Mix intimately by rotating
gently, allow to harden, and incubate.

Those bacteria which ferment lactose with the production of acid
appear as red colonies. Non-lactose-fermenting organisms appear as
blue colonies.

Blood Agar. Blood is drawn with aseptic precautions from the
carotid or femoral artery of a dog or rabbit into a sterile flask con-
taining beads. The blood is defibrinated by prolonged agitation and
added to plain (not dextrose) nutrient agar previously melted and
cooled to 45 C., in the proportion of 2 c.c. of blood to 10 c.c. of agar.
Small amounts of blood may be withdrawn directly from the heart
of an animal without difficulty, provided a small hypodermic needle
is used. The blood may be injected directly into the melted agar
without defibrination.

Occasionally human blood is added to agar; if a series of agar slants
are prepared it is possible to convert them into blood agar with a small
amount of blood, as follows :

Withdraw 10 c.c. of blood, using aseptic -precautions, from the
median basilic vein, in a large syringe. Inject the blood at once into
four times the volume of plain nutrient agar melted and cooled to 45
C. Mix at once and run 2 c.c. over the slanted surface of each agar
slant, and allow to harden in the inclined position in such a manner
that a uniform layer of the blood-agar mixture is obtained. Incubate
to prove sterility.

1 Kendall and Day, Jour. Med. Res., 1911, xxv, 95.

METHODS FOR THE MICROSCOPIC STUDY OF BACTERIA 203

Ascitic and Hydrocele Fluid Media. Ascitic and Hydrocele Agar. 1
Collect ascitic or hydrocele fluid in a sterile bottle, using aseptic
precautions. Allow to stand in the ice-box until clear, and heat to
50 C. for half an hour to destroy enzymes. Two parts of hydrocele
or ascitic fluid to eight or ten parts of plain nutrient agar previously
melted and cooled to 45 C. make a medium especially adapted to the
growth of many of the more fastidious pathogenic bacteria. 2

Ascitic broth is prepared by adding 20 to 50 per cent, by volume of
sterile ascitic fluid to plain nutrient broth. Incubate to prove sterility.

Egg Media. Eggs are a very good substitute for blood serum in
Loffler's medium. Eggs are carefully broken into a clean beaker
stirred gently with a rod (avoiding the formation of air bubbles) until
homogeneous, and mixed with dextrose broth in the proportion of
one part by volume of broth to three volumes of egg. The medium
is coagulated in a slanted position and sterilized precisely as Loffler's
blood serum is coagulated and sterilized.

Egg Medium. No. 1. Mix four to six volumes of thoroughly homo-
genized eggs with one volume of nutrient broth, and add sufficient
glycerin to make the concentration of the latter 3 per cent, by weight.
Coagulate and sterilize in the slanted position precisely as Loffler's
blood serum is coagulated and sterilized. This medium is excellent
for the cultivation of tubercle bacilli.

No. 2. Add one volume of physiological salt solution to ten volumes
of egg which have been lightly stirred with a rod until the yolks and
whites are intimately incorporated. Coagulate and sterilize in a
slanted position in test tubes.

Milk and Litmus Milk. One liter of fresh milk is thoroughly mixed
and tubed in the ordinary manner. Litmus milk is prepared by adding
sufficient litmus solution to impart a clea'r blue color. It is tubed,
using 10 c.c. to each tube, and sterilized in the autoclave.

For some purposes it is desirable to remove the cream before tubing,
but for cultural work the color of the cream ring in litmus milk is of
some diagnostic importance. Thus, members of the paratyphoid
group of bacilli almost invariably show a blue-green cream ring; the
colon bacillus colors the cream ring red brown. It should be remem-
bered that litmus milk does not coagulate as readily or as rapidly as
plain milk.

1 Ascitic and hydrocele fluids may be sterilized by passage through an unglazed porce-
lain filter.

2 It should be remembered that ascitic and hydrocele fluids usually contain about
0.08 per cent, dextrose.

204 MICROSCOPIC AND CULTURAL STUDY OF BACTERIA

Potato. New potatoes have an acid reaction, as a rule, and old
potatoes are slightly alkaline.

Large potatoes are thoroughly scrubbed, the skin removed, and
cut into cylinders with a cork-borer. The cylinders, which should be
at least 1.5 cm. in diameter, are divided into equal parts by a diagonal
cut. The pieces are placed in running water overnight so that they
will not darken, and are inserted, base downward, in large test tubes.
It is advisable to add about 1 c.c. of water to each tube to prevent
drying. Sterilize in the autoclave.

Hiss's 1 Semisolid Medium.

FORMULAE

Water . . -. 1000 c.c.

Agar 8 grams

Peptone 10 "

Meat extract 3 "

NaCl 5 "

Gelatin 2 40 "

When all the ingredients are dissolved, adjust the reaction to +0.5
(phenolphthalein), filter, and add sufficient litmus solution to impart
a clear blue. Dissolve 1 per cent, of dextrose, lactose, saccharose,
mannite, or other carbohydrate in the medium, and fill test-tubes
with it. Sterilization of lactose and saccharose semisolid media is
preferably carried out in the Arnold sterilizer. Dextrose and mannite
media may be sterilized in the autoclave.

Semisolid media are inoculated by the stab method. A change in
reaction is indicated by the litmus; gas-forming organisms form bubbles
in the depth of the medium.

Russell Double Sugar Medium. To 1 liter of nutrient agar, slightly
alkaline to litmus, add sufficient sterile 5 per cent, litmus solution to
impart a distinct clear blue color. Add 1 per cent, of C. P. lactose
and 0.1 per cent, dextrose, and distribute in test-tubes.

Sterilize in the Arnold sterilizer for three successive days, and allow
to harden in a slanted position.

Media for the Cultivation of Aciduric Bacteria. Acid Broth. Add
sufficient glacial acetic acid to a liter of 2 per cent, dextrose broth
to make the reaction equal to 50 c.c. of normal acid. A precipitate
forms, which will settle out, leaving a clear supernatant fluid that may
be removed to sterile test tubes with a sterile 10 c.c. pipette.

Oleate Agar. The addition of 0.2 per cent, sodium oleate to dextrose
agar makes a favorable medium for the cultivation of aciduric bacteria.

1 Jour. Exp. Med., 1897, ii, 677.

2 Add after the other ingredients are in solution.

METHODS FOR THE MICROSCOPIC STUDY OF BACTERIA 205

V. The Cultivation of Bacteria. Inoculation of Culture Media.
A platinum wire of 24-gauge is generally used to transfer bacteria from
medium to medium. A piece of platinum wire 1 three inches in length
is fused into the end of a glass rod 5 mm. in diameter and about 15
cm. in length. Metal handles are preferred by many; they possess
the great advantage of not breaking, but become heated during the
process of sterilization. The straight wire or " needle" is commonly
used for the inoculation of slant and stab cultures in solid media;
for the inoculation of fluid media a loop is formed on the end of the
wire. The use of the loop permits of the transfer of a greater amount

FIG. 23. Needle sterilizer. (A. de Khotinsky.)

of material. It is occasionally necessary to transfer more material
than a drop or two obtained with a loop in order to insure growth,
and for this purpose sterile capillary pipettes are very convenient.
Many anerobic bacteria and organisms which grow poorly in artificial
media must be transferred with the pipette.

The transfer of bacteria from media to media involves the following
steps :

(a) Flame cotton plugs to destroy molds and spores of bacteria;
extinguish flame.

(6) Twist cotton plugs to ^destroy adhesion to the neck of the
tube. The plugs may then be removed intact.

(c) Sterilize platinum wire in Bunsen flame. Heat wire white hot
and pass that portion of the handle adjoining the wire through the
flame, rotating it between the fingers while doing so. Allow the wire
to cool.

(d) Grasp the tubes in the left hand and remove plugs from the
tubes, holding one between the third and fourth fingers of the right

1 A cheap and efficient substitute for platinum wire is "Nichrome" wire. It is rather
less durable than platinum, and melts at a lower temperature.

206 MICROSCOPIC AND CULTURAL STUDY OF BACTERIA

hand, the other between the second and third fingers, the plugs pro-
jecting outward. Flame the mouths of the tubes and test coolness
of the platinum wire by plunging it for a distance of about a centi-
meter into the sterile medium.

(e) Remove some material from the infected tube by dipping the
tip of the wire in it, and transfer to the sterile tube.

(/) Replace the plugs in their respective
tubes and sterilize the wire before laying it
down.

II. The Isolation of Pure Cultures of Bacteria.
A. Aerobic Organisms. It is the exception
rather than the rule that bacteria exist in nature
or in many pathological processes in pure cul-
ture, that is, that a single kind of organism
alone is present. From such mixtures of bac-
teria it is frequently necessary to isolate one or
more organisms in a pure state, uncontami-
nated by other microorganisms. A common
and efficient method of separating bacteria from
mixtures is to distribute them in melted gelatin
or agar, 1 in such a manner that individual cells
are somewhat widely separated. The medium
is then allowed to harden. The organisms are
immobilized in or upon the medium and sur-
rounded by nutrients; the descendants of each
individual organism thus develop locally and

apart from the descendants of other organisms. Under favorable
conditions the descendants of individual cells become so numerous
they may be seen with the unarmed eye as spots or colonies, each
of which is made up of the progeny of a single organism. It is a
simple matter to touch such a colony with a sterile, cool platinum
needle, and infect sterile media with the adherent bacteria. In this
manner pure cultures are obtained. The technic of the isolation of
aerobic and facultatively anaerobic bacteria is technically termed
plating, or streaking, depending upon the apparatus used.

1. Plate Method. Three tubes of nutrient agar or gelatin are melted
and cooled to 42 to 45 C. A platinum wire, previously sterilized

1 Agar melts at about 95 C. and solidifies at about 40 C. It is necessary to work
rapidly with melted and cooled agar, to carry out the technic of inoculation before
solidification takes place.

FIG. 24. Platinum needle
and platinum loop.

METHODS FOR THE MICROSCOPIC STUDY OF BACTERIA 207

and cooled, is dipped to a depth of about 0.5 c.c. in a mixture of bac-
teria in fluid media, or touched to a growth in solid media, and then
rotated two or three times in a tube of the sterile melted medium.
Without sterilizing the needle, the process is repeated in the second
and third tubes. Each tube is then rotated between the palms of the
hands, to distribute the organisms thoroughly, and poured individually
into the sterile Petri dishes. The medium is flowed uniformly over
the bottom of the dish and set aside to harden.

It will be seen that the first tube inoculated contains the greatest
number of organisms, and that the third tube would theoretically
contain but few. The colonies in one of the plates will be so widely
separated that they can be "fished" with the platinum wire without
the danger of touching other colonies, and transferred to fresh, sterile
media. The success of this procedure depends largely upon a rigorous
observance of details. The mouths of the tubes and the cotton plugs
should be flamed thoroughly before inoculation is practiced, and the
transfer of the contents of the tube to the Petri dish must be done
carefully to prevent contamination. *The cover of the Petri dish should
be raised with the left hand, but directly over the bottom, to prevent
the entrance of adventitious bacteria from the air. The mouth of the
tube should not touch the bottom or edge of the Petri dish and, finally,
the cover of the latter should be replaced at once.

After the medium has hardened the plates are incubated gelatin
plates at 20 C., agar plates at 37 C. It is customary to invert agar
plates during incubation; when agar cools and becomes solid a con-
traction takes place which squeezes out some fluid. (This is well
defined in slanted agar as the water of condensation.). If the fluid
were allowed to remain on the surface of the agar plate it would con-
vert the surface potentially into a broth culture, in which the various
organisms would mix in hopeless confusion. Inversion of the plates
prevents the accumulation of moisture on the surface to a large degree;
the water of condensation collects on the cover instead. The porous
tops recommended by Hill may advantageously be used they absorb
moisture as it is formed. Gelatin plates are not inverted; fluid is
not expressed as the medium solidifies, and liquefied gelatin formed
during the growth of actively proteolytic organisms would collect on
the cover and probably contaminate the entire plate.

2. Streak Method. The isolation of pure cultures of bacteria by the
streak method differs from the plate method in that the medium
(gelatin, agar, blood serum) is not inoculated in the fluid state; the

208 MICROSCOPIC AND CULTURAL STUDY OF BACTERIA

necessary dilution to secure isolated colonies is attained by drawing
a platinum needle infected with bacteria several times across the sur-
face of sterile, slanted gelatin, agar, blood serum, blood agar or^ other
solid medium, each time covering an area not previously touched.
Eventually a degree of dilution is reached where discrete colonies are
discernible.

The plating method and streak method possess advantages and
disadvantages. A considerable proportion of the growth in plates
inoculated in the fluid state is beneath the surface, where it
is less characteristic than surface colonies. The distribution of
organisms, however, is more uniform, and small numbers of bacteria
occurring in mixture with larger numbers of undesirable organisms
are somewhat less likely to be overlooked. It is possible, moreover,
to obtain a quantitative estimation of the numbers of bacteria in
mixtures by the plate method. The streak method is advantageous
both with respect to the economy of time necessary to inoculate the
medium, and in that the colonies are wholly upon the surface of the
medium. There is less danger of contamination when u fishing" from
streak plates than from the regular method of plating, because there
is no chance for submerged colonies to underlie those upon the surface.

The use of certain kinds of media, as that of Endo, of blood agar,
and Loffler's blood serum, requires that surface inoculation shall be
made. The possibility of missing or overlooking small numbers of
the less hardy types of bacteria is greater with the streak method of
isolation.

3. The Barber Method for the Isolation of a Single Cell. It is occa-
sionally necessary, in very refined bacteriological studies, to be abso-
lutely certain that the starting point of a pure culture is a single
organism. Theoretically, single cells are the progenitors of the colonies
observed in media inoculated by the plate or the streak method, and
such is usually the case. Undoubtedly it may happen that a chain of
streptococci may remain adherent and their descendants appear as a
single colony, and it is equally certain that two alien bacteria may
occasionally become adherent by intertwining of flagella or adhesion
of viscid capsular substance and develop into a mixed colony. The
apparatus of Barber, 1 which consists essentially of a delicate capillary
pipette mounted in the substage of the microscope, and capable
of upward and downward motion in the optical axis of the instrument,
is designed to circumvent this possibility. In practice a very thin

1 Univ. Kansas Science Bull., No. 1, March, 1907.

METHODS FOR THE MICROSCOPIC STUDY OF BACTERIA 209

emulsion of bacteria in a fluid medium is placed on the surface of a
sterile thin plate of glass in such a manner that a drop of the con-
taminated fluid hangs in the opening in the stage ordinarily occupied
by the condenser. The drop is manipulated by a mechanical stage,
guided by direct observation with a one-sixth-inch lens until a single
organism appears in the field of vision. The sterile capillary pipette
is carefully brought upward until the tip engages the dependent drop;
the organism will be seen to enter the pipette, which is then lowered and
removed from its attachments to the microscope. The single cell is
transferred to a suitable medium and incubated in the usual manner.

B. Anaerobic Bacteria. 1. Plating Methods. The cultivation of
anaerobic bacteria which do not grow in the presence of atmospheric
(free) oxygen requires special apparatus and technic. The simplest
method, and one which is successful if gas-forming bacteria are absent;
is to make dilutions in dextrose agar precisely as described under
Plating in the preceding paragraphs. The tubes should be filled to a
depth of 10 cm. with the medium, and tubes of relatively large dia-
meter 2 to 3 cm. are preferable. The tubes, previously heated to
the boiling point, and rapidly cooled to 43 to 45 C. to prevent reabsorp-
tion of oxygen, are inoculated by the dilution method, rotated between
the hands to distribute the organisms uniformly, and cooled rapidly
in an upright position.

Colonies appear within the depths of the media after incubation;
in the thinly seeded tubes these colonies are discrete, and they may
be removed without contamination, either in sterile capillary pipettes
introduced through the surface of the medium, or after breaking the
tubes from the side. It is, of course, necessary to sterilize the outside
of the tube if it is to be broken. A greater degree of anaerobiosis may
be obtained within the tubes if after solidifying they are placed neck
downward with the cotton plugs removed, in a beaker containing
freshly prepared alkaline pyrogallate solution. 1 Growth of anaerobic
bacteria upon the surface of agar or blood serum may be obtained in
this manner. 2 Those bacteria which produce gas during their growth
cannot be isolated in pure culture in deep agar tubes; the liberation
of gas bubbles fragments the medium and permits the various colonies
to coalesce.

1 Five grams of dry pyrogallic acid are placed in a beaker and covered with 15-25 c.c.
of water: when dissolved a layer of kerosene or paraffin oil about 1 cm. in depth is added,
and a 10 per cent, solution of sodium hydroxide is introduced below the oil layer with
a pipette.

2 See Rickards, Cent. f. Bakt., 1904, I Abt., xxxvi, 557.

210 MICROSCOPIC AND CULTURAL STUDY OF BACTERIA

The " bottle-plate" method of Simonds and Kendall 1 overcomes this
difficulty to a considerable degree through the use of simple appli-
ances.

Sixteen-ounce French square tincture-mouth bottles are plugged
with cotton and sterilized with dry heat. With the bottles lying on
their sides, sufficient blood agar is poured in to form a layer 5 to 10

mm. deep, and allowed to harden. Dorset's
egg medium, dextrose agar or other media
may be substituted for the blood agar if
desired.

As soon as the medium has hardened the
bottles should be turned on the opposite
side, thus bringing the medium uppermost
and preventing condensation water from
adhering to it. Inoculation is made with a
bent glass rod infected with bacteria from
a thin suspension in a liquid medium, and
rubbed over the surface of the agar within
the bottle. A partial vacuum is next cre-
ated within the bottle, and residual oxygen
dissolved in alkaline pyrogallate solution
in the following manner: A closely fitting
rubber stopper with one hole carrying a
glass tube four inches in length is inserted
in the bottle. The outer end of the glass
tube projects three-quarters of an inch
beyond the stopper and is fitted with a
rubber tube three inches in length. That
portion of the glass rod within the bottle
is bent at an angle of 45 and the stopper
is turned in such a manner that the end of
the glass tube points toward the side of the
bottle opposite the layer of agar. As much

air as possible is aspirated from the bottle, and the rubber tube
closed with a pinch-cock to prevent reentrance of air.

The bottle is now placed on its side, with the medium uppermost,
and with a pipette, 10 c.c. each of a 50 per cent, solution of pyrogallic
acid and 10 per cent, sodium hydrate are run in through the rubber
tube, avoiding the entrance of air. A few cubic centimeters of clean

1 Jour. Inf. Dis., 1912, xi, 207.

FIG. 25. Wright's method of
making anaerobic cultures in
fluid media. (Mallory and
Wright.)

METHODS FOR THE MICROSCOPIC STUDY OF BACTERIA 211

water are also run in, to wash the rubber tube free from caustic alkali.
The apparatus is working properly when the rubber tube between the
pinch-cock and the bottle is collapsed, indicating a partial vacuum
within the bottle itself. Residual oxygen is rapidly absorbed within
the pyrogallate solution, leaving an inert atmosphere of nitrogen.

The bottle is incubated, medium uppermost, for the required time.
Inspection of the surface of the medium will show the colonies. After
incubation the pinch-cock is carefully opened, admitting air very gently
to avoid spattering the medium, and the stopper is removed. The
pyrogallate solution is poured out and residual traces removed with
clean water. The bottle is drained standing upon end, mouth down,
and then the colonies are ready for fishing. The colonies which develop
are all surface growths : the isolation of gas-forming anaerobic bacteria
is as readily accomplished as the isolation of non-aerogenic types.

FIG. 26. Novy jar for anaerobic cultures. (Park.)

Pure cultures of anaerobic bacteria may be obtained in an atmos-
phere of hydrogen; plates prepared in the usual manner are placed
on a rack in a Novy jar or other similar vessel provided with a tightly
fitting stop-cock, through which hydrogen can be admitted in sufficient
volume to displace the air. The stop-cock must be hydrogen-tight.
The procedure is to place inoculated plates without covers on a rack
within the jar in an inverted position, one above the other. A few
grams of pyrogallic acid are placed on the bottom of the jar with a
small piece of solid sodium hydroxide. At the last moment, when
everything is in readiness, 20 to 30 c.c. of water are gently poured
down the side of the jar to prevent spattering, and the cover quickly
clamped down. A current of hydrogen gas, either from a cylinder or
from a Kipp generator, is passed through the jar at a fairly rapid rate.

212 MICROSCOPIC AND CULTURAL STUDY OF BACTERIA

The hydrogen should enter at the top, and the outlet for the gas should
be as near the bottom of the apparatus as possible. A sample of the
escaping gas, collected in a test-tube by downward displacement,
will ignite without an explosion when all oxygen is displaced. The
inlet tubes are closed, and incubation practiced in the usual manner.
An atmosphere of nitrogen is to be preferred to an atmosphere of
hydrogen whenever it is practicable.

b c d

FIG. 27. Koch i&fety
burner. (Park.)

FIG. 28. Dunham thermo-
regulator. (Park.)

FIG. 29. Roux Bimetallic
regulator. (Park.)

2. Anaerobic Cultures in Fluid Media. A simple method of main-
taining anerobiosis in fluid media, sufficiently effective for ordinary
usage, is to overlay a flask or test tube containing dextrose broth
with a layer of albolene about 1 cm. in depth. Immediately before
inoculation all residual oxygen in the medium should be removed
by an exposure of half an hour in the Arnold sterilizer, or ten minutes
in an autoclave. The liquid is cooled rapidly to minimize reabsorp-
tion of oxygen. Wright 1 has maintained anaerobic conditions in test-
tube cultures with alkaline pyrogallate solution. Test tubes are
prepared with absorbent cotton plugs, which are made tighter than
ordinary usage demands. After the culture medium (freed from dis-
solved oxygen by heating and rapid cooling) is inoculated, the cotton

1 Mallory and Wright, Pathological Technic, 4th ed., p. 126.

METHODS FOR THE MICROSCOPIC STUDY OF BACTERIA 213

plug is pushed into the tube until the upper end is about 15 mm.
below the top. The space above the cotton plug is filled loosely with
dry pyrogallic acid and a strong solution of sodium hydroxide, 2 to
3 c.c., is added to dissolve the acid. Immediately a tightly fitting
rubber stopper is inserted into the mouth of the tube. The alkaline
pyrogallate solution absorbs the oxygen within the tube, leaving an
atmosphere of nitrogen.

FIG. 30. Incubator. (Park.)

The addition of bits of fresh, sterile tissue, 1 fresh, sterile defibrinated
blood, or of the coagulum which is formed during the coagulation of
meat infusion adds greatly to the nutritional value of cultures for the
growth of anaerobic bacteria.

1 Theobald Smith, Cent. f. Bakt., 1890, vii, 502.

214 MICROSCOPIC AND CULTURAL STUDY OF BACTERIA

Special mention of the preparation of tissue media is made in the
sections on Specific Anaerobic Organisms.

The Incubation of Bacterial Cultures. The growth of bacteria in
artificial media is markedly influenced by the temperature to which
they are exposed. A majority of those organisms parasitic upon or
pathogenic for man develop most luxuriantly at the temperature of
the human body, 37 C. Exposure to temperatures but slightly above
37 C. leads to rapid death of these organisms, consequently incuba-
tors must be available within which cultures may be safely exposed
to a uniform and constant degree of heat equal to that of the human
body. Gelatin cultures must be maintained at temperatures not
exceeding 22 C.

Incubators are single- or double-walled chambers of various sizes,
heated directly by gas or electricity, or indirectly through a water
jacket. The latter run more uniformly, because water receives and
imparts heat more slowly than air. On the other hand, large incuba-
tors cannot be surrounded with water jackets because of mechanical
difficulties. The regulation of temperature within incubators is con-
trolled by bimetallic regulators which actuate valves or electromagnets
controlling the supply of gas or electricity which heats the chamber,
or by mercurial thermoregulators working upon the principle of the
mercury thermometer. Bimetallic regulators, in which the movement
imparted to the regulator of the source of supply of heat is due to the
differential expansion or contraction of two dissimilar metals, are more
sensitive to slight variations in heat and they possess the additional
advantage of being less fragile than mercurial regulators. Various
patterns of thermoregulators of tried efficiency are on the market and
a selection between them is largely a matter of mechanical adaptability
to local needs.

VI. The Study of Bacterial Cultures. I. Growth in Solid Media.
(a) Colonies. The macroscopic appearance of bacterial colonies upon
solid media is of considerable value for the differentiation and recog-
nition of the various types; in a similar manner their microscopic
appearance, stained or unstained, permits of some differentiation.

The aspect of a colony is influenced.

1. By the kind of organism Streptococcus colonies, for example,
are habitually small and nearly transparent; anthrax colonies are
habitually larger and opaque.

2. By the consistency of the medium in firm, dense media the
growth of bacteria is limited and relatively dry; in moist, semisolid

METHODS FOR THE MICROSCOPIC STUDY OF BACTERIA 215

media the growth of the -same organism is usually luxuriant, moist,
and spreading.

3. By the composition and reaction of the medium the addition
of specifically nutritive substances, as of fresh sterile tissue to media
for the cultivation of anaerobes; of utilizable carbohydrates to media
for the cultivation of carbohydrophilic bacteria; of fresh, defibrinated
blood to media for the cultivation of hemoglobinophilic organisms;
these may improve conditions otherwise unfavorable for bacterial
development.

The reaction of the medium, furthermore, is important; many
bacteria are extremely sensitive to slightly acid media; the aciduric
bacteria thrive in media too acid for the existence of other organisms.
Even the ordinary laboratory media, made according to a definite
formula, vary sufficiently in chemical and physical properties to
influence materially the appearance of bacterial colonies. The degree
of influence is more pronounced in the feebly growing forms, but it
may affect the appearance of colonies of the more hardy types as well.

4. The rate of growth of bacteria also affects the appearance of
colonies.

It is useless, as a scientific procedure, to attempt to recognize dif-
ferences of greater refinement than the accuracy of the method permits
of, and for this reason the descriptions of bacterial colonies should
not be carried to extremes. In general, bacterial growths on solid
media are described as solids in space the average size, form, color,
lustre and texture. This applies equally well to colonies, slant and
stab cultures. The really valuable information gleaned from a study
of bacterial growths is the recognition of types of growth. For example,
spore-forming bacteria (aerobic) produce rather heavy, opaque, floc-
culent colonies; members of the Alcaligenes dysentery, typhoid,
paratyphoid group grow characteristically as rather small, round,
transparent colonies.

(b) The Enumeration of Bacteria. A very practical application of
the plating method for the isolation of bacteria is the enumeration of
bacteria in water, milk and other similar substances. The principle
involved depends upon the development of colonies of bacteria from
single cells. If a definite volume of water, 1 c.c. for example, is dis-
tributed in melted agar, thoroughly mixed in the tube by rotation
between the hands, and poured carefully into a sterile Petri dish, the
number of colonies which develop within a definite period of incuba-
tion may be regarded as a measure of the number of living bacterial

216 MICROSCOPIC AND CULTURAL STUDY OF BACTERIA

cells in a cubic centimeter of water. Experience has shown that the
accuracy of the method is influenced somewhat by the number of
organisms in the sample. A large number of bacteria, by mutual
antagonism, will fail to develop into a proportionate number of colonies.
The most accurate results are obtained when the bacterial content of
the sample as plated lies between fifty and two hundred individual
organisms. If more than two hundred bacteria are probably present,
a dilution of the sample with sterile water is made before plating, to
reduce this source of error. It is convenient in making dilutions to
use a multiple of ten, because the subsequent calculation is much
simplified. A dilution of 1 to 10 is made by adding 1 c.c. of the sample
to 9 c.c. of sterile water, shaking thoroughly and plating 1 c.c. If
the technic is all right, each colony on the plate represents one-tenth
the number of living bacteria in the original sample. The total number
of colonies multiplied by ten gives the theoretical bacterial count of
the sample. A dilution 1 to 100 is made by adding 1 c.c. of the sample
to 99 c.c. of sterile water. The plating method is inexact, partly because
an unknown proportion of organisms in the original sample will fail
to develop for various reasons in the cultural medium; furthermore,
certain types of organisms, as streptococci, may remain adherent
in chains of greater or lesser length and develop as a single colony.
Anaerobic bacteria do not develop under aerobic conditions.

A template of paper or glass ruled in square centimeters is used
to facilitate the enumeration of colonies; for densely colonized plates,
each centimeter square of the template is subdivided into smaller
units, usually one-ninth of a square centimeter. The Petri dish con-
taining colonies is placed upon the template in such a manner that
the colonies appear superimposed upon the rulings. It is a simple
matter, with the lines as a guide, to count either the entire number of
colonies in the Petri dish, or a few representative areas, which can be
multiplied by a factor. (The average Petri dish contains about 63
square centimeters.)

Example. A sample of milk diluted 1 to 100 shows a large number
of colonies after forty-eight hours' incubation. The total count of
nine squares (each a square centimeter) is 180 colonies, an average
of twenty colonies per square centimeter. The colonies upon the
entire plate (63 square centimeters) is 63 x 20, or 1260. The number
of living bacteria in 1 c.c. of the sample of milk would be 1260 x 100
or 126,000, because the number of colonies upon the plate is T^TT the
entire number in 1 cm.

METHODS FOR THE MICROSCOPIC STUDY OF BACTERIA 217

The value of the method as a convenient means of comparison of the
bacterial content of various samples of milk, water, sewage, and the
like depends largely upon the supposition that the same types of bac-
teria present in different samples will grow quantitatively under like
conditions. The comparison of bacterial counts is therefore a com-
parison of a section of the total bacterial flora, not an absolute measure
of the number of living organisms. The method of counting bacterial
colonies has been highly developed for the regulation of water and milk
supplies of cities. (See section Water and Milk.)

(c) Growth of Bacteria in Gelatin. Gelatin is added to cultural
media both to confer upon the media the property of solidifying, and
to enrich the content in nitrogenous substances.

Pure gelatin does not contain tyrosine and it is relatively rich in
diamino acids; according to Hausmann, 1 nearly 36 per cent, of the
nitrogen in gelatin is diamino nitrogen about 63 per cent, in the form
of mono-amino acids. Chemically, gelatin media are convenient for
the demonstration of soluble, proteolytic enzymes. 2 In the absence of
utilizable carbohydrate, several types of bacteria "liquefy" gelatin,
that is, through the activity of their proteolytic enzymes the gelatin
molecule is split by hydrolytic cleavage to molecules so simple in
their state of aggregation that they can no longer produce a "gel."
The presence of utilizable carbohydrate prevents the liquefaction of
gelatin by many bacteria. 3

Formerly the morphology of the liquefied zone in gelatin stab cul-
tures was regarded as distinctive for individual organisms; thus, the
napiform liquefaction produced by cholera vibrios was supposed to
be sufficiently constant to possess diagnostic value. It is now generally
conceded that this morphological characteristic is of comparatively
little importance for the identification of the organism. On the other
hand, the liquefaction of gelatin and of coagulated blood serum and
casein as well is important from a chemical viewpoint, because it
indicates the activity of a soluble proteolytic enzyme. 4

II. Growth of Bacteria in Fluid Media. (a) Plain Broth. Plain
broth, prepared from meat infusion and peptone in the usual manner,

1 Zeit. f. physiol. Chem., 1899, xxvii, 95.

2 Kendall, Boston Med. and Surg. Jour., 1913, clxviii, 825.

3 Kendall and Walker, Jour. Inf. Dis., 1915, xvii, 442.

. 4 The enzyme may be obtained sterile and in an active state in the filtrates of liquefied
gelatin, blood serum, casein, and from plain broth cultures as well, if the bacteria are
removed by filtration through unglazed porcelain: Auerbach, Arch. f. Hyg., 1897,
xxxi, 311; Berghaus, ibid., 1906, Ixiv, 1; Kendall and Walker, Jour. Inf. Dis., 1915,
xvii, 442.

218 MICROSCOPIC AND CULTURAL STUDY OF BACTERIA

and freed from sugar with Bacillus coli, 1 contains, on the average,
about 300 milligrams of nitrogen per 100 c.c. A small but variable
amount of the nitrogen exists as free ammonia. 2 Less than 10 per
cent, of the total nitrogen, as a rule, exists as aminonitrogen (deter-
mined by the method of Van^Slyke) . 3

The visible changes in the appearance of broth cultures incidental
to the development of bacteria are not of great importance; they
consist essentially of turbidity, sediment, and occasionally a ring
or pellicle. The development of a pellicle is of importance in the pro-
duction of toxin by the diphtheria bacillus, however, because it indi-
cates the maximum oxygenation of the bacteria. The character of
the turbidity and sediment the viscidity and color may afford
some information of the character of the organism. Products of
importance are frequently detected by chemical or physiological
examination in plain broth cultures of bacteria. Thus, in the absence
of utilizable carbohydrate, diphtheria and tetanus bacilli produce
their very potent toxins; 4 proteolytic bacteria elaborate soluble
enzymes; 5 Bacillus coli, Bacillus proteus and other bacteria form indol
and phenolic bodies from tryptophan and tyrosine respectively; the
cholera vibrios form nitroso indol, 6 and in sugar-free broth containing
freshly drawn, sterile, defibrinated blood, various bacteria produce
hemolysis.

The rate of decomposition of the protein constituents of the broth
may be measured by the Folin microscopic method for ammonia; the
increase in ammonia indicates the extent of deaminization. 7 The
rate of hydrolysis of protein is conveniently estimated with the Van
Slyke 8 amino-acid apparatus, after removal of ammonia from the
medium. 9 A combination of these methods affords an approximate
analysis of plain broth media before and after bacterial growth. Un-
doubtedly the application of the Emil Fischer esterification method
of aminonitrogen determination will throw much light upon the utili-
zation of various amino acids by specific bacteria during their growth
in artificial media. Amino acids containing aromatic nuclei as tryp-

1 Theobald Smith, Cent. f. Bakt., 1897, xxii, 45.

2 Determined by the Folin Method, Jour. Biol. Chem., 1912, xi, 523.

3 Jour. Biol. Chem., 1912, xii, 275; 1913, xvi, 121.

4 Theobald Smith, Tr. Assn. Am. Phys., 1896; Jour. Exp. Med., 1899, iv, 373.

5 Kendall and Walker, Loc. cit.

c Kendall, Jour. Med. Res., 1911, xxv, 117.

7 Kendall and Farmer, Jour. Biol. Chem., 1912, xii, 13, 215, 219, 465; xiii, 63; Kendall,
Day and Walker, Jour. Am. Chem. Soc., 1913, xxxv, 1201; 1914, xxxvi, 1937.

8 Van Slyke, Loc. cit.

9 The rate of hydrolysis may also be estimated by Sorenson's formol titration method,
but this is less accurate for small amounts than Van Slyke's method.

METHODS FOR THE MICROSCOPIC STUDY OF BACTERIA 219

tophan, tyrosine, histidin give colored compounds with various
reagents because they contain the chromophoric group, C = C. The
formation of indol from tryptophan (see page 74 for chemistry) has
long been used as a diagnostic test for Bacillus coli and other bacteria.
The test depends upon the removal of alanin from the tryptophan
molecule by the activity of the organism, and the addition of an auxo-
chromic group, NO 2 in the beta position of the pyrrol ring, previously
occupied by alanin. In an acid medium the compound, betanitroso-
indol, is brownish red.

Procedure, Indol Test. To a three-day plain broth culture of Bacil-
lus coli (or other organism) add 1 c.c. of concentrated hydrochloric
acid. 1 Mix thoroughly and overlay the acid broth with 1 to 2 c.c. of a
0.1 per cent, solution of sodium nitrite. 2 At the junction of the two
solutions a brownish-red ring of nitroso indol develops.

(b) Carbohydrate Broths. The addition of sugars, as dextrose, lac-
tose, saccharose, or of alcohols, as glycerol, to plain broth media,
greatly enriches the medium in non-nitrogenous substances which
may be readily utilizable sources of energy for bacteria. It is hardly
necessary to emphasize the importance of purity in all sugars and
other carbohydrates intended for bacterial purposes, nor the fallacy of
attempting to determine the action of bacteria upon specific carbohy-
drates in media not freed from muscle sugar (dextrose). The use of
serum as a basis for fermentation media frequently introduces a source
of error, because blood serum normally contains about 0.08 per cent,
of dextrose, an amount quite sufficient to give rise to considerable
amounts of acid. 3

The observations made in carbohydrate media are usually restricted to :

(a) Change in reaction.

(b) Gas formation, and in fermentation tubes, to growth in the
closed arm as well.

1 Any strong mineral acid will answer the purpose.

2 Best accomplished by running the nitrite solution carefully down the side of the
tube held in a slanting position; a stratification of the two liquids should be obtained.

3 The significance of fermentation media for the classification and identification of
bacteria depends upon their content both of protein and carbohydrate. Bacteiia derive
their energy requirements from carbohydrate, if it is utilizable, but of course they must
obtain their "Bausteine" from the nitrogenous constituents. If the carbohydrate
cannot be utilized, both structural and energy requirements are derived from the protein
constituents. Bacteria vary greatly in their ability to ferment carbohydrates; some
types, as Bacillus alcaligenes, do not appear to ferment even dextrose. Bacillus lactis
aerogenes, on the contrary, can ferment hexoses, bioses, and even starches. The fer-
mentability of a carbohydrate depends apparently upon its stereo-isomeric configuration,
and relatively slight differences in the configuration of similar carbohydrates may
determine their value for specific organisms as sources of energy. This point is discussed
somewhat later in this section.

220 MICROSCOPIC AND CULTURAL STUDY OF BACTERIA

Bacteria which can utilize carbohydrates for their energy require-
ments usually produce acid; many types produce gas as well. The
acid, which is commonly lactic, together with small amounts of acetic
and other fatty acids may be estimated by titration with standard
alkali. A more accurate estimation is based upon the determination
of the hydrogen ion concentration. 1 The gases formed are usually
carbon dioxide and hydrogen. An approximate ratio of the proportion
H/CO 2 is conveniently made in the Smith Fermentation Tube, 2 in the
following manner:

The level of the gas in the closed arm is marked with a wax pencil.
The bulb of the fermentation tube is then completely filled with
a 1 per cent, solution of sodium solution, and the gas brought
into contact with the alkaline solution by inverting the tube several
times. The gas is then entirely run back into the closed arm, and the
volume again measured. The volume is diminished proportionately
to the absorption of CO 2 by the caustic alkali.

Smith 3 has determined the "gas ratio" for the principal aerogenic
bacteria as follows :

Organism. Dextrose. Lactose. Saccharose

H C0 2 H CO 2 H C0 2

B. coli 63 37 63 37 63 37

Hog cholera 66 34

B. lactis aerogenes .... 65 35 62 38 80 20

Friediander bacillus ... 67 33 86 14 67 33

B. edematis maligni ... 67 33 ? . . ?

B. proteus 72 28 67 33

B. cloacse 70 30 37 63 58 42

Bacteria -which ferment sugars grow in the closed arm of the fer-
mentation tube; those organisms, with very few exceptions, which
cannot utilize the carbohydrate of a fermentation medium fail to
grow in the closed arm where free oxygen is not available; growth
appears only in the open arm.

Occasionally a very slight change in the stereo-isomeric formula of
a carbohydrate, or a very small change in its terminal groups will
determine its fermentability by various organisms. Thus dextrose,
mannose, and their respective alcohols, sorbite and mannite, according
to Emil Fischer, 4 have the following stereo-isomeric formulae:

1 Clark, Jour. Inf. Dis., 1915, xvii, 109.

2 Theobald Smith, The Fermentation Tube, Wilder Quarter Century Book, 1895,
187 et seq.

3 LOG. cit.

4 Untersuchungen iiber Kohlenhydrate und Fermente, 1884-1908, Berlin, 1902.

METHODS FOR THE MICROSCOPIC STUDY OF BACTERIA 221

H H

I I

H C=O H C=O H C OH H C OH

I I I I

H C OH HO C H H C OH HO C H

I I I I

HO C H HO C H HO C H HO C H

till

H C OH H C OH H C OH H C OH

I I I I

H C OH H C OH H C OH H C OH

I I I I

H C H H C H H C H H C H

I I I I

o o o o

H H H H

D. Glucose. D. Mannose. D. Sorbite. D. Mannite

The fermentation of these hexoses and their respective alcohols
by certain bacteria is shown in the accompanying table:

Organism. D. Dextrose D. Mannose. D. Sorbite. D. Mannite

B. dysenterise Shiga + +

B. dysenteiise Flexner . + + +

B. Morgan No. 1 . . . + +

B. paratyphosus Beta + + + +

An explanation for the phenomenon set forth in the table does not
readily suggest itself. Somewhat similar selective action upon specific
amino acids by other bacteria is known, qualitatively at least. Thus,
members of the Hemorrhagic Septicemia Group, particularly those
derived from animal sources, produce indol in plain broth media.
Typhoid bacilli, diphtheria bacilli and many other pathogenic organ-
isms usually fail to produce indol in ordinary media under similar con-
ditions. It is possible that this noteworthy action of members of the
Hemorrhagic Septicemia Group upon tryptophan may be related to
the fact that this amino acid is an important constituent of the hemo-
globin, the coloring matter of the blood; the Hemorrhagic Septicemia
Bacteria are particularly likely to grow in the blood stream of infected
animals.

Fermentation reactions of bacteria in varied carbohydrate media
are of importance in their cultural identification. The table on page
316 illustrates the separation of members of the Intestinal Group of
Bacteria by their fermentation reaction in various carbohydrates.

Milk. Milk is an important natural medium for bacterial growth.
It contains protein, carbohydrate and fat, together with inorganic
salts. A variety of reactions and changes in milk are produced by
bacterial development.

222 MICROSCOPIC AND CULTURAL STUDY OF BACTERIA

(a) Change in Reaction. Milk contains, in addition to protein, two
carbohydrates, which play a prominent part in determining the reac-
tion of the medium. The principal carbohydrate is lactose, but fresh
milk contains in addition, a small amount about 0.08 per cent. of a
sugar reacting like dextrose. 1 Changes in the reaction of milk caused
by bacterial activity, therefore, may be of several types. An initial
acidity followed by an alkaline reaction, as exhibited by the dysentery
bacilli and other organisms, is probably due to the initial fermentation
of the small amount of dextrose, which results in the formation of
acid and then the production of alkaline products from the decom-
position of protein when the dextrose is exhausted. These organisms
do not ferment lactose.

A permanent acid reaction is induced either by bacteria which fer-
ment lactose, or less commonly, by the decomposition of fat with the
liberation of fatty acids. Bacillus typhosus and Bacillus paratyphosus
alpha produce a permanently acid reaction in milk, but do not ferment
lactose. The exact chemistry of their activity in the medium is not
known. An alkaline reaction in milk is usually an indication of proteo-
lytic action with the formation of basic products of protein decom-
position.

The accumulation of acid incidental to the fermentation of lactose,
as, for example, by B. coli, may be sufficient in amount to cause an
acid coagulation of the casein. 2 An acid coagulation can be distin-
guished from an enzyme (lab or rennin) coagulation; the acid coagulum
will redissolve in alkali, but an enzyme coagulum fails to redissolve
by merely neutralizing the reaction.

Some types of bacteria, as Bacillus aerogenes capsulatus, ferment
lactose energetically, liberating a large amount of gas, and forming
butyric acid as well. For some unknown reason, Bacillus coli and
allied organisms, which ferment lactose in fermentation tubes with
the liberation of considerable amounts of gas, fail to produce gas from
the lactose as it exists in milk. It has been shown, however, 3 that the
colon bacillus will liberate gas from lactose if the milk is first acted upon
by a strongly proteolytic organism, as B. mesentericus.

Proteolytic bacteria, which are unable to utilize either the small
amount of dextrose, the lactose or the fats of milk, usually produce

1 Theobald Smith, Jour. Boston Soc. Med. Sci., 1897, ii, 236; Jones, Jour. Inf. Dis.,
1914, xv, 357.

2 It must be remembered that bacteria grown in litmus milk frequently fail to cause
coagulation unless the medium is heated to boiling.

3 Kendall, Boston Med. and Surg. Jour., 1910, clxiii, 322.

METHODS FOR THE MICROSCOPIC STUDY OF BACTERIA 223

an alkaline reaction which may be a simple alkalinity without obvious
change in the appearance of the medium (as, for example, B. alkali-
genes) or a deep peptonization of the casein, as illustrated by B.
pyocyaneus. B. mesentericus peptonizes casein energetically, but
the reaction of the medium is persistently acid. The initial acidity
is probably due to the formation of acid from the dextrose of the
milk; the residual acid may be associated with the activity of an
esterase which certain strains of this organism appear to elaborate.
Fatty acids are formed by hydrolysis of the glycerides of the cream
by the soluble esterase, while the metabolic activities of the organism
appear to be largely directed to the proteins of the medium. 1

It is apparent, therefore, that the chemical and physical changes
induced in milk incidental to bacterial development in the medium
are, or may be, complex in their origin. A knowledge of the proteo-
lytic and fermentative activities of bacteria in the simpler media,
however, will frequently furnish an explanation for the more involved
reactions in the highly complex medium, milk.

1 Kendall, Day and Walker, Jour. Am. Chem. Soc., 1914, xxxvi, 1937.

CHAPTER X.

BACTERIOLOGICAL EXAMINATION OF MATERIAL FROM
THE PATIENT AND THE CADAVER.

MATERIAL FROM THE LIVING SUBJECT.
Blood Culture.

Technic of Blood Cultures.
Bacteriological Examination of Cere-

brospinal Fluid.
The Examination of Peritoneal, Pleu-

ral and Pericardial Fluids.
Pus.

Examination of Urine.
Examination of Feces.
Examination of Sputum, of Buccal

and Pharyngeal Material.

Examination by Staining.
Cultural Methods.

Bacteriological Examination of the
Eye.

Bacteriological Examination of the

Ear and Nose.

THE UTILIZATION OF ANIMALS FOR BAC-
TERIAL DIAGNOSIS AND EXPERI-
MENTATION.

The Inoculation of Animals.

THE successful outcome of a bacteriological examination of material
from a patient or a cadaver depends to a large degree upon the appli-
cation of proper technic at the time of collection. Naturally this is
varied according to the nature of the case.

Postmortem cultures are taken from organs or tissues usually
indicated by the nature of the infection, and a choice of media for the
isolation of a specific bacteria, or types of bacteria, is made with this
information in view. The value of a postmortem bacteriological
examination is frequently measured by the promptness with which
it is made after death; postmortem invasion of tissues, organs, and
even the heart and larger bloodvessels by bacteria from the mouth
and gastro-intestinal tract takes place very quickly. Even if the
cadaver is placed at once in a cold room, some time must elapse before
the internal organs are cooled sufficiently to arrest bacterial growth.

The spleen, liver, kidneys, and bloodvessels are more commonly
examined for evidence of pathogenic microorganisms. The surface
of the undisturbed organ is first seared with a hot iron, then incised
through the sterile area, and some of the contents withdrawn in a
platinum loop or with a sterile capillary pipette and introduced at once
into suitable media. (The kind of media to be used is clearly set forth
for each organism, in succeeding chapters.) Blood may be obtained
from the heart, after searing the surface of the organ, or from the
larger veins of the extremities. Exudates from the pleural, peritoneal

MATERIAL FROM THE LIVING SUBJECT 225

or pericardial cavities may be removed with sterile pipettes and trans-
ferred temporarily to sterile test-tubes or flasks. Purulent discharges
are, if small in amount aspirated directly into sterile capillary pipettes ;
if in considerable quantity, removed to test tubes or flasks, and inocu-
lated as soon as practicable into suitable media.

MATERIAL FROM THE LIVING SUBJECT.

Blood Cultures. The organisms of septicemia may be numerous
or few in number in the blood stream furthermore, they may be
associated with specific lysins and agglutinins, as occasionally hap-
pens in typhoid fever. For these various reasons, experience has shown
that from 5 to 15 c.c. of blood, drawn aseptically, should be discharged
at once with aseptic precautions, into at least 100 c.c. of 0.1 per cent,
meat infusion dextrose broth, 1 and evenly distributed by careful agita-
tion. The degree of dilution attained practically renders lytic action
and agglutination ineffective; the enrichment of the medium by the
relatively large proportion of blood creates a very favorable medium
for the development of the organisms.

Technic of Blood Cultures. 1. Apparatus. An all-glass syringe with
a platinum-iridium needle of moderately large bore is sterilized in the
autoclave, preferably enclosed in a large test tube. A syringe cannot
be sterilized for bacterial purposes by boiling in water.

As an alternate apparatus, a 250 c.c. Ehrlenmeyer flask fitted with
a rubber stopper containing two glass tubes bent at right angles may
be used. The flask contains 100 c.c. of 0.1 dextrose meat infusion
broth. One tube is connected with a platinum-iridium needle by a
short length of rubber tubing, and the needle is protected during steril-
ization by a small test tube slipped over it and extending its full
length. The test tube is removed when the blood is to be taken. The
other tube is protected by a short length of rubber tubing containing
a small filter of absorbent cotton. Suction is applied through the
latter tube. It will be seen that blood may be drawn directly into
the broth in this apparatus, and in practice it has been found con-
venient to replace the rubber stopper with a sterile cotton plug after
the blood is mixed with the media.

2. Collection of Blood. The skin over the median basilic vein is
cleansed with green soap and alcohol, dried, and sterilized by the
application of tincture of iodine, which is allowed to act for two to

i See Media.
15

22(5 BACTERIOLOGICAL EXAMINATION OF MATERIAL

three minutes. Then the point of the needle is gently inserted into
the vein (which may be made prominent by gentle pressure with a
tourniquet applied to the arm above the elbow), and from 5 to 20 c.c.
of blood withdrawn. This is introduced at once into broth, as outlined
above. 1

It may be desirable to estimate the number of bacteria in the blood :
1 c.c. of blood is mixed at once with 10 c.c. of agar previously melted
and cooled to 42 C., and plated in a Petri dish. If desired, dilution
may be made 1 to 10, 1 to 100 in succeeding tubes of agar.

Typhoid and paratyphoid bacilli grow readily in the broth cultures.
They may be identified by their cultural and agglutination reactions
with highly potent specific sera. Streptococcus and pneumococcus
cultures are obtained in a similar manner from the blood stream in
blood bouillon. The organisms are isolated in pure culture by smear-
ing the broth, after incubation, upon the surface of freshly prepared
blood agar plates. The Streptococcus colonies usually exhibit a wide
clear zone of hemolysis. Pneumococcus colonies are characterized by
a narrower green zone of altered blood pigment around them. Plague
bacilli and Micrococcus melitensis are frequently detected in the
blood stream; occasionally the organisms are present in sufficient
numbers to develop in blood agar plates inoculated with 1 to 2 c.c.
of blood. The former produces characteristic lesions in guinea-pigs;
the latter develops very slowly, frequently becoming visible only after
five to seven days' incubation.

Bacteriological Examination of Cerebrospinal Fluid. Spinal fluid
for bacteriological examination is obtained by lumbar puncture with
a sterile hypodermic needle, or fine trochar about 8 cm. long and 1
mm. in bore. The needle is introduced preferably in the fourth intra-
vertebral space; the fasciculi of the cauda equinum are not tense at
this level and are readily pushed aside by the needle without injury.
An imaginary line touching the crests of the ilia intersects the spinous
process of the fourth lumbar vertebra; the sterile needle is inserted
through the previously sterilized skin at a point 1 cm. to the right
(or left) of the lower rim of the spinous process, and directed obliquely
upward and inward to such a degree that the point of the needle will
reach the median line at a depth of 5 to 6 cm. The subarachnoid space
is reached at this level and resistance to the passage of the needle

1 Occasionally circumstances arise which make it necessary to send the blood to a
distance for examination; mixing the blood with an equal volume of glycerine bile (one
part glycerin, ten parts ox bile; sterilize in autoclave) is said to be an efficient method
for preserving the bacterial content of blood practically unchanged for several hours.

MATERIAL FROM THE LIVING SUBJECT 227

ceases, and spinal fluid should flow at once. The fluid should be col-
lected in a sterile test tube. Usually from 20 to 30 c.c. of fluid flow
spontaneously; the flow may be much greater, 75 c.c. or even more.
Rarely but a few drops, or even none at all may be obtained. Normal
spinal .fluid is clear and practically colorless. Only a few cells, chiefly
lymphocytes, may be found in the sediment obtained by centrifugaliza-
tion. Pathologically the fluid may contain numerous cellular elements.
A blood-stained spinal fluid may be due to injury to bloodvessels dur-
ing the passage of the needle, or to blood from hemorrhage in the brain
or upper levels of the cord. In the former case the blood will clot if
the spinal fluid is allowed to stand; in the latter case the blood settles
to the bottom, but fails to clot. A turbid spinal fluid is indicative of
an inflammatory process in the cerebrospinal axis. If the turbidity
is uniform, pus cells are almost invariably present. Occasionally the
fluid appears clear, but upon standing, solitary, cobweb-like coagula
appear, which enmesh cellular elements and bacteria that may be
present. Sometimes an artificial stimulus to coagulation is produced
by adding a fibre or two of sterile cotton.

The spinal fluid should be centrifugalized and some of the sediment
stained with Wright's stain to determine the types of leukocytes and
organisms present. Polymorphonuclear leukocytes indicate an infec-
tion with meningococcus, parameningococcus, streptococcus, staphy-
lococcus, typhoid, colon, influenza or plague bacilli. The fluid is
usually more or less turbid. Tubercular infection, which, next to
meningococcus infection, is the most common, is usually accompanied
by a clear spinal fluid from which the cobweb coagula mentioned above
may be obtained upon standing. About 75 per cent, of cases of tuber-
cular meningitis may be diagnosed through the recognition of acid-
fast bacilli in the stained smears from these coagula. It is essential,
in doubtful cases, to inject 1 to 2 c.c. of spinal fluid subcutaneously
into guinea-pigs. If the inguinal glands are injured mechanically by
squeezing them between thumb and index finger before the injection
is made, and the material is introduced as near the glands as possible,
a definite diagnosis of tuberculosis may frequently be made within
two weeks; ordinarily four to six weeks are required for the develop-
ment of tuberculosis in the guinea-pig.

For the diagnosis of acute infections of the cerebrospinal axis, about
10 c.c. of spinal fluid should be withdrawn with aseptic precautions
into a sterile test tube. If this fluid is visibly turbid, direct smears
stained by Gram's stain and with Wright's method will furnish valuable

228 BACTERIOLOGICAL EXAMINATION OF MATERIAL

evidence of the etiological organism, and will indicate the medium to
use for its isolation and identification. Blood agar is best suited for
the meningococcus, parameningococcus, streptococcus and influenza
bacillus. The staphylococcus, typhoid, colon and plague bacilli are
less fastidious in their requirements. Less commonly, bacteria other
than those described above are found in the cerebrospinal fluid follow-
ing infection of the sinuses, otitis media, mastoid infection or septi-
cemia. The virus of anterior poliomyelitis is also found in the spinal
fluid. The most practical method of diagnosis for the latter is to filter
the clear spinal fluid through a Berkefeld filter to eliminate all bac-
teria, and to inject 5 to 10 c.c. of the filtrate intraspinously into
monkeys. The animal usually will exhibit symptoms within two weeks
if the virus is present.

The Examination of Peritoneal, Pleural and Pericardial Fluids.
Fluids or exudations from the peritoneum, pericardium or pleurae
should be stained by Gram's method to determine the type of organism,
and by Wright's method to distinguish the types of cellular elements
and their relation to the microorganisms. If the fluid is clear, or if
lymphoid cells predominate, an infection with the tubercle bacillus is
immediately suggested. Sediment from such a fluid should be injected
into a guinea-pig, using the method outlined for suspected spinal fluid.
A turbid fluid usually indicates an infection with the streptococcus,
pneumococcus, staphylococcus or pneumobacillus, if the material is
from the pleura? or pericardium; an infection with the streptococcus
or members of the intestinal group if the source is the peritoneal cavity.
Rarely the gonococcus has been found. An examination of the Gram-
stained smear will indicate the proper medium to use for the isolation
of the organisms in pure culture.

Pus. A Gram stain of pus will indicate, as a rule, the proper medium
to use for the isolation and identification of the organisms. Pus from
'* cold" abscesses frequently contains no organisms recognizable either
by Gram or acid-fast stains; experience has clearly demonstrated,
however, that a small amount of the material injected subcutaneously
into guinea-pigs will cause their death, frequently within three weeks.
At autopsy, tubercles and tubercle bacilli are found in abundance.
Much and others believe that tubercle bacilli found in the pus from
cold abscesses do not exist in their normal form, but appear as gran-
ules the so-called Much granules which are, however, viable and
virulent for guinea-pigs. In this animal the organisms regain their
normal morphology and staining reactions. The possibility of Hypho-

MATERIAL FROM THE LIVING SUBJECT 229

mycetes in the pus from old cavities in the lungs should be borne in
mind. Aetinomyces are usually visible to the naked eye as minute, yel-
lowish granules which exhibit the characteristic club when viewed under
the microscope in properly stained specimens. Pus from abscesses
in the cervical region may contain spiral organisms. The occurrence
of these organisms should suggest the possibility of a sinus connecting
the abscess with the mouth. Frequently such a sinus originates at
the base of a carious tooth.

Examination of Urine. A bacteriological examination of the
urine is of value not only in the diagnosis of infection of the genito-
urinary system; it may afford information of the causative organisms
in septicemia, and occasionally those concerned in the more chronic
heart or joint lesions as well.

The external genitalia are usually contaminated with B. smegmatis,
which resembles the tubercle bacillus, and with various adventitious
organisms as well. Prominent among the latter is Bacillus coli. A
satisfactory sample of urine for bacteriological examination may be
obtained from males if the glans and meatus are thoroughly cleansed
with soap and water. The greater amount of urine passed should be
rejected, and the last portion should be collected in a sterile, wide-
mouthed bottle. It is necessary to catheterize females after a pre-
liminary cleansing with soap and water, to obtain a satisfactory
specimen for bacteriological examination. A sterile catheter must be
used, and the first portion of the urine should be discarded. Under
ordinary conditions, except in tubercle infections the causative
organisms will be present in sufficient numbers so that a direct smear
of the sediment, stained by Gram's method, will furnish a valuable
clue to the method and media to be used for the isolation and identifi-
cation of the organism.

Blood agar is a favorable medium for the isolation of the streptococ-
cus, pneumococcus, gonococcus and staphylococcus. The gonococcus
is usually recognized by a Gram-stained smear without further attempt
at isolation. It is a Gram-negative diplococcus which, in acute infec-
tion, usually appears both intra- and extracellularly among polymor-
phonuclear leukocytes. Micrococcus catarrhalis, which might easily
. be confused with the gonococcus, occurs very rarely in genito-urinary
infections; ordinarily it may be disregarded. Micrococcus melitensis
grows very slowly upon ordinary media. Its very small size together
with the deliberateness of growth usually suffice to attract attention
to its presence. An agglutination with a specific serum completes the

230 BACTERIOLOGICAL EXAMINATION OF MATERIAL

diagnosis. Streptococci and pneumococci produce distinctive changes
in the hemoglobin of blood a'gar plates. Their final identification
is discussed in the section devoted to these organisms. Bacillus coli
and Bacillus proteus are common incitants of cystitis; they grow readily
upon ordinary media and their recognition depends upon the changes
pure cultures induce in artificial media. (See table, page 316.)

Bacillus typhosus and members of the Paratyphoid Group are occa-
sionally found in the urine of patients and convalescents. The organ-
isms are readily obtained in pure culture by plating upon nutrient
agar, or, better, upon Endo medium (see page 201). Their cultural
characteristics and agglutination with specific sera establish their
identity. Tubercle bacilli may be found in the urine; the only satis-
factory and trustworthy diagnosis is made by injecting the sediment
of a twenty-four-hour sample of urine subcutaneously into a guinea-
pig. The animal will succumb to infection if tubercle bacilli are
present, but will fail to react to smegma bacilli, which are acid-fast
and resemble tubercle bacilli morphologically.

Examination of Feces. (See also Special Section, Bacteriology of
the Feces.) The isolation and identification of pathogenic microorgan-
isms from the feces is frequently a difficult task because the normal
intestinal bacteria preponderate even in severe infections. Never-
theless, the use of special media has greatly reduced the difficulties
and a search for specific microorganisms is now possible with a very
favorable outlook for success.

For convenience, intestinal infections may be divided into those
caused by cocci, by bacilli, and spiral organisms. Of the spherical
organisms or cocci, the streptococcus is by far the most common
pathogenic organism encountered in intestinal infections, although an
overgrowth of Micrococcus ovalis may be associated with a distinct
symptomatology. The streptococcus is a common inhabitant of the
intestinal tract, and for this reason streptococcus infection of the
alimentary canal is denied by many observers. The streptococcus is
frequently an important secondary invader of the intestinal mucosa
in bacillary dysentery, and possibly in typhoid and paratyphoid infec-
tions as well. It is also frequently associated with an overgrowth of
the ''gas bacillus' 5 (Bacillus aerogenes capsulatus) in intestinal infec-
tion with the latter organism. The occasional acute enteritis observed
both sporadically and epidemically among young children is also
incited by streptococci. The distinction, if any exists, between the
intestinal streptococcus and Streptococcus pyogenes is not clearly

MATERIAL FROM THE LIVING SUBJECT 231

established. The isolation of streptococci from intestinal contents
is made either by direct plating upon dextrose agar, or by inoculation
of feces into dextrose broth. The streptococcus, as a general rule,
produces enough acid in the medium after one or two days' growth
at body temperature to seriously restrain the development of the
intestinal bacteria. A Gram stain prepared from the sediment of the
fermentation tube will frequently reveal a nearly pure culture of the
organism. A direct smear from the feces, stained by Gram's method,
also will indicate the unusual preponderance of streptococci in acute
streptococcus enteritis.

The members of the alcaligenes, dysentery, typhoid, paratyphoid
group comprise the more important bacilli ordinarily sought for in
the intestinal contents. Their isolation upon ordinary media is diffi-
cult because Bacillus coli, the most important of the intestinal organ-
isms, greatly outnumbers the more delicate pathogenic bacteria; its
colonies on ordinary media are not readily distinguished from typhoid
colonies. The Endo medium (see page 201) however, affords a ready
means of identification between the pathogenic bacteria and Bacillus
coli. The Endo medium is essentially lactose agar containing a small
amount of basic fuchsin decolorized with sodium sulphite. Organic
acids including lactic acid restore the color to fuchsin. None of the
members of the Alcaligenes-typhoid Group ferment lactose, therefore
no lactic acid is formed in and around colonies of these bacilli. Bacillus
coli, on the other hand, ferments lactose, and consequently the colonies
of this organism are colored red. The lactic acid resulting from the
fermentation of the lactose locally restores the color to the fuchsin.

Procedure. A thin suspension in plain broth, prepared from a freshly
passed specimen of feces, is incubated if possible, for an hour at 37
C., then rubbed gently over the surface of an Endo plate with a sterile
bent-glass rod or platinum needle. At the end of eighteen to twenty-
four hours, small colorless transparent colonies are removed to 0.1
per cent, dextrose meat infusion broth for further development.
Inasmuch as colonies of B. alcaligenes, dysenterise (Flexner, Shiga
and other strains) typhosus, paratyphosus alpha and beta, and the
Morgan bacillus are practically identical in appearance, a final iden-
tification must depend upon their cultural characteristics (see page
316 for table) and their agglutination with specific sera of high potency.

Members of the Mucosus Capsulatus Group are occasionally found
in acute and subacute diarrheas. They grow readily upon the surface
of Endo plates as very viscid, slimy colonies which are readily recog-

2:52 BACTERIOLOGICAL EXAMINATION OF MATERIAL

nized by their macroscopic appearance. Bacillus pyocyaneus is an
occasional incitant of intestinal disturbance. Its colonies upon
ordinary agar are surrounded by a yellowish or greenish halo. The
same general appearance characterizes its growth upon Endo medium.
Among the anaerobic bacilli, the 'gas bacillus" (Bacillus aerogenes
capsulatus) is the most important. The organism is present in variable
but small numbers in the feces of healthy adults, and occasionally in
young children as well. It may occasionally become a very prominent
organism among the fecal flora. The isolation and recognition of the
gas bacillus from the intestinal contents depends primarily upon the
energetic fermentation in milk cultures inoculated with feces and
heated to 80 C. for twenty minutes prior to incubation. (See Chapter
XXV for details.) Members of the spiral group, including the highly
pathogenic cholera vibrio, are readily isolated and identified by the
procedure described in the section on Vibrio Choleras (Chapter XXVI).

Tubercle bacilli are not infrequently found in the feces of individuals
who have advanced pulmonary tuberculosis. It is almost certain
that the organisms have been swallowed in a majority of such cases.
Occasionally a diagnosis of tuberculosis may be made thus in young
children from whom it is difficult or impossible to obtain a satisfactory
specimen of sputum. Tubercle bacilli are also found in the feces,
derived from tuberculous ulcerations. A diagnosis of tubercle bacillus
cannot safely be made from a demonstration of acid-fast organisms in
the fecal contents, because acid-fast bacteria other than tubercle
bacilli may be present. A guinea-pig furnishes the only reliable
method of distinguishing tubercle bacilli from adventitious non-patho-
genic acid-fast organisms.

Examination of Sputum, of Buccal and Pharyngeal Material. 1
A sample of sputum suitable for bacteriological examination should be
collected with care. The mouth should be clean, the receptacle should
be sterile, and the material should be raised by a deep pulmonary
cough, not by a superficial effort. Buccal and pharyngeal material
for bacteriological examination is usually obtained upon sterile cotton
swabs. Bits of membrane may be removed with sterile forceps.

Examination by Staining. A Gram-stained preparation of sputum,
buccal or pharyngeal material usually contains a variety of micro-
organisms comprising cocci, spiral forms, and even fungi and yeasts.
Many of the organisms may be normal inhabitants of the buccal

1 An excellent discussion of Infections of the Respiratory Tract and of Sputum as a
Moans of Diagnosis is that of Leutscher, Arch. Int. Med., 1915, xvi, 657.

MATERIAL FROM THE LIVING SUBJECT 233

cavity, and of the pathogenic organisms, pneumococci, streptococci,
and occasionally diphtheria bacilli are found. Usually clinical signs
or an abnormal appearance of the sputum, mouth, or throat lead
to a microscopic examination of the material from this region and, as
a rule, the nature of the symptomatology is a reliable guide to the
stain to be used. Among the organisms which stain by Gram's method,
pneumococci, streptococci, staphylococci, Micrococcus tetragenus,
and occasionally Diplococcus crassus are the more common spherical
organisms. Micrococcus catarrhalis, the meningococcus and para-
mcningococcus are the only Gram-negative cocci, so far as is known.

Of the Gram-staining bacilli, the diphtheria and pseudodiphtheria
bacilli together with Bacillus subtilis and rarely Bacillus anthracis
may be found. The bacillus of Friedlander, typhoid, influenza, pertus-
sis, plague and glanders bacilli are Gram-negative, Bacillus fusiformis
and Vincent's spirillum are Gram-negative as well. They color some-
what indistinctly with Lofflers methylene blue and very distinctly
with Wright's or Giemsa's stain. Mouth spirals and Treponema
pallidum are best stained with the latter method. Tubercle, leprosy
and nasal secretion bacilli (Karlinski) stain with the acid-fast stain.

Higher bacteria and moulds are occasionally identified in material
from the buccal cavity. Actinomyces, Oi'dium albicans, aspergillus,
mucor, streptothrix, and yeasts have been detected. The virus of
poliomyelitis has also been demonstrated in material from the naso-
pharynx which has been freed from bacteria by passage through a
Berkefeld filter and injected into a monkey.

For the routine examination of sputum, three stains are ordinarily
employed Ziehl-Neelsen for tubercle bacilli, Loffler's alkaline methyl-
ene blue for diphtheria, pseudodiphtheria, and fusiform bacilli (and
Vincent's spirillum), and the Gram stain, using dilute carbol fuchsin
as a counterstain for pneumococci, streptococci, influenza, and per-
tussis bacilli principally. Smith's stain for sputum (see page 186)
is advantageous for pneumonic sputum.

The organisms mentioned previously but not detailed in the routine
examination of sputum are of comparatively rare occurrence. They
must be studied by purely cultural methods.

Cultural Methods. Antiseptic gargles should not be used before
collecting sputum or material from the mouth or pharynx for cultural
examination. Sputum or exudate, obtained in a suitable manner, is
first washed through six or seven portions of sterile salt solution, if its
cohesiveness permits, to remove or diminish surface contamination.

234 BACTERIOLOGICAL EXAMINATION OF MATERIAL

For a majority of bacteria, freshly prepared blood agar plates are the
most satisfactory media to employ. 1 Hemolytic streptococci, pneumo-
cocci, Pneumococcus mucosus and influenza bacilli grow upon this
medium.

Diphtheria bacilli are grown upon Loffler's blood serum, as described
in the section on diphtheria.

Tubercle bacilli can be readily distinguished from lepra bacilli,
nasal secretion bacilli and adventitious acid-fast organisms by the
injection of washed, cheesy particles from sputum into guinea-pigs.

The organism commonly found in Vincent's angina (Bacillus fusi-
formis) is not readily cultivated upon ordinary media. Its recognition
usually depends upon its demonstration in smears prepared directly
from the lesions.

Bacteriological Examination of the Eye. The normal conjunctival
sac frequently contains Staphylococcus albus and Bacillus xerosis;
indeed these organisms are so commonly found in this region that they
are regarded as normal inhabitants. Abnormally a variety of bacteria
may develop on the conjunctiva, frequently causing a violent inflam-
mation. Material for bacteriological examination is best obtained
after gently flooding the conjunctival sac with a few drops of sterile
salt solution, which are removed with a sterile cotton swab. Then a
small sterile cotton swab is gently rubbed over the conjunctival sur-
face and inoculated into suitable media after a Gram-stained smear
has been examined.

The gonococcus, Koch- Weeks bacillus, and the pneumococcus are
more commonly the incitants of acute inflammation of the conjunctiva;
less frequently hemoglobinophilic bacilli (B. influenzse particularly)
or Bacillus pyocyaneus may be found. An examination of Gram-
stained smears will indicate the media to be employed if isolation of
the organisms in pure culture is desired. The meningococcus is occa-
sionally found in conjunctival inflammations in cases of cerebrospinal
meningitis; it must not be confused with the gonococcus. Micrococcus
catarrhalis, which resembles both the gonococcus and meningococcus
in its morphology and staining reactions, does not produce an acute
conjunctival inflammation with a profuse purulent discharge rather,
this organism usually gives rise to a slight reaction, even though the

1 Several drops of sterile blood, obtained from the finger or the lobe of the ear after
a preliminary sterilization, are placed in the centre of an agar plate. The material to
be studied is streaked out radially from the blood. Enough blood can be moved with
the organisms by this method to insure growth.

MATERIAL FROM THE LIVING SUBJECT 235

organisms are numerous. 1 Blood agar plates are preferable for the
cultivation of bacteria from the eye. Not only do the hemoglo-
binophilic organisms and the gonococcus grow in this medium the
less fastidious forms also develop rapidly.

Subacute Conjunctivitis. The Morax-Axenfeld bacillus is a common
excitant of subacute conjunctivitis, particularly when the internal
angle is involved. The secretior is meagre and best obtained in the
morning. The bacilli are short aud thick, Gram negative, and occur
singly and in pairs, both free and in leukocytes. They must be dis-
tinguished from members of the Mucosus Capsulatus Group, which are
comparatively common in ozena which involves the nasal ducts. The
latter are capsulated, which distinguishes them from the Morax-Axen-
feld organism.

Corneal ulcerations may be caused by pneumococci, streptococci,
leprosy bacilli, and rarely by tubercle bacilli. The latter organism
is best detected by animal inoculation.

Pseudomembranous conjunctivitis is frequently the result of a
localization and development of diphtheria bacilli, less commonly of
streptococci. The etiology of phlyctenular conjunctivitis is still
unknown.

Bacteriological Examination of the Ear and Nose. The middle
ear normally is sterile, but bacteria may reach it either by extension
of growth from the nasopharynx through the Eustachian tube, or
directly from the blood and lymph channels. By far the most com-
mon incitant of infection of the middle ear is the streptococcus alone
or less frequently in association with other organisms. This organism
is also commonly isolated from thrombosed sinuses. The pneumococcus
and Pneumococcus mucosus are also frequently isolated from otitis
media. Bacillus pyocyaneus or Bacillus proteus are not uncommonly
found in middle ear infections, particularly those containing fetid pus.
Bacillus coli has also been detected in foul-smelling pus from the
middle ear. Staphylococci, Micrococcus catarrhalis, Micrococcus
tetragenus, influenza bacilli, members of the Mucosus Capsulatus
Group of bacilli, typhoid and diphtheria bacilli have also been isolated
from otitis media.

Infection of the external auditory meatus, which contains cerumen,
is frequently the result of an overgrowth of various moulds, particularly
Aspergillus and Mucor.

1 For a discussion of Gram-negative diplococci found in the eye, see Blue, Arch.
Ophthal., 1915, xliv, No. 6.

236 BACTERIOLOGICAL EXAMINATION OF MATERIAL

The normal nasal cavity, although freely exposed to the exterior
and theoretically, at least, continually contaminated with bacteria
both from the inspired air and the microorganisms washed from the
eyes in the lachrymal secretions, is relatively free from microorganisms.
Staphylococcus albus, non-hemolytic short-chain streptococci and
pseudodiphtheria bacilli appear to be the more common organisms
isolated from the healthy nasal cavity. Material for examination is
obtained after cleaning the external nares with sterile salt solution
upon swabs of sterile cotton.

Diphtheria, leprosy, ozena, rhinoscleroma and various coryzas are
the common types of nasal infection, but a variety of organisms may
be present there either transiently, or somewhat more permanently
during bronchial infections. Thus pneumococci, influenza and per-
tussis bacilli have occasionally been isolated from the nasal secretion
during pneumonia, influenza or whooping cough respectively. Menin-
gococci and parameningococci have been demonstrated both in patients
and carriers during epidemics of cerebrospinal meningitis. It is not
unlikely that Micrococcus catarrhalis has been incorrectly diagnosed
as the meningococcus in the past, because both organisms are Gram-
negative diplococci. Microcococcus catarrhalis is occasionally found
in large numbers in the nasal secretion of acute coryza.

The bacteriology of ozena is a subject of controversy. Bacillus
ozaense and Bacillus rhinoscleromatis, both members of the Mucosus
Capsulatus Group of bacteria, have been regarded as the etiological
agents in the past.

The earliest lesion of leprosy appears to be a nasal ulcer, more
frequently located at the junction of the bony and cartilaginous septum,
hence an examination of the nasal cavity is of paramount importance
for the early diagnosis of this disease.

Tuberculous ulcerations of the nose are comparatively infrequent;
the tubercle bacillus is readily distinguished from the lepra bacillus
by injection of suspected material into a guinea-pig. The animal
is very susceptible to infection with the tubercle bacillus, but refrac-
tory to lepra bacilli. Occasionally acid-fast bacilli, which are neither
lepra nor tubercle bacilli, have been reported as occurring in the
nasal secretion. Karlinski's nasal secretion bacillus is the best known
of the Saprophytic Acid-fast Group. It grows promptly and with con-
siderable luxuriance upon glycerin agar, which at once distinguishes
it from the pathogenic acid-fast bacilli.

Nasal diphtheria is not an uncommon type of infection with the

UTILIZATION OF ANIMALS FOR BACTERIAL DIAGNOSIS 237

diphtheria bacillus. The organism is readily distinguished by its
morphology with the methylene blue stain both from the nasal secre-
tion and from cultures upon Loffler's blood serum. When the nasal
secretion is profuse, as, for example, in acute or subacute coryza,
saprophytic bacteria, as Bacillus proteus, may develop in the nasal
secretion, causing extremely offensive odors. There is little evidence
that the organism is exciting inflammation, however; it would appear
that the secretion is p favorable medium for the development of the
organism.

The virus of poliomyelitis may be found in the nasal secretion. Its
identification has been discussed above.

THE UTILIZATION OF ANIMALS FOR BACTERIAL DIAGNOSIS
AND EXPERIMENTATION.

Pasteur's brilliant animal experiments led Koch to formulate* his
Postulates for the etiological relationship of bacteria to disease. A
rigorous demonstration of the etiological relationship of bacteria to
specific disease, said Koch, must fulfill the following conditions:

1. A specific microorganism must be constantly associated with
the disease.

2. The organism must be isolated from the lesion and cultivated
outside the body of the host.

3. A pure culture of the organism must incite the disease when
introduced into a normal animal.

4. The organism must be isolated from the experimental animal
again in pure culture.

Experience has shown that many diseases of man cannot be exactly
reproduced in experimental animals and Koch's Postulates, therefore,
cannot be fulfilled with exactitude in these instances. Nevertheless,
experimental animals are indispensable both in diagnostic and
experimental bacteriological laboratories. They are used:

1. As culture media for certain types of bacteria which grow slowly
or feebly upon artificial media, particularly when the number of such
organisms is too small to permit of cultivation under artificial condi-
tions. The isolation of tubercle bacilli from urine, of glanders bacilli
from the lesions of glanders are illustrative.

2. To obtain pure cultures of bacteria from mixtures, as the inocu-
lation of white mice with pneumonic sputum for the pneumococcus,
or rubbing mixtures containing plague bacilli upon the shaved abdo-
men of a guinea-pig to obtain pure cultures of B. pestis.

238 BACTERIOLOGICAL EXAMINATION OF MATERIAL

3. To study experimentally the lesions incited by specific micro-
organisms.

4. To distinguish sharply between closely related bacteria, as for
example, between bovine and human tubercle bacilli. Thus, rabbits

FIG. 31. Guinea-pig dissection to show anatomical relations of internal organs and
important lymph glands. (From Eyre, Bacteriological Technique, Saunders & Co.)

are susceptible to infection with bovine, but not with human tubercle
bacilli. Guinea-pigs are susceptible to infection with both types.

5. To study the virulence of various microorganisms.

6. To test the toxicity of bacterial toxins and other products, and
to measure the potency of curative sera,

UTILIZATION OF ANIMALS FOR BACTERIAL DIAGNOSIS 239.^

7. For the production of various antibodies, as antitoxins, agglu-
tinins, precipitins and lysins.

The choice of animals depends chiefly upon the nature of the obser-
vation to be made. Rabbits, guinea-pigs, white rats and mice, dogs
and cats are more commonly made use of for these various examina-
tions. The method and site of inoculation, as well as the dosage, may
influence the course of the infection.

The Inoculation of Animals. Animals may be inoculated through
natural channels, as by inhalation into the respiratory tract, or inges-
tion into the alimentary tract. More frequently, however, material is
introduced parenterally into the tissues direct. The site of inoculation
is usually the skin, the body fluids or body cavities. The skin must
necessarily be entered to reach the deeper tissues. For this reason
the site of injection should be shaved and sterilized with tincture of
iodin. 1

Cutaneous Inoculation. (a) Cutaneous: Material is rubbed upon a
shaved area of skin.

(6) Intracutaneous : Injection is made directly into the skin.

(c) Subcutaneous: Material is introduced beneath the skin. A
pocket is sometimes made by separating the skin from the cellular
subcutaneous tissue, into which solid fragments of tissue are placed.
The skin over the abdomen is a common site for inoculation with fluid
cultures; the hypodermic needle is introduced at one side of the
median line and forced through the subcutaneous tissue in a trans-
verse direction, to a point well beyond the median line on the opposite
side. The abdominal wall becomes somewhat tense and does not
permit leakage to the outside if this procedure is followed.

Intravenous inoculations are made either into the blood stream
through a vein, or directly into the heart. Rabbits are readily injected
through the marginal ear veins; the vein is pinched close to the head
of the animal and gently massaged; this causes distention and makes
the vein prominent. A hypodermic needle will then readily enter
the vein; it should be gently forced along its course for a centimeter
or two before injection.

Body Cavities. The peritoneal cavity is commonly selected, but
intrapleural injections are readily made. Before introducing a hypo-
dermic needle into the peritoneal cavity, the animal, guinea-pig or

1 Tincture of iodin should be freshly prepared and painted upon the dry surface it is
desired to sterilize. Sterilization is usually accomplished after two or three minute's
exposure to the iodin solution.

240 BACTERIOLOGICAL EXAMINATION OF MATERIAL

rabbit, is held head downward to permit the intestines to pass ante-
riorly as far as possible. The needle is first introduced somewhat
obliquely through the abdominal skin posteriorly, then directly into
a fold of the abdominal wall pinched between the fingers. The needle
should be pressed in until resistance to its passage has ceased. Unless
the precaution is taken to dip the point of the needle in sterile vase-
line, some of the contents will be introduced into the cutaneous or
subcutaneous tissues as well as the peritoneal cavity. The u Plitchens"
syringe with its side-arm containing salt solution to rinse the entire
charge from the needle before withdrawal from the animal is highly
recommended for this purpose.

Intracerebral injections are made either through the optic foramen,
or through the dura after trephining the skull.

Intratracheal injections are occasionally made, but more commonly
the material is introduced deep into the bronchi through a flexible
rubber cannula. The animal should be anesthetized for this operation.

White mice and rats are usually inoculated in the loose subcutaneous
tissue at the base of the tail. The needle should pass somewhat
obliquely to avoid the spinal cord.

Care of Animals. Guinea-pigs and rabbits are very susceptible to
"snuffles" and frequently perish from contagious pneumonia and
other epizootics of the respiratory tract. 1 The first symptoms are
usually nasal discharge and a mucopurulent exudation from the eyes.
Such animals should be killed at once and their cages thoroughly
sterilized. Animals in adjacent cages should be quarantined.

Inoculated animals are best kept in separate cages apart from the
healthy stock. If they become moribund it is better to chloroform
them and perform the autopsy at once; fresh, uncontaminated cul-
tures may be obtained only at this time. If animals are permitted
to die, frequently several hours intervene before an autopsy is per-
formed, and postmortem bacterial invasion of the tissues and blood
stream is usually a disturbing factor. Infected material is obtained
from animals with the same precautions and technic as those for a
human autopsy.

1 Theobald Smith, Jour. Med. Research, xxix, 291, for discussion.

CHAPTER XL

PRACTICAL STERILIZATION, ANTISEPSIS AND
DISINFECTION.

LABORATORY STERILIZATION.
Physical Agents.

Heat.

Live Steam.

Fractional Sterilization.

Boiling Water.
Chemical Solutions.

Sajts of Heavy Metals.

Oxidizing Solutions.

Phenols, Cresols.

Tincture of lodin.

Boric Acid.

Formaldehyde.

Essential Oils.

Soaps.
Testing and Standardizing

Disinfectants.

Liquid

Gaseous Disinfectants.
Formaldehyde .
Paraform.
Sulphur.
Chlorine Gas.
Ozone.
PRACTICAL DISINFECTION.

Sputum:

Vomitus.

Feces and Urine.

Fomites.

Bath Water.

Skin and Hand.

Instruments.

Clinical Thermometers Dental Instru-
ments.

THE terms sterilization, disinfection, antisepsis and deodorization
are frequently used indiscriminately, but it is important to distinguish
between them. Sterilization and disinfection imply the destruction
of microorganisms, the latter being restricted largely to hygienic
procedure, as the disinfection of excreta, etc. A restriction of bac-
terial growth not necessarily involving the death of microorganisms
'is properly termed antisepsis. Deodorants, as the term signifies, are
those substances which destroy or mask odors; deodorants may or
may not destroy bacteria.

LABORATORY STERILIZATION.

The many kinds of apparatus and media used in the study of bac-
teria must be freed from adventitious organisms before they are
applicable to bacteriological investigation. Physical and chemical
agents are commonly made use of for this purpose.

Physical Agents. 1. Heat. (a) Incineration. Incineration is a
most efficient method of sterilizing articles of little value. The free
flame is commonly used for sterilizing platinum needles and platinum
loops. If the latter are charged with pathogenic bacteria, and par-
ticularly bacteria which contain fats, as the tubercle bacillus, it is

242 STERILIZATION, ANTISEPSIS AND DISINFECTION

necessary to dry the material by holding the loop near the flame before
incineration to prevent "spattering." The "bacteria incinerator"
made by de Khotinsky is particularly to be recommended for this
purpose. 1

(6) Dry Heat. Test tubes, flasks, Petri dishes, pipettes and other
laboratory glassware are sterilized in the hot-air sterilizer an oven
heated with a gas flame. An exposure of one and a half hours at 160
C. or one hour at 180 C. will effectually kill all spores. The heat
should be applied gradually and reduced gradually to diminish the
danger of cracking. Dry heat has but little power of penetration.
Glassware is conveniently wrapped in paper before sterilization to
protect it from dust prior to its use. The cotton plugs of flasks and
beakers are also covered with paper before sterilization, for the same
reason.

(c) Moist Heat. 1. The most satisfactory agent for the steriliza-
tion of articles uninjured by moisture is steam under pressure. Many
kinds of media and laboratory apparatus, and fomites as well are
quickly and completely sterilized by steam. The autoclave is com-
monly used for laboratory purposes. It consists essentially of a
double-walled chamber with close-fitting cover, into which steam
may be introduced. There are many patterns, but the essential
features are the steam should enter the chamber from the top, and
the bottom of the chamber should be provided with a stop-cock,
through which the residual air and condensation can escape.

Operation. A single layer of apparatus should be sterilized at one
time. If several layers are introduced, condensation water from the
upper layer may collect on the lower layers, permitting of subsequent
contamination. Steam is admitted to the chamber to displace the
air, and the air-cock should remain open until live steam flows freely
from it, because hot air is far less efficient than steam for sterilization.
Also, the condensed steam escapes through the same orifice. When
all the air is replaced by dry steam the pressure is gradually in-
creased until fifteen pounds are recorded on the pressure gauge.
This pressure is maintained from ten to twenty minutes, depending
upon the nature of the material to be sterilized. In general, media

1 It consists essentially of a tube about 12 cm. in length and 1 cm. in diameter, of
fire clay surrounded by a resistance coil of sufficiently fine wire and numerous layers
to heat the interior of the tube to a white heat. The charged platinum wire is placed
in the tube, and within a few seconds it becomes white-hot. There is absolutely no danger
from "spattering," because the extruded organisms fall upon the hot walls of the tube
(see Fig. 23, page 205).

LABORATORY STERILIZATION 243

in test-tubes is more quickly sterilized than media in flasks. At the
end of the allotted time, the pressure is gradually reduced until
equilibrium is reached with the atmospheric pressure; a sudden release
of pressure would cause violent ebullition of fluid, and a wetting or
even expulsion of cotton plugs from test-tubes or flasks.

TABLE OF PRESSURE AND TEMPERATURE.

Pressure, Temperature,

pounds. Centigrade.

100.0

5 107.7

10 115.5

15 121.5

20 126.5

2. Live Steam. Many solutions are injured by temperatures above
100 C. Media containing sugars (particularly bioses) milk and gela-
tin are partly decomposed by prolonged sterilization in the autoclave.
An exposure to live steam at lOQ^.C. for thirty minutes on each of
three successive days usually suffices to effect sterilization of these
media without injury to the constituents of the medium. This method
of fractional sterilization depends upon the destruction of all vegetative
cells during the heating process, and the germination of spores into
vegetative organisms between heatings. It is assumed that all viable
spores will have germinated before the third exposure to heat, but
Theobald Smith 1 has shown that spores of anaerobic bacteria may not
vegetate within the specified time. A fourth exposure to heat after
two or three days may be required to insure sterilization. The Arnold
sterilizer is widely used for fractional sterilization with live steam.
It consists essentially of a double-walled copper chamber surmounting
a double-bottomed water reservoir, the lower compartment of which
is shallow and contains but little water. A flame applied to this
shallow reservoir soon generates steam, which rises through a central
passage to the chamber in which the material to be sterilized is placed.
Condensed steam flows by gravity to the upper water compartment,
and from thence to the lower heated reservoir to replace the evapora-
tion. It takes but a few minutes to generate sufficient steam to fill
the sterilizing chamber. The sterilizing process begins when the
contents of the sterilizing chamber have reached 100 C.

3. Fractional Sterilization at temperatures from 60 to 80 C. is fre-
quently made use of for materials such as blood serum, which would
be injured by exposure to 100 C. The sterilizing process is repeated

1 Jour. Exp. Med., 1898, iii, 647.

244 STERILIZATION, ANTISEPSIS AND DISINFECTION

for an hour daily over a period of five to seven days. The sterilization
of Loffler's blood serum in a Koch inspissator is carried out at this
lower temperature.

4. Boiling Water. Petri dishes, culture tubes and other apparatus
containing pathogenic bacteria may be freed from bacteria by boiling
in water for five minutes. Practically no pathogenic bacteria form
spores. If tetanus, anthrax or gas bacillus cultures are to be destroyed,
the autoclave is necessary.

Chemical Solutions. Chemical disinfectants are most efficient in
aqueous solutions, and they must therefore be soluble in water.
Moisture is also essential for gaseous disinfectants.

The theory of the germicidal action of disinfectants is not well
understood; apparently the efficiency of salts of heavy metals is
associated with their noteworthy affinity for proteins, with which they
form firm combinations. It must be remembered that these salts
react more quickly with animal proteins than bacterial proteins,
therefore greater concentrations of metallic salts are required to kill
bacteria suspended in protein solutions than to destroy the same
organisms in aqueous suspension. Thus, typhoid bacilli may be
killed by 1 to 500,000 bichloride of mercury if they are suspended in
water, but a concentration of at least 1 to 1500 is required to sterilize
the same organism in blood serum. Absolute alcohol does not
appear to be a very powerful germicide; possibly its rather limited
germicidal value is associated with its dehydrating properties. Dilute
solutions of alcohol, 20 to 30 per cent., are practically as destructive
of bacteria as absolute alcohol is. Phenols are excellent germicides
in aqueous solutions, but their tendency to go into solution in oils
(which do not readily penetrate the ectoplasm of cellular structures)
makes them unreliable germicides in oily menstrua.

Salts of Heavy Metals. 1. Mercuric Chloride, HgCI 2 . Mercuric
chloride or bichloride of mercury is a powerful germicide, very soluble
in hot water, less soluble in cold water. 1 It is usually dispensed in
tablet form mixed with NaCl, which increases its solubility and also
prevents somewhat its marked tendency to unite with proteins. This
is of importance in the treatment of wounds and secretions of wounds
with this germicide. A 1 to 1000 solution of bichloride in water is
the dilution commonly used for practical purposes. This strength

1 One part of bichloride will dissolve in 3 to 4 parts of boiling, distilled water; upon
cooling, much of the bichloride becomes insoluble; one part of the salt will dissolve in
16 to 18 parts of water at room temperature.

LABORATORY STERILIZATION 245

of solution will kill all pathogenic bacteria in a very short time; a
solution of 1 to 500 strength will even kill anthrax spores within a
few hours.

The advantage of bichloride of mercury as a germicide resides in
its great bactericidal powers. Its disadvantages are: its marked
affinity for protein which, in the case of wounds, may lead to local
necrosis of tissue, or in greater concentrations, by absorption, to
toxic action on the kidneys, intestinal tract, and even the central
nervous system. It is unreliable for the disinfection of sputum, feces,
urine, purulent discharges, and other excreta, and it should never be
employed in the sterilization of instruments or eating utensils. Linen
soiled with blood or stained in any way should not be immersed in
bichloride, for it acts as a mordant and "sets" the stain.

2. Silver Salts. Silver nitrate is a much less efficient germicide
than mercuric chloride, but it is quite extensively used upon mucous
membranes. The soluble organic compounds of silver, as Protargol,
are less irritating than the inorganic salts and apparently nearly as
efficient.

Oxidizing Solutions. 1. Potassium Permanganate, KMn0 4 . Potas-
sium permanganate is a strong disinfecting agent, but it is almost
instantly reduced and rendered inert by organic substances. This
greatly impairs its practical value. Nevertheless, it is used in surgical
asepsis and also in wells and cisterns which are to be freed from
pathogenic bacteria. A strong solution is thrown into the well or
cistern, enough to impart a very pronounced pink color to the water,
and left for several hours. The water is fit for use when the last traces
of color are removed by dilution or emptying and washing out the
reservoir. This process is spoken of as "pinking" a well.

2. Hydrogen Peroxide, H 2 O 2 Hydrogen peroxide is a valuable
germicide, applicable to the cleansing of mucous surfaces and wounds.
It is readily reduced to H 2 O and nascent oxygen in contact with
organic substances, and its efficiency is attributable to the latter
element. It is essential that the peroxide actually reach the organism
to be destroyed in order to be effective. Usually hydrogen peroxide
is quite acid in reaction and irritating for this reason.

3. Chlorinated Lime or " Bleach." Chlorinated lime is an excellent
deodorant and germicide when it is fresh, but it soon loses chlorine
when exposed to the air. Nascelit chlorine is liberated from aqueous
solutions, and reacts with water to form nascent oxygen and hydro-
chloric acid, according to the equation 2C1 + H^O = 2HC1 + O.

246 STERILIZATION, ANTISEPSIS AND DISINFECTION

One part of nascent chlorine to 1,000,000 parts of water a milligram
to a litre in other words will kill moderate numbers of bacteria
within a few minutes. For this reason, chlorinated lime is extensively
used in the treatment of swimming pools to reduce the bacterial count.
It is also used for the practical sterilization of urine, bath water, feces,
and in the solid state, in privies, cellars, and stables.

Phenols, Cresols. Phenol, popularly known as carbolic acid, and
cresols, of which three are known ortho, meta, and para are powerful
germicides :

OH OH OH OH

\/CHs \/

CH 3
Phenol Ortho cresol Meta cresol Para cresol

Phenol and the cresols are somewhat sparingly soluble in water.
A 6 per cent, aqueous solution of carbolic acid, and 5 per cent, solu-
tions of the cresols are about the limits of solubility; 3 to 5 per cent,
solutions are used for most practical purposes. Phenol and cresols
are not only very toxic for bacteria, they are caustic and poisonous
for human tissue as well. Stronger solutions are anesthetic, sugges-
tive of a definite action upon nervous tissue. These substances appear
to be readily absorbed from mucous surfaces, the skin, and wounds.
They are excreted, in part at least, through the kidneys. "Smoky
urine," indicating an irritation of the kidney tissue, is a not uncommon
sequel of carbolic acid poisoning.

A 3 per cent, solution of phenol is approximately equivalent in its
disinfectant value to a 1 to 1000 solution of bichloride of mercury,
but it does not unite readily with proteins to form insoluble, inert
compounds, and it is not destructive of fabrics, metals and articles
of every-day use. 1 For sputum, urine, feces, purulent discharges, and
for stained and soiled linen, a 5 per cent, solution, equal in volume to
the bulk of the material to be disinfected, is used and allowed to remain
at least one hour before being disturbed.

Cresols form soaps with caustic solutions, which are strongly ger-
micidal. An excellent cresol soap may be made by adding one part
by volume of cresols to an equal amount of soft soap (potash soap).
This is stirred thoroughly and allowed to stand twenty-four hours.
A 5 per cent, aqueous solution of this preparation is nearly three times
as efficient in its disinfectant value as a 5 per cent, solution of carbolic
acid.

1 Hamilton, Therapeutic Gazette, 1914, xxxviii, 311.

LABORATORY STERILIZATION 247

Tincture of lodin. In vitro, tincture of iodin is of little value as a
germicide, but freshly prepared tincture of iodin applied to the skin
appears to possess very considerable germicidal value. This solution
seems to be most effective when it is freshly prepared and works most
satisfactorily when the part upon which it is to be used has been
cleaned with alcohol and allowed to dry. Nascent iodin is liberated,
and it is stated that iodin in statu nascendi is the active germicidal
factor. Tincture of iodin is rather widely used as a skin disinfectant
for minor operations, for sterilizing the epidermis prior to spinal
puncture, collecting blood for cultural purposes, and for operations
upon laboratory animals. Iodin is absorbed through the skin, and
in large amounts it is toxic.

Boric Acid. Boric acid is frequently used upon mucous surfaces
and other exposed parts when a very mild antiseptic solution is
required. Boric acid is rather an antiseptic than a germicide: its
chief advantage lies in the fact that 1 to 3 per cent, aqueous solutions
have but little action on the tissues.

All disinfectants appear to be cellular poisons to a greater or lesser
degree; in lesser concentrations they are without marked effect upon
microorganisms; in effective concentrations they appear to form
combinations with tissues if they are used in or on man.

Disinfection of the tissues has been attempted with specific bac-
tericidal sera, which are without noteworthy harmful effects upon
the patient. At the present time immune sera are not wholly satis-
factory for this purpose, but sufficiently encouraging results have
been obtained to justify their present use and to afford promise of
their improvement in the future.

A majority of chemical disinfectants are, to use Ehrlich's termin-
ology* organotrophic rather than parasitotrophic, that is, they have a
greater affinity for the tissues of the host than for the parasite. Quinine,
on the contrary, appears to be parasitotrophic it is almost a specific
for malarial parasites. Ehrlich's brilliant researches in chemotherapy
have added organic compounds containing arsenic to the list of para-
sitotrophic substances; they have a very direct and inimical action
upon trypanosomes and the Treponemata, and but minimal action
upon the tissues of the host.

Formaldehyde. A solution of formaldehyde gas in water, commer-
cially known as formalin, is a powerful disinfectant; it does not react
as strongly as mercuric chloride with protein solutions; 1 it does not

1 Formaldehyde unites with ammonia and with the amino-nitrogen of amino acids to
form stable compounds; there is relatively little action upon native proteins, however.

248 STERILIZATION, ANTISEPSIS AND DISINFECTION

injure metals or ordinary fabrics. The commercial solution con-
tains about 35 per cent, of formaldehyde, hence a 10 per cent, solu-
tion of "formalin" will contain but 3.5 per cent, of " formaldehyde,"
which is, of course, the reactive substance. Formaldehyde is an
excellent disinfectant for sputum, urine and feces, and other excre-
tions; a 5 per cent, solution of formalin (corresponding to about 2
per cent, formaldehyde) in the proportion of two volumes of the disin-
fectant to one of the excretion will effect practical sterilization of
feces within an hour. Fomites are sterilized in the same manner.
The fumes are irritating, and disinfection should not be practiced
in the sick-room.

Essential Oils. Essential oils have been used extensively in the
past, particularly in the treatment of nasal and pharyngeal infections,
and for mouth-washes. Menthol, thymol and eucalyptol, the active
principles of oil of peppermint, thyme and eucalyptus respectively,
undoubtedly possess antiseptic and feebly germicidal properties.
Cloves, cinnamon and other spices have been used for the preserva-
tion of certain types of foods; their efficiency probably depends largely
upon their content of essential oils. The expense of these substances
compared with their efficiency as antiseptics makes their use practically
prohibitive.

Soaps. Cleanliness is a very important barrier to the spread of dis-
ease. Very few pathogenic bacteria upon exposed surfaces of rooms
can survive an application of hot soap suds applied with a vigorous
arm and a scrubbing brush. A 5 per cent, solution of washing soda
(commercial sodium carbonate) is even more efficient if applied hot,
but there are limitations to its use. Fine furnishings and hangings,
wall paper and similar objects cannot ordinarily be treated with liquid
disinfectants.

Testing and Standardizing Liquid Disinfectants. The first satis-
factory method of comparing the disinfectant value of chemical disin-
fectants was that of Rideal and Walker, 1 widely known as the "Car-
bolic Coefficient" method. A modification of this method, proposed
by Anderson and McClintic, 2 is widely used in the United States.
Briefly, the method as modified by Anderson and McClintic consists
in comparing the activity of the unknown disinfectant solution in
various dilutions with a standard solution of carbolic acid; Bacillus

1 Jour. Sanitary Institute, London, xxiv.

2 Bull. Hyg. Lab., Washington, D. C., April, 1912, No. 82. Full details of method and
the disinfectant value of a large number of substances are given.

LABORATORY STERILIZATION 249

typhosus is the organism selected for the purpose, and the strength
of solution of both the unknown and known solutions are carefully
measured. The time and temperature of exposure of the organism to
the disinfectant solutions and the nature of the medium in which the
exposure is made are carefully controlled. Even with the most rigor-
ous attention to details, the carbolic coefficient of the same disinfectant
determined by this method varies nearly 50 per cent, in the hands of
different observers; 1 for the present, the standards of the Public
Health and Marine Hospital Service 2 are regarded as official for the
United States.

Gaseous Disinfectants. Pathogenic bacteria which are known or
suspected to be present upon fabrics or furnishings injured by chemical
disinfectant solutions, as well as bacteria promiscuously distributed
in rooms through droplet infection and by dust may be killed by
gaseous disinfectants, of which several are available.

1 . Formaldehyde. Formaldehyde is the most efficient of the gaseous
disinfectants for superficial disinfection, but its limited power of
penetration must be borne in mind. Formaldehyde is dispensed com-
mercially under the name "formalin," which signifies a 40 per cent,
volume solution of the gas (formaldehyde) in water. Commercial
formalin rarely contains more than 36 per cent, of formaldehyde by
volume, however, and in practice it is well to estimate 35 per cent, as
a working basis. Commercial solutions, it must be remembered, are
always acid, and the gas itself in small amounts is irritating to mucous
membranes. Prolonged exposure to concentrations of the gas suffi-
cient to kill bacteria may be fatal to animals. The gas has practically
no insecticidal value. In sufficient concentration the gas is inflam-
mable and may be ignitecrby any free flame.

In the past formaldehyde was liberated from its aqueous solution
in the gaseous state in complicated retorts, autoclaves or lamps of
special design. Much simpler methods have been evolved, which are
now used almost exclusively in practical gaseous disinfection. Of
these the permanganate method and the " sheet- volatilization method"
are the most widely used; the former possesses the dual advantage
of a quick liberation of the entire available amount of disinfectant,
and very simple apparatus ; the latter is advantageous when a gradual
evolution of gas and a prolonged exposure to its action are desired.

The Permanganate Method. When formalin is poured upon crystals'
of potassium permanganate, an energetic reaction with the evolution

1 Hamilton and Ohno, Jour. Pub. Health, 1913; ibid., 1914, iv, 163.

2 Bull. Hyg. Lab., Washington, D. C., April, 1912, No. 82.

250 STERILIZATION, ANTISEPSIS AND DISINFECTION

of sufficient heat to boil the liquid takes place. Formaldehyde gas and
heated water vapor are evolved. The entire process requires but a
few minutes, and when two parts of formalin to one part of perman-
ganate are used the residue is small in amount and practically dry
and free from reactive substances.

Ten ounces of formalin and five ounces of permanganate of potash
crystals are required for each thousand cubic feet of space to be disin-
fected. The temperature must be not less than 60 F., and the
humidity must be at least 60 per cent, for successful results. It is
convenient to place the permanganate in a three-gallon, galvanized-
iron pail with flaring sides, because the reaction between permanganate
and formalin is attended with considerable spattering. It is also
advisable to place two or three layers of heavy paper under the pail,
of sufficient size to project two feet at least in all directions, or better,
to place a galvanized-iron plate of similar dimensions under the pail
to catch all the liquid which is ejected from the pail during the process
of evolution of the gas. For successful disinfection, all closets, drawers
and alcoves should be opened as freely as possible; doors, windows
and fireplaces leading to the exterior should be tightly closed. The
room should be left closed and undisturbed for at least four hours.

The Sheet Volatilization Method. This method requires no appara-
tus except sheets, and some mechanical device for spraying formalin
upon the sheets. The conditions of moisture and humidity and the
same general preparation of the room as for the potassium perman-
ganate formalin method must prevail.

Sheets are hung upon tightly stretched cords or other similar sup-
port, in such a manner that they rest at an angle of about 45 with
the perpendicular. They are wet with warm water, are "wrung out"
to remove the excess, and sprayed with formalin in the proportion of
ten ounces to each thirty square feet of surface. One sheet (thirty
feet square) is sufficient for each thousand cubic feet of room space.

The evolution of formaldehyde is slower with the sheet method than
with the permanganate method, but equally efficient disinfection is
obtained if the room is kept closed eight hours.

2. Paraform. Paraform is a polymer of formaldehyde; it is a white
solid which is readily ignited, and burns with a bluish flame. It offers
no advantages over formaldehyde, except that it occupies much less
space. Special lamps have been devised to liberate formaldehyde
from it in the gaseous state, but the efficiency of the method is not
greater than the permanganate method, and the apparatus is some-

LABORATORY STERILIZATION 251

what more expensive, and bulky to transport. Paraform dissolved
in warm w r ater, in the proportion of two ounces of the former to half
a pint of the latter may be used in place of formalin either in the
permanganate method or the volatilization method described in the
foregoing.

Attempts have been made to combine paraform and sulphur in the
form of candles or pastilles for purposes of disinfection. Such pre-
parations are valueless so far as the generation of formaldehyde is
concerned, because the products of combustion of this substance are
carbon dioxide and water.

3. Sulphur. Sulphur was formerly highly regarded as a gaseous
disinfectant, but it is now used chiefly for insecticidal fumigation".
The products of combustion are SO 2 and SO 3 , both gases; in the
presence of moisture they have considerable germicidal activity, but
little penetrating power.

Sulphur dioxide and trioxide are vigorous bleaching agents; they
destroy fabrics, fine furnishings, and are injurious to painted or var-
nished surfaces. Consequently, the usefulness of sulphur as a germi-
cide is restricted to the holds of ships, to warehouses and similar struc-
tures, where the destruction of vermin is an important factor in the
disinfecting process.

At least 5 pounds of sulphur for each 1000 cubic feet of space to be
disinfected are placed in a broad, shallow iron pot, preferably from
one to two feet in diameter and from three to six inches high. These
are placed in pans containing about two inches of water, both to
prevent damage if the pot cracks during the burning process, and to
supply moisture essential to the success of the disinfection. The
sulphur should be not more than three inches deep in the pot and
should slope gently from the edges of the pot to the center, where
a crater is hollowed out and filled with an ounce of alcohol to start
combustion. The sulphur burns slowly, and all cracks, doors and
windows should be sealed with paper and paste to prevent escape of
the fumes. At least twelve hours should be allowed before the room
is opened.

Liquid sulphur dioxide is sometimes used in place of burning sul-
phur; the cost is several times that of burning sulphur, and for the
practical disinfection of rooms it is rarely used.

4. Chlorine Gas. Chlorine gas, particularly in humid atmospheres,
possesses considerable germicidal power, but its extremely corrosive
action upon fabrics and furnishings has materially restricted its field
of usefulness for practical disinfection.

252 STERILIZATION, ANTISEPSIS AND DISINFECTION

5. Ozone. Nascent oxygen in actual contact with bacteria is a
powerful germicide, and aside from the cost of production, it is of
value for the purification of water for domestic purposes. As an aerial
disinfectant, however, it has been disappointing.

PRACTICAL DISINFECTION.

Sputum. The bacteria and other microorganisms which incite
disease of the mouth, nose and respiratory tract leave the patient
chiefly in the nasal secretion and sputum. They are eliminated in
"droplets" of sputum during violent expulsion of the expired air, as
in coughing and sneezing. The patient, therefore, should be instructed
to cough or sneeze into paper or cloth napkins, to prevent the escape
of infected droplets of sputum, and to expectorate into a sputum box
provided with a cover. The paper napkins should be placed in a
covered receptacle and eventually burned. Cloth napkins may be
satisfactorily treated by complete immersion in boiling water for at
least fifteen minutes.

Sputum may be disinfected with 5 per cent, carbolic or cresol solu-
tion, or with a 5 per cent, solution of formalin. At least one hour's
exposure to the disinfectant is required.

Vomitus. An elimination of pathogenic bacteria from the body
in vomitus is by no means impossible, although relatively little atten-
tion has been paid to this subject in the past. The cholera vibrio s
probably the most formidable organism to be reckoned with, but the
possibility of typhoid bacilli must be borne in mind. Vomitus should
be handled with the same precautions as infected feces.

Feces and Urine. Those organisms which are the etiological agents
of infections involving the gastro-intestinal tract, as typhoid, dysen-
tery, paratyphoid bacilli and cholera vibrios, amoebae, and probably
the unknown excitants of the intestinal disorders escape from the
diseased host chiefly in the feces, and occasionally in the urine.

The feces and urine should be received in porcelain or metal con-
tainers of appropriate pattern to prevent mechanical loss of material
and immediately mixed with twice the volume of carbolic acid or
cresol solution, an equal volume of 5 per cent, formalin solution, or
with chloride of lime in the proportion of 10 per cent, of the total
volume of feces and urine. The fecal mass, unless completely fluid,
should be intimately mixed with the disinfectant solution and allowed
to remain in contact with it at least an hour. The soiled parts of the

PRACTICAL DISINFECTION 253

patient should be wiped with a cloth dipped in 2 per cent, carbolic
acid or cresol solution, then with water to remove the disinfectant.
The cloths should be either placed at once in briskly boiling water, or
in the bedpan, and treated with the feces.

Fomites. Soiled linen, clothing and bedding should be immersed
in a liberal amount of 2 or 3 per cent, carbolic acid solution and left
at least two hours. An exposure of fifteen minutes in briskly boiling
water, provided a considerable volume is used, is also sufficient to
disinfect soiled fomites.

Bath Water. The water in which patients suffering from intestinal
infections have bathed should be disinfected before it is discharged
into a drain. An ounce of chlorinated lime thoroughly mixed with
the bath water will disinfect it within an hour. The sides of the bath-
tub above the level of the water must be disinfected as well as the
water itself.

Skin and Hands. Infection of the skin and -hands, both of the
patient and attendants, is frequently unavoidable in intestinal diseases.
A vigorous application of a scrubbing brush and green soap and a
thorough cleansing of the nails frequently suffices for the hands. An
application of 2 to 3 per cent, carbolic acid, or 1 to 1000 bichloride of
mercury for several minutes will remove all danger of infection.

Sterilization of the hands for surgical operations is still a subject
of debate; there is little uniformity in the methods advocated by
leading surgeons. Wearing sterilized rubber gloves during operations
is a common practice.

Instruments. The preparation of instruments for surgical use,
often erroneously called "sterilization," must be sharply distinguished
from true sterilization in the bacteriological sense. Simple boiling of
surgical appliances in soda solution does not necessarily render them
free from bacterial spores, although the method is efficient for surgical
technic because the residual bacteria which may survive this treatment
do not germinate in the tissues. It is frequently deemed sufficient to
boil syringes and other appliances used for removing blood or other
material for bacteriological study; the only trustworthy method for
this purpose is the autoclave or the hot-air sterilizer, depending upon
the nature of the appliance.

The use of carbolic acid is not recommended for bacteriological
syringes and other apparatus used in collecting material for bacterio-
logical examination; it is difficult to remove the last traces of the disin-
fectant without contaminating the instrument itself.

254 STERILIZATION, ANTISEPSIS AND DISINFECTION

Clinical Thermometers, Dental Instruments. Clinical ther-
mometers and dental instruments are ethically on a par with the
common drinking cup and the common towel. Barbers' razors and
brushes also belong to this group. The cost of such instruments is
prohibitive for individual use, however, and their disinfection appears
to be the practical solution of the problem. In hospitals the ther-
mometers can be sterilized readily, first, by wiping them carefully
to remove adherent mucus, then immersing them in 5 per cent, for-
malin solution, 1 where they remain until wanted again. A thorough
rinsing in water will remove the formalin. The clinician who has an
extensive visiting practice cannot afford individual thermometers;
for practical purposes his thermometer can be kept free from bacteria
if it is washed each time in running water until free from mucus, and
kept in a metallic case containing 10 per cent, formalin solution pre-
pared daily. Running water will remove all traces of formalin before
use. At least two hours should be allowed before sterilization is
regarded as complete. Several thermometers may be required to
permit of this period of sterilization for each individual instrument.

Dentists' instruments almost without exception can be safely
sterilized in a boiling 5 per cent, solution of washing soda (sodium
carbonate) within five minutes' exposure. If they are then wiped
dry there is little danger of rusting. The sterilization of dental mouth
mirrors is a problem which would appear to require special investiga-
tion.

1 A covered container is required; the fumes of formaldehyde are very irritating to
the patient.

SECTION II.

PATHOGENIC BACTERIA.

CHAPTER XII.
THE PYOGENIC COCCI.

THE BACTERIA OF INFLAMMATION.
THE STAPHYLOCOCCUS GROUP.
Micrococcus Aureus.

Staphylococcus Pyogenes Albus.
Staphylococcus Epidermidis Albus.
Micrococcus Tetragenus.

Staphylococcus Pyogenes Citreus. ; Micrococcus Ovalis.

THE BACTERIA OF INFLAMMATION.

THERE is a group of bacteria which possesses in common the ability
to incite that type of infection which is commonly spoken of as inflam-
mation. A majority of these organisms are habitual parasites of man
living upon the exposed surfaces of the body, the skin and mucous
membranes chiefly: with respect to their pathogenic properties they
may be regarded as "opportunists," not as a rule requiring a well-
defined portal of entry through definite tissues to become invasive.
Any break in the continuity of the skin or a weakening or change in
the physiological state of a mucous membrane (frequently caused by
intracurrent infection) provides the necessary atrium for invasion of
the underlying tissues.

Not only are these bacteria ordinarily unable of themselves to
locate and force an entrance to the tissues of their host; after invasion
is accomplished they are unable to escape from the tissues in suffi-
cient numbers to cause progressive disease of like nature in other
hosts. They are locked up in the body, as it were, and eventually
perish. They have not perfected their pathogenic mechanism. (See
chapter on Parasitism.)

Bacteria of the "opportunist" type may be raised to very con-
siderable pathogenic powers if artificially created atria of entrance to
and escape from the tissues are provided, as for example, by passage

256 THE PYOGENIC COCCI

through suitable animals, but they soon tend to lose their artificially
acquired pathogenic properties under ordinary conditions and return
again to a parasitic existence.

Prominent among these habitually parasitic bacteria which occur
on the skin and mucous membranes of man are the various members
of the Staphylococcus and Streptococcus Groups.

THE STAPHYLOCOCCUS GROUP.

Micrococcus Aureus. Synonyms. Staphylococcus pyogenes aureus;
Staphylococcus aureus; Micrococcus pyogenes aureus; Micrococcus sali-
varius aureus.

Historical. Staphylococci probably were first seen by Klebs, some-
what later by Billroth, in unstained pus. Pasteur 1 repeatedly isolated
them from the pus of furuncles, and in one case of osteomyelitis, and
suggested their etiological relationship to these lesions, but to Rosen-
bach 2 belongs the priority of growing them in cultures of undoubted
purity.

Morphology. The organisms in the free state are spherical, measur-
ing from 0.7 to 0.9 micron in diameter. Those just about to divide
are frequently oval. They occur singly, in pairs, or in irregular masses,
both in culture and in pus; rarely chains of four to six cocci are found.

Staphylococci are non-motile and possess no flagella; they do not
form capsules, and spores have not been observed. They siain readily
with ordinary anilin dyes, some individuals more intensely than their
fellows. They are Gram-positive.

Isolation and Culture. Staphylococci are readily obtained in pure
culture by plating or streaking the suspected material directly upon
agar or gelatin. The colonies on gelatin after thirty to forty-eight
hours' incubation at room temperature become visible as gray, glis-
tening growths 0.5 to 1 mm. in diameter; somewhat later the colonies
sink into saucer-shaped depressions of liquefied gelatin, the bacteria
collect at the bottom of the depression and soon become golden-yellow
in color. The growth upon agar plates at 37 C. is more rapid: at
the end of forty-eight hours' incubation the colonies are golden-yellow
and have attained a diameter of 1 to 3 mm.

Staphylococci grow readily in the ordinary cultural media. Gela-
tin, coagulated blood serum (sugar-free) and casein are liquefied.

1 Compt. rend. Acad. Sci., 1880, xc, 1035.

2 Mikroorganismen bei den Wundinfektionskrankheiten des Menschen, Wiesbaden,
1884, 19.

THE STAPHYLOCOCCUS GROUP 257

Acid is produced in dextrose, lactose, saccharose and mannite broths.
Milk is coagulated, usually within three days at 37 C.; many strains
subsequently partially digest the coagulum. In plain and dextrose
broths a turbidity is produced after twelve to fourteen hours' incuba-
tion at 37 C.; after forty-eight hours' growth a golden-yellow sedi-
ment collects in the bottom of the tubes. t Growth on slanted agar is
golden-yellow in color, moist and spreading. Pigment production is
especially luxuriant on slanted potato.

The organisms are aerobic, facultatively anaerobic. The optimum
temperature of growth lies between 28 and 38 C.; growth ceases
below 8 C. and above 43 C.

FIG. 32. Staphylococcus. X 1000.

Staphylococci are among the most resistant of the non-spore-form-
ing bacteria to physical agents. An exposure of one hour at 80 C
or two hours at 70 C. moist heat is usually fatal. Several minutes'
exposure at 100 C. (flowing steam) or twelve hours' exposure to direct
sunlight may fail to kill them. Indirect daylight may fail to destroy
their vitality even after two weeks; three months' continuous drying
(on cloth or paper) is equally ineffective; 0.001 per cent, mercuric
chloride and 5 per cent, carbolic acid usually kill the naked germs in
about ten minutes.

Products of Growth. Acids, chiefly lactic, but with demonstrable
amounts of propionic, butyric, and valerianic, are formed during the
fermentation of ordinary sugars. No gas is produced. The pus of
Staphylococcus abscesses is usually acid in reaction; the organisms
appear to form limited amounts of acid from protein. 1 Emmering 2

1 Kendall, Day and Walker, Jour. Am. Chem. Assn., 1913, xxxv, 1246.

2 Berlin, deut. chem. Gesellsch., 1896, 2721.
17

258 THE PYOGENIC COCCI

has identified indol, phenol, skatol, and trimethylamine among the
decomposition products of staphylococci grown anaerobically in
protein media. Cacace 1 has shown that the earlier decomposition
products produced from gelatin and coagulated blood serum are
chiefly proteoses and peptones; as proteolysis proceeds, these products
are degraded to simpler amino acid compounds.

Pigment. Staphylococci isolated directly from severe inflamma-
tions usually produce a golden-yellow pigment, but prolonged cul-
tivation upon artificial media may result in a partial or complete loss
of chromogenesis. Armand 2 has isolated non-chromogenic strains
of staphylococci directly from typically chromogenic cultures by the
plate method. The yellow pigment, which is produced most abun-
dantly in media containing carbohydrates (particularly on potato)
in the presence of free oxygen, appears to lie between the individual
organisms, not within their substance. It is insoluble, or nearly so,
in water, readily soluble in alcohol. It is related to the lipochromes.
The pigment can be saponified readily, and it evolves an odor of acro-
lein when it is dry-heated. Strong acids, notably sulphuric, change
the yellow color to a green-blue (lipocyanin) . Lugol's solution
(iodin-potassium iodide) turns it green.

Enzymes. 1. Proteolytic. Old sugar-free broth and gelatin cul-
tures of staphylococcus contain a proteolytic enzyme which will liquefy
gelatin a gelatinase. This enzyme may be obtained in an active
state free from bacteria by filtering either broth or liquefied gelatin
cultures of the organism through unglazed porcelain. 3 An enzyme
which liquefies casein is demonstrable in milk cultures; whether the
latter enzyme is identical with the gelatinase has not been determined.

2. Amylolytic. According to Buxton, 4 staphylococci produce a
maltase which hydrolyzes maltose; no other inverting enzymes have
been observed.

3. Lipolytic. Wells and Corper 5 have demonstrated a lipase of
moderate activity in autolyzed agar slant cultures of staphylococci.

4. Hemolytic. Neisser 6 and Wechsberg 7 have shown that old
(7- to 14-day) broth cultures of staphylococci, particularly the more
virulent strains, contain a soluble enzyme which hemolyzes blood

1 Cent. f. Bakt., 1901, xxx, 244.

2 Quoted by Lehmann and Neumann, Bacteriology, 1904, 3d ed., 193.

3 Loeb, Cent. f. Bakt., 1902, xxxii, 471.

4 Am. Med., 1903, vi, 137.

s Jour. Inf. Dis., 1912, xi, 388.

6 Zeit, f. Hyg., 1901, xxxvi, 299.

7 Cent. f. Bakt., Orig., 1903, xxxiv, 857.

THE STAPHYLOCOCCUS GROUP 259

both in vivo and in vitro. In vitro this enzyme, staphylolysin, appears
to digest the stroma of red blood cells, liberating hemoglobin from
them. A quantitative measure of the activity of this hemolysin can
be made by adding gradually decreasing amounts of broth culture
(filtered through unglazed porcelain) to well-washed red blood cells
suspended in salt solution; the mixtures are incubated at 37 C.
for one hour, then kept in the ice box twenty-four hours. The greatest
dilution of broth showing hemolysis is considered the unit. 1 This
enzyme is destroyed or inactivated at a temperature of 60 C. in
twenty minutes. Whether this hemolysin is identical with or produced
parallel to the proteolytic enzyme of the staphylococcus has not been
determined. Burckhardt 2 believes the staphylolysin is a true hemolytic
bacterial toxin; from his observations it appears to be non-protein
in nature, not giving the biuret reaction. It is soluble in ether.

Leucocidin. Van de Velde 3 has obtained an enzyme which destroys
leukocytes by injecting virulent staphylococci into the pleural cavities
of rabbits; the exudate, freed from cellular detritus by filtration
through unglazed porcelain, rapidly kills and even dissolves fresh leuko-
cytes. Neisser has shown that fresh leukocytes will reduce the color
of dilute methylene blue solutions to the point of extinction; if dilute
methylene blue is added to tubes containing leukocytes and leuko-
cidin, no reduction occurs, thus indicating that the leukocytes are in-
jured or destroyed. Leukocidin solutions alone fail to remove the color.

Thrombokinase. Loeb's observation 4 that the products of growth
of staphylococci cause blood to coagulate more rapidly than normal
.has been interpreted by Much 5 to be due to a substance reacting like
a thrombokinase.

Distribution in Nature. Staphylococci are found widely distributed
in nature, but associated rather closely with man and the higher
domestic animals. These organisms do not appear to be adapted to
a purely saprophytic existence. They are found in dust, particularly
that of stables, houses, and hospitals; they are common on the skin,
the mucous membranes of the nose, mouth, and to a lesser extent in
the gastro-intestinal tract, 6 the eye, the external ear, and nearly always

1 It must be remembered that the sera of normal men and of animals frequently
exhibit antibemolytic powers, hence the necessity of washing red blood cells thor-
oughly before testing the activity of staphylolysin upon them.

2 Arch, exp: Path. u. Pharm., 1910, Ixiii, 107.

3 Ann. Inst. Past., 1896.

4 Jour. Med. Res., 1903, x, 407.
6 Biochem. Zeit., 1908, xiv, 143.

6 Moro, Jahrb. f. Kinderheilk., lii, 530; Streit, Inaug. Diss., Bonn, 1897.

260 THE PYOGENIC COCCI

under the finger nails and in the hair follicles in man, which makes
sterilization of the skin and hands difficult.

Chemotaxis. The bodies of staphylococci appear to contain sub-
stances of unknown composition which attract leukocytes; the cell
substance of killed cocci injected in the cornea frequently causes an
accumulation of leukocytes in the anterior chamber of the eye
hypopyon.

Pathogenesis. Man. Ordinarily the organisms exist on the intact
surfaces of man as "opportunists," occasionally gaining entrance to
the underlying tissues through abrasions, chiefly in the skin, causing
localized abscesses, furuncles, or metastatic inflammations. Of the
metastatic inflammations, acute osteomyelitis and endocarditis are
the more common; less commonly generalized purulent pyemias
develop. It is assumed that metastatic pyemias are caused either by
direct invasion of the blood stream or less commonly by transmission
of staphylococci in leukocytes to remote parts of the body; there
they escape from the leukocytes and set up new foci of infection.
Suppurative pleurisy and pericarditis are not uncommon. The occur-
rence of furunculosis in diabetics is so frequent as to lead to the sup-
position that not only is the general average resistance to invasion by
staphylococci reduced in this disease, there may be a peculiar local
lack of resistance in the skin itself. Occasional individuals exhibit
a certain vulnerability to infection in particular regions; the neck
and buttocks are more frequently affected. One invasion appears to
predispose to subsequent infection. Staphylococci frequently are
secondary invaders in pulmonary tuberculosis, diphtheria and other
severe infections. Generally speaking, staphylococci cause acute
focal inflammations. Generalized infections of staphylococcus causa-
tion are relatively uncommon. Prolonged infections frequently result
in profound generalized symptoms; chills with intermittent fever are
the more common clinical signs. Parenchymatous or even amyloid
degeneration of certain glandular organs, notably the kidneys, is the
more common pathological lesion in such cases.

Experimental Reproduction of Lesions. A satisfactory explanation
of the pathogenesis of staphylococci for man is not available. Neither
the staphylolysin nor the leukocidin appears to play a prominent part
in the morbid process. There is little definite evidence that the cell
substance of the organisms themselves is the important factor. Never-
theless, the etiological relationship of staphylococci to furunculosis

THE STAPHYLOCOCCUS GROUP 261

has been definitely established by the experiments of Carre 1 and Engels, 2
both of whom rubbed virulent cultures of these organisms upon their
skin, producing there typical furuncles.

Animals. Rabbits are the best of the laboratory animals for
experimental inoculation. Subcutaneous inoculations of virulent
strains frequently result in abscess formation and the development of
a febrile reaction. These abscesses commonly ulcerate, discharge and
heal spontaneously. By no means do all virulent strains induce lesions,
however; there is great difference between them in this respect. Intra-
peritoneal injections frequently cause a rapidly fatal peritonitis with
or without septicemia. The intravenous injection of 0.25 to 1 c.c.
of an eighteen-hour broth culture usually causes a generalized pyemia
with septic foci, particularly frequent in the kidneys and liver. Orth 3
and Wyssokowitsch 4 have shown that mechanical injury to the heart
valves prior to the intravenous injection of staphylococci usually
causes a localization of the organisms there, producing an endocarditis.
If a bone is injured prior to an intravenous injection, a typical osteo-
myelitis frequently results. It should be remembered that the pus
produced by staphylococci in rabbits is more dry than that produced
in man. Guinea-pigs are less susceptible than rabbits to infection
with the staphylococcus.

Immunity and Immunization. Staphylococci do not ordinarily
exhibit invasive powers for man or animals; they are usually parasitic.
Whenever the continuity of the skin is destroyed, as by abrasion, or
weakened, as in diabetes, the organisms reach the underlying tissues
and induce inflammatory reactions. Repeated injections first of
killed then live staphylococci will frequently raise the threshold of
infection in experimental animals to a very considerable degree, but
the process of immunization can not be always relied upon many
animals die rather abruptly with rather extensive amyloid degenera-
tion, particularly of the kidneys. Leukocytes, particularly the poly-
morphonuclear leukocytes, appear to play a prominent part in the
immunity against staphylococci; it can be shown by experiment that
the leukocytes are more active phagocytically in immunized than in
non-immunized animals.

Similarly, the resistance to staphylococcus infection, which appears

1 Fortschritt d. Med., 1885, 170.

2 Cent. f. Bakt., Orig., 1903, xxxiv, 96.

3 Cent. f. d. med. Wissensch., 1905, No. 33.

4 Virchow's Arch., 1886, ciii.

262 THE PYOGENIC COCCI

to be rather marked in the average normal man, seems to depend
largely on the phagocytic activity of leukocytes in the last analysis;
and the efficiency of vaccines, particularly the autogenous vaccines,
in the treatment of furunculosis has focused attention sharply upon
the part played by opsonins in these infections. Generally speaking,
injections of killed cultures of staphylococci in graduated doses
beginning with one hundred millions and increasing to a thousand
millions or more at appropriate intervals exert a favorable influence on
the course of the infection. The efficiency of this vaccination (active
immunization) is attributed to the gradual development of specific
opsonins (bacteriotropins) which reenforce the action of normal
opsonins, whose activity is somewhat below normal. In practice
this is accomplished in the following manner: the organism is isolated
on agar slants in pure culture, washed off, after twenty-four hours'
incubation, in normal salt solution, thoroughly emulsified, and stan-
dardized so that each cubic centimeter contains the requisite number
of bacteria. They are killed either by heating to 80 C. for one hour,
or, better, by the addition of Ot5 per cent, carbolic acid, and incubation
at 37 C. for twenty-four hours. The sterility of the culture must be
demonstrated before it is used. This vaccine is inoculated subcutan-
eiously, with surgical precautions, using the dosage mentioned above
as a routine. The inoculations are repeated at intervals of from five
to eight days. The duration of the immunity induced by vaccination
is not known. Vaccines are less effective in pyemia and metastatic
staphylococcus infections than in the localized infections.

The lessened lipase activity of the blood, manifested by a decreased
splitting of ethyl butyrate, is a frequent result of staphylococcus
invasion, according to Clerc; 1 according to V. Dungern, 2 the blood
serum from cases of extensive osteomyelitis is several times as inhib-
itory to the staphylococcus enzymes as is that of normal individuals.

Antibodies. The cell substance of staphylococci does not appear
to be very poisonous to experimental animals, 3 and although an anti-
staphylolysin and an antileukocidin are relatively easily produced in
experimental animals, they do not appear to confer any consider-
able degree of immunity. Agglutinins do not appear to have been
demonstrated in the blood serum of man and animals suffering from
staphylococcal infections, but Kolb and Otto, and Proscher 4 claim

1 Compt. rend. Soc. de biol., 1901, liii, 1131.

2 Munchen. med. Wchnschr., 1898, xlv, 1040.

3 Kruse, Allgemeine Mikrobiologie, Leipzig, 1910, p. 968.

4 Cent. f. Bakt., 1903, xxxiv: quoted by Besson, Practical Bacteriology, 1913.

THE STAPHYLOCOCCUS GROUP 263

to have prepared sera of marked agglutinating value, which clump
virulent strains in higher dilution than non- virulent strains.

Precipitins. Specific precipitin reactions appear to have been
demonstrated in animals infected with staphylococci.

Bacteriological Diagnosis. (a) Microscopic. A Gram stain of the
suspected material usually suffices to establish a diagnosis. It must
be remembered, however, that staphylococci from pus and exudates
may occur in pairs and even in short chains; they may, therefore,
be mistaken for streptococci. An absolute diagnosis can be made
only by the identification of pure cultures.

(6) Cultural. Pure cultures of staphylococci are usually obtained
readily by "streaking out" or plating the organisms on agar. Blood
agar is preferable, if streptococci or pneumococci % are also suspected

FIG. 33. Micrococcus tetragenus. X 800.

to be present, otherwise the latter may be overlooked. The identi-
fication of the colonies on agar usually can be made by the examina-
tion of a Gram-stained preparation. Staphylococci are common on
the skin, and precautions must be taken to eliminate this source of
error before making cultures.

(c) Animal Inoculation. The virulence exhibited by staphylococci
for animals is not a reliable index of their virulence for man.

Dissemination and Prophylaxis. The wide distribution of staphy-
lococci on the mucous membranes, particularly on the skin and in
the hair follicles, makes the prevention of their introduction to under-
lying tissues through cuts and abrasions difficult. The customary
procedures of aseptic surgery are the best preventatives of infection.
The skin may be sterilized for operation (after thorough cleansing
and drying, which is imperative) by painting with freshly prepared

264 THE PYOGENIC COCCI

tincture of iodin or iodoform. Sterilization is usually accomplished
within ten minutes after the iodin is applied.

Staphylococcus Pyogenes Citreus. This organism differs from
Staphylococcus aureus chiefly in the color of the pigment it produces,
a lemon yellow, and a lessened ability to liquefy gelatin.

Staphylococcus Pyogenes Albus. In many instances this organism
is an achromogenic variant of Staphylococcus aureus: it produces
white colonies on agar and gelatin, it liquefies gelatin slowly, and it
is somewhat less pathogenic for rabbits; many strains do not ferment
mannite.

Staphylococcus Epidermidis Albus. Welch first described this
organism, which appears to be a degenerate Staphylococcus albus;
it does not liquefy gelatin and its pathogenic powers are practically
nil. It frequently causes the troublesome but relatively benign

stitch abscesses." It appears to be a very constant parasite on the
skin.

Micrococcus Tetragenus. Micrococcus tetragenus was first de-
scribed by Gaffky; 1 he found it in cavities of the lung in pulmonary
tuberculosis. It occurs but rarely in pure culture in abscesses either
in man or animals, 2 but it is often present in the saliva; occasionally
it has been recovered from dento-alveolar abscesses. 3

Morphology. The organism occurs typically in tetrads, enclosed in
transparent gelatinous capsules which require special staining methods
for their demonstration. The individual cells are about 1 micron in
diameter. In artificial media the tetrad arrangement may disappear
and the cocci occur chiefly in pairs and groups of three or four pairs.
The tetrad arrangement and the capsule are restored by passage
through animals. The organism is non-motile, and possesses no fla-
gella. It forms no spores and stains readily with ordinary anilin dyes.
It is Gram-positive.

Isolation and Culture. Micrococcus tetragenus grows rather slowly
in all ordinary media, particularly the first transfers from the tissues
to artificial media. It can be isolated readily in pure culture in gelatin
or agar plates; the colonies are small, round and grayish, 0.5 to 0.75
mm. in diameter.

Growth in Media. The organism does not liquefy gelatin, casein,
or blood serum. Acid is produced in dextrose, lactose, saccharose, and

1 Mitt. a. d. kais. Gesamte, i, p. 1.

2 Miiller, Wien. klin. Wchnschr., 1904, xvii, 1815.

3 Goadby, Mycology of the Mouth, 1903, p. 101.

THE STAPHYLOCOCCUS GROUP 265

mannite broths. A uniform turbidity is produced in plain and sugar
broths; the growth is more luxuriant in the latter. Milk is slightly
acidulated, but no coagulation or peptonization takes place. Micro-
coccus tetragenus is aerobic, facultatively anaerobic. The optimum
temperature of growth is 37 C., the maximum about 44 C., the mini-
mum about 12 C. The resistance to physical and chemical agents
is undetermined.

Products of Growth. Unknown: no toxin has been described.

Pathogenesis. The frequent occurrence of the organism in the
sputum of the tuberculous and its occasional isolation from tuber-
culous cavities has led to the theory that Micrococcus tetragenus may
play a secondary part in the destruction of lung tissue. This is not
definitely determined, however. It is also found in the saliva of healthy
individuals. Less commonly it has been found in the pus of empyemas
which follow pneumonia; but the organism can hardly be regarded
as a human pathogen.

Injected subcutaneously into white mice, Micrococcus tetragenus
usually causes a fatal septicemia; the organism may be recovered
from the heart blood, spleen and liver. House and field mice appear
to be relatively immune. Intraperitoneal injection into guinea-pigs
may cause a fatal peritonitis with much pus in which typical tetrads
are found. Rabbits and dogs are not .susceptible. Infections with
the organism in man are so uncommon that nothing is definitely known
of human susceptibility and immunity. Vaccines have been tried
in a very few cases with somewhat promising results.

Bacteriological Diagnosis. The finding of Gram-positive cocci about
1 micron in diameter in pus, which occur habitually in tetrads, usually
suffices to establish a satisfactory bacteriological diagnosis. The
saliva occasionally contains tetracocci which resemble Micrococcus
tetragenus very closely, but it is claimed by many that these organisms
are not necessarily Micrococcus tetragenus, Isolation and identifica-
tion by cultural methods must be resorted to in suspected cases.

Micrococcus Ovalis. Synonym. Enterococcus. 1

Historical. Micrococcus ovalis was described by Escherich, 2 who
found it very commonly in the intestinal tracts of nurslings and bottle-
fed infants.

Morphology. The organism is oval in outline, measuring 0.6 to
0.9 microns in the lesser diameter, and it occurs habitually in pairs,

1 Thiercelin, Th&se de Paris, 1894.

2 Darmbakterien des Sauglings, Stuttgart, 1886, p. 89.

266 THE PYOGENIC COCCI

with a tendency for the proximal ends to be slightly flattened and the
distal ends to be somewhat pointed. In this respect Micrococcus
ovalis resembles the pneumococcus very closely. In fluid media, par-
ticularly sugar broths, the pairs of organisms remain adherent in chains
of greater or lesser length giving rise to a diplostreptococcus arrange-
ment which is precisely like that exhibited by the pneumococcus under
the same conditions.

Micrococcus ovalis is non-motile and possesses no flagella. It forms
no spores. According to Lewkowicz 1 and others, capsules are produced
when the organism is isolated directly from lesions. The organism
stains readily with ordinary anilin dyes, and it is Gram positive.

Isolation and Culture. Micrococcus ovalis grows with moderate vigor
on agar plates, better in dextrose or lactose agar. The colonies after
forty-eight hours' incubation at 37 C. are round, translucent, color-
less, and measure about 1 to 2.5 microns in diameter. They are not
distinctive. Colonies on gelatin plates are very small and develop
slowly. The medium is not liquefied. Blood agar appears to be a
better medium for isolation of Micrococcus ovalis than any other;
the colonies are 1 to 3 mm. in diameter even after eighteen hours'
incubation, grayish and succulent. No hemolysis takes place. A
slight turbidity, which soon settles, forms in plain broth; the addition
of dextrose or lactose greatly enriches the growth. Milk is usually
coagulated in one to three days (acid coagulation), but the coagulum
does not become digested.

Micrococcus ovalis is an aerobic, facultatively anaerobic organism.
The lower limit of growth is about 8 C., the optimum from 37 to
39 C., and the maximum about 45 C. Its resistance to chemical
and physical agents is about the same as that of the streptococcus.

Products of Growth. Chemical. The organism exhibits no evidence
of proteolysins ; it is relatively inert in protein media. No indol,
skatol or volatile sulphur compounds are produced. Acid is produced
in dextrose and lactose broths; the action on other sugars is yet to
be determined.

Enzymes. No enzymes are known.

Toxins. No toxic products have been detected in cultures of
Micrococcus ovalis.

Distribution. The normal habitat of Micrococcus ovalis appears

to be the intestinal tract of man; it occurs in the meconium frequently, 2

1 Cent. f. Bakt., 1901, xxix, 635.

2 Escherich, loc. cit.

THE STAPHYLOCOCCUS GROUP 267

and it is a constant inhabitant of the intestinal flora of artificially fed
infants; it also occurs commonly, but in lesser numbers, in the intes-
tinal flora of the normal nursling. The organism has been repeatedly
isolated from the feces of adults, and it has also been isolated from the
intestinal tract of cattle. 1

Pathogenesis. Man. Micrococcus ovalis is ordinarily a harmless
parasite of the intestinal tract; occasionally it becomes invasive
(usually secondarily) and produces various inflammations, according
to the tissues invaded and its association with other bacteria. Lewko-
wicz 2 isolated Micrococcus ovalis in nearly pure culture from three
cases of severe dysentery; the organisms were found to be capsulated
and resembled pneumococci in a striking manner. Jouhaud, 3 Thier-
celin, 4 Ramonovitsch, 5 and Gilbert and Lippman 6 have isolated the
organism either in pure culture or in association with other bacteria
from cases of cholecystitis, puerperal fever, appendicitis, various
intestinal inflammations, and even from the cerebrospinal canal in
cases of meningitis. The close resemblance of the organism to the
pneumococcus, which has been observed by Kruse, 7 Sittler 8 and others,
has doubtless led to confusion; many cases of "pneumococcus" infec-
tion of the stomach, gall-bladder, appendix and other intestinal adnexa
are probably infections with Micrococcus ovalis, and vice versa.

Animal. Wilhelmi 9 has isolated Micrococcus ovalis from enteritides
of young cattle; Lewkowicz 10 has found the organism isolated directly
from human inflammations to be pathogenic for white mice. It exhibits
no pathogenicity as it occurs normally in the intestinal tract. 11

Bacteriological Diagnosis. 1. Microscopical. The presence of con-
siderable numbers of diplococci in the feces with their approximated
ends slightly flattened, their distal ends somewhat pointed, staining
intensely with the Gram stain, is frequently sufficient evidence to
establish a tentative diagnosis of Micrococcus ovalis.

2. Cultural. Various dilutions of feces or products of inflammation
are plated either on dextrose agar or "streaked out" on blood agar.

1 Wilhelmi, Landwirthschaft. Jahrb. f. Schweiz., 1899, xiii.

2 Cent. f. Bakt., 1901, xxix, 635.

3 These de Paris, 1903.

4 Comp. rend. Soc. de biol., 1902, No. 27; 1908, Ixiv, 76.
6 Ibid., 1911, Ixx, 122.

6 Ibid., 1902, No. 30.

7 Cent. f. Bakt., Orig., 1903, xxxiv, 737.

8 Die wichtigsten Bakterientypen der Darmflora beim Sauglinge, u. s. w., Wurzburg,
1909.

9 Landwirthschaftl. Jahrb. f. Schweiz., 1899, xiii.

10 Loc. cit.

11 Thiercelin, These de Paris, 1894; Compt. rend. Soc. de biol., April 15, 1899. Jou-
haud, These de Paris, 1903.

268 THE PYOGENIC COCCI

The morphology and cultural reactions outlined above suffice to estab-
lish a diagnosis. The absence of hemolysis or of green discoloration
of the hemoglobin separates the streptococcus and pneumococcus
from Micrococcus ovalis.

3. Serological. Not practicable.

Dissemination and Prophylaxis. Micrococcus ovalis does not cause
progressive disease from man to man; it is an intestinal parasite
habitually and only occasionally becomes invasive. No precautions
other than the careful sterilization of dejecta are necessary. The
hands of attendants should be kept surgically clean when caring for
intestinal disturbances incited by Micrococcus ovalis, or, indeed, by
any microorganism.

CHAPTER XIII.
THE STREPTOCOCCUS-PNEUMOCOCCUS GROUP.

THE STREPTOCOCCUS GROUP.
Streptococcus Pyogenes.

Streptococcus Einheit or Vielheit.
THE PNEUMOCOCCUS.

THE STREPTOCOCCUS GROUP.

THE Streptococcus Group comprises those spherical bacteria in
which as multiplication proceeds the successive planes of division are
parallel and the individual cells remain adherent in longer or shorter
chains. The limits of the group are poorly defined, both morphologi-
cally and pathogenically. It includes organisms which occur habitually
in chains, both in culture and in the animal body, and its limits have
been extended to enclose types which exhibit chain formation only
in fluid media. The latter, of which Micrococcus ovalis and the
pneumococcus are examples, occur in the animal body as diplo^^i,
and grow thus on solid media; in fluid media they grow habitually
in chains of greater or lesser length, in which, however, the typical
diplococcal arrangement persists. The term streptococcus, there-
fore, is a purely morphological one; it includes organisms which excite
various types of inflammation in man and in animals, together with
those which are ordinarily saprophytic.

The most important members of the group exist on the skin, and
particularly on the mucous membranes of man, as habitual parasites
or "opportunists." Streptococcus pyogenes and its variants are the
most common of these and the most versatile in their pathogenesis.

Streptococcus Pyogenes. Synonyms. Streptococcus erysipelatos ;
Streptococcus scarlatinosus; Streptococcus septicus.

Historical. Streptococci were seen in unstained pus by Klebs in
1872. Several years later Koch 1 demonstrated them in stained sec-
tions and in inflammatory exudates. Pasteur 2 appears to have been
the first to cultivate streptococci from cases of puerperal fever and to
differentiate them from staphylococci, both morphologically and by

1 Untersuchungen liber Wundinfektion, 1878.

2 Compt. rend. Acad. sci., 1880, xc, 1035.

270 STREPTOCOCCUS-PNEUMOCOCCUS GROUP

the character of the lesions which they excite. Ogsten 1 independently
confirmed Pasteur's observations. Fehleisen, 2 using more exact cul-
tural methods, isolated streptococci from a case of erysipelas; Rosen-
bach 3 studied the organism in great detail and introduced the name,
Streptococcus pyogenes.

Morphology. The individual cells are spherical, less commonly
oval, measuring from 0.5 to 1 micron in diameter. The size of
individual cells varies somewhat even in the same culture. The
organisms remain adherent in chains which vary in length from four
to twenty or more elements, in which a definite association of cocci
in pairs with their proximate sides flattened is occasionally observed.
The number of elements in the chain varies somewhat according to the
origin of the culture; it has been observed that streptococci freshly
isolated from lesions tend to occur in longer chains, while those organ-
isms which grow habitually upon the normal surfaces and mucous
membranes of the body appear more frequently in shorter chains.
V. Lingelsheim 4 has designated those strains which form chains of
eight or more cocci, Streptococcus longus; the short-chain types are
called Streptococcus brevis. Notwithstanding the frequent parallel-
ism of pathogenesis and development of long chains of cocci in artificial
media, in contradistinction to the lesser virulence of the short-chain
types, experience has shown that the length of the chains may also
be influenced directly by variations in the culture media. 5 This dis-
tinction, therefore, is untenable. Streptococci grown on solid media
are prone to group themselves in pairs, or even irregular masses,
resembling staphylococci. Similarly, the typical streptococcal arrange-
ment is frequently lacking in purulent inflammations of streptococcal
causation. Occasional cells in a chain of streptococci, especially in
old cultures, are met with which are distinctly larger than their fellows ;
they color somewhat differently and were formerly regarded as spores
arthrospores. 6 It is now known that they are not noticeably more
resistant than the more typical cells, and they are probably to be
regarded as involution forms.

Streptococcus pyogenes is non-motile, non-flagellated, and does
not produce true endospores. Occasional strains, isolated directly

1 Brit. Med. Jour., 1881.

2 Aetiol. d. Erysipelas, Berlin, 1883.

3 Mikroorganismen bei Wundinfektions-Krankh. des Menschen, Wiesbaden, 1884.

4 Zeit. f. Hyg., 1891, x, 331.

5 Hueppe, Die Methoden der Bakterien-Forschung, Wiesbaden, 1889, 24, 130.

6 See Aronson (Berl. klin. Wchnschr., 1896, No. 32; 1902, No. 42) and Vincent (Arch,
de med. exp.,-etc., 1902) for details.

THE STREPTOCOCCUS GROUP 271

from lesions or from animals, exhibit a delicate stainable zone around
individual organisms or pairs of organisms, which suggests capsules.
Howard and Perkins 1 have isolated such an organism which exhibited
a very definite capsule. It grew habitually in short chains in fluid
media, the individuals occurring typically in pairs. The organism
is closely related to the pneumococcus, and Dochez and Gillespie 2
have named it Pneumococcus mucosus.

Streptococcus pyogenes stains readily with ordinary anilin dyes.
It is typically Gram-positive, although old cultures may fail to retain
the Gram stain. The saprophytic types frequently are Gram-negative.

Isolation and Culture. Streptococci may be isolated directly from
inflamed areas and from pus upon agar plates, better upon dextrose
agar plates. The colonies are minute, gray and transparent, and
may be readily overlooked; if they occur in association with staphy-
lococci or other rapidly growing organisms, they are readily over-
grown. The more virulent varieties develop less readily, and require
the addition of blood or ascitic fluid to ordinary media for their initial
growth outside the body. On blood agar plates (one part human blood,
two parts of nutrient, sugar-free agar) the majority of virulent strep-
tococci produce a wide, clear zone of hemolysis 4 to 8 mm. in diameter
around each colony. This medium is particularly valuable for the
isolation of streptococci. 3 On Loffler's blood serum growth is mod-
erately luxuriant; typical chains are found in the condensation water
of solid media, but not as a rule upon the surface. The organisms
grow feebly in gelatin stab cultures producing a few small discrete
gray colonies along the line of inoculation. Little or no growth is
found on the surface of the medium. Liquefaction does not take place.

A slightly alkaline reaction (neutral to phenolphthalein) is most
favorable for the growth of streptococci; the addition of sugars, par-
ticularly dextrose, to ordinary media (but not blood agar) increases
the rate and extent of development, which, however, are soon limited
by the accumulation of acid products of fermentation. The addition

1 Jour. Med. Research, 1901, vi, 163.

2 Jour. Am. Med. Assn., 1913, Ixi, 727.

3 Schottmuller (Munch, med. Wchnschr., 1903, xx, 849) has classified streptococci
according to the changes they produce in blood agar as follows:

I. Streptococcus longus pyogenes seu erysipelatis (Streptococcus pyogenes) produces a
wide, clear zone of hemolysis around the colony; in blood broth the color changes to a
burgundy red. Long-chained streptococci.

II. Streptococcus mitior seu viridans (Streptococcus viridans) produces a greenish area
around the colony; a brownish color in blood broth. Short-chained streptococci.

III. Streptococcus mucosus. No hemolysis on blood agar. Colonies viscid. Organisms
distinctly encapsulated.

272 STREPTOCOCCUS-PNEUMOCOCCUS GROUP

of solid calcium carbonate (marble) to sugar media is important;
it neutralizes the excess of acid, and also appears to add somewhat to
the nutritive value of the medium. 1

Streptococci grow slowly in plain broth, producing a sediment after
twenty-four to forty-eight hours' incubation. A flocculent sediment
consisting of long chains of organisms is characteristic but not distinc-
tive of many virulent strains (Streptococcus conglomeratus) ; a
granular sediment usually contains short-chain streptococci almost
exclusively.

Streptococcus pyogenes ferments dextrose, lactose, maltose and
saccharose and sorbite with the formation of considerable amounts of
acid. Mannite is not as a rule attacked. Milk is coagulated in from
three to five days, the coagulum resulting from the accumulation of
the acid fermentation of the lactose. The coagulum is never dissolved.
Andrewes and Horder 2 state that Streptococcus pyogenes does not
coagulate milk, although the organism produces a considerable amount
of acid in this medium. Smith and Brown 3 have shown that boiling
the milk may be necessary to make the coagulum visible.

Streptococcus pyogenes is an aerobic, facultatively anaerobic
organism. Pathogenic strains do not as a rule grow below 16 to 18
C. The optimum temperature lies between 35 and 39 C., the maxi-
mum about 44 C. The parasitic types are not long-lived away from
the human body. Exposure to 60 C. for one hour will kill most
streptococci; a longer time is required if the organisms are exposed
in albuminous media. Five per cent, carbolic acid and 1 to 1000 mer-
curic chloride will kill the naked germs in from five to ten minutes.
Streptococci dried in sputum will resist a temperature of 100 C.
(in flowing steam) for several minutes, and drying at ordinary tem-
peratures in the dark for several weeks. Direct sunlight kills them in
about ten hours. The organisms survive and retain their virulence
if they are suspended in sterile, defibrinated blood and kept in the
ice box for several weeks.

Products of Growth. Chemical. Streptococci exhibit but little
evidence of proteolytic activity. No indol, skatol, phenol or other
aromatic derivatives of amino acids have been detected in cultures;
gelatin is not liquefied and casein and coagulated blood serum are
not visibly changed. Emmerling 4 found peptone, leucin, ty rosin,

1 Bolduan, New York Med. Jour., 1905, May 13.

2 Lancet, 1906, ii, 708.

3 Jour. Med. Research, 1914, xxxi, 455.
* Berl. chem. Gesell., 1897, 1863.

THE STREPTOCOCCUS GROUP 273

ammonia, methylamine, propyl pyridin, succinic acid, butyric acid
and other volatile acids among the anaerobic decomposition products
of fibrin by this organism, but no aromatic derivatives.

Toxin. A soluble toxin has not been demonstrated in cultures of
streptococci, although substances have been isolated by Marmorek 1
and others which will kill guinea-pigs. These substances do not
exhibit sufficient potency to warrant the assumption that they are
important factors in the production of the grave symptoms charac-
teristic of severe streptococcus infections. Attempts to demonstrate
endotoxin have also been unsuccessful; the bodies of the organisms
are but slightly toxic to experimental animals. The manifestations
of toxemia in streptococcal infections, however, are too striking to

FIG. 34. Streptococcus in pus. X 800.

be reconciled with the negative results of these investigations; the
nature of the mechanism of streptococcus infection remains to be
elucidated.

Hemolysin Streptocolysin. Bordet 2 and Besredka 3 have shown
that filtered broth cultures of streptococci will dissolve red blood
corpuscles, liberating hemoglobin, and that this hemolytic substance
streptocolysin is active both in mw and in vitro. Frequently the
blood of rabbits injected with streptocolysin was found to be u laked"
just before death. Besredka's observations would indicate that the
substance is rather firmly bound to the organisms and does not appear
in the medium to any considerable degree. M'Leod, 4 M'Leod and

1 Berl. klin. Wchnschr., 1902, xiv, 253.

2 Ann. Inst. Past., 1897, xi, 177.
a Ibid., 1901, xv, 880.

4 Jour. Path, and Bact., 1912, xvi, 321.

274 STREPTOCOCCUS-PNEUMOCOCCUS GROUP

M'Nee, 1 and Lyall 2 have studied the conditions favoring the formation
of the hemolysin and find that sugar-free ascitic broth is suitable for
this purpose. The substance is thermolabile and is found in an active
state only during the first twelve to twenty-four hours of culture,
at which time small amounts of sterile (filtered) broth, 0.01 to 0.10
c.c., are strongly hemolytic. The hemolysin does not induce antibody]
formation when it is injected into susceptible animals. Hemoglobin-
emia and hemoglobinurea are produced in rabbits that are very sus-
ceptible to the hemolysin; less susceptible rabbits- react but slightly.
There is no definite evidence that streptocolysin plays a prominent
part in the streptococcus infections of man. Virulence and hemolytic
activity are frequently, but by no means necessarily, parallel pheno-
mena.

Distribution in Nature. Streptococci are widely distributed in nature,
always, however, in rather intimate association with man or the
higher animals. They are found in the soil, water, milk, and they
exist as "opportunists" on the exposed surfaces and mucous mem-
branes of man. They are common in the mouth, nose and throat, the
intestinal tract, and rare in the normal vagina.

Pathogenesis. Human. Streptococci excite both local inflam-
matory and suppurative processes and generalized septicemic infec-
tions, the latter being the more common and characteristic. Super-
ficial lesions may be mild in character, resembling those caused by
staphylococci. The organisms may, and frequently do, enter the
blood or lymph channels, and spread rapidly through the body, incit-
ing the most severe generalized infections. Streptococci are the etio-
logical agents of erysipelas, frequently of general and puerperal sepsis
and phlebitis, and inflammations of the internal organs; of these,
the middle ear, the endocardium, the peritoneum, the.meninges or
joints are more commonly involved. 3 Escherich 4 and others have
described a severe type of enteritis, particularly of young children
streptococcus enteritis which occasionally exhibits an epidemic
tendency in the summer months. 5 Attention has been directed in
recent years to severe epidemics of septic sore throat in which the

1 Ibid., 1913, xvii, 524.

2 Jour. Med. Research, 1914, xxx, 487.

3 Menzer, Deut. med. Wchnschr., 1901, 97. Meyer, Zeit. f. klin. Med., 1902, xlvi,
311; Internal. Beitrage zur inn. Med., 1902, ii, 443. Philipp, Deut. Arch. f. klin.
Med., 1903, Ixxvi, 150. Poynton and Payne, Cent. f. Bakt., Orig., 1902, xxxi, 502. Cole,
Jour. Inf. Dis., 1904, i, 714. Rosenow, Jour. Inf. Dis., 1910, vii, 411; ibid., 1912, xi,
210; Jour. Am. Med. Assn., 1913, Ix, 1223.

4 Jahrb. f. Kinderheilk., 1899, xlix, 137.

5 Kendall, Day and Bagg, Boston Med. and Surg. Jour., 1913, clxix, 741.

THE STREPTOCOCCUS GROUP 275

evidence points to streptococci transmitted through milk as the
etiological agent. The type of streptococcus involved has been a
subject of controversy, but the extensive studies of Smith and Brown 1
show clearly that Streptococcus pyogenes is by far the most common
organism found. They demonstrated that the streptococcus which
is isolated from bovine mastitis is not, except possibly in rare instances,
a causative factor in epidemic sore throat.

Streptococci occur frequently as secondary invaders in diphtheria,
many gastro-intestinal diseases, and diseases of the lungs, where they
may be at times even more formidable than the primary infecting
organism. As Theobald Smith has admirably expressed it, they are
''organisms of the diseased state." The virulence exhibited by strep-
tococci varies considerably, as does the type of lesions they excite. This
variation in virulence is not at all well understood at the present time,
but experiments indicate that the site of infection and the past history
of the organism exercise some influence. Rosenow 2 has isolated
streptococci, using special methods, from the regional glands in arth-
ritis, gall-bladders, and gastric ulcers. He states that the freshly-
isolated strains exhibit rather marked tendencies to localize in the
homologous tissues of experimental animals. This specific tissue
affinity is rapidly lost during cultivation of the organisms in artificial
media, however.

Animal. Frankel, 3 Petruschky, 4 and Koch and Petruschky 5 showed
that the virulence of the same strain of streptococcus varied materially
according to the conditions of culture, and that the lesions produced
in rabbits varied likewise; thus the descendants of the same culture
would produce variously a rapidly fatal septicemia, erysipelas, arth-
ritis, endocarditis or peritonitis. Marmorek has shown that the viru-
lence of streptococci for animals may be greatly increased by repeated
passage; after a series of passages an incredibly small amount of cul-
ture, even one one-hundred-millionth of a cubic centimeter of a forty-
eight-hour broth culture introduced intraperitoneally may cause death
within two days. Streptococci which are virulent for man frequently
exhibit but little virulence for animals; it is essential, therefore,
that large amounts of material be injected into experimental animals
to obtain infection. Rabbits are more susceptible than other labora-

Jour. Med. Research, 1914, xxxi, 455.

Jour. Am. Med. Assn., 1913, Ix, 1223; Ixi, 1947; 1914, Ixiii, 1835. Jour. Inf. Dis.,
1915, xvi, No. 2.

Cent. f. Bakt., 1889, vi, 671.
Zeit. f. Hyg., 1896, xxiii, 144.
Ibid., p. 478.

276

STREPTOCOCC US-PNE UMOCOCC US GRO UP

tory animals. Subcutaneous injections of morbid material into
rabbits result variously, depending upon the virulence of the strain
for this animal (not necessarily upon its virulence for man) ; a localized
abscess may form or an erysipelatoid inflammation may occur, which
is usually somewhat localized, but may develop into a wide-spread
cellulitis. Intraperitoneal injections are usually followed by rapidly-
fatal peritonitis. Death may occur within twenty-four hours. Intra-
venous injections may cause a rapidly fatal generalized septicemia, or,
if the strain is less virulent and death does not occur during the first
three to four days, the serous surfaces may be violently inflamed. Less
virulent strains which do not cause acute death usually lead to endo-
cardial or joint involvement. Mice are nearly as susceptible to strepto-

FIG. 35. Streptococci in liver, section stained by Gram's method. X 800. (KoJle and

Hetsch.)

coccus infection as rabbits. Guinea-pigs are less susceptible; subcu-
taneous inoculations usually lead to abscess formation, which soon
heals, but intraperitoneal injections may result in peritonitis and
death. Horses are quite susceptible to infection with streptococci,
particularly with Streptococcus equi (Streptococcus coryzse contagiosse
equorum), which causes equine distemper or strangles. The udders of
milch cattle occasionally become infected with streptococci result-
ing in a severe inflammation, mastitis or garget, which may lead to
loss of function of one or more quarters of the udder. It is probable
from the investigations of Smith and Brown 1 that streptococci of
bovine origin are not commonly the etiological agents of septic sore
throat in man.

1 Loc. cit.

THE STREPTOCOCCUS GROUP 277

Immunity and Immunization. Streptococcus infections, mild or
severe, do not appear to induce any considerable degree of active
immunity. Not infrequently recovery is a matter of some time; the
acute symptoms may abate and the organisms disappear from the
blood stream, only to localize in some internal organ, a structure as for
example, a joint, where they may cause a chronic, obstinate arthritis.
It is possible that various strains of streptococci which can not be
differentiated by our somewhat artificial cultural criteria may exist,
and that subsequent infection may be with another strain. A similar
condition exists in lobar pneumonia. Van de Velde 1 has stated that
the serum of an animal immunized against one strain of streptococcus
will protect against the homologous strain, but not against hetero-
logous strains of streptococci, a somewhat parallel situation. On the
other hand, experiments are recorded which are not in accord with
this hypothesis. A patient suffering from an inoperable tumor was
inoculated subcutaneously with a culture of streptococcus; the
inoculation resulted in a moderately severe erysipelas which per-
sisted for about ten days; when the inflammation had subsided a
second reinoculation was made in the same place, and a secondary
erysipelatoid inflammation spread over the same area. A third
inoculation resulted similarly. These experiments indicate that this
patient did not develop immunity at the site of infection. 2

Rabbits have been actively immunized to streptococci through
repeated vaccination, first with killed cultures, then gradually
increasing doses of living, virulent organisms; eventually the animals
will resist successfully several times the original fatal dose of the
homologous strain. Active immunization with polyvalent vaccines
containing many strains of streptococci from lesions is considerably
more efficient in protecting the animal against subsequent infection
with heterogeneous strains. The sera of such actively immunized
animals do not possess noteworthy antihemolytic properties; their anti-
toxic content, if indeed there be any, is unknown. The chief demon-
strable change in the serum appears to be an increased phagocytic
power, causing Jeukocytes in vitro to take up more streptococci than
they would normally. The injection of sera of actively immunized
animals appears to increase the resistance of non-immunized animals
to otherwise fatal amounts of streptococci.

1 Cent. f. Bakt,, 1898, xxiv, 688.

2 Coley has injected streptococci into malignant tumors with occasional beneficial
results; the observations are too few to warrant any definite statement of the efficiency
of the procedure.

278 STREPTOCOCCUS-PNEUMOCOCCUS GROUP

Marmorek, 1 Tavel 2 and others have prepared antistreptococcic
immune sera on a large scale by immunizing horses first with killed
cultures, then with increasing amounts of living cultures. Marmorek,
a staunch supporter of the "Einheit" theory that all streptococci
were identical, used a single strain of organism, whose virulence was
greatly increased for rabbits prior to injection into horses. Immuniza-
tion requires several months. He found that for some days following
each injection the horse exhibited a febrile reaction, and during that
period the serum was toxic for rabbits; streptococci may be found in
the blood stream during this period. After the temperature has
reached normal three weeks or more after the injection the toxic
properties disappear and the serum exhibits protective powers when
it is introduced into rabbits with a lethal dose of streptococci. This
serum has been used extensively in the treatment of erysipelas, puer-
peral fever, and scarlet fever, but its curative value is still a matter
of discussion.

Tavel's serum is essentially like that of Marmorek, except that a
polyvalent vaccine is used for immunization. Besredka also uses a
polyvalent vaccine for immunizing horses, but the organisms are
not exalted in virulence for rabbits by passage through a series of
them before inoculating horses. Besredka believes that passage
through rabbits may modify the virulence of the streptococci for
man, from whom the organisms are obtained for immunizing the
horses, and for whom the serum is to be used. Streptococcal sera are
as yet of debatable value; in localized lesions they have frequently
exhibited some therapeutic value; in the severe generalized infections
in man they are usually either irregular in their action or inactive.
Somewhat more encouraging results have been reported where the
specific immune serum is used in connection with autogenous vaccines
of streptococci.

Antibodies. Agglutinins are present in the sera of animals immu-
nized with streptococcus vaccines, and the degree of agglutinating
power may be very considerable for homologous strains. The results
are usually less definite with heterologous strains, and agglutinins
developed during immunization with streptococci are of no consider-
able value in prognosis. The part they may play in immunization
is problematical.

Complement fixation has not been found a satisfactory method for

1 Ann. Inst. Past., 1895, ix, 593.

2 Loc. cit.

THE STREPTOCOCCUS GROUP 279

identifying streptococci; the results are occasionally variable without
apparent cause.

Bacteriological Diagnosis. 1. Microscopical Examination. Smears
from abscesses or inflammatory areas usually exhibit pairs and short
chains of cocci which retain the Gram stain. Occasionally the organ-
isms can not be distinguished with certainty from staphylococci.
Frequently, when microscopic examination fails (and this is usually
the case when blood is examined), streptococci are found by cultural
methods.

2. Cultural Examination. If the material is purulent, it may be
streaked or plated out on 0.1 per cent, dextrose agar; the colonies are
small and transparent, and may be easily overlooked. Blood, lymph
or serum should be plated on blood agar. If the material is blood, one
part may be added to two parts of melted plain agar, and the whole,
after thorough mixing, may be poured into sterile Petri dishes. Usually
small, gray colonies with relatively broad, clear areas of hemolysis
appear within forty-eight hours. If lymph and serum be. the sus-
pected material, blood agar should be used for plating out. Hemolytic
colonies, as above, appear usually within two days. It is always
well to inoculate 1 or 2 c.c. of blood serum or lymph into broth and
maintain it at 37 C. for twenty-four hours to enrich the culture, then
plate on blood agar; also inoculate a like amount into a rabbit.

3. Animal Inoculation. The intraperitoneal injection of suspected
fluids into rabbits frequently results in a fatal perftonitis, from which
the organism may be recovered from the blood stream. Relatively
large amounts should be used.

The detection of streptococci in the blood of a patient is frequently
an unfavorable clinical sign; it does not necessarily, however, justify
a grave prognosis. Cases are met with which present symptoms of
septicemia, yet the organisms may not be obtained from the blood.
Occasionally the patient dies from toxemia, due apparently to the
absorption of toxic substances from the local infection. Streptococci
from erysipelas, septicemia, scarlet fever, and even from articular
rheumatism are so similar culturally and morphologically that the
various strains can not be differentiated with certainty; slight varia-
tions in cultural reactions are exhibited by all these strains. Neither
does animal experimentation afford definite criteria for the estab-
lishment of types. Even one passage through an animal may modify
the pathogenicity greatly.

In the light of our present knowledge the resistance of different

280 STREPTOCOCCUS-PNEUMOCOCCUS GROUP

tissues and the portal of entry play a prominent part in determining
both the type of lesion which will result from invasion of the body
by streptococci, and the modification in virulence they may undergo
in man or animal as the struggle between host and invader is extended
in time.

Prophylaxis. General surgical aseptic methods. Autogenous vac-
cines have been extensively used in streptococcus infections, but with
less favorable results than autogenous staphylococcus vaccines.

The Streptococcus Einheit or Vielheit. Considerable discussion
has arisen concerning the unity or the plurality of types included
within the organism known as Streptococcus pyogenes. Marmorek 1
and others have stoutly maintained the Einheit theory. Considerable

FIG. 36. Pneumococcus mucosus showing capsule. X 1000.

evidence in favor of this view was advanced by Koch and Petruschky, 2
who showed that a streptococcus obtained from a fatal puerperal
sepsis caused erysipelas in a rabbit when it was injected subcutaneously,
peritonitis when injected intraperitoneally, and septicemia when
introduced intravenously. The organisms freshly isolated caused a
rapidly fatal septicemia when introduced through the blood stream,
but the virulence was gradually lost following cultivation on artificial
media; the septicemic phenomena diminished in intensity and there
was evidence of a localization of the organisms. Their conclusions
were that the type of lesion produced by Streptococcus pyogenes
depended largely upon the virulence of the culture, the tissue invaded,
and the number of organisms. With a comparatively slight loss in
virulence the endocardium appeared to be somewhat more frequently

1 Berl. klin. Wchnschr., 1902, xxxix, 299.

2 Loc. cit.

THE STREPTOCOCCUS GROUP 281

the site of the focal infection; with a greater loss of virulence, the
joints. It must be remembered in this connection that the virulence
of a streptococcus for man does not necessarily determine the virulence
for animals.

It is possible to raise the virulence of streptococci very materially
by artificially creating portals of entry and of escape which are not
usually available to the streptococcus. This is accomplished by
passage through experimental animals. By passage it is possible
to reproduce with considerable accuracy the various reactions men-
tioned above, depending upon the virulence of the organism, the
tissue into which the injection is made, and the number of organisms
introduced. It is also important to remember than an increase in
virulence for one animal, attained by frequent passages, frequently
results in a loss, partial or complete, of the virulence of the organism
for another animal. Too little is known of the mechanism of virulence,
however, to place a final interpretation upon the biological signifi-
cance of changes in pathogenic powers.

Additional evidence of the Einheit of streptococci has been brought
forward by Rosenow, 1 who states that he has changed streptococci
to pneumococci and back again by special methods of culture and
animal inoculation. Two possibilities present themselves to explain
this phenomenon, if Rosenow's claims are substantiated. First, the
streptococcus-pneumococcus complex is a single organism which
exhibits nodes of relative cultural stability (assuming that present-
day methods for the recognition of bacterial types are fundamentally
sound), and the organism may pass from one node to another under
the stress of environmental stimuli. The second possibility is that
the streptococcus and pneumococcus are in reality distinct biological
entities and that an actual discontinuous mutation has occurred.
The many variables to be considered in this connection variations
in virulence, adaptability to various hosts, and changes in appearance
in different media, all of which may change independently of or parallel
to each other complicate the problem to a considerable degree;
final judgment must await the establishment of authoritative standards
for bacterial diagnosis of unquestioned fundamental stability.

Neufeld, and Cole and his associates have presented a new aspect
of the problem. They found that the older conception of the unity
of the pneumococcus type was untenable. They found there were
four distinct types of pneumococcus which were recognizable both

1 Loc. cit.

282 STREPTOCOCCUS-PNEUMOCOCCUS CROUP

by serological and pathological methods, and that these types were
mutually stable, for long-continued passage through animals failed
to alter or modify their general cultural and agglutinating properties,
although the virulence of the respective types for one or another
animal could be increased or decreased. It is not improbable that a
thorough study of the streptococcus group may reveal similar sero-
logical variance and that in the type now designated Streptococcus
pyogenes several individual types parallel to those of the pneumococcus
may be demonstrated.

The important question for the moment is, do these changes of
virulence, et cetera, exhibited by the streptococcus influence the diag-
nostic aspect of the question? Theobald Smith has admirably summed
up the present status of the subject in the following words : " Spon-
taneous changes in the cultural characters of the streptococcus do not
proceed rapidly enough, if they go on at all, to interfere with current
bacteriological methods. Tendencies toward slow changes may be
used as further valuable distinguishing characters." 1

THE PNEUMOCOCCUS.

Synonyms. Micrococcus pasteuri, Diplococcus pneumonia, Diplo-
coccus lanceolatus, Streptococcus lanceolatus.

Historical. Although the pneumococcus was observed by Stern-
berg 2 and independently by Pasteur 3 in the blood of rabbits inoculated
with sputum, the etiological relationship of the organism to lobar
pneumonia was not established until 1886, when Frankel 4 and Weich-
selbaum 5 published their respective studies upon lobar pneumonia.

Morphology. Viewed under the microscope, the pneumococcus
presents two distinct appearances, depending upon the source of the
culture. Observed in human or animal tissues, exudates or body
fluids, or in media containing non-coagulated albuminous fluids, as
blood serum, ascitic or hydrocele fluids, the organisms occur typically
in pairs surrounded by a definite capsule, or less commonly in short
chains enclosed in a capsule. The individual cells are typically lanceo-
late in shape with the apposed surfaces of each pair flattened, and the
distal ends somewhat pointed. Less commonly the organisms are
oval, or nearly spherical. The paired arrangement is maintained
when the organisms remain adherent to form short chains. Cultures

1 Smith and Brown, Jour. Med. Research, 1914, xxxi, 501.

2 National Bureau of Health, 1881.

3 Compt. rend. Acad. Sci., 1881, xcii, 159.

4 Zeit. f. klin. Med., 1886, x, 401. Ibid, xi, 437.
6 Wien. med. Jahrb., 1886, p. 483.

THE PNEUMOCOCCUS

in artificial media which do not contain albuminous fluids are not
encapsulated, and the distinctive lanceolate shape is frequently lost;
the organisms become more nearly oval or spherical in outline, but
the tendency to remain adherent in pairs is usually maintained. Chains
of from four to eight elements are developed in broth cultures, which
has led many observers to include the pneumococcus in the strepto-
coccus group. The size of the organisms varies considerably; ordi-
narily the lesser diameter measures from 0.5 to 0.8 microns, and the
longer diameter from 1 to 1.3 microns.

The pneumococcus is non-motile and possesses no flagella. The
capsule, which surrounds pairs of organisms derived from sputum,
tissue, body fluids and exudates of man and animals, as well as those

FIG. 37. Pneumococcus showing capsules.

cultivated in milk or media containing uncoagulated albuminous sub-
stances, is readily demonstrated by the methods of Welch, 1 Hiss 2 and
Rosenow. 3 The capsule is poorly formed or absent from pneumo-
cocci derived from chronic processes or from mucous surfaces where
the organisms are growing as parasites or "opportunists."

The ordinary anilin dyes stain pneumococci readily, and they are
Gram-positive when freshly isolated, but tend to become Gram-
negative during cultivation in artificial media.

Isolation and Culture. Pneumococci grow slowly and feebly upon
ordinary laboratory media, and they soon perish. Cultures may
be obtained from the blood stream in a large percentage of cases from
the fifth day of the disease to the crisis 4 by inoculating 5 to 10 c.c. of

1 Bull. Johns Hopkins Hospital, 1892, xiii, 128.

2 Cent. f. Bakt., Ref., 1902, xxxi, 302.

3 Jour. Infec. Dis., 1911, ix, 1.

4 Rosenow, Jour. Inf. Dis., 1904, i, 280,

284 STREPTOCOCCUS-PNEUMOCOCCUS GROUP

blood into 100 to 150 c.c. of 0.1 per cent, dextrose broth, and incubating
for twenty-four hours at 37 C. Isolation of pneumococci from sputum
by cultural methods is practically hopeless; but pure cultures may
be obtained from the heart blood of white mice inoculated subcutan-
eously with sputum.

The organisms may be obtained from inflammatory exudates and
pus either by inoculation of the material into white mice or infecting
the surface of blood agar, serum, ascitic or hydrocele agar plates.
Colonies on blood agar plates are minute, gray, and surrounded by a
greenish halo which Butterfield and Peabody 1 and Cole 2 have shown
to be methemoglobin. Colonies on ascitic agar are small, transparent
and colorless. The growth upon plain nutrient agar or gelatin is very
scanty. Gelatin is not liquefied. The addition of dextrose to agar
increases the nutritive value of the medium, but the acid formed by
the fermentation of the dextrose soon kills the bacteria unless calcium
carbonate is added to neutralize the acid. Many strains of pneumo-
cocci grow in milk, producing as a rule sufficient acid to cause coagula-
tion. The coagulum is never liquefied. Growth upon Loffler's blood
serum is moderately luxuriant, particularly for subcultures; initial
development of the organisms directly from human or animal sources
is not extensive upon this medium. The colonies are small, clear and
colorless, and not distinctive. Growth is more rapid in fluid than
in solid media. Secondary inoculations into plain broth or broth
containing utilizable carbohydrates result in a clouding of the medium
and extensive development, more luxuriant in the latter than the
former. The addition of blood, blood serum or ascitic fluid to media
increases the nutritive value greatly. The organisms die within a
few days, and even after twenty-four hours' incubation degenerative
forms appear, and they become Gram-negative. Transfer at frequent
intervals to fresh media is essential to maintain viable cultures of the
pneumococcus.

The pneumococcus is an aerobic, facultatively anaerobic organism
whose limits of growth lie between 25 C., below which development
ceases, and about 42 to 43 C.; the optimum temperature of growth
is 37 C. The organisms are not resistant to heat, being killed by an
exposure of ten to fifteen minutes to 55 C. 3 Chemical disinfectants,
as 5 per cent, carbolic acid or 1 to 1000 bichloride of mercury, destroy
pneumococci readily. Dried rapidly in sputum, they retain their

1 Jour. Exp. Med., 1913, xvii, 587.

2 Ibid., 1914, xx, 363.

3 See Wood, Jour. Exp. Mod., 1905, vii, 592, for literature.

THE PNEUMOCOCCUS

285

viability for nearly two weeks, but sunlight is rapidly fatal. The
virulence is rapidly lost during cultivation in artificial media, but it
may be retained practically unimpaired for weeks if the organisms
suspended in blood are sealed in glass tubes and maintained in the
dark at ice-box temperature. Pneumococci obtained from sputum,
either of healthy individuals or from the "rusty sputum" character-
istic of the earlier stages of lobar pneumonia, possess sufficient viru-

FIG. 38. Pneumococcus in sputum. X 1000.

lence to kill white mice. The original virulence may frequently be
restored to cultures on artificial media by passage through white
mice, provided large doses are administered at the start. Repeated,
rapid inoculations of virulent pneumococci frequently lead to a decided
increase of virulence above that originally exhibited by the organisms.
Products of Growth. Chemical. The pneumococcus produces acids,
chiefly lactic, but smaller amounts of formic acid, in hexoses, bioses,
and many starches. Hiss 1 has shown that the fermentation of inulin
by the pneumococcus is a very constant cultural differentiation of
the organism from the streptococcus, which is unable to ferment this
starch. Another important method of distinguishing between pneumo-
cocci and streptococci is the solubility of the former in bile or a freshly
prepared solution of sodium chlorate. 2 3 Colonies of the pneumococcus
on blood agar are surrounded by a greenish zone of methemoglobin. 4

1 Jour. Exp. Med., August, 1905, vii., 547.

2 Neufeld, Zeit. f. Hyg., 1900, xxxiv, 454. Wadsworth, Jour. Med. Research, 1904,
x, 228.

3 The test is made as follows: 1 c.c. of a twenty-four-hour broth culture of the sus-
pected organism is mixed with 0.1 c.c. of a freshly-prepared 2 per cent, solution of sod-
ium chlorate and maintained at 37 C. Clearing of the solution indicating solution of
the organisms does not take place uniformly; some cultures dissolve more rapidly than
others. Cole, Jour. Exp. Med., 1912, xvi, 658. Acids interfere with the success of the
test.

4 Butterfield and Peabody, loc. cit. Cole, loc. cit.

286 STREPTOCOCCUS-PNEUMOCOCCUS GROUP

Enzymes have not been demonstrated in cultures of the pneumo-
coccus.

Toxins. Soluble toxins have not been detected in cultures of
pneumococci, although the filtrates obtained by Klemperer, 1 Wash-
bourn, 2 and Isaeff 3 were toxic for small laboratory animals. The
toxicity observed in these preparations was probably due to the
liberation of endotoxins as the result of autolysis of pneumococci in
the medium. Macfadyen 4 has obtained toxic substances from two-
to three-day agar cultures of virulent pneumococci, which were
ground finely after freezing with liquid air (method of Macfadyen
and Roland), then extracted with 1 to 1000 potassium hydrate,
centrifugalized to remove fragments of the organisms and filtered.
A small amount of the filtrate, 0.5 to 1 c.c. in rabbits, 0.1 to 1 c.c. in
guinea-pigs, produced death when injected intravenously or intra-
peritoneally. The toxicity of the filtrate was roughly proportional
to the virulence of the organisms for rabbits. Heating the filtrate
to 55 C. for an hour, or exposure to chloroform vapor for the same
time reduced the toxicity of the preparation very considerably. Neu-
feld and Dold 5 and Rosenow 6 obtained toxic substances from pneumo-
cocci, the former by extraction of the organisms in 0.1 per cent,
lecithin in physiological salt solution, the latter by simple autolysis,
which induced symptoms in guinea-pigs suggesting acute anaphy-
laxis. Cole 7 has repeated these experiments with results that were
irregular: thus, of 213 guinea-pigs injected with extracts of pneumo-
cocci in salt solution, 8 died acutely with symptoms resembling acute
anaphylactic shock, 83 died within twelve hours, the remainder were
negative. Cole concludes that extracts of pneumococci in salt solu-
tion may be toxic, but not uniformly so. The exact conditions under
which these solutions become toxic are unknown. Solutions of
pneumococci dissolved in dilute solutions of bile salts were found to
be very constantly toxic. 8 The intravenous injection of these solu-
tions into rabbits and guinea-pigs elicits symptoms resembling closely
those of acute anaphylaxis. Many of the animals die acutely.

1 Zeit. klin. Med., 1891, xx, 165.

2 Jour. Path, and Bact., 1897, iii, 214.
. 3 Ann. Inst. Past., 1892, vii, 259.

4 Cent. f. Bakt., Orig., 1907, xliii, 30.

5 Berl. klin. Wchnschr., 1911, xlviii, 1069.

6 Jour. Infec. Dis., 1911, ix, 190.

7 Jour. Exp. Med., 1912, xvi, 644.

8 Casagrandi (quoted by Pribram: Kolle and Wassermann Handb., 2 ed., 1913, iia,
1350) states that normal rabbit blood contains antihemolysins.

THE PNEUMOCOCCUS 287

Hemotoxin. Recently Cole 1 has shown that solutions obtained by
dissolving pneumococci in dilute solutions of bile salts, or by tritura-
tion, are hemolytic for rabbits, guinea-pigs, sheep and human red blood
cells, and that their activity is inhibited by minute amounts of choles-
terin. The injection of these solutions in gradually increasing amount
leads to an inhibition of their action; in other words, this "hemolytic
endotoxin" appears to act as an antigen.

Pathogenesis. Human. At least 90 per cent, of all cases of lobar
pneumonia, one of the most prevalent and fatal of human diseases,
is caused by the pneumococcus, but this disease is by no means the
only one in which the organism is an etiological factor. Many
bronchopneumonias which follow acute infections, as typhoid, diph-
theria, so-called " aspiration pneumonia," are also of pneumococcic
causation. Pleurisy, a frequent complication of both types of pneu-
monia, is quite commonly a pneumococcus infection, and a majority
of sporadic cases of meningitis, particularly in children, are also caused
by the organism. Indeed, in children the pneumococcus is rather
more commonly isolated from suppurative processes than any other
organism; in adults the incidence of pneumococci in suppurations is
on the whole considerably less. Middle ear involvement, inflamed
mastoids, endo- and pericarditis are all frequently caused by the
pneumococcus. The channel of infection appears to be through the
blood stream, and pneumococci have been isolated from the blood
stream in a very large percentage of all cases of lobar pneumonia. 2
Less commonly the organisms become localized in joints, causing
arthritis, and around the shafts of bones, causing osteomyelitis.
Conjunctival inflammation of varying degrees of severity which
occasionally leads to ulcer formation is frequently a pneumococcus
infection.

It was formerly stated that virulent pneumococci could be obtained
from the sputum of fully 30 per cent, of normal individuals. The
supposition was that the patient became the victim of his own
organisms. Recent studies by Dochez and Avery 3 suggest strongly
that the pneumococci found in the sputum during pneumonia are
commonly replaced by pneumococci of a less virulent type soon after
convalescence. Their observations, furthermore, make it justifiable
to consider those patients who harbor the more virulent types after

1 Jour. Exp. Med., 1914, xx, 346.

2 Rosenow, loc. cit.

3 Quoted by Cole, New York Med. Jour., January 2 and 9, 1915.

288 STREPTOCOCCUS-PNEUMOCOCCUS GROUP

recovery as carriers, precisely as typhoid carriers harbor typhoid
bacilli after recovery from typhoid fever.

Animal. Mice are the most susceptible of laboratory animals
to infection with the pneumococcus. Small amounts of pneumonic
sputum, exudate or pus injected subcutaneously lead to a rapidly fatal
septicemia. Encapsulated pneumococci are found in the blood and
visceral organs, particularly the spleen, which is enlarged, and the
peritoneal fluid. Rabbits are somewhat less susceptible and the
results of inoculation of pneumococcic exudates or cultures depend
upon the virulence of the organisms, the size of the dose, and the
method of inoculation. 1 The intravenous or subcutaneous inoculation
of virulent cultures leads to a fatal septicemia, death occurring within
five days as a rule. The less virulent organisms, which do not kill
the animal within a few days after inoculation, frequently cause
localized abscess formation with a fibrinous exudate. The nature
and extent of the lesions induced depend largely upon the time which
elapses between inoculation and the death of the animal. In general
it may be stated that localized lesions appear when less virulent
organisms are injected. Intravenous injections are more effective
than subcutaneous inoculations of the same amount of organisms.
Guinea-pigs are relatively non-susceptible to pneumococcus infection.

Many attempts have been made to reproduce the typical patho-
logical lesions of lobar pneumonia in experimental animals. Wads-
worth 2 succeeded in reproducing typical lobar pneumonia in rabbits
by first partially immunizing them to the organism in order to localize
the lesions in the lungs. Lamar and Meltzer, 3 and Wollstein and
Meltzer 4 produced lobar pneumonia in dogs by the method of tracheal
insufflation devised by Meltzer; and Winternitz and Hirschf elder 5
have been equally successful in producing lobar pneumonia in rabbits.
The method consists essentially in forcing suspensions of pneumo-
cocci deep into the terminal bronchioles and their alveoli. Cole 6
has shown that the strain of organism influences the results ; organisms
of slight virulence give negative results, and organisms possessing too
great virulence cause a generalized septicemia with congestion and
edema of the lungs as the only local pulmonary manifestations.

1 Kruse and Pansini, Zeit. f. Hyg., 1892, xi, 279 et scq.

2 Am. Jour. Med. Sc., 1904, cxxvii, 851.

3 Jour. Exp. Med., 1912, xv, 133.

4 Ibid., 1913, xvii,,353, 424.

5 Ibid., 1912, xvii, 657.

6 Arch. Int. Med., 1914, xiv, 56.

THE PNEUMOCOCCUS

289

Types of Pneumococci. Kruse and Pansini 1 as early as 1891 called
attention to the differences, both cultural, morphological and in
virulence, which they observed in studying eighty-four strains of
pneumococci isolated from many cases of pneumonia. They believe
that there was no sharp line of demarcation between the pneumo-
coccus and Streptococcus pyogenes, because their various strains
included all variants between the two types of organisms. Recently
Rosenow 2 has reported the transmutation of typical pneumococci to
Streptococcus pyogenes by a series of animal passages and cultural
manipulations. Cole 3 has been unable to confirm this observation
in any one of several hundred strains, but it should be stated that
he has not employed Rosenow's procedure in detail.

Much light has been shed upon the apparent variability of strains
of pneumococci by the observations of Neufeld and Handel, 4 and
Dochez, 3 and Dochez and Gillespie. 6 These observers have shown
by serological reactions that pneumococci may be divided into four
groups or types, each of which fails to agglutinate with sera other
than the homologous serum. These groups have been tentatively
designated I to IV inclusive. Groups I and II are typical virulent
pneumococci. Group III comprises the organism formerly known
as Streptococcus mucosus, now called Pneumococcus mucosus; and
Group IV includes relatively avirulent strains which are commonly
found in the mouths of healthy persons. Group IV is somewhat
more heterogenous, judging from agglutination reactions, than Groups
I to III. Group III contains the most virulent organisms. A study
of the distribution of the various types in seventy-two cases of pneu-
monia illustrates this point. 7

Infection type.
1

2 ..

Total

No. cases.
34
13
10
15

No. deaths.
8
8
6

Per cent.
24
61
60

It is possible that " mixed infections" will be found when more
cases are carefully studied. The same general types have since been
reported in Europe and in Philadelphia. 8

1 Loc. cit. 2 Jour. Am. Med. Assn., 1913, Ixi, 2007. 3 Loc. cit.

4 Zeit, f. Immunitatsforsch., 1909, iii, 159; Berl. klin. Woch., 1912, xlix, 680.

5 Jour. Exp. Med., 1912, xvi, 680. 6 Jour. Am. Med. Assn., 1913, Ixi, 727.
* Cole, Arch. Int. Med., 1914, xiv, 33.

8 Cole, New York Med. Jour., January 2 and 9, 1915.
19

290 STREPTOCOCCUS-PNEUMOCOCCUS GROUP

Immunity and Immunization. Relatively little is known of the
nature and extent of immunity following recovery from an attack
of pneumonia. One attack appears to predispose somewhat to a sub-
sequent attack, which was explained formerly on the basis that little
or no immunity was conferred on the patient. The extensive work
of Cole and his associates suggests that a second attack of the disease
may be caused by a different type of pneumococcus; their experiments
indicate that antibodies specific for one type are not protective against
infection with the other types.

The serum of convalescent pneumonia patients exhibits relatively
feeble bactericidal activity, even upon the homologous strain of the
pneumococcus, and the mechanism which leads to recovery is not
definitely known. Neufeld 1 and others have advanced the hypothesis,
based upon careful observation, that the crisis in pneumonia, which
usually marks the end of the prominent clinical symptoms, is asso-
ciated with a somewhat abrupt increase in the amount of specific
opsonin of the blood an increase in bacteriotropins in Neufeld's
terminology. This theory assumes that leukocytes play a prominent
part in the healing process, and that phagocytic activity becomes
efficient at or about the time of the crisis.

Neufeld and Handel, 2 and Cole and his associates 3 have produced a
serum which protects susceptible animals, as mice, against many
times the fatal dose of the homologous strain of organism by injecting
gradually increasing doses of very virulent pneumococci into horses.
Cole has used these sera clinically in the treatment of pneumonia
with promising results in infections caused by Types I and II of the
pneumococcus. The serum appears to destroy or greatly reduce the
number of pneumococci in the blood, and to be of material benefit in
reducing the severity of the infection. At present a satisfactory
serum for infection with Type III, Pneumococcus mucosus, has not
been prepared. Cole specifically directs attention to the necessity
of identifying the type of infecting organism (by agglutination reac-
tions) before administering the serum. It is imperative that the
homologous serum be used.

Bacteriological Diagnosis. Pneumococci are found in* the healthy
throats of a very considerable percentage of adults, consequently the
identification of pneumococci in the sputum is of little clinical signifi-

1 Zeit. f. Immunitatsforsch., 1909, iii, 159.

2 Arb. a. d. kais. Gesamte, 1910, xxxiv, 169.

3 Jour. Am. Med. Assn., 1913, Ixi, 663; New York Med. Jour., January 2 and 9,
1915.

THE PNEUMOCOCCUS 291

cance unless the type of the organism is determined. Dochez and
Avery 1 have found that the common mouth pneumococcus is usually
the avirulent type (Type IV); convalescents from pneumonia usually
exhibit the virulent types, I to III, as a rule. These types can be
identified by agglutination reactions with the specific sera prepared by
Cole.

Pneumococci isolated from pleural and pericardial exudates, middle-
ear infection, empyema and pneumococcic cerebrospinal meningitis
can be identified morphologically by their lanceolate shape and Gram-
positiveness ; the type of organism, however, must be determined by
serological reactions.

They are best obtained in pure culture, if they are mixed with other
bacteria, upon blood agar plates. A green halo surrounds the typical
pneumococcus colony.

The prophylaxis is the same as for any acute respiratory disease.

1 Quoted by Cole, loc. cit.

CHAPTER XIV.
THE MENJNGOCOCCUS GONOCOCCUS GROUP.

THE MENINGOCOCCUS GROUP. THE GONOCOCCUS GROUP.

Micrococcus Meningitidis. Micrococcus Gonorrhrae.

Parameningococcus. Micrococcus Catarrhalis.

THE MENINGOCOCCUS GROUP.

Micrococcus Meningitidis. Synonyms. Diplococcus intracellularis
meningitidis; Diplococcus weichselbaumii, Meningococcus.

Historical. Micrococcus meningitidis was isolated in pure culture
by Weichselbaum 1 from purulent cerebrospinal fluids of several typical
cases of cerebrospinal meningitis. The injection of pure cultures of
the organisms directly into the meninges of dogs resulted in well-
marked meningeal inflammation and encephalitis. Other organisms,
pneumococci, streptococci, Bacillus influenzse, for example, may incite
inflammations of the cerebrospinal membranes, but these bacteria
do not ordinarily cause epidemics of the disease. The meningococcus
frequently causes wide-spread infection, and, unlike the organisms
just mentioned (except the pneumococcus occasionally) the typical
lesions are primarily of the cerebrospinal axis.

Morphology. Meningococci obtained directly from the cerebrospinal
fluid or from meningeal exudates occur characteristically in pairs with
their apposed sides flattened and somewhat elongated. They measure
about one micron in diameter, although the size varies even in the
same culture. The individuals are fairly uniform in size and shape
in very young, fresh cultures, but in older cultures considerable
variations in size are met with. Examined directly in inflammatory
exudates from the spinal fluid or meninges during the acute stages of
the disease, the organisms occur typically and characteristically as
intra- and extracellular diplococci and tetrads. They are found in
polymorphonuclear leukocytes, but never in lymphocytes or other
body cells. 2 They are intracellular but never intranuclear, according
to Councilman, Mallory and Wright.

1 Fortschr. d. Med., 1887, Nos. 18 and 19.

2 Councilman, Mallory and Wright, Epidemic Cerebrospinal Meningitis. A Report
to the Mass. St. Bd. of Health, 1898, p. 75.

THE MENINGOCOCCUS GROUP 293

The organisms are non-motile and possess no flagella. No spores
are forjned and no capsules have been demonstrated. (Jaeger 1
believed that the organisms produced capsules, but his observations
are unconfirmed.) Ordinary anilin dyes stain meningococci, but quite
irregularly. Occasionally one element of a pair stains intensely
while its fellow stains faintly or .not at all. Relatively large oval
or round forms are frequently seen in cultures and in purulent exudates
as well, which exhibit a brightly staining point in the centre of the
organism; the remainder of the cell is scarcely colored. 2 Carbol-
thionin is one of the best stains for the organism. The meningococcus

FIG. 39. Meningococci in pus. X 1000.

is Gram-negative. Meningococci obtained from purulent exudates
or from cultures on artificial media can not be definitely differentiated
from gonococci or even from Micrococcus catarrhalis by any known
staining methods. The source of the material should be known before
even a tentative morphological diagnosis is attempted.

Isolation and Culture. The meningococcus grows feebly or not at
all upon ordinary artificial media. Growths may be obtained upon
agar containing animal protein, as defibrinated blood or ascitic fluid,
or upon Loffler's blood serum by smearing cerebrospinal fluid (drawn
with aseptic precautions by lumbar puncture) in liberal amounts
upon the surface of these media. 3 The addition of 1 per cent, of
dextrose to the media favors development of the cocci. If the fluid
obtained is not turbid, centrifugalization should be resorted to and

1 Zeit. f. Hyg., 1895, xix, 358.

2 Councilman, Mallory and Wright, loc. cit., p. 74.

3 For technic of lumbar puncture, see page 226.

294 THE MENINGOCOCCUSGONOCOCCUS GROUP

the sediment distributed as densely as possible in the manner indi-
cated. A few small, transparent, round colonies are usually obtained
when relatively large amounts of material are inoculated. The first
growth upon artificial media is difficult to obtain; secondary trans-
fers, if made within three days from initial cultivations, are usually
successful and development is somewhat more vigorous. It should
be emphasized that relatively large amounts of cocci must be inocu-
lated to insure growth in artificial media. 1 Little or no growth occurs
in plain broth; the addition of calcium carbonate 2 to dextrose broth
makes a favorable medium for the development of the organism.
Ascitic and serum broths are suitable media for the meningococcus.
A coherent sediment gradually accumulates in these media and a
delicate pellicle usually forms on the surface after a few days. Secon-
dary transfers in milk usually grow, but there is little or no detectable
change in the physical properties of the medium.

The meningococcus is essentially an aerobic organism, at least in
its development outside the human body. The optimum tempera-
ture of growth is 37 C., and growth ceases when the temperature
exceeds 42 C. or falls below 25 C. The organism is soon killed by
low temperatures. Stock cultures can not be maintained at the
temperature of the ice-box; they should be kept at temperatures
between 32 and 38 C. Frequent transfers (every two or three days)
must be made to maintain the viability of the organism ; exceptionally
strains are met with which become acclimatized to the conditions
obtaining in artificial media to such a degree that transfers made at
less frequent intervals suffice to maintain the viability of the culture.

The meningococcus exhibits little resistance to heat, drying or the
action of chemical agents. Five minutes' exposure to 65 C. or two
minutes' exposure at 80 C. suffices to sterilize the culture. Drying
for a few hours at 20 C. is likewise fatal to the organism. Exposure
of the organism to carbolic acid broth (1 to 800) inhibits development,
and drying in the dark for seventy- two hours is fatal; sixty hours'
exposure to drying is insufficient to kill the organisms. 3

Products of Growth. Meningococci are culturally very inert. No
proteolytic enzymes have been demonstrated; gelatin and blood
serum are not liquefied, and no coagulation or peptonization of milk
occurs. Indol, skatol, phenol or other products of similar nature are

1 The organisms, like gonococci, degenerate rapidly in artificial media. This may
explain the necessity of transferring the organisms at frequent intervals.
* Bolduan, New York Med. Jour., 1905, May 13.
3 Councilman, Mallory and Wright, loc. cit., p. 78.

THE MENINGOCOCCUS GROUP 295

not demonstrable in cultures of Jhe organism. Acid, but no gas, is
produced with considerable regularity in dextrose and maltose broths; 1
other ordinary carbohydrates are unattacked. These fermentation
reactions are of considerable value in the cultural differentiation of
meningococci from other organisms which may readily be confused
with them.

Toxins. Soluble exotoxins have never been demonstrated among
the products produced by the meningococcus; killed cultures of the
organism appear to be as fatal for ordinary experimental animals
as the living organisms. This would suggest that the toxic phenomenon
may be attributable to the liberation of endotoxins rather than to a
soluble toxin.

FIG. 40. Meningococci from^ferebrospinal fluid. X 1200. (Kolle and Hetsch.)

Pathogenesis. The meningococcus possesses but feeble pathogenic
powers for guiiffea^pigs ; all attempts to induce infection by subcuta-
neous injections, according to Councilman, Mallory and Wright, 2
were negative. Occasionally successful results were obtained from
intraperitoneal and intrapleural inoculation. A slight fibrinopurulent
exudate was found postmortem in the peritoneal or pleural cavities
in the fatal cases. Intracranial inoculations were uniformly negative.
One successful infection, of a goat by spinal canal inoculation was
obtained by these observers; the animal died within twenty-four
hours, and autopsy revealed intense congestion of the meninges of
the cord and brain. A small amount of purulent spinal fluid was

1 Kopetsky, Meningitis, The Laryngoscope, 1912, xxii, 797, has called attention to the
early disappearance of the reducible substance (dextrose?) normally present in the
spinal fluid in cerebrospinal meningitis. It is possible that the action of the organism
upon this substance explains the phenomenon.

2 Loc. cit., p. 76.

296 THE MENINGOCOCCUS GONOCOCCUS GROUP

obtained containing but little fibrin. Small numbers of cocci were
found within the polymorphonuclear leukocytes. Flexner 1 and Von
Lingelsheim and Leuchs 2 have reproduced the essential lesions of
cerebrospinal meningitis in monkeys by the subdural injection of
suspensions of the organisms. The organisms were recovered in pure
culture at autopsy.

The evidence of the etiological relation of the meningococcus to
cerebrospinal meningitis in man is essentially the common, almost
constant demonstration of meningococci in the cerebrospinal fluid
and exudates antemortem, and from the tissues of the brain and cord
postmortem. It must be remembered that other organisms can
produce essentially the same lesions, however. The nature and extent
of the lesions observed in fatal cases varies somewhat with the time
which elapses between the onset of symptoms and death. The rapidly
fatal cases frequently exhibit intense congestion of the membranes
of the cord and brain; usually a fibrinopurulent exudate forms,
more extensive as a rule at the base of the brain but readily demon-
strable in the spinal fluid obtained by lumbar puncture. According
to Westenhoffer, 3 there is commonly a swelling of the tonsils and
pharynx in the early stages of the disease; middle ear involvement
is comparatively frequent. It is probable that the organism passes
from the nose and nasopharynx through the lymphatics to the base
of the brain. The accessory sinuses of the nasal cavity appear to be
inflamed in a majority of cases, particularly during the initial clinical
period of the disease. There is a thickening of the meninges in those
cases which run a more chronic course, frequently with considerable
distention of the ventricles. Intracranial pressure is usually a promi-
nent symptom. The organism has been isolated from the blood by
Jacobitz, 4 Dieudonne, 5 Elser, 6 Elser and Huntoon, 7 the latter in 25
per cent, of their large series of cases.

Immunity and Immunization. Little is definitely known of man's
immunity to the meningococcus. One of the surprising results of the
intensive study of the epidemic disease is the occurrence of the organ-
ism in the nasopharynx in a very considerable number of apparently
healthy individuals, chiefly among those in actual contact with
patients, less commonly among those not intimately in association
with cases but in regions where the disease is epidemic, and rarely

1 Cent. f. Bakt., 1907, xliii, 99. 2 Klin. Jarhb., 1906, xv, 489.

3 Berl. med. Gesellsch., 1905, May 17; abstr. Cent. f. Bakt., Ref., 1905, xxxvi, 754.

4 Munchen. med. Wchnschr., 1905. * Cent. f. Bakt., Orig., 1906, xli, 420.
6 Jour. Med. Research, 1906, xiv, 89. 7 Jour. Med. Research, 1909, xx, 371.

THE MENINGOCOCCUS GROUP 297

among individuals residing in areas where but few sporadic cases
have been reported. The percentage of positive examinations varies
considerably. Dieudonne 1 found about 12 per cent, of normal soldiers
in a garrison at Munich, where an outbreak occurred, gave positive
cultures from the nasopharynx. Bruns and Hohn 2 found 465 carriers
among 3154 healthy individuals in a community where the disease
was epidemic. They also found the percentage of carriers was great-
est when the epidemic was at its height. Usually these carriers are
temporary carriers; smaller numbers become permanent carriers or
periodic carriers. 3

Serum Therapy. Many attempts have been made to prepare sera
for the treatment of epidemic cerebrospinal meningitis, and two
preparations have stood the test of actual practice, Kolle and Wasser-
mann's 4 serum and the serum prepared by Flexner and Jobling. The
method of immunization adopted by Flexner and Jobling appears
from available data to be essentially that of Wassermann. It is as
follows: horses are injected subcutaneously, first with dead cultures
of meningococci, secondly with live cultures, and finally with auto-
ly sates of cultures. The latter are prepared by suspending virulent
meningococci in sterile water for two days at 37 C. and injecting
the supernatant fluid. The serum thus produced appears to combine
phagocytic properties, increasing the destruction of the organisms by
leukocytes; bacteriolytic properties, killing and dissolving the cocci,
and possibly some antitoxic properties as well. It is essential, as
Flexner has pointed out, to inject the serum directly into the spinal
canal. This is accomplished by lumbar puncture. The turbid spinal
fluid is allowed to escape through the needle with which the puncture
is made until symptoms of intercranial pressure are reduced. An
additional amount of fluid is then withdrawn to make way for the
serum which is injected directly, 15 to 20 c.c. for young children and
20 to 40 c.c. for adults. The treatment is repeated from two to several
times, until the spinal fluid is clear and has a normal appearance and
cellular content. The serum must be used early in the disease to
obtain the best results. Flexner and Jobling 6 have analyzed 328
cases with the following mortality :

Per cent.

Injection during first to third day of disease mortality 19.9

Injection during fourth to seventh day of disease .... mortality 22.0
Injection after seventh day of disease mortality 36.4

1 Loc. cit. 2 Klin. Jahrb., 1908, xviii, 285.

3 Mayer and Waldmann, Munch, med. Wchnschr., 1910, 475. Mayer, Waldmann,
Furst and Gruber, Munchen. med. Wchnschr., 1910, 1584.

< Deut. med. Wchnschr., 1906. 5 j our . Am. Med. Assn., 1908, li, No. 4.

298 THE MENINGOCOCCUSGONOCOCCUS GROUP

Similar results have been obtained in Germany with Wassermann's
serum. 1 Later observations by Flexner 2 confirm these results. The
mortality has been reduced from about 70 per cent, to about 20 to
25 per cent.

Bacteriological Diagnosis. (a) Morphological. The demonstration of
Gram-negative, biscuit-shaped diplococci in purulent spinal fluid from
patients exhibiting the characteristic clinical symptoms is sufficient
to establish a diagnosis of the meningococcus. It is to be remem-
bered that the spinal fluid is clear for the first twenty-four hours of
the disease, and usually clear after the tenth day to the fourteenth
day even in untreated cases. Centrifugalization in sterile tubes must
be resorted to in such cases; the sediment is examined as above.
Smears from the nasopharynx, from middle-ear infections, and from
suspected carriers can not be definitely diagnosed upon morphological
characters alone. Cultural characteristics must be studied as well.

Cultural Characters. Spinal fluid removed aseptically (and cen-
trifugalized if the fluid is clear) and material from the nasopharynx,
nasal cavity, or accessory nasal sinuses 3 is spread upon Loffler's
blood serum and incubated at 37 C. After twenty-four to forty-eight
hours' incubation, small, clear, round colonies develop in the majority
of cases in which meningococci are present. These should be trans-
ferred to ascitic broth (preferably containing 1 per cent, of dextrose
and a small piece of calcium carbonate) and examined after twenty-
four hours' incubation at body temperature. If growth occurs, inocu-
lation should be made in ascitic fluid dextrose and ascitic fluid maltose
broths to determine if acid is produced. Several diplococci have been
found which resemble the meningococcus microscopically but which
differ from it in their fermentation reactions. A negative result does
not exclude the possibility of an infection with the meningococcus;
negative cultures occur quite frequently. Von Lingelsheim 4 and Elser
and Huntoon 5 have studied these organisms carefully and give the
following differential table :

1 Wassermann, Deut. med. Wchnschr., 1907, 1585; Wassermann and Leuchs, Klin.
Jahrb., 1908, xix, Heft 3.

2 Jour. Am. Med. Assn., 1909, liii, 1443.

3 Material for examination from the nasopharynx is best obtained upon sterile swabs;
the infected swab should be immediately rubbed over the surface of a series of blood
serum tubes or ascitic agar plates. This method is particularly adapted for the exami-
nation of suspected carriers.

4 Klin. Jahrb., 1906, xv, Heft 2.

6 Jour. Med. Research, 1909, xx, 377.

THE MENINGOCOCCUS GROUP 299

O Q 3 O S ^ %

Meningococcus + +

Pseudomeningococcus + +

Gonococcus +

Micrococcus catarrhalis

Diplococcus crassus 3 + + + + + + +

Diplococcus flavus + + +

Micrococcus pharyngis siccus . + + +

Pigmented coccus I + + + +

II. .-""'." ... : .'', . + + +

" III. . . . . . + +

Micrococcus cinereus 4

It will be seen that the meningococcus produces acid in dextrose
and maltose. A differentiation between the gonococcus, Micrococcus
catarrhalis and the meningococcus can frequently be made by their
growths upon cultural media. The gonococcus grows poorly or not at
all upon blood serum (Loffler's), the meningococcus grows with mod-
erate rapidity upon it, and Micrococcus catarrhalis grows even upon
plain agar.

The final diagnosis of the meningococcus depends upon its agglu-
tination with specific sera. Positive agglutination will take place in
dilutions of 1 to 500, even in 1 to 2000. Kutscher 5 has isolated strains
of the organism which failed to agglutinate (macroscopic method) at
37 C., but agglutinated typically at 55 C. This should be tried in
doubtful cases.

Serological Diagnosis. Bettencourt 6 and Franca, 7 von Lingelsheim,
Elser and Huntoon 8 and others have shown that the sera of convales-
cent cases of cerebrospinal meningitis very frequently exhibit specific
agglutinins for the meningococcus. Of 593 tests, von Lingelsheim
found 24.1 per cent, positive during the first five days of the disease,
56.7 per cent, positive from the sixth to the tenth day. Normal sera
did not agglutinate with the organism in dilutions greater than 1 to
25; the sera of patients agglutinated in dilutions as high as 1 to 200.
Elser and Huntoon have obtained agglutination in dilutions as high
as 1 to 400.

The method of complement-fixation has not been satisfactory in
the diagnosis of cerebrospinal meningitis. 9

1 + = Gram-positive 2 + = acid produced

- = Gram-negative - = no acid produced.

3 Jaeger's meningococcus. 4 Micrococcus catarrhalis?

5 Kolle and Wassermann, Handb. d. path. Mikroorganismen, I. Erganzbd., 1907, 518.

6 Zeit. f. Hyg., 1904, xlvi, 463. 7 Klin. Jahrb., 1906, xv, Heft 2.

8 Loc. cit.

9 Von Lingelsheim XIV Cong, for Demog. and Hyg., Berlin, September, 1907.

300 THE MENINGOCOCCUSGONOCOCCUS GROUP

Dissemination and Prophylaxis. The disease is usually more fre-
quent in children and young adults, usually in the winter and spring
months. Frequently a nasal inflammation is prevalent before the
disease begins to spread. The disease spreads by contact; as the organ-
isms die out rapidly away from the human body. Many cases do not
progress beyond the stage of nasal pharyngitis and sore throat, and
it is probable that these cases are potentially carriers. According
to Bruns and Hohn, 1 there may be from ten to twenty times as many
carriers as cases. The disease is very likely to occur in barracks and
boarding houses. Many people may be exposed to infection but
comparatively few acquire the disease, suggesting a rather high
natural resistance to the organism. The meningococcus may remain
for months in the nasal passages of carriers, although ordinarily
they remain less than a week.

Ward attendants should be segregated and quarantined, and nasal
sprays used on the patients and attendants. It is quite probable that
infected handkerchiefs or inhalation of infectious droplets are impor-
tant in spreading the organism. It should be treated like any other
acute infectious disease of the respiratory tract.

Parameningococcus. In a critical discussion of the treatment of
epidemic cerebrospinal meningitis with a specific antimeningococcus
serum, Flexner 2 had directed attention to a relatively small group of
cases which either failed to respond favorably to the serum, or reacted
for a short time and later failed to improve. The spinal fluid of these
cases contained organisms microscopically indistinguishable from
typical meningococci. It was assumed tentatively that there might
be two types of meningococcus, one of which was naturally "serum-
fast," the other acquired " serum-fastness" during the course of the
treatment with the serum. Dopter 3 has described an organism the
parameningococcus apparently identical with the typical meningo-
coccus in its morphological and cultural characteristics, but specifi-
cally different in its serological reactions. The parameningococcus,
like the meningococcus, has been isolated from the nasal and oral
cavities of man, and, in a few cases, from the blood stream and the
meninges as well. The clinical manifestations incited by the para-
meningococcus are indistinguishable from those of epidemic cerebro-
spinal meningitis, but they fail to respond favorably to the adminis-
tration of meningococcus serum. Dopter 4 has prepared a specific

1 Loc. cit. 2 Jour. Exp. Med., 1913, xvii, 553.

3 Compt. rend., Soc. de Biol., 1909, Ixvii, 74. 4 Semaine m6d., 1912, xxxii, 298.

THE GONOCOCCUS GROUP 301

parameningococcic serum which is stated to have effected rapid
improvement in the few cases of parameningococcus infection in which
it was tried. These cases failed to respond to injections of meningo-
coccus. serum.

Wollstein 1 has made careful comparative studies of the morpholog-
ical, cultural and serological reactions exhibited by a series of meningo-
cocci and parameningococci; her conclusions, which follow, summarize
the available information of the relationship between these two
organisms :

'"The parameningococci of Dopter are culturally indistinguishable
from true or normal meningococci, but serologically they exhibit
differences as regards agglutination, opsonization, and complement
deviation.

"Because of the variations and irregularities of serum reactions
existing among otherwise normal strains of meningococci, it does not
seem either possible or desirable to separate the parameningococci
into a strictly definite class. It appears desirable to consider them
as constituting a special strain among meningococci, not, however,
wholly consistent in itself.

The distinctions in serum reactions between normal and paramen-
ingococci are supported by the differences in protective effects of the
monovalent immune sera upon infection in guinea-pigs and monkeys.

" It is therefore concluded that it is highly desirable to employ strains
of pararneningococcus in the preparation of the usual polyvalent
antimeningococcus serum. It remains to be determined where it is
better to employ the parameningococci along with normal meningo-
cocci in immunizing horses, or to employ normal and para strains
separately 'in the immunization process and to combine afterward, in
certain proportions, the sera from the two kinds of immunized horses."

THE GONOCOCCUS GROUP.

Micrococcus Gonorrheas. Synonyms. Diplococcus gonorrhese, gon-
ococcus.

Historical. The gonococcus was first observed by Neisser 2 in puru-
lent urethral and vaginal discharges. Some years later Bumm 3 grew
the organism in pure culture upon coagulated human blood serum and
reproduced acute gonorrhea in men by urethral injections.

1 Jour. Exp. Med., 1914, xx, 201. 2 Cent. f. d. med. Wise., 1879, No. 28.

3 Die Mickroorganismen des gonorrhoischen Schleimhauterkrankungen Gonococcus,
Neisser, Wiesbaden, 1885, No. 28.

302

THE MENINGOCOCCUSGONOCOCCUS GROUP

Morphology. The gonococcus occurs typically as a diplococcus, the
proximated surfaces of pairs of cocci being flattened and elongated;
they resemble coffee beans in shape. The longer diameter measures
about 1.5 microns, the shorter diameter about 0.8 micron. The
polymorphonuclear leukocytes of pus from cases of acute gonorrhea
usually contain from one to several pairs of gonococci which are within
the cytoplasm of the leukocyte but rarely or never within the nuclei.
The organisms are also found within desquamated epithelial cells and
occur free in pus as well. Gonococci are less numerous in the subacute
and chronic stages of the disease, and they occur chiefly extracellu-

FIG. 41. Gonococcus smear of pus from acute case. Methylene blue stain. (Warden.)

larly, with occasional pairs or clusters of gonococci in epithelial cells,
less commonly in polymorphonuclear leukocytes. The organisms
undergo degeneration rapidly, and even in pus from the more acute
cases many large faintly staining cocci are found in association with
those which are more typical in morphology and staining. In the
chronic stage of the disease degenerated forms are very common.

The gonococcus is non-motile, and possesses no flagella; it forms no
spores and capsules have not been detected. It stains with ordinary
anilin dyes, but with some difficulty. It is Gram-negative.

Isolation and Culture. The organism does not grow upon ordinary
media; for the first growths outside the human body media con-
taining uncoagulated protein, preferably that of human origin, is

THE GONOCOCCUS GROUP 303

required. Agar 1 smeared with sterile defibrinated blood, 2 or agar mixed
with hydrocele or ascitic fluid (one part fluid, two parts agar) furnishes
a satisfactory nutrient substrate. Pus from acute cases (after pre-
liminary cleaning and sterilization of the external genitalia) spread
upon one of the media described above, should exhibit colonies after
twenty-four hours' incubation at 37 C. The colonies are minute,
clear and colorless; they resemble small dewdrops and exhibit a ten-
dency to coalesce. Organisms stained from these colonies remain
typical in morphology only for one or two days. Very soon degen-
eration (autolysis) commences, and in a very short time the organisms
are dead 3 and partially dissolved. Secondary growths may be obtained
from colonies, provided the inoculations are made within twenty-four
to forty-eight hours from the time of plating. Ascitic broth is an
especially favorable medium for this purpose.

The gonococcus is markedly aerobic; little or no growth occurs in
media from which oxygen is excluded. The temperature limits are
very restricted; growth ceases below 30 C. and above 40 to 42 C.
The optimum temperature is 37 C. The organism is extremely
sensitive to desiccation, and cultures die spontaneously within six
to eight days. Repeated transfers of the cocci at intervals of two to
three days will prolong the life of the culture almost indefinitely, pro-
vided they are maintained at 37 C. The organisms are very readily
killed (outside the body) by the^usual disinfectants. Gonococci in
the urethra can not be killed readily by chemical disinfectants; the
organisms penetrate rather deeply into the walls and the disinfectant
can not reach them in sufficient concentration to be effective. This
is particularly true of the sub acute and chronic stages of the disease.

Products of Growth. No enzymes have been detected in cultures of
gonococci. Culturally the organism is inert; no development occurs
in ordinary media. Acid is produced in dextrose-ascitic broth, but no
other sugars are fermented. (See page 299 for comparison of cultural
characters of gonococcus and similar Gram-negative diplococci.)

Toxins. No soluble (exo-) toxin has been demonstrated in cultures
of gonococci.

Finger, Ghon and Schlagenhaufer, 4 Nicolaysen, 5 Wassermann 6 and
de Christmas 7 have shown that the cell substance itself is toxic.

1 Glycerin agar is better than ordinary agar for this purpose.

2 The blood agar should be heated to 56 C. for thirty minutes to destroy its bacteri-
cidal properties before use.

3 Warden, Jour. Infec. Dis., 1913, xii, 93.

4 Arch. f. Derm. u. Syph., 1894, xxviii, Nos. 1 and 2; Cent. f. Bakt., 1894, xvi, 350.

5 Cent. f. Bakt., 1897, xxii, 305. 6 Zeit. f. Hyg., 1898, xxvii, 307.
7 Ann. Inst. Past., 1900, 349.

304 THE MENINGOCOCCUS GONOCOCCUS GROUP

De Christmas has shown that the poisonous substance (endotoxin)
diffuses readily into the culture medium, probably because of the
rapid autolysis which is a noteworthy feature of the organism. The
endotoxin is fairly resistant to heat; a brief exposure to 120 C. fails
to entirely destroy its potency.

Pathogenesis. Experimental. Bumm 1 and Finger, Ghon and
Schlagenhaufer 2 have reproduced typical urethritis in man with
pure cultures of the gonococcus. The latter successfully infected the
urethras of six healthy men with the organism (serum agar culture).
The incubation period was from two to three days, and the clinical
picture was typical in each instance. The organism was recovered
in pure culture from each patient.

Animal. Laboratory animals are not susceptible to urethral
infection with the gonococcus. Intraperitoneal injections of cultures
into 'white mice produce a purulent peritonitis, but there is little
evidence that the organisms multiply there. Acute joint inflammations
with purulent exudation follows the inoculation of the cocci into the
joints of rabbits, and purulent conjunctivitis can be produced in
young rabbits by rubbing gonococci on the conjunctiva. There is no
evidence that the organisms multiply in these sites ; the reverse appears
to be the case for the cocci disappear rather rapidly. The endotoxins
are responsible for the local reactions.

Human. Man is very susceptible to infection with the gonococcus.
The usual portals of entry are the mucous membranes of the urethra,
vagina, and the conjunctiva. The urethral mucous membrane is
particularly susceptible and it is commonly the primary site of inva-
sion. The uterine mucosa and adnexa are also readily infected in
adults; in young children the cervix is closed and infection of the
uterus by continuity of growth from the vagina is rare in them, but
vulvovaginitis is common, especially in hospital wards where infec-
tion is readily transmitted by thermometers, hands of ward attendants,
and by direct contact.

The initial development of the organisms is upon the surface of the
mucosa, then they penetrate to the deeper layers, infecting the pros-
tate, and by continuity the epididymis in the male. Infection may
spread from the vagina to the uterus in the female, then by continuity
of growth to the Fallopian tubes, the ovaries, and the peritoneum,
causing endometritis, salpingitis, oophoritis, and peritonitis. Sterility
is usually the result. Cystitis and arthritis are not uncommon sequelse

1 Loc. cit. 2 Loc. cit

THE GONOCOCCUS GROUP 305

of infection with the gonococcus, and the mucosa of the rectum is
occasionally involved. Serous surfaces are rarely involved. Occa-
sionally a generalized invasion takes place frequently resulting in
septicemia with endocarditis. Ophthalmia neonatorum is a particu-
larly common infection of the newborn of infected mothers. The
conjunctive become contaminated with gonococci as the child passes
through the vagina. A large percentage of the blind have lost their
eyesight in this manner at birth. The instillation of silver prepara-
tions, required by law in many States, has greatly reduced this form
of infection.

Immunity. Man exhibits little or no resistance to infection with
the gonococcus and the mucous membranes may actually be more
susceptible to reinfection than they were originally. 1 In the chronic
cases, where the organisms lie dormant for months, even years, the
tissues appear to be somewhat less suited for growth of the organisms,
but the patient can infect others even at this stage of the disease.
Various attempts to prepare sera for curative purposes have not been
generally successful, although Rogers 2 has reported cures in cases of
gonorrheal rheumatism and chronic gonorrheal urethritis by the
injection of the serum prepared by Torrey. 3

Vaccines have been used with variable success. The injection of an
autogenous vaccine, containing from five to ten million gonococci
from a twenty-four-hour ascitic fluid agar culture, appears to give the
best clinical results. Probably the extremely rapid autolysis of the
gonococci plays a prominent part in the ineffectual attempts to induce
improvement by the use of vaccines. 4

Bacteriological Diagnosis. (a) Microscopical. Pus from the urethra
of acute cases of gonorrhea should be dropped upon a cover glass or
slide and spread by gently pressing a second cover glass or slide
upon the first, then sliding them apart. By so doing the organisms
remain in the polymorphonuclear leukocytes and epithelial cells, a
very important diagnostic point. A Gram stain and a methylene-
blue stain should be made. The former reveals intracellular and
intercellular bean-shaped diplococci which are Gram-negative. Occa-
sionally leukocytes contain as many as twenty pairs of the cocci.
Dilute methylene blue 1 to 10 (Loffler's) usually stains gonococci
intensely; the remainder of the cellular elements are faintly colored.
The morphology of the organisms is clearly shown by this procedure.

1 It is uncommon, however, to find auto eye infections from venereal lesions; even
in cases of gonorrheal vulvovaginitis the eyes are rarely infected with the gonococcus.

2 Am. Jour. Med. Sc., 1906, xlvi, No. 4.

a Ibid. 4 Lespinasse; Illinois Med. Jour., April, 1912.

306 THE MENINGOCOCCUSGONOCOCCUS GROUP

In chronic cases the discharge is scanty and it is better to receive
the morning urine in a sedimentation glass containing a crystal or two
of thymol. After a short time threads of mucus separate out; these
should be removed with a capillary pipette and examined as above.
The pus from old cases of gonorrhea frequently contains but few gono-
cocci, which are difficult to find. It has been found that the local
injection of silver nitrate (properly diluted) will usually cause an
elimination of pus which frequently contains the organisms in some-
what larger numbers. Drinking beer is said to produce the same
result. Vaginal smears may be obtained from swabs which are intro-
duced into the vagina, or by means of long pipettes with rounded ends,
containing a few drops of 1 to 1000 mercuric chloride, which are
expressed and drawn up into the pipette several times deep in the
vagina. The material thus removed is stained in the usual manner.
Smears from the conjunctiva should be diagnosed very conservatively;
Micrococcus catarrhalis and other Gram- negative diplococci which
may occur within polymorphonuclear leukocytes are occasionally
associated with an inflamed conjunctiva. The clinical picture should
be considered in making the final diagnosis in such cases, and whenever
possible cultures should be made to confirm the results.

(b) Cultural. Cultures of the gonococcus are best obtained early
in the disease, when secondary infection with staphylococci or other
organisms has not taken place. The external genitalia should be
carefully cleaned as for a surgical operation, and pus collected on a
sterile swab which is rubbed over the surface of blood- or ascitic agar.
The isolation of gonococci from pus of the subacute and chronic stages
of the disease is extremely difficult; indeed, it is practically a matter
of chance if pure cultures are obtained at this time. Vaginal cultures
may be obtained upon sterile swabs which are inoculated in the same
manner.

The gonococcus does not grow upon ordinary media, not even
Loffler's blood serum, which distinguishes it from the meningococcus
and from other Gram-negative cocci, including Micrococcus catar-
rhalis. (For fermentation reaction of the gonococcus, see page 299).

Serological Diagnosis. Agglutinins. The diagnosis of gonorrhea
by agglutination of the gonococcus with the serum of the patient has
not been successful. 1

1 Torrey (Journ. Med. Res., 1908, xx, 771) has isolated ten strains of gonococcus
identical morphologically and culturally, but distinct serologically. This may explain
in part the irregularity of the reaction of agglutination provided but a single strain of
organism is used.

THE GONOCOCCUS GROUP 307

Complement-fixation Reaction. Diagnosis of gonococcus infection
by the method of complement fixation has been shown to be of con-
siderable value, particularly in the more chronic cases, provided an
homologous strain of the organism is used for the antigen. A mixed
antigen composed of several strains is frequently employed in prac-
tice. 1 Much additional work is required, however, to determine the
limits of variability of the various strains of the organism before the
method is placed upon a thoroughly satisfactory basis for routine work.

Shattuck and Whittemore 2 have prepared concentrated polyvalent
glycerin extracts and autolysates of gonococci to test the value of the
skin reactions in gonococcus infections. The tests were made intra-
dermally and by the von Pirquet method. Their results were
in) satisfactory diagnostically.

FIG. 42. Micrococcus catarrhalis and staphylococcus.

The medicolegal aspects of gonorrheal infections make it incumbent
upon the examiner to be very cautious in diagnosing the organism.

Dissemination and Prophylaxis. The common towel has in the past
been responsible for many cases of gonorrheal ophthalmia, but laws
forbidding its use have largely removed this danger. It is certain
that ordinary care will prevent infection of the innocent with th^
organism. Ophthalmia neonatorum is prevented by the instillation
of silver salts in the manner indicated above.

Micrococcus Catarrtiklis. Historical. Micrococcus catarrhalis ap-
pears to have been described first by Seifert 3 and by Kirchner; 4 the

1 Lespinasse and Wolff, Illinois Med. Jour., January, 1913. Torrey's ten strains
should be used in preparing the gonococcus antigen.

2 Boston Med. and Surg. Jour., 1913, xlxix, 373.

3 Volkmann's Sammlung klinischer Vortrage, No. 240.
Zeit. f. Hyg., 1890, ix, 528.

308 THE MENINGOCOCCUSGONOCOCCUS GROUP

name first appears in Die Mikroorganismen (Fliigge), 3d edition, in
1896, credited to R. Pfeifl'er.

Morphology. Micrococcus catarrhalis occurs typically as a diplo-
coccus with the apposed surfaces of adjacent cocci flattened and
somewhat elongated. It measures about one micron in diameter.
Occasionally the organisms are arranged in tetrads, particularly in
young, active cultures in artificial media; in older cultures a tendency
toward short chain formation is frequently observed. Degenerated
cocci occur in older cultures. In sputum, bronchial secretions and
other material from inflammation of the upper respiratory tract, in
which Micrococcus catarrhalis is a primary or accessory factor, the
organisms occur both within and without the pus cells. In the acute
stages they are usually extracellular. 1 The organism is non-motile,
and it has no flagella. It forms neither spores nor capsules. It colors
readily with ordinary anilin .dyes, some cells more intensely than their
fellows, and it is Gram-negative.

Isolation and Culture. The organism grows with moderate vigor
upon agar; after twenty-four hours' incubation the colonies are
small, translucent and gray. After three to four days the colonies
are larger with an opaque centre, the periphery being translucent.
Old colonies tend to become somewhat brownish. Development is
more vigorous in media containing blood, blood serum, or ascitic
fluid. Hemolysis of the blood does not occur. The growth in gelatin
is slow, and usually feeble. A slight turbidity develops in broth.
Moderate development occurs in milk. Micrococcus catarrhalis
grows best at 37 C.; restricted development takes place at 16 C.;
no growth can be detected at 43 C.

Products of Growth. The organism is culturally inert. It does not
produce any demonstrable proteolytic enzymes, and it produces no
acid in any sugar. No toxic products are known. Filtrates of broth
cultures have no apparent action upon white mice. No pathogenesis
for laboratory animals has been detected.

Human Pathogenesis. Micrococcus catarrhalis has occasionally been
reported as a causative factor in catarrhal inflammations of the upper
respiratory tract, and even in atypical pneum6nia 2 and in bronchitis. 3
Ordinarily it is an opportunist found in the upper respiratory tract.

Bacteriological Diagnosis. The organism is of importance chiefly
through its striking resemblance to the meningococcus and the gono-

1 Ghon, Pfeiffer and Sederl, Zeit. f. klin. Med., 1902, xliv, 262.

2 Bernheim, Deut. med. Wchnschr., 1900. 3 Ritchie, Jour. Path, and Bact., 1900

THE GONOCOCCUS GROUP 309

coccus. It differs from these diplococci both in its relatively luxuriant
growth upon artificial media and its ability to grow at room tem-
perature. It resembles them in its intracellular disposition and in its
staining reactions.

Droplet infection and transmission by contact are possible means
of dissemination, and appropriate precautions should be taken to
prevent this.

CHAPTER XV.

MICROCOCCUS MELITENSIS.

Historical. The organism was discovered by Bruce. 1

Morphology. Micrococcus melitensis is a very small oval coccus,
occurring singly or in pairs, rarely in short chains; the individual
cells measure about 0.3 to 0.4 micron in diameter. Some observers
declare the organism to be a short bacillus, a view which is perhaps
based upon its appearance in old cultures, where various involution
forms are readily observed. The coccus form almost invariably pre-
dominates in fresh material. The organism is non-motile, possesses
no flagella and forms no capsule. Spore formation has never been
observed. It stains readily* with ordinary anilin dyes, and is Gram-
negative.

Isolation and Culture. One of the noteworthy cultural characters
of Micrococcus melitensis is its slow growth on artificial media, even
at 37 C. Suspected material, either blood, urine, milk, or material
from splenic puncture, should be spread upon the surface of slightly
acid agar and examined after three or four days' incubation for very
minute white colonies which have a darker center. The organism
grows slowly in gelatin, without producing liquefaction, and it pro-
duces a slight turbidity in broth. Milk appears to be a good medium
for its development, goats' milk being better than cows' milk for this
purpose.

The coccus is aerobic, facultatively anaerobic. The minimum tem-
perature of growth is about 8 C., the optimum 37 C., and the
maximum about 44 C. Direct sunlight kills it in a few hours; an
exposure to 55 C. is usually fatal within an hour; 1 per cent, carbolic
acid kills it in ten to fifteen minutes. 2 It resists drying in the cold
and in the dark for several weeks.

Products of Growth. ^Micrococcus melitensis is culturally inert; 3 it
produces no proteolytic enzymes and it produces neither acid nor gas
in any sugars. Milk, particularly goats' milk, becomes progressively
alkaline in reaction. No toxins have been demonstrated.

1 Practitioner, September, 1887, xxxix, 161.

2 Mohler and Eichhorn, Bureau of Animal Industry, 1911, xxviii, 125.

3 Kendall, Day and Walker, Jour. Am. Chem. Soc., 1913, xxxv, 1247.

IMMUNITY AND IMMUNIZATION

311

Pathogenesis. Animal. Apes are susceptible to Micrococcus meli-
tensis; the subcutaneous inoculation of cultures of the organism
leads to definite clinical symptoms parallel to those observed in
man. The disease usually runs a prolonged course and is often fatal.
Monkeys are somewliat less favorable subjects than apes. (Goats,
sheep, cattle, and horses are also susceptible to infection, although
the disease is rarely generalized; the presence of the virus in the urine
of the males, the milk and urine of the females of these species is the
principal indication of infection.) The incubation period is from
five to fourteen days. Eyre 1 spates that rabbits and guinea-pigs may
be infected, but not rats and mice.

FIG. 43. Micrococcus melitensis and staphylococcus. X 1000. (Kolle and Hetsch.)

Milk appears to be the chief source of infection; on the Island of
Malta, where Malta fever was first described, fully 10 per cent, of
the female goats contained the organism in their milk. Monkeys
readily contracted the disease by drinking this milk. The urine of
both male and female goats was shown to be infected as well.

Immunity and Immunization. The blood and urine of infected indi-
viduals contain the virus of the disease and specific agglutinins are
present in the blood even early in the disease. The agglutinins may
persist for years after convalescence. Dilutions of ^ to joioo
are made from the blood serum with suitable controls. A small amount
of growth from a three-day agar culture of the organism is thoroughly
emulsified in each dilution of serum and in the controls; the emulsions
are incubated at 37 C. for two hours, then placed in the ice-box for
twenty-four hours before the readings are made. A control with a

1 Kolle u. Wasserman, Handb. d. Path. Mikroorganismen, I. Erganzband.

312 MICROCOCCUS MELITENSIS

non-specific serum (g$) and the organism should be made at the
same time and incubated in the same manner, for experience has
shown that the serum of normal individuals may agglutinate Micro-
coccus melitensis in moderately high dilution. Wright has immunized
horses with repeated injections of Micrococcus melitensis. The blood
serum agglutinated the organism in high dilution; it was claimed by
'him that the serum possessed curative value, the chief phenomena
following its administration being a fall in temperature and a shorten-
ing of the course of the disease. This is still debatable.

Bacteriological Diagnosis. A. Blood, 1 urine, milk, or material from
splenic puncture is plated out as outlined above. The organisms are
agglutinated with a serum of high potency.

B. The blood of the patient should be examined in high dilution
(lUo) f r specific agglutinins.

Dissemination and Prophylaxis. The organisms leave the body
through the milk or urine. Pasteurization of the milk and disinfection
of the urine of infected animals is the best prophylaxis. It should
be remembered that the organisms can enter the body through
cutaneous wounds.

1 The organisms are not always present in the blood of patients in demonstrable
numbers; a negative culture is not conclusive.

CHAPTER XVI.

THE ALCALIGENES DYSENTERY TYPHOID PARA-
TYPHOID GROUP.

BACILLUS ALCALIGENES.

BACILLUS alcaligenes was first isolated by Petruschky 1 from the feces
of a patient presenting the clinical symptoms of typhoid fever. The
serum did not agglutinate the typhoid bacillus and no typhoid bacilli
were recovered from the blood or dejecta. Several similar cases are
now on record in which Bacillus alcaligenes has been isolated both
from the blood stream and the intestinal contents; the sera of these
cases agglutinated the specific organism in dilutions of 1 to 50
or even higher, and Bacillus typhosus was not found. Bacillus
alcaligenes occurs occasionally in acute intestinal disturbances of
young children, not infrequently in association with organisms of the
dysentery and paratyphoid groups. 2 Less commonly it is found in
the dejecta of normal children, adults 3 and in water.

Morphology. The organism both in size and shape resembles the
typhoid bacillus very closely. It is actively motile and has peritrichic
flagella. It does not form spores, and so far as is known, does not
exhibit a capsule. Ordinary anilin dyes color it readily and it fails to
retain the Gram stain.

Isolation and Cultures. The organism grows readily in ordinary
media. On agar the colonies are transparent, colorless, and round,
and after eighteen hours' incubation at 37 C. attain a diameter of
from 1 to 3 mm. The organism grows with moderate luxuriance on
gelatin, but produces no liquefaction. In broth theie is a uniform
clouding, and after a few days a delicate pellicle usually forms. Bacillus
alcaligenes grows fairly readily in milk; the reaction becomes progres-
sively alkaline. In sugars no acid or gas is developed.

The organism is aerobic, facultatively anaerobic. The minimum
temperature of growth is about 6 C., the optimum 37 C., and the

1 Cent. f. Bakt., 1896, xix, 187.

2 Kendall, Day and Bagg, Boston Med. and Surg. Jour., 1913, clxix, 741.

3 Ford, Studies from the Royal Victoria Hospital. Montreal, 1903, i, No. 5.

314

THE ALCALI&ENES DYSENTERY TYPHOI I)

maximum about 44 C. The resistance of Bacillus alcaligenes to
physical and chemical reagents is similar to that of the typhoid bacillus.
Products of Growth. Bacillus alcaligenes is characterized culturally
by its inertness. Neither acid nor gas is produced from any known
sugar. A moderate amount of proteolysis similar in degree to that
of the typhoid bacillus in sugar-free broth is characteristic of the
development of this organism in all the ordinary media. 1 Milk is
not coagulated nor peptonized, but a progressive alkalinity develops,
associated with the liberation of small amounts of ammonia. 2 No
enzymes have been detected, and no toxins have been demonstrated
in cultures of the organism.

FIG. 44. Bacillus alcaligenes; bouillon culture. X 1000.

Pathogenesis. The comparatively few cases of infection with
Bacillus alcaligenes have not been studied in sufficient detail to throw
any light upon the character of the lesions produced by the organism.
The disease resembles typhoid fever clinically, and it is possible that
in the past occasional typhoidal fevers have been incorrectly diagnosed.
Animal experimentation has been uniformly negative.

Immunity. Nothing definite is known of the immunological rela-
tions of Bacillus alcaligenes. Specific agglutinins have been demon-
strated in a few instances where infection with the organism has been
confirmed bacteriologically.

Bacteriological Diagnosis. The organism may be isolated occasion-
ally from the blood; ordinarily, however, the diagnosis is made by
the isolation of the bacilli from the feces. Upon the Endo medium the
organism grows precisely like the typhoid bacillus. It is readily dif-

1 Kendall, Day and Walker, Jour. Am. Chem. Assn., 1913, xxxv, 1216.

2 Ibid., 1914, xxxvi, 1940.

THE GROUP OF THE DYSENTERY BACILLI 315

ferentiated from the typhoid bacillus by cultural reactions, Bacillus
alcaligenes forming neither acid nor gas in dextrose, lactose, saccharose,
or mannite. It does not liquefy gelatin, and it produces a permanent
alkalinity in milk. The differential cultural reactions are shown in
the table (page 316).

Dissemination and Prophylaxis. Nothing is known of the method of
dissemination of Bacillus alcaligenes. It appears to be an organism
whose portal of entry is the gastro-intestinal tract. Carriers have
never been satisfactorily demonstrated. Prophylaxis is precisely the
same as that for other intestinal organisms.

THE GROUP OF THE DYSENTERY BACILLI.

The term dysentery as it is used in the clinical way includes at
least two entirely distinct entities: amebic dysentery, a semi-acute
or chronic infection caused by an ameba, which is usually restricted
to the tropics and subtropics; and an acute type caused by members
of the dysentery bacillus group, more frequently encountered in
temperate zones. The latter type not uncommonly assumes epidemic
proportions, but occurs sporadically as well. Japan has suffered greatly
in the past from the ravages of bacillary dysentery. Ogata and
Eldridge 1 state that 1,136,067 cases with 257,289 deaths occurred in
that country during the period between 1878 and 1899 inclusive.
The mortality, which varied markedly from year to year, averaged
22.6 per cent, of all cases. The disease appears to be rare in England,
but it has been reported in Germany. 2 The Atlantic seacoast cities
of the United States have experienced epidemics of the disease, but
the inland cities appear to have been relatively free from it. During
inter-epidemic years mild, atypical, sporadic cases and moderate
numbers of bacilli carriers (both of the Shiga and Flexner types of
organisms) have been discovered. 3

The most virulent of the dysentery bacilli was isolated and described
by Shiga 4 during the great epidemic of 1897 to 1898 in Japan. Flexner 5
recovered an organism which he believed was identical with the Shiga
bacillus from cases of dysentery in the Philippines. Later studies of
this organism by Martini and Lentz 6 revealed specific differences in

1 Quoted in Public Health Reports, 1900, xv, 1.

2 Kruse, Deut. med. Wchnschr., 1900, vol. xxvi.

3 Kendall, Boston Med. and Surg. Jour., 1913, clxix, 754; May 20, 1915.

4 Cent. f. Bakt., 1898, xxiii, 599; xxiv, 817, 870, 913.
B Ibid., 1900, xxviii, 625.

6 Zeit. f. Hyg., 1902, xli, 540.

316

THE ALCALIGENES DYSENTERY TYPHOID

agglutinins from the Shiga bacillus, and Lentz 1 showed that the Shiga
bacillus did not ferment mannite; the Flexner bacillus ferments this
alcohol with the production of acid. Later intensive studies of bacil-
lary dysentery bacilli by Park and Dunham, Hiss and Russell, and
others confirmed the work of the earlier observers and added several
strains to the group, which differ from the Shiga and Flexner strains
both with respect to their specific agglutinating powers and their
cultural reactions. The principal cultural reactions of the more
prominent Gram-negative intestinal bacteria, including not only the
pathogenic organisms but the habitually parasitic organisms as well,
follow :

Motility.

Dextrose.

Lactose.

Mannite.

Levulose.

Galactose.

Maltose.

Gelatin
liquefaction.

B. alcaligenes
B. dysenterise Shiga ....
B. dysenterise Flexner
B. dysenterise Hiss- Russell .
B. dysenterise Rosen ....
B. pyogene^ fcetidus . .
B. typhosus a- . . . . . .
B. typhosus b
B. para typhosus alpha
B. paratyphosus beta
B. Morgan No. 1

+
+
+
+
+
+
-f-

-h
+'
+
+
+
+
+
g
g

+
+

+
+
+
+
+
+
g
g

+
+
+
. +
+
+
g
g
?

+
+
+
+
+
g

g
?

+
+
+
+
+
g

g
?

+
+

-4-

B. coli a
B. coli b
B. proteus
B. cloacae

+
+
+
+

g
g

g
g
g

g
g

g
g
g
g

C 1

p
c/p

Legend: carbohydrate solutions: - = no fermentation, + = acid produced, g = gas

produced.

milk: = no fermentation, alkaline reaction, =*= = initial acidity, terminal
alkalinity, + = acid, c = coagulation, p = peptonization.

Morphology. The morphology of the members of the dysentery
group of bacilli is practically identical; they are medium-sized, rod-
shaped organisms, measuring from 0.8 to 1 micron in diameter, and
from 1.5 to 3 microns in length. They have rounded ends and occur
singly or in pairs, rarely in short chains. Frequently elongated
somewhat irregular involution forms are found in old broth cultures.
The bacilli are non-motile (except the "Rosen" strain, which is slug-
gishly motile), possess no flagella, form no capsules and produce no
spores. They stain fairly readily with ordinary anilin dyes; frequently,

1 Zeit. f. Hyg., 1902, p. 559.

THE GROUP OF THE DYSENTERY BACILLI 317

the ends of the organisms stain somewhat more heavily than the
centre. All the organisms comprising this group are Gram-negative.

Isolation and Culture. The dysentery bacilli grow well on ordinary
laboratory media. Colonies on agar, after eighteen to twenty-four
hours' incubation at the body temperature, are round, transparent
and colorless; frequently they attain a diameter of from 1 to 3 mm.
The colonies are indistinguishable from those produced by bacilli of
the typhoid and paratyphoid groups. There is moderate growth
along the line of inoculation in gelatin, but no liquefaction. In broth
after eighteen to twenty-four hours' growth a uniform turbidity
develops, somewhat more luxuriant in dextrose than in plain broth.
After several days' growth in plain broth a delicate pellicle frequently

FIG. 45. Bacillus dysenterise. Shiga type, bouillon culture. X 1000. .

appears on the surface of the latter medium. In milk moderate devel-
opment takes place with no coagulation. There is 'an initial acidity
followed after from two to five days by an alkaline reaction, which
increases somewhat in intensity with the age of the culture. On potato
the growth is very similar to that of the typhoid bacillus; on acid
potato the growth is almost invisible; on alkaline potato the growth
is brownish and of moderate luxuriance.

The dysentery bacilli are aerobic, facultatively anaerobic bacilli
whose limits are approximate^fcthe following; minimum temperature
of growth 8 C.; maximum 42 to^ C.; optimum 37 C.

Cultures of dysentery bacilli varyWimewhat in their resistance to
heat. The majority of cultures are killed by an exposure of ten min-
utes at 65 C. Some strains, however, are only killed by an exposure
of ten minutes at 70 C. The organisms are moderately resistant to

318 THE ALCALIGENES DYSENTERY TYPHOID

cold. Cultures may retain their viability in the ice-box, 6 to 10 C.,
for nearly two months. In sterile water the organisms at ordinary
temperatures do not as a rule survive more than a week. Pfuhl 1 has
found that dysentery bacilli may remain alive for 101 days in moist
soil protected from sunlight; in dry soil under otherwise the same
conditions they do not survive more than thirty days. In cheese and
in butter they remain alive for at least nine days, and in sterile milk for
about three weeks. Dried on linen, they also survive about three weeks.

Products of Growth. Chemical Products. Plain broth cultures of
Shiga and Flexner bacilli do not contain indol or phenols, even after
prolonged incubation. The statements with reference to indol produc-
tion in the group, however, are somewhat conflicting, particularly
with reference to the Flexner type of organism. Morgan and others 2
have stated that Flexner bacilli produce indol; on the other hand,
Kendall, Bagg, Day and Walker 3 have isolated over 200 strains of
Flexner bacilli from dysenteric cases and have found almost without
exception that indol is not formed. These strains were identified by
their cultural reactions and by agglutination with specific Flexner
serum of high potency. Dopter 4 has found that strains of Flexner
bacilli obtained from different sources, which were identical culturally
and agglutinated the same with specific sera, vary in indol production
some producing indol, others not producing it.

Acid Production in Carbohydrate Media. All members of the dysen-
tery group agree in two important characteristics: they do not form
gas in carbohydrate media, and form acid in dextrose. Lentz 5 has
called attention to an important cultural differentiation of the Flexner
and Shiga bacillus, the former producing acid in mannite, the latter
not fermenting this alcohol. Further study has shown that the fer-
mentation of various carbohydrates is important in the recognition
of the various types. The fermentation and other cultural reactions
of members of the dysentery bacillus group are shown in the table on
page 316. The members of the dysentery group produce an initial
acidity in milk; fermentation of the small amount of dextrose, amount-
ing to about 0.1 per cent., which is found in fresh milk (Theobald
Smith 6 ) followed by an .alkaline reaction (action of the organisms
upon protein when the utilizable carbohydrate is exhausted). 7

i Ztschr. f. Hyg., 1902, xl, 555. 2 Brit. Med. Jour., April 6, 1907, 908; July 6, 16.

3 Boston Med. and Surg. Jour., 1911, clxiv, 301; 1913, clxix, 741, 753; Jour. Am.
Chem. Soc., 1913, xxxv, 1211.

4 Les Dysenteries, Paris, 1909, 36. 6 Ztschr. f. Hyg., 1902, xli, 559.

5 Boston Jour. Med. Sci., 1897, ii, 236; Jones, Jour. Inf. Dis., 1914, xv, 357.

7 See Kendall, Day and Walker for essential analytical details, Jour. Am. Chem.
Assn., 1914, xxxvi, 1940,

THE GROUP OF THE DYSENTERY BACILLI 319

Enzymes. Dysentery bacilli do not appear to produce extracellular
proteolytic enzymes. They do not liquefy gelatin, blood serum or
fibrin, and do not coagulate milk. Wells and Corper 1 have demon-
strated a lipase of moderate activity in the autolysates of dysentery
bacilli.

Toxins. (a) Exotoxin. The nature of the poison produced by
the Shiga bacillus, the most virulent of the dysentery bacilli, is a
matter of debate. Todd, 2 Ludke, 3 Doerr, 4 and Kraus and Doerr 5
state that the organism produces a soluble (exo-) toxin which
stimulates antibody formation in suitable animals; the sera are
specifically antitoxic and protect laboratory animals against several
times the fatal dose of the toxin. According to Kraus and Doerr, 6
this toxin acts somewhat like that of the diphtheria bacillus; the
lesions observed in the large intestine are comparable to the lesions
of the diphtheria bacillus on the tonsils and pharynx. The nervous
lesions are somewhat like those of poliomyelitis. Intravenous injec-
tion of large doses in rabbits causes death in from six to eight hours;
smaller doses cause paresis, diarrhea, which is frequently bloody,
paralysis of the bladder, hypothermia and death in one to four weeks.
Postmortem there is a mucohemorrhagic enteritis, usually localized
in the cecum. It is stated that the entire intestinal tract is involved
in dogs, with the duodenum particularly affected. Intraperitoneal
and subcutaneous injections give a much milder reaction with a pro-
longed incubation period. The toxin is inactivated by acids, but its
potency may be partially restored when the acid is neutralized with
alkali. Conradi 7 and others find dead cultures almost as toxic as the
living bacilli ; they call attention to the toxic properties of autolysates
(in sterile water) of the Shiga bacillus, a fact which was pointed out
by Gay 8 some time before. It is probable that both soluble and
autolytic poisons are concerned in the toxicity of filtrates of broth
cultures of the organism. The toxic substances may be obtained in
dry form by saturating the broth (freed from bacilli by filtration
through unglazed porcelain) with ammonium sulphate, dialyzing the
precipitate to remove the ammonium salts, and evaporation of the

1 Jour. Infec. Dis., 1912, xi, 388.

2 Brit. Med. Jour., December 5, 1902, ii; October 4, 1903, ii.

3 Jour. Path, and Bact., 1905, x, 328.

4 Cent. f. Bakt., Orig., 1905, xxxviii, 420, 511.
s Ztschr. f. Hyg., 1906, Iv, 1.

6 Loc. cit.

7 Deutsch. med. Wchnschr., 1903, xxix, 26.
sPenna. Med. Bull., 1902.

320 THE ALCALIGENES DYSENTERY TYPHOID

dialyzed solution to dry ness in vacua. The dried residue is very toxic
for rabbits; 0.002 to 0.005 grams dissolved in a small amount of
sterile salt solution will usually kill these animals when injected
intravenously. Smaller amounts gradually increased stimulate anti-
body formation. 1 The antitoxin, however, has little curative value,
for the toxin appears to have a greater affinity for the epithelium of the
intestinal mucosa and central nervous system than it has for the anti-
toxin. The other members of the dysentery group do not produce
soluble toxic substances in demonstrable amounts.

(b) Endotoxin. Neisser and Shiga 2 have found that autolysates
of Shiga bacilli produce a mucohemorrhagic enteritis in rajbbits.
Besredka, 3 Conradi 4 and others have also extracted substances from
the organisms by grinding them with sand, by alternate freezing and
thawing (method of MacFadyen and Roland), or by autolysis, which
in small amounts will kill experimental animals when injected intra-
venously, intraperitoneally, or subcutaneously. Administration by
mouth is without noteworthy effect. The potency of the endotoxin
is not appreciably impaired by an exposure to 70 C. for an hour; an
exposure to 80 C. renders it inactive. Conradi 5 has shown that
occasional strains of dysentery bacilli (Shiga type) produce small
amounts of soluble hemotoxin.

Pathogenesis. Experimental. Direct experimental evidence of the
etiological relationship of the dysentery bacillus to bacillary dysentery
is afforded by a few laboratory accidents in which the clinical disease
has followed the accidental ingestion of cultures of dysentery bacilli.
The most conclusive experiment, however, is that reported by Strong
and Musgrave. 6 A forty-eight-hour broth culture of B. dysenteric
(Shiga type) was swallowed by a condemned criminal after a dose of
sodium hydrogen carbonate was given to neutralize the gastric acidity.
The initial symptoms of a typical attack of bacillary dysentery fol-
lowed after an incubation period of thirty-six hours. The organisms
were isolated from the mucopurulent, bloody feces*. Ravant and
Dopter 7 produced clinical dysentery in an ape by feeding it Shiga
bacilli.

Human. Infection with the Shiga bacillus is somewhat less com-
mon in the United States than infection with the Flexner and other

1 Todd, loc. cit. ; Kraus and Doerr, loc. cit.

2 Deutsch. med. Wchnschr., 1903, No. 4. 3 Ann. Inst. Past., April, 1906, vol. xxv.
4 Loc. cit. 6 Loc. cit.

6 Report of the Surgeon-General, United States Army, 1900.

7 Quoted by Kolle and Hetsch, Die experimentelle Bakt., II. Aufl., i, 304.

THE GROUP OF THE DYSENTERY BACILLI 321

types of the dysentery group, but far more fatal. Mixed infections
in which both Shiga and Flexner bacilli are present are occasionally
seen. 1 Among adults infection with the Flexner type of organism
tends to be sporadic in distribution and less severe than infections
with the Shiga type which more commonly assume epidemic tendencies.

The incubation period of bacillary dysentery may be as brief as
forty-eight hours, or even less, and as a rule there are no distinctive
prodromal symptoms. The feces, at first watery, may be very fre-
quent, as many as twenty to thirty per diem, and become muco-
purulent with considerable amounts of fresh blood mixed in them.
The organisms are present in variable numbers. Dysentery bacilli
do not as a rule appear to invade the blood stream, but at least three
instances are on record where pure cultures of the Shiga bacillus have
been isolated antemortem from the general circulation; 2 occasionally
pure cultures of dysentery bacilli may be obtained from mesenteric
lymph nodes postmortem.

Lesions. The lesions, which are found chiefly in the large intestine,
vary with the severity and duration of the disease. In the early stages
of the disease there is a severe catarrhal inflammation of the mucous
membrane of the large intestine with some necrosis of the epithelium,
associated with hyperemia of the mucosa of the small intestine as well.
The mesenteric glands are usually swollen and hyperemic. Later the
inflammation may become very severe; a pseudomembrane may form
in the large intestine with extensive superficial ulceration of the
mucosa. The ulcers do not extend as a rule to the submucosa; conse-
quently, perforation is rare in uncomplicated cases. The submucosa,
however, may be swollen and somewhat edematous.

The nervous symptoms which are a feature of severe dysentery
infections would suggest that in addition to the intestinal lesions there
may be involvement of the nervous system. Southard, McGaffin
and Richards 3 have shown that in addition to lesions of the intes-
tinal tract, the Shiga toxin has a special affinity for the anterior horn
ganglion cells, thus explaining on a definite anatomical basis the ner-
vous symptoms which are a feature of fatal cases of bacillary dysentery.
Dopter 4 has expressed the same opinion. He believes the toxin of the

1 Kendall, Bagg, Day and Walker, loc. cit.

2 Rosenthal, Deutsch. med. Wchnschr., 1903, No. 6; Kendall, Bagg and Day, Boston
Med. and Surg. Jour., 1913, clxix, 741; Darling and Bates, Am. Jour. Med. Sc., 1912,
clxiii, No. 1.

3 Boston Med. and Surg. Jour., 1909, clxi, 65, 108.

4 Loc. cit., p. 77.

322 THE ALCALIGENES DYSENTERY TYPHOID

Shiga bacillus has an elective affinity for the intestinal mucosa of
the large intestine, and it is the toxin secreted by the dysentery bacilli
during their multiplication in the intestinal mucous membrane which
induces the anatomical and nervous lesions characteristic of the
disease.

Animals. Typical bacillarv dysentery has not been produced in
laboratory animals by feeding the organisms. The intravenous inocu-
lation or intraperitoneal injection of living or killed broth cultures of
Shiga or Flexner bacilli, however, are usually fatal, particularly to
rabbits. Vaillard and Dopter, 1 and Flexner 2 have shown that small
amounts of forty-eight-hour broth cultures of Shiga bacilli introduced
intravenously into young rabbits frequently lead to diarrhea, which
at first is mucous in character; later it becomes mucosanguineous.
After two or three days symptoms of paraplegia develop. At autopsy
the large intestine is swollen and frequently edematous. The mesen-
tery is hyperemic with enlarged glands. The intestinal contents are
mucosanguineous in character and the intestinal wall is considerably
thickened. If the animal survives for several days, more advanced
lesions are sometimes seen, particularly beginning ulcer ation and
necrosis. Flexner states that the intestinal lesions of bacillary
dysentery in man and in animals are probably due, in part at least,
to the direct action of the dysentery toxin.

Immunity and Immunization. Shiga 3 and others have succeeded in
immunizing laboratory animals, particularly rabbits, guinea-pigs, and
horses, with dysentery bacilli, beginning by injecting killed cultures
of these organisms, first with very small amounts which are slowly
and cautiously increased, finally with living bacilli. It is difficult to
immunize animals because of the toxicity of the organism. The sera
of these animals contain specific agglutinins, lysins, precipitins, and
opsonins, frequently of high potency. According to Todd, and Kraus
and Doerr, 4 specific antitoxins are also demonstrable in the sera of
these animals, particularly in animals immunized to the Shiga bacillus.
The agglutinins which are specific for the type of organism used in
immunization are, according to Dopter, 5 as a rule of greater potency
when killed cultures exclusively are used for immunizing. In .thor-
oughly immunized animals the agglutinins may be active even in
dilutions of 1 to 5000.

1 Ann. Inst. Past., 1903, p. 472.

2 Jour. Exp. Med., 1906, vol. viii.

3 Ztschr. f. Hyg., 1902, xli, 355.

4 Loc. cit. 5 Loc. cit., p. 84,

THE GROUP OF THE DYSENTERY BACILLI 323

Specific bacteriolysins have been demonstrated in immune sera
in vitro by Shiga 1 and in vivo by Kruse. 2 Specific precipitins, which
in dilutions of 1 to 10 or greater will produce a precipitate in broth
filtrates of the homologous strain, but not, as a rule, for other types
of the dysentery bacilli are also found. The sera of patients who
have recovered from attacks of bacillary dysentery usually contain
specific agglutinins which are active even in dilutions of 1 to 50.
Specific precipitins, lysins, and opsonins are also demonstrable in the
sera of these patients.

Therapy. Attempts to immunize man with vaccines, both mono-
and polyvalent, 3 sensitized vaccines (bacteria which have been in
contact with antidysentery serum, then centrifugalized, washed, and
suspended in salt solution, according to the method of Besredka and
of Gay), and the use of antisera, usually derived from immunized
horses, have not been generally successful, although a few favorable
results have been recorded.

Bacteriological Diagnosis. (a) Agglutinin Reaction. The sera of
normal individuals rarely agglutinate dysentery bacilli in dilutions
greater than 1 to 10, although Dopter 4 states that the Flexner organ-
ism may be clumped with the serum of apparently normal individuals
in a dilution greater than 1 to 10. For this reason agglutination tests
should be made in a dilution of 1 to 20 to 1 to 30 with the Shiga
organism, and 1 to 80 to 1 to 100 with the Flexner strain in each
case examined, since one or the other organism, or both, may be
present in typical cases of bacillary dysentery. Agglutinins do not
as a rule appear in mild cases, and in severe cases they are not
demonstrable until from the seventh to the tenth day on the average.

The serum of dysentery carriers, both those giving a history of a
previous attack and those with the negative dysentery history, fre-
quently agglutinates either with Shiga or Flexner bacilli. The agglu-
tination reaction, therefore, is not conclusive for clinical diagnosis
unless a negative reaction is obtained early in the disease followed
by a positive reaction on or after the seventh to the tenth day.

(6) Isolation of Dysentery Bacilli from the Feces. Dysentery bacilli
do not invade the blood stream as a rule, and they are not found
in the urine. The bacteriological diagnosis, therefore, depends upon

1 Loc. cit.

2 Deutsch. med. Wchnschr., 1902.

3 Shiga, Deutsch. med. Wchnschr., 1901, Nos. 43 and 45; Kruse, ibid., 1903, Nos. 1
and 3.

4 Loc. cit., p. 91.

324 THE ALCALIGENES DYSENTERY TYPHOID

the isolation of the organisms from the feces and their identification
by cultural and serological reactions.

A bit of blood-stained mucus offers the best material for isolation
of the organisms: it should be washed two or three times in sterile
salt solution to remove extraneous organisms as far as possible, for
experience has shown that dysentery bacilli are frequently enclosed
in mucus. The mucus is then macerated in sterile broth, and if
possible incubated for one or two hours at 37 C. It is then spread
upon the surface of Endo-plates and incubated for eighteen to twenty-
four hours at 37 C. The colonies are precisely similar to those of
typhoid and paratyphoid bacilli; the final identification of the dysen-
tery bacilli is made by their cultural reactions (see page 316) and by
agglutination with specific sera of high potency. The rapid method
of isolating and identifying typhoid bacilli described on page 338 is
equally applicable to dysentery bacilli. The possibility of carriers
should be borne in mind when mild and atypical cases are under
consideration .

Dissemination and Prophylaxis. Dysentery bacilli appear to be
widely distributed in certain areas of the temperate zone, and out-
breaks occur at varying intervals. Interepidemic years are occa-
sionally characterized by considerable numbers of atypical, mild
cases, and carriers are not uncommon. 1

The organisms enter the body through the mouth and intestinal
tract, and leave it in the feces; consequently the method of trans-
mission of the disease is similar to that of typhoid and other excre-
mentitious disorders. There is some evidence that the disease may
be milk-borne; exclusively breast-fed infants are rarely or never
infected; bottle-fed babies of the same age may be infected in relatively
large numbers during years which exhibit an epidemic tendency of
bacillary dysentery. Zinsser 2 has produced evidence in favor of the
occasional milk transmission. The organism may also reach the body
by direct transmission through carriers, in hospitals, and through
contaminated water and food. Flies may also play a part in the spread
of the disease.

The precautions to be observed are those for any intestinal infec-
tion.

1 Kendall, Boston Med. and Surg. Jour., 1913, clxix, 7493; ibid., May 20, 1915.

2 Proc. New York Path. Soc., 1907.

TYPHOID BACILLUS 325

TYPHOID BACILLUS.

Historical. Typhoid bacilli were first seen in sections of tissue
from autopsies by Klebs in 1876. Somewhat later Eberth 1 success-
fully demonstrated them in sections of mesenteric glands, lymph
nodes and the spleen by the use of the recently introduced tissue
stains. Gaffky 2 first isolated the organisms in pure culture and
established their probable etiological relationship to typhoid fever.
Later investigations with more refined methods have completely
substantiated Gaffky's observations.

Morphology Typhoid bacilli are rod-shaped organisms of moderate
size, measuring from 0.5 to 0.8 microns in diameter and from 1
to 3 microns in length. The dimensions vary within the limits
given upon different media, the organisms being as a rule somewhat
longer in fluid media than upon solid media. Elongated rods and even
filaments are occasionally found in old gelatin and potato cultures.
The bacilli have rounded ends and occur as a rule singly or in pairs.
They are actively motile, particularly in young cultures grown in
0.1 per cent, dextrose broth; plain broth cultures are usually more
sluggish. Each organism possesses characteristically from eight to
ten peritrichic flagella; rarely as many as twenty may be attached to
a single organism. The flagella are somewhat wavy in outline and
measure from 6 to 8 microns in length. No spores are produced
It was formerly held that typhoid bacilli formed no capsules. Car-
pano, 3 and Gay and Claypole, 4 however, have demonstrated capsules
around typhoid bacilli grown in blood media.

The organisms stain readily with ordinary anilin dyes and they
are Gram-negative.

Isolation and Culture. The typhoid bacillus grows readily upon the
ordinary media. Colonies on agar plates are round, colorless, flat, and
nearly transparent; they attain a diameter of from 0.5 to 1.5 mm.
after eighteen to twenty-four hours' incubation at 37 C. Devel-
opment in gelatin is less rapid, and the colonies after two to three
days' incubation at 20 C. are somewhat brownish in color. A uniform
turbidity is produced in plain broth after eighteen hours' growth at
37 C.; development in dextrose broth is more intense, but after five
to seven days it ceases and the organisms die, due to the accumula-

1 Virchows Arch., 1880, Ixxxi, 58; 1881, Ixxxiii, 486.

2 Mitt. a. d. kais. Gesamte, 1884, ii, 370.

3 Cent. f. Bakt., Orig., 1913, Ixx, 42.

4 Arch. Int. Med., 1913, xii, 624.

326 THE ALCAL1GENES DYSENTERY TYPHOID

tion of acid. Growth is luxuriant in milk, but there is little chemical
change in the composition of the medium as the result of the growth. 1
Two types of reaction are observed in litmus milk: (a) The reaction
becomes slightly acid, turning the litmus to a lilac color which per-
sists. This is much more common than (6) ; the milk becomes slightly
acid, as in "a," then it becomes slowly but progressively alkaline.
Relatively few authentic strains of typhoid bacilli appear to produce
the transient acidity in this medium. At one time potato was regarded
as an important differential medium for the recognition of the typhoid
bacillus. The "invisible growth" described by Gaffky 2 is now known
to be dependent largely upon the reaction; potatoes having an acid
reaction give this invisible growth; old potatoes which usually have

FIG. 46. Bacillus typhosus, flagella stain.

a slightly alkaline reaction give a heavy, brownish growth much like
that of the colon bacillus. The addition of small amounts of alkali,
as sodium carbonate, to potato prior to inoculation makes the growth
visible and brown; the addition of a small amount of organic acid to
the medium usually results in the development of the invisible type
of growth.

The typhoid bacillus is an aerobic, facultatively anaerobic organism,
whose minimal temperature of growth is about 8 C.; development is
maximal at 37 C., and ceases when the culture is exposed to tem-
peratures above 43 to 44 C. An exposure of ten to twenty minutes
at 60 C. will kill the naked organisms; a longer exposure at a higher
temperature is required to kill them when they are suspended in organic

1 Kendall, Day and Walker, Jour. Am. Chem. Assn., 1914, xxxvi, 1958,

2 Loc. cit.

TYPHOID BACILLUS 327

matter, as feces. Cultures exposed to temperatures from C. to
10 C. for three months occasionally contain viable organisms.
Alternate freezing and thawing is more fatal than simple freezing.
The typhoid . bacillus dies out rather rapidly in potable water, less
rapidly in sterilized potable water. The addition of organic matter,
particularly of fecal origin, appears to promote longevity somewhat.
The observations of Jordan, Russell and Zeit 1 would indicate that a
large percentage of organisms exposed in potable water die within
three days. Kersten 2 has shown that typhoid bacilli will develop
with considerable rapidity in raw milk. The bacilli may remain alive
in soil for several months, provided they are shielded from direct
sunlight, and they may resist drying under similar conditions for

FIG. 47. Bacillus typhosus, bouillon culture. X 1000.

several weeks. A maximum exposure of from four to eight hours to
direct sunlight in the months of June, July and August (Northern
Hemisphere) usually kills the organisms. Mercuric chloride 1 to 1000
kills the naked germs in about ten minutes; 5 per cent, carbolic acid
kills them in from five to ten minutes, as a rule.

Products of Growth. The typhoid bacillus liberates ammonia from
protein in sugar-free media, and forms small amounts of non-volatile
alkaline products as well. The reaction, therefore, becomes progres-
sively alkaline. A radical change in the nature of the products of
metabolism occurs when the bacilli are grown in protein media con-
taining utilizable carbohydrates, as dextrose or mannite. The reaction
becomes strongly acid, due to the fermentation of the sugar. The

1 Jour. Infec. Dis., 1904, i, 641.

2 Arb. a. d. kais. Gesarat., 1909, xxx, 341.

328 THE ALCALIGENES DYSENTERY TYPHOID

protein under these conditions is left unattacked except for minute
amounts necessary to supply the nitrogenous requirements of the
organism. The acids formed are chiefly lactic acid, together with
smaller amounts of formic acid. 1 Indol or phenols are not formed
in ordinary media, but Peckham 2 has shown that indol may be pro-
duced in protein media of special composition.

The essential cultural characters of B. typhosus are indicated in
the table on page 316. Culturally Bacillus typhosus. is relatively
inert; it does not produce proteolytic enzymes which liquefy gelatin,
blood serum or fibrin. A fat-splitting ferment has been demonstrated
in autolyzed typhoid bacilli by Wells and Corper. 3 An esterase
which liberates butyric acid from ethyl butyrate is detectable in sterile
filtrates of plain and dextrose broth cultures of the organism. 4

Typhohemolysin (typholysin) . Castellani, 5 and E. Levy and P.
Levy 6 have found that filtrates of (sugar-free) broth cultures of typhoid
bacilli are hemolytic. They appear to have demonstrated specific
antihemolytic properties in the blood of animals injected with hemo-
lytic filtrates, thus meeting the objection that the hemolysis might be
due to the alkalinity of the medium itself. There is no evidence at
present which would suggest that this hemolysin plays any impor-
tant part in typhoid infections of man. The typholysin is relatively
thermostabile.

Toxins. A soluble toxin has never been satisfactorily demon-
strated among the products of growth of the typhoid bacillus, and
the consensus of opinion at the present time is in favor of the view
that the principal toxic substance of the organism is an endotoxin.
The endotoxin has been studied with special thoroughness by Mac-
Fadyen and Roland, 7 and Besredka. 8 It has been obtained in various
ways: by grinding the organisms with sand, by freezing in liquid air
and triturating, or by autolysis of the bacilli in sterile distilled water.
Relatively small amounts of endotoxin obtained by any of these
methods will usually kill guinea-pigs. No antitoxin has been produced
in the sera of animals inoculated with gradually increasing amounts
of this endotoxin.

1 Kendall, Jour. Med. Research, 1911, xxiv, 411; 1912, xxv, 117. Boston Med. and
Surg. Jour., 1911, Ixiv, 288. Kendall, Day and Walker, Jour. Am. Chem. Assn., 1913,
xxxv, 1214.

2 Jour. Exper. Med., 1897, ii, 549. 3 Jour. Infec. Dis., 1912, xi, 388.
4 Kendall and Simonds, Jour. Infec. Dis., 1914, xv, 354.

6 Lancet, February 15, 1902. Cent. f. Bakt., 1901, xxx, 405.

7 Cent. f. Bakt., Orig., 1903, xxxiv, 618, 765; MacFadyen, ibid., 1903, xxxv, 415.

8 Ann. Inst. Past., 1905-1906.

TYPHOID BACILLUS 329

Typhoid Fever. Pathogenesis. Experimental Typhoid fever is a
disease of man only, and until recently rigorous experimental proof
that the typhoid bacillus is the specific cause of this infection has been
lacking. The evidence of the etiological relationship of the typhoid
bacillus is of two kinds: (1) a few cases where laboratory attendants
have accidentally or purposely swallowed cultures of typhoid fever
and have developed the disease ; (2) experiments of Metchnikoff and
Besredka. 1

The experiments of Metchnikoff and Besredka appear to be con-
clusive. They produced typhoid fever in anthropoid apes by feeding
the animal food infected with fecal material containing typhoid bacilli.
The animals (fifteen in all) developed fever and diarrhea after eight
days, and typhoid bacilli were isolated from the blood stream on the
tenth day. Three died. Specific agglutinins were demonstrable in
the blood serum, and the clinical picture was essentially that of typical
typhoid fever. These observers ruled out the possibility of a filterable
virus.

Pathogenesis in Man. Portal of Entry. Typhoid bacilli enter the
body through the mouth and pass through the gastro-intestinal tract.
They lodge in lymphatic tissue of the intestines, particularly Peyer's
patches, then invade the general lymphatic system and spleen, and
are found in the blood, especially during the first week of the clinical
disease. Typhoid fever, therefore, is a bacteremia. Rose spots, which
are frequently found on the abdomen during the first week of the
clinical disease, contain colonies of typhoid bacilli which are localized
in the subcutaneous tissue. 2 Characteristic lesions are found in Peyer's
patches which at first are swollen and hyperemic. After a few days
the glands become rather pale, caused, in part at least, by hyperplasia
of the lymphoid and endothelioid cells, which cuts off the blood supply
in whole or in part, leaving these areas even more prominent (medul-
lary swelling). 3 Necrosis then commences and the glands gradually
become yellowish in color and 'softer in consistency. Soon the necrosis
ceases rather abruptly as immunity checks the process and the necrotic
tissue then sloughs away, leaving a somewhat irregular elongated ulcer
which usually extends to or through the muscular layer of the intestine.
About the end of the third week scar tissue begins to appear in these
ulcers, which in time practically fills up the original area, leaving the

1 Ann. Inst. Past., March 25, 1911; xxv, 193, 865.

2 Richardson, Philadelphia Med. Jour., March, 1900. (Special Typhoid Fever Number.)
8 Mallory, Jour. Exp. Med., 1898, iii, No. 6, p. 611.

330 THE ALCALIGENES DYSENTERY TYPHOID

site of the ulcer marked by a somewhat depressed cicatrix. Occasion-
ally secondary infection of the ulcers results in perforation or hemor-
rhage, and sometimes an uninfected ulcer may erode through a
blood vessel, causing hemorrhage. It should be remembered that
typhoid ulcers tend to run along the long axis of the intestine, whereas
tuberculous ulcers, on the contrary, run transversely, following the
course of the lymphatics.

In addition to the intestinal lesions, there is in typhoid fever an
acute splenic tumor with a great proliferation of typhoid bacilli in
this organ. Foci of typhoid bacilli are commonly found also in the
kidneys and the liver, mesenteric lymph nodes, less commonlv in
lungs, meninges, bone marrow, certain muscles and the tonsils. Paren-
chymatous degeneration of the heart, liver and kidneys is common,
as is a catarrhal inflammation of the respiratory tract and a severe
inflammation of the entire intestinal mucous membrane. Somewhat
uncommonly, typhoid cases have been recorded in which there are no
intestinal lesions. In these cases it would appear that the disease is
septicemic in character. 1 In typhoid fever there is leucopenia, due
apparently to some interference with the activity of the bone marrow.
The febrile reaction is usually attributed to the liberation of endotoxin
from typhoid bacilli, which are dissolved in the blood stream by
specific lysins. This toxin exhibits both a general and local reaction.
The general reaction is characterized chiefly by fever and symptoms
of generalized toxemia; the local reaction is particularly marked in
those areas where typhoid bacilli undergo solution, as in the spleen
and Peyer's patches.

Various complications of typhoid fever are occasionally reported,
caused by the localization of typhoid bacilli either alone or in
association with other bacteria, as the streptococcus, staphylococcus,
or pneumococcus, in various organs. Peritonitis, usually following
perforation of an ulcer in the intestinal wall, is one of the most severe
of these complications. Abscess formation in various deep-seated
organs, as the spleen and psoas muscle, is not uncommon. Broncho-
pneumonia, pleurisy, pericarditis, osteitis, and inflammation of the
membranes of the cord (meningitis) and brain have also been attributed
to the typhoid bacillus.

Carriers. Typhoid bacilli can not be isolated from the majority
of typhoid patients after the fifth week of the disease. In a small

1 Possett, Atypische Typhusinfektion. Lubarsch and Ostertag, Ergebn. d. allgem.
Pathol., 1912, xvi, 184.

TYPHOID BACILLUS 331

percentage of cases, however, the organisms may be excreted in the
urine, or more commonly in the feces, for months or even years after
recovery. Thus, Philipowicz 1 isolated typhoid bacilli from a case
of cholecystitis who had had typhoid fever thirty-eight years previous
to the operation. In this case very few typhoid bacilli were present
in the feces, and it is probable that the few organisms were over-
whelmed by the intestinal bacteria during their passage through the
intestinal tract. From 1 to 4 per cent, of all typhoid cases which
recover appear to become fecal typhoid carriers; a smaller percentage
become urinary carriers. No history of typhoid fever can be elicited
from some of these carriers, and the supposition is that either the
carrier had in the past a mild unrecognized case, or less commonly
that the organism had become acclimatized in the intestinal tract
without inducing disease. Many carriers give a positive Widal reaction.

The residual focus of typhoid bacilli in carriers is usually the gall-
bladder and the ducts of the gall-bladder, less commonly the urinary
bladder. From the gall-bladder the organisms pass in irregular num-
bers into the intestinal tract; occasionally in sufficient numbers to be
demonstrable in the feces. A considerable proportion of operations
for cholecystitis and gall-stones the greater majority being among
women give positive typhoid cultures when the contents are examined
bacteriologically.

Pathogenesis in Animals. All animals, except possibly anthropoid
apes, are naturally immune to typhoid fever, and inoculation of old
laboratory cultures of typhoid bacilli into laboratory animals is
usually without noteworthy effect; virulent cultures of typhoid bacilli,
particularly those produced by repeated passage through laboratory
animals, may produce peritonitis and death when they are introduced
into the animals by the intraperitoneal route. The infection, how-
ever, does not resemble typhoid fever. The lesions observed post-
mortem are marked congestion of the abdominal organs, particularly
the spleen, kidneys and liver, as well as involvement of the intestinal
lymph apparatus; the thoracic organs are less involved as a rule.

The organisms may be recovered from the peritoneal fluid, the blood
stream, and from various abdominal organs. Gay and Claypole 2
have succeeded in inducing with great regularity the carrier state in
rabbits by injecting into them typhoid bacilli which have been grown
for several successive transfers on agar overlaid with fresh defibrinated

1 Wien. klin. Wchnschr., 1911, 1802.
2 Arch. Int. Med., December, 1913.

332 THE ALCALIGENES DYSENTERY TYPHOID

rabbit's blood. They found that the typhoid bacilli localize them-
selves in the gall-bladders of the rabbits, and that they may from time
to time invade the blood stream. In a more recent communication 1
they have shown that the carrier state occurs much less frequently
if the animals are immunized with their dried sensitized vaccine.

Antibody Production. Animals may be immunized by repeated
injections of typhoid bacilli to such a degree that they will successfully
resist several times the original fatal dose of these organisms. 2 Suc-
cessive injections of typhoid bacilli stimulate antibody formation
in horses, rabbits, guinea-pigs, and other animals. Of these anti-
bodies, the lysins and agglutinins may be produced in high potency
if the injections are continued long enough. Other antibodies, opsonins
and precipitins particularly, are also produced. Gay and Claypole 3
have produced experimental evidence indicating that the titre of the
specific agglutinins which develop during the process of immunization
of rabbits affords no indication of the degree of protection attained by
the immunizing process.

Protective Immunization. As a rule, one attack of typhoid fever
confers immunity; subsequent attacks are unusual.

During the last few years definite progress has been made in the
protective immunization of human beings, both by the use of killed
cultures of typhoid bacilli and by live cultures. The vaccine treat-
ment for typhoid fever is the best known and the most widely prac-
ticed. The procedure is to grow typhoid bacilli on agar slants, wash
them off with sterile physiological salt solution, kill them by heating
to 60 C. for an hour, standardizing the suspension of typhoid bacilli,
and injecting as a first dose five hundred million killed typhoid organ-
isms. After an interval of seven to ten days a second injection of a
billion killed typhoid bacilli is made, and after an equal interval a
third and last injection of a billion killed typhoid bacilli is made.
In about 20 per cent, of the cases injected general symptoms which
consist of a febrile reaction and malaise develop, accompanied by
local symptoms of pain, redness, and swelling at the site of inoculation.
These symptoms may appear after the second or even after the third
injection. It is customary to make the inoculation about four o'clock
in the afternoon, so that the patient in the majority of cases sleeps
through the general symptoms.

1 Arch. Int. Med., 1914, xiv, 671.

2 See Gay and Claypole, Arch. Int. Med., 1914, xiv, 671, for essential details.

3 Loc. cit.

TYPHOID BACILLUS 333

The immunity produced is generally considered to be relatively
complete for from six months to a year. It must be remembered that
for at least three weeks following the vaccination there is a diminution
in the resistance of the individual to typhoid fever; consequently,
typhoid vaccination should not be undertaken if there is a possibility
of exposure to typhoid during this period. Vaccination is also very
undesirable if it is performed during the incubation period of typhoid
fever. It should be practiced only on perfectly healthy subjects free
from all general and local organic defects or infections, particularly
tuberculosis. Nurses, ward orderlies, doctors, and those engaged in
the care of typhoid patients are particularly likely to benefit by these
inoculations. Gay and Claypole 1 have demonstrated experimentally
that a satisfactory degree of protection may be attained in animals
by three injections, at intervals of two days each, of a dried sensitized
vaccine. Observations upon man immunized with this vaccine
indicate that the reactions are milder and the whole process can be
completed within a week, thus diminishing very materially the time
element which has been an important factor in the past. It is very
probable that the period of increased susceptibility to infection may
be decidedly shortened as well.

Vaccination with Living Cultures. Metchnikoff and Besredka 2
found that the subcutaneous injection of living sensitized cultures
produced an immunity in anthropoid apes which was apparently
as definite as that produced by an actual attack of typhoid fever.
The organisms were shown not to appear in the urine or feces or blood
when introduced subcutaneously. They were unable to induce
immunity in the chimpanzee with killed cultures of typhoid bacilli
or with autolysates of killed cultures. Having in mind the efficiency
of living cultures, they 3 attempted the vaccination of man with living
cultures of the typhoid bacillus. They used sensitized cultures which
appeared to cause only a feeble local reaction and no general reaction
in the chimpanzee, in preference to non-sensitized living cultures,
which they found produced rather intense local and general reac-
tions. The vaccine was prepared by emulsifying agar cultures of
typhoid bacilli in normal salt solution and permitting the organisms
to remain in contact with antityphoid serrm for twenty-four hours
at 37 C. The organisms are then removed by centrifuging, washed

1 Loc. cit.

2 Ann. Inst. Past., 1913, xxvii, 597. Besredka, Ann. Inst. Past., 1913, xxvii, 607.

3 Semaine Med., July 24, 1912, 355.

334 THE ALCALIGENES DYSENTERY TYPHOID

repeatedly, then re-emulsified in normal saline solution and heated
to 50 C. for thirty minutes, then standardized in the usual manner.
Nearly eight hundred people have been vaccinated with these sen-
sitized living cultures; the^ local reaction was slight in each instance,
and only exceptionally was there any general reaction. A careful
examination of the blood, urine and feces of sixty-four of these cases
failed to show typhoid bacilli, which would suggest that individuals
vaccinated with living typhoid bacilli neither develop typhoid fever
nor become carriers. The cases are too few in number to compare
statistically with the cases vaccinated with killed cultures. Gay and
Claypole 1 have taken issue with Metchnikoff upon this point and their
experiments indicate that their sensitized dried vaccine may be
equally or more efficient without the theoretical dangers which attend
the use of living bacilli.

Various attempts have been made to induce passive immunity to
typhoid infection by the injection of sera obtained from horses which
have received numerous injections of typhoid bacilli or their soluble
products. The results have on the whole not been encouraging. Gay
and Force 2 have applied a preparation of typhoid bacilli (" typhoidin")
made like Koch's old tuberculin, by the von Pirquet method, to
patients that have recovered from typhoid fever and to those who have
been vaccinated with typhoid bacilli. They find that 95 per cent, of
recovered cases from typhoid (20 cases out of 21 examined) gave a
clear-cut cutaneous reaction. One case had typhoid forty-one years
previously. The reaction was negative in 85 per cent, of individuals
not -giving a history of typhoid (and presumably not vaccinated) 41
cases tested. The 9 cases (15 per cent.) that gave a positive reaction
were suspected to have had a mild undiagnosed attack. Several,
but not all, of those vaccinated within four years (9 out of 15) gave
a positive reaction. Gay and Force suggest that the test is of
presumptive value as an index of protection against typhoid by
vaccination. Later observations by them confirm this view.

Diagnosis. The diagnosis of typhoid fever in the living subject may
be made either by the isolation and identification of the specific organ-
ism, Bacillus typhosus, or by the demonstration of antibodies specific
for this organism in the body fluids of the patient.

(a) BACTERIOLOGICAL DIAGNOSIS. 1. Isolation of typhoid bacilli
from the blood stream and from rose spots.

1 Loc. cit.

2 University of California Publications in Pathology, 1913, ii, No. 14; Arch. Int. Med.,
1914, xiii, 471.

TYPHOID BACILLUS 335

Typhoid bacilli are found in the peripheral blood of a large percen-
tage of typical cases of typhoid fever during the first week of the
clinical disease. The organisms are found less frequently in the later
stages. The statistics reported by Coleman and Buxton, 1 covering
1137 cases, show this clearly.

Positive,
Cases. per cent.

First week of clinical disease 224 89

Second week of clinical disease 484 73

Third week of clinical disease ....... 268 60

Fourth week of clinical disease 103 38

Fifth week of clinical disease 58 26

The organisms have also been isolated from rose spots (which
appear as a rule early in the clinical course of the disease) by Richard-
son and others. From these observations typhoid fever may be
regarded primarily as a bacteremia. 2 It should be remembered, how-
ever, that the organisms are destroyed in the blood stream by specific
lysins, and that their presence in the circulating fluids of the body
are partly caused by an overflow of organisms from foci in the spleen
and other organisms.

Method of Collecting Blood. The skin of the elbow is thoroughly
cleansed as for a surgical operation, a tourniquet is applied, and a
large hypodermic needle is introduced into a vein, preferably the
median basilic. From 5 to 15 c.c. of blood are removed, discharged
at once into a flask containing 150 to 250 c.c. of dextrose broth (0.1
per cent.), and mixed thoroughly before clotting takes place. This
considerable dilution of the blood is important, partly because clotting
takes place more slowly and thus favors the escape of the organisms
into the broth, and also because it dilutes the lysins which are usually
present in the blood of typhoid patients. It is necessary to reduce
the concentration of lysins, for lysins dissolve typhoid bacilli. Incu-
bation of the culture at 37 C. for twenty-four hours usually results
in a growth of bacteria in which the specific organisms are present,
either alone or mixed with skin cocci.

Coleman and Buxton 3 recommend an ox bile glycerin peptone
medium for the isolation of typhoid bacilli. The medium as prepared
by them has the following composition: Ox bile, 900 c.c.; glycerin,
100 c.c.; peptone, 20 grams. This is sterilized in the autoclave and

1 Am. Jour. Med. Sc., 1907, cxxxiii.

2 Brion and Kayser, Deut. Arch. f. klin. Med., 1906, Ixxxv, 552. Coleman and
Buxton, Jour. Med. Research, 1909, xxi, 83. Kolle and Hetsch, Experimentelle Bakt.
und. Infektionskrank., 1911, 3ed., i, 250,

3 Loc. cit.

336 THE ALCALIGENESDYSENTERY^TYPHOID

distributed in flasks, 25 c.c. to a flask. The ox bile prevents the
coagulation of the blood. Three c.c. of blood, according to the Cole-
man technic, are added to the flask of this medium, incubated for
eighteen to twenty-four hours, then plated out on agar. Experience
has shown that larger amounts of blood are more satisfactory, for
it has been found that not infrequently 5 c.c. of blood will not give
a growth of typhoid bacilli, whereas 10 c.c. or, better, 15 c.c. will give
a growth. The organisms obtained in pure culture are identified by
agglutination with a known specific typhoid serum of high potency.
Such a serum used in high dilution reduces the possibility of "group
agglutinins" which might otherwise give an erroneous diagnosis.
It must be remembered that occasional strains of typhoid bacilli are
isolated from the body which are typical culturally, but which are
non-agglutinable. Frequently a few successive transfers of these
organisms on artificial media will restore their agglutinating properties;
occasionally, however, a strain is met with which will not agglutinate
with specific typhoid serum even after long-continued transfer on
artificial media. Mclntosh and McQueen 1 have found that at least
certain strains of these non-agglutinable typhoid bacilli will stimulate
the production of typical typhoid agglutinins if they are injected into
animals. The agglutinins developed in these animals will promptly
clump agglutinable typhoid bacilli, but will not agglutinate the non-
agglutinable strains which incited the production of these agglutinins.
These non-agglutinable strains, however, will absorb the agglutimns
apparently as readily as the agglutinating strairs. Gay and Claypole 2
have found similarly that occasional strains of typhoid bacilli isolated
from "typhoid carrier" rabbits may be non-agglutinable. They
absorb agglutinin, however. They suggest the use of sera obtained
from animals immunized with cultures of typhoid bacilli grown upon
agar containing the blood of man. The isolation of typhoid bacilli
from the blood stream and their identification establishes the diag-
nosis of typhoid fever beyond question of doubt.

The isolation of typhoid bacilli from rose spots is performed in essen-
tially the same manner, except that fluid is expressed from the rose
spot after the skin is sterilized over it, and the expressed fluid is grown
either in the dextrose broth or in the bile medium. Neufeld 3 and
Richardson 4 have successfully isolated typhoid bacilli from the roseola

1 Jour. Hyg., 1914, xiii, 409.

2 Jour. Am. Med. Assn., 1913, Ix, 1141; Arch. Int. Med., 1913, xii, 613.
'Ztschr. f. Hyg., 1899, xxx, 498.

4 Philadelphia Med. Jour., March 3, 1900.

TYPHOID BACILLUS 337

of typhoid fever in a considerable number of cases. Thus, Neufeld 1
obtained cultures in 13 of 14 cases examined, and Richardson obtained
them in 5 out of 6 cases. Both Neufeld and Richardson emphasize
the importance of incising several spots. The technic developsd by
Richardson is as follows: the skin over several rose spots is cleaned
as for a surgical operation and then frozen by a spray of ethyl chloride.
This procedure drives out most of the blood, as well as making the
operation practically painless. A small incision is then made with
a sterile knife and the substance of the rose spot is removed with a
small skin curette and at once placed in 0.1 per cent, dextrose broth,
and incubated for eighteen to twenty-four hours. The identification
of the bacilli which develop in the broth is made by the usual cultural
and agglutination reactions.

2. Isolation of Typhoid Bacilli from the Urine. Typhoid bacilli
have been found in the urine in from 25 to 35 per cent, of the cases
examined. Such urines frequently contain albumin. The organisms
do not as a rule appear until the third week of the disease, conse-
quently their isolation is of comparatively little value diagnostically,
although their recognition is of great importance for the prevention
of secondary cases. The organisms may exist in the urine for a few
weeks after recovery. Rarely they persist for months or very rarely
for years after recovery. Frequently their presence is not mani-
fested by clinical symptoms, but occasionally persistent cystitis may
be caused by their continued growth in the urinary bladder. Usually
the bacilli present in the urine are found in pure culture. Occasionally
colon bacilli are found either in association with typhoid bacilli or
even in pure culture after the typhoid bacilli have disappeared.

3. Isolation of Typhoid Bacilli from Feces. Typhoid bacilli are
usually found in pure culture or nearly pure culture in the blood, and,
if the proper precautions are observed, in the urine as well. In the
feces, on the contrary, they are usually in the minority and their
isolation presents certain difficulties. It has been claimed by many
authorities that typhoid bacilli are not found in the feces in demon-
strable numbers, at least until about the middle of the second week.
Klinger 2 has collected statistics from 812 contact cases which indicate
the danger of infection from feces even before the development of
clinical symptoms.

1 Loc. cit.

2 Public Health Reports, 1911, xxvi, 319.

338 THE ALCALIGENES DYSENTERY TYPHOID

SECONDARY CASES INFECTED FROM PRIMARY CASES.

First week of incubation period 33

Second week of incubation period 150

First week of disease 1S7

Second week of disease 158

Third week of disease 116

Fourth week of disease 59

Fifth week of disease 34

Sixth week of disease 22

Seventh week of disease 14

Eighth week of disease 16

Ninth week of disease 15

The isolation and identification of typhoid bacilli from the feces
is by no means proof that the case under consideration is typhoid fever;
the patient may be a carrier.

Technic of Isolation of Typhoid Bacilli from Feces. A thin uniform
emulsion of feces suspected to contain typhoid bacilli is made in 0.1
per cent, dextrose broth and incubated, if time permits, for one hour
at 37 C.

The emulsion is best made by repeatedly running a rather heavy
platinum needle through the fecal mass to insure a representative
sample. The process is continued until the desired density of bacteria
in the broth tube is attained. Incubation of one hour permits of a
slight development of all the organisms; it particularly acclimatizes
the typhoid bacilli to artificial media. The emulsion is then spread
with a bent sterile glass rod on the surface of Endo medium previously
prepared in large Petri dishes. 1 The Petri dishes after inoculation
are inverted and placed in the incubator at 37 C. and examined eigh-
teen to twenty-four hours later for clear, colorless, transparent colonies
which rarely attain a diameter exceeding 2 mm. These colonies are
transferred to 0.1 per cent, dextrose broth and after incubation for
eighteen to twenty-four hours at 37 C. are mixed with a high potency
antityphoid serum and examined for agglutination.

Rapid Method of Isolating Typhoid Bacilli. 2 It is frequently pos-
sible to identify typhoid bacilli (and paratyphoid and dysentery bacilli
as well) in feces within twenty-four hours by taking advantage of the
microscopic agglutination method with a high potency serum in the
following manner: Endo plates are inoculated as indicated above
and incubated at 37 C. for fifteen to eighteen hours. Typical colonies
are removed entire to small test-tubes containing 1 c.c. of 0.1 per
cent, dextrose broth which have been kept at incubator temperature.

1 For preparation and use of the Endo medium, see page 201.

2 Kendall and Day, Jour. Med. Research, 1911, xx, 95.

TYPHOID BACILLUS 339

Incubation of these infected tubes for one to two hours almost invari-
ably gives sufficient numbers of organisms to make a microscopic
agglutination. A confirmatory cultural diagnosis may be obtained
by the inoculation of small tubes of semi-solid media and milk with
the remainder of the troth culture. This method differs from the one
usually employed merely in the small amount of broth used, which
requires less bacteria to produce turbidity, and in the fact that the
growth is practically continuous from the Endo medium to the tube,
the broth being warmed to the body temperature at the start. Taking
advantage of these factors cuts down the time required for diagnosis
nearly twentv-four hours.

(6) SEROLOGICAL DIAGNOSIS. The blood serum of patients who
have recovered from a typical attack of typhoid fever contains elements
which give specific reactions with the typhoid bacillus or its products;
of these, lysins, agglutinins, opsonins and precipitins have been
carefully studied. The method of fixation of complement and the
ophthalmo reaction have received less attention.

The lysins, which appear early in the course of the disease, dissolve
typhoid bacilli, but not other bacteria, at least in the dilutions ordi-
narily used. It is probable that the lysins not only dissolve typhoid
bacilli in vitro, they destroy the organisms in the blood stream as
well, 1 liberating endotoxins which play a prominent part in the produc-
tion of the febrile reaction.

Agglutinins are formed in the majority of cases, which will clump
typhoid bacilli. The significance of agglutinins in the typhoid complex
is not definitely established.

The opsonic index of the serum of immunized animals and of clinical
cases of typhoid fever in man appears to be increased, but available
methods of measuring the opsonic index do not furnish information
consistent enough to warrant definite conclusions.

The reaction of fixation of complement has been used diagnostically
in a limited number of cases. The technical skill required to elicit
satisfactory results has doubtless interfered with its general application.

The agglutination reaction is by far the most commonly used anti-
body reaction employed in the diagnosis of typhoid fever.

The Widal Reaction. Historical. Gruber and Durham appear
to have first demonstrated that the sera of animals immunized to
typhoid bacilli would agglutinate the typhoid bacilli, even if the

1 Coleman and Buxton, Medical and Surgical Report of Bellevue and Allied Hospitals,
1909-10, iv, 46.

340 THE ALCALIGENES DYSENTERY TYPHOID

serum were diluted many times. Griinbaum and later Widal applied
this principle in the diagnosis of typhoid fever. It is now recognized
that the principle involved is a general one for certain kinds of bac-
teria, and the Gruber-Durham-Gru'nbaum-Widal reaction is used
practically in the diagnosis of several diseases. The sera of such
animals frequently contain agglutinins which are active even in
dilutions of 10^00 or even higher. Specific lysins are also produced,
which in dilutions of igg to nk) will dissolve (and kill) typhoid bacilli.
1. Collection of Blood for the Agglutination Test. Dried blood, blood
serum, blister fluid, or whole blood may be used for this reaction.

(a) Dried Blood. A generous drop of blood is dropped upon a
thin sheet of aluminum or upon clean, glazed paper, and allowed
to dry. The advantages of dried blood are: (1) it is easily obtained
by making a puncture in the ear of the patient and collecting a drop
of blood; (2) it does not lose its agglutinating properties readily;
(3) it is not readily contaminated; and (4) the blood may be
removed quantitatively after it is dried (scaled off), weighed and
then diluted to the desired degree as accurately as blood serum. The
disadvantages are: (1) flies will readily remove a film of dried blood;
and (2) typhoid bacilli are rarely found in blood clots. There is,
however, very little danger of spreading typhoid in this way. In
practice dried blood is diluted with physiological normal saline solu-
tion to a pale rose color, which corresponds to a dilution of 1 to 20.
This dilution is somewhat inaccurate and anemic bloods introduce a
disturbing factor. This method of dilution, however, is sufficiently
accurate for all except unusual cases, and it is a method generally
used in routine board of health examinations.

(b) Blood Serum. A few drops of blood are collected in a capillary
pipette or small test-tube and allowed to clot. The serum is removed
and diluted accurately with salt solution. The advantages are: (1)
the accuracy with which dilution may be made; and (2) the ease
with which serum is obtained. The disadvantages are: (1) that
blood serum is readily contaminated; and (2) it does not keep well,
it deteriorates. Blood serum is the best for accurate work.

(c) Blister Fluid. This possesses no advantages over blood serum.
It is somewhat more difficult to obtain and probably somewhat less
accurate than blood serum.

(d) Whole Blood. Aside from clotting, whole blood is as reliable
as blood serum, so far as accuracy of dilution and potency of agglu-
tinins is concerned. It must be remembered, however, that the red

TYPHOID BACILLUS 341

blood cells appear in the field viewed under the microscope. Fresh
whole blood presents one great disadvantage the fibrin in it may
cause a pseudoagglutination, for the fibrin network that forms as
coagulation proceeds entangles typhoid bacilli in its meshes, giving
the appearance of a true agglutination. Whole blood can be con-
veniently drawn into a blood-counting pipette and diluted accurately
and immediately.

The Culture to be Used. Old stock cultures of typhoid bacilli usually
give the best results. Freshly isolated cultures not infrequently agglu-
tinate less readily than those which have been on artificial media for
some time. The organisms should be grown in 0.1 per cent, dextrose
broth for eighteen hours at 30 to 32 C. It has been found that
typhoid bacilli grown at this temperature agglutinate somewhat
better than those grown at 37 C. Killed cultures are frequently
used, but the results obtained are somewhat less accurate than those
with living cultures. In rare instances it has been found that killed
cultures will agglutinate with typhoid sera at 45 C. when living
cultures fail to agglutinate. Controls must always be made: the
typhoid culture is diluted with an equal volume of salt solution.
Spontaneous agglutination sometimes takes place when no serum is
present. This is shown in the control and at once invalidates the
agglutination which may be obtained with the serum.

Technic of Test. (A) Microscopic Method. Dried blood, blood
serum, blister fluid, or whole blood is diluted 1 to 20 with physiological
salt solution. A loopful of this diluted fluid is mixed intimately with
a loopful of typhoid broth culture on a coverglass and suspended in
a hanging drop slide ringed with vaseline to prevent evaporation.
The final dilution of the blood is 1 to 40 by this procedure. A control
is made using a loopful of salt solution and a loopful of typhoid culture
prepared in the same manner. Both the serum and the control are
kept at room temperature. A preliminary examination should show
actively motile bacteria in the control preparation and usually actively
motile bacteria in the serum preparation. It sometimes happens that
agglutination takes place in the serum preparation almost immediately.
If the preliminary examination is satisfactory, the final examination
is made at the end of an hour. Both preparations are examined and
the controls should show actively motile unclumped organisms. A
positive agglutination is recorded if the control is as stated and the
organisms in the serum preparation are non-motile and gathered
together in clumps with few or no free-swimming bacteria between
the clumps.

342 THE ALCALIGENES DYSENTERY TYPHOID

(B) Macroscopic Method. Various dilutions of serum are placed
in small sterile test-tubes, 1 c.c. in each test-tube. As a routine, a
dilution of 1 to 20 is used, but a series of dilutions up to the limits
of the serum are frequently made. To each tube is added 1 c.c. of a
broth culture of typhoid bacilli. A control is made by adding 1 c.c.
of a broth culture of typhoid bacilli to 1 c.c. of salt solution. These
mixtures are respectively shaken and incubated together with the
control at 37 C. for two hours, then they are placed in the ice-box,
and examined eighteen to twenty-four hours later. A positive agglu-
tination is indicated when the supernatant fluid of the serum typhoid
mixtures is clear, while the control containing no serum remains
uniformly cloudy.

The microscopic method is much more rapid than the macroscopic
method and is sufficiently accurate for ordinary purposes. The macro-
scopic method requires a much longer time, but it is more accurate,
for the dilutions can be made carefully with graduated pipettes.

Discussion. Available statistics show that about 20 per cent, of
typhoid patients exhibit a positive agglutination reaction at the end
of the first week; 60 per cent, at the end of the second week; 80 per
cent, at the end of the third week; and 90 per cent, at the end of the
fourth week. These agglutinins persist; about 75 per cent, of all
patients exhibit a positive agglutination after two months. Occa-
sionally agglutinins may persist for several years. 1 The amount of
agglutination present, as indicated by the degree of dilution which
will still clump typhoid bacilli, has no known relationship to the
severity of the attack. An occasional mild case of typhoid may be
accompanied by the appearance of agglutinins of great potency;
severe attacks may exhibit little or no agglutinin in the blood. Occa-
sionally, agglutinins are not demonstrable in the blood serum of
undoubted cases of typhoid fever. This has been found to be the case
by Moreschi 2 in several cases of chronic leukemia. Moreschi 3 has made
the interesting observation that even the vaccination of these leukemics
with killed cultures of typhoid bacilli may not lead to the development
of agglutinins. In icterus an agglutination is not infrequently encoun-
tered even if the serum is highly diluted. It is very probable that at
least some of these cases are typhoid carriers, having typhoid bacilli in
the gall-bladder. They may be ambulatory cases. It has been claimed

1 An initial negative reaction (first week) followed by a positive reaction is conclusive.
It rules out the possibility of persistent agglutinins from previous cases, and those due
to protective vaccination.

2 Ztschr. f. Immunitatsforsch., 1914, xxi, 410. 3 Loc. cit.

TYPHOID BACILLUS 343

by some observers that the agglutination seen in icteric patients
is due to bile in the blood stream. This, however, has not been
proven. A negative agglutination, when the clinical symptoms suggest
typhoid fever, should suggest the possibility of a paratyphoid infection.

Ophthalmo Reaction. Chantemesse 1 has found that an ophthalmo
reaction may be elicited in typhoid patients similar to that produced
by the introduction of tuberculin in the eye of the tuberculous patient.
Broth cultures of typhoid bacilli are precipitated with alcohol; the
precipitate is dried and pulverized; ^ milligram of the powder is
dissolved in a few drops of sterile saline solution and introduced into
the eye. A transient redness with a flow of tears occurs in normal
individuals; a severe reaction (even accompanied by a serofibrinous
exudate in unusual cases), which reaches its maximum intensity
within twelve hours, is elicited in typhoid patients, and, occasionally,
in individuals who have recovered from the disease. The diagnostic
value of the reaction is as yet undetermined.

Dissemination and Prophylaxis. The disease typhoid fever occurs
only by transmission of typhoid bacilli directly or indirectly from
preexisting cases. The disease is acquired only by the ingestion of
the specific organisms, and infection by any other channel than the
alimentary canal has not so far been satisfactorily demonstrated.

Prophylactic measures, therefore, should begin with the isolation
of the patient and disinfection of all excreta and all utensils which
have been in contact with the patient. The organism may occur in
the fecal discharges of patients before clinical symptoms develop, in
patients recently recovered from the disease, in carriers (which number
about 2 per cent, of all cases diagnosed), and probably in a relatively
few individuals in whom the organism may gain a temporary foothold
without producing symptoms. The bacilli may be transmitted to
others by the hands of those who care for the patients, and the hands
of carriers. Fecal matter containing typhoid bacilli may be trans-
ferred by flies, by water, through milk, and perhaps by vegetables
which are eaten uncooked. The water in which typhoid patients have
bathed is frequently grossly contaminated with the organisms. Rarely,
wells and water supplies are contaminated by urinary typhoid car-
riers, in which event the colon bacillus, which is ordinarily relied upon
for evidence of contamination, may be absent. A thorough disin-
fection of excreta including urine will prevent spread of the disease
from known cases.

1 IV. International Cong, of Demog. and Hyg., Berlin, September 26, 1907.

344 THE ALCALIGENES DYSENTERY TYPHOID

THE PARATYPHOID GROUP.

There is a group of closely related bacilli which exhibit cultural
and pathogenic characters intermediate between those of the typhoid,
dysentery and colon groups of bacteria, respectively. These organ-
isms are variously known as the hog cholera, Salmonella, Gartner,
enteritidis, intermediate, paracolon or paratyphoid group.

Smith and Salmon 1 isolated the type organism of the group from
the intestinal contents of swine infected with hog cholera. They
named their organism the hog cholera bacillus. 2 Three years later
Gartner 3 described an organism, B. enteritidis, recovered by him both
from the spleen and blood of a fatal case of meat-poisoning, and from
the suspected meat (beef) itself. Numerous epidemics of meat poison-
ing 4 have been studied bacteriologically during the years following
Gartner's discovery, and very similar, if not identical, bacilli have
been recovered from many of the patients.

In 1893 Smith and Moore 5 made the important observation that
organisms culturally indistinguishable from the hog cholera bacillus
could be isolated not infrequently from the intestinal contents of
normal cattle, swine, sheep, cats and dogs. The significance of this
discovery from the view-point of meat poisoning was not understood
at that time.

In 1896 Achard and Bensaude 6 described paratyphoid fever and
outlined the essential clinical and bacteriological diagnostic differences
between this disease and typhoid fever. They obtained paratyphoid
bacilli from the urine and blood stream of several cases, and recovered
the organism from a secondary purulent arthritis in one of them as
as well. Schottmuller 7 also obtained cultures of paratyphoid bacilli
both from the feces and the blood stream of several cases of para-
typhoid fever. Brion and Kayser 8 separated these organisms into two
types: B. paratyphosus alpha, which produced a slight permanent
acidity in litmus milk and gave an "invisible" growth on potato

1 Ann. Rep. United States Bur. Animal Ind., 1885, vol. ii.

2 A year earlier Klein (Virchows Arch., 1884, xcv, 468) obtained a bacillus from
diseased swine which he regarded as the causative factor of hog cholera, but his organism
produced spores, which at once distinguished it from the paratyphoid type. Neither the
Klein bacillus nor the Smith-Salmon bacillus cause hog cholera; a filterable virus is the
probable infecting agent.

3 Correspondz.-Blatt des allgem. arztl. Vereins von Thuringen, 1888, No. 9.

4 Not to be confused with botulismus (see B. botulinus).

5 Additional investigations concerning swine diseases, Washington, D. C., 1893.

6 Soc. Med. des Hop. de Paris, 1896, 3d Sens, xiii, 679.

7 Deutsch. med. Wchnschr., 1900, p. 511.

8 Munchen. med. Wchnschr., 1902, p. 611.

THE PARATYPHOID GROUP 345

like the typhoid bacillus; and B. paratyphosus beta, which produced
an initial acidity in litmus milk followed by a progressively alkaline
reaction. These observations, both clinical and bacteriological, have
been confirmed by later investigations.

Morphology. The members of the intermediate group are indistin-
guishable morphologically. They are rod-shaped bacilli with rounded
ends, measuring from 0.8 to 1 micron in diameter, and 1.5 to 3.5
microns in length, occurring singly or in pairs, seldom in chains. In
actively-growing cultures the organisms may be short, almost ovoid.
In old cultures the organisms may be elongated; filamentous forms
are more commonly seen in old gelatin cultures. The members of the
group are actively motile and possess from four to twelve peritrichic
flagella. Motility is greater in dextrose broth than in plain broth;
this is particularly the case in young cultures. The organisms form
no spores and appear to possess no capsules. They stain readily with
ordinary anilin dyes; occasionally organisms from cultures several
days old exhibit a tendency toward bipolar staining. They are Gram-
negative.

Isolation and Culture. The organisms of the paratyphoid group grow
readily upon ordinary artificial media, B. paratyphosus alpha somewhat
less luxuriantly than the remaining members. The colonies produced
on agar after eighteen hours' incubation at 37 C. resemble those
of the typhoid-dysentery group small, round, and transparent
measuring from 1 to 3 mm. in diameter. On Endo medium the colonies,
like those of B. typhosus and the dysentery bacilli, are clear and
colorless and somewhat smaller than those developing upon plain
agar. They usually measure from 0.75 to 2 mm. in diameter. The
organisms grow well in gelatin, but do not cause liquefaction. They
produce acid and gas in dextrose and mannite; lactose and saccharose
are not fermented.

Milk. Plain milk is not coagulated. All the members of the group
except B. paratyphosus alpha cause a slow change in this medium,
which becomes thin, brownish, and almost opalescent after two or
more weeks' incubation. In litmus milk the cream ring is colored
a deep blue-green, which is so constant as to be suggestive diagnos-
tically. B. paratyphosus alpha produces a slight acidity which is
permanent; the milk assumes a lilac color. B. paratyphosus beta
and other members of the group produce a transient acidity 1 which

1 For an explanation of the phenomenon, see page 222.

346 THE ALCALIGENES DYSENTERY TYPHOID

is followed by a progressive alkalinity, associated with the liberation
of small amounts of ammonia. 1

All members of the intermediate group produce considerable tur-
bidity in plain and sugar broths. A pellicle may develop in plain
broth after several days' incubation. Potato: B. paratyphosus
alpha grows much like the typhoid bacillus on potato; the growth is
nearly invisible on acid potato, but comparatively luxuriant. On
alkaline potato the growth is brownish. B. paratyphosus beta pro-
duces a brownish growth even on slightly acid potato, which resembles
that characteristic of B. coli.

The members of the intermediate group are all aerobic, facultatively
anaerobic. The minimum temperature of growth is about 6 to 8 C.,
the optimum 37 C., and growth ceases at approximately 44 C. The
resistance of the members of the intermediate group to environmental
conditions, drying and to chemicals is similar to that of the typhoid
bacillus. They are, however, somewhat more resistant to heat; an
exposure of fifteen minutes at 70 C. or of five minutes at 75 C.,
kills the bacilli. This is a point of importance in meats infected
with the organisms; temperatures lower than 75 C. in the centre of
the meat can not be relied upon to remove danger of infection. Higher
temperatures, 100 C., are preferable to remove all danger from the
poisonous substances of the bacilli, which are not destroyed by gastro-
intestinal digestion.

Products of Growth. (a) Chemical. Paratyphoid bacilli are rather
more active proteolytically than typhoid and dysentery bacilli, but
they produce neither phenols nor indol. 2 Dextrose and mannite are
fermented with the formation of carbon dioxide and hydrogen, lactic
acid, and smaller amounts of acetic and formic acids. Lactose and
saccharose are not fermented. Numerous attempts have been made
to classify the paratyphoid bacilli into several varieties upon the basis
of the fermentation of carbohydrates other than those mentioned
above, but the lack of agreement has proved an insurmountable
obstacle to their general acceptance.

(b) Enzymes. The members of the paratyphoid group do not pro-
duce soluble proteolytic ferments, and they do not liquefy coagulated
blood serum, gelatin, fibrin or egg albumen. Neither lipolytic nor
amylolytic enzymes have been demonstrated in cultures of these
organisms.

1 Kendall, Day and Walker, Jour. Am. Chem. Soc., 1914, xxxvi, 1943.

2 Ibid., 1913, xxxv, 1221.

THE PARATYPHOID GROUP 34?

(c) Toxins. Soluble toxins have not been demonstrated in cul-
tures of paratyphoid bacilli. Cathcart 1 and Franchetti 2 have shown
that minute amounts of autoly sates of the organisms are rapidly* fatal
to rabbits and other small laboratory animals. According to Cath-
cart, 3 the poisonous substance (endotoxin) liberated from the organ-
isms during autolysis is relatively thermostabile ; a brief exposure of
it to 100 C. does not completely destroy its potency.

Classification and Identification of the Paratyphoid Group. It is pos-
sible to divide the Paratyphoid Group into two distinct types by their
reaction in milk: the alpha type, of which several strains have been
described, differing somewhat in their serological reactions; and the
beta type. The former appears to be limited to man, but the latter
comprises organisms which are rather widely distributed not only
in man but in the lower animals as well. The better known strains
of the beta type comprise not only B. paratyphosus beta, B. enteritidis
and the hog cholera bacillus (B. choleras suis, B. suipestifer) mentioned
above, but B. psittacosis, obtained from infectious enteritis of parrots,
which produces a pneumonic infection in man. 4 B. icteroides, San-
arelli, originally supposed to cause yellow fever, but now known to
be indistinguishable from the hog cholera bacillus, the Danysz bacillus
of rat plague, and B. typhi murium, Loffler, obtained from epizootics
of rodents, B. sertrycke, de Nobele, and B. moorseele, van Ermengem,
from epidemics of meat poisoning, and B. morbificans bo vis, Basenau,
isolated from a diseased cow, all belong to the same group. They
possess in common cultural characteristics which differ somewhat
quantitatively, but not qualitatively. Bainbridge and O'Brien 5 have
attempted to classify the organisms by agglutination and absorption
tests; they recognize four groups as follows: (1) B. paratyphosus
alpha; (2) B. paratyphosus beta; (3) B. suipestifer (hog cholera bacil-
lus), including B. psittacosis, B. sertrycke and some strains of B.
typhi murium; (4) B. enteritidis, including the Danysz bacillus, B.
morbificans bovis, and some strains of B. typhi murium. This clas-
sification, if substantiated, possesses the advantage of separating
those organisms which cause paratyphoid fever, the alpha and beta
types, from the bacilli more commonly associated with the lower

1 Jour. Hyg., 1906, vi, 112.

2 Ztschr. f. Hyg., 1908, Ix, 127.

3 Loc. cit.

4 Nocard, Conseil d'hygiene pub. et Salubrite du Dept. du Seine, S6ance, March 24,
1893.

5 Jour. Hyg., 1911, xi, 68.

348 THE ALCALIGENES DYSENTERY TYPHOID

animals, of which the hog cholera bacillus and B. enteritidis are the
types. This classification has not been universally accepted, how-
ever. Doubtless the multiplicity of strains which have received the
same name has led to confusion in standard type organisms which are
especially essential in this line of investigation. It is not an assured
fact that the paratyphoid bacilli, alpha and beta, are restricted to the
production of paratyphoid fever in man; nor can it be stated definitely
that B. enteritidis and the hog cholera bacillus consistently cause
meat poisoning. Available information suggests that occasionally
the choleraic symptoms of meat poisoning may be elicited by para-
typhoid bacilli, and that the symptoms of paratyphoid fever may
follow infection with B. enteritidis or B. suipestifer.

Pathogenesis. Animal. The members of the Paratyphoid Group
are, as a rule, very pathogenic for small laboratory animals. The intra-
peritoneal injection of very minute amounts of bacilli usually causes
acute death in guinea-pigs and mice. Rats are somewhat more
resistant. B. typhi murium and other a rat viruses" produce a fatal
enteritis in mice and rats ; the bacilli are present not only in the intes-
tinal contents, they may be obtained from the tissues and organs post-
mortem as well. Bacilli belonging to the Paratyphoid Group have
been isolated from epizootics and sporadic cases of enteritis in cattle,
parrots, and rodents. The organisms appear to be widely distributed
among the lower animals.

Human. Three types of disease are produced in man by the bac-
teria of the paratyphoid group: (a) meat poisoning: the symptoms
are choleraic in character, and they may be severe enough to be
confused with true cholera; 1 infection usually follows the ingest ion
of imperfectly cooked beef or pork contaminated with B. enteritidis
or the hog cholera bacillus. Somewhat similar symptoms have
resulted from the accidental ingestion of the "rat virus" of Danysz
and others; 2 (b) paratyphoid fever, a disease clinically resembling
mild typhoid fever, usually caused by B. paratyphosus alpha or B.
paratyphosus beta; (c) a rare type of disease, pneumonic in character,
produced by B. psittacosis, which produces an epizootic disease among
parrots.

(a) Meat Poisoning. The disease is more prevalent in summer and
fall than it is in winter and spring, probably due in part to decreased

1 Hetsch, Klin. Jahrb., 1907, xvi, 267.

2 Mayer, Miinchen. med. Wchnschr., 1906, No. 47; Shibayama, Miinchen. med.
Wchnschr., 1907, 979.

THE PARATYPHOID GROUP 349

efficiency of refrigeration of meats in the warmer months. The incu-
bation period may be as brief as four to six hours, or as long as twenty-
four to seventy-two hours after ingestion of the infected food. The
initial symptoms are usually a severe headache and chill, rapidly
followed by acute gastro-intestinal disturbances, dizziness, nausea
and vomiting, abdominal pain and diarrhea. Nervous symptoms and
marked restlessness are characteristic of the severe and fatal cases.
Usually the symptoms and fever abate within a week; they may
persist for several weeks. The mortality is, as a rule, low, averaging
from 1 to 2 per cent. The conspicuous lesion observed at autopsy
is an intense hyperemia of the gastro-intestinal mucosa, usually with-
out noteworthy involvement of Peyer's patches. Fatty degeneration
of the liver is common. Bacilli (usually B. enteritidis or B. cholerae
suis, 1 less commonly B. paratyphosus beta) may be isolated from the
feces and blood stream in many of the acute cases during the first
few days of the disease. They are almost invariably recovered from
the heart blood and spleen at autopsy. Serum reactions, especially
specific agglutinins, may be demonstrated at the end of the first week
in many but not all cases.

An epidemic of meat poisoning is characterized by the sudden, prac-
tically simultaneous onset of symptoms in those who have eaten the
contaminated food, and the limitation of the disease to the primary
cases. Secondary infection is uncommon. It should be remembered
that not all epidemics of meat poisoning are caused by members of
the paratyphoid group of bacteria.

Distribution of Organisms. The hog cholera bacillus (B. cholera?
suis, B. suipestifer) is frequently found in the intestinal tracts of
swine, rats and mice; probably somewhat less commonly in cattle.
B. enteritidis is a frequent inhabitant of the intestinal contents of
rats and mice, and relatively uncommon in healthy cattle. 2 It is
suspected that a postmortem infection of beef is more common than
an antemortem invasion; this is reasonably suggested by the wide
distribution of rats and mice in slaughter houses. The organisms
possess the somewhat unusual property of rapidly diffusing them-
selves through the substance of meat after they have been distributed
on the surface of it by careless handling. Unless infected meat is
thoroughly cooked, the organisms are not killed, and they may not
be even weakened if the degree of heat and time of exposure is insuffi-

1 Bainbridge, Lancet, March 16, 23, 30, 1912.

2 Ibid.

350 THE ALCALIGENES DYSENTERY TYPHOID

cient. The endotoxins of the bacilli, furthermore, are relatively
thermostabile. Thorough cooking of such meat is essential to insure
safety.

(b) Paratyphoid Fewr. Bacteriologically, paratyphoid fever may
be caused either by B. paratyphosus alpha or B. paratyphosus beta.
Clinically there is little or no difference between the two infections.
According to Bainbridge, 1 paratyphoid fever in Asia, particularly
in India, is more frequently an infection with the alpha organism; in
Europe the beta organism is much more frequently reported. Both
types are found in the United States. 2 The organisms are occasionally
found in the intestinal contents and feces of young children and adults
who give no history of infection.

The incubation period of paratyphoid fever varies from eight to
twenty days; the average is about two weeks. The onset is gradual;
the usual prodromal symptoms are severe head- and backache, malaise
and anorexia. Bronchitis and sore throat are common. There may
be an initial chill, then the temperature rises rather rapidly to a maxi-
mum of 103 to 105 C.; after the fifth to the seventh day it falls
slowly; it is normal by the end of the second week. Rose spots are
occasionally seen early in the disease. Less commonly acute gastro-
enteric symptoms, resembling those of meat poisoning, complicate the
clinical picture. Paratyphoid fever is a bacteremia, very similar to
typhoid fever in this respect. The mortality is low, averaging from
1 to 2 per cent, of all cases. The lesions observed postmortem are
intense hyperemia of the gastro-intestinal tract, usually with superficial
ulcerations in the ileum and cecum, not necessarily, however, involv-
ing Peyer's patches. Acute splenic tumor is usually not a feature of
paratyphoid infections. The bacilli may be isolated from the heart
blood and visceral organs.

Bacterial Diagnosis. (a) Isolation of Bacilli. Blood cultures made
during the first week are frequently positive. The organisms are
usually present in the feces, occasionally in the urine. The identifi-
cation of the bacilli depends upon the cultural characters outlined
above; gas production in dextrose and mannite, no liquefaction of
gelatin, and a permanent acidity in litmus milk (alpha type) or a
transient acidity followed by a progressively alkaline reaction in this

1 Loc. cit.

2 Gwyn, Bull. Johns Hopkins Hospital, 1898, vol. ix. Gushing, ibid., 1900, vol. xi;
Buxton and Coleman, Proc. Path. Soc. New York, February, 1902; Proescher and
Roddy, Jour. Am. Med. Assn., 1909, lii, No. 6; Kendall, Bagg and Day, Boston Med.
and Surg. Jour., 1913, clxix, 741; Kendall and Day, ibid., 1913, clxix, 753.

THE PARATYPHOID GROUP 351

medium (beta type). Isolation from the feces is made upon Endo-
plates in the same manner that dysentery and typhoid bacilli are
obtained. The final diagnosis depends upon the agglutination of the
bacilli with specific agglutinating sera of high potency. 1

(b) Serological. As a routine measure the diagnosis of paratyphoid
fever by the agglutination test is unreliable. Not infrequently the
blood serum of a patient agglutinates typhoid bacilli in dilutions
approaching those ultimate for the homologous organism. The para-
typhoid bacilli and B. typhosus possess in common group agglutinins
which greatly vitiate the value of the test. The same objection does
not hold for the diagnosis of typhoid fever by the agglutination
reaction, however.

The isolation of B. paratyphosus (alpha or beta) from the blood
stream during life, or from the internal organs at autopsy is the only
reliable method of diagnosis. Carriers are not uncommon, and like
typhoid bacillus carriers the organisms frequently remain in the
gall-bladder, consequently isolation of the bacilli from feces does not
necessarily establish a correct clinical diagnosis. Paratyphoid bacilli
have been isolated occasionally from gall-stones and from cases of
cholecystitis, particularly in women.

SUMMARY.

THE MORE IMPORTANT DIFFERENTIAL DETAILS OF PARATYPHOID FEVER AND
OF MEAT POISONING.

Meat Poisoning. Paratyphoid Fever.

Organism .... Hog cholera bacillus. B. paratyphosus alpha.

B. enteritidis. B. paratyphosus beta.

Habitat of organism . Intestinal canal of lower ani- Chiefly intestinal tract -

mals chiefly: hog cholera of man.
in swine, enteriditis com-
mon in rodents.

Mode of infection . . Usually contaminated meat Usually human bacilli

(human carriers rare). carriers.

Incubation period . . Six to forty-eight hours. Eight to twenty days.

Symptoms .... Choleraic. Typhoidal.

Pneumonic Infection with B. Psittacosis. B. psittacosis causes a
fatal enteritis in parrots, and it has been noticed, particularly in
France, that coincidently with enteric disease in parrots a pneumonic
infection has appeared in those associated with them. The disease
in man presents no definite clinical features which would differentiate
it from typhoid fever complicated by pneumonia. Tke incubation

1 Sera that will agglutinate homologous strains in dilutions of 1 to 40,000 are readily
prepared; such sera in dilutions of 1 to 10,000 may be regarded as specific for the identi-
fication of members of the group, if typical agglutination occurs.

352 THE ALCALIGENES-DYSENTERY TYPHOID

period varies from five days to three weeks, usually, however, less
than ten days. The onset is gradual in some cases, like typhoid, but
it may be abrupt with an initial chill, as in pneumonia. The spleen
is enlarged, but rose spots are rarely found. The mortality varies;
it may be as high as 30 per cent. The postmortem lesions have not
been established. In one case the bacillus was isolated from the heart's
blood postmortem. Specific agglutinins in the patient's blood serum
have not been satisfactorily studied, and the disease as a clinical
entity is yet to be defined. The principal evidence of the causative
relationship of B. psittacosis to the disease rests at present upon the
occasional household epidemics following closely upon the presence of
a diseased parrot.

Immunity and Immunization to Paratyphoid Infection. The duration
of immunity following recovery from an attack of paratyphoid fever
or of meat poisoning is as yet undetermined. The brilliant results
of protective immunization against typhoid fever with vaccines or
residues of the typhoid bacillus have led to similar vaccination against
paratyphoid infection with polyvalent vaccines composed of the
principal strains of the paratyphoid group. Combined protective
vaccination against typhoid and paratyphoid by the use of com-
pound vaccines has also been attempted. The efficiency of the
immunization can not be stated at the present time because statistics
are unavailable.

Dissemination and Prophylaxis. Paratyphoid fever appears to be
spread by mild unrecognized cases, by carriers, and by the occasional
transmission of bacilli through food, water or milk. Flies may also
be a factor in the dissemination of the organisms. Meat poisoning
is chiefly disseminated by infected meats, more frequently that of
cattle or swine. The customary precautions appropriate for excremen-
titious diseases, including the restriction of carriers, may be con-
fidently relied upon to prevent the spread of paratyphoid fever.
Thorough cooking will largely reduce the occasional danger from
contaminated meats.

CHAPTER XVII.
THE COLI CLOACA PROTEUS GROUP.

BACILLUS COLL

Historical. Bacillus coli was isolated in pure culture from the
feces of infants, and its important cultural characters determined by
Escherich in 1886. 1 It is very probable, as Escherich suggested, 2
that Emmerich's B. neapolitanus, Brieger's "propionic acid bacillus,"
and Frankel's bacilli 3 are identical with the colon bacillus.

Morphology. Bacillus coli is a rod-shaped organism which varies
in shape from oval organisms resembling cocci to bacilli of moderate
length. The organism varies in size from 0.5 to 0.8 micron in dia-
meter and from 1 to 3 microns in length. The bacilli occur singly and
in pairs; in older cultures short chains and elongated organisms are
frequently observed. The ends are distinctly rounded. Motility is
variable; many strains are non-motile except during the earlier hours
of growth. Young cultures on gelatin are said to exhibit motility
when older growths even in the same medium are motionless except
for Brownian movement. Very commonly only a very few organisms
in a microscopic field exhibit motion, the remainder being without
movement. Four to eight peritrichic flagella are commonly attached
to each bacillus; less frequently as many as twelve may be demon-
strated. The flagella are somewhat shorter than those of the typhoid
bacillus and they are more difficult to stain. Bacillus coli forms no
spores nor capsules. It stains readily with the ordinary anilin dyes,
ard it is uniformly Gram-negative.

Isolation and Culture. The colon bacillus grows readily on the ordi-
nary media; the superficial colonies on agar plates are clear and color-
less and attain a diameter of from 2 to 5 mm. after eiiteen hours'
incubation at 37 9 C. If the surface of the medium is mowt the edges
of the colonies are somewhat irregular in outline; on dry surfaces the
colonies are round and slightly convex in section. Viewed by trans-

1 Die Darmbakterien des Sauglings, Stuttgart, 1886, 63J

2 Loc. cit., 73, 74.

3 Deutsch. med. Wchnschr., 1885, Nos. 34 and 35.
23

354 THE COL/ CLOACA PROTEUS GROUP

mitted light the growths are yellowish-brown; by reflected light they
are colorless. Colonies on gelatin develop more slowly and become
somewhat brownish in color. The medium is not liquefied. Rapid
development occurs in plain and sugar broths. A heavy, brownish
spreading growth occurs on the surface of slanted potato.

Bacillus coli is an aerobic, facultatively anaerobic organism which
grows best at 37 C. Growth ceases below 8 to 10 C., and above
43 to 45 C. An exposure of fifteen minutes at 75 C. kills them. In
general the colon bacillus is somewhat more resistant to physical and
chemical agents than the typhoid bacillus.

Products of Growth. (a) Chemical. Bacillus coli produces indol
from tryptophan in sugar-free media, and phenolic bodies from

FIG. 48. Bacillus coli flagella. X 1500. (Kplle and Hetsch.)

tyrosine under the same conditions. Hydrogen sulphide and ammonia,
the latter resulting largely from deaminization of proteins and
protein derivatives, are also produced in considerable amounts in
media containing no utilizable carbohydrates. 1 Similar products
may be formed in the intestinal tract under certain conditions. The
addition of utilizable carbohydrates to protein media changes the
character of the products of metabolism in a noteworthy manner.
Under these conditions the protein constituents of the media are
practically unchanged; the sugars are fermented with the production
of carbon dioxide and hydrogen, 2 lactic acid and smaller amounts of

1 Kendall, Day and Walker, Jour. Am. Chem. Soc., 1913, xxxv, 1228.

2 In the proportion H : CO 2 = f . Theobald Smith, The Fermentation Tube. The
Wilder Quarter Century Book, 1893, p. 202. Very exact determinations of the gaseous
products of fermentation of B. coli have been made by Harden and Walpole, Proc. Roy.
Soc., 1906, 77, 399.

BACILLUS CO LI 355

acetic acid and formic acid. Dextrose, lactose and mannite are thus
fermented; saccharose is not decomposed by the strains of the colon
bacillus commonly found in the intestinal tract. Occasionally a sac-
charose-fermenting strain is encountered in the feces. 1

The reactions of the colon bacillus in milk are variable; typical
strains produce enough acid from the fermentation of the lactose to
cause an acid coagulation in one to three days at 37 C. Neutraliza-
tion of the acid by alkali redissolves the coagulum and the medium
resumes its normal appearance. Occasional strains do not cause
coagulation even after boiling the milk. 2 Gas is not produced in
appreciable amounts in milk by B. coli, and the organism leaves the
milk proteins practically intact even after prolonged incubation

FIG. 49. Bacillus coli, broth culture.

the carbohydrate constituents alone 'are acted upon. 3 Coagulation
does not as a general rule occur in litmus milk, but boiling the medium
usually causes rapid clotting. The ordinary litmus of commerce
contains considerable amounts of calcium carbonate. This may
neutralize seme of the acid products of fermentation, reducing the
acidity below the coagulation point. This explanation does not
account for the same phenomenon in milk colored with pure litmus
or azolitmin. Gelatin is not liquefied by B. coli. Nitrates are reduced
to nitrites.

(6) Enzymes. Soluble proteolytic and lipolytic enzymes have not

been detected in cultures of Bacillus coli. Buxton 4 has demonstrated

1 Theobald Smith, Am. Jour. Med. Sc M September, 1895.

2 Ibid., Fermentation Tube, p. 201.

3 Kendall, Day and Walker, Jour. Am. Chem. Soc., 1914, xxxvii, 1945,
<Am. Med., 1903, vi, 137.

356 THE COLI CLOACAE PROTEUS GROUP

both a maltase and a lactase in maltose and lactose cultures of the
organism respectively. The investigations of Franzen and Stuppuhn 1
would suggest that the liberation of gas in sugar broth cultures of
B. coli and other aerogenic bacteria depends upon the production
of formic acid from the carbohydrate and its subsequent decomposi-
tion into carbon dioxide and hydrogen by the action of an enzyme,
formiase, in accordance with the equation H.COOH = CO 2 + H 2 .

(c) Toxins. Bacillus coli does not produce a soluble toxin. The
injection of killej cultures into laboratory animals frequently causes
death; if large amounts are introduced intravenously into rabbits
there is usuaUy a lowering of the body temperature, diarrhea, collapse
and death even within three hours. 2 If the animals survive for a
longer time a purulent peritonitis may develop. Living cultures of
colon bacilli derived from inflammatory processes in man are gen-
erally virulent for guinea-pigs. Old stock cultures ark less virulent
as a rule. The symptoms of toxemia which are exhibited by labor-
atory animals following the injection of colon bacilli are probably
caused by the liberation of endotoxins from the bacilli.

Pathogenesis. The colon bacillus is a normal inhabitant of the
intestinal tracts of man and the higher animals. Ordinarily it is a
harmless parasite, but it may become invasive if conditions arise
which weaken the intestinal mucosa. In peritonitis, purulent per-
forative appendicitis, angiocholitis, and even in occasional cases of
pancreatitis the organism is frequently isolated, either in pure culture
or in association with other bacteria, as streptococci, typhoid bacilli,
or staphylococci. It is difficult to determine with precision the part
played' by Bacillus coli in these conditions. Occasional cases of enter-
itis are encountered which appear to be caused by this organism,
other bacteria having been ruled out. The careful studies of Coleman
and Hastings 3 are of great importance in this connection. They
isolated colon bacilli from the blood stream in a small series of cases
which presented symptoms indistinguishable from those of typhoid
fever. No typhoid bacilli were ever found in these patients, and no
specific agglutinins for the typhoid bacillus were demonstrable.
Specific agglutinins for the homologous strains of B. coli persisted
until recovery. Cystitis and pyelonephritis, particularly the former,
are frequently found to be a pure colon infection. B. coli is occa-

1 Ztschr. f. physiol. Chem., 1912, Ixxvii, 129.

2 Escherich, Fort. d. Med., 1885, 521.

3 Med. and Surg. Report of Bellevue and Allied Hospitals of the City of New York,
1909-1910, iv, 56.

BACILLUS COLI 357

sionally isolated from the centre of gall-stones; it is surmised that
the organism, or clusters of them, act as nuclei around which the
cholesterin is gradually deposited. Colon bacilli have been isolated
in rare instances from purulent cerebrospinal fluids, and they may
cause bronchopneumonia. Perirectal abscesses also may contain pure
cultures of colon bacilli.

Immunity and Immunization. The constant occurrence of B. coli in
large numbers in the normal intestinal tract is an index of the rela-
tive immunity of man to infection with this organiaji. Occasionally
very small numbers of bacilli may gain entrance to the tissues, par-
ticularly in young children. The blood serum usually contains agglu-
tinins in small amounts for the organism. In practice no attempt is
made to increase the immunity to colon bacilli, except in cases of
cystitis or other local infection. Vaccines of the homologous strain
of B. coli are occasionally administered in such instances. The results
have been variously interpreted.

Bacteriological Diagnosis. The methods of isolation, identification
and significance of B. coli in water supplies will be discussed in the
chapter on water. Isolation of colon bacilli from the intestinal con-
tents or feces is readily accomplished by plating methods. The
organisms far outnumber any others normally present, and even in
severe diarrheal disorders colon bacilli do not entirely disappear.
Prolonged starvation does not eliminate B. coli from the intestinal
canal. 1 The morphology and staining reactions are not distinctive.
Plating methods principle involved: lactose agar, containing lit-
mus or decolorized fuchsin (Endo medium) as an indicator is infected
with material suspected to contain B. coli. The organism ferments
the lactose with the production of acid; the acid changes the color
of the indicator immediately surrounding the colon bacilli, red if
litmus is used, pink if fuchsin is employed. The red colonies are
inoculated into broth and incubated to obtain sufficient organisms
for their identification by cultural methods.

Cultural Identification. A Gram-negative bacillus which produces
gas in dextrose, lactose and mannite (optionally in saccharose), coagu-
lates but does not 'peptonize milk, does not liquefy gelatin, and is
without action upon starches is Bacillus coli.

1 At the end of thirty-one days' abstinence from all food, typical colon bacilli were
present in the lower part of the large intestine. Kendall, Observations upon the Bacterial
Intestinal Flora of a Starving Man, Publication No. 203 of the Carnegie Institute of
Washington, 1915, p. 232. This experiment emphasizes the fallacy of "starving out"
intestinal bacteria by withdrawing food.

358 THE COLI CLOACA PROTEUS GROUP

BACILLUS CLOAOffi.

Bacillus cloacae was isolated from sewage and polluted water by
Jordan. 1 The organism appears to be relatively abundant some years
and comparatively uncommon other years. When it is abundant in
sewage it is found occasionally in the intestinal tract of man.

Morphology. The bacillus is of moderate size, measuring from 0.6
to 0.8 micron in diameter and from 1 to 2 microns in length. It
occurs singly or in pairs, uncommonly in short chains. Young cultures
exhibit motility, and the organisms possess peritrichic flagella. No
spores or capsules have been demonstrated. Ordinary anilin dyes
color the bacilli readily, and they are Gram-negative.

Isolation and Culture. The colonies on agar plates after eighteen
hours' incubation are round, clear and colorless, and measure from
1 to 3 mm. in diameter. There is nothing distinctive in the appear-
ance of the growths.

Products of Growth. (a) Chemical. Indol, phenol, hydrogen sul-
phide and ammonia are produced in sugar-free broth. The ammonia
production is greater than that characteristic of B. coli and less than
that ordinarily produced by B. proteus. 2 Acid and gas are produced
in dextrose, lactose, saccharose and mannite broths. The gas ratio
is somewhat variable, but distinctive ; the proportion of carbon dioxide
to hydrogen is greater than that produced by other closely-related
bacteria. 3 The action of the organism upon lactose is slow, and less
gas is produced from this sugar. The amount of gas produced from
dextrose and saccharose is greater than that produced by other aero-
genie members of the paratyphoid-pro teus group. B. cloacae forms
but little acid from the fermentation of sugars, and after one to three
days the reaction, even in sugar broth, becomes alkaline, due to the
exhaustion of the sugar and the subsequent decomposition of the
protein constituents of the broth. 4 Indol and other products of putre-
faction are formed as soon as the sugar is exhausted.

Milk is coagulated and slowly peptonized. Freshly isolated cultures
usually liquefy gelatin, but this property is lost after prolonged artifi-
cial cultivation.

The organism is ordinarily non-pathogenic for man.

1 Annual Report of Massachusetts State Board of Health, 1890, p. 836.

2 Kendall, Day and Walker, Jour. Am. Chem. Soc., 1913, xxxv, 1230.
3 H:CO 2 = - i Theobald Smith, Fermentation Tube, 1893, p. 215.
4 Kendall, Day and Walker, loc. cit.

BACILLUS PROTEUS GROUP 359

BACILLUS PROTEUS GROUP.

Synonyms. Proteus vulgaris, Proteus mirabilis, Proteus Zenkeri,
Proteus Zopfii, Proteus fluorescens.

Historical. The proteus group comprises several closely-related
bacilli found commonly in soil, in water rich in organic matter, as
sewage, in human feces, and associated with the decay of organic
matter. The important members of the group were first isolated
in pure culture and described by Hauser. 1

Morphology. The proteus bacilli are rod-shaped organisms of vari-
able length which occur singly and in pairs as a rule; less commonly
they remain adherent in short chains. The size of individual cells
varies considerably, even in the same culture. The limits of varia-
tion are comprised within the following dimensions: diameter from
0.6 to 0.8 micron, length from 1.0 to 3.5 microns. Proteus bacilli
are actively motile and possess a large number of peritrichic flagella 2
which are frequently seen as a tangled filamentous mass surrounding
each individual cell. 3 Special staining methods are required for the
demonstration of these flagella. The organisms produce no spores
and form no capsules. They stain with ordinary anilin dyes, but
somewhat faintly, and they are Gram-negative.

Isolation and Culture. The members of the proteus group develop
rapidly on gelatin at room temperature; the organisms typically
liquefy the medium with great rapidity. Some strains liquefy gelatin
but slightly or even not at all. The colonies of rapidly liquefying
strains in 5 per cent, gelatin are^ frequently very characteristic; the
organisms tend to remain adherent, forming masses of bacilli which
slowly move around in an area of liquefied gelatin. Hauser 4 recognized
four types of proteus bacilli classified according to their ability to
liquefy gelatin: Proteus vulgaris liquefies gelatin rapidly; Proteus
mirabilis liquefies gelatin slowly; Proteus zenkeri and Proteus zopfii
do not liquefy this medium. The latter, Proteus zopfii, exhibits
negative geotropism on slanted solid media. It is now recognized
that cultures of B. proteus may gradually lose their gelatin-liquefying
power after prolonged cultivation, so that a cultural transition from
B. proteus to B. zenkeri may be observed in the laboratory. A dis-

1 Ueber Faulnisbakterien und deren Beziehungen zur Septikamie, Leipzig, 1885.

2 Zettnow, Centralbl. f. Bakt., 1891, x, 689.

3 Massea (Centralbl. f. Bakt., 1891, ix, 106) states that young bacilli may possess
from 60 to 100 flagella.

4 Loc. cit.

360

THE COL/ CLOACA PROTEUS GROUP

tinction between the three types is no longer made. It is not deter-
mined whether B. zopfii is a separate variety of B. proteus.

The organisms grow vigorously in milk, causing slight acidification
and peptonization. The development in broth is equally vigorous;
acid and gas are produced in dextrose and saccharose broths. 1 Neither
acid nor gas is formed in lactose broth. 2

Proteus bacilli grow slowly at C. 3 and at temperatures not
exceeding 43 to 45 C. The optimum temperature is about 25 C.
but development is rapid at 37 C. Strains obtained from putrefying
organic matter are tolerant of considerable degrees of alkalinity 4
and acidity; 5 those from the human body are somewhat less tolerant.
The growth of B. proteus at low temperatures is of considerable prac-

FIG. 50. Bacillus proteus, flagella stain. X 1500. (Gunther.)

tical importance; several cases of ptomain poisoning have been
attributed to foods decomposed by this organism at the temperature
of the ice-box. The resistance of the organisms to heat is not great.
According to Meyerhof, 6 an exposure of twenty-five to thirty-five
minutes at 54 C., five to ten minutes at 56 C., and of one-half a
minute at 60 C., kills them. Their resistance to disinfectants is
similar to that of B. coli.

Products of Growth. (a) Chemical. Proteus bacilli decompose
proteins and protein derivatives energetically. The following sub-
stances have been detected among the cleavage products: trimethy-
lamine, betain, phenol, hydrogen sulphide ; 7 from the decomposition of

1 Theobald Smith, Fermentation Tube, Wilder Quarter Century Book, 1893, p. 213.

2 The bacilli may gradually lose their ability to ferment saccharose; strains which
do not ferment this sugar may be mistaken for paratyphoid bacilli, particularly if the
gelatin-liquefying power disappears simultaneously. The very considerable production
of ammonia in sugar-free broth readily distinguishes the proteus bacilli. Kendall, Day,
and Walker, Jour. Am. Chem. Soc., 1913, xxxv, 1231.

3 Levy, Arch. f. offentl. Gesundhpf. in Els. Lothr., 1895, xvi, Heft 3.

4 Deelman, Arb. a. d. kais. Gesamte, 1897, xiii, 374.
6 Fermi, Centralbl. f. Bakt., 1898, xxiii, 208.

6 Centralbl. f. Bakt,, 1898, xxiv, 20.
7 Emmerling, Ber. chem. Gesell., 1896, 2711.

BACILLUS PROTEUS GROUP 361

casein, deuteroalbumose, peptone, mono- and diamino-acids (histidin
and lysin), tyrosin, indol, and skatol. 1 An extensive liberation of
ammonia takes place in protein media free from sugars. 2 Ammonia
is also formed from the proteins of milk, but more slowly, and in
smaller amounts. 3 Carbon dioxide and hydrogen (H : CO 2 f) are
formed in dextrose and saccharose broths, together with lactic acid
and small amounts of formic acid. Lactose is unfermented. 4 Urea
is actively decomposed, ammonia and carbon dioxide being liberated. 5
The addition of dextrose prevents the liberation of ammonia and
carbon dioxide. 6

(b) Enzymes. B. proteus produces a soluble proteolytic enzvme
in protein media containing no utilizable sugars, which liquefies egg
albumen, fibrin, blood serum, and gelatin. This enzyme is not pro-
duced when utilizable sugars are present in the medium. No other
enzymes are known.

(c) Toxins. A soluble toxin has not been demonstrated in cultures
of B. proteus. At one time "sepsin" (see page 75) was supposed to
be an important factor in "ptomain poisoning." This substance is
produced in but minute amounts by proteus bacilli, however, and
no importance is attached to it. The nature of the poisonous substance
produced by B. proteus is unknown.

Pathogenesis. Several types of disease have been attributed to
members of the proteus group. Meat poisoning and ptomain poison-
ing epidemics caused by eating meats decomposed by the organisms
have been reported by Levy, 7 Wesenberg, 8 Silberschmidt, 9 and Pfuhl. 10
Dieudonne 11 has described an epidemic which originated in a potato
salad from which proteus bacilli were isolated. B. proteus is one of
the very few bacteria which will cause cystitis when it is injected
into the urinary bladder. Cystitis in man is frequently caused by
B. proteus. 12 Pyelonephritis, frequently of a very purulent type, and
abscesses are occasionally caused by members of the group. The
organisms do not as a rule grow in normal tissues, but they grow

1 Taylor, Ztschr. f. physiol. Chem., 1902, xxxvi.

2 Kendall, Day and Walker, Jour. Am. Chem. Soc., 1913, xxxv, 1232^

3 Ibid., 1914, xxxvi, 1945.

4 Theobald Smith, Fermentation Tube, Wilder Quarter Century Book, 1893, p. 213.

5 Schnitzler, Centralbl. f. Bakt., 1893, xiv, 219.

6 Brodmeier, Centralbl. f. Bakt., 1895, xviii, 380.

7 Arch. f. exp. Path. u. Pharm., 1895, xxxiv, 342.

8 Ztschr. f. Hyg., 1898, xxviii, 484.

9 Ibid., 1899, xxx, 328.

10 Ibid., 1900, xxxv, 265.

11 Miinchen. med. Wchnschr., 1903, 2282.

12 See Meyerhof. Centralbl. f. Bakt., 1898, xxiv, 18, 55, 148.

362 THE COL/ CLOACA PROTEUS GROUP

readily in necrotic tissues, forming much pus which has a foetid odor.
Middle ear infections, characterized by very foul-smelling pus, have
been reported.

Bacillus proteus fluorescens, an organism exhibiting many charac-
teristics of the proteus group, has been isolated from several cases of
Weil's disease (infectious jaundice) by Jaeger, 1 Conradi and Vogt, 2
and Bruning. 3 Bar and Renon 4 isolated a similar bacillus from a case
of jaundice in the newborn. Booker 5 has isolated B. proteus from
the feces of a large number of cases of acute summer diarrhea in
children. It would appear from his studies that the organisms played
a prominent part in the causation of certain types of this illness, par-
ticularly those characterized by choleraic symptoms.

Bacillus proteus is not very pathogenic for laboratory animals.
The injection of large doses usually causes death.

Bacteriological Diagnosis. Bacillus proteus is readily isolated upon
gelatin plates: the bacilli grow rapidly at room temperature and
liquefy the medium around each individual colony. Subcultures in
sugar media, gelatin and milk produce the changes outlined above.
B. proteus mav be confused with B. cloacae, because the latter organ-
ism ferments lactose more slowly than other sugars. 6 B. cloacae, how-
ever, is distinctly less proteolytic than B. proteus, 7 and it produces
less acid and more gas from dextrose.

1 Zeit. f. Hyg., 1892, xii. 2 Ibid., 1901, xxxvii, 283.

3 Deut. med. Woch., 1904, 1269. 4 Sem. med., 1895, 234.

5 Johns Hopkins Hospital Reports, vi.

6 Theobald Smith, Fermentation Tube, 1893, 215.

7 Kendall, Day and Walker, loc. cit., 1230.

CHAPTER XVIII.

THE MUCOSUS CAPSULATUS GROUP.

THE Mucosus CAPSULATUS GROUP.
Bacillus Rhinoscleromatis.

Bacillus Ozsense.
Bacillus Lactis Aerogenes.

THE first member of the bacteria commonly known as the pneumo-
Bacillus Group or the Mucosus Capsulatus Group was isolated by
Friedlander 1 from pneumonic lungs. At that time he believed his
"pneumonia micrococcus" was the causative agent of lobar pneu-
monia, and it was so regarded until Frankel 2 and Weichselbaum 3
pointed out its comparative infrequency in lobar pneumonia, and
differentiated it clearly from the pneumococcus, the true etiological
organism of this disease. Weichselbaum also correctly interpreted
its morphology and conferred upon it the name Bacillus pneumonise.
Subsequent investigations by many observers have added several
closely-related bacteria to the group which at the present time com-
prises the following somewhat imperfectly-differentiated types:
Bacillus mucosus capsulatus (Friedlander's pneumobacillus), Bacillus
rhinoscleromatis, 4 Bacillus ozfense, 5 Bacillus lactis aerogenes, 6 and
Bacillus acidi lactici. 7

Morphology. The members of the Mucosus Capsulatus Group are
bacilli which vary in size and shape in the same culture from oval
almost coccoid elements to distinctly elongated rods. The limits of
size are comprised practically within the following dimensions : diam-
eter, 0.5 to 1.5 microns, length, 0.6 to 3.5 microns. They occur
typically singly or in pairs, less commonly united in short chains.
Motility is not observed in cultures of any members of the group and
they appear to be devoid of flagella. Spores have not been detected.
A well-defined capsule, readily demonstrable by capsule stains, sur-
rounds each organism if it is examined in tissues or secretions of the
animal body, or in albuminous media. It tends to disappear during

1 Virchows Arch., 1882, Ixxxvii, 319; Fort. d. Med., 1883, i, 719.

2 Ztschr. f. klin. Med., 1886, x, 401.

3 Wien. med. Jahrb., 1886.

* V. Frisch, Wien. med. Wchnschr., 1882, No. 32.

8 Abel, Ztschr. f. Hyg., 1896, xxi, 89; Centralbl. f. Bakt., 1893, xiii, 161.

6 Escherich, Darmbakterien des Sauglings, Stuttgart, 1886, p. 57.

7 Hueppe, Deutsch. med. Wchnschr., 1884, p. 778.

364 THE MUCOSUS CAPSULATUS GROUP

prolonged cultivation in the usual artificial laboratory media. Ordi-
nary anilin dyes color the organisms readily, and they are Gram-
negative.

Isolation and Culture. The members of the Mucosus Capsulatus
Group grow readily on artificial media. The colonies on agar are white
or gray, from 1.5 to 3 mm. in diameter, very viscid, and raised; they
tend to become confluent. When touched with a platinum needle
the growth may be drawn away as a tenacious, sticky filament. In
gelatin, a non-characteristic filamentous growth occurs along the line
of inoculation and the surface becomes covered with a white, glisten-
ing raised colony. The gelatin is not liquefied. Milk is acidified, and

FIG. 51. Bacillus mucosus capsulatus. X 1000.

frequently the accumulation of acid leads to coagulation. A light-
pink color is imparted to litmus milk and coagulation is irregular in
this medium. Broth is clouded, and a slimy, viscid sediment collects
at the bottom of the tube. A majority of strains produce gas bubbles
on potato.

The organisms are aerobic, facultatively anaerobic. Growth takes
place at 8 to 10 C., but 37 C. is the optimum temperature. Little
or no growth occurs above 43 C.

Products of Growth. The majority of strains do not form indol,
but occasional cultures give a marked reaction for this substance. 1
Practically all strains form a mucinous substance on artificial media.

The reactions of fermentation have been used as a basis for separa-
tion into types by Perkins, 2 who groups the organisms in the following
manner :

1 Kendall, Day and Walker, Jour. Am. Chem. Soc., 1913, xxxv, 1237.

2 Jour. Infec. Dis., 1904, i, 241.

BACILLUS OZMNM 365

Type I. All carbohydrates and starch fermented with the produc-
tion of gas (H:CO 2 = r) and acid; Bacillus lactis aerogenes.

Type II. All carbohydrates except saccharose fermented; starch
fermented Bacillus mucosus capsulatus, Bacillus rhinoscleromatis,
Bacillus ozsenae.

Type III. All carbohydrates except saccharose and starch fer-
mented; Bacillus acidi lactici.

Enzymes and toxins have not been demonstrated in cultures of any
members of the group.

Pathogenicity. Human. Bacillus mucosus capsulatus has been
isolated in a considerable proportion of cases of lobular pneumonia,
but it practically never is the sole incitant of lobar pneumonia. It
is occasionally detected in purulent inflammations of the respiratory
tract not pneumonic in character, in the purulent secretions of the
nasal and frontal sinuses, in occasional cases of pericarditis and
pleurisy, stomatitis and otitis media. The normal sputum occasionally
contains the organism.

Animal. Subcutaneous inoculations into mice, rabbits or guinea-
pigs frequently lead to abscess formation characterized by thick,
viscid pus. Occasionally a generalized infection which results fatally
takes place.

Bacillus Rhinoscleromatis. Rhinoscleroma, characterized by
indurated granulomatous nodules of the mucous membrane of the
nose, is ascribed to Bacillus rhinoscleromatis by v. -Frisch, 1 Paltauf
and v. Eiselsberg, 2 and others. A satisfactory demonstration of the
etiology of this infection is wanting, but organisms culturally like
Bacillus rhinoscleromatis have been isolated from the cells of Miculicz,
large, swollen cells with crescentric nuclei characteristically present in
rhinoscleroma and demonstrated within them on section.

Bacillus Ozsense. Ozena, a disease of the nose characterized by a
fetid catarrhal inflammation, is very frequently associated with the
presence of large numbers of a member of the Mucosus Capsulatus
Group to which Abel 3 gave the name Bacillus ozsense. The organism
has not been sharply separated from Bacillus rhinoscleromatis and
Bacillus mucosus capsulatus, and its etiological relationship to ozena
is still sub judice. Autogenous vaccines of the organism have been
used with varying success in the treatment of the disease.

1 Loc. cit.

2 Fort. d. Med., 1886, Nos. 19 and 20.

3 Loc. cit.

366 THE MUCOSUS CAPSULATUS GROUP

Bacillus Lactis Aerogenes. This organism is an almost constant
inhabitant of the upper part of the intestinal tract of nurslings; it is
common in the intestinal contents of bottle-fed infants, and it fre-
quently persists in small numbers in the adult intestinal tract. A
closely related organism, Bacillus acidi lactici, is found fairly widely
distributed in milk, water, and sewage. A sharp differentiation between
the two organisms is difficult to establish. There is evidence that
the organism, ordinarily a harmless intestinal parasite, may become
temporarily pathogenic and incite intestinal disturbance varying in
intensity from slight diarrhea to severe enteritis. 1 Occasional cases
of cystitis in infants are also associated with the presence of Bacillus
lactis aerogenes in pure culture.

It is obvious that the interrelations of the Mucosus Capsulatus Group
are at present in an unsatisfactory state attempts to separate the
organisms on the basis of serological reactions have been unsuccessful,
partly because of the difficulty of removing the capsules which appear
to be somewhat impervious to antibodies. A final arrangement of
the group and an ultimate differentiation of the various organisms
comprising it awaits future elucidation.

1 Kendall and Day, Boston Med. and Surg. Jour., 1913, clxix, 753; Kendall, ibid.,
May 20, 1915.

CHAPTER XIX.

GLANDERS, ANTHRAX, PYOCYANEUS, INFECTIOUS
ABORTION: ACIDURIC BACTERIA.

BACILLUS MALLEI.

Historical. Glanders is a disease primarily of animals having an
undivided hoof: horses, asses, and mules. It may be acute or chronic,
and two clinical types are recognized: glanders, an initial infection
of the nasal mucosa and regional lymphatic glands, later an involve-
ment of the internal organs, more commonly the lungs; and farcy,
a cutaneous glanders, in which the cutaneous lymphatics are involved
with the formation of nodules (farcy buds) which frequently ulcerate
and discharge a cohesive sticky secretion. Man is occasionally
infected, the disease being one of the most fatal known. The causa-
tive organism, Bacillus mallei, was described by Loffler and Schiitz
in 1882. 1

Morphology. Bacillus mallei is a small bacillus with rounded or
somewhat attenuated ends, measuring from 0.5 to 0.75 micron in
diameter, and from 2 to 5 microns in length. The organisms occur
singly and in pairs in culture media, although long filamentous forms
are not uncommon on potato. In pus and from tissues the bacilli
occur in groups or clusters. The bacilli frequently appear as short,
almost coccoid elements, both in culture and in vivo. Older cultures
frequently contain many branched forms. The glanders bacillus is
non-motile, and possesses no flagella. Capsules and spores have not
been observed. The organism stains faintly with ordinary anilin
dyes^ better with those having an alkaline reaction. It is Gram-
negative ^ t Stained with LofHer's alkaline methylene blue, the
organism exhibits irregularity of colorable material; the bacilli may
even resemble groups of cocci with faintly stain able substance con-
necting the deeply stained, round granules. Zeit 2 has called atten-
tion to the resemblance of B. mallei in pus and tissue to staphylococci
when stained with methylene blue, and the possibility of error in diag-

1 Deutsch. med. Wchnschr., 1882, No. 52.

2 Jour. Am. Med. Assn., 1909, lii, 181.

368 GLANDERS ANTHRAX PYOCYANEUS

nosis upon morphological examination alone. The Gram stain will
distinguish between the two, however.

Isolation and Culture. Bacillus mallei grows well upon ordinary
laboratory media, better if glycerin is added, and upon blood serum
and potato. The first growth outside the animal body may be diffi-
cult to obtain. Colonies on glycerin agar are small, yellowish and
round. At first the growths are translucent, later they become nearly
opaque and more deeply colored. Qrowth in gelatin is slow and not
distinctive; no liquefaction takes place. A uniform turbidity appears
in broth after twenty-four hours' incubation at 37 C., which gradually
settles out as a tenacious, slimy sediment. If the culture is undis-
turbed, a pellicle gradually forms on the surface of the medium.
Litmus milk is slowly acidified, and coagulation may occur after
seven to fourteen days' incubation. Growth on old alkaline potato
is distinctive; after twenty-four to forty-eight hours' incubation a
light brown, translucent layer appears, which has been likened in
color and general appearance to a layer of honey. Later the growth
becomes darker, even brownish-red in color, and the underlying potato
becomes greenish or even brown. Potato that is acid does not exhibit
the typical honey yellow growth.

The glanders bacillus is aerobic, facultatively anaerobic; the opti-
mum temperature of development is 37 C., growth ceases above 43
C., and is extremely slow below 25 C. The resistance of the organism
to chemical agents is not great, but it remains viable for several weeks
when dried in pus or blood and maintained in a cool, dark place. An
exposure of naked bacilli to 55 C. for five to seven minutes kills the
organisms.

Products of Growth. Chemical. Bacillus mallei is culturally inert
in purely protein media: indol, skatol and other products of degrada-
tion of amino acids are not produced. Acid, but no gas, is formed in
dextrose broth, and acid is produced in milk.

Enzymes. No enzymes have been demonstrated in cultures of B.
mallei.

Toxins. Soluble toxins have not been isolated from growths of the
glanders bacillus; the poison of the organism belongs to the group
of the endotoxins. A substance analogous to tuberculin has been
prepared from four to five weeks' glycerin broth cultures of Bacillus
mallei, mallein or morvin. The preparation of mallein is essentially
the same as for tuberculin. The injection of mallein in moderate
doses into normal animals may lead to transient fever and a slight

BACILLUS MALLEI 369

local swelling which quickly subsides. In horses infected with B.
mallei a swelling appears within a few hours which is painful and
inflamed ; it gradually enlarges for twenty-four hours or more, and the
lymphatics of the area usually become prominent. The swelling may
persist for several days, but gradually diminishes and usually disap-
pears within ten days. The temperature rises with the local swelling
and reaches a point 1 to 2 or even 3 above the normal within twenty-
four hours. The animal usually exhibits all the signs of a generalized
reaction; it becomes listless, the coat roughens, and there is greater
or lesser generalized weakness. The temperature usually persists for
forty-eight hours or more. The reaction is specific but requires
experience for its interpretation. Variations in temperature are
caused by strangles, .bronchitis and other inflammatory infections,
hence the temperature should be observed for some hours before the
injection of the mallein. A positive reaction is of more diagnostic
value than a negative reaction. It should be borne in mind that
mallein interferes with serologic tests, hence the latter should be made
before the injection of mallein*.

Pathogenesis. Animal. Cattle appear to be immune to glanders;
swine are but slightly susceptible; cats, sheep, goats, field mice and
guinea-pigs' are susceptible, but white mice are refractory.

Acute glanders in horses and asses begins after an incubation period
of from three to six days with an abrupt rise of temperature and a
viscid, purulent nasal discharge. The nasal mucosa, at first deeply
congested, becomes ulcerated; the regional lymph glands enlarge
and may suppurate. The lungs become involved and death usually
occurs within six to fourteen days; occasionally the animal lives
several weeks. The onset of the chronic form is somewhat more insid-
ious, and the symptoms are less violent. There is usually a nasal
discharge which may be blood-streaked, and the superficial glands
of the neck are palpable. The cutaneous lymph glands and usually
the lymph channels as well become generally enlarged, and they may
break down and suppurate. The disease may run a very mild course,
hardly noticeable, and frequently terminates in a cure after months
or years.

The injection of material from ulcers, nasal secretion, or lymph
glands into male guinea-pigs leads, usually within two or three days,
to a characteristic lesion, unless the material is grossly contaminated
with other organisms, namely, a purulent orchitis; the testicle
enlarges until it can not be retracted, and the inflammation spreads

370 GLANDERS ANTHRAX PYOCYANEUS

from the tunica vaginalis to the epididymis. The peritoneum is
inflamed, and if the organism is not very virulent there is joint
involvement and gradual emaciation and death. This is known as the
Straus reaction.

Human. The essential lesion in man is similar to that in the horse
a granulomatous nodule made up chiefly of epithelioid cells and
many lymphoid cells. The bacilli occur in these nodules in large
numbers as a rule. The nodules occur chiefly in the nasal mucosa,
or in cutaneous infections under the skin; they break down readily,
causing ulceration or abscess formation. A crop of papules, which
soon break down, appears on the face, around joints, and frequently
upon the arms. The disease terminates fatally in about 65 to 70 per
cent, of all cases.

Immunity and Immunization. Recovery from an attack of glanders
does not appear to confer immunity to subsequent infection, and
attempts to induce immunity in susceptible animals by vaccines, by
the use of mallein, or by sera have been unsuccessful. Specific agglu-
tinins and precipitins are present in the blood serum of infected animals
and a diagnosis can be made by the method of complement fixation.
The latter procedure, important in horses and other domestic animals,
has not been tried very extensively in man, partly because of the
comparative rarity of cases.

Bacteriological Diagnosis. 1. Microscopical Examination. Material
from the purulent discharges of the nose or scrapings from cutaneous
nodules are stained by Gram's method and by Loffler's alkaline
methylene blue. The organism is Gram-negative, and frequently
exhibits a beaded appearance not unlike the diphtheria bacillus. A
diagnosis based upon purely morphological characters is not reliable.

2. Cultural. Scrapings from unopened granulomata or from the
organs postmortem should be inoculated upon potato having an
alkaline reaction. The characteristic appearance of the growth upon
this medium is suggestive", but not conclusive. Bacillus pyocyaneus
grows very similarly. Pus must be plated upon glycerin agar or blood
serum, because the discharges from ulcers and abscesses are almost
invariably contaminated with other organisms. Pure cultures are
examined microscopically and injected into male guinea-pigs intra-
peritoneally.

3. Animal Injection. The intraperitoneal injection of suspected
material into the peritoneal cavity of male guinea-pigs leads, in the
absence of organisms capable of causing a violent peritonitis, to the

BACILLUS MALLEI 371

localization of the bacilli in the testes, which become inflamed and
swollen the Straus reaction. The animal usually dies within a
week. Potato cultures and microscopical examination of the purulent
material in the testes usually .suffices to establish the diagnosis. In
case the material for examination is contaminated with other bacteria,
it is advisable to inoculate it into the subcutaneous tissues of one
guinea-pig, and to inoculate a second male pig with material from
an enlarged lymph gland of the first pig. A negative examination
is inconclusive.

Serological Diagnosis. (a) Mallein. Discussed above.

(6) Ophthalmo Reaction. The instillation of a few drops of mallein
into the conjunctival sac of a glanderous horse leads to a reaction
very similar to the ophthalmo-tuberculin reaction in man, except
that in positive cases a purulent discharge as well as a red inflamed
conjunctiva results.

(c) Agglutination Test. Specific agglutinins for Bacillus mallei
appear in the blood of infected animals usually within four to seven
days in acute glanders, and there is a rough parallelism between the
severity of the disease and the development of the immune bodies.
The agglutinins as a rule diminish considerably if the disease becomes
chronic, and may become reduced to such a degree that the reaction
becomes unreliable. The sera of normal horses frequently contain
non-specific agglutinins which may clump glanders bacilli in dilutions
of 1 to 100 to 1 to 300. Injections of mallein appear to influence
antibodies specific for the glanders bacillus adversely, consequently
serological examinations should be made before mallein is injected.

Serum for agglutination tests should be withdrawn in a sterile syringe
from the jugular vein in the horse, and from the median basilic vein
in man. The serum, separated from the clot, is diluted with a suspen-
sion of glanders bacilli to the following degrees: 1 to 500, 1 to 1000,
1 to 2500, 1 to 5000, 1 to 7500. Glanders bacilli, virulent for guinea-
pigs (obtained by passing glanders bacilli through a series of animals
until the organism kills the animal within five days intraperitoneal
injection), from glycerin agar slants are emulsified in physiological
salt solution containing 0.5 per cent, carbolic acid, thoroughly shaken
and filtered through a thin layer of absorbent cotton to remove clumps.
Salt-phenol solution is added to the suspension until a moderately
turbid suspension is obtained. Decreasing amounts of serum from
the suspected animal are added to obtain the dilutions mentioned
above. A normal serum and a known positive serum are diluted in

372 GLANDERS ANTHRAX PYOCYANEUS

the same manner to serve as controls. Incubation is continued at
37 C. for seventy-two hours, because the reaction is usually slow in
developing. Sterility must be maintained throughout. Strongly
positive sera may give a definite clumping in twenty-four hours or
less; the supernatant fluid becomes clear, and the organisms collect
as a diffuse sediment at the bottom of the tube. A negative reaction
is indicated by a turbid supernatant fluid. The reaction may be
made microscopically or macroscopically, the latter being preferable.

Attempts have been made to shorten the reaction time by aiding
sedimentation with the centrifuge. The various dilutions are incu-
bated for a full hour at 37 C., allowing fifteen minutes for the tubes
to reach 37 C. in the incubator; then they are whirled for fifteen
minutes at a speed with a twenty-four inch radius not exceeding 1500
revolutions, placed in the ice-box and examined after three hours.
The slowly developing reactions may not be definitely positive for
twenty-four hours.

A reaction in a dilution of 1 to 500 (horse, ass or mule) is the lowest
limit to which a definite reaction may be attributed, and the result
should be controlled with a mallein test. Dilutions of 1 to 750 or
higher are usually safely regarded as diagnostic. In human cases a
positive reaction in a dilution of 1 to 100 is diagnostic.

The method of complement fixation (see page 164 for details) is
rapidly becoming a general method for the diagnosis of glanders.

Dissemination and Prophylaxis. Glanders is transmitted by direct
contact, by infection through cutaneous abrasions and cuts, and by
feeding paraphernalia, watering troughs and buckets. In man cuta-
neous infection is more common.

BACILLUS ANTHRACIS.

Bacillus anthracis was first seen by Davaine 1 in 1863, in the blood
of animals infected with anthrax. Koch 2 confirmed Davaine 's obser-
vation, obtained the organism in pure culture, and reproduced the
disease with these cultures in other animals, thus establishing the
etiology of anthrax. He also demonstrated spore formation by B.
anthracis upon artificial media.

Morphology. Bacillus anthracis is a rod-shaped organism measur-
ing from 1 to 1.50 microns in diameter and from 2 to 4 microns in

1 Compt. rend. Acad. Sci., 1863, Ivii.

2 Cohn's Beitr. z. Biol. der Pflanzen, 1876, ii, 277.

BACILLUS ANTHRACIS 373

length. Occasionally filaments 20 to 25 microns in length are
observed, which exhibit no demonstrable septation; these long rods
may be single cells or chains of cells in which septation is imper-
fect. The ends of the bacilli are square cut and often" appear
to be concave, particularly when the organisms are examined in a
strained preparation made directly from the blood of an infected
animal. Occasionally the ends are somewhat thickened, giving the
bacillus an appearance which suggests a segment of bamboo. Bacillus
anthracis produces short chains of three to eight elements in the
bloodvessels of infected animals, and in artificial media it produces
long, coiled chains of bacilli which give a characteristic filamentous
appearance to the colonies upon solid media. The organism is non-
motile, and possesses no flagella. A capsule 1 is formed around the

FIG. 52. Bacillus anthracis, spore formation. X 1000. (Gunther.)

bacilli in the animal body and also in cultures containing albuminous
substances, as uncoagulated blood serum. 2 Spores are produced in
media freely exposed to the air between the temperatures of 15 C. and
40 C. The lower limit of spore formation has a practical bearirtg
upon the presence of anthrax spores in soil. In the temperate zones
a temperature exceeding 15 C. in midsummer is not found at depths
greater than five feet, hence anthrax carcasses buried deeply are not
likely to cause infection of the soil. It has been stated that earthworms
may carry infected material from the deeper layers of the soil to the

1 The capsule was first seen by Serafini (Progress Medico, 1888), but Johne (Deutsch.
Ztschr. f. Tiermed. u. vergl. Path., 1893, xix, 244; 1894, xx, 426) first called attention
to the diagnostic importance of the capsule in the diagnosis of anthrax of the domestic
animals.

2 Haase, Deutsch. Ztschr. f. Tiermed. u. vergl. Path., 1894, xx, 429; Johne, ibid.,
1894, xxi, 142.

374 GLANDERS ANTHRAX PYOCYANEUS

surface, where speculation may occur. If these temperatures are
exceeded in either direction, spore formation does not occur. The
spores, which are oval, are situated at or near the centre of the cell
and measure about 0.8 micron in diameter and from 1.2 to 1.4 microns
in length. Occasional asporous strains 1 are met with, and spore
formation may be suppressed by cultivating the bacteria at 42 C.
for several hours or in fluid media containing potassium bichromate
in dilutions from 1 to 5000 to 1 to 2000, or small amoimts of phenol. 2
Lehmann 3 states that long-continued transfer of cultures from gelatin
to gelatin frequently leads to a suppression of spore formation. Some
strains become asporeless much more readily than others. 4 Spores

FIG. 53. Bacillus anthracis, showing capsule formation. X 1000. (Kolle and Hetsch.)

are not formed in the intact animal body. Mature vegetative bacilli
emerge from the spores in. the presence of oxygen, if the temperature
is maintained between 15 and 40 C. The spore membrane merges
imperceptibly into the newly formed vegetative cell; no visible rup-
turing of the spore membrane is detectable.

Bacillus anthracis stains well with ordinary anilin dyes and young
cultures are Gram-positive. Older cultures may gradually lose their
ability to retain Gram's stain. Spores may be stained with the Ziehl-
Neelsen stain. (See Staining of Spores.)

Isolation and Culture. Bacillus anthracis grows readily upon any
artificial media. Material is best obtained from the spleen or liver

1 Asporous cultures do not necessarily become avirulent (Chamberland and Roux,
Compt. rend. Acad. des Sci., 1883, xcvi, 1090).

2 Roux, Ann. Inst. Past., 1890, 25.

3 Milnchen. med. Wchnschr., 1887, No. 26.

4 Surmont and Arnould, Ann. Inst. Past., 1894, p. 832.

BACILLUS ANTHRACIS . 375

of dead animals, or from the blood of an infected animal. Gelatin
is rapidly liquefied; colonies appear in gelatin plates within eighteen
hours after inoculation, which are from 1 to 2 mm. in diameter. They
are gray, opaque, and somewhat irregular in size. The organisms
develop rapidly, and liquefaction commences within thirty hours as a
rule. At this stage of development the edges of the colonies are com-
posed of tangled, radiating chains of bacilli which extend into the
surrounding medium, and the colony itself is composed of a mass of
twisted filaments which has been likened to a Medusa head. Few,
if any, pathogenic bacteria present such an appearance. The growth
in stab cultures in gelatin is also characteristic; the organisms grow

FIG. 54. Bacillus anthracis, section from kidney, semi-diagrammatic. X 500.
(Kolle and Hetsch.)

away from the line of inoculation into the medium as spikelets which
resemble an "inverted pine tree." Liquefaction soon takes place.
Milk is rendered acid, and the casein precipitated and slowly liquefied.
A pellicle forms~~upon the surface of broth which readily becomes
detached from the sides of the tube and settles to the bottom. No
turbidity is produced in fluid media.

Bacillus anthracis is a strongly aerobic bacillus, but growth will
take place under anaerobic conditions. Growth is very slow at 18
C., and ceases below 15 C. The optimum is about 37 C., and
development does not take place at 45 C.

The vegetative (asporeless) organisms are not resistant to heat or
drying. The spores are very resistant. Dried spores have remained
viable and virulent for eighteen years. 1 Fresh blood containing anthrax

1 V. Szekely, Ztschr. f. Hyg., 1903, xliv, 363.

376 GLANDERS ANTHRAX PYOCYANEUS

bacilli may remain viable for two months if relatively thick layers
are prepared. Dry heat at 160 C. kills anthrax spores within one
and a half hours; live steam (100 C.) kills them within ten minutes.
Carbolic acid is not very effective as a germicide, but 1 to 1000 bichlo-
ride of mercury kills the spores within half an hour. Direct sunlight
kills them within six hours. 1

Products of Growth. Chemical. Martin 2 found protoalbumose ,
deuteroalbumose, a trace of peptone, an alkaloidal substance, and
small amounts of leucin and tyrosin in a serum culture of B. anthracis.
Nojacid or gas is produced in any sugar media. The albumoses and
peptone caused a febrile reaction in animals, and the alkaloidal sub-
stance (anthrax-alkaloid) caused edema and congestion. These results
have never been repeated. 3

Enzymes. Bacillus anthracis produces a proteolytic enzyme which
liquefies gelatin, blood serum and casein. No other enzymes are
known.

Toxins. Soluble toxins have not been demonstrated in cultures
of anthrax bacilli, and the nature of the endotoxin is unknown the
cellular substance of the organism is not as toxic as that of many other
pathogenic bacteria, and the nature of the action of the bacillus is
not clearly determined.

Pathogenesis. Animal. Anthrax is a disease of cattle, sheep 4 and
horses. Swine are less susceptible. Guinea-pigs, rabbits and white
mice are very susceptible to inoculation. Rats and dogs succumb to
large doses. Birds and cold-blooded animals are naturally immune,
although, as Pasteur showed, the immunity may be overcome by
reducing the body temperature of birds and by raising the body tem-
perature of cold-blooded animals.

The artificially-induced disease in small laboratory animals is
usually a rapidly fatal septicemia; the organisms swarm in the blood-
vessels and appear upon section to almost occlude the capillaries.
The spleen is greatly enlarged and there is congestion of the other
glandular organs. Cattle and sheep readily succumb to infection
with pure cultures of the organism. The natural infection in cattle
and sheep appears to be chiefly through the intestinal tract. In horses

1 Moment, Ann. Inst. Past., 1892, 23.

2 Proc. Royal Soc., London, May 22, 1890. Brit. Med. Jour., March 26, April 2, 9,
1892. Animal Report Local Government Board, Supplement, 1890-91, xx, 255-266.

3 It is probable that these substances were produced from the serum by the action of
the organism ; they cannot be regarded as specific toxic products.

4 Algerian sheep are said to be more resistant to infection than ordinary sheep.

BACILLUS ANTHRACIS 377

infection may take place through the skin as well. Less commonly
cutaneous infection may occur through wounds in cattle and sheep.
A localized severe inflammation results which may heal spontaneously
or lead to a generalized infection. It is stated that flies, particularly
the horse flies (Tabanidse) may transmit the virus to animals. The
disease may also be transmitted experimentally by the inhalation of
spores; this method of infection is probably not common in animals.

Human. Anthrax bacilli or their spores may cause disease in man
either by gaining entrance to the body through abrasions of the skin,
by inhalation, or by ingestion. Inoculation through the skin may
give rise to malignant pustule, characterized by a small papule at the
site of infection, which soon becomes vesicular. The process may
stop spontaneously with the formation of a scab and the gradual
drying up of the vesicle, or the inflammation may spread, producing
a wide area of induration in which vesicles appear, often in consider-
able numbers. The involved area becomes edematous, and the
regional glands become enlarged. Death may ensue within five to
seven days, or the inflamed area slowly returns to normal. Less com-
monly edema is the prominent symptom, pustule formation being
absent or not conspicuous. The edematous area spreads rapidly and
it may be extensive enough to interfere with the nutrition of the part
and lead to gangrene. The head, the arms, or the hands are more
frequently involved than the lower extremities.

Intestinal anthrax and pneumonic anthrax or woolsorters' disease
are usually caused by the ingestion or inhalation of anthrax bacilli
or their spores. Intestinal anthrax is uncommon; it is supposed to
be an infection through the gastro-intestinal tract resulting from the
ingestion of meat or milk of diseased animals. The symptoms are
essentially those of meat-poisoning: chill, vomiting, and nausea,
diarrhea, and some .fever. Woolsorters' disease prevails where hides
and wool, particularly from South America, Morocco and Russia, are
handled. The symptoms are: a sudden chill, immediate great pros-
tration, intense pain, bronchial irritation, and occasionally death
within twenty-four hours. Cerebral symptoms frequently are promi-
nent in those cases which are more protracted. There are no distinc-
tive postmortem changes; the lungs may be edematous and there
are scattered patches of lobular pneumonia with inflammation of the
regional bronchi.

Immunity and Immunization. The vulnerability of human tissues
to anthrax infection is varied; the skin appears to be relatively resis-

378 GLANDERS ANTHRAX PYOCYANEUS

tant, but the lungs are very susceptible. The disease resulting from
infection of the lungs by anthrax bacilli is one of the most rapid and
fatal known to man. Practically no attempt has been made to
immunize man to anthrax, but Sobernheim has prepared a serum
obtained by injecting animals immunized by Pasteur's method with
virulent anthrax bacilli, which is said to be of some value as a curative
agent in malignant pustule.

Animal Immunization. Pasteur protected animals against anthrax
infection by vaccination with attenuated anthrax bacilli. Two vac-
cines were used; they were prepared in the following manner: Vaccine
A was obtained by growing anthrax bacilli at 42.5 C. for six weeks.
The organisms are asporeless after this treatment, but they grow
luxuriantly. They are avirulent for rabbits and guinea-pigs, but kill
mice. Vaccine B was obtained by growing anthrax bacilli at 42.5
C. for two weeks. The organisms kill mice and guinea-pigs, but do
not kill rabbits. Vaccine A is injected, and after two weeks Vaccine
B is injected, both subcutaneously. The animals are immune two
weeks after the last injection to cutaneous infection with anthrax
bacilli, but are somewhat less resistant to infection by way of the
alimentary tract. The immunity is of about one year's duration,
and it must be renewed at the end of that time. Sobernheim 1 has
attempted to increase the immunity to ingestion anthrax by injecting
his serum (5 to 15 c.c.) and Vaccine B of Pasteur simultaneously. He
states that this combined immunizing process brings the resistance
of the animal to such a level that ingestion infection rarely or never
occurs.

Bacteriological Diagnosis. The diagnosis of anthrax in man depends
wholly upon the identification of the anthrax bacillus.

(a) Morphological Diagnosis. Smears from the blood or tissues of
animals stained by Gram's method show large, square-ended, Gram-
positive bacilli, which occur singly, in pairs, or short chains. The
organisms are encapsulated but require special capsule stains for their
demonstration. In man similar examination is made from the serous
fluid expressed from the malignant pustule, the blood (best obtained
from the ear), fluid from edematous areas, sputum from woolsorters'
disease, and feces from intestinal cases.

(b) Cultural. The material collected aseptically is inoculated into
ordinary media. It is well to examine the media after two to three
days' incubation for spores if the culture is impure; if spores are

1 Ztschr. f. Hyg., 1899, xxxi, 89.

BACILLUS PYOCYANEVS 379

found heating the culture to 80 C. for fifteen minutes will destroy all
vegetative forms leaving the anthrax spores in excess and frequently
in pure culture. The growth on gelatin is fairly distinctive.

(c) Inoculate a guinea-pig or a mouse with a small amount of blood
or fluid from a suspected lesion; if bacilli are not numerous, incubate
the material in broth for twenty-four hours, then inject the enriched
culture. The occurrence of typical large Gram-positive bacilli in the
blood stream postmortem is sufficient to establish the diagnosis in
the light of the clinical history. The principal organisms likely to
cause confusion are: B. subtilis and members of the mesentericus
group, which do not produce acute death in guinea-pigs by generalized
septicemia, and B. edematis maligni and B. aerogenes capsulatus,
both of which are obligate anaerobes.

Dissemination. The spores of anthrax bacilli are extremely resis-
tant to dessication, and they remain alive for years in the soil. Once
a pasture or other enclosure is infected with the organisms it is unsafe
to permit cattle, sheep or other domestic animals to graze there.
The washings from such infected lands may convey infection to other
lands.

Prophylaxis in man consists essentially in preventing contact infec-
tion with diseased animals or infected material, and particular care
in preventing the inhalation of dust from hides or wool of cattle or
sheep from countries where the disease is prevalent; this applies par-
ticularly to South American, Moroccan and Russian hides and wool.

BACILLUS PYOCYANEUS.

Historical. Surgeons for many years have noticed that occasional
suppurating wounds discharge pus which stains bandages a green or
green-blue color. Gessard 1 demonstrated the specific organism, Bacil-
lus pyocyaneus, in pure culture and described it in considerable detail.
Somewhat later Charrin 2 studied the pathogenesis of the organism for
rabbits (maladie pyocyanique), setting forth clearly the importance
of the bacillus as a disease-producing microorganism.

Morphology. Bacillus pyocyaneus is a moderate-sized organism
with rounded ends, usually occurring singly or in pairs, less commonly
in short chains. The dimensions vary considerably even in the same
culture; the diameter averages about 0.6 micron, although some

1 These de Paris, 1882.

2 La maladie pyocyanique, Paris, 1889.

380 GLANDERS ANTHRAX PYOCYANEUS

bacilli measure but 0.3 micron and others as much as 1 micron. The
length varies between 1.5 and 4 microns, the average being about 2
microns. The organism is actively motile, and possesses a terminal
polar flagellum (monotrichic flagellation). Capsules and spores have
not been observed. Ordinary anilin dyes color the bacillus with mod-
erate intensity, and it^ is Gram-negative, although the gentian violet
is somewhat less readily removed by alcohol than from a majority
of Gram-negative bacilli, as Bacillus coli for example.

Isolation and Culture. The organism grows readily and rapidly
upon ordinary artificial media, producing the characteristic pigments
in the presence of oxygen. The colonies on agar are round and measure
from 1 to 3 mm. in diameter after eighteen to twenty-four hours'
incubation at 37 C. The growth spreads rapidly, and the pigment
which becomes visible within eighteen hours dissolves in the medium
imparting a blue-green color to it. Gelatin colonies are not charac-
teristic in outline, but rapidly liquefy the medium, which becomes
green. A turbidity is visible within eight hours in broth and a pellicle
usually forms on the surface. A viscous, gray-brown sediment collects
at the bottom of the tube, and an ammoniacal odor is noticeable even
within twenty-four hours. The medium, particularly the upper layers
in contact with oxygen, becomes blue-green. Milk is coagulated,
the coagulum being slimy, and eventually partly or even completely
dissolved; the medium, at first yellowish, becomes green, then blue,
particularly in the Upper layers.

Bacillus pyocyaneus is aerobic, facultatively anaerobic. The opti-
mum" temperature is 37 to 38 C.; development is sluggish below 18
C. and practically ceases at 43 to 44 C.

Products of Growth. Chemical. The organism produces a relatively
large amount of ammonia from pr.oteins and protein derivatives, 1 and
in milk. 2

Pigments. Two pigments are produced by Bacillus pyocyaneus:
a water-soluble, green, fluorescent pigment similar in physical proper-
ties to that found in cultures of other fluorescent bacteria; and a
specific pigment, pyocyanin, which is insoluble in water but soluble
in chloroform. Pyocyanin, to which the empirical formula Ci4Hi 4 NO 2
has been ascribed by Ledderhose, 3 crystallizes from chloroform solu-

1 Armaud and Charrin, Compt. rend. Ac. sc., 1891, cxii, 755, 1157; Kendall, Day,
and Walker, Jour. Am. Chem. Soc., 1913, xxxv, 1243.

2 Kendall, Day and Walker, Jour. Am. Chem. Soc., 1914, xxxvi, 1948, 1963.

3 Deutsch. Ztschr. f. Chir., 1888, xxviii, 201.

BACILLUS PYOCYANEUS 381

tion as blue needles. It forms salts with acids, and exists as a leuco-
base in cultures from which oxygen is excluded. The color changes
to a brownish-red in old cultures.

Enzymes. One of the noteworthy products of Bacillus pyocyaneus
is a soluble proteolytic enzyme, a protease, which dissolves gelatin,
casein, coagulated blood serum and fibrin. 1 Breymann 2 showed that
the bodies of the bacteria, freed from culture media, contained the
same or a similar enzyme. Emmerich and Low 3 isolated a proteolytic
enzyme, called by them pyocyanase, which possessed the remarkable
property of dissolving alien bacteria. This enzyme has been used
therapeutically with some success. Whether pyocyanase is identical
with the protease mentioned above has never been clearly determined.

No diastatic enzymes have been detected in cultures of Bacillus
pyocyaneus. 4

Toxins. Wassermann 5 found that filtered cultures of Bacillus
pyocyaneus or cultures killed with toluol would kill guinea-pigs when
injected intraperitoneally in amounts of 0.2 to 0.5 c.c. The organisms
themselves were decidedly less toxic. The toxicity is not attributable
to the specific pigment, pyocyanin, but to substances of unknown
composition.

Pathogenesis. Animal. Bacillus pyocyaneus is pathogenic for
small laboratory animals, guinea-pigs being the most susceptible.
A cubic centimeter or less of an actively growing broth culture intro-
duced into the peritoneal cavity causes death within twenty-four hours
as a rule. There is edema, leukocytosis, and the peritoneal fluid
increased in amount swarms with the bacilli. Rabbits are less sus-
ceptible; rats and mice are relatively refractory. The subcutaneous
injection of cultures of the organism, especially if the virulence is not
great, leads to a chronic, wasting infection which usually terminates
fatally. The subcutaneous tissue becomes edematous and necrotic,
and ulceration frequently occurs.

Human. Besides the focal lesions, abscesses, ulcers, otitis media,
less commonly liver abscesses, and bronchopneumonia, Bacillus pyo-
cyaneus occasionally produces severe gastro-intestinal infection,
especially in young children, generalized sepsis, and inflammation of
serous surfaces, the pleura, pericardium, and peritoneum.

1 Jakowski, Ztschr. f. Hyg., 1893, xv, 474; Fermi, Centralbl. f. Bakt., 1891, x, 401;
Kendall, Day and Walker, Jour. Am. Chem. Soc., 1914, xxxvi, 1966, and others.

2 Centralbl. f. Bakt., Orig., 1902, xxxi, 481.

3 Ztschr. f. Hyg., 1899, xxxi, 1.

4 Fermi, loc. cit. s Ztschr. f. Hyg., 1896, xxii, 263.

382 GLANDERS ANTHRAX PYOCYANEUS

Immunity and Immunization. It is possible to immunize animals
both by the cautious injection of the bacilli which stimulate the for-
mation of specific bacteriolysins, and by filtrates of broth cultures of
the organisms, which incite the formation not only of bacteriolytic
substances, but antitoxic substances as well. No practical use is
made of these antibodies in human infections, however.

Bacteriological Diagnosis. Wounds infected by Bacillus pyocyaneus
are usually diagnosed by the blue-green color of the dressings. The
bacilli are readily isolated upon gelatin plates, where the development
of the blue-green color is very characteristic.

BACILLUS ABORTUS.

Historical. Infectious abortion is a disease which has for many
years been recognized as an important economic one in the cattle
industry. Later it was found that the same disease also exists among
horses, goats and sheep. The organism was first isolated by Bang. 1

Morphology. B. abortus is a small pleiomorphic bacillus, measuring
0.4 to 0.6 micron in diameter, by 0.6 to 2.5 microns in length. It
occurs singly and in pairs; rarely short chains of three to six elements
are found. The shape varies: some organisms are almost spherical,
others are distinctly rod-shaped, the latter being more frequently
found in broth cultures and in vivo. According to Priesz, 2 branched
forms may be found in older cultures. It is non-motile and possesses
no flagella, although Brownian movement may be fairly active. It
possesses no capsules, and no spores have been demonstrated. It
stains readily with ordinary anilin dyes, but somewhat irregularly,
some areas staining more intensely than others. Occasionally with
the methylene blue stain the organisms may present a bipolar appear-
ance. The organism is Gram-negative.

Isolation and Culture. Initial growths on artificial media outside
the animal body are somewhat difficult to obtain. The organism
appears to grow best in a somewhat rarified atmosphere. This has
been obtained by Fabyan 3 by growing the organism on an agar slant
which is connected by a narrow tube with an agar slant on which
B. subtilis is growing. B. subtilis appears to so change the percentage
composition of the air in the two tubes that B. abortus grows fairly
readily. He also found that a pressure of three to five atmospheres

1 Ztschr. f. Thiermedizin, 1897. i, 241 2 'S.

2 Ccntralbl. f. Bakt., Orig., 1903, xxxiii. LJO

3 Jour. Med. Research, 1912, xxvi, 441.

BACILLUS ABORTUS 383

would facilitate the growth of the organism. On dextrose agar the
colonies are round, normally colorless and transparent, and have a
very glistening, pearly sheen. The colonies attain a diameter of from
0.5 to 2.5 mm. The organism grows well on blood serum. On gelatin
the growth is usually very slow, probably because of the lowered tem-
perature of incubation. No liquefaction takes place. In milk there
is a moderate growth; no acid is formed, and no coagulation or pep-
tonization takes place.

Conditions of Growth. The organism is killed by an exposure of 59
C. for ten minutes. 1

Products of Growth. The organism produces no known ferments
and it produces no acid in dextrose or other sugars; on the contrary,
the reaction on artificial media in which the organism is growing
becomes slightly alkaline. The organism forms no extracellular toxins.

Pathogenesis. Infectious abortion appears to be an infection of the
fetus in utero and its membranes, which results in the death of the
fetus and its expulsion, or less commonly its expulsion in a living
and enfeebled state. The time of expulsion is not definite; it may
occur early during the period of gestation, or it may not take place
until the normal completion of pregnancy. Ordinarily there is no
direct evidence of disease in the mother.

The lesions in experimental guinea-pigs, which have been described
very carefully by Fabyan, 2 resemble both macroscopically and his-
tologically those of tuberculosis. As a rule, the "muscles are free from
lesions and there is a tendency for the organism to localize itself in
the perivascular or subcapsular regions of various abdominal organs.
The organism may persist in experimental animals for very consider-
able periods of time without producing manifest symptoms. Fabyan
has shown 3 that the organism may remain alive but latent in guinea-
pigs for over a year.

Immunity. Cows which have aborted once or twice appear to acquire
an immunity which is supposed to be due to the formation of anti-
bodies in the blood. Although no extracellular toxins have been
demonstrated as yet, it is probable that the infected animal is sensitized
by endotoxins of the abortion bacillus, for such animals injected with
"Abortin" (an extract of the abortus bacillus) usually give a definite
reaction.

Of extreme importance is the frequent occurrence of the organism

1 Fabyan, loc. cit., p. 481. 2 Loc. cit.

3 Jour. Med. Research, 1913, xxviii, 81.

384 GLANDERS ANTHRAX PYOCYANEUS

in milk. Melvin 1 has found B. abortus in eight out of seventy-seven
samples of market milk and in the milk of six dairies out of a total of
thirty-one examined. As early as 1894 Theobald Smith 2 called atten-
tion to peculiar tubercle-like lesions induced in guinea-pigs following
the injection of cow's milk. He recognized that the disease was not
tuberculosis; later Schroeder 3 made similar observations. In the
same year Smith and Fabyan 4 showed that, the tubercle-like lesions
were caused by B. abortus, and in 1913 Fabyan 5 demonstrated con-
clusively the extremely important fact that B. abortus is very fre-
quently found in the milk of cows that have aborted. He also showed
that pasteurization of milk, if carried out in the proper manner, will
certainly destroy the bacillus. Whether certain cases of abortion
observed in man are due to the organism is not yet proven. 6

Bacteriological Diagnosis. The bacteriological diagnosis is best
made by injecting guinea-pigs with suspected milk or material from
a diseased animal and observing the development of the characteristic
tubercle-like lesions. If the animal does not die within a reasonable
time it should be killed and autopsied.

Serological Diagnosis. The blood serum of infected cattle usually
agglutinates B. abortus in dilutions greater than 1 to 50. The value
of the agglutination reaction as a method of diagnosis is as yet debat-
able. The extensive statistics of MacFadyen and Stockmann 7 upon
this phase of the subject are representative. An agglutination with
B. abortus in a dilution of 1 to 50 was obtained with the sera of 526
out of a total of 535 apparently healthy cows;* in the remainder (9)
agglutination took place in dilutions greater than 1 to 50. Of 127
cattle, either infected or suspects, an agglutination was not obtained
in a dilution of 1 to 50; in 11 agglutination was positive, 1 to 50; in

19 a positive reaction was obtained in a dilution of 1 to 100; and in

20 a reaction in a dilution of 1 to 200. Holth 8 tested the sera of 7
normal cattle with negative results. The sera of 38 animals out of a
total of 39, which were plainly infected with B. abortus, gave positive
agglutination with the specific organism in a dilution of 1 to 100.

1 Vet. Jour., 1912, Ixviii, 526.

2 Bureau of Animal Industry, 1894, Bull. 7, 80.

3 Bur. Animal Industry, Circ. 198, November 2, 1912.
4 Centralbl. f. Bakt., Orig., 1912, Ixi, 549.

6 Jour. Med. Research, 1913, xxviii, 85.

6 Recently Laisen and Sedgwick (Am. Jour. Dis. of Child., 1913, vi, 326) have exam-
ined blood serum from 425 children by the method of complement-fixation; 73 were
positive, 325 weie negative.

7 Jour. Compt. Path, and Therap., 1912, xxv, 22.

8 Berl. tierarztl. Wchnschr.. 1909, 686.

ACIDURIC BACTERIA 385

The method of complement fixation, precipitin test, ophthalmo reac-
tion and intracutaneous reaction with various preparations of B.
abortus have been tested for their diagnostic value, but the results
are not clear cut and definite. 1

Prophylaxis and Dissemination. The infection of market milk with
B. abortus focuses attention sharply upon the transmissibility of the
organism to man. Definite details are lacking, but pasteurization of
milk should remove all practical danger from this source.

ACIDURIC BACTERIA. 2

There is a somewhat poorly defined group of bacilli, chiefly found
in the intestinal contents of man and animals, 3 which possesses the
unusual property of growing in fermentation media of a degree of
acidity incompatible with the development of all other known bacteria.
The aciduric bacteria are of two kinds : the true acidurjc bacilli, of
which Bacillus acidophilus is the best known, and facultatively
aciduric bacteria, 4 which are occasionally detected in the intestinal
contents of man and animals fed for some time upon carbohydrate.
The facultative organisms rapidly lose their acid tolerance upon cul-
tivation in ordinary media, and they are probably to be regarded as
examples of bacterial adaptation.

Rahe 5 distinguishes three types of aciduric bacilli, depending upon
their action upon carbohydrates. Acid, but no gas is formed, as
follows :

Type I. Bacillus bulgaricus (not an intestinal organism) coagulates
m ilk, but does not ferment mannite.

Type II. Coagulates milk and ferments mannite.

Type III. Does not coagulate milk, but ferments mannite.

Bacillus Acidophilus. Bacillus acidophilus, described by Moro 6
and independently by Finkelstein 7 is a somewhat pleiomorphic bacillus
of varying length, which occurs singly or in pairs as a rule. Chain
formation is not uncommonly observed in cultures on artificial media.
The organism forms no spores or capsules and it is typically Gram-

1 See Klimmer, Ergebnisse der Immunitatsforsch. u. experimentelle Therap., 1914, i,
143-188, for details.

2 Kendall, Jour. Med. Research, 1910, xxii, 153 for resume and literature to 1910.
See also Rahe, Jour. Inf. Dis., 1914, xv, 141.

3 Mereschkowsky, Centralbl. f. Bakt., Orig., 1905, xxxix, 380, 584, 696; 1906, xl, 118.

4 Kendall, loc. cit., p. 165.

5 Loc. cit.

6 Wien. klin. Wchnschr., 1900, v, 114.

7 Deutsch. med. Wchnschr., 1900, xxii, 263.

386 GLANDERS ANTHRAX PYOCYANEUS

positive, although in old cultures a majority of the bacilli are fre-
quently Gram-negative.

Isolation and Culture. The organism may be isolated directly from
suspected material in 2 per cent, dextrose broth containing 0.25 per
cent, acetic acid. After two or three days growth becomes apparent
and a few loopfuls of the well-shaken culture are transferred to a second
and then a third tube of the same medium. Usually the third transfer
contains either a pure culture or it is greatly enriched with the specific
organism. Pure colonies are obtained by plating upon 2 per cent,
dextrose agar unadjusted for reaction, or better, upon dextrose agar
containing 0.2 per cent, sodium oleate according to the procedure of
Salge. 1

The colonies are of two types a round, smooth-edged compact
colony, and a thin, semi-translucent colony with delicate filamentous
edges.

Products of Growth. Bacillus acidophilus is carbohydrophilic in its
activities; it does not grow well in media containing proteins and
protein derivatives only. Indol, phenols and similar products of
protein degradation are not found in cultures of this organism. Gelatin
is not liquefied and growth is feeble in this medium.

Frequently cultures on sodium oleate agar slants exhibit a clouding
of the medium; 2 the cause of the clouding is not known.

Pathogenesis. Escherich 3 and Salge 4 have described acute diarrheas
in young children, characterized bacteriologically by large numbers
of Gram-positive bacilli in the feces which are strongly acid in reac-
tion; Escherich applied the name "Blaue Bazillose" to this type of
intestinal disturbance, because of the great preponderance of Gram-
positive bacilli in Gram-stained preparations prepared directly from
the feces. Subsequent investigations have shown that the u blue
bacilli" were in all probability Bacillus acidophilus, and it has been
shown that a condition apparently identical with that described by
Salge may develop in young children fed with too much maltose or
malz suppe. 5

Bacillus Acidophil-aerogenes. Torrey and Rahe 6 have described a
member of the aciduric group of bacteria which produces acid and gas

1 Jahrb. f. Kinderheilk., 1904, lix, 399.

2 Kendall, loc. cit., p. 156; Rahe, loc. cit., p. 9.

3 Jahrb. f. Kinderheilk., 1900, Hi, 1.

4 Kie akute Diinndarmkatarrh des Sauglings, Leipzig, 1906.

5 Kendall, Boston Med. and Surg. Jour., 1910, clxiii, 322.

6 Jour, of Inf. Dis., 1915, xvii, 437.

ACIDURIC BACTERIA 387

in dextrose, lactose, saccharose and maltose; mannite was not fermented.
The morphology of the organism, the types of colonies produced on
dextrose agar, and its staining reactions resemble those of Bacillus
acidophilus. The production of gas in the sugar mentioned and
the relatively feeble growth in milk are its distinguishing cultural
characteristics.

Sera of animals immunized to Bacillus acidophil-aerogenes failed
to agglutinate Bacillus acidophilus, and vice versa, indicating that the
two organisms are quite distinct entities.

CHAPTER XX.

THE DIPHTHERIA BACILLUS GROUP.

THE DIPHTHERIA BACILLUS.

BACILLI SIMILAR TO THE DIPHTHERIA

BACILLUS.

Bacillus Hofmanni.
Bacillus Xerosis.
Bacillus Hodgkini.

THE DIPHTHERIA BACILLUS.

Synonyms. Corynebacterium diphtherise, Klebs-Loffler bacillus.

Historical. A small group of bacteria excrete soluble extracellular
toxins which produce specific disease. The first member of the
group to be isolated and studied was the diphtheria bacillus. Klebs 1
called attention to the very general occurrence of a bacillus of unusual
and characteristic appearance in the gray membranes usually present
in the throats of severe and fatal cases of diphtheria, and a year
later Loffler 2 isolated the organism in pure culture from several cases
of the disease. Loffler also obtained the diphtheria bacillus from the
throat of an apparently normal child, which led him to be very guarded
in attributing a specific relationship of the organism to the disease.
Subsequent studies by innumerable investigators have corroborated
these observations in every essential detail, and have demonstrated
conclusively that the diphtheria bacillus is the specific etiological
organism of diphtheria. Roux and Yersin 3 discovered the soluble
toxin of the diphtheria bacillus and reproduced the essential systemic
phenomena of the disease in experimental animals by injecting the
toxin freed from bacteria by filtration through porcelain. V. Behring
and Kitasato 4 made the very important discovery that the blood
serum of animals injected with gradually increasing amounts of diph-
theria toxin contained a specific antitoxin which would neutralize the
toxin; diphtheria antitoxin is one of the very few specific sera possess-
ing curative properties.

Morphology. The diphtheria bacillus is one of the very few bacteria
which possess a characteristic morphology. The organisms are

1 Verhandl. Kong. Inn. Med., Wiesbaden, II. Abt., 1883, 143.

2 Mitt. a. d. kais. Gesamte, 1884, ii, 451.

3 Ann. Inst. Past., 1888, 642.

4 Deutsch. med. Wchnschr., 1890, xvi, 1113.

THE DIPHTHERIA BACILLUS 389

highly pleiomorphic bacilli, usually slender straight or slightly curved
rods with rounded and frequently swollen ends. The size and shape
of the individual organisms vary greatly even in the same culture;
they are not uniformly cylindrical as a rule, but have club-like thick-
enings at one or both ends, or they are swollen in the middle and more
or less pointed at the ends. Occasionally one end only is thickened,
giving rise to a long, somewhat wedge-shaped rod. The distinctive
morphology is best seen in eighteen- to twenty-four-hour growths on
Loffler's coagulated blood serum; organisms from growths on agar
are more uniform in appearance. Diphtheria bacilli observed directly
in diphtheritic membranes are also less pleiomorphic than those from
blood serum cultures. The organisms occur singly or in pairs, very
uncommonly in short chains. The size is very variable, ranging from
0.3 to 0.8 micron in diameter and from 1.5 to 6 microns in length.
The organism as ordinarily seen in diphtheritic membranes is about
0.6 micron in diameter and about 4 microns in length. Branched
forms are occasionally seen, particularly in the membrane which
forms on old plain broth cultures.

The stainable substance of the organisms is not uniformly dis-
tributed, but occurs in somewhat irregular concentration, giving rise
to three rather distinct types of bacilli: the granular, the barred, and
the solid. 1 Metachromatic granules (Ernst-Babes granules) are also
present, and, according to Williams, the diphtheria bacillus reproduces
by fission at one of these granules. It was originally supposed that
the metachromatic granules were only found in virulent strains and
that the non-virulent strains had no granules. Neisser 2 invented a
stain which brings out these granules very sharply. 3 It is now known
that the granules are not necessarily related to virulence, conse-
quently the Neisser stain is rarely used. Diphtheria bacilli stain
well with the ordinary anilin dyes and very characteristically with

1 Wesbrook, Wilson, and McDaniel, Jour. Boston Soc. Med. Sc., 1900, iv, 75; Trans.
Assn. Am. Phys., 1900.

2 Ztschr. f. Hyg., 1897, xxiv, 443.

3 The stain is prepared in the following manner:

A. Methylene-blue (Griibler's) 1 gram

Alcohol, 96 per cent 20 c.c.

Glacial acetic acid 50 c.c.

Distilled water 950 c.c.

B. Bismarck brown . 1 gram

Distilled water . . 500 c.c.

The smear, fixed in the flame in the usual manner, is covered with solution A for three
to five seconds, washed in water, then covered with B for three to five seconds. After
thorough washing in water, the preparation is ready for microscopic examination.
The granules are stained blue, the bodies of the bacilli brown.

390

THE DIPHTHERIA BACILLUS GROUP

Loffler's methylene blue. With methylene blue the granules above
mentioned are brought out very sharply, and it is observed that these
granules exhibit the phenomenon known as metachromatism, that is,
they stain mahogany red while the rest of the organism stains blue.
Diphtheria bacilli are Gram-positive, but prolonged washing with
alcohol removes the Gram-positive stain. Cultures prepared directly
from diphtheritic membranes stain more uniformly than organisms
obtained from cultures on Loffler's blood serum.

Diphtheria bacilli are non-motile, possess no capsules and form no
spores. Very frequently the organisms are arranged in a definite
and characteristic manner, occurring very commonly in pairs, each
pair of organisms forming a configuration very similar to a capital
"L," and a series of these angulated pairs are arranged in parallel,

FIG. 55. Bacillus diphtherias, methylene-blue stain. ( X 1000.)

very much like chevrons. This angular arrangement of the organisms
is due to their method of reproduction.

Isolation and Culture. The diphtheria bacillus grows best on Loffler's
alkaline blood serum and this medium is almost specific for the
organism, which during the first nine to eighteen hours' incubation
outgrows all other organisms with which it is usually associated in
characteristic lesions, except staphylococci. Colonies of diphtheria
bacilli on this medium after eighteen hours' incubation at 37 C. are
gray-white, round, rather dull, with darker centres, and may attain
a diameter of 1 to 1.5 mm. Diphtheria bacilli grow somewhat more
slowly on plain agar, forming small, non-characteristic colonies. The
organisms produce a well-marked zone of hemolysis around the
individual colonies on blood agar, but the hemolytic area is smaller

THE DIPHTHERIA BACILLUS 391

than that characteristic of the streptococcus. Pseudodiphtheria bacilli
do not produce hemolysis on this medium 1 The growth of diphtheria
bacilli in gelatin is slow, and the organisms do not produce liquefaction
of the medium. In plain broth the organism grows rather slowly;
repeated transfers are usually followed by the development of a pel-
licle which floats on the surface. 2 This pellicle may sink, but a new
one usually takes its place. The growth in dextrose broth is more
luxuriant than in plain broth, but no pellicle forms. There is, however,
a well-marked turbidity. The diphtheria bacillus grows well in milk,
producing an initial acid reaction during the first two or three days

FIG. 56. Bacillus diphtherias, branching. ( X 800.)

of incubation, followed by the gradual development of an alkaline
reaction. 3 No gross changes, however, are produced in the milk, even
with prolonged cultivation. The growth on potato is very slight
provided the reaction of the potato is alkaline; no growth at all
takes place on acid potato.

The diphtheria bacillus is an aerobic, facultatively anaerobic organ-
ism. Its limits of growth are 17 C. as a minimum, 43 C. as a maxi-
mum, with the optimum at 37 C. Ten minutes exposure to 60
C., five minutes at 70 C., or one minute at 100 C. readily kills diph-
theria bacilli. The organisms are occasionally transmissible through
milk, and in this connection it should be remembered that the ordinary
method of heating milk in an open vessel will not certainly kill diph-

1 Mandelbaum and Heinemann, Centralbl. f. Bakt., Orig., 1910, Hii, 356; Rankin,
Jour. Hyg., 1911, xi, 271.

2 It is essential for the production of toxin that the organisms be cultivated until
they produce a pellicle, leaving the underlying medium perfectly clear and free from
bacilli.

3 Kendall, Day and Walker, Jour. Am. Chem. Assn., 1914, xxxvi, 195.

392 THE DIPHTHERIA BACILLUS GROUP

theria bacilli, for, as Theobald Smith 1 pointed out many years ago,
a scum forms on the free surface of the milk, consisting of casein and
lime salts, which is a non-conductor of heat. Within this membrane
the diphtheria bacilli may resist a long period of heating. Diphtheria
bacilli exposed to heat enclosed in a false membrane, as for example,
those taken from the throat, may resist an exposure of 95 to 100 C.
even for an hour. Organisms dried in this membrane may remain
viable at low temperatures, protected from sunlight, from three to
five months. Naked germs are readily killed by antiseptics in the
ordinary concentrations, but those exposed to the action of antiseptics
protected in membranes may resist for some time. Hydrogen peroxide
is said to be particularly germicidal for the diphtheria bacillus.

Products of Growth. Chemical. Bacillus diphtheria produces acids,
chiefly lactic, together with smaller amounts of formic acid from the
fermentation of dextrose and maltose. Lactose, saccharose and
mannite are not fermented. Neither indol nor phenols are formed
in sugar-free broths, 2 but small amounts of ammonia are produced
in this medium, the amount increasing with the age of the culture. 3

Enzymes. No enzymes acting upon proteins, carbohydrates or
fats have been detected in cultures of the organism.

Toxin. The most important and characteristic of the products
formed by the diphtheria bacillus is a potent, soluble (extracellular)
toxin. The potency of the toxin varies somewhat with the culture
used, some strains producing more than others. An occasional strain,
typical in other respects, fails to form toxin. Prolonged cultivation
of the organism in artificial media may lead to a diminution in toxin-
producing capacity, but this is by no means a general rule. Williams 4
isolated a diphtheria bacillus from a mild case of tonsillar diphtheria
(No. 8) which has retained its toxin-producing power unimpaired up
to the present time. This culture is widely used throughout the world
in the commercial preparation of diphtheria antitoxin.

Conditions Favoring the Production of Diphtheria Toxin. 1. Com-
position of the Medium. Park and Williams 5 found that 2 per cent,
of peptone added to meat infusion broth increased the yield of toxin
very materially, and Theobald Smith 6 made the very important

1 Theobald Smith, Jour. Exp. Med.. 1899, iv, 233.

2 Ibid., 1897, ii, 543.

3 Kendall, Day and Walker, Jour. Am. Chem. Soc.. 1913. xxxv, 1210.

4 Jour. Med. Research, June, 1902.
6 Jour. Exp. Med., 1896, i, 164.

Tr. Assn. Am,. Phys., 1896: Jour. Exp. Med., 1899, iv, 373.

THE DIPHTHERIA BACILLUS 393

observation that the presence of muscle-sugar (dextrose), commonly
found in small amounts in meat-juice, prevented the formation of
diphtheria toxin; he demonstrated conclusively that small amounts
of dextrose (less than 0.2 per cent.) delay the appearance of toxin;
in sugar-free broth toxin production increases with the growth of the
organisms. Diphtheria toxin is formed from the protein constituents
of the medium; when utilizable carbohydrate (dextrose) is present
in the medium, the bacilli ferment it instead of attacking the protein. 1
It is customary to add 0.1 per cent, of dextrose to broth for the pro-
duction of diphtheria toxin; the initial development of the bacilli is
greater, and this amount of dextrose is rapidly used up, leaving greater
numbers of organisms to form toxin from the protein constituents.
The culture must be grown at 37 C. to insure a potent toxin.

2. Oxygen. Free oxygen is an essential factor in the production
of toxin. It is customary to distribute the broth in shallow layers
with a relatively large surface exposed to the air.

3. Pellicle. Cultures of diphtheria bacilli which grow habitually
on the surface of fluid media must be used for the preparation of
toxin. Diphtheria bacilli can be "trained" to develop on the surface
by repeated transfers in broth. 2 Surface development insures a
maximal exposure of the bacilli to the air.

4. Incubation. It requires from seven to ten days' incubation at
37 C. for the maximum accumulation of toxin. Deterioration of the
toxin after this time sets in, and the formati6h of new toxin fails to
keep pace with the recession in potency of the toxin already formed. 3

Storage of Toxin. At the end of the period of incubation carbolic
acid or other preservative is added to the broth to kill the bacilli;
they rapidly settle out, leaving a clear supernatant fluid free from
bacteria, containing the toxin, which is either decanted off from the
bacilli or filtered through unglazed porcelain to remove the bacteria.
It is then stored in amber bottles which are completely filled and
kept in cold storage. Under these conditions the toxin deteriorates
comparatively slowly.

Testing Toxin. Toxin produced by the diphtheria bacillus kills
the ordinary laboratory animals, guinea-pigs, rabbits, dogs, and birds;
but it is practically without effect upon rats and mice, unless the
toxin is injected directly into the nervous system. The general method

1 Kendall, Boston Med. and Surg. Jour., 1913, clxviii, 825.

2 Theobald Smith, Jour. Exp. Med., 1899, iv, 392.

3 Ibid.

394 THE DIPHTHERIA BACILLUS GROUP

of testing the potency of the toxin is to inject successively smaller
graduated doses of it subcutaneously into guinea-pigs of two hundred
and fifty grams weight and observe the results. The smallest amount
of a toxin which will kill a guinea-pig weighing two hundred and fifty
grams in four days is designated the minimal lethal dose (M. L. D.).
The minimal lethal dose varies considerably with different strains
of bacilli; in general it varies from 0.25 c.c. to 0.001 c.c. The injec-
tion of a M. L. D. of toxin leads to an edematous swelling at the site
of inoculation and the animal soon exhibits generalized symptoms as
well; the temperature rises, the respirations are hurried, and death
ensues from the results of the toxemia. The more acute the death,
the less striking the symptoms and lesions. Guinea-pigs which have
died on the fourth day exhibit a marked congestion of the abdominal
and thoracic viscera and of the colon. A hemorrhagic infiltration and
enlargement of the suprarenals is almost pathognomonic. Frequently
the stomach wall is markedly injected with blood and small ulcera-
tions are demonstrable in the mucosa. 1 The lesions present the same
general appearance when both toxin and bacilli are injected, but a
false membrane, composed of bacteria and a fibrinopurulent exudate,
forms at the site of inoculation. The bacilli do not spread to other
parts of the body, however, but remain strictly localized. The changes
in the visceral organs are attributable to the absorption of the toxin.
A sub-lethal dose of toxin or an attenuated culture of diphtheria bacill i
does not cause death; an ulcer forms at the site of inoculation which
eventually sloughs away and is completely replaced by scar tissue.

Constitution of Diphtheria Toxin. The composition of diphtheria
toxin is unknown, although many investigations have been made upon
it. Attempts to demonstrate that the toxin is non-protein in nature
by growing the organisms in protein-free media have not been con-
vincing. Small amounts of toxin have been detected in these cultures,
but the well-recognized synthetic powers of bacteria make this line
of evidence inconclusive. There are two current theories which receive
serious consideration. One theory maintains that diphtheria toxin is
enzymic in nature, the other theory assumes that the toxin is related
to the proteins, particularly the globulins. The toxin is readily
destroyed by exposure to light, heat, protoplasmic poisons and to peptic
digestion, consequently moderate amounts of it may be swallowed
without apparent harm. Acids destroy the toxin slowly, and oxidizing
agents, as hydrogen peroxide, iodin and iodin trichloride, reduce the

1 Rosenau and Anderson, Jour. Inf. Dis., 1907, iv, 1.

THE DIPHTHERIA BACILLUS 305

toxicity very materially. An exposure to 60 C. for ten hours, or at
70 C. for two hours, attenuates the toxin, and it is rapidly inactivated
or destroyed at 100 C.

Protein precipitants, as ammonium sulphate and alcohol, precipitate
the toxin from the broth in an insoluble state with but little reduction
in potency. The precipitate, after dialysis to remove the salts, is
soluble in water. A further reduction in volume and partial puri-
fication can be attained by evaporating the broth to one-tenth its
original volume in vacuo at a temperature not exceeding 25 C. (in
the dark), precipitating with alcohol, filtering and dissolving in water,
and again precipitating, then drying the precipitate in vacuo.

Physiological Action. The chemical constitution of the toxin mole-
cule is unknown, and toxin can not be detected or assayed chemically.
It provokes, however, a definite physiological response in susceptible
animals, as guinea-pigs, and its presence is detected and its strength
determined by injecting graduated doses into them, as mentioned above.
From its physiological action the toxin molecule appears to consist
of three components in varying amounts: (a) Toxin, which causes
the acute symptoms of intoxication, parenchymatous degeneration
and death when injected into susceptible animals. This fraction of
the toxin, according to Ehrlich, has a special affinity for the antitoxin.
(b) Toxone: the toxone causes edema at the site of inoculation and
the postdiphtheritic paralyses. It combines with antitoxin more
slowly than the toxin, (c) Toxoid: diphtheria toxin rather readily
loses its toxic properties on standing, retaining its power of com-
bining with antitoxin unimpaired, however. Toxin which is devoid
of toxic power but which combines with antitoxin is called "toxoid."

Antitoxin. Preparation. The injection of the soluble toxin of the
diphtheria bacillus in sublethal doses into experimental animals stimu-
lates the formation of specific antitoxin which has both curative and
prophylactic value. Antitoxin is obtained from horses because they
are less susceptible to the action of the toxin than smaller animals.
The serum of horses, at least in single doses, is innocuous for man, and
horses furnish large amounts of blood (containing antitoxin) without
injury to the animal. Young animals free from glanders, tuberculosis
and other diseases are used for the purpose. Several methods are
available for immunization, but the one commonly selected is carried
out in the following manner: an initial injection of diphtheria toxin,
either mixed with an excess of antitoxin or attenuated by iodin
trichloride, is made and about a week later a second injection con-

396 THE DIPHTHERIA BACILLUS GROUP

taining an increased amount of toxin follows. At regular intervals
the injections are repeated, each time increasing the amount of toxin
in regular progression until after three to four months as much as
250 to 300 c.c. of unaltered toxin is introduced at one time. After
about two weeks following the last injection the animal is bled and
the potency of the serum tested. If it contains one hundred and fifty
units or more of antitoxin to the cubic centimeter, from two to five
liters of blood are removed from the jugular vein with sterile precau-
tions into sterile receptacles, and the animal is again treated with
toxin to induce further immunization. As a rule, about two-thirds
of the volume of blood taken is regained in antitoxin-containing serum.
It is customary in large establishments to immunize several horses
at the same time and mix this serum, for experience has shown that
the serum of certain animals contains substances which cause erythe-
matous rashes in man which are disagreeable and irritating although
not necessarily harmful. Pooling the blood reduces this possibility.
The serum is stored in sterile containers in a dark cold place and
retains its antitoxic properties well. It deteriorates less rapidly than
toxin.

Concentration. Atkinson 1 noticed that the globulins of the horse
serum increased and the albumins diminished as the antitoxin content
of the blood increased, and he effected a partial purification of the anti-
toxin fraction by removing the albumin with protein precipitants .
Gibson 2 carried the process further and obtained a serum which was
about three times as rich in antitoxin per unit volume as the original
horse serum. Banzhaf 3 has reduced the proportion of non-specific
protein as far as is practical by purely physical agents. This reduction
of non-specific proteins is important for two reasons: first, because
it reduces the danger of anaphylaxis due to sensitization of the patient;
and, secondly, because the rashes and joint swellings are notably
reduced when the concentrated antitoxin is used instead of the whole
horse-serum. It is possible to obtain the same therapeutic effect by
the injection of about one-third the amount of solution when con-
centrated antitoxin is administered.

Properties. Diphtheria antitoxin specifically neutralizes diphtheria
toxin both in vitro and in vivo. It has little neutralizing value for
the toxone, however ; consequently in severe cases when it is used late,

1 Jour. Exp. Med., 1899, iii, 649.

2 Jour. Biol. Chem., 1906, i, Nos. 2 and 3.

. 3 Collected studies from the Research Laboratory, New York City Board of Health,
vols. v and vi.

THE DIPHTHERIA BACILLUS 397

it will not prevent the development of postdiphtheritic paralyses. It
has both prophylactic and curative properties. It is not bacteriolytic
and exhibits no agglutinins for diphtheria bacilli. Nothing is definitely
known of the nature of diphtheria antitoxin. If the diphtheria toxin
is a ferment, the antitoxin would appear to be an antiferment. The
fact that it is precipitated with the globulin fraction of the blood serum
would suggest that it may be either closely related to the proteins or
a true protein itself.

Diphtheria toxin varies considerably in its potency due to the fact
that it deteriorates; the antitoxin, on the contrary, is more stable.
Consequently for purposes of comparison and standardization a
standard antitoxin is used. Two such standard antitoxins are
recognized officially: one prepared by Ehrlich in Germany; the other
prepared by the United States Public Health Laboratory in Washing-
ton, D. C. Both of these antitoxins were prepared on a very large
scale and preserved in a cold, dark, dry place in packages of conveni-
ent size. When the supply of one or the other of the standards is
nearly exhausted a new lot of antitoxin will be prepared and carefully
compared with the old. Small amounts of the standard antitoxin
containing a definite number of antitoxin units are sent out regularly
by the central laboratories to interested laboratories for testing
purposes.

Standardization of Antitoxin. The antitoxin unit may be defined
as "that amount of antitoxin which just suffices to protect a guinea-
pig of 250 grams weight against 100 times the minimal fatal dose of
diphtheria toxin." The process of standardization of antitoxin of
unknown potency is carried out in the following manner: diphtheria
toxin, prepared as described above, is mixed in gradually diminishing
amounts with a definite amount of the standard antitoxin (containing
a known number of antitoxic units) and allowed to stand for twenty
to thirty minutes to permit union of the toxin-antitoxin to take place.
The mixtures are then injected subcutaneously into guinea-pigs of
250 grams weight. The greatest dilution of toxin which kills a guinea-
pig in four days is said to be the L+ dose that amount of toxin
which will neutralize (say) 100 antitoxin units and leave an excess of
toxin just sufficient to kill the animal. Having found the L+ dose
of toxin (which standardizes its toxicity in terms of standard anti-
toxin), the same process is repeated, using this L+ dose of toxin mixed
with gradually diminishing amounts of the antitoxin to be stan-
dardized. That dilution of antitoxin of unknown potency which will

398 THE DIPHTHERIA BACILLUS GROUP

neutralize all except sufficient toxin to kill a 250 gram guinea-pig
in four days contains 100 antitoxin units in the example cited.
Knowing the dilution of the antitoxin, it is a simple problem to deter-
mine the number of units in 1 c.c. A good antitoxic serum should
contain from 200 to 700 units per cubic centimeter of the unconcen-
trated product.

Curative Value of Diphtheria Antitoxin. Diphtheria antitoxin
should be used as early as possible in order to obtain the maximum
curative effect. This is clearly set forth in the following table. 1

Day of Cures.

tess.
1 . . .

Treated.

Cured.

Died.

Per cei
100

. . 71

3 .

4
5
6

, , . . 39
. 1 . . 25
, .- 17

30
15
9

9
10

77
60
47

7-14

, . . 41

Indefinite

Totals 233 179 54 77

According to Donitz and others, the initial dose of antitoxin should
be large; it is believed that with large doses of antitoxin even some
of the toxin attached to the tissue cells can be neutralized. For this
purpose 4000 units is a minimal initial dose, and severe or desperate
cases are given 10,000 to 100,000 units. The antitoxin should be
repeated the following day if necessary. It is better to administer
too much than too little antitoxin. Antitoxin given subcutaneously
is least dangerous so far as danger from anaphylaxis is concerned
but the absorption is slow. Intramuscular injections, particularly
in the gluteal region, are said to be more efficient curatively. In des-
perate cases intravenous injections of antitoxin (without carbolic
acid as a preservative if possible) are indicated.

Active Immunization with Toxin- Antitoxin Mixtures. Following a
suggestion of Theobald Smith, 2 Von Behring 3 has attempted to create
active immunity to diphtheria in man by subcutaneous injections of
toxin-antitoxin mixtures which are neutral or but slightly toxic for
guinea-pigs. The few observations which are available are on the
whole encouraging, but do not justify a formal opinion of the practical
value of this procedure.

1 Quoted from Citron, Immunity.

2 Jour. Med. Research, 1907, xvi, 359.

3 Deutsch. med. Wchnschr., 1913, p. 873.

THE DIPHTHERIA BACILLUS 399

The Schick Reaction. Available evidence indicates that immunity
to infection with the diphtheria bacillus depends largely upon the
antitoxin content of the blood, and systematic studies of the anti-
toxin content of the blood of infants, children and adults by Schick, 1
Park, Zingher and Scrota, 2 Park and Zingher, 3 Kolmer and Moshage, 4
Bundesen, 5 and Moody 6 indicate that a large percentage nearly 80
per cent, of young infants, 50 per cent, of children, and nearly 90
per cent, of adults exhibit sufficient antitoxin to protect them against
the disease. The demonstration of antitoxin in the blood has been
simplified greatly by Schick, and modified somewhat by Park. 7 It
is made in the following manner: an amount of diphtheria toxin
equivalent to one-fiftieth the minimal fatal dose for a guinea-pig is
made up to a volume of 0.2 c.c. in sterile salt solution and is injected
subcutaneously, or preferably intracutaneously, in the flexor surface
of the forearm. Immediately the skin is raised somewhat as the fluid
enters the tissues. The reaction elicited depends upon the antitoxin
content of the blood, a positive reaction indicating that antitoxin
is absent, or present in minimal amounts appears within twenty-four
hours as a circumscribed area of redness and a more diffuse area of
induration measuring from one-half an inch to more than an inch
in diameter. The maximum reaction appears within forty-eight hours
and disappears within a week. The blood of a patient reacting in
this manner contains less than one-thirtieth of a unit of antitoxin
per cubic centimeter. A fainter reaction is frequently exhibited,
which is interpreted to mean that the antitoxin content of the blood
lies approximately between one-fortieth and one twenty-fifth of an
antitoxin unit per cubic centimeter. If the antitoxin content is at
least one-twentieth of a unit per cubic centimeter, the reaction is
negative; only a slight reaction results due to the wound itself.

Practically, it has been found that individuals giving a negative
reaction possess sufficient antitoxin to protect them from infection;
nurses, doctors, ward orderlies, and patients who react negatively do
not need to be immunized with antitoxin if they have been, or are,
exposed to the infection. Persons giving a mild or severe reaction
should be immunized with prophylactic doses of antitoxin.

1 Miinchen. med. Wchnschr., 1913, Ix, 2608.

2 Arch. Pediatrics, July, 1914.

3 Proc. New York Path. Soc., N. S. 1914, xiv, 151.

4 Am. Jour. Dis. Child., 1915, p. 189.

5 Jour. Am. Med. Assn., 1915, Ixiv, p. 1203.

6 Ibid., 1915, Ixiv, p. 1206.

7 Loc. cit.

400 THE DIPHTHERIA BACILLUS GROUP

Pathogenesis. Experimental Evidence of Pathogenesis. Loffler 1 ap-
pears to have been the first to attempt to establish the etiological
relationship of the diphtheria bacillus to the disease. He succeeded
in producing diphtheritic membranes on the mucous surfaces of
animals by rubbing cultures on the previously injured surface. Num-
erous laboratory accidents, where the organisms have been inadver-
tently swallowed with the subsequent development of typical clinical
diphtheria, complete the proof of the* etiological relationship of the
organism to the disease.

Animal Pathogenesis. Laboratory animals, excepting mice and
rats, are very susceptible to the diphtheria toxin. Guinea-pigs are
particularly susceptible, and the subcutaneous injection of fatal or
nearly fatal doses of broth cultures is followed after one to three days
by the appearance at the site of inoculation of a membrane, edema,
and a serosanguineous exudate. A pleuritic' and frequently a peri-
cardial exudate are found as well. There is hyperemia of the abdom-
inal organs and a very characteristic swelling and hyperemia of the
adrenals. The kidneys are also usually hyperemic. Often there are
ecchymoses and even ulcers in the gastric mucosa. No bacilli are
found in the internal organs. Intraperitoneal injections are less severe
as a rule than subcutaneous inoculations of the same dose. There
is usually some peritoneal effusion which frequently contains diph-
theria bacilli. Intratracheal inoculation after mechanical injury is
commonly followed by the appearance of a false membrane and the
animal dies of toxemia; 2 intravaginal injection after injury of the
mucosa frequently leads to a necrotic inflammation with membrane
formation. 3 Repeated applications of diphtheria toxin to the con-
junctiva of rabbits cause a marked conjunctivitis with membrane
formation. 4

Human. In man diphtheria bacilli are usually localized in the
false membranes, chiefly on the tonsils or pharynx, and these mem-
branes may extend to the nose, larynx, and mouth. The organisms
occasionally invade the blood stream. They may even extend into
the lungs causing a true bronchial pneumonia. Occasionally diph-
theria bacilli may cause rhinitis fibrinosa or simple rhinitis. 5 They
also are found in occasional cases of vulvitis gangrenosa and noma

1 Loc. cit.

2 Fraenkel, Deutsch. med. Wchnschr., 1895, 176.

3 Roux and Martin, Ann. Inst. Past., 1894, p. 625.
4 Morax and Elmassian, Ann. Inst. Past., 1898, p. 219.
5 Neumann, Centralbl. f. Bakt., 1902, xxxi, 33.

THE DIPHTHERIA BACILLUS 401

faciei. 1 Rarely, false membranes are found on the genitalia or in
cutaneous wounds, in the latter case producing a true wound diph-
theria. 2 The association with certain other organisms, particularly the
streptococcus, the staphylococcus, and B. coli, appears to increase the
virulence of the diphtheria bacillus. 3

Diphtheria is a generalized toxemia with a local infection. The
bacilli cause coagulation necrosis of the superficial cells, and an inflam-
matory membrane consisting of a serofibrinous exudate in which
fibrin and leukocytes are prominent, together with epithelial cells,
pyogenic cocci, and diphtheria bacilli in the deeper layers adjacent
to the denuded epithelium. At times the membrane strips off without
serious injury to the underlying epithelium, but in severe cases the
membrane tears away, leaving a bleeding raw surface.

There are three principal types of diphtheria: the faucial, laryngeal,
and tracheal. The incubation period is from two days to a week. An
important sequela is the postdiphtheritic paralysis, which is sup-
posed to be caused by the toxone component of the diphtheria toxin.
This is anatomically a toxic neuritis and it occurs in from 10 to 20
per cent, of all cases of diphtheria from 2 to 4 weeks after the attack.
There is no apparent relation between the severity of the attack and
the paralysis. The pharynx is most commonly affected, next in order
the eyes, leading to strabismus (ptosis). In a smaller number of cases
the heart is affected. When the heart is affected the patients not
infrequently drop dead as the result of cardiac failure. The early
Use of antitoxin usually prevents or greatly modifies the development
of postdiphtheritic paralysis.

Bacteriological Diagnosis. The principle involved in the bacterio-
logical diagnosis of diphtheria (and the diagnosis can only be definitely
established by bacteriological examination) is to make cultures from
the suspected lesions on Loffler's alkaline blood serum, to incubate
the culture from twelve to eighteen hours at 37 C., to stain the
resulting growth with Loffler's methylene blue, and to diagnose the
organisms by their characteristic morphology.

The Technic of Inoculation. Rub a sterile swab on the under surface
of the diphtheritic membrane, avoiding extraneous organisms and
avoiding touching the tongue or other parts of the mouth. Smear
this infected swab gently over the surface of the S3rum, rotating the

1 Freymouth, Deutsch. med. Wchnschr., 1898, No. 15.

2 Schottmiiller, Deutsch. med. Wchnschr., 1895, p. 273.

3 Theobald Smith, Medical Record, May, 1896.
26

402 THE DIPHTHERIA BACILLUS GROUP

swab while doing so to bring every part of it in contact with the
medium. The serum is incubated at 37 C. for twelve to eighteen
hours. It is customary in many laboratories to make a preliminary
examination of the growth on the serum after five hours' incubation,
and also to make a smear from the swab itself after the serum has
been inoculated with it. By these preliminary examinations from 30
to 60 per cent, of diagnoses may be correctly anticipated. During the
first eighteen hours of incubation diphtheria bacilli outgrow practically
all other organisms. After this time the other organisms tend to out-
grow the diphtheria bacillus.

Results. 1. Negative. Negative results may be due to several
factors: (a) the absence of diphtheria bacilli; (b) lack of care in
taking the culture, either failure to touch the infected membrane, or
making preparations immediately after the use of antiseptic gargles;
(c) improper smears and improper stains; (d) poor media; (e)
improper interpretation.

2. Positive. Positive results do not necessarily prove that the
patient has diphtheria for carriers of diphtheria bacilli are fairly
numerous and appear to be responsible, in part at least, for the
spread of the disease. From 1 to 3 per cent. 1 of healthy people harbor
fully virulent bacilli in their mouths, and about 2 per cent, of all
school children in large cities have them. Positive results may also
be obtained with avirulent strains of diphtheria bacilli. In order to
determine the virulence it is necessary to isolate the organism in
pure culture and to inject two guinea-pigs respectively with a forty-
eight-hour broth culture. The isolation is best made from cultures
on Loffler's blood serum which microscopic examination has shown
to contain diphtheria bacilli. Such a culture is emulsified in broth
and streaked out on an agar plate, or, better, upon blood agar plates.
After twenty-four hours' incubation diphtheria colonies are removed
to plain (sugar-free) broth and incubated two days. One-half a cubic
centimeter of this forty-eight-hour broth culture per 100 grams weight
of guinea-pig is injected into Pig A, and a similar amount of the broth
culture, mixed prior to inoculation with an excess of antitoxin, allow-
ing half an hour for the antitoxin to unite with the toxin prior to
inoculation, injected into guinea-pig B. Guinea-pig A should die
in from one to five days, and an autopsy should present a typical

1 Recent observations by Moss, Guthrie, and Gelien (Tr. XV Congress on Hyg.
and Demog., 1912, iv, 156) indicate that the number of carriers of virulent diphtheria
bacilli may greatly outnumber the actual cases of the disease. Their observations
showed that carriers were about four times as numerous as the cases.

THE DIPHTHERIA BACILLUS 403

picture of diphtheria poisoning. Guinea-pig B should live because
the diphtheria toxin is neutralized by the antitoxin.

The diagnosis of diphtheria by serological methods is not practical.

Dissemination and Prophylaxis. Diphtheria bacilli are spread chiefly
by contact or by carriers. Occasionally milk appears to be a vehicle
of transmission. As a prophylactic agent for destroying diphtheria
bacilli, antitoxin is one of the greatest blessings which bacteriology has
conferred on medicine. Diphtheria antitoxin is used in two ways : (a)
prophylactically ; (6) curatively. If diphtheria breaks out in a house-
hold or a hospital, those in contact with the patient should receive pro-
phylactic doses of antitoxin: that is, from 500 to 1500 units of antitoxin
repeated after fourteen days or until all danger is over, provided
the Schick test is faintly or markedly positive. (See Schick test.)
Curatively, from 3000 to 15,000 units, or in severe cases 20,000 units,
or even more, are used. In severe and desperate cases the antitoxin
should be introduced intravenously, preferably using antitoxin pre-
pared without preservatives for this purpose. Antitoxin must be
used early. If it is used early the mortality is reduced more than 50
per cent. If the serum is used within the first twenty-four hours,
the prognosis is favorable in at least 95 per cent, of the cases. The
general death rate prior to the introduction of antitoxin was from
25 to 33 per cent.; since the use of antitoxin it varies from 3 to 14 per
cent.

In a certain proportion of cases of diphtheria treated with anti-
toxin, usually from eight to fourteen days after the administration
of antitoxin, rashes and painful joints develop, together with fever,
angioneurotic edema, swollen lymph glands, and albuminuria. This
is the so-called serum disease, which is usually particularly severe in
asthmatics, in whom there occasionally develops a true bronchial
spasm with respiratory embarrassment. In a few cases, less than one
in ten thousand, sudden death may occur within five to fifteen minutes
after the injection. At autopsy there is usually found a persistent
thymus. These are cases of status lymphaticus. This sudden death
is not due to the antitoxin, but to the proteins in the horse serum in
which the antitoxin is contained. 1 If there is reason to suspect that
the administration of antitoxin will result seriously, a few drops (not
more than a quarter of a cubic centimeter) should be injected, and

1 It should be remembered in this connection that man is less susceptible than a
guinea-pig to serum diseases, and, furthermore, it ordinarily takes about 5 c.c. of horse
serum to bring about the anaphylactic reaction in sensitized guinea-pigs. Proportion-
ately, it would take 200 c.c. to induce the same symptoms in man.

404 THE DIPHTHERIA BACILLUS GROUP

the remainder after half an hour. The first small injection indicates
the susceptibility of the patient; if no symptoms appear the full dose
may be given with impunity; even if symptoms do appear the anaphy-
lactic shock is aborted by the first injection and the remainder may
be given at the end of an hour.

BACILLI SIMILAR TO THE DIPHTHERIA BACILLUS.

There is a group of bacteria closely related to Bacillus diphtherise,
but differing from it either in virulence, morphology, or both. Certain
of these organisms exhibit the characteristic morphology, staining and
cultural reactions of the diphtheria bacillus, but do not form toxin;
these strains, which are occasionally found in healthy and diseased
throats, may be tentatively regarded as non-toxin-producing variants
of the type organism.

In addition to the non-virulent but morphologically typical diph-
theria bacilli, other bacteria have been described which resemble
Bacillus diphtheriae superficially, but differ from it in certain impor-
tant details. Two principal types have been recognized: Bacillus
hofmanni and Bacillus xerosis.

Bacillus Hofmanni. Bacillus hofmanni appears to have been
first observed by Loffler; 1 somewhat later Hofmann 2 studied it in
considerable detail.

Morphologically the Hofmann bacillus is somewhat shorter and
relatively thicker than Bacillus diphtherise, and more uniform in size
and shape. Stained with Loffler 's methylene blue, but a single
unstained area is observed typically, the organism being somewhat
diplococcoid in form under these conditions.

Growth is relatively more luxuriant in artificial media than that
of the diphtheria bacillus, and no toxin is produced in sugar-free broth.
The organism ferments no sugars, not even dextrose.

Bacillus hofmanni is found not infrequently in normal and diseased
throats, and occasionally in the nasal secretion.

Bacillus Xerosis. Bacillus xerosis, first observed by Bezold, 3 was
obtained in pure culture from several cases of a chronic type of con-
junctivitis known as xerosis by Kirschbert and Neisser. 4 Recently
the organism has been isolated repeatedly from the healthy conjunc-

1 Centralbl. f. Bakt., 1887, ii, 106.

2 Wien. klin. Wchnschr., 1888, Nos. 3 and 4.
sBerl. klin. Wchnschr., 1874, p. 408.

4 Breslauer arztl. Ztschr., 1883, No. 4.

BACILLI SIMILAR TO THE DIPHTHERIA BACILLUS 405

tiva and the nasal secretion. The morphological similarity between
Bacillus hofmanni and Bacillus xerosis has doubtless led to confusion
in the past. Knapp 1 has studied the fermentation reactions of the
group and has shown that within the diphtheria group three cultural
types are recognizable, as follows:

Dextrose. Saccharose.

Bacillus diphtheria? .. . U . 1 . V . acid" alkaline

Bacillus hofmanni . ' . . ... . . alkaline alkaline

Bacillus xerosis .- . . . . . - : - acid acid

Bacillus Hodgkini. Hodgkin's disease, a malignant granulomatous
lymphatic infection long regarded as a special type of infection with
the tubercle bacillus, is now generally regarded as an infectious entity

FIG. 57. Pseudodiphtheria bacilli. (Park.)

quite apart from tuberculosis. The etiology remained obscure until
Negri and Mieremet 2 published a description of a pleiomorphic, diph-
theroid bacillus obtained from two undoubted cases. The organism,
which was found to be Gram-positive, received the name Corynebac-
terium granulomatis maligni. Bunting and Yates 3 have recovered
a similar pleiomorphic bacillus from several cases of Hodgkin's disease.
Initial cultures were obtained upon Dorset's egg medium. Subse-
quent development upon ordinary media gave the following cultural
reactions: gelatin not liquefied; little or no change in litmus milk;
an adherent growth in broth tubes with the gradual accumulation of
a slimy sediment. The colonies upon serum and agar are not
characteristic.

1 Jour. Med. Research, November, 1904, vol. xii, 475.

2 Centralbl. f. Bakt., Orig., 1913, Ixviii, 292.

3 Arch. Int. Med., 1913, xii, 236. See also Bull. Johns Hopkins Hosp., 1915, xxvi,
376, for relation of pseudodiphtheria bacilli to leukemia, pseudoleukemia, and Banti's
disease.

406 THE DIPHTHERIA BACILLUS GROUP

Morphologically the organism is variable in shape. Bacillary
forms predominate in young cultures, but the bacilli exhibit a marked
tendency toward coccoid elements after prolonged cultivation.

The etiological relationship of the organism, 1 which received the
name Corynebacterium hodgkini, to Hodgkin's disease is as yet un-
determined, but vaccines injected into several typical cases caused
a definite recession in the size of the enlarged glands. The permanence
of this recession must await final reports.

Bunting and Yates 2 injected their organism into monkeys, and a
chronic lymphadenitis with an increased mononu clear and eosinophile
count resulted. A polymorphonuclear leukocytosis was not observed.
They conclude that the anatomical lesions were very similar to those
observed in the early stages of Hodgkin's disease in man.

1 See excellent resume by Bloomfield, Arch. Int. Med., August, 1915, p. 197.

2 Jour. Am. Med. Assn., 1913, Ixi, 1803; ibid., 1914, Ixii, 516.

CHAPTER XXI.

THE HEMORRHAGIC SEPTICEMIA GROUP.

THE Hemorrhagic Septicemia or Pasteurella Goup of bacilli com-
prises a number of organisms which possess in common peculiarities
of morphology, similarity of cultural characters and great patho-
genicity for animals.

Morphologically they are short, ovoid bacilli of relatively large
diameter, measuring about 0.5 to 0.8 micron in the widest part, and
varying in length from 0.8 to 1.5 microns. The organisms usually
exhibit marked pleiomorphism in old lesions and in old cultures.
They are non-motile, uniformly Gram-negative, and exhibit a marked
tendency to bipolar staining; the stainable substance is collected at
the ends of the bacillus, separated by a central, faintly stainable area.

The hemorrhagic septicemia bacilli grow well upon ordinary cul-
tural media, and they are chemically relatively inert. Indol is pro-
duced by certain types, but not by all. Gelatin is not liquefied. Acid,
but no gas, is formed in dextrose, lactose and many hexoses. The
fermentation of other sugars and starches has not been thoroughly
studied.

The type of infection induced is usually an acute generalized
septicemia, which, because of punctate hemorrhages on serous sur-
faces, and in the internal organs, is called hemorrhagic septicemia.
Inflammation of the intestinal tract and frequently the respiratory
tract is usually an important feature of the infection.

The most important animal diseases are chicken cholera, swine
plague, rabbit septicemia, and a similar disease of cattle and wild
herbivora. Plague is the disease of man which most closely approaches
hemorrhagic septicemia of the lower animals.

The lesions caused by Bacillus pestis in experimental animals and
in the naturally occurring disease in rodents present many similarities
to the hemorrhagic septicemias, and occasionally a distinction must
be made between the plague bacillus and other members of the group.
The Indian Plague Commission state that Bacillus pestis may be
differentiated from the other members of the group by its ability to
develop and produce acid (but no gas) in dextrose and mannite media

408 THE HEMORRHAGIC SEPT ICE MI A GROUP

containing bile salts, particularly sodium taurocholate ; the other
organisms will not grow in this medium.

Bacillus Pestis. Plague, the most dreaded of the acute epidemic
diseases, has at somewhat irregular intervals swept over parts of the
Orient, and during the earlier centuries of the Christian era even
invaded Europe. The great epidemics of the third and the fourteenth
centuries caused widespread death; literally millions perished, and
the effect upon the residual population was most distressing. The
disease has recently become endemic on the western coast of the
United States, the reservoir of infection being certain rodents.

FIG. 58. Plague bacillus, bouillon culture, methyl ene-blue stain showing bipolar
staining. X 1000. (Kolle and Hetsch.)

The causative organism, Bacillus pestis, was isolated and described
almost simultaneously by Kitasato 1 and Yersin 2 from the purulent
contents of buboes, the lymph glands, the blood and the cerebrospinal
fluid. Later the specificity of the organism was established by labora-
tory accidents and by the very comprehensive studies of the British
Indian Plague Commission.

Morphology. Bacillus pestis is a small thick bacillus with rounded
ends, wilich occurs singly or in pairs as a rule, although short chains
of three to six elements are occasionally seen. The organism is not
characteristically rod-shaped, rather it approaches in outline a some-
what ovoid cell. The size varies within the comparatively narrow
limits of 0.5 to 0.7 micron in diameter at the widest part and 1.5
to 1.8 microns in length. The bacilli are very pleiomorphic and exhibit
great variation in size and shape according to the medium and age

1 Lancet, 1894. 2 Ann. Inst. Past., 1894, p. 662.

BACILLUS PEST IS 409

of the culture. In young cultures and fresh lesions the typical ovoid
shape predominates, but in older cultures and lesions considerable
variation in size and outline is very common. The addition of 2 to
3 per cent, of salt to artificial media greatly increases the proportion
of involution forms. Bacillus pest is is non-motile and possesses no
flagella. Spores are not produced. Zettnow 1 and Albrecht and Ghon 2
state that the organism forms a capsule. The organism stains readily
with anilin dyes, and it is Gram-negative. Dilute methylene blue
colors the bacilli in a characteristic manner. This is best observed
when the bacilli are fixed with absolute alcohol for thirty minutes in
place of heating. 3 The centre of the cell is practically uncolored and
the stain able substance is seen as a deeply colored granule at each

FIG. 59. Plague bacillus. Involution forms from culture on 3 per cent, salt agar.
X 1000. (Kolle and Hetsch.)

end of the rod bipolar staining. Pleiomorphic forms are usually
stained faintly or scarcely at all by this method.

Isolation and Culture. The plague bacillus grows readily on ordinary
media and pure cultures are usually readily obtained from the
aspirated contents of unopened buboes or other lesions, and frequently
from the blood stream in septicemic cases. Colonies on agar after
twenty-four hours' incubation are small, somewhat irregular in out-
line, translucent, and not distinctive. Similar growths appear upon
gelatin after two to three days' incubation. The medium is not
liquefied. The bacilli develop with considerable luxuriance in broth
forming a granular sediment in the bottom of the tube and frequently
adhering to the sides. The addition of a drop of neutral oil as cocoa -

1 Ztschr. f. Hyg., 1896, xxi, 165. 2 Centralbl. f. Bakt., 1899, xxvi, 362.

3 Kossel and Overbeck, Arb. a. d. kais. Gesamte, 1901, xviii, 117.

410 THE HEMORRHAGIC SEPTICEMIA GROUP

nut oil provided the culture is maintained free from all vibration,
causes a characteristic "stalactite" growth; the organisms grow
down from the oil droplets as filaments (which have been likened to
stalactites) until they even reach the bottom of the tube. Chains of
bacilli are most characteristically developed in this medium. The
addition of 2 to 3 per cent, common salt to broth or agar stimulates
the formation of very irregular involution forms. Milk is not coagu-
lated, but a slight permanent acidity gradually develops. Growth on
coagulated blood serum or ascitic agar, although somewhat more
luxuriant than on ordinary laboratory media, is not characteristic.

Bacillus pestis is an aerobic organism; it fails to develop with its
customary vigor in the absence of oxygen. Unlike a majority of
pathogenic bacteria, the optimum temperature of growth is about 30
C.; growth ceases below 10 C. and above 40 C. The viability of the
organism in cadavers is considerable; they may remain alive for
several weeks. In pus and sputum viable cultures may be obtained
after one or even two weeks. Exposure to sunlight kills the bacilli
within a few hours, and naked germs (unprotected by mucus or
other protein envelope) are rapidly killed by drying. An exposure
to 58 C. for an hour, or 100 C. for a few minutes is fatal: 5 per cent,
carbolic acid and 1 to 1000 bichloride of mercury kill the organisms
within fifteen minutes. The virulence of the bacilli diminishes rather
rapidly in artificial media as a rule.

Products of Growth. Indol is not produced in sugar-free broth.
Acids, but no gas, are produced in dextrose, lactose, galactose, mannite
and maltose, but not in saccharose, sorbite, dulcite, and inulin.

No enzymes have been demonstrated in cultures of plague bacilli.

Toxins. Filtered cultures of plague bacilli possess little or no
toxicity, although old broth cultures, freed from bacteria by filtration
through unglazed porcelain, may exhibit slight toxic action. It is
probable that this toxicity is referable to some endotoxin which has
been liberated in the medium during the gradual autolysis of the
organisms. The symptoms of plague are attributed to the action of
endotoxins which are liberated within the host as the organisms dis-
integrate. The virulence of plague cultures is variable. Freshly
isolated strains may occasionally exhibit almost no virulence for
experimental animals, although as a rule they are very virulent.
Prolonged cultivation upon artificial media usually results in a decided
lowering of virulence, although here again exceptions are met with. 1

1 McCoy and Chapin, Pub. Health Bull.. January, 1912, No. 53, p. 1.

BACILLUS PESTIS 411

Pathogenesis. Animal. "Plague is primarily a disease of rodents,
and secondarily and accidentally 1 a disease of man." 2 The reservoir
of plague appears to be certain rodents; the disease exists in chronic
form in the marmot (Arctomys bobac) of India, and has recently been
discovered in the western United States as a chronic disease in the
ground squirrel (Citellus beechyi) by Wherry, 3 whose observations
have been confirmed by McCoy. McCoy 4 and Chapin 5 have found
that a smaller member of the squirrel family, Ammospermophilus
lecurus) is also susceptible to infection with Bacillus pestis. The
various members of the genus Mus Mus norwegicus, Mus rattus,
Mus alexandrinus, and probably Mus musculus, "are the producers
of acute outbreaks, the conduit for the carriage of the virus from its
perpetuating reservoir to the body of man." Guinea-pigs are some-
what more susceptible to infection with the plague bacillus than
rodents; the disease appears to have a seasonal distribution among
rodents 6 in India, and these epidemic periods coincide in time with
epidemics in man. Immediately following epidemic periods con-
siderable numbers of rodents appear to be relatively non-susceptible
to infection. Rabbits 7 and monkeys 8 are also susceptible. Dogs
and cats are more refractory; herbivora appear to be practically
immune, at least to natural infection.

The lesions observed in rats are striking and important because
plague epidemics usually appear about two weeks earlier among
these animals than in man. Infection may take place through infected
fleas from other rats, from ingestion of dead plague-infected animals,
or by inhalation. The lesions are those of an hemorrhagic septicemia;
upon laying open the animal, 9 the inguinal and axillary glands are
usually enlarged (buboes), markedly injected and frequently hemor-
rhagic. The contents may be firm, or, less commonly, purulent.
The peritoneal and pleufal surfaces are red and injected, and there
is an excess of fluid in both cavities. The spleen is enlarged, engorged
and moderately soft, and the liver usually presents a mottled appear-
ance, due to punctiform hemorrhages and areas of necrosis which

1 This statement may possibly require modification in connection with the pneumonic
type of plague in man (see human pathogenesis) .

2 Rucker, Public Health Reports, July 19, 1912, p. 1130.

3 Public Health Reports, 1908, xxiii, 1289; Jour. Inf. Dis., 1908, v, 485.

4 Public Health Bull., April, 1911, No. 43.

5 Ibid., January, 1912, No. 53, p. 15.

6 See Jour. Hyg., 1908, viii, 266, for details.

7 McCoy, Public Health Bull., April, 1911, No. 43.

8 Wyssokowitsch and Zabolotny, Ann. Inst. Past., 1897, xi, 665.

9 Which should be done after treatment with an insecticide to kill ecto-parasites.

412 THE HEMORRHAGIC SEPTICEMIA GROUP

appear yellowish in contrast to the hemorrhagic points. A simple
inspection suffices as a rule to establish a correct diagnosis, although
cultures and smears should be prepared. Occasionally rats are sub-
mitted for diagnosis which are badly decomposed. The rapid over-
growth of adventitious bacteria makes the isolation of Bacillus pestis
difficult by ordinary methods. Albrecht and Ghon 1 have shown that
the plague bacillus, even in the presence of large numbers of con-
taminating bacteria, may be obtained in pure culture by rubbing the
suspected material upon the freshly shaved abdomen of a guinea-
pig. The plague bacillus readily penetrates the skin and causes a
rapidly fatal generalized infection with characteristic lesions. It may
be obtained in pure culture from the internal organs. Fritsche 2 has
found that other bacteria, even of the hemorrhagic septicemia group,
fail to penetrate the skin and infect the animal. This cutaneous test
is of great diagnostic importance.

McCoy and Chapin 3 have described a disease superficially resem-
bling plague in its pathological anatomy caused by Bacillus tularense.
The disease is readily transmitted to guinea-pigs, rabbits and mice,
less readily to rats. Wherry and Lamb 4 have recently isolated the
organism from an epizootic among wild rabbits and from a human
case presenting corneal ulcerations and lymphadenitis.

Man. The atria of infection are chiefly the skin and the respira-
tory tract, giving rise to two general types of the disease, glandular
and pneumonic plague. Rarely a localized cutaneous lesion, plague
carbuncle, is met with where the focus of localization of the organisms
appears to be very circumscribed. Cases of pneumonic plague which
develop sporadically during epidemics of the bubonic type do not as
a rule appear to spread rapidly; on the contrary, during epidemics
in which the pneumonic type predominates the infectivity from man
to man is very great. No theory has been presented in explanation
of this very unusual phenomenon, and the origin of the pneumonic type
of the disease is not definitely known. 5 Typical pneumonic plague
resembles lobar pneumonia in its symptomatology, and the fatalities

1 Denkschrift der math-Naturw. Klasse der kaiserl. Akad. d. Wissensch., Wicn.,
1898, Ixvi.

2 Arb. a. d. kais. Gesamte, 1902, vol. xviii.

3 Jour. Inf. Dis., 1912, x, 61; Public Health Bull., April, 1911, No. 43; ibid., January,
1912, 53.

4 Jour. Inf. Dis., 1913, xv, 331 ; Jour. Am. Med. Assn., 1914, Ixiii, 2041.

6 Animal experiments suggest that attenuated cultures of Bacillus pestis which fail
to kill guinea-pigs by subcutaneous inoculation may give rise to a fatal infection when
inoculated into the respiratory tract, and that the virulence of cultures may be recovered
by this process. The high mortality observed during epidemics of pneumonic plague
in man may be a similar phenomenon.

BACILLUS PESTIS 413

are very great. Death usually intervenes in less than a week. The
marked cardiac depression which is a feature of this type of plague
is of some differential diagnostic value. One or more lobes are infected,
the inflammation being catarrhal in nature. Enormous numbers of
plague bacilli are coughed up with the sanguinoserous exudate, which
are readily transmitted to doctors, nurses and attendants by droplet
infection. Very frequently a generalized invasion of the blood stream
occurs.

Bubonic plague, the most common type of the disease in man, is
essentially a localization of plague bacilli which have gained entrance
to the tissues through the skin in the regional lymph glands. The
inguinal glands are more commonly invaded; next in order of fre-
quency are the axillary, then the cervical glands. The glands
are violently inflamed, and not infrequently soften and caseate. A
generalized blood infection septicemic plague may occur either
secondarily following the development of a bubo or of pneumonic
plague, or, less commonly, as an initial generalized invasion following
very shortly after infection and before the bubo becomes conspicuous.
In such cases the organisms may be obtained in pure culture from the
blood stream, and in about 20 per cent, of cases may be actually
demonstrated in stained preparations of the blood by microscopical
examination.

Immunity and Immunization. Recovery from one attack of plague
in man almost always confers lasting immunity. Immunity has been
induced in animals monkeys, rats and guinea-pigs by inoculation
with living avirulent cultures. Usually a moderate reaction is noticed,
and a bubo may even form, but the animal recovers, and even after
months successfully resists several times the fatal dose of virulent
organisms. This method is far too dangerous for human practice.
Bacillus pseudo tuberculosis rodentium, an organism that occasionally
is found in diseased rats producing lesions superficially not unlike
plague, will immunize rats against Bacillus pestis and vice versa. This
bacillus must be sharply differentiated from the plague bacillus in
the microscopic diagnosis of plague in rodents. It fails to infect guinea-
pigs by the cutaneous method, however, and is less virulent for labora-
tory animals. Cultures of plague bacilli heated to 50 C., or killed
with chemicals, as phenol or alcohol, have also been employed success-
fully, but the degree of resistance to subsequent infection is less. 1
Specific bacteriolysins and agglutinins develop during the immunizing
process. .

1 Kolle and Otto, Ztschr. f. Hyg., 1903, xlv, 507.

414 THE HEMORRHAGIC SEPTICEMIA GROUP

Active immunization of man against plague has been accomplished
by Haffkine, using broth cultures of plague bacilli grown in shallow
layers of broth containing droplets of cocoanut or other neutral oil
on the surface to increase the development of the organisms by stalac-
tite formation; after about six weeks' incubation, during which time
several crops of bacilli develop and sink to the bottom, the culture
is heated to 60 to 65 C. for an hour, and 0.5 per cent, phenol is
added. 2 to 3.5 c.c. of the killed culture are injected subcutaneously
into adults, proportionately smaller amounts into children. Usually
a second injection is given, somewhat larger in amount, after ten
days. The German Plague Commission 1 used forty-eight-hour agar
cultures of virulent plague bacilli emulsified in salt solution and
sterilized at 65 C.; 0.5 per cent, phenol was added as a preservative.
The amount for injection into an adult was the equivalent of one
agar culture of the organism. Available evidence indicates that
prophylactic inoculation against plague reduces materially both the
morbidity and mortality of the disease. The protection, as the
statistics show, is by no means absolute, and ' it appears that the
duration of resistance to infection is indeterminate, probably on the
average several months. A serum obtained from horses immunized
against plague bacilli has been prepared, but its use in man has on
the whole been irregular and disappointing; the chief practical use
appears to be in those cases where exposure to infection is reason-
ably certain, as for example, in those attending plague patients. The
excessive cost of the serum is prohibitive for general use.

Transmission and Plague. The Interim Report of the Advisory
Committee for Plague Prevention in India 2 contains a very excellent
summary of the mechanism of plague transmission by the flea. In
bubonic plague, the most common type seen in man, the plague bacilli
are locked up in the body, as it were, and can not of themselves escape
to other hosts. The rat is usually the source of infection in plague
epidemics, and rat fleas, Xenopsyllus cheopis, transmit the disease
from rat to rat and from rat to man. When the host dies (rat or man)
its ectoparasites escape if possible to living hosts. It was shown by
the Indian Commission 3 that fleas from plague-infected rats frequently
contained large numbers of plague bacilli in their intestinal tracts,

1 Gaffky, Pfeiffer, Sticker, and Dieudonne, Bericht u. d. Thatigkeit dcr zur Erfor-
schung der Pest im Jahre 1897, etc., Berlin, 1899.

2 Jour. Hyg., 1910, x, No. 3.

3 See Jour. Hyg., 1906, 1907, 1908, 1910, for a most complete discussion of the relation
of fleas to the transmission of plague.

BACILLUS PESTIS 415

and that the bacilli were present at least three weeks after the last
feeding. In the absence of fleas no infection takes place, at least in
man. The bite of the infected flea may result in infection, or, since
the feces of the flea are usually deposited during feeding, laden with
plague bacilli, the irritant flea bite may lead to scratching of the area,
resulting in the "rubbing in" of the bacilli deposited with the rat
feces. Epidemics of pneumonic plague are spread by droplet infection.
Preventive measures include the appropriate care of the patient and
measures to reduce the rat population. This is accomplished by careful
disposal of all garbage, rat proofing all houses and granaries, and an
active campaign against rats by poison, destruction of nests and run-
ways, and the creation of rodent-free zones of considerable magnitude
around settlements.

FIG. 60. Influenza bacillus from sputum. X 1200. (Kolle and Hetsch.)

Bacteriological Diagnosis. Human The juice of buboes, 1 of lymph
glands, of petechiae, the blood, the sputum from pneumonic cases,
and occasionally the urine contain plague bacilli in large numbers.
They may be obtained readily from the spleen, liver, lungs and kidneys
of the cadaver.

Animal. The postmortem appearance of plague-infected rats is
very characteristic.

Microscopical Diagnosis. The presence of Gram-negative ovoid
short bacilli in considerable numbers in films prepared from material
outlined above is very suggestive, but not conclusive evidence of
infection with Bacillus pestis. In man the evidence is stronger than

1 Buboes which have suppurated frequently do not contain plague bacilli, or plague
bacilli in association with extraneous organisms. Even if buboes have not developed,
the lymphatic glands usually contain the bacilli.

416 THE HEMORRHAGIC SEPTICEMIA GROUP

in rats, for in the latter Bacillus pseudotuberculosis rodentium, Bacil-
lus tularense, and other organisms may be present, which produce
lesions superficially not unlike those of plague.

Cultural Diagnosis. Prepare agar plates from the contents of
enlarged glands or other material incubate at 30 C. (or 37 C.),
and isolate colonies in pure culture. Blood, collected aseptically,
should be plated out on agar. From the pure colonies inoculate 3
per cent, salt agar and examine after twenty-four hours for involution
forms; make the "stalactite" test in broth containing .a few drops of
neutral oil. (The culture must be kept in an absolutely quiet environ-
ment to obtain stalactite growth.) This reaction is not absolutely
distinctive for other members of the Hemorrhagic Septicemia Group
may also develop in this manner.

Animal Inoculation. A small amount of culture inoculated at the
root of the tail of a rat subcutaneously or intranasally in a guinea-
pig will cause death within three to five days with characteristic
lesions. The organism may be recovered from the internal organs.
If the material for inoculation be mixed with adventitious bacteria,
the cutaneous method of inoculation 1 of guinea-pigs will give positive
results and the organisms may be recovered in pure culture from the
internal organs postmortem.

1 Albrecht and Ghon, loc. cit.

CHAPTER XXII.

HEMOGLOBINOPHILIC BACILLI: KOCH-WEEKS,
MORAX-AXENFELD AND DUCREY BACILLI.

BACILLUS INFLUENZA.

BACILLUS INFLUENZA was isolated in pure culture and described by
Pfeiffer. 1

Morphology. It is an extremely small bacillus, one of the smallest
known, measuring from 0.2 to 0.3 micron in diameter and from 0.5
to 1 micron in length. The ends are rounded and it occurs singly or
in pairs, rarely in short chains. The organism is non-motile, and no
flagella have been demonstrated. Spores and capsules are not pro-
duced. Ordinary anilin dyes do not color the organism readily,
but Pfeiffer 2 has shown that dilute carbol-fuchsin 3 stains it readily.
StaineoTwith methylene blue or dilute carbol-fuchsin, the ends of the
bacilli are colored somewhat more deeply than the centre, suggesting
a bipolar distribution of the cytoplasm similar to that exhibited by
the bacteria of the hemorrhagic septicemia group. The organism is
Gram-negative.

Isolation and Culture. Bacillus influenzse is an obligately hemoglo-
binophilic organism; it does not grow outside the body in the absence
of hemoglobin, although the amount of this substance required to
encourage development may be so small in amount that it is invisible
to the eye. 4 The organism may be isolated from bronchial mucus
by Pfeiffer's method. The mucus is washed several times with sterile
water to remove extraneous bacteria, then spread upon blood agar
plates. Human, pigeon or rabbit blood added to neutral plain agar
creates a favorable medium for the bacillus. The colonies which
appear after twenty-four to forty-eight hours' incubation at 37 C.
are very minute, clear and colorless. They may require a lens for
their recognition. The hemoglobin is not visibly changed in appear-
ance and no hemolysis occurs. Massive cultures of influenza bacilli

1 Deutsch. med. Wchnschr., 1892, No. 2.

2 Ztschr. f. Hyg., 1893, xiii, 357.

3 One part carbol-fuchsin, 9 parts water.

*Ghon and Preyss, Centralbl. f. Bakt., 1902, xxxii, 90; 1904, xxxv, 531.

418 HEMOGLOBINOPHILIC BACILLI

may be obtained in blood bouillon. One c.c. of sterile defibrinated
pigeon or rabbit blood is added to 50 c.c. of neutral nutrient broth.
After incubation to demonstrate its sterility, the medium is ready
for inoculation. 1 Attempts to grow the bacilli on hemoglobin-free
media have been uniformly negative; no development occurs in
ordinary media.

Bacillus influenzse is an aerobic bacillus. It has not been grown in
the absence of oxygen. Growth does not take place below 25 C. nor
above 42 to 43 C. The optimum temperature is 37 C. It is possible
to maintain cultures by transplanting them upon fresh hemoglobin
media at intervals not greater than five days. Drying is rapidly fatal
to influenza bacilli; dried in mucus the organisms are not viable
after one to three days. They may remain alive in moist mucus for
nearly two weeks, however. Ten minutes' exposure at 57 C. kills
them and ordinary chemical disinfectants, bichloride of mercury 1 to
1000, and 5 per cent, carbolic acid, destroy them in a few minutes.

Products of Growth. The nature of the products of metabolism of
the influenza bacillus are unknown. Enzymes have not been detected
in cultures of tjhe organism and soluble toxins have not been demon-
strated. There is evidence that the cell substance of the bacilli is
toxic; it is probable that this toxic substance is endotoxic in character.

Pathogenesis. The direct evidence of the etiological relationship
of B. influenza? to the disease influenza rests upon a single laboratory
infection with a pure culture of the organism. The hands were con-
taminated, and within twenty-four hours a typical attack of influenza
developed. The organisms persisted in the sputum for two months. 2

Animal. Influenza is essentially a disease of man. Pfeiffer 3 has
shown that mice, rats, guinea-pigs, swine, dogs, and cats are refrac-
tory to infection with the living organisms. The introduction of
hemoglobin broth cultures of B. influenzse through the chest wall of
monkeys frequently causes a transient febrile reaction, and a catarrhal
bronchitis which, however, is hot clinically comparable to influenza
of man. The animals recovered rapidly. Rabbits are susceptible to
the endotoxin of the influenza bacillus. 4 The injection of large num-
bers of living or killed organisms causes dyspnea and a paralysis of the
leg muscles. Frequently the animals die.

1 Delius and Kolle, Ztschr. f. Hyg., 1897, xxiv, 327.

2 Tedesco, Centralbl. f. Bakt., Orig., 1907, xliii, 323.

3 Ztschr. f. Hyg., 1893, xiii, 357.

4 Pfeiffer, loc. cit.; Cantani, Ztschr. f. Hyg., 1896, xxiii, 265.

BACILLUS INFLUENZA 419

Man. Influenza occurs . pandemically at infrequent intervals:
during interpandemic intervals the organism causes somewhat local-
ized epidemics of "grippe," and not infrequently appears to be the
causative factor in "grippe colds." The bacilli persist in the respira-
tory tract as "opportunists" and are frequently detected in the lungs
of consumptives. Invasion takes place through the respiratory tract,
usually by droplet infection, and frequently spreads by continuity
to the lungs, where a purulent broncho- or lobular pneumonia develops
in typical cases. Pleurisy is a frequent complication, usually caused
by a secondary infection with pneumococci or streptococci. The in-
fluenza bacillus rarely causes pleurisy. Enormous numbers of bacilli
are coughed up in the sputum. The incubation period is brief, from
one to three days as a rule. Influenzal meningitis, 1 pharyngitis and
laryngitis 2 and conjunctivitis 3 are not uncommon. The occurrence
of influenza bacilli in the blood has been a matter of controversy.
Canon, 4 Bruschettini 5 and Ghedini 6 have isolated bacilli from the
blood of patients at the height of the disease which they believe to be
B. influenzse. Slawyk 7 has reported a case of generalized infection
with the influenza bacillus which would appear to confirm these
observations. Other investigators have questioned the accuracy of
this work and lay stress upon the incomplete diagnosis of the organisms
obtained from the patients. The question can not be regarded as
definitely settled at the present time.

Immunity. Attempts to induce immunity in experimental animals
have been unsuccessful. Relapses are common in man, and there is
no evidence of immunity as the result of recovery from the disease.

Bacteriological Diagnosis. 1. Sputum raised from the deeper air
passages is spread upon slides, air dried, fixed, and stained with dilute
carbol-fuchsin. Large numbers of minute organisms colored pink
with a tendency toward bipolar staining are suggestive of the influenza
bacillus. There is no tendency toward a definite arrangement of the
bacilli. They are frequently found in leukocytes.

2. Cultural. Blood agar plates are made by depositing a generous
drop of human, rabbit or pigeon's blood in the centre of an agar plate.

'Pfuhl, Ztschr. f. Hyg., 1897, xxvi, 112; Frankel. Ztschr. f. Hyg., 1898, xxvii, 329;
Jundell, Jahrb. f. Kinderheilk., 1904, lix, 777.

2 Treitel, Arch. f. Laryngol., 1902, xiii, 147.

'Pretori, Arch. f. Augenheilk., 1907, Ivii, 97; Possek, Wien. klin. Wchnschr., 1909,
No. 10.

4 Deutsch. med. Wchnschr., 1892, No. 3; Arch. f. Anat. u. Phys., 1893, cxxxi, 401.

5 Riforma Med., 1893, viii, 783. Centralbl. f. Bakt., Orig., 1907, xliii, 407.
7 Ztschr. f. Hyg., 1899, xxxii, 443.

420 HEMOGLOBINOPHILIC BACILLI

Mucus raised from the deeper air passages is thoroughly washed in
sterile water and emulsified in broth or water, selecting for the pur-
pose purulent masses by preference, and streaked out radially from the
drop of blood. After twenty-four to forty-eight hours' incubation
the plate is examined with a lens for very minute, clear, homogeneous
colonies which should be removed to blood agar slants. When growth
occurs transfer some of it to plain agar. No further development
occurs unless some blood has been removed with the organisms. The
failure of the bacteria to develop on media free from hemoglobin is
distinctive.

3. Serological. The serum diagnosis of influenza has been unsuc-
cessful.

Dissemination and Prophylaxis. Influenza bacilli are distributed
chiefly by droplet infection. Carriers are said to be common. Prophy-
laxis is the same as for any respiratory disease.

BACILLUS PERTUSSIS.

The etiology of pertussis (whooping-cough) has been a subject of
controversy for several years. The problem is complicated by the
rather general occurrence of influenza-like bacilli in the sputum and
bronchial exudate from cases of whooping-cough. A clean-cut dif-
ferentiation between these influenzoid bacilli and Bacillus pertussis
described by Bordet and Gengou 1 has been difficult and has doubtless
led to confusion in the past. It is now generally conceded that the
Bordet-Gengou bacillus is worthy of serious consideration as the
etiological factor of whooping-cough.

Morphology. B. pertussis is somewhat larger than B. influenzse,
measuring 0.3 micron in diameter and varying in length from 0.5 to
1.5 microns, the average length being about 1 micron. It occurs singly
and in groups, less commonly in pairs. The organism has rounded
ends; frequently it is almost ovoid in shape. The organism is non-
motile and possesses no flagella. Neither capsules nor spores have
been demonstrated. It stains poorly with ordinary anilin dyes and
is Gram-negative. Carbol methylene blue, carbol toluidine blue and
dilute carbol-fuchsin stain it readily. Methylene blue is also a satis-
factory stain. The organisms stain irregularly, particularly when
grown in artificial media. In young cultures and in sputum they
appear frequently with the ends stained more deeply than the centre,
resembling in this respect the influenza bacillus.

1 Bull. Acad. de Med. Belgique, July, 1906; Ann. Inst. Past., 1906, xx, 731.

BACILLUS PERTUSSIS 421

Isolation and Culture. Unlike the influenza bacillus, B. pertussis
can be made to grow in media which do not contain hemoglobin. For
initial growths outside of the human body, however, Bordet and
Gengou have recommended a potato-glycerin-blood agar medium
which is claimed to be far more efficient than blood agar. 1 The Bor-
det-Gengou bacillus is more readily isolated from the bronchial secre-
tion during the first paroxysms 2 than later in the disease. Cultures
are obtained from bronchial mucus which has been washed several
times in sterile water, then spread on the surface of the potato medium
and incubated at 37 C. After twenty-four to forty-eight hours'
incubation colonies appear as very minute, transparent growths
which resemble dew drops; colonies of B. influenzse frequently develop
at the same time, but the colonies of the latter are somewhat larger
than those of B. pertussis. Secondary transplantations of B. pertussis
upon fresh potato-glycerin-blood agar grow more luxuriantly than
of B. influenzse, however, and after repeated transfers the Bordet-
Gengou bacillus will grow upon ascitic agar. The influenza bacillus
will not grow in media free from hemoglobin. Ordinary media, unless
ascitic fluid or blood serum is added, are wholly unsuited for the
growth of B. pertussis.

B. pertussis, like the influenza bacillus, is an aerobic organism.
Anaerobic development has not been obtained. The optimum tem-
perature of growth is 37 C.; Wollstein 3 states that slight development
takes place even at 5 to 10 C. An exposure of thirty minutes at
57 to 60 C. prevents further development in artificial media. The
organisms may remain viable upon the potato-glycerin-agar medium
for two months.

Products of Growth. According to Wollstein, 4 no acid is produced
in dextrose, lactose, saccharose or mannite serum broth. No enzymes
have been demonstrated in cultures of the organism, and it produces
no visible changes in hemoglobin. Extracellular toxins have never
been demonstrated, but autolyzed cultures introduced intravenously

1 It is prepared as follows: 100 grams finely chopped potatoes are boiled in 200 c.c.
of 4 per cent, glycerin for a short time, then cooled. To every 100 c.c. of the potato
glycerin extract there is added 300 c.c. of 7.5 per cent, agar containing 0.5 per cent.
NaCl. The glycerin potato extract replaces the usual peptone-meat-juice nutrients of
nutrient agar. The mixture is heated to boiling, filtered, and sterilized in test-tubes
about 3 c.c. of medium per tube. (Old potatoes which are slightly alkaline in reaction
are much better than new potatoes which are usually acid in reaction, in preparing the
glycerin extract.) To each tube of the sterilized glycerin potato agar medium is added
an equal volume of sterile, defibrinated human or rabbit's blood, while the medium is
still warm, 45 to 50 C. Then the mixture is cooled in an inclined position.

2 Wollstein, Jour. Exp. Med., 1909, xi, 41.

3 Loc. cit. 4 LOC. cit.

422 HEMOGLOBINOPHILIC BACILLI

into rabbits frequently kill them within twenty-four to forty-eight
hours. Subcutaneous injections of autolysates may cause local
necrosis, but generalized symptoms fail to appear. Similar results
have been obtained with endotoxin obtained by grinding the bacilli
to an impalpable powder and injecting a saline suspension of it. 1

Pathogenesis. Animal. Klimenko 2 and Franker 3 produced a catarrh
of the respiratory mucosa of monkeys and young dogs by intratracheal
injections of B. pertussis suspended in salt solution. A febrile reac-
tion appeared after three to four days and several of the animals died
within two to three weeks. The bacilli were recovered from the
bronchial mucus, bronchi, and from the areas of bronchopneumonia
which developed in the lungs. No characteristic paroxysms were
induced, although Klimenko stated that sneezing and coughing were
noticed. Wollstein 4 has pointed out a possible source of error in the
dog experiments: she finds that those dogs which die after injections
of B. pertussis succumb to canine distemper; the lesions of the respira-
tory tract are readily accounted for on this basis, and the blood of
the animals fails to react specifically with the Bordet-Gengou bacillus.

Human. There are no postmortem lesions characteristic of whoop-
ing-cough. Bronchopneumonia is the most common complication
seen at autopsy. Mallory and Hornor, 5 and Mallory, Hornor and
Henderson 6 have advanced an interesting explanation for the parox-
ysms of whooping-cough. They find that the ciliated epithelium of
the respiratory tract is denuded in places and the cilia plastered down
to such an extent as to interfere with the free removal of mucus by
the mechanical action of the bacteria. When mucus accumulates in
sufficient amount, jt is forcibly expelled by a prolonged violent parox-
ysm of coughing. These experiments were made upon animals; the
frequent occurrence of B. bronchosepticus or a closely-related bacillus
in the respiratory tracts of laboratory animals, which produces similar
lesions to those seen in canine distemper, should be borne in mind in
interpreting these results.

Immunity. Whooping-cough is more commonly a disease of chil-
dren, and recovery from one attack appears to confer lifelong immunity
as a rule.

1 Bordet and Gengou, Centralbl. f. Bakt., Ref., 1909, xliii, 273.

2 Centralbl. f. Bakt., Orig., 1909, xlviii, 64.

3 Munch, med. Wchnschr., 1908, p. 1683.

4 Loc. cit.

6 Jour. Med. Research, 1912, xxvii, 115.
6 Ibid., 1913, xxvii, No. 4.

THE KOCH-WEEKS BACILLUS 423

Bacteriological Diagnosis. 1. Morphological. The diagnosis of
whooping-cough by a microscopical examination of bronchial dis-
charges is not satisfactory. Influenza bacilli are frequently present
in the mucus and sputum from cases of pertussis, and no method is
available at the present time which will distinguish with certainty
between the two bacilli.

2. Cultural. The isolation of B. pertussis from bronchial mucus
upon potato-glycerin-blood agar and its ability to grow upon ascitic
media free from hemoglobin separates the Bordet-Gengou organism
from B. influenzas

3. Serological. The sera of animals highly immunized to B. per-
tussis agglutinate the organism in high dilution, but fail to agglu-
tinate B. influenzse and vice versa. The serum of patients during and
after recovery from whooping-cough, however, agglutinates B. per-
tussis irregularly, and the method has no general diagnostic impor-
tance. The method of complement fixation similarly has not been
successful as applied to the diagnosis of the disease in man, although
the reaction is clear-cut when applied to the sera of immunized
animals. 1

The etiology of whooping-cough has not been definitely established;
the Bordet-Gengou bacillus, however, is found in the majority of cases
of pertussis. Up to the present time it has not been isolated from
healthy subjects.

THE KOCH-WEEKS BACILLUS.

Acute contagious conjunctivitis or, as it is popularly known,
pink-eye is generally considered to be an infection of the conjunctiva
by a small bacillus first described by Koch. 2 Somewhat later Weeks 3
described the organism anew and succeeded in growing it in artificial
media, probably in association with other organisms. Kartulis 4
isolated it in pure culture on blood serum, and Kamen 5 published a
more complete study of the cultural characters of the organism.

Morphology. The Koch-Weeks bacillus is a small rod-shaped
organism resembling the influenza bacillus. It is of about the same
diameter as the influenza bacillus, 0.25 micron, but somewhat longer,
measuring from 1 to 2 microns in length. It occurs singly and in

1 Wollstein, loc. cit.

2 Wien. klin. Wchnschr., 1883, 1550; Arb. a. d. kais. Gesamte., 1887, iii, 19.

3 Arch, of Ophthalmology, 1886, xv, No. 4.

4 Centralbl. f. Bakt., 1899, i, 449.
' Ibid., 1889, xxv, 401, 449.

424 HEMOGLOBINOPHILIC BACILLI

pairs, but short chains of bacilli are not uncommonly seen in growths
on artificial media. Involution forms, which are atypical in form and
size, are also found in cultures outside of the body. The organism is
non-motile and it has no flagella. Capsules and spores- have not
been demonstrated. The Koch-Weeks bacillus stains with ordinary
anilin dyes, but not intensely. It is Gram-negative.

Isolation and Culture. The organism grows best in a medium of
semi-liquid consistency. 5 per cent, agar containing blood or ascitic
fluid appears to be the best for this purpose. Material for inoculation
is conveniently obtained first by flushing the conjunctiva thoroughly
with sterile salt solution then removing some of the secretion which
soon accumulates with a sterile swab which is immediately rubbed
upon the surface of the blood agar. After twenty-four to forty-eight
hours' incubation at 37 C. colonies usually appear which are very
minute and colorless. They die rapidly.

Resistance. The Koch-Weeks bacillus is very susceptible to drying
and to heat; chemical disinfectants very rapidly destroy the organism
outside the human body.

Nothing is known of the products of growth.

Pathogenesis. Attempts to produce conjunctivitis in animals with
the organism have been uniformly negative but inoculations upon
the healthy conjunctiva of man usually reproduce the disease.

The disease is very contagious; it is spread chiefly by contact.

MORAX-AXENFELD BACILLUS.

In 1896 Morax 1 described a diplobacillus which he observed repeat-
edly during an epidemic -of subacute conjunctivitis. The year follow-
ing, Axenfeld 2 published an excellent description of the same organism,
which is commonly referred to as the Morax-Axenfeld bacillus or the
diplobacillus of subacute conjunctivitis.

Morphology. The organisms, as the name implies, occur typically
in pairs; less frequently they may remain adherent to form short
chains. The individual bacilli are of average size, measuring from
1 to 2 microns in length, and about 1 micron in diameter as an average.
The ends of the bacilli are rather square cut. Cultures on artificial
media are somewhat variable in size and shape; chain formation is not
uncommon and involution forms are frequent. The organisms are
non-motile and possess no flagella. Neither spores nor capsules have

1 Ann. Inst. Past., June, 1896. 2 Centralbl. f. Bakt., 1897, xxi, 1.

BACILLUS OF DUCREY 425

been demonstrated. Ordinary anilin dyes color the bacilli readily,
and the Gram stain is negative.

Isolation and Culture. Growth occurs only in media containing
blood serum or ascitic fluid; Loffler's alkaline blood serum is a favorable
substrate. The colonies on blood serum after twenty-four to forty-
eight hours' development at 37 C. are slightly sunken, due to the
liquefaction of the medium. After several days the serum is almost
completely liquefied. Colonies grown on ascitic agar are small, color-
less and transparent even after several days' incubation. Oxygen
is essential for the growth of the bacilli; no growth occurs when
oxygen is excluded from the media.

Prognosis of Growth. A proteolytic enzyme which liquefies coagu-
lated blood serum is the only enzyme which has been described. No
other products of growth are known.

Pathogenesis. Attempts to reproduce the characteristic subacute
conjunctivitis in experimental animals have utterly failed. In man
the typical disease is a subacute catarrhal conjunctivitis with com-
paratively little pus formation, differing in this respect sharply from
the acute conjunctivitis which is produced by the Koch-Weeks bacil-
lus. The angles of the eye are inflamed, particularly the caruncles.
The organisms are best detected in the secretion which collects during
the night. They occur both in pus cells and free, frequently in con-
siderable numbers. Morax appears to have reproduced the essential
lesions by inoculating a drop of culture upon a healthy conjunctiva.

BACILLUS OF DUCREY.

The soft chancre, chancroid, or soft sore must be sharply differen-
tiated from the hard chancre with which it has nothing in common.
The soft chancre is a non-specific, ulcerating sore common to both
sexes, particularly among the unclean. It begins as a small red spot
which rapidly develops into a pustule. This pustule soon breaks down,
leaving a spreading ulcer in which necrosis is a prominent feature.
The ulcer spreads with considerable rapidity and is difficult to control.
The adjacent and regional lymph glands usually become involved
and they soon soften and ulcerate.

Ducrey 1 first called attention to a bacillus (which bears his name),
which he found regularly in chancroids. In 1900 Besancon, Griffon
and Le Sourd 2 succeeded in growing the organism in pure culture upon

1 Monatsch. f. prakt. Dermat., 1889, ix, No. 9.

2 Gaz. dcs Hop., 1900, No. 14; Ann. de Dermat. et Syph., 1901, ii, 1.

426 HEMOGLOBINOPHILIC BACILLI

blood agar. The organism is often referred to as the Streptobacillus
of Ducrey.

Morphology. The bacillus of Ducrey is a small bacillus, measuring
about 0.5 micron in diameter and from 1 to 2 microns in length. It
occurs characteristically both in chancroids and in culture in chains
of considerable length. Frequently these streptobacilli are found in
dense masses. The organism stains with ordinary anilin dyes; usually
the stain is more intense at the ends of the rod, the centre being nearly
devoid of color. This gives the organism a diplococcoid appearance.
It is Gram-negative.

Isolation and Culture. The Ducrey bacillus does not grow upon
ordinary media, but cultures may be obtained by the method of
Davis, 1 which consists essentially in sterilizing the skin over an
unbroken bubo and aspirating the contents with a hypodermic needle
which has been maintained at body temperature. 2 The material is
introduced directly upon the surface of blood agar. 3 If the ulcers or
buboes have opened, they may be cleaned with sterile gauze, dried,
then painted with tincture of iodin and covered with sterile gauze.
Inoculations upon blood agar are made from the pus which collects
under the gauze within twenty-four hours.

The colonies are usually visible after twenty-four hours' incubation
at 37 C. They appear as raised, shining, grayish droplets with a
pearly lustre. They die out rapidly at room temperature, but may be
kept alive at body temperature for some days. The colonies are
removed from the medium with some difficulty, for they tend to slip
away from the platinum needle. Subcultures tend to increase some-
what in luxuriance of growth, and by frequent transfer the organism
may be kept alive for weeks provided the growths are maintained at
37 C.

The bacillus of Ducrey is an aerobic organism which is not resistant
to drying. The pus becomes non-infective after one or two days'
desiccation. Weak antiseptics quickly destroy it.

The products of growth are unknown.

Pathogenesis. It is non-pathogenic for ordinary laboratory animals.
Tomasczewski 4 claims to have reproduced a chancroid in a monkey

1 Jour. Med. Research, 1893, ix, 401.

2 Both the syringe and the blood agar should be warmed to body temperature before
use, because the organism rapidly loses its viability at room temperature.

3 Blood agar for this purpose is prepared by adding defibrinated human or rabbit
blood to agar; the medium is heated to 56-60 C. for thirty minutes to destroy natural
bactericidal substances, and incubated for twenty-four hours to insure sterility.

4 Deutsch. med. Wchnschr., 1903, No. 26.

BACILLUS OF DUCREY 427

(Macacus) by the injection of a blood agar culture obtained from a
bubo. This same culture also produced a chancroid when inoculated
into a man. Several successful inoculations in a man are recorded
which appear to establish satisfactorily the etiological relationship
of the bacillus of Ducrey to the lesion.

Bacteriological Diagnosis. 1. Microscopical. If material be removed
carefully from the base of an ulcer and spread gently upon a glass
slide to prevent the breaking up of the characteristic arrangement of
the bacilli in long intertwined chains, a definite diagnosis may fre-
quently be made by direct observation of the Gram-stained prepara-
tion under the microscope.

2. Cultural. Material preferably obtained from an unopened bubo
should be spread upon the surface of blood agar, employing the technic
outlined above. As much material as possible should be inoculated
to insure growth of the bacilli.

3. Inoculation of Patient. The forearm of the patient is thoroughly
cleaned, then scarified with a platinum needle infected with material
from the ulcer or from a pure culture. The lesion appears within
twenty-four hours and it is typically developed in from three to five
days. It is obvious that little or no immunity is produced, because
autoinoculation results in infection. The possibility of syphilis
must be borne in mind in inoculation experiments, particularly in
transferring material from one subject into another. Syphilis and
chancroid may exist in the same patient.

CHAPTER XXIII.

THE TUBERCLE BACILLUS GROUP: HUMAN, BOVINE,

AND AVIAN.

THE ACID-FAST GROUP.

THERE is a well-defined group of bacteria characterized by the
relatively large amounts of lipoidal substances contained within their
bodies. These lipoidal or waxy substances confer upon the members
of the group distinctive staining reactions; ordinary dyes do not
stain them at all, or at best slowly. The more intense stains con-
taining a mordant, as carbol-fuchsin, penetrate the waxy envelope,
especially when heat is applied; once stained the bacteria retain
the dye tenaciously even after treatment with mineral acids. This
resistance to decolorization with acids has led to the name the
Acid-fast group.

Included within the group of acid-fast bacteria are saprophytic
types found rather commonly in hay and manure; parasitic organ-
isms found upon the surface of the body, as the smegma bacillus and
the nasal secretion bacillus of Karlinski; and exquisitely pathogenic
bacteria, Bacillus tuberculosis and Bacillus leprse. The basis o*
classification therefore is chemical rather than morphological, and in
this sense the definition of the acid-fast organisms does not follow a
strictly natural system.

Types of Tubercle Bacilli. Four types of tubercle bacilli are
recognized which are pathogenic respectively for man, cattle, birds,
and for fishes and reptiles; the human, bovine, avian, and ichthic
varieties. Considerable discussion has arisen concerning the identity
of the human, bovine, and avian varieties, some authorities claiming
that they are identical, although modified somewhat by their con-
tinuous sojourn in a series of animals of the same kind. The evidence
for this view is arrayed around the observation that tubercle bacilli
of undoubted bovine type occasionally are isolated from tuberculous
lesions in man (chiefly in children, infrequently in adults). On the
other hand human bacilli are less commonly found in progressive
tuberculous lesions of cattle. In spite of many attempts to change

PLATE II

Tubercle Bacilli; Ziehl-Neelsen Stain.

TUBERCLE BACILLUS 429

one type into the other, no experiments have been reported up to the
present time which are sufficiently conclusive and extensive to war-
rant the assumption that one variety has been permanently changed
into the other. Loss or increase of pathogenic properties of one
strain does not suffice to bridge the gap between it and another strain
habitually pathogenic for another animal. It is very probable that
the human, bovine, and avian strains had a common ancestor and
that acclimatization in different animals has led to the perpetuation
of three culturally and pathogenically stable varieties. The ichthic
type is much more closely related to the non-pathogenic grass and
dung bacilli than to the true tubercle bacilli.

TUBERCLE BACILLUS.

Historical. One of the greatest chapters in the history of medicine
was inaugurated by the isolation of the tubercle bacillus in pure cul-
ture, and the demonstration of its etiological relationship to tuber-
culosis. The credit for this work, which in every detail marks an
important epoch in bacteriology, belongs to Robert Koch. 1

Morphology. The tubercle bacillus is a slender, straight or slightly
curved rod measuring from 0.2 to 0.6 micron in diameter and from
1.5 to 6 microns in length. The size and appearance of the organism
varies somewhat according to the source. In sputum it frequently
occurs in small clumps, often with the long axes of the bacilli parallel.
Occasionally a pair of bacilli are arranged at an angle like the letter
"V." The bacilli are typically isodiametric, but irregularities of
outline are not uncommon; these irregularities are due to nodules
which cause the organism to swell or bulge wherever they occur.
These nodules frequently stain deeply, and between them are areas
which stain lightly or not at all, thus giving the rod a beaded or vacuo-
lated appearance which may be so marked that the organism resembles
a chain of cocci. These "beaded" forms are frequently observed in
the sputum of consumptives and occasionally in old growths on
artificial media.

True branching is also occasionally exhibited by tubercle bacilli
derived both from the sputum and from culture. 2 Some observers
have classed the tubercle bacillus with the group of Actinomyces on
the basis of this branching. 3

1 Berl. klin. Woch., 1882, No. 15; Mitt. a. d. Kais. Ges.-Amte, 1884, ii, 1.

2 Klein, Centralbl. f. Bakt., 1890, vii, 794.

3 Babes and Levaditi, Arch, de med. exper. et d'anat. path., 1897, ix, 1041.

430 THE TUBERCLE BACILLUS GROUP

The tubercle bacillus is non-motile and possesses no flagella. It
forms no capsule but possesses a waxy envelope which confers upon
the organism unusual resistance to desiccation and to the action of
chemicals. No spores have been definitely demonstrated, but Koch 1
believed that the deeply staining granules found in the bacillus might
be true endospores. The generally accepted view is opposed to this
supposition. 2

Staining. Tubercle bacilli and closely related organisms possess
in common a relatively large amount of waxy substance 3 which is
relatively impervious even to the more intense stains, as carbol-fuchsin.
Ordinary anilin dyes do not stain members of the tubercle bacillus

FIG, 61. Tubercle bacilli, beaded forms.

group. They are Gram-positive, but it requires several hours for the
anilin-oil gentian violet to color the organisms. When a stain has
penetrated the substance of the tubercle bacillus it is retained with
great tenacity; alcohol and even rnineral acids in moderate concen-
tration fail to remove it except after long exposure. The members
of the tubercle bacillus group vary somewhat in this resistance to
decolorization; the true tubercle bacilli are both "alcohol-" and
"acid-fast;" other organisms in the group may be either "alcohol-"
or "acid-fast." Young tubercle bacilli are frequently non-acid-fast. 4

1 Mitt. a. d. Kais. Gesamte, 1884, ii, 22.

2 See Wherry, Centralbl. f. Bakt., Orig., 1913, Ixx, Heft 3-4. Conditions which favor
the formation of "spores" in certain acid-fast bacteria.

3 For chemical composition of fatty substance of the tubercle bacillus see: de
Schweinitz and Dorset, Jour. Am. Chem. Soc., 1898, xx, No. 8, p. 618; 20th Annual
Report, Bur. Animal Ind., 1903. Levene, Med. Record, December 17, 1898, 873;
Jour. Med. Research, 1901, vi, 120; 1904, xii, 251. Kresling, Centralbl. f. Bakt., 1901,
xxx, 897.

4 Wolbach and Ernst, Jour. Med. Research, 1903, x, No, 3.

TUBERCLE BACILLUS

431

The best and most universally applicable stain for the tubercle
bacillus is the Ziehl-Neelsen stain. 1 It is used as follows:

1. A thin smear of the material to be examined for tubercle bacilli
is prepared and fixed in the usual manner, then flooded with carbol-
fuchsin and steamed gently (not boiled) for five minutes. The pre-
paration must be flooded continuously with the stain.

2. Wash thoroughly with water to remove the excess of stain.

3. Decolorize with 90 per cent, alcohol containing 3 per cent,
hydrochloric acid until the pink color has practically disappeared.

4. Wash with water.

5. Counterstain lightly with Loffler's alkaline methylene blue.

6. Wash, dry, examine.

* 'l

FIG. 62. Tubercle bacillus showing branching. X 1800. (Wolbach and Ernst.)

Tubercle bacilli are colored red; non-acid-fast bacteria are colored
blue. It should be remembered that spores may also be stained red
by this method, but they are not likely to be confused with tubercle
bacilli; they are round or oval; tubercle bacilli are much longer.

The decolorization and counterstaining may be accomplished by
one operation, according to the Frankel-Gabbett method. 2 The
preparation of the smear and staining with carbol-fuchsin is carried
out as above (Steps 1 and 2). Decolorization and counterstaining are
accomplished by flooding the preparation with the Frankel-Gabbett
solution (100 c.c. water, 25 c.c. sulphuric acid, 50 c.c. saturated alco-
holic solution of methylene blue) for three to five minutes, then wash

1 Ziehl, Deutsch. med. Wchnschr., 1882, 451; Neelsen, Fortschr. d. Med., 1885, 200.
'Frankel, Beil. klin, Woch., 1884; Gabbett, Lancet, 1887.

432 THE TUBERCLE BACILLUS GROUP

with water, dry and examine. Acid-fast bacteria are stained red,
all other organisms are blue.

Much Granules. Certain granules are found in old caseous foci
and occasionally in the pus of cold abscesses which do not contain
tubercle bacilli demonstrable by the acid-fast stain. Material con-
taining these granules introduced into guinea-pigs causes a rapidly
fatal tuberculosis. Much 1 states that these granules are living frag-
ments of tubercle bacilli which develop into the typical bacillus when
environmental conditions are optimum. They are Gram-positive and
non-acid-fast, but may regain their acid-fastness. When they are
non-acid-fast they do not multiply. The exact significance of these
granules (Much granules) is as yet undetermined; whether they are
identical with the "splinters" described by Spengler 2 is problematical.
The "splinters" are usually colored red with fuchsin, and they frequently
appear in tubercle bacilli that do not stain uniformly, appearing as
rows of red, acid-fast granules. According to Spengler they may be
found in sputum or other tuberculous material as heaps of small
granules, even if the bacilli themselves cannot be demonstrated.

Isolation and Culture. It is difficult to cultivate the tubercle bacillus
directly from lesions upon artificial media and it is even more diffi-
cult to obtain pure cultures directly from sputum, feces, or lung
cavities where tubercle bacilli are growing in the presence of other
organisms which develop much more rapidly on artificial media. The
initial growth on artificial media is the most difficult to obtain. Either
coagulated dog's serum 3 or the Dorset egg medium 4 is best for this
purpose. Tissue containing tubercles is removed from the animal
with aseptic precautions to sterile Petri dishes. The tissue is minced
somewhat and then distributed over the slanted surface of either the
serum or the egg medium. At the end of a week or ten days the bits
of tissue are moved around to fresh surface areas; at the end of two
to four weeks the tubercle bacilli appear as minute gray nodules
which gradually spread, forming eventually a wrinkled dull gray-
yellow growth covering the greater part of the medium. Subcul-
tures from the original culture grow better on artificial media than
the original culture, although even subcultures grow very slowly.

1 Beitr. z. Klinik d. Tuberkulose, 1907, viii, 85, 357, 368; 1908, xi, 67; 1913, Supp.
Bd. vi.

2 Deutsch. med. Wchnschr., 1907, p. 337.

3 Coagulated at 75 C.; Theobald Smith, Jour. Exp. Med, 1898, iii. 647; Trans.
Assn. Am. Phys., 1898, xiii, 417.

4 Bureau of Animal Industry. Annual Report, 1902, p. 574.

TUBERCLE BACILLUS 433

The coagulated serum or the egg medium may be used for subcultures ;
glycerin agar or glycerin potato is also suitable for this purpose. It
is essential to protect the cultures from evaporation and to incubate
them in a slanting position. This is best accomplished by sealing the
slant cultures after they are made, either with paraffin or with corks
which have been charred to kill off moulds or other organisms, then
covered with lead foil. Tubercle bacilli grow fairly readily on the
surface of glycerin broth after they have become accustomed to
artificial media. A fresh thin film from egg medium floated on the
surface of the broth is the best method of obtaining the growth in
this medium. The organisms must be floated on the surface of the
broth, otherwise growth does not take place. If the growth sinks to
the bottom all development ceases. Tubercle bacilli do not grow
readily in gelatin or other artificial media not containing glycerin
or proteins derived from blood serum or egg.

Cultures of tubercle bacilli which have been grown on artificial
media for some time may be gradually accustomed to develop in media
of simple composition. Proskauer and Beck 1 grew the organism upon
the Uschinsky medium to which glycerin was added: Wherry 2 and
Lowenstein 3 have employed media in which ammonium salts were
the only source of nitrogen. Kendall, Day and Walker 4 have corro-
borated these results. Tuberculin appears to be produced even in
these simple media. The tubercle bacillus grows in milk, producing
a gradual solution of the casein. 5

The tubercle bacillus is aerobic, although it will develop slowly
anaerobically. Its temperature range is rather limited, the organisms
growing between 30 C. and 42 C., with an optimum temperature
of 37 C. Growth below 35 C. is slow. Tubercle bacilli are fairly
resistant to drying, naked germs being killed by dry heat at 100 C.
only after forty-five minutes. With moist heat an exposure to 60
C. kills them in thirty minutes, 65 C. in fifteen minutes, 70 C. in
five minutes, 80 C. in one minute, and 100 C. in half a minute.
The organisms enclosed in mucus are much more resistant, dry heat
(100 C.) killing them only after an exposure of from two to three

1 Ztschr. f. Hyg., 1894, xviii, 128.

2 Jour. Inf. Dis., 1913, xiii, 144; Centralbl. f. Bact., Orig., 1913, Ixx, 115.

3 Centralbl. f. Bakt., Orig., 1913, Ixviii, 591.

4 Jour. Inf. Dis., 1914, xv, 428.

6 Klein, Centralbl. f. Bakt., Orig., 1900, xxviii, 111. Monvoisin, Compt. rend. Acad.
Sci., October, 1909, xxvi; Rev. de Med. vcterin., 1910, Ixxxvii, 16. Mossu and Mon-
voisin, Compt. rend. Soc. Biol., 1907, Uii, No. 26. Kendall, Day and Walker, Jour,
Am. Chem. Assn., 1914, xxvi, 1959.
28

434 THE TUBERCLE BACILLUS GROUP

hours, 70 C. after seven hours, and 60 C. after ten hours. In sterile
water the organisms may remain alive for over two months. They
are quite resistant also to putrefaction. Instances are on record
where tuberculous lungs have been buried for six months and yet
contained virulent organisms. Schottelius claims that a tuberculous
lung buried two years contained virulent tubercle bacilli at the end
of that time.

The thermal death point in milk is 60 C. for thirty minutes. There
is d source of error in determining the thermal death point of the
tubercle bacillus or of any other organism in milk. If the experiment
is carried out in milk which is not enclosed in such a manner as to
prevent surface evaporation the results are inaccurate; the scum
which forms on the surface of the milk as the result of evaporation
contains casein and salts; they are non-conductors of heat and protect
the organisms so that they apparently resist a much higher tempera-
ture than would otherwise be the case. 1

Tubercle bacilli in sputum are killed in twenty-four hours by mixing
the sputum with an equal volume of 5 per cent, carbolic acid. Mer-
curic chloride is not suitable for this purpose because it precipitates
mucus, forming a compound with it which renders its germicidal
action nil. Rooms containing tubercle bacilli may be disinfected either
by burning four pounds of sulphur to 1000 cubic feet in a moist atmos-
phere, or by evaporating 500 c.c. of formaldehyde to every 1000 cubic
feet under the same conditions. The room should not be opened up
until after eight hours have elapsed.

Direct sunlight kills tubercle bacilli even when they are enclosed
in sputum, but the rapidity with which they are killed depends some-
what upon the season; a longer exposure is required in winter than
in summer. Sputum exposed out of doors in indirect light may remain
infectious for some time. In order to determine that tubercle bacilli
are killed it is necessary to inoculate the material containing them
into guinea-pigs, the guinea-pig being far more sensitive than artificial
media for this purpose. Theobald Smith 2 has shown that it takes at
least 1500 times as many tubercle bacilli to infect artificial media
as it does to infect a guinea-pig. It must be remembered that even
killed tubercle bacilli, as Prudden and Hodenpyl 3 have shown, produce
tubercles in guinea-pigs, but that these tubercles are not transmissible

1 Theobald Smith, Jour. Exp. Med., 1899, iv, 233.

2 Jour. Med. Research, 1913, xxviii, 91.

3 New York Med. Jour., Jun e6, 1891, p. 20.

TUBERCLE BACILLUS 435

to other guinea-pigs; consequently it is necessary to inoculate a
second set of guinea-pigs from the tubercles developing in the first
set of pigs in order to be certain that the bacilli are killed.

Products of Growth. Enzymes. Tubercle bacilli do not produce
soluble proteolytic enzymes. No carbohydrate-splitting enzymes
have been observed.

Carriere, 1 Wells and Corper 2 have shown that the bodies of tubercle
bacilli contain a lipase of moderate activity. Kendall Walker and
Day 3 have demonstrated that the filtrates of cultures of human and
bovine tubercle bacilli contain a soluble esterase; the action of the
enzyme upon fats is relatively slight. This esterase is produced in
an active form in media of very simple composition. 4

Winternitz and Meloy 5 have shown that the lipase (esterase ?)
activity of the blood is decreased in tuberculosis. Bauer states it is
increased in the early stages of the disease.

Hemolysis. Raybaud and Hawthorn 6 state that cultures of tubercle
bacilli will not hemolyze the erythrocytes of normal guinea-pigs; the
erythrocytes of tuberculous pigs are hemolyzed.

Tubercle bacilli do not form indol or the ordinary products of bac-
terial decomposition in ordinary media. They do not liquefy gelatin
nor do they coagulate milk. Theobald Smith 7 has called attention
to a very constant differential character between the human and
bovine types of the tubercle bacillus. In glycerin broth the human
tubercle bacillus causes a permanent acid reaction, while the bovine
bacillus under the same conditions causes the medium to become
alkaline if the growth conditions are suitable. Tuberculin prepared
from human cultures consequently is acid in reaction, while that
prepared from bovine cultures is alkaline. The organism liberates
a moderate amount of ammonia incidental to its metabolism of pro-
teins or amino acids. 8 Old cultures of tubercle bacilli occasionally
are very gelatinous. 9 Vaughan, 10 White and Avery 11 and White, 12 using

1 Compt. rend. Soc. Biol., 1901, liii, 320.

2 Jour. Inf. Dis., 1912, xi, 388.

3 Ibid., 1914, xv, 443.

4 Kendall, Walker and Day, Jour. Inf. Dis., 1914, xv, 455.

6 Jour. Med. Research, 1910, xxii, 107.

Compt. rend. Soc. Biol., 1903, No. 55.

7 Trans. Am. Phys., 1903, xviii, 108; Am. Jour. Med. Sc., 1904, cxxviii, 216; Jour.
Med. Research, 1905, xiii, 253, 405.

8 Kendall, Day and Walker, Jour. Inf. Dis., 1914, xv, 417, 423, 428, 433.

9 Weleminsky, Berl. klin. Wchnschr., 1912, xlix, 1320. Gotzl, Wien. klin. Wchnschr.
1913, 1614. Kendall, Day and Walker, Jour. Inf. Dis., 1914, xv, 428.

10 Protein Split Products, 1913.

11 Jour. Med. Research, 1912, xxvi, 317.

12 Trans. 9th Ann. Meet. Nat'l. Assn. Study and Proven. Tuberculosis.

436 THE TUBERCLE BACILLUS GROUP

the method of Vaughan, have isolated a non-specific poisonous sub-
stance from fat-free tubercle bacilli which kills guinea-pigs with
symptoms typical of anaphylaxis. The mineral constituents of
tubercle bacilli have been determined by de Schweinitz and Dorset. 1

Toxins. The tubercle bacillus appears to elaborate both an endo-
toxin and an extracellular toxin. 2 The endotoxin causes necrosis,
caseous degeneration and general cachexia and stimulates tubercle
formation. The extracellular toxin causes fever and the acute inflam-
matory reaction observed around tubercles and tuberculous tissue
in tuberculous animals. Little or no effect is produced in healthy
animals except emaciation. The toxins liberated by the tubercle
bacillus are apparently on the whole rather mild, because they produce
as a rule only local lesions. This would indicate that the diffusion
of toxin is somewhat limited. Furthermore, the kidneys do not ordi-
narily exhibit anatomical changes which could be definitely ascribed
to the elimination of a tuberculous toxin through them. Whether
the cachexia, which is a prominent feature of advanced cases of
tuberculosis, is to be regarded as a purely toxic phenomenon is not
clear. Holmes 3 has suggested that the fatty acids of the tubercle
bacillus cause a lymphocytosis.

Pathogenesis. Human. According to Naegeli, 4 rather more than
90 per cent, of adults who come to autopsy show scar tissue at the
apices of the lungs, which he believed were healed tubercles. Later
observations have not fully confirmed these figures, but it appears
that fully 50 per cent, of adults have healed tubercles at this site. 5
Frequently virulent tubercle bacilli have been isolated from the
centre of this scar tissue, but it should be remembered that occasion-
ally virulent tubercle bacilli have been isolated from bronchial lymph
nodes which appear to be normal.

Modes of Infection. Hereditary Transmission. Transmission of
the tubercle bacillus through the sperm has never been established;
transmission through the ovum is also not definitely established.
The maternal blood, on the contrary, appears to be a vehicle through
which tubercle bacilli may pass, or grow through the placental barrier
and thus reach and infect the fetus. 6

1 Jour. Am. Chem. Assn., 1898, xx, 618.

2 Armand-Delille, Monographies Cliniques en Medicine, etc., 1911, No. 06, Paris.

3 Guy's Hospital Reports, 1909, lix, 155.
* Virchow's Arch., 1900, clx, 426.

6 Lubarsch, Virchow's Arch., 1913, ccxiii, 417.

6 See Gartner, Ztschr. f. Hyg., 1893, xiii, 126-139, for summary and discussion of
early literature. Also, Schmorl and Geifel, Miinchen. med. Wchnschr., 1904, 1676.

TUBERCLE BACILLUS 437

Latency Theory. Baumgarten 1 believes that tubercle bacilli may
lie dormant in the body for months or years and become active when
the u resistance" of the body is lowered. The evidence is on the whole
opposed to this view, partly because congenital tuberculosis is uncom-
mon, chiefly because the organs of fetuses of tuberculous mothers
do not cause infection in guinea-pigs. More recently v. Behring 2 has
advanced the theory that infection takes place in childhood, probably
by ingestion of milk containing tubercle bacilli, and that the manifes-
tations of infection become apparent later in life.

Inoculation Theory. Direct inoculation through the skin is rare.
Inhalation Theory. Droplet infection and infection by dust con-
taining viable tubercle bacilli appear to be the most common methods
of transmission of the organism.

Ingestion of milk or meat containing tubercle bacilli must be con-
sidered as important methods of transmission of the organism.

Trauma, establishing a locus minoris resistentiae to which tubercle
bacilli lying dormant in the body may be transported and set up infec-
tion, is probably uncommon.

Conditions Favoring Infection. Overcrowding with its attendant
evils of dark, damp rooms, poor food and general unhygienic condi-
tions appears to be a most potent factor in the spread of tuberculosis.
No age is exempt, although the disease is somewhat less frequent
between the ages of five and ten years, greatest between sixteen and
thirty-five years. The sexes are about equally infected. Negroes
are especially prone to the disease, possibly because of their surround-
ings and manner of living rather than any inherent lack of resistance.
Tuberculosis is relatively uncommon among the aboriginal negro
races in Africa. Jews appear to be relatively immune to the disease.
Those occupations in which dust is generated in large amounts exhibit
a higher incidence of the disease than occupations in which dust is
not a feature. Catarrhal infections of the respiratory tract appear
to predispose to pulmonary tuberculosis as do measles, whooping-
cough and influenza.

Atria of Invasion. The respiratory and digestive tracts (including
the tonsils) and the skin are the three portals through which tubercle
bacilli enter the tissues of the body. Of these the respiratory tract
is more frequently involved. Droplet infection is by far the most
common method of transmission of tubercle bacilli; dust-borne infec-
tion is probably relatively uncommon.

1 Deutsch. med. Wchnschr., 1882, No. 22.

2 Ibid., 1903, 692; 1904, 194.

438

THE TUBERCLE BACILLUS GROUP

Tubercle bacilli also enter the intestinal tract and they may pass
through the intestinal mucosa without leaving any trace of their
passage, particularly if they be suspended in fatty menstrua, as butter
or cream. 1 The bovine type of the tubercle bacillus may enter through
the tonsils, or the digestive tract occasionally. Rarely, tubercle bacilli
enter through the skin, usually causing somewhat localized epidermal
proliferations containing tubercle bacilli in small numbers, which
are sometimes called butcher's warts, postmortem warts or verruca
necrogenica. Usually they remain localized.

COMBINED TABULATION, CASES REPORTED AND OWN SERIES OF
CASES. (PARK AND KRUMWIEDE.)

Diagnosis.

Adults 16 years
and over

Children

5 to 16 years.

Children
under 5 years.

Human.

Bovine.

Human.

Bovine.

Human.

Bovine.

Pulmonary tuberculosis ....
Tuberculous adenitis, axillary or
inguinal
Tuberculous adenitis, cervical
Abdominal tuberculosis ....
Generalized tuberculosis, alimentary
origin .
Generalized tuberculosis ....
Generalized tuberculosis, including
meninges, alimentary origin
Generalized tuberculosis, including
meninges .

568

2
22
15

18
11
1

1
3

1
1

4
33

2
4

7
2
26
1
1

3
1

1
1

2
15
6

45
14
21

1
1

20
13

10
5

Tubercular meningitis ....
Tuberculosis of bones and joints
Genito-urinary tuberculosis .
Tuberculosis of skin ....

Miscellaneous cases:
Tuberculosis of tonsils .
Tuberculosis of mouth and cervi-
cal nodes

Tuberculous sinus or abscesses .
Sepsis, latent bacilli ....

Totals

677

161

Mixed or double infections, 4 cases.
Total cases, 1042.

For some years much discussion has centred upon the incidence of
bovine tubercle bacillus infection in man. Koch was inclined to the
view that infection with this organism was so rare as to be practically
negligible. Later he modified his opinion. Weber 2 studied 628 cases

1 Nicolas and Descos, Jour. Physiol. et Path, gen., 1902, iv, 910; Ravenel, Jour. Med.
Research, 1903, x, 460.

2 Tuberkulose, Arbeiten a. d. kais. Gesamte, 1910, Heft x.

TUBERCLE BACILLUS 439

(284 children, 335 adults, 9 age unstated), all of whom had drunk
the milk of cows having tuberculosis of the udder, or had consumed
uncooked products made from the milk. Only two patients, both
very young children, were definitely shown to be infected with bovine
tubercle bacilli. Both had enlarged caseous cervical glands from which
the organism was isolated and identified. Six children and one adult
had glandular swelling in the neck, but the evidence was not conclusive
that bovine infection had taken place. The general conclusion was
that there was relatively little danger from drinking milk containing
viable bovine tubercle bacilli.

Much more convincing are the studies of Park and Krumwiede. 1
The accompanying table (see preceding page), which is their sum-
mary of their own extensive investigation and a recapitulation of
authentic observations of others, shows very definitely that infection
with bovine bacilli is relatively common in children and young adults
up to sixteen years of age, but relatively uncommon in adults.

Bovine bacilli are found not only in unpasteurized market milk, 2
but also in the glandular organs of a considerable proportion of cattle
and swine. The muscles are usually not invaded. No meat from
tuberculous animals can be offered for sale in the public markets,
however.

Lipschutz 3 has reported a case of cutaneous infection by the avian
tubercle bacillus in man which resembled leprosy anatomically. The
diagnosis was arrived at only after an exhaustive study of the organism.
Infection of man with the avian tubercle bacillus is uncommon.

The mechanism of infection with the tubercle bacillus has been the
subject of much controversy. It is apparent that the acid-fastness
of the organism per se does not confer pathogenic properties upon the
organism because other non-pathogenic acid-fast bacteria are unable
to induce progressive disease from man to man or from animal to
animal. Acid-fastness, however, may be an initial factor in patho-
genism, an opening wedge as it were, for it appears to be well estab-
lished that acid-fast bacteria are relatively insoluble in body fluids
and remain unchanged for considerable periods of time when they
are introduced into the animal body. Theobald Smith 4 has advanced
a tentative hypothesis which explains satisfactorily many of the

1 Trans. 6th Ann. Meet. Nat'l. Assn. Study and Preven. Tuberculosis*

2 See Kober, Trans. Am. Phys. for literature to 1903. Hess, Jour. Am. Med. Assn.,
1909, Hi, No. 13, 1011. Moore, Jour. Med. Research, 1911, xxiv, 517.

3 Arch. f. Dermat. u. Syph., June, 1914, cxx.

4 Jour. Am. Med. Assn., April 28, 1906.

440 THE TUBERCLE BACILLUS GROUP

phenomena. As tubercle bacilli reach the body (and as they escape
from the body) they are surrounded by a protective envelope which
causes the organism to behave somewhat as an inert foreign body
until it finally settles down in some structure where it can grow.
The envelope is then slowly removed or modified by the action of
normal tissue fluids and growth commences. In this connection it is
interesting to note that young tubercle bacilli are frequently non-
acid-fast, 1 and that the tissues usually invaded by the bacilli lym-
phoid tissue and the lungs contain active lipase. 2 If this supposition
is correct, the tubercle bacillus may remain latent in the body until
the fatty capsule is removed or modified, perhaps by a fat-splitting
enzyme (lipase) ; then development takes place. It should be remarked
parenthetically that polymorphonuclear leukocytes which occasion-
ally engulf tubercle bacilli do not contain lipase; 3 these leukocytes
may transport the organisms to lymphoid tissue or other tissue where
eventually the bacilli escape, thus establishing new foci. Mono-
nuclear leukocytes appear to contain lipase, as do certain fixed phago-
cytic cells in the alveoli of the lungs.

Lenk and Pollak 4 and Wiener 5 appear to have found active proteo-
lytic ferments in tuberculous exudates. Opie and Barker 6 have shown
that the mononuclear epithelioid cells contain an enzyme which digests
protein in a slightly acid medium; it is practically inert in an alkaline
medium. Jobling and Petersen 7 have found that the inhibition of
enzyme action in caseous tubercle foci is apparently due to unsaturated
fatty acids. Saturation of these acids with iodin causes an accelera-
tion of the activity of the ferments.

The primary lesions usually tend to progress slowly. Secondary
invasion by tubercle bacilli through the lymph and bloodvessels
frequently occurs, causing tuberculous foci in various ducts and
glands of the body, as the bronchi, alveoli of the lungs, spleen, liver,
tubules of the kidney, and in the genito-urinary system, particularly
the epididymis and testicle of the male and the Fallopian tubes in the
female. The glandular organs are those most commonly infected,
and of these the lungs and lymph nodes are most frequently involved;

1 Wolbach and Ernst, Jour. Med. Research, 1903, x, 313.

2 Bradley, Jour. Biol. Chem., 1913, xiii, No. 4. Briscoe, Jour. Path, and Bact., 1907,
xii. Bartel and Neumann, Centralbl. f. Bakt., Orig., 1909, xlviii, 657. Zinsser and
Carey, Jour. Am. Med. Assn., 1912, Iviii, 692.

3 Fiessinger and Marie, Compt. rend. Soc. Biol., 1909, Ixvii, 177. Bergell, Miinchen
med. Wchnschr., 1909, Ivi, 64.

4 Deutsch. Arch. klin. Med., 1910, cix, 350. 5 Biochem. Ztschr., 1912, xli, 149.
6 Jour. Exp. Med., 1908, x, 645; 1909, xi, 686. 7 Ibid., 1914, xix, 383.

TUBERCLE BACILLUS 441

also the spleen, kidneys, liver, meninges both of the cord and brain,
the pleural and pericardial cavities, the genito-urinary apparatus,
and, less frequently, joints and bones. The muscles are only very
rarely invaded. Various clinical names have been applied to tuber-
culosis of different tissues: tuberculosis of the lungs is commonly
designated consumption; of the spine, Pott's Disease; of the cervical
lymph glands, scrofula; and of the skin, lupus. The characteristic
initial lesion is a small nodule or tubercle which may undergo secon-
dary changes, as caseation, calcification, ulceration, or various types
of sclerosis. In the lungs the first organisms that reach the alveoli
may leave no trace. They are dissolved there apparently, but may
produce no progressive lesion. A second invasion in the same area
frequently causes a local inflammation which usually results in infec-
tion, apparently because the body has been sensitized by the first
bacilli that entered, and in some way is rendered locally susceptible
to the organism.

The irritation caused by the extracellular toxin excreted by the
tubercle bacillus brings about a response on the part of the tissues
which is protective, as is manifested by a walling off of the bacilli.
First there is a proliferation of the connective tissue which forms a
spherical mass of epithelioid cells around the focus of infection. Out-
side of the epithelioid cells there is usually an infiltration of lympho-
cytes. The tissue is avascular and the young tubercles contain little
or no fats. 1 The central part of the tubercle soon begins to undergo
coagulation necrosis, probably due to the action of the intracellular
toxin, and it is gradually converted into a homogeneous, cheesy mass.
In many tubercles giant cells are found, which are formed either by
the coalescence of several epithelioid cells, or by atypical cell divi-
sion, the nucleus dividing faster than the cytoplasm. The nuclei of
the giant cell are arranged peripherally as a rule, either completely
around the cell, or in the shape of a horseshoe. The centre of the
giant cell likewise may undergo caseous degeneration, and tubercle
bacilli are not infrequently found in the middle of these cells. 2 Accord-
ing to Zeit, giant cells are essentially blind blood capillaries which
have extended into the tuberculous area, but have not become true
vessels because the toxins of the organisms have prevented the
final development of functional blood channels. Besides these small

1 Joest, Virchow's Arch., 1911, cciii, 451.

2 See Evans, Bowman, and Winternitz, Jour. Exp. Med., 1914, xix, 283, for a critical
experimental study of the histogenesis of the miliary tubercle in vitally stained rabbits
for the finer details of the process.

442 fH% TUBERCLE BACILLUS

miliary tubercles, larger areas of caseation may develop; epithelioid
cells, lymphocytes, and giant cells are usually found closely packed
around these areas.

The destruction of the capillaries and the resulting avascular tissue
helps in the necrosis of the tubercle by cutting off the blood supply.

What is generally known as consumption or destruction of the lung
tissue is probably not due to the action of the tubercle bacillus alone,
but to secondary infection and liquefaction of tissue by other organ-
isms, as the streptococcus, staphylococcus, pneumococcus, or Micro-
coccus tetragenus. If a caseous necrotic tubercle located near a
bronchus ruptures into this bronchus, a large amount of tuberculous
material is suddenly swept into the regional areas of the lung, over-
whelming it and setting up a rapidly fatal infection which is known
as galloping consumption or phthisis florida. If a caseous tubercle
ruptures into a lymph or bloodvessel, the material may be carried
very widely through the body, causing generalized miliary tuber-
culosis, which resembles typhoid fever clinically. Hemorrhage not
infrequently takes place from the lung, due to the erosion and subse-
quent bursting of a bloodvessel which may have been included in the
caseous area. In the human lung it is practically always possible to
find old lesions at the apices when the infection is due to the human
type of the tubercle bacillus. Uncommonly no old healed tubercles
can be found, and the lungs are filled with miliary tubercles, in which
case the infection is usually caused by the bovine type of the tubercle
bacillus. Tubercle bacilli