LIBRARY

TOXINES AND ANTITOXINES.

QHARLES GRIFFIN & CO., LTD., PUBLISHERS.

FERMENTSANDTHE1RACTIONS. A Text-Book on the Chemistry and Physics
of Fermentative Changes. By CARL OPPENHEIMER, PH.D., M.D. Translated
from the German by C. AINSWORTH MITCHELL, B.A., F.I.C., F.C.S. In Crown
8vo, Cloth. 7s. 6d. net.

" Such a veritable multum inparvo has never yet appeared." — Brewers' Journal.

TECHNICAL MYCOLOGY. The Utilisation of Micro-organisms in the Arts and
Manufactures. By DR. FRANZ LAFAR. Translated by CHARLES T. C. SALTER.

In Large 8vo, Handsome Cloth. Fully Illustrated. VOL. II.— PART I. 7s. 6d.

" The first work of the kind which can lay claim to completeness in the treatment of a fascinating
subject."— Lancet.

FLESH FOODS: With Methods for their Chemical, Microscopical, and Bacterio-
logical Examination. By C. AINSWORTH MITCHELL, M.A., F.I.C. Fully
Illustrated. 10s. 6d.

" A book which no one whose duties involve considerations of food supply can afford to be without."
Municipal Journal.

POISONS: THEIR EFFECTS AND DETECTION. By A. WYNTER BLYTH,
M. R. c. s. , F. i. c. , F. c. s. FOURTH EDITION, Revised. In Large 8vo, Cloth. With
Tables and Illustrations. 21s.

' ' Undoubtedly the most complete work on Toxicology in our language." — Analyst.

OUTLINES OF PRACTICAL PHYSIOLOGY. Being a Manual for the Physio-
logical Laboratory, including Chemical and Experimental Physiology, with
Reference to Practical Medicine. By WM. STIRLING, M.O., SC.D. FOURTH
EDITION, Thoroughly Revised. With over 400 Illustrations. 15s. net.

"The best Text-Book on Practical Physiology." — Lancet.

ELEMENTARY TEXT-BOOK ON BIOLOGY. By J. R. AINSWORTH DAVIS,
M.A. , F. z. s. SECOND EDITION. In Two Volumes.

VOL. I.— Vegetable Morphology and Physiology. 8s. 6d.
VOL. II.— Animal Morphology and Physiology. 10s. 6d.

" Certainly the best ' Biology ' with which we are acquainted."— British Medical Journal.

ATLAS OF URINARY SEDIMENTS. With Special Reference to their Clinical
Significance. Edited by SHERIDAN DELEPINE, M.B., C.M.EDIN. Translated by
FRED. C. MOORE, M.SC.. M.B., from the German of DR. HERMANN RIEDER.

With Thirty-Six Coloured Plates, Comprising 167 Figures. 18s.

"A success in every way ... by far the best we have seen. " — Lancet.

PHYSIOLOGY AND PATHOLOGY OF THE URINE. By J. DIXON MANN,
M.D., F.R.C.P. Fully Illustrated. 8s. 6d. net.

" Cannot fail to be of inestimable value." — British Medical Journal.

THE WORK OF THE DIGESTIVE GLANDS. By PROF. PAW LOW Of St. Peters-
burg. Translated by W. H. THOMPSON, M.D.. M.CH., F.R.C.S. 6s.net.

"Full of new and interesting facts."— Lancet.

TREATMENT OF DISEASES OF THE DIGESTIVE SYSTEM. By ROBERT

SAUNDBY, M.D., M.SC., LL.D., F.R.C.S. In Crown 8vo. Handsome Cloth.

OUTLINES OF BACTERIOLOGY. A Practical Handbook on the Basis of
"Precis de Microbie." By DR. L. H. THOINOT and E. J. MASSELIN. Trans-
lated by WM. ST. CLAIR SYMMERS, M.B.ABERDEEN. Pocket Size, Leather. With
Illustrations. 10s. 6d.

London: Charles Griffin & Co., Ltd., Exeter St., Strand, W.Q.

TOXTNES

AND

ANTITOXIN ES.

BY

CARL OPPENHEIMER, M.D., Pn.D.

TRANSLATED FROM THE GERMAN

BY

C. AINSWORTH MITCHELL, B.A.(OxoN.), F.I.C.

LONDON :

CHARLES GRIFFIN & COMPANY, LIMITED;

EXETER STREET, STRAND.

1906.

[All Rights Reserved.}

> rf

PREFACE TO THE GEEMAN EDITION.

THE author's object in writing this work was to bring
together all that was known on the subject of toxines,
the definition of a toxine being based on chemical grounds
and on the side-chain theory, quite independently of the
origin of the poison.

Hence it follows that although this book may be
regarded as complete in one respect — viz., that it does, in
fact, give a comprehensive idea of toxines as a whole — yet,
at the same time, it was not part of my plan to deal with
all the toxic substances of a doubtful nature that are
formed in animals and plants. For this reason many
animal and vegetable poisons, the toxine character of
which is open to question (e.g., fish venoms), have only
been described very cursorily, while no mention has been
made of others (e.g., bee-poison, which is probably of a
basic nature).

In my attempt to give an exact outline of the toxines
themselves and their antitoxines, I have had practically
no previous work of importance at my disposal.

Therefore, so far as it was possible, I have obtained
my facts solely from the original papers. Yet I cannot
but fear that my book remains a torso, since much that
has been written on toxines is scattered through different
journals, many of which seem unlikely to deal with such
a subject, and it is therefore probable that many facts
may have escaped my notice.

CARL OPPENHEIMER.

692232

NOTE BY THE TRANSLATOR.

THIS book is in many respects a companion volume to
" Ferments and their Actions," by the same author. As
in that case, I have made numerous additions to the text,
so as to include the more important facts that have been
made known since the appearance of the German Edition,
and the selection from the voluminous mass of new
matter has been made by Dr. Oppenheimer himself.

C. A. M.
GRAY'S INN,
LONDON, W.C., June, 1906.

TABLE OF CONTENTS.

GENERAL PART.

PAGES

Introduction: Definition of a " Toxine," Chemically, Physio-
logically, and Theoretically — Side-Chain Theory — Mode of
Formation — Nutrient Media — Means of Promoting the For-
mation of Toxines — Filtration — Purification — Chemical
Nature — Influence of External Factors (Oxidising Agents,
Poisons) — Fate in the Organism — Formation of Antitoxine —
Immunity — Fate in the Digestive Tract — Mode of Action of
Toxines, Specific Nature, Incubation Period — Constitution
of Toxines — Physiological Action— Toxoids and Toxones —
General Characteristics of Antitoxines, ..... 1-29

Behaviour of Toxines towards Antitoxines : Haptophore

and Toxophore Groups — Specific Combination — Units of
Measurement — Toxoid Formation — L0 and L+ — The Value
"D" as a Measure of Toxones — Incomplete Saturation — The
Value "200" — Poison "Spectra" — Physico-Chemical Meas-
urements— Speed of Reaction — Conditions of Equilibria —
Heat Manifestation of the Action of Toxine upon Antitoxine, 29-62

Endotoxines and Bacterial Proteins, 62-68

SPECIAL PART.
I. THE TRUE TOXINES.

Diphtheria Toxine: Secretion— Nutrient Media— Admission
of Air — Preservation — Purification — Properties — Effects —
Toxoids and Toxones, 69-91

Diphtheria Antitoxine: Production — Preservation of Serum
— Concentration of the Antitoxine — Chemical Nature —
Combination with Proteids of the Serum, .... 91-102

Tetanus Toxine : Nutrient Media — Properties— Concentration
— Action — Relation to Central Nervous System — Fermenta-
tion Theory — Tetanus antitoxine, 102-125

Botulism Toxine — Pyocyaneus Toxine — Toxine of
Symptomatic Anthrax,

Bacterial HsemolysineS : Tetanolysine — Pyocyanolysine —
Colily sine— Staphylolysine - Other Lysines— The Leucocidine
of the Staphylococci, 133-144

V111 CONTENTS.

II. ENDOTOXINE8 AND OTHER BACTERIAL POISONS.

PAGES

Cholera ViPUS : Poisonous Substances Isolated — Is it a True

Toxine combined with the Cells ? — Antitoxine, . . . 145-154

Typhoid Virus - Bacillus Coli Communis — Dysentery —
Plague Toxine— Pneumotoxine— Gonotoxine— Strep-
totoxine, 154-166

Poisons Of the Tubercle Bacillus : Tuberculine— Behring's

Tubercle Poisons-Malleine 166-176

Anthrax Poison— Other Bacterial Poisons— Hog Cholera
—Malignant (Edema— Swine Plague— Metsehnikoff's
Vibrio— Dysentery Toxine, 176-184

III. THE VEGETABLE TOXIN ES (Phytotoxines).

Ricine : Preparation — Chemical Nature — Properties — Action on

the Blood - Immunity — Antiricine, ..... 185-201

Abrine— Cro tine— Robine— Hay Fever Toxine, . . . 201-208

IV. THE ANIMAL TOXINES (Zootoxine.s).

Snake Toxines : Preparation — Chemical Nature— Properties —
Mode of Action— Neurotoxine — Hsemorrhagine — Toxoids —
Hsemolysine of Snake Poisons— Complex Nature — Action —
Lecithin as Complement — Lecithides — Immunity — Anti-
toxines, 209-233

Toad Toxine (Phrynolysine)— Salamander Poison— Spider

Venom: Arachnolysine— Scorpion Venom, . . . 233-238

Fish Venoms : Trachinus Poison— Eels' Blood Poison (Ichthyo-

toxine), 238-245

Toxine of Fatigue, 245

BIBLIOGRAPHY AND INDEX or AUTHORS' NAMES, ... 246-265
SUBJECT INDEX, , 267-274

TOXINES AND ANTITOXINES,

GENERAL.

Introduction. — Not long after the principles of bacteriology had
been established by ROBERT KOCH, it became the general belief
that it was not, in the main, to the bacteria themselves that the
chief symptoms of infectious diseases were to be attributed. It
was soon recognised that the living micro-organisms were, for
the most part, only indirectly harmful, and that their chemical
products were to be regarded as the immediate cause of the
disease.

BRIEGER, in particular, pointed out in the very early days of
bacteriology that search ought to be made for the specific poisons
of bacteria, and he himself endeavoured to discover and isolate
these hypothetical poisons.

He first separated from culture media that had been altered by
the growth of bacteria, and notably from the mixed products of
putrefaction, a series of well-defined chemical substances which he
termed ptomaines — nitrogenous bases, some of which were virulent
poisons. These substances, however, proved not to be the real bac-
terial poisons. They were not the weapons used by the parasites in
living bodies. They were not the specific bacterial poisons, which
were first grouped under the collective name of " Toxines." Then
by degrees the conception of a toxine underwent a natural process
of specialisation, so that the term began no longer to connote the
poison isolated from any given products of decomposition formed
in the vital processes of bacteria, but to be limited to the specific
bacterial poison that caused specific illness. Hence this tendency
to limit the connotation developed without being the result of
conscious thought, or much less of being formulated in set terms.
The confusion in the definition was still further increased by the
fact that a series of bacterial poisons, apparently related to the
proteids, were termed toxalbumins. tinder this name were in-
cluded, not only certain poisons that we must to-day regard as

2 TOXINES AND ANTITOXINES.

true toxines, but also others that had nothing in common with
them except their apparent proteid nature.

On the other hand, one very important and far-reaching theo-
retical result of these researches was that a comparison was drawn
between these bacterial toxalbumins and other toxalbumins of
the animal and vegetable kingdom, such as snake poison and the
like on the one hand, and on the other the poisonous vegetable
^rjoteids r^tw,; abnirte, and crotine, first investigated by ROBERT
anti'his pupili. V* £Tn.e* most important deduction from this point
• of ^yi&w^as. that. fir,st. drawn at a later period by EHRLICH, CAL-
'HJfejf-E; krid'e^heps^yiz.^ that the specific bacterial poisons do, as
a matter of fact, enter* into a fundamental relationship with the
cell products in question of higher organisms to form specific
" anti-bodies " in the organism of the attacked animal, so that
EHRLICH'S side-chain theory was applicable in general to their
action. These poisons are haptines in EHRLICH'S conception,
and, as regards the theory, it matters little what was their origin.
BRIEGER'S great service lies in the fact that he was the first to
show a relationship between bacterial poisons and other known
poisonous substances. Hence it is not a very essential point
that BRIEGER'S views require modification in two respects, in
consequence of later researches, of which his own were not the
least important. In the first place, the relationship towards
other toxalbumins does not hold good in the case of all the bac-
terial poisons isolated by BRIEGER, for very many of these are
substances of non-specific character, not comparable with ricine,
&c., since they are not haptines. In point of fact, there remain
practically only the poisons of diphtheria, tetanus, Bacillus
botulinus, and B. pyocyaneus as typical true toxines (e.g.,
tetanolysine, staphylolysine, and staphyloleucocidine, with pro-
bably also the blood-solvent poisons of other bacteria), and some
others, such as the poisons of cholera and typhus. Moreover,
BRIEGER himself, at a later period, stated that his diphtheria
toxine was not a proteid, and the proteid nature of ricine is also
very doubtful, whereas snake poisons, even in the present condi-
tion of our knowledge, appear to be proteins. In this respect,
however, toxines resemble the closely analogous enzymes, among
which, in addition to, presumably, true proteids (trypsin, diastase
(?) ), there are substances of high molecular weight that belong to
another class (pepsin, invertase). The definition of a true
"toxine" is, therefore, arrived at in the following manner in
the case of bacterial poisons : —

All bacteria produce certain chemical substances in the media
in which they develop.

TOXTNES. 3

Although many of the bacterial substances formed in the most
different ways are products of secondary decomposition due to
obvious chemical reactions, yet beyond doubt some of them are
primary products of bacterial metabolism.

Some of these metabolic products are more or less virulent
poisons. In this respect there is, in general, no difference
between pathogenic and non-pathogenic micro-organisms.

Thus, even when substances of this kind are poisonous, they
have certainly nothing to do with the poisoning of an organism
through the invasion of bacteria, even assuming they are pro-
duced by pathogenic micro-organisms. Such poisons, for instance,
as neurine have their specific effects, whether produced by bac-
teria or by purely chemical reactions. Hence they must be
eliminated first of all from the definition of a toxine.

In the second place, a series of substances has been prepared
by different processes from the dead cells of many species of
pathogenic micro-organisms. These are of a proteid nature, as,
for example, BUCHNER'S bacterial proteins, and are more or less
poisonous. But their toxic effects differ but little whatever their
origin. They never show a specific character, and never produce
symptoms resembling those of specific diseases. In addition to
this, the cell protoplasm of many bacteria contains toxic proteids
that cannot be isolated from the protein, and these are also non-
specific in their action.

Excluding these, what is left on which to base the definition
of a toxine ? Certain pathogenic bacteria, when grown in a pure
cultivation, produce poisons that dissolve in the liquid nutrient
media, and can only be obtained in an undecomposed, concen-
trated, and imperfectly purified condition by the most careful
treatment — substances which are not ptomaines and not proteids
(vide infra}.

Such substances have been isolated, in particular, from pure
cultivations of the bacilli of diphtheria and tetanus, and they are
the true bacterial toxines in the narrower sense of the word. In
like manner all those more definite chemical poisonous substances
formed in the cells of higher animals and plants must be separated
from the toxines.

The latter form a class of substances whose definition is found
in their nature and characteristic mode of action, irrespective of
their origin.

Toxines are characterised in the first place by a sum of external
properties. Their chemical structure is absolutely unknown.
They are extremely unstable, and very sensitive towards even
slight chemical influences, and especially towards the action of

4 TOXINES AND ANTITOXIN ES.

heat. They are not proteids and not toxalbumins, and they show
many remarkably close analogies with ferments.

Physiologically, they are distinguished by being under suitable
conditions extremely poisonous, far more so than any other
known poisons. Nearly all toxines have also the peculiar
characteristic of not acting immediately, but only after a latent
period, a time of incubation, in which respect their action is
analogous to the poisoning by living bacteria. In spite of their
being so extremely poisonous to many animals, exceeding in this
respect the most active of the simple poisons, such as hydrocyanic
acid, they yet exhibit in only a few cases (e.g., snake poison) the
immediate action that is characteristic of the simpler poisons.
Above all, they are characterised by the specific nature of their
action. Toxines have a peculiar form of activity characteristic of
the whole group, as we shall subsequently see. In addition to
this, each individual toxine has its own particular mode of action,
which, in the case of bacterial toxines, shows a close relationship
with the disease produced by the parent cells, being completely
analogous in the case of tetanus poison. They are also strictly
specific in the narrower sense of the word — i.e., they are only
able to injure certain -living organisms, whilst they have abso-
lutely no effect upon other organisms (some of which are closely
allied to them), with which they stand in the fundamental highly
important relationship of natural immunity. Nor are their
relationships of acquired immunity less important, for it is a
fundamental property of toxines to produce in the organism
attacked antidotes of a strictly specific nature, which render the
poison harmless in vivo, and, when separated from the parent
organism, can also exercise their neutralising activity in vitro,
each on its respective toxine, and only on that. Thus, each true
toxine has also its corresponding true antitoxine.

Incidentally, it may be mentioned that hitherto all attempts to prepare
true antitoxines to the simple crystalloid poisons have been unsuccessful.
Even the recent statement by HmscHLAFF,1 who claimed to have prepared
an antimorphine serum, has been shown by MoRGENROTH2 to be completely
unsupported by the facts, and to have been due to want of accuracy in the
preparation of the minimum lethal dose.

Not only have we data for forming a conception of what a
toxine is chemically and physically, but we also have a theory
based upon them. According to EHKLICH'S side-chain theory, a
toxine is a poison which possesses at least two specific atomic

1 Hirschlaff, "Antimorphinserum," Berl. Idin. Woch., 1902.

2 Morgenroth, "Zur Frage d. Antimorphinserum," ibid., 1903, 21.

TOXINES. 5

groups, — a haptophore group, whose function it is to enter into
combination with the attacked cells, and a toxophore group, which
produces the toxic effects. Every substance that possesses a
specific affinity, a suitable haptophore group, for definite com-
plex groupings of protoplasm, is a haptine, and every poisonous
haptine that also possesses a toxophore group is a toxine. We
must therefore base our definition of a toxine in accordance
with this, and rigorously exclude from the toxines any poisonous
substance that is not a haptine and that does not produce an
antitoxine.

This has been much more easily done in the case of mineral
and vegetable poisons, for no one would dream of including
among the poisons, in the narrower sense, the alkaloids, &c.,
of plants, or the few poisonous crystalline substances that can
be isolated from the secretions and organs of animals, such as,
for example, the alkaloids in the skin of the toad, adrenaline, &c.

But it is of much more importance to effect a classification
of bacterial poisons by means of this sharp definition. We
must, in the first place, separate from the toxines all non-specific
substances as defined above, from whatsoever bacteria obtained.

But a still further classification is introduced by the fact that
there are probably specific bacterial poisons, which are only pro-
duced by definite groups of bacteria, and exert specific activity,
but which are not haptines, do not form anti-bodies, and are
therefore not toxines. Poisons of this nature appear to play a
part — e.g., in tuberculosis. They will be shortly discussed in
the special part. In addition to these there is a whole series
of very imperfectly known bacterial poisons, whose specific
nature and toxine character there are strong reasons for
doubting.

Most difficult of all is the question of the poisons produced
by certain bacteria, notably those of cholera and typhus. Their
toxines are scarcely known in the free state, and the poisons
excreted by them do not appear to be the true toxines. On the
other hand, we find among them another type, which presents
great difficulties in the way of closer investigation. These are
the endotoxines, which are firmly attached to the living cells,
and are thus comparable with the endo-enzymes of yeast and
animal organs. We shall discuss these more fully in a subse-
quent chapter.

On Toxines in General. — The true toxines, as defined above,
are characterised, we repeat, by their physical and chemical
properties in the aggregate, which we have now to describe
more closely, combined with their fundamental property of

6 TOXINES AND ANTITOXINES.

splitting off free haptophore side-chains in suitable organisms
— i.e., of producing antitoxines.

Although each individual toxine possesses its own character-
istics, which can only be dealt with properly in the special part,
yet all true toxines have a number of properties in common
which justify us in speaking of them collectively.

Bacterial toxines share these properties with all the other
toxines known to us, such as snake venoms, the poison of the
blood of the eel and murcena, spider and toad venoms, ricine,
abrine, crotine, &c.

In the first place, the mode of production is common to the
bacterial toxines. They are to be regarded not, it would seem,
as the products of culture media altered by the invasion of
bacteria, but, as was demonstrated by BucHNER,1 as the actual
products of the cell protoplasm, as secreted products of bac-
terial cells. Just as the cell of the pancreatic glands produces
and secretes its trypsin and the starch cell of wheat endosperm
its diastase, so do the bacterial cells secrete their specific toxines.
The fact that in the case of certain micro-organisms — e.g., of
cholera, &LG. — the toxines may, under suitable conditions, be
retained firmly in the protoplasm, has also its analogy among
ferments, for we find the yeast enzymes possessing exactly the
same characteristic.

When grown upon suitable culture media, pathogenic micro-
organisms that produce toxines usually develop their charac-
teristic poisons within a very short time. Thus, SpRONCK2
obtained a very active diphtheria toxine within forty-eight hours.

The virulence increases, however, with the age of the culti-
vation. Roux and YERSINS found that a filtered diphtheria
cultivation of seven days' growth killed a rabbit in six days,
but that after forty-two days' growth an equal dose of the same
cultivation caused death in a much shorter time. SPRONCK'S
diphtheria toxine was ten times more virulent after five or six
days' than after forty-eight hours' growth. Still, the virulence
attains its maximum after a certain time, and then begins to
diminish owing to the decomposition of the toxine (vide infra,
Toxoids), so that old cultivations become less poisonous. Then,
after a fairly long time, the degree of virulence usually becomes
constant.

1 Buchner, "Die Bedeutung der aktiven loslichen Zellprodukte, &c.,"
Munch, med. Woch., 1897, 12.

2 Spronck, "Prepar. de la tox, dipth.," Ann. Past., xii., 701, 1898.

3 Roux and Yersin, "Contribution a 1'etude de la diphtheric," ibid., iii.,
273, 1889; iv., 385, 1890.

TOXINES. 7

The kind of culture medium naturally has the greatest influence
on the production of the poison.

In general, bouillon cultivations are employed, usually with
the addition of a certain proportion of peptone ; but culture
media are also frequently prepared from meat extracts, yeast
extracts, &c.

Agar and other nutrient media can hardly be advantageously
employed. Interesting experiments have been made to produce
toxines in proteid-free culture media, as, for instance, in aspara-
gine solutions containing suitable salts (ARMAND and CHARRIN 1),
and in dialysed urine. Satisfactory results, however, have not
yet been obtained (GuiNOCHET,2 UscHiNSKY3). ZINNO'S* explana-
tion of apparently successful attempts to produce toxines in such
proteid-free culture media is, that only traces of proteid are
necessary for the production of detectable amounts of poison.
As a matter of fact, the quantities of toxine thus produced are
extremely small. He concludes, from his own experiments, that
the presence of a small amount of proteid is indispensable.

Speaking generally, this factor varies so greatly with the kind
of toxine that the reader must be referred to the special part,
where a full description is given of the different methods that
have been employed to obtain the largest possible yield of
toxines.

Here we will only point out briefly that too great acidity or
alkalinity of the medium must always be avoided, and that, in
general, the same precautions with regard to temperature, &c.,
must be observed as are usually taken to obtain the most active
and virulent cultivations of bacteria.

One point, however, of special interest may be mentioned here.
The production of a very active and virulent growth of the
bacteria themselves is not invariably accompanied by the de-
velopment of very energetic cultures of their toxines. For, on
the one hand, it would seem that the production of poison by the
bacteria is not a direct function of their activity of growth or
their high degree of virulence. Indeed, in the case of diphtheria,

1 Arraand and Charrin, "Transformation de la Matiere organique azote,
&c.," Bull. Med., 1891, 356; 1892, 957. Absts. in Centralbl. f. JBakt., xi.,
248, 1892. Cf. Buchner, " Bakteriengifte und Gegengifte," Munch, med.

Woch., 1893, 449.

2 Guinochet, " Contrib. a 1'etude de la toxine du bacille de la diph-
therie," ibid., 1893, 293.

3 Uschinsky, " Les Poisons de la diphthe"rie et du cholera," ibid., 1893,
293

4 Zinno, "Beitr. z. Stud. d. Entstehung der Toxine," Centralbl. f. JBakt.,
xxxi., 42, 1902.

8 TOXINES AND ANTITOXINES.

there may be very energetic growth of bacteria, whilst the
culture medium is absolutely non-toxic (LUBOWSKI 1).

Moreover, in the case of vegetable and animal toxines, the
production of poison depends on manifold physiological condi-
tions, such as age, nutrition, &c., with which points we will deal
more fully in the special part.

On the other hand, it is beyond question that certain agencies
increase the growth, and eventually the virulence, of the bacteria,
but reduce the yield of toxines. This is due to the fact that they
partially destroy the toxine already formed. Even when agencies
of this kind increase the production of toxine simultaneously with
the vigour of growth of the bacteria, yet, if employed too freely,
the amount of toxine destroyed exceeds that of new-formed
toxine, so that the final result is a diminished yield of toxine.
Thus, when such agents are employed (e.g., the admission of air
to diphtheria cultivations), it is possible to plot a curve, the
abscissae of which represent the increasing quantities of the
agent, and the ordinates the final yields of toxine. So long as,
for example, the introduction of air causes the diphtheria bacilli
to produce an abundance of toxine and the simultaneous destruc-
tion by the current of air of toxine already formed keeps within
narrow limits, the curve will rise. But by degrees the destruc-
tive effect of the air outweighs its favourable influence on the
production, and the curve falls again. There is thus an inter-
mediate point at which, with a definite intensity of air current,
there is a maximum yield of toxine, and its position obviously
depends on numerous conditions, such as the nature of the
growth, the culture medium, temperature, &c. This optimum is
hardly ever realised in practice, the result being, as we shall see
later, that conflicting statements are made as to the benefit or
injury produced by the same agents.

In addition to the introduction of air, other factors may have
a similar effect. Thus, an increase of temperature may influence
both the production and the decomposition of a toxine. On the
other hand, there are apparently agents that do actually increase
the final yield of toxine. A considerable amount of work has
been done in these experiments to obtain larger quantities of
toxines by the use of the most suitable culture media and tem-
peratures, by the addition of special substances, &c., so that
highly poisonous cultivations of the most important toxines can
now be prepared. These methods, however, are of an entirely
special character, and it is hardly possible to give at this stage

1 Lubowski, " Ueber einen atoxischen und avirulenten Diphtheriestamra,"
Zeit.f. Hyg., xxxv., 87, 1900.

TOXINES. 9

any general method of importance that is universally applicable
to all toxines.

On the other hand, we must not omit to mention briefly here
that toxine solutions do not invariably have a uniform strength.
This phenomenon manifests itself notably in the case of tetanus
poison. Apart from the fact that Nicolaier's bacillus produces
two totally different poisons — viz., that which causes the charac-
teristic symptoms of tetanus, and tetanolysine (q.v.) — very great
differences are also to be observed in individual solutions of the
poison as regards their specific activity. Although, as a general
rule, tetanus poison is considerably more toxic for guinea-pigs
than for rabbits, yet there are also some preparations of the
poison that are almost as deadly for rabbits as for guinea-pigs

(TlZZONl).

BEHEING has confirmed this statement with regard to Tizzoni's
poison, and has also found in the case of his own cultivations
that certain parts were, relatively, extremely poisonous to rabbits.
We must therefore conclude that tetanospasmine is not an indi-
vidual substance, but consists of different active constituents.
Similar phenomena have also been observed in the case of the
poisons of diphtheria. Over -neutralised poisons are known
which are absolutely without effect upon guinea-pigs, but are
still poisonous to rabbits.1

Assuming now that we have prepared cultivations of living
bacteria rich in toxines, it is necessary to treat them in such
a way as to eliminate the action of the living cells, so as to be
able to examine the poisons separately. For this purpose we can
either kill the bacteria, or attempt to remove them completely
from the poison.

The first method, in which no attempt is made to remove the
dead cells, cannot, by itself, lead to any definite conclusion as to
the action of the poison, inasmuch as the dead bodies still possess
definite chemical and physiological properties, which must render
all deductions uncertain. Fortunately, this method, which was
formerly employed, can be dispensed with in the investigation of
true toxines, and, as a matter of fact, has fallen into complete
disuse.

For it is- possible to separate true toxines from their parent
cells by filtration through filters impervious to bacteria. The chief
substances used for the filtration, in addition to infusorial earth
and chalk, are porcelain filters or Chamberland's candles. During
the filtration the greater portion of the toxine passes completely
into the filtrate, the residual cells retaining only as much viru-
1 For further particulars see Ehrlich, Munch, med. Woch., 1903, 33.

10 TOXINES AND ANTITOXINES.

lence as corresponds to the amount of toxine mechanically
adhering to them, and from this they can be freed by washing
with a physiological solution of salt. But the cells no longer
contain any true toxine, which might presumably be extracted
from them by breaking up the cell structure (treatment with
alkali), as H. KossEL,1 for example, was able to prove in the case
of diphtheria bacilli. It is true that these dead cells may still
contain poison of quite another kind (bacterial proteins), but
these have nothing to do with the specific poisoning (vide infra").

The phytotoxines are found in the most different organs of
plants, and especially in the seeds, from which they can be
isolated by extraction with dilute solutions of salt.

The zootoxines are produced in the secretions and blood of
animals. It follows from all this that typical toxines are free
secretions — substances that are phsiologically thrown off by the
cells into the surrounding media. In this respect they are
analogous to the true enzymes. Just as the pancreatic glands
secrete trypsin, or the glandular cells of the starch layer secrete
diastase, so does the cell of the diphtheritic micro-organism
secrete diphtheria toxine.

This can only be stated with certainty, however, for the typical
toxines, notably diphtheria and tetanus. In other cases the
facts are much more obscure.

As we shall see later, it is still quite open to discussion
whether, for example, the micro-organisms of cholera and typhus
produce true toxines in the sense of our definition. Even grant-
ing that this is the case, the toxines are certainly not secreted in
a free state in any appreciable amount, but are firmly retained,
at least, by the living cell. Only when the cell decomposes after
death are they liberated to a limited extent, as in the case of
older cultivations. But even then the poisonous substances
have undergone radical changes, having been converted into
secondary, more stable products that no longer show the
characteristics of a true haptine. We shall return to this
point later.

A close analogy to this retention of the active substances to
the living cells may be observed in the case of certain ferments?
We know that the yeast cell, in addition to secreting a small
amount of free diastase, also contains a series of other enzymes
— invertase, maltase, &c. — which can only be set free after the

1H. Kossel, "Zur Kenntnis des Diphtheriegiftes," Centralbl.f. Bakt.,
xix., 977, 1898.

2 Oppenheimer, Ferments and their Actions, English translation by C.
Ainsworth Mitchell. London, 1902 (Griffin & Co.).

TOXINES. 1 1

death or injury of the cell protoplasm, or, as in the case of
zymase, after breaking down the cell wall ; and we also know
that Monilia Candida will not give up its invertase to the
surrounding medium, under any conditions whatsoever.

When toxine solutions have once been obtained by nitration
of the cultures, or analogous preliminary means, such as extrac-
tion of seeds, &c., they can at once be used, although containing
numerous impurities, for physiological experiments. Moreover,
even in this impure condition some rough experiments can be
made with them to determine the behaviour of the respective
toxines towards physical and chemical agents.

Moreover, in order to have it in a more suitable form for
keeping, this solution may also be concentrated with various
precautions, or may even be evaporated to dryness without
materially injuring the toxine. The main conditions are that
the temperature must never exceed 45° C. (whence it is best to
carry out the evaporation in vacuo), and that any strong acids or
bases that may be present must be partially neutralised.

For a closer investigation of toxines, however, tedious processes
of purification are necessary to remove all impurities as com-
pletely as possible. The simplest method is dialysis, which,
however, only separates the toxine from the salts and peptones
also present in the solution, but does not eliminate the proteids.

Hence, complicated methods have had to be devised to isolate
the toxines in the purest possible condition. The favourite
methods employed consist of precipitation by means of am-
monium or magnesium sulphate with subsequent dialysis, and
of precipitation by means of the salts of heavy metals with
subsequent decomposition of the resulting double compounds.
The first of these methods yields, eventually, only solid concen-
trated preparations of toxines which are still impure, although
capable of being used for practical purposes, and the second
method is the only one that gives approximately pure toxine
preparations. A fuller account of its details, which have been
worked out, notably by BRIEGER and his pupils, is given in the
special part. They include exceedingly tedious manipulations
requiring close attention, and consisting, in the main, of precipi-
tation with zinc, lead, or mercury salts. The double compounds
of the toxines with these salts that are precipitated are again
decomposed by means of hydrogen sulphide, or of alkali carbon-
ates, or phosphates. Then by means of filtration or dialysis
solutions are obtained, which on evaporation in vacuo yield
preparations which in favourable cases are very rich in toxine.
They invariably, however, still contain considerable quantities

12 TOXINES AND ANTITOXINES.

of impurities, either of inorganic nature (ash) or of organic nature
(albuminous substances). A pure toxine is, up to the present, as
little known as a pure enzyme, and it is scarcely to be expected
that it will be successfully prepared in the immediate future.
Even in the case of preparations, which were still not pure
although containing relatively very few impurities, BRIEGEE and
BOER obtained such minute quantities that further purification
was out of the question. Moreover, attempts to prepare toxines
on proteid-free culture media (UscniNSKY, loc. cit.) have given
very unpromising results.

As regards the chemical nature of toxines there is, therefore,
practically nothing known. Just as in the case of enzymes, to
which they stand in such close relationship, they were at first
regarded as albuminous substances and termed toxalbumins. The
more thoroughly, however, the preparations were purified, the
stronger became the idea that the albuminous substances,
although very difficult of removal, were only impurities, and
that the pure toxines themselves were in all probability not
proteids in the ordinary sense. And BRIEGER himself, to whom
we owe the conception of toxalbumins, succeeded in preparing
toxine preparations which no longer gave the ordinary reactions
of proteids (cf. Tetanus Poison) any more than did toxines
produced in albumin-free culture media.

In the case of other toxines, too, successful attempts have
been made to reduce, very considerably at least, the albuminous
impurities.

JACOBY1 succeeded by means of digestion with trypsin in
obtaining preparations of ricine that were practically free from
albuminous substances. The active principle itself is not at-
tacked by this enzyme, whereas the albuminous substances
present are decomposed by it. Since then these decomposition
products, like trypsin itself, cannot be precipitated by ammonium
sulphate, even in a 50 per cent, solution, whereas ricine is readily ,
precipitated at that degree of concentration, it is possible by this
means to obtain from the mixed products of the digestion pre-
parations of ricine that no longer give the proteid reactions.

That is the only — negative — knowledge that we have of the
constitution of toxines. With this exception we have to content
ourselves with stating that they are bodies of high molecular
weight, probably allied to the proteids with which they corre-
spond in certain properties, and still more closely related to the
ferments, whose constitution is equally a matter of hypothesis,

1 Jacoby, "Die chemische Natur des Ricins," Arch. exp. Path., xlvi.,
98, 1901 ; Hofm. JSeitr., i., 51, 1901.

TOXINES. 1 3

and to whose properties they offer the closest analogies in their
reactions and activity.

These analogies stand out in the sharpest relief, when we
compare the influence of external factors on bacterial toxines
with their effect upon ferments. There is the closest corre-
spondence in almost every particular.

A special characteristic of toxines is their extreme sensitive-
ness to the action of heat. In their normal solutions they soon
perish at temperatures above 50° C. ; at 80° C. their activity
is at once destroyed, and even at 45° C. they are slowly
decomposed. Individual ferments differ but little in this
respect. On the other hand, they offer great resistance to the
action of dry heat. Solid preparations can be heated to over
100° C. without suffering injury ; but even these appear to be
destroyed by a temperature of 150° C.

It is interesting to note that in liquids devoid of water, such
as amyl alcohol, they can also be heated far above 80° C., and that
many salts, such as, e.g., anhydrous sodium sulphate, increase
their resistance to the acfcion of heat (BuCHNEE1). Low tem-
peratures check their activity, but do not otherwise injure them.

They dialyse to a slight extent through parchment, but not
through collodion (RoDET and GUECHOFF 2), and through animal
membranes, such as those of the oesophagus, bladder, large intes-
tine, and especially the small intestine (CHASSIN and Moussu 3).

In all these respects they behave exactly like enzymes.

Toxines are more sensitive than enzymes to the action of light.
Diphtheria or tetanus poison in aqueous solution is very rapidly
destroyed by both sunlight and diffused daylight (tetanus poison
within forty-eight hours according to KiTASATO4). Light is
without action upon them when they are in the dry condition,
or suspended in liquids devoid of water.

An electric current can also be injurious to toxines, though
only when continuous, for alternating currents of high tension
are quite harmless to tetanus poison (MARMIER 5). Even when
merely allowed to stand in solution in the dark, with every
precaution, the poisons gradually become weaker, being decom-
posed, at any rate, in the case of diphtheria and some other

JBuchner, " Bakteriengifte und Gegengifte," Munch, med. Woch., 1893,
449.

2Rodet and Gue"choff, Soc. Biol., Hi., 965, 1900.

3Chassin and Moussu, "Influence de la dialyse, &c.," ibid., lii., 694,
1900.

4Kitasato, "Exper. Unt. iib. d. Tet. Gift.," Zeit. f. Hyg., x. 287, 1891.

5 Marmier, " Les toxines et 1'e'lectricite', " Ann. Past., x., 469, 1896.

14 TOXINES AND ANTITOXINES.

poisons, into toxoids (vide infra). In the case of other poisons
the existence of toxoids has not been established with certainty.

All toxines are very sensitive to the influence of nearly all
chemical agents.

Oxygen, even when as dilute as in the atmosphere, has a pro-
nounced injurious effect. When exposed to the air, especially
when light is also present, toxines speedily lose their virulence,
notably so in the case of tetanospasmine, which is greatly
weakened by simple nitration.

Speaking generally, all oxidising agents, including hydrogen
peroxide, are very injurious.

SiEBER1 found that calcium peroxide completely destroyed the
virulence of diphtheria and tetanus poison within a few hours
(1,000 times the lethal dose), and also that of abrine (5,000 times
the lethal dose being rendered inert by 0-5 gramme of calcium
peroxide). He also discovered that the oxydases of animal
tissues acted upon bacterial toxines, but not upon abrine. More-
over, on simultaneous injections of oxydase and toxine, the
animal on which the experiment was made remained healthy.
A vegetable oxydase (from the yam) also proved effective,
although peroxydases, which only turn guaiacum blue when
hydrogen peroxide is also present, were inert. An interesting
statement made by him is that fibrin from the blood of a highly-
immune horse contains an oxydase that destroys diphtheria virus,
which is not the case with ordinary fibrin. But it is open to
question whether this was not some remnant of unseparated
antitoxine.

A little is also known with regard to the action of other
chemical substances. Strong bases and acids naturally have a
destructive effect, and weak bases are injurious, while very
dilute acids, especially those of organic nature, have probably a
stimulating influence. The influence of neutral salts and of various
other substances, notably upon tetanus poison, has been studied
by FERMI and PERNOSSi.2 Some have a stimulating and others
an injurious effect upon the toxic activity.

Indifferent gases, such as carbon dioxide, hydrogen, and carbon
monoxide, have no influence. Only in the case of hydrogen sul-
phide did BRIEGERS observe any injurious effect upon tetanus

1 Sieber, " Ueber die Entgiftung der Toxine durch die Superoxyde," Z.
physiol. Client., xxxii., 573, 1901.

2 Fermi and Pernossi, " Ueber das Tetanusgift," Zeit.f. Hyg., xvi., 385,
1894.

3 Brieger, « ' Weitere Erf ahrungen iiber Bakteriengifte," ibid., xix., Ill,
1895.

TOXINES. 1 5

poison after the toxine and gas had been kept in contact for four
days in a sealed tube.

Protoplasmic poisons, such as carbolic acid, chloroform, &c.,
have no materially injurious effect. Alcohol is very injurious.
According to SALKOWSKi,1 salicylic aldehyde is extremely harm-
ful, and he makes the same statement as to the action of chloro-
form and formalin.

Iodine and carbon bisulphide have probably a quite distinctive
mode of action, inasmuch as they appear to attack only the toxo-
phore group, and to tend to the formation of toxoids (EnELiCH2).
Thymus extract has apparently a similar action (BRIEGER,
KITASATO, and WASSERMANN.S

Fate of Toxine s in the Organism. — Toxines disappear fairly
rapidly after their introduction into the circulatory system of
susceptible animals. In a short time the blood is completely
free from toxines, as has been proved by the experiments of
BOMSTEIN, CROLY, and BRUNNER on diphtheria (q.v.), the virus
having entered into close combination somewhere or other in the
latent stage of its activity, as has been shown by the researches
of DONITZ 4 and others. DONITZ found that he could not save
infected animals by the injection of antitoxines, even when only
a few minutes had elapsed after the poisoning, since the toxine
was no longer in a free condition when it encountered the anti-
dote. Only by the injection of very large doses is it possible
after a certain time to break up the combination of the toxine
with the cells of the body, and so destroy the effects of the latent
poisoning. Yet e\ien then there is a time limit, and after its
expiry, notably in the case of tetanus, even huge doses of anti-
toxine are no longer of any avail. Herein lies one of the causes
of the defective therapeutical results in the serum treatment of
tetanus. According to the latest views on tetanus, the antitoxine
cannot follow the toxine in the nerve tracts. (For further par-
ticulars see Tetanus.)

A toxine escapes detection as such when small doses are
injected into the body. Thus, if a single lethal dose, or a small
multiple thereof, be injected into a susceptible animal, the poison
rapidly disappears from the blood, nor can it then be detected in

1 SalkoAvski, "Ueber die Wirkung der Antiseptica auf Toxine," BerL
Uin. Woch., 1898 (No. 25), 545.

2 Ehrlich, "Die Wertbemessung des Diphtherieheilserums," Klin.
Jahrbuch, vi.

3 Brieger, Kitasato, and Wassermann, "Immunitat u. Giftfestigung,"
Ze.it. f. ffyg., xii., 137, 1892.

4 Donitz, "Ueber des Tetanus-antitoxin," Deutsche med. Woch., 1897,
428.

16 TOXINES AND ANTITOXINES.

the organs of the animal.1 The poison has then become firmly
attached to the organs specifically susceptible to it. Moreover,
it is not excreted with the urine (GOLDBERG 2).

When, however, the doses are very large, the poison takes
some time to disappear, and can then appear in the urine.3 This,
too, is perfectly explicable by the theory that the receptors are
not adapted for such a sudden invasion of enormous quantities
of poison, and hence a small proportion of the toxine breaks
through the barrier of the kidneys and appears in the urine.
The fact that the toxine disappears in the system furnished one
of the supports for the conception of the so-called "fermenta-
tion theory " of tetanus. According to this, the true toxine first
splits off by a secondary reaction within the organism another
poison, upon which the anti-body can no longer act (whence
tetanus is not curable after the poisoning [vide supra]), and this
having no incubation period, acts with the rapidity of an alkaloid,
such as strychnine. COURMONT 4 and others claim to have some-
times detected such a poison in the organs of the victims of
tetanus. We will critically examine this theory in its proper
place, and endeavour to show that it is at least superfluous.
The disappearance of the toxines on the one hand, and the
incurability on the other, can be readily explained by the
side-chain theory without the aid of additional hypotheses.

But all this only applies to susceptible animals. The fate of
toxines introduced into the circulatory system of refractory,
naturally immune, animals is materially different.

The question of natural immunity has not yet been completely
elucidated in all its details. It is, undoubtedly, an extraordinarily
complex phenomenon, and, in particular, its forms of manifesta-
tion and its causes show essential differences as regards natural
immunity against poisons on the one hand, and against living
bacteria on the other. In the case of toxines, we have only to
deal with natural antitoxic immunity.

This can be due a priori to two causes. Either the body of
the naturally-immune animal contains antidotes which neutralise
the action of the intruding poison, or the cells of the animal are

1 Salter's assertion (Lancet, 1898, i., 152) that toxines pass into the sweat
is not supported by sufficient evidence.

2 Goldberg, " Ueber Ausscheidung des Tetanusgiftes durch die Nieren-
sekretion," Centralbl. f. BaU., xxvi., 547, 1899. But cf. Cobbett (" Excre-
tion of Diphtheria Toxine in the Urine," Brit. Med. Journ., 1900, i., 21),
who claims to have detected the toxine in the urine.

3 See in particular Brunner, "Z. Kenntnis d. Tetanusgiftes," Z. f. 1dm.
Med., xxxi., 367, 1897.

4 Of. v. Leyden-Blumenthal, Der Tetanus, Vienna, 1901.

TOXINES. 17

immune against the attack of the poison, which is thus for them
a completely indifferent substance.

Both cases occur. As we shall see later, normal sera, and
especially horse serum, contain antitoxines which afford pro-
tection against small doses of toxines. And it is particularly
interesting that, according to WASSERMANN,1 from 80 to 85 per
cent, of the human race have appreciable amounts of a diphtheria
antitoxine virus in their serum. These facts, however, are in-
sufficient in themselves to explain the origin of natural antitoxic
immunity, for such antitoxines are found exclusively in the
sera of susceptible living organisms. On the other hand, the
normal sera of refractory animals contain absolutely no trace of
antitoxines.

It was difficult to account for this fact until EHRLICH suc-
ceeded, by means of his side-chain theory, not only in explaining
it, but in making it one of the chief arguments in support of his
view. Where there are no corresponding receptors (receptive
groups in the body cells), there can be no attack on the part of
a toxine, and the result is immunity against the poison. But
in such cases the splitting off of side-chains to form an antitoxine
is equally out of the question. Hence, according to this view,
it is impossible for the blood of absolutely refractory animals to
contain antitoxines.

But it is interesting to know what happens to the toxines
introduced into the circulatory system of such organisms. It
is quite conceivable that such unstable and extraordinarily
sensitive substances might be rapidly decomposed in the blood
without having produced their injurious effects, or that they
might be very rapidly expelled from the body with the
excretions.

Neither is the case. It is a remarkable phenomenon that
these extremely active substances, which under favourable
conditions can produce results of astounding intensity, should,
when introduced into the blood of refractory animals, behave
like the most harmless inert chemical substances remaining
unaltered for a relatively long period, until finally they are
slowly drawn into the metabolic changes, and gradually undergo
complete oxidation.

Hence it follows that, in the case of these animals, the mutual
attraction between the poison and body cell must be much smaller
than in the case of susceptible animals. There is no absolute
difference, however, between susceptible and refractory animals,

1 Wassermann, "Ueber die personliche Disposition u. Prophylaxe gegen
Diphtherie," Zeit. f. Hyg., xix., 408, 1895.

2

18 TOXINES AND ANTITOXINES.

but only a relative one. The attractive power of the body cells
(receptors) shows a gradual decrease from the most susceptible
to the least susceptible animal. Thus tetanus poison, in a dose
many hundred times greater than is sufficient to kill a mouse,
circulates unchanged through the blood of a pigeon. If, how-
ever, still larger doses are given, the pigeon becomes ill. In this
case then there is not complete immunity, but only a very slight
power of attraction on the part of the receptors. According to
METSCHNIKOFF,1 and FERMI and PERNOSSI (loc. cit.), certain cold-
blooded animals have still smaller powers of attraction.

METSCHNIKOFF found that in the case of fishes, tortoises, and
alligators, and also of arthropoda, the toxine remained unaltered
in the blood without producing antitoxine. Only in the case of
alligators, after longer action (fifty-eight days), did he obtain any
antitoxine, while he was able to bring about this formation of
antitoxine in old crocodiles by keeping the reptiles at a tem-
perature of 30° C.

Even then, however, he could not observe symptoms of illness
in the animals. Similar results were obtained by FERMI and
PERNOSSI in experiments on snakes, tritons, and turtle doves.

Moreover, METSCHNIKOFF detected active tetanus toxine after
the lapse of a month in the livers of scorpions, in which there
had been neither symptoms of poisoning nor any formation of
antitoxine.

The hen has been a particularly favourite subject for experi-
ments with tetanus poison, because although offering extreme
resistance she is not completely immune against it. METSCHNI-
KOFF asserts that the toxine can be recovered from the blood and
the ovaries, and that eventually slight traces of antitoxine appear.
ASAKAWA 2 found that the toxine introduced into hens' blood re-
mained almost unaltered until the seventh day, and then slowly
disappeared without being excreted.

ASAKAWA could not detect any toxine in the brain or spinal
cord of the hen, although he found it in all the other tissues.
This may, of course, be due to the presence of blood in these
tissues leading to some erroneous conclusion as to presence of
toxine, while there is but little blood in the central nervous
system. But, on the other hand, it is also very probable that
slight traces of toxines disappear there by combining with

1 Metschnikoff, "Influence de 1'organisme sur les toxines," Ann. Past.,
xi., 801, 1897; xii., 81, 1898. Also, Immunitat, Germ, translation by
Meyer, p. 264. Jena, 1902.

2 Asakawa, " Die Basis der natiirl. Immun. des Huhnes gegen Tetanus,"
CentrcdU.f. Bakt., xxiv., 166, 1898.

TOXINES. 19

isolated receptors. For the hen is not absolutely proof against
tetanus, and it is also possible to demonstrate a slight antitoxic
activity in the brain, &c., of that bird. Corroborative evidence
of this is afforded by the fact that direct intercerebral injection
of tetanus poison produces the symptoms of tetanus in the hen.

Hence, according to the views of EHRLICH and WASSERMANN,
the defective attraction of the toxine for the body cell (receptor) is
the main cause of natural antitoxic immunity. If the poison
circulates in the free state and does not combine with the
receptors, or does so only to an insignificant extent, the toxo-
phore group is unable to act energetically, and so no serious
injury is done.

Defective attraction, however, is not invariably the cause of
natural immunity. Thus, MORGENROTH * found that in the case
of the frog tetanus poison enters into firm combination even in
the cold, without making the animal ill. Under these conditions
the toxophore group is inactive ; but it acts immediately when
the frog is warmed to about 30° C.

These theories of the attraction of the poison for the living
cell and their specific combination are supported by experi-
mental proofs. WASSERMANN 2 found that fresh substance from
the central nervous system of susceptible animals combined with
considerable quantities of tetanus poison. This was in agree-
ment with the results of METSCHNIKOFF and ASAKAWA, who
found that, in the case of less susceptible animals, the smaller
the degree of susceptibility the less, too, was the amount of
combination in the brain, &c. Thus, the brain of the hen only
enters into feeble combination with the toxine, and that of the
tortoise not at all. An additional support of this theory is
furnished by experiments which show that, in the case of those
animals (e.g., rabbits) in which the tetanus poison combines intra
vitam with receptors which are not attached to the cells of the
central nervous system, emulsions of other organs — e.g., the
spleen — also combine with the tetanus poison (WASSERMANN).

It is not a general rule that receptors only occur in the
particular organs in which the poison produces its injurious
effects. It frequently happens that combination and formation
of antitoxine take place in other organs where the toxine does
little injury. The action of the poison and formation of the

1 Morgenroth, "Zur Kenntnis des Tetanus des Frosches," Arch. Inter-
nat. d. Pharmacodyn. , vii., 265, 1900.

2 Wassermann and Takaki, " Ueber tetanusantitoxische Eigenschaften
des Centralnervenssystems, BerL klin. Woch., 1898, 5; Wassermann,
" Weitere Mitt, iiber Seitenkettenimmunitat," ibid., 1898, 209.

20 TOXINES AND ANTITOXTNES.

antitoxins are two distinct processes which may, under certain
conditions, follow widely different courses.

Fate of Toxines in the Digestive Tract. — The question of what
becomes of toxines when they are introduced into the stomach
or intestinal canal has a special interest. It is the unanimous
opinion of observers that all toxines, including snake poisons,
&c., with the single exception of ricine, are absolutely without
action upon the system from the stomach. And it has been
shown by GiBiER1 that these toxines do not act by way of the
anus.

CHARRIN and CASSIN2 assert that absorption of toxines
through the intestine takes place when the mucous membrane
is injured.

It was proved by NENCKI and SCHOUMOW-SIMANOWSKI 3 that
even large doses were not absorbed in the least through the
digestive tract, and that only when huge doses, more than
100,000 times the lethal amount, were given, did symptoms of
poisoning finally appear.

These results conclusively show that toxines are not absorbed
into the system through the normal intestinal tract. They must
therefore either pass unchanged into the excreta, or they must
be completely destroyed. RANSOM 4 concluded that the first
alternative happened in the case of tetanus poison, but later
workers, notably NENCKI and SCHOUMOW-SIMANOWSKI (loc. cit.)
and CARRIERS,5 were unable to detect any trace of toxine in the
excreta, even when doses as large as 100,000 times the lethal
amount were given ; whilst CARRIERE was also unable to
discover any antitoxic function in the serum after the intro-
duction of toxine per os. Although KEPIN 6 identified abrine
in the faeces he was unable to detect either diphtheria poison or
cobra venom.

Hence, toxines are destroyed ; and in this process three factors
must be taken into account — viz., the living intestinal wall, the

1 Gibier, " Effets produits par les toxines, &c., injectees dans le rectum,"
Sem .Med., 1896, 202 (abst.).

2 Charrin and Cassin, ' ' Fonctions protectrices actives de la muqueuse
intestinale," ibid., 1895, 545.

3 Nencki and Schoumow-Simanowski, " Die Entgiftung der Toxine durch
die Verdauungssafte," Centralbl.f. Bakt., xxiii., 840, 1898.

4 Ransom, "Das Schicksal des Tet. Giftes nach seiner intestinal Einver-
leibung," Deutsch. med. Woch., 1898, 117.

5 Carriere, "Toxines et digestion," Ann. Past., xiii., 435, 1899 (gives
Bibliography); also, "Du sort de la toxine tetanique introduit dans le
tube digestif," Soc. BioL, li., 179, 1899.

8 Repin, " Sur 1'absorption de 1'abrine par les muqueuses," Ann. Past.,
ix., 517, 1895.

TOXINES. 21

intestinal bacteria (FERMI and PERNOSSI), and the secretions of
the intestine.

From the concordant results of experiments made by NENCKI
(loc. cit.) on PAWLOW'S dogs and of those by CARRIERS (he. cit.)
with preparations of enzymes, it is beyond question the digestive
ferments that render toxines innocuous. Thus, CARRIERS (loc.
cit.) found that even the saliva diastase had an injurious influence,
that pepsin was not very active, but that trypsin, and, above all,
the bile, had the greatest effect. NENCKI obtained the following
results in experiments with pure sterilised fluid from fistulas : —
Pepsin, by itself, destroyed bacterial poisons (but not abrine).
The presence of acid was not essential, since the same result was
obtained after nearly complete neutralisation, as had also been
stated previously by CHARRIN.1 Pancreatic juice, by itself, had
a greater destructive effect upon diphtheria toxine than upon
tetanus toxine, the latter being particularly sensitive to the
action of a mixture of 3 parts of pancreatic juice and 1 part
of bile.

Attempts to produce immunisation by the simultaneous injec-
tion of bile with the toxine were unsuccessful. CHARRIN and
LEVADITI 2 injected diphtheria poison (100 times the lethal dose)
into freshly extirpated pancreas, and found that it was completely
destroyed after twenty-two hours. Muscle plasma or pancreas
heated to 70° C. were without action. According to CARRIERS,
no destructive effect is to be attributed to the mucous membrane
of the intestine or the intestinal bacteria.

BALDWIN and LEVENES also proved that pepsin, trypsin, and
papayotin destroyed diphtheria toxine.

Notwithstanding this, CANO-BRUSSO4 still asserts that the
destruction of the tetanus poison in the intestine is to be
attributed to the action of the mucous membrane itself.

Mode of Action of Toxines. — Toxines, as shown above, do not
act through the digestive tract, and it is necessary to introduce
them by other means into the organism. The most convenient
method is subcutaneous injection, just as in cases of poisoning by
living bacteria.

1 Charrin, "Action des sues digestifs sur les poisons microbiens," Arch,
de Phyx., 1898, 67. Charrin and Lefevre, "Action de la pepsine sur la
toxine diphth.," Soc. BioL. xlix., 830, 1897 ; Sem. Med., 1897, 296.

2 Charrin and fLevaditi, "Action de pancreas sur la toxine diphth.,"
Soc. BioL, li., 215, 1899.

3 Baldwin and Levene, "Action of proteolytic ferments on bacterial
toxines," J. Med. Research, vi., 120 ; Malys Jb., 1901, 953.

4 Cano-Brusso, "Untergang d. Tetanusgiftes im Darm," Malys Jb.,
1901, 914.

22 TOXINES AND ANTITOXINES.

Direct introduction into the circulatory system (intravenous)
is still more effective than intraperitoneal, intercerebral, or sub-
dural inoculation, as has been used in the case of tetanus and
gonococci poisons, or yet HOMEN'S method of injection into the
nerves, which is sometimes employed.

Intercerebral injection is of special importance in cases where
either the poison when distributed throughout the body is
seized by the receptors or other less susceptible organs (tetanus
in the rabbit), or where the receptors in the brain are not very
numerous, so that only concentrated toxine can effect any serious
injury (tetanus in the hen).

As regards the activity of toxines, there are two factors of
fundamental importance — their specific character, and the time
of incubation.

The specific character is one of the most prominent character-
istics of true toxines. Although more or less thorough resistance
to the action of crystalloid poisons is known, cantharidine, for
instance, being relatively innocuous to the hedgehog and atropine
to pigeons ; yet in these cases there is only a weakening of the
activity of the poison, and not an absolute resistance. Certain
bacterial poisons, however, are completely harmless to refractory
animals, while they act with the greatest energy upon susceptible
animals.

But the most important point in this connection is that, in the
case of refractory animals, the toxine is by no means destroyed,
but that it circulates unchanged in their blood as a completely
inert substance.

Thus arises the paradoxical phenomenon which we have de-
scribed at some length above, that it is possible to kill a mouse
with tetanus, by means of the blood of an apparently healthy
hen which has previously been inoculated with large doses of
tetanus poison. When the toxine cannot find corresponding
receptors it is unable to make its attack. The toxophore group
remains inactive, with the result that the toxine is a completely
inert substance which the body regards as of so little importance
that it does not even make speedy attempts to destroy it. This
phenomenon, too, can readily be explained by the side-chain
theory. According to EHRLICH, all nutritive substances, so far
as they are not merely chemically changed by the secretions and
their enzymes, enter into combination as haptines, and are thus
brought within the power of the destructive and assimilative
forces of the protoplasm. But, since the toxine does not enter
into such combination, it also is not destroyed — not even treated
in the same way as nutritive substances.

TOXINES. 23

The specific nature of true toxines completely corresponds with
that of the living bacteria.

It is also characteristic of most of the known toxines that,
unlike most simple crystalloid poisons, they do not exert their
activity instantaneously or after a very short time, but that their
toxic effects only appear after a definite time — the period of
incubation. In this respect, too, bacterial toxines behave exactly
like their parent living cells. The period of incubation varies,
not only with the nature of the toxine, but is also dependent
on other factors, such as the amount of the dose, the body tem-
perature, &c. There is a limit, however, to this dependency,
particularly as regards the amount of dose. The period of in-
cubation is not in any way inversely proportional to the quantity
of toxine. For example, according to COURMONT and DoYON,1
the period of incubation for guinea-pigs after injection of a lethal
dose of diphtheria toxine is fifteen hours, and this cannot be
reduced below twelve hours even by the introduction of enormous
doses (90,000 times the lethal amount). A very interesting
phenomenon in connection with this point is that in tetanus of
the frog the incubation period can be delayed to any desired
extent, by keeping the temperature low. Thus, if a frog into
which the poison has been injected be kept at 8° to 10° C., it
remains healthy, but at 30° C. tetanus begins after a definite
time, and the frog dies. If, again, the warming be interrupted,
the frog can be kept at 8° C. as long as desired, but after com-
pletion of the remainder of the incubation period at a higher
temperature, tetanus symptoms appear (MORGENROTH).

In the case of some other toxines, however, notably snake
venom, there is not this incubation period. These act with
extreme rapidity.

KRAUS2 found a true antitoxine-producing toxine in the filtrates of
cultivations of Naskin's vibrio, which resembles the cholera vibrio. It
killed rabbits without an incubation period when intravenously injected in
doses of 0'5 to 1 c.c. There was an incubation period after subcutaneous
injection.

Constitution of Toxines. — The characteristic of every toxine is
that it is a haptine — in other words, that it contains a haptophore
and a toxophore group.

Taking their constitution as thus fixed, most bacterial toxines

1 Quoted by Deutsch and Feistmantel, Die Impfstqffe und Sera, Leipzig,
1903, 40.

2 Kraus, " Ueb. ein acut. Wirkendes Bakterientoxin," Centralbl. /,
Bakt., xxxiv., 488, 1904.

24 TOXINES AND ANTITOXINES.

are, according to EHRLICH'S terminology, simple hapiines of the
first class. Other toxines, however, have a more complicated
structure. Thus, ricine and abrine possess, in addition to their
toxic properties, the further characteristic of agglutinating the
red corpuscles of the blood.

The question now arises whether this second property is to be
ascribed to some peculiar active principle invariably present in
preparations of ricine, or whether the ricine itself is endowed
with this double power. One of the proofs that the toxic and
agglutinating properties are separable was afforded by the fact,
that digestion with pepsin-hydrochloric acid soon destroyed the
toxicity, but did not affect the power of agglutination (JACOBY,1
MICHAELIS and OppENHEiMER2). It was thus possible that
ricinus toxine and ricinus agglutinine might be quite distinct
substances, analogous to tetanospasmine and tetanolysine, the
two separate poisons of tetanus cultivations.

While, however, it could be proved that these were separate
substances, which occurred in quite different proportions in
different cultivations, and produced two distinct anti-bodies,
similar experiments with ricine gave opposite results. The
toxine and the agglutinine possess the same haptophore group,
since they produce only one antiricine (JACOBY).

The structure of ricine, therefore, is complex. In additon to
its haptophore group it possesses two ergophore groups, with
toxic and agglutinating powers respectively ; and it is thus a
complex haptine of the first class. (For further particulars
see Ricine.^

In the case of other toxines the constitution is still more
complex and uncertain. Here, too, other activities (e.g., hsemo-
lytic), in addition to the toxic function, are frequently met with,
notably in the case of eel's blood and snake poisons. In such
instances, it still remains to be determined with certainty
whether the disintegrating principles are distinct substances.
This is probably the case, however, with snake poison, for
example. And, if so, the further question arises whether the
ly sines, like the toxines, are hap tines of the first class — i.e.,
contain the haptophore and ergophore groups in one complex
combination ; or whether they do not rather approximate in
constitution to the haptines of the second class, such as the
hsemolysines of normal and immune sera. The latter contain

1 Jacoby, "Ueber Ricimmmunitat," Hofm. Beitr. z. Chem. Physiol.,
i., 57, 1901.

2 Michaelis and Oppenheimer, " Ueber Immunitat geg. Eiweisskorper,"
Engelmanris Arch., 1902, Suppl. H.

TOXINES. 25

haptophore and ergophore groups which are not united in one
complex grouping ; but they have an amboceptor with two hapto-
phore groups and a complement, which attaches itself to the
complementophile group of the amboceptor after the latter has
combined with the receptor of the cells, and thus brings about
the hsemolytic action.

The latest results obtained with cobra hsemolysine lead to the
conclusion that this poison actually has some such constitution
— that it possesses an amboceptor which is seized upon and
brought into action, not only by an endo-complement that occurs
in the red blood-corpuscles, but also by lecithin (KYES and SACHS).
We shall return to this in discussing snake poison.

Physiological Action of Toxines. — Although the primary char-
acteristic of each of the bacterial toxines is its specific action upon
the organism, as will be described more fully in the special part,
they have also certain reactions in common, of which a brief
mention should be made. They act primarily upon the general ,
health, producing weakness, prostration, and eventually collapse, y
These phenomena are undoubtedly due, in the main, to a dele-
terious effect of the toxine on the heart's action, this effect being |
manifested in a reduction in the pressure of the blood, frequently
accompanied by a diminution in the rate of pulsation and eventu- \
ally by paralysis of the heart. According to BARDiER's1 compara-
tive experiments, however, individual toxines differ in their action
upon the heart.

Hypothermia, with or without preceding fever ', is an almost »
invariable symptom. When injected into the skin and the |
subcutaneous tissue toxines frequently give rise to infil-
trations, abscesses, and necroses. Loss of hair is frequently
observed.

Internal organs, too — e.g., the intestinal tract — are usually
injured (diarrhoea, &c.). Nor does the liver remain unaffected,
as has been demonstrated, notably by TEISSIER and GuiNARD.2
CLAUDE 8 has observed bleeding of the gall-bladder. According
to PADOA,* the liver acts differently upon different toxines.
Thus, while it combines with and lessens the virulence of the
diphtheria toxine when introduced into the mesenteric vein, it
allows typhus toxine to pass, which acquires a higher virulence

1 Bardier, "Toxine et Coeur," Soc. Biol., xlix., 311, 1897.

2 Teissier and Guinard, "Effets des toxines microbiennes," Arch. Med.
Exper., ix., 994, 1897.

3 Claude, "Deux cas d'h^morrhagie de la v^sic. biliaire, &c.," Soc. Biol.,
1896, 169; Sem. Mtd., 1896, 62.

4 Padoa, "Ueber d. verschied. Wirkung. des Diphtheric- u. Typhus-
toxins," Riform. Med., 1899, No. 26; MalysJb., 1899, 921.

26 TOXINES AND ANTITOXINES.

when thus introduced into the system than when injected sub-
cutaneously or into the veins. Degenerative changes also take
place in the kidneys.

Alterations of the blood (hcemoglobincemia) and vascular system
are, in like manner, of frequent occurrence. Many toxines appear
to have a solvent action upon the blood-corpuscles. According
to KAEFUNKEL,1 they reduce the alkalinity of the blood, but
this effect can be partially prevented by the application of
artificial heat. Lastly, the nervous system and, in particular,
its central organ are usually attacked, but in very different
ways ; this will be dealt with more fully in the special part.

CoNSiGLio2 states that he has observed a peculiar charac-
teristic of diphtheria toxine. He has found that in small
doses it stimulates the process of fermentation, but in larger
doses checks it ; whilst, on the other hand, it has invariably a
very injurious influence on the germination process of seeds.

Toxoids and Toxones. — According to the side-chain theory,
we must regard toxines as bodies that possess two distinct
stereo groups — the haptoplwre group and the toxophore group.
Assuming, now, that under certain conditions the toxophore
group may be so changed as to lose its characteristic activity
while the haptophore group remains unaltered, new substances
will be formed which still have the power of attaching them-
selves to receptors whether free (antitoxines) or combined (body
cells), but without being poisonous. EHRLICH has investigated
substances of this kind in the case of diphtheria toxine, and has
established their great influence upon the toxicity of solutions
of poisons and in the preparation of curative sera, with which
we shall deal at length in the next chapter. These "toxoids,"
then, are non-poisonous, but are still haptones with specific powers
of combination. When they are secondary decomposition pro-
ducts of true toxines they are termed " toxoids," in the narrower
sense of the word ; but there are also primary bacterial products,
which are able to seize upon the same haptophore group as the
toxine, but which have a different and much weaker toxophore
group, capable of producing slight characteristic effects of its
own, as has been demonstrated by EHRLICH and MADSEN in
the case of diphtheria. These primary substances, which thus
represent a second secretion product of the bacteria, are known as
"toxones"

1 Karf unkel, ' ' Schwankungen des Blutalkaleszenzgehaltes nach Einver-
leibung von Toxinen, &c.," Zeit. f. Hyg., xxxii., 149, 1899.

2 Consiglio, " Azione di alcune tossine, &c.," Arch, di Farm., vi., No. 3,
1898; MalysJb., 1898,634.

ANTITOXINES. 27

has shown that toxoids of tetanus poison which have become
quite non-poisonous no longer induce the formation of antitoxine. Accord-
ing to his views, when side-chains are broken off there must be present in
addition to the haptophore group a certain "combining ring" derived
from the toxophore group.

Toxones are not absolutely innocuous, but eventually produce
temporary symptoms of poisoning (paralysis, &c.), which, how-
ever, are absolutely different in kind from the effects of small
doses of toxines.

Toxoids have been identified with certainty in the case of
diphtheria (EHRLICH), of tetanolysine (q.v.) (MADSEN) and of
staphylotoxine (NEISSER and WECHSBERG, q.v.), and JACOBY (loc.
cit.) has shown that in all probability they accompany ricine.

There is also, however, a considerable weight of evidence that
some at least of the other bacterial poisons have the power of
producing secondary toxoids, in the case of tetanus, for example,
to which we shall return in the special part.

VAN CALCAR2 succeeded in separating the toxones from the toxine by
taking advantage of the fact that animal membranes stretched to a certain
extent allow the toxine to pass, but not the toxones. Hence, toxones
have a greater molecular volume than toxines. When the membrane was
stretched to a greater extent the toxones also diffused through it.

This is the extent of our knowledge of toxoids and toxones.
Since they, too, are specific haptines, they also produce anti-
toxines as MADSEN and DREYER 3 succeeded in demonstrating in
the case of diphtheria toxones.

MADSEN and DREYER were also able to prove the existence
of toxones which were poisonous to rabbits, but without action
upon guinea-pigs. A poison that was without action upon
guinea-pigs after neutralisation in the ratio of 200 : 200, did not
become innocuous to rabbits until the ratio of the neutralising
antitoxine reached 240 : 200. EHRLICH has given the name
" toxonoids " to this variety of poison.

On Antitoxines in General. — Our definite knowledge of anti-
toxines is even smaller than our knowledge of toxines.

They occur in the fluids of the body, notably in the blood
serum and milk of immunised animals. Slight traces are also

, "Beitr. z. Theorie d. Immunitat," Zeit. f. Hyg., xlvi., 176,
1904.

2 Van Calcar, "Ueb. die Constit. des Diphtheriegiftes," Berl. klin.
Woch., 1904, No. 39.

3Madsen and Dreyer, "Ueber Imrnun. mit den Toxonen d. Diphtherie-
giftes, Zeit.f. Hyg., xxxvii., 249, 1901.

28 TOXINES AND ANTITOXINES.

of frequent occurrence in nornal sera, as, for instance, diphtheria
antitoxins in the serum of horses (in the case of about 30 per
cent.). Other anti-bodies are also very frequently present in
normal sera, notably anti-ferments, &c. They can be separated
from the fluids of the body by precipitation methods similar to
those employed for the concentration of toxines.

BLUM l found that lymph glands that had undergone auto-digestion were
antitoxic to tetanus poison, but not to diphtheria virus or cobra venom.

Precipitation with alum and ammonia (ARONSON), with mag-
nesium sulphate (TizzoNi), and with solid sodium or magnesium
chlorides (BRIEGER and BOER), followed by a method of puri-
fication, yield dry preparations of antitoxines.

The nature of antitoxines is unknown. They are probably
albuminous substances, but this has not been definitely proved.
The fact that they offer considerable resistance to the action of
trypsin is against this view, although they are very sensitive to
the action of pepsin-hydrochloric acid. Experiments to obtain
further proof that diphtheria is an albuminous substance, or to
determine to which albuminous constituent of the blood it is
related, have not as yet led to any conclusive result.

Antitoxines, like toxines, are sensitive to the action of heat,
acids, &c., although, in general, they offer far greater resistance
(see Diphtheria). CAMUS 2 found that anti-snake venom and
anti-diphtheria poison could be heated for thirty minutes at
120° C. and for fifteen minutes at 140° C. without injury,
provided they had been dried at a lower temperature, and then
heated at 100° C. in a current of air.

According to EHRLICH antitoxines are normal constituents of
the cells — broken-off receptors — and as such possess relatively
little physiological or chemical activity. They are not invariably
produced in the organs where the poison exerts its specific action,
but are also developed in other groups of cells. This appears to
be especially the case in tetanus of the rabbit (q.v.). They are
secreted under the stimulus of the haptophore group. EHRLICH
regards them as "simple uniceptors'' — i.e., as substances with only
one haptophore group, which coincides with the corresponding
group of the toxine. Hence, nothing is more erroneous than to
ascribe to antitoxines an activity similar to that of the toxines —

1Blum, "Ueb. Antitoxinbildung bei der Autolyse," Hofmeister's Beit-
rage, v., 142, 1904.

2 Camus, "Resistance aux temper, elevees des vaccins desseche"s," Soc.
BioL, L, 235, 1898.

RELATION OF TOXINES TO ANTITOXINES. 29

to represent them as " substances akin to ferments" There is no
support for the unthinking transference "of this notion from the
toxines, and its only result would be to weaken the conception
that toxines are closely related to ferments. Toxines, and
probably also some ferments, have a haptophore and at least
one " ergophore " group. They can not only combine, but also
attack. Antitoxines do not possess the latter characteristic;
they can only combine, and so ward off the toxophore group
from the threatened cell, but not injure it. WASSERMANN'S
experiments on pyocyaneus poison and CALMETTE'S on snake
poison prove that the toxine remains intact in a mixture of
toxine and antitoxine, and that after destruction of the anti-
toxine the toxine can again manifest its activity (vide infra).
Antitoxines, therefore, are not active substances, not "ferments."
Antitoxines, as such, are physiologically completely inert, and
are unable to produce any toxic effects.

This, of course, only holds good for antitoxines by themselves
and not for the sera in which they are present. It is true that
unlimited quantities of horse serum containing antitoxine can be
injected into a horse without producing any by-reactions. But,
on the other hand, albuminous substances foreign to the body
are in a certain sense invariably poisonous. They, too, give rise
to protective substances — the precipitines described by MYERS
and others.1 It is thus evident that unlimited quantities of
horse serum cannot be injected into animals of different species.
In fact, disturbances due to this cause have frequently been
observed in the therapeutic use of diphtheria serum (q.v.). In
such cases, however, the disturbances must be attributed to the
serum rather than to the antitoxine itself.

There is, doubtless, a connection between the struggle of the
organism to eliminate foreign albuminous matters and KNORR's2
observation that antitoxines injected with sera foreign to the
body speedily disappear, whilst antitoxines introduced in the
sera of the same species remain for a very long time in the
organism.

Behaviour of Toxines towards Antitoxines. — We have already
stated in the introduction that an essential characteristic in the
definition of a toxine is that true toxines produce an antidote, an
antitoxine in the body of the attacked organism. This fact,

1 For further particulars and bibliography of precipitines cf. Michaelis
and Oppenheimer, "Ueber Immunitat geg. Eiweisskorper," Engelmanris
Arch., 1902, Suppl. H.

2 Knorr, "Das Tetanusgift u. s. Bezieh. zum tier. Organismus," Munch,
med. Woch., 1898, 321, 362.

30 TOXINES AND ANTITOXINES.

which was first discovered by EHRLICH * in his fundamental ex-
periments with ricine, a vegetable poison closely akin to the
bacterial toxines, has now become so firmly established that we
even have to regard the formation of antitoxine as a radical
property of the true toxine. It is not possible in this place to
deal with the significance that this formation of antitoxine in
the organism has for the disappearance of infectious diseases, and
for the development of the state of acquired immunity, or with
the way in which these phenomena have been used as supports
of the monumental side-chain theory.

Here it is only necessary to describe the experimentally-
determined relation between the toxine and its antitoxine as
exactly as is possible in the present state of our knowledge,
which, for the most part, we owe to the unwearying classical
researches of EHRLICH.

According to the side-chain theory, the only poisons that can
act as true toxines are those that possess a specific affinity for the
definite cells. For the representation of this specific affinity,
EHRLICH assumes that both sides, the toxine on the one hand,
and the attacked cell on the other, each have in their protoplasm
an atomic group which reciprocally coincide, and thus enter into
combination, and bring the toxine within the immediate reach of
the cells. The first step in the action of the toxine is thus a concen-
tration of the poison upon the cell by means of the reciprocal
" haptophore " groups. The cell is now through this concentra-
tion brought within the sphere of action of the toxine, and then
follows as the second phase the specific action of the poison upon
the cell — a function of the second specific the "toxophore" group.2

Toxines thus combine with the haptophore groups of cells,
which are active as regards their " side-chains." If now, as in
artificial immunisation, such side-chains provided with hapto-
phore groups are produced in excess, and separated in a free
state in the fluids of the body, especially the blood serum, these
haptophore groups retain their capacity of entering into combina-
tion with the corresponding haptophore groups of the toxine.
Hence these brok en-offside-chains represent the specific antitoxine
to the toxine.

1 Ehrlich, " Experimentelle Unters. iiber Immunitat," Deutsch. med.
Woch., 976, 1218, 1891; also " Zur Kenntnis der Antitoxinwirkung,"
Fortsch. d. Med., 1897, 41.

2 The theory only speaks of atomic groupings in a substance, and Ehrlich
has never asserted that a toxine consists of two substances — a haptophore
and a toxophore substance — as Danysz (Ann. Past., 1899, 581) credits him
with stating. Danysz, who confuses the process of plasmatolysis with
toxine activity, has misunderstood Ehrlich.

RELATION OF TOXINES TO ANTITOXINES. 31

This conception at once gives us two highly important points
of view regarding the reciprocal relationship of the toxine to the
antitoxine.

Its acceptance immediately renders untenable two possible
modes by which the poison might be influenced by its specific
anti-body — viz., by direct destruction of the poisonous substance
in its entirety, as it might, for instance, be destroyed by, say, a
strong acid ; and secondly, by the action of the antitoxine on the
specifically injurious toxophore group of the poison, just as, for
instance, the poisonous property of aniline is considerably reduced
by the introduction of acetic acid into its toxic amino-group.
Neither is reconcilable with the side-chain theory, for there can
only be any influence in the sense that the antitoxine saturates
the haptophore group of the toxine, and thus prevents the possi-
bility of its bringing its toxophore group into action by seizing
upon the cell, while in reality its toxic force remains unchanged.

While we have deduced this fundamental point of view as a
consequence of the side-chain theory accepted by us as a heuristic
principle, the actual development of the theory has naturally
proceeded in a converse manner. Tedious experiments were
first made to establish the correctness of these facts in order to
use them as important supports of the theory. EHRLICH and
BEHRING were at first of opinion that the poisonous property of
the toxine was affected by the antitoxine, but subsequently came
to the conclusion that it was a question of simple combination.

The facts that led to this now generally accepted conclusion
were of different kinds.

At first it was the general view that the action of an anti-
toxine was only an indirect one, its function being to render the
organism "proof" against the toxine.

This opinion was afterwards abandoned when it was found
that toxine and antitoxine combined in accordance with definite
arithmetical laws (law of multiples), with which we shall deal
more fully later on. Then results of the greatest importance
were obtained by a closer study of processes in which the inter-
vention of the living organism could be absolutely excluded, and
in which plainly visible reactions in the test tube were used as
indicators of the influence of the antitoxine upon the toxine.

The earliest of these were EHRLicn's1 celebrated experiments
upon the agglutinating action of ricine upon the red corpuscles
of the blood, the results of which showed that there were fixed
numerical relationships between ricine and antiricine, inasmuch

1 Ehrlich, " Zur Kenntn. d. Antitoxin wirkg.," Fortschr. d. Med., 1897,
41.

32 TOXINES AND ANTITOXINES.

as that the action of definite doses of ricine upon the blood was
invariably exactly neutralised by corresponding doses of anti-
ricine. The action of other hsemolysines, such as snake poison
(KANTHACK), eel's blood (KOSSEL), and tetanolysine (EHRLICH).
<fec., can be counteracted by similar means.

Direct relationships between toxine and antitoxine were thus
established. But these might still depend upon a direct destruc-
tion of the poison by the antidote. The facts, however, showed
that this was not the case, but that there was a simple combina-
tion of the two constituents.

In particular, the fact that it is possible, when this compound
has only been formed a short time, to break it up in such a way
that the original toxic activity is restored, has afforded the
strongest support to the theory that there is a loose form of
combination. CALMETTE was the first to demonstrate this fact
with certainty in the case of the animal toxine, snake poison, the
antitoxine of which is much more unstable than the toxine.

It has also been found by WASSERMANN that the antitoxine of
B. pyocyaneus (q.v.) is much more readily destroyed than the
toxine, so that we are justified in assuming that there is also a
simple combination in the case of bacterial toxines.

The gist of these experiments was that a neutral mixture of
toxine with antitoxine required a large proportion of the original
toxic capacity on being heated, owing to the fact that the anti-
toxine, being the more readily attacked constituent of the loose
combination, was destroyed by the higher temperature. The
two components, however, must only have been mixed a short
time, since otherwise it was no longer possible to separate the
compound. A process of diffusion can also be used to separate
the two components in the case of snake poison, the toxine of
which is much more diffusible than the antitoxine (MARTIN and
CHERRY) (cf. Snake Poison).

On the other hand, similar experiments with diphtheria poison have
proved unsuccessful (DziERZGOWSKi1). Now, here the conditions are quite
different. In the first place, as EHRLICH has shown, the combination is
exceedingly stable. But, apart from that, the toxine in this case is the
more destructible, so that on heating the mixture it is not the toxine but
the free antitoxine that ought to be regenerated. The reason why this
does not happen is manifest a priori; for in the change that takes place
in the toxine on heating the poison does not disappear, but is only trans-

1 Dzierzgowski, " Zur Frage liber die Beziehungen zwischen dem anti-
diphth. Heilserumu. d. Diphtherietoxin," Arch. Internal. dePharmacodyn.,
v., 1, 1898. Cf. Marenghi, " Ueber d. gegens. Wirkg. antidiphth. Serums
und des Diphth. -Toxins," Centralbl.f. Bakt., xxii., 520, 1897.

RELATION OF TOXINES TO ANTITOXINES. 33

formed into toxoids, whilst, the combination not being broken up, no free
antitoxine can be detected. Thus, the negative results of these experi-
ments prove nothing, since they might have been predicted beforehand
to be theoretically very probable.

This combination is a chemical reaction, and as such obeys
the laws of chemical kinetics. Very considerable differences
may be shown as regards firmness of combination and speed of
reaction. Thus, diphtheria antitoxine has a far greater affinity
for its toxine, and combines with it much more rapidly (five to
ten seconds) than is the case with tetanus antitoxine and its
toxine (EHRLICH).

The degree of attraction for reciprocal saturation and the
speed of the reaction also depend, to a very considerable extent,
upon the temperature (increasing with its rise) as well as upon
the concentration (EHRLICH, KNORR 1). Combination takes place
much more rapidly in concentrated solutions.

We must defer consideration of the question of the condition
of equilibrium between toxine and antitoxine.

The view that antitoxines do not combine with toxines in
fixed numerical proportions, but that their action depends upon
a protective influence upon the cells, has not yet been entirely
abandoned, notwithstanding all proofs to the contrary. In
particular, attempts have been made to base this conclusion on
the alleged facts that when the dose of toxine is multiplied the
same multiple of antitoxine is insufficient — i.e., that its "pro-
tective " power fails with large doses of poison. This view has
recently been advanced again — e.g., by BOMSTEIN.

But EHRLICH'S experiments on ricine, and the absolutely ana-
logous results of CALMETTE with snake poison^ of CAMUS, KOSSEL,
&c., with eeVs Hood poison, and of many others with erythrocytes,
lead us to exclude any intervention on the part of the organism,
and are only to be explained on the assumption of a direct
fixation of the poison by the antitoxine. And apart from this,
the assertion that there is a discrepancy in the numerical pro-
portions rests upon very insufficient data.

COBBETT and KANTHACK2 were able to demonstrate that the
multiples exactly corresponded with theory, provided a large
multiple of the lethal dose was used at the very beginning of
the experiment. They give a simple reason for concluding that
when a quantity of poison approximating a single lethal dose is

1 Knorr, * ' Die Entstehung des Tetanusantitoxins, " Fortschr. d. Med.,
1897, 657.

2 Cobbett and Kanthack, "Ueber das Schicksal d. Diphtherietoxins im
Tierorganismus," Centralbl.f. Bakt., xxiv., 129, 1898.

3

34 TOXINES AND ANTITOXIN ES.

used at the beginning, it is extremely probable that symptoms
of poisoning will appear on multiplying the respective doses. For
it is quite possible, when a single lethal dose is neutralised, for
a slight excess of poison to remain unnoticed in the mixture,
since it may not even reach the minimum amount necessary to
cause illness ; but if now the relative quantities of poison and
antitoxine be increased tenfold, this excess of poison is also
multiplied by ten, and the poisonous action of the mixture is
manifested. Ehrlich's theory cannot be upset by such proofs
as these.

Taking all things into consideration, we are justified in
assuming, on practical and theoretical grounds, that the action
of an antitoxine upon a toxine consists essentially of a reciprocal
combination of two groups endowed with specific affinity.

This at once leads to the fundamental conclusion that the
reciprocal action of the two substances must follow the laws
that hold good for the reciprocal saturation of two simple
chemical substances possessing atomic groupings specifically
adapted to each other — i.e., that they must combine in fixed
quantitative proportions. Just as the same amount of pure
sodium hydroxide invariably requires the same amount of
pure hydrochloric acid for neutralisation, so must the propor-
tion between a definite dose of toxine and the amount of
antitoxine that exactly " neutralises " it be an absolute constant.
A given quantity of pure toxine must invariably require the
same dose of pure antitoxine for its activity to be exactly
neutralised, provided that the combination is stable, and does
not lead to a state of dissociated equilibrium, a question with
which we shall deal later.

The difficulty of establishing this extremely important fact
is enormously increased by the circumstances. In the first
place, neither toxines nor antitoxines are known in the free
state. We are not dealing here with substances that can
be isolated as chemical entities, and to which the balance can
be applied in determining whether x grms. of diphtheria anti-
toxine invariably neutralise y grms. of diphtheria toxine. No ;
the only measurement that is applicable to these poisonous sub-
stances is the physiological one — the determination of the "single
lethal dose" which we are compelled to adopt as the fundamental
unit of measurement for toxines, or, in the case of hcemolysines,
the measurement of the amount of solvent action.

This drawback, however, would not be so very serious if we
could only establish a constant relationship between every toxine
solution of definite strength and a given antitoxine solution, so

RELATION OF TOXINES TO ANTITOXINES. 35

that eventually every " single lethal dose " of poison would cor-
respond to a definite number of "antitoxine units" For this
purpose the strength of the solution might be readily calculated
upon the known toxicity of 1 c.c. of a solution to be taken as
unity (normal poison). Unfortunately, this is beyond the range
of practicability. Almost every solution of toxine stands in a
different proportion toward the amount of antitoxine required
for its neutralisation, when the ratio of a "lethal dose" to the
number of " antitoxine units " is calculated.

Here we are face to face with the most extraordinarily com-
plicated conditions, and there still remain certain obscurities and
difficulties to be explained, although the confusion has been
for the most part cleared up by the painstaking researches of
EnRLiCH.1 In the first place, the rapidity with which toxine
and antitoxine combine depends not only on their nature but
also on the concentration of the two components (vide supra}.
But, above all, it has been found that every bouillon of diph-
theria virus contains, in addition to the specifically active toxine,
varying proportions of other substances which possess the hapto-
phore, though not the toxophore, group of the true toxines, and
which, therefore, make the same demand upon the antitoxine as
the toxine itself, although they have no influence upon the lethal
dose — the toxic effect of the poison. Thus, in determining the
unit of measurement, — the single lethal dose, — these substances
escape observation, but their influence is immediately manifested
when an attempt is made to determine the amount of a given
antitoxine solution required to neutralise this single lethal dose.

If a pure solution of toxine would require a given number of
c.c. of a standard solution of antitoxine, this number would be
increased in proportion to the quantity of these non-poisonous,
though antitoxine-consuming, substances in the impure toxine
solution. Hence, the varying amount of these substances in
every solution of toxine enormously increases the difficulty of
establishing the absolute constancy of these proportions, as
demanded by the side-chain theory ; and these difficulties have
not yet been completely overcome in every case.

In order to obtain a clear idea of the immediate factors in this
question it is necessary for us to begin with the physiological
units of measurement devised by BEHRING and EHRLICH for the
study of the action of antitoxines. The numerical definitions
fixed for diphtheria virus are as follows : —

1 Ehrlich, "Die Wertbemessung des Diphtherieserums," Klin. Jahrb.,
vi., 299, 1899; "Ueber die Constit. des Diphtheriegiftes," Deutsch. med.
Woch., 1898, 597.

36 TOXINES AND ANTITOXINES.

The single lethal dose is defined by EHRLICH as that amount of
toxine, expressed in c.c. of the poison solution or in grammes
of the solid poison, that is just sufficient to kill a guinea-pig
weighing 250 grms. (an animal about six weeks old) within
four or five days. This dose is the physiological toxic unit.

The definition of normal toxine, as fixed by v. BEHRING, is a
solution of poison that contains 100 lethal doses in 1 c.c. The
term "normal toxine" is abbreviated by v. BEHRING into
DTN, M250 (Dijihtherietoxin normal einfach, Meerschweinchen
von 250 grms. — i.e., single normal diphtheria toxine, guinea-
pigs of250 grms.).

Now the measurement of the strength of antitoxine solutions
is based upon this arbitrary toxic unit. A " single " serum is
one, 1 c.c. of which is capable of neutralising 1 c.c. of the normal
toxine — i.e., 100 toxic units. This amount (1 c.c. of the single
serum) is the antitoxine unit, the so-called immunity unit, briefly
designated I.E. (Immunitats-Einheit)1 and has thus been em-
pirically established and retained (vide infra).

When, then, a serum is first prepared against a new toxine, it
is invariably the rule in all the experiments made at the same
time to express the relationship of the toxine to the antitoxine
solution in terms of cubic centimetres. And since in the case
of this fresh toxine the ratio of toxic activity to the number of
c.c. used remains constant, there is also a constant relationship
between the toxic activity and the amount of antitoxine — i.e., every
lethal dose invariably corresponds exactly with the same amount
of antitoxine solution, expressed in cubic centimetres.

If, however, this toxine be allowed to stand for some time
and its relative value towards the serum be again determined, it
will be found that the conditions of quantitative combination
have materially changed in one respect. It is true that the
ratio of toxine to antitoxine solution, expressed in c.c., will
still remain constant — i.e., that every c.c. of the toxine solution
requires the same amount of antitoxine solution as the fresh
toxine, but at the same time this quantity of toxine solution,
expressed in c.c., has considerably smaller toxic activity than an
equal amount of the fresh poison. If, on the other hand, a
determination be made of the amount of antitoxine required to
saturate one toxic unit, it will obviously be found that a con-

1 Mad sen ("Constitution du poison diphth.," Ann. Past., xiii., 568,
1899) has introduced several other abbreviations. T = toxine unit; (T)
the amount of toxine bouillon in c.c. containing T ; I = immunity unit (in
German abbreviated to I.E.); and (I) the amount of serum in c.c. that
contains I. We shall occasionally use these abbreviations.

RELATION OF TOXINES TO ANTITOXINES. 37

siderably greater quantity is necessary than was the case with
the fresh toxine.

Hence, it follows that the toxine solution becomes weaker on
keeping — i.e., that the toxophore group has become inactive in part
of the toxine ; but since the toxine solution still requires the
same number of c.c. of the serum for neutralisation as before, it
is plain that the haptophore groups have remained intact during
this weakening process. From this it follows that there must be
present in this weakened toxine solution, substances which have
become non-poisonous through loss of their toxophore group,
although, owing to their possessing intact haptophore groups,
they are as capable as before of entering into combination with
antitoxine.

These substances are termed toxoids l by EHRLICH.

EHRLICH has also introduced two values of limitation which
he terms L0 (limes, "Nil") and L+ (limes, "Death"). The
numerical significance of these terms is as follows : —

L0 is that quantity of the toxine solution under examina-
tion, expressed in toxic units (lethal doses), which, when mixed
with an immunity unit, is completely neutralised thereby — so
neutralised that absolutely no symptom of poisoning appears.
This mixture of one immunity unit with the maximum of toxine
solution that can be added without the production of any
physiological toxic effect, is physiologically neutral.

It is not easy to determine the point L0 beyond doubt, since
it is difficult to determine with absolute certainty whether a
toxine solution does or does not still exercise a slight action.
Hence, EHRLICH has intoduced his second value : L+ is that
quantity of the toxine solution under examination expressed in
toxic units (lethal doses) which, when added to an antitoxine
unit, is still sufficient to kill a guinea-pig of 250 grms. in four
or five days. This mixture then contains a lethal dose in the
free state. This point can be determined with ease and certainty.
The difference L0 - L+ is designated D by EHRLICH. In the
case of pure poisons it must obviously equal one lethal dose, but
in practice it is invariably greater — which is a point of great
significance in the investigation of the constitution of toxines
(vide infra).

These limitation values, then, apart from their practical use
in testing serum, have been of extreme value in the investigation
of the constitution of solutions of toxines such as diphtheria

(quoted by Ehrlich, Deutsch. med. Woch., 1891, 978) and
Aronson (Berl. klin. Woch., 1893, 625) inter alias had already indicated
the existence of such non-poisonous, though immunising, bacterial products.

38 TOXINES AND ANTITOXINES.

virus. For by their aid EHRLICH has succeeded in determining
the composition of toxine bouillon as regards the toxines and
toxoids. He has discovered the existence of extremely complex
relationships, and the final solution of the problems involved
will doubtless still furnish work for the next generation.

The value of these limitation values in the determination of
the amount of toxine and toxoids in solutions of poisons depends
upon the following facts : —

The antitoxine unit is a measure based upon a definite toxine
solution. The original cannot as yet be reproduced, and its
value has only been preserved by EHRLICH having kept an
originally standardised serum of 1,700 times the normal strength
under special precautions (vacuum, darkness, ice, dryness), and
using this serum after suitable dilution to standardise new
toxine solutions, which then in turn can be used to test fresh
sera. In the original normal toxine an antitoxine unit corre-
sponded to 100 lethal doses ; hence, to neutralise an immunity
unit 100 lethal doses of this normal toxine of BEHRING would be
necessary — i.e., L0 is exactly 100. But this numerical relationship
is not invariably necessary with all toxines, for there can also
be some in the case of which a large proportion of the immunity
unit would be fixed by non-poisonous haptines, toxoids, whose
quantity is expressed in toxic units — i.e., L0 then becomes less than
100. This is not the case, however, with most fresh toxines,
for with them L0 is really 100, or, in other words, these fresh
toxines have a constitution exactly like that of Behring's normal
toxine.

But the case is different when we deal with older solutions of
toxines, as we have already pointed out above. Here part of the
toxine has been converted into toxoids — i.e., L0 has become
smaller, and a smaller number of toxic units in an equal volume
are required to reach the stage of neutrality. On the other hand,
there are also toxines whose relative proportion in true toxine is
greater than in the case of Behring's normal toxine, so that we
are driven to the conclusion, which is difficult to follow, that
even fresh toxines may contain not only poisonous haptines, true
toxines, but also relatively non-poisonous substances, which
EHRLICH has designated toxones, to distinguish them from the
toxoids, which are only produced as secondary products. We
thus arrive at the following definition of L0 : —

Every solution of toxine, even in the fresh condition, contains,
in addition to the true toxine, non-poisonous haptines, toxones,
which usually stand in such relationship to the true toxines that,
in the case of most fresh toxine preparations, L0 is equal to 100.

RELATION OF TOXIN ES TO ANTITOXINES. 39

In certain toxine solutions, however, there may be relatively
more or less toxones, so that L0, even in the case of fresh pre-
parations, may sometimes be greater or smaller than 100. At
the same time, secondary derivatives, toxoids, formed from the
true toxine are produced in all toxine solutions on keeping, and
these invariably lower the value of L0.

We must reserve until afterwards a fuller discussion of the
mode in which toxines are converted into toxoids, and, above all,
of the quantitative relationships of this conversion. Here we
have next to deal with the significance of the limitation value
L+. Whilst the toxoids formed as secondary products have a
decided influence upon the amount of the L0 dose, they are
absolutely without influence upon the value L+ (and, therefore,
also upon D), as has been shown by EHRLICH.

Now there are, a priori, three conceivable kinds of toxoids : —
Firstly, those that possess a greater affinity than the toxine for
the antitoxine, which combine with it sooner, and are, under
certain conditions, capable of dissolving existing combinations
between toxine and antitoxine in their own favour. These are
the protoxoids.

A second category consists of the syntoxoids, which have the
same affinity as the toxine for the antitoxine, and thus have no
influence upon the combination of toxine with antitoxine, just as
their own combinations with the antitoxine are not affected by
the toxine. Lastly, there remain the epitoxoids, which have a
weaker affinity than the toxine for the antitoxine, and can be
liberated again by the toxine from their combination with the
antitoxine. Such epitoxoids occur, as EHRLICH was able to
demonstrate, not as secondary products, but are already present
in the fresh toxine solutions ; they are identical with the toxones
mentioned above.

This consideration leads us to the following points of view as
regards the influence of toxoids and toxones upon the value of L +
or D:—

The toxoids formed as secondary products — viz., pro- and
syntoxoids — have no influence at all upon L+, as can very easily
be demonstrated.

Assuming that we have a neutral mixture of antitoxine with
toxine and protoxoid, we can represent this condition of
equilibrium graphically by means of the equation —

90 Toxine- Antitoxine +10 Protoxoid- Antitoxine
= Physiological Neutrality (L0).

Additional quantities of the same poison are now added in order

40 TOXINES AND ANTITOXINES.

to find L+. This can cause no alteration in the compounds
already formed between toxine and antitoxine or protoxoid and
antitoxine, which might make them yet combine with fresh doses
of toxine added, and so cause them to disappear as regards L+ —
i.e., as soon as the amount of poison added to the neutral mixture
reaches one lethal dose, L+ is reached, as is demanded by theory
for pure poisons ; this may be expressed graphically by the
equation —

90 T-A + 10 P-A + 1 Toxine = L+.

Thus the presence oj protoxoid cannot cause L0 - L+ (D) to be
greater than one lethal dose.

The syntoxoids have just as little influence in raising the value
of D. They, too, are not influenced in their combination by the
addition of toxine, and L+ is reached by the addition of new
poison so soon as one lethal dose is added to L0.

Thus all the secondary toxoids are without influence upon D.

We find quite a different state of things when we examine the
behaviour of toxones in this respect.

Let us now leave out of consideration the toxoids, which have
no bearing upon this question, and represent a neutral mixture
of toxine and toxone with antitoxine (L0) by the following
equation : —

90 T-A + 10 Toxone-A = L0.

We then add fresh quantities of poison. If, in the first place,
we add a quantity containing exactly one toxic unit, we find
that L+ is by no means reached by that addition ; for the toxine
liberates a toxone unit from its combination with the antitoxine ;
and, instead of the free toxine added, we find a free toxone, as
represented in the following equation : —

90 T-A +10 Toxone-A + 1 Toxine
= 91 T-A + 9 Toxone-A + 1 Toxone (free) = L0.

And so the process continues until the whole of the toxones are
free. Then, and not till then, the next toxic unit produces L+.

100 T-A + 10 Toxones (free) = L0.

100 T-A + 10 Toxones + 1 Toxine = L+.

According to this equation, we should thus have to add, not
one, but eleven, toxic units to L0 before reaching L+, D being
thus equal to 11.

We see then that toxones have the property of raising the

RELATION OF TOXINES TO ANTITOXINES. 41

difference D above the amount "one" theoretically required by
pure toxines.

The relative amount of such toxones varies very considerably,
and hence D is also a very variable value. EHELICH found it to
vary from 1*7 to 28 toxic units in the case of 11 toxine solutions.

The value D, after reduction of the finally active one toxic unit
(D - 1), thus affords a measure of the quantity of toxones present
in toxine solutions. The fact that these variations in the value
of D occur even in fresh toxines, and do not alter as the poison
becomes old, when L0 decreases, shows that toxones are not
secondary decomposition products of the toxine, but are primary
bacterial products — non-poisonous haptines.

As regards other physiological characteristics, toxones are not absolutely
inactive. Their action can be studied in what EHRLICH has termed the
"differential zones" — i.e., between L0 and L+, where, according to his
view, free toxones are present. They produce slight toxic symptoms
(which we shall discuss later on), which are essentially different from the
effects of small non-lethal doses of toxines.

A special importance attaches itself to the discoveries of MADSEN, that it
is possible to produce antitoxic immunity with the mixed poisons in the
differential zones, which thus only contain toxones in the free state. We
shall discuss this more fully in its proper place (see Diphtheria Virus in the
special part).

EHRLICH next endeavoured, by means of elaborate, tedious,
and difficult processes, to obtain a clearer view of the quantita
tive relationships of the poisons and the numerical conditions of
their decomposition. These relationships proved to be of an
extremely complex character, and we do not wish to do more
than briefly touch upon them here.

In the first place, EHRLICH fixed the following formula for
any given poison : —

x Toxoid + y Toxine + z Toxone.

The value of y must be found by a physiological method (deter-
mination of the lethal dose), and is then represented by a. z, the
toxine value, is a function (F) of the value D - 1, which can
also be numerically expressed. It is designated j3 by EHRLICH.
Thus, the formula of every known solution of the poisons can be
expressed as follows : —

x Toxoid + a Toxine + F(j3) Toxone.

The formation of toxoids is illustrated by the following experi-
ment : —

42 TOXINES AND ANTITOXINES.

As already mentioned, the dose L? is equal to 100 in the case
of most toxines in the fresh condition. Thus, EHRLICH found
that one of his freshly-prepared diphtheria virus solutions was
of such composition that one I.E. (Immunity Unit) neutralised
0'31 c.c. of the poison. Hence, it followed that the lethal dose

0*31

of the poison solution had to be = O'OOSl c.c., and this

10U

was found to be actually the case. Thus, with this poison, too,
L0 = 100. After three-quarters of a year the poison showed
the same neutralisation value in c.c., but the single lethal dose had
risen to 0-009 c.c., so that L0 ivas equal to about 38 — i.e., that
33 toxic units (contained in 0*31 c.c.) corresponds to the dose
L0. The toxic value and L0 dose then remained constant.

Other poisons decompose in such a way that L0 is equal to 50,
while others, again, become finally constant with an L0 dose of
25, &c.

It would seem, then, that the toxines either decompose in
such a manner that half of them becomes inactive, or that they
undergo a tripartite change so as to consist of two parts of toxoid
to one of unaltered toxine.

EHRLICH'S chief endeavour has been to establish the absolute
value of an I.E. (Immunity Unit) — i.e., to determine to how
many saturation units the I. E. corresponds in poisons consisting
of toxines, toxones, and toxoids, or, to express it quite crudely,
how many haptophore groups correspond with the number in one
I.E. He is strongly inclined to fix their number at 200, and
he bases this conclusion on the following premises : — The value
L0 is usually equal to 100 in the case of fresh poisons, and these
subsequently decompose in such a manner that their L0 values
stand in a very simple proportion to 100. From this he con-
cludes that the absolute force of combination must also stand
in some very simple relationship to the number 100. Now,
no poison has yet been prepared, in spite of all attempts at
purification, with a value L0 exceeding 200, the highest observed
being 160 in the case of a poison which was certainly not per-
fectly pure (MADSEN). Hence, EHRLICH concludes that every
toxine bouillon must contain 200 saturation units — i.e., that
the I.E. is equivalent to 200 saturation units. An absolutely
pure toxine (without toxones) would thus, in the fresh condition
(i.e., without toxoids), show an L0 value of 200 and an L+ value
of 201.

In this case x 4 y + z, in the general formulae given above,
is equal to 200, and from this it is possible to calculate the
quantity of toxones by the aid of the values a and /3, where

RELATION OF TOXINES TO ANTITOXINES. 43

a represents the number of toxic units and /3 the value D — 1.
If z represent the amount of toxones, 200 - z represents the
amount of toxines and toxoids, on the assumption of 200 units
of combination. Thus, the formula for the value L0 of every
mixture of the poisons becomes

L0 = (200 - z) Toxine-Toxoid + z Toxone,
all of which enter into combination with the antitoxine. Hence,

in order to liberate one toxine it is necessary to add -^-r - ,

^ 200 — z

of which ^- — - a represents the proportion of toxine. And

this is, therefore, the amount of the poison solution, expressed
in toxic units, which must be added to (I.E.) + L0 to obtain a
mixture in which the whole of the toxones are free, so that
the addition of a single toxic unit produces L+ ; the amount

— — — - a is thus equivalent to D — 1 -- /?.

To find the amount of toxone in a poison bouillon, we have
thus an equation containing one unknown value :

whence is derived

a + j8

With the aid of this formula EHRLICH has calculated the
proportion of toxones in the poisons examined by him, and
has found that in the case of toxones, too, the values stand in
a very simple relationship towards 100 — e.g., 100, 50, 25, or 33,
66, &c. Owing to these simple relationships it is possible, by
determining the values of a and ]8, always on the basis of the
assumption of 200 saturation units, to reproduce the immunity
unit, which was previously only an empirical unit of measure-
ment, since it is possible by this means to determine the
proportion of toxine and toxone — i.e., in the case of fresh
poisons — which do not contain toxoids, to explain their whole
constitution. Most of the poisons in the fresh condition appear
to consist of 100 parts of toxine and 100 parts of toxone.

-z) = az-, 200/8 - 0s= 02; oz + |8 z = 200 |8; z (a + /3)=200j3;
200/3

44 TOXINES AND ANTITOXINES.

The conversion of toxine into toxoids usually takes place by merely
keeping the poisons. The L0 value then generally becomes constant after
some time, and new toxoids are no longer produced.1 This rule, however,
does not appear to be without exceptions ; at least MADSEN,S whose work
has completely confirmed EHKLICH'S results, has described a poison which
appears to decrease continuously in its toxic activity. He found in the
last determination that L0 had already sunk to 10 and L+ to 15, and hence
considers it possible that the bouillon might eventually become completely
non-toxic and contain nothing but toxoids.

A further interesting point is that the toxones, too, do not remain un-
changed, as has been proved by EHRLICH and also by MADSEN. In their
case it is the haptophore group that suffers, with the result that toxonoids
are formed. This is made manifest by the increase in the value L0, for
when in a mixture consisting of 100 parts of toxine : 100 parts of toxone,
part of the toxone that combines with the antitoxone changes as regards
its haptophore group, so that it can no longer be identified by reason of
its combining, it obviously follows that the proportion of toxine to the
200 saturation units is greater than 100. MADSEN regards the fact that
L0 values of, e.g., 133 (which he found in one case) can occur even with
fresh poisons, as evidence that the formation of toxonoids may probably
be going on spontaneously the whole time during the production of the
toxine.

Light, according to MADSEN, has an injurious effect upon both the
haptophore and the toxophore groups. He found, on exposing a poison
mixture to sunlight, that, although the toxicity diminished greatly, yet
the values L0 and L+ showed a simultaneous increase. Eventually the
specific character of the toxine completely disappeared, but the mixture
still remained poisonous. It caused animals to die of cachexy, but none
of the characteristic effects of diphtheria poison were found on section.
Thus toxic toxoids, if the term be permissible, are formed under the influence
of light.

Not content with all these tedious experiments to determine
the constitution of diphtheria virus by repeated determinations
of the L0 and L+ values, EHRLICH has obtained a further insight
into their nature by the aid of a second and still more ingenious
method.

If we grant the hypothesis that toxoids, toxines, and toxones
have a different degree of affinity for the antitoxine, the postulate
follows that they do not combine in equal proportions with a
given quantity of antitoxine. This is already shown to be
probable by a determination of the L+ value, but it can only
be proved by a direct quantitative estimation of the different

1 This condition appears usually to be reached in a year. Hence, in
testing sera, only such poisons are used as have been kept for a year, in
considerable quantities (4 to 5 litres of bouillon), under a thick layer of
toluene (Donitz, "Ber. iib. die Thatigkeit des kgl. Instituts f. serum-
forschung, &c.," reprint. Klin. Jahrb., vii., 1899).

2 Madsen, " Constitution du poison diphtherique, Ami. Past., xiii., 568,
1899.

RELATION OF TOXINES TO ANTITOXINES. 45

degrees of affinity. This has been done by EHRLICH and
subsequently by MADSEN (loc. cit.) in the following manner : —
If an immunity unit were added to 200 saturation units (i.e.,
100 toxic units in the case of fresh poisons), the mixture would
be physiologically absolutely neutral (L0). If now the quantity
of antitoxine added to the same solution of mixed poison were
to be reduced, a measured fraction of one immunity unit (200
units of combination) being added, the toxic characteristics
would gradually appear again, since there would be an excess
of free toxine. If the poison bouillon consisted of pure toxine,
one toxic unit would be liberated by the reduction of the
antitoxine to the extent of one unit of combination, two toxic
units if the reduction amounted to two units of combination,
and so on, until eventually the whole of the toxine had been
liberated. If the poison bouillon contained, in addition to the
toxine, only non-poisonous haptines of equal affinity, every
diminution to the extent of one unit of combination would
liberate a fraction of a toxic unit ; but this phenomenon would
occur quite regularly ; so that if 10 toxic units were set free by
a reduction to the extent of 20 units of combination, 50 would
be liberated if the units of combination were reduced by 100.
The result would be totally different, however, when substances
of different affinity were present, tinder such conditions the
first effect of the reduction in the units of combination would be
to liberate the haptines possessing the smallest affinity (toxones],
next those with medium affinity (toxines and syntoxoids), and,
last of all, those with the greatest degree of affinity (protoxoids).
Or, to express it differently, if a given solution of the mixed
poison were to be treated with increasing amounts of antitoxines,
the protoxoids would be saturated first, then the toxines, and
finally the toxones.

Now these theoretical requirements can be proved by experi-
ment. If we proceed from 200 : 200 no toxine activity appears
up to a certain stage, but only the quite distinct toxone activity
mentioned above (zone of free toxones). But if we go beyond
this limit the results are different according as the poison still
contained only toxines (fresh poisons), or also syntoxoids and
protoxoids in addition to toxines.

In the first more simple case each reduction by -^-^ immunity

ZOO

unit (one unit of combination) then liberates one lethal dose, and

this continues to the end. Usually this limit is at -^— - ; thus

ZOO

we have then —

46 TOXINES AND ANTITOXINES.

x c.c. of poison ) ^00
(100 lethal doses) / + 200 L'

x c.c. of poison + ^^. = toxone activity

x c.c. of poison + . = toxone activity

99
x c.c. of poison + ^Q = 1 toxine activity

(1 lethal dose)
x c.c. of poison + ™r = 30 lethal doses

x c.c. of poison + -^-^ = 90 lethal doses,

and so on.

The "Spectrum" (EHRLICH) of this simplest conceivable poison
may be thus represented —

O 10 ZO 3O 40 50 60 1O 8O OO WO J5O 2OO

Fig. 1. — "Spectrum" of a Fresh Poison.

But the conditions would certainly never be as simple as these.
For, in the first place, the toxines themselves are not uniform in
their affinity (to which point we shall have occcasion to return),
and, in the second place, they very soon form protoxoids, which
alter the curve. For instance, take the following series of
values : —

200
x poison + ^r = 0

180
x poison + ^— = tree toxone

160
x poison + --Q = free toxone

RELATION OF TOXINES TO ANTITOXINES.

47

159 ..

x poison + T^T = 1 tree toxine

100
x poison + ^r = 60 T. free

x poison +

50

= 100 T. free.

And next the non-poisonous protoxoids : —

59

x poison + -- n = 100 T. free
200

30
x poison + j^j

1

x poson +

200

= 100 T. free
= 100 T. free.

Protcocoid,

Toocone

O W ZO OO 4O 50 GO IOC 16O 1GO 20O

Fjg> 2.—" Spectrum " of a Poison in the Protoxoid State.

A further complication is introduced by the formation of
syntoxoids (hemitoxine formation).

Assuming that the toxine decomposes into equal parts of
toxine and syntoxoid, the saturation takes place as follows : —

200 .
x c.c. of poison + r = 0

x c.c. of poison +

160

158

Toxone

x c.c. of poison + -^ = 1 T. free
'ZOO

156
x c.c. of poison + 2QQ =21. free

x c.c. of poison + -^-^ = 30 T. free,

and so on.

48

TOXINES AND ANTITOXINES.

The " spectrum " (for the same poison), after the formation of
syntoxoid, could then be represented thus —

ProtojcoicL

Synboocouci

Toocorte

GO WO 16O ZOO

Fig. 3. — The same Poison in the Hemitoxine condition.

As a matter of fact, however, the diagrams are still much more complex.
It is not my intention to deal specially with the individual " spectra" that
EHRLICH and MADSEN have published. I only wish to give the principles
of these methods, and can now content myself with a simple statement of
the results that have been obtained in these analyses.

The quantitative decomposition of poisons on standing takes place in the
following manner : — At first only toxines and toxones are present. The
toxines consist of three distinct varieties, each of which has a different
affinity for antitoxine — viz. , proto-, c£ew£ero-,and trito -toxines — the last being
most nearly related to the toxones. Moreover, each of these varieties of
toxines is composed of two modifications — a- and p-toxine — in equal pro-
portions. The a- modification of all three toxines decomposes very rapidly,
losing its toxophore groups — formation of syntoxoids and development of
the above-mentioned hemitoxine condition.

Then begins, at an early stage, the destruction of the toxophore group of
the /3-tritotoxine, which, however, never continues to the complete replace-
ment of the toxine by toxoid. Small amounts of toxine invariably remain
in this zone — e.g., 3 : 7, 2 : 8, or 1 : 9 of toxoid — which can be recognised
in the "spectra" by the fact that toxic action still takes place here, that,
for example, in the case of the proportion 1 : 9, a reduction of the units of
combination by 10 liberates 1 lethal dose.

Not until a later period does the /3-prototoxine also disappear (develop-
ment of protoxoid zone). Thus, there finally remains unchanged only the
ft-deuterotoxine, in addition to a small quantity of /3-tritotoxine ; and at
this stage the decomposition usually stops, the poison remaining unaltered
for a long period in this form.

There appear, however, to be occasional exceptions to this rule. Both
EHRLICH and MADSEN have described "spectra" in which, even in the
case of very fresh poisons, it is possible to detect the development of the
protoxoid zone, although even the a-deuterotoxine is still intact, so that
there is still a zone of unaltered toxine present. MADSEN (loc. cit. ), how-
ever, considers that, taking into account the continued decrease in his pure
poison mentioned above, we must conclude that the /3-deuterotoxine, too,
is not of uniform structure, but that it possesses more readily decomposable
constituents which change into tritotoxine.

A very interesting confirmation of these extraordinarily
complicated relationships is afforded by the fact that MADSEN
(loc. cit.) frequently succeeded in again finding with sufficient

RELATION OF TOXINES TO ANTITOXINES. 49

accuracy in these the quantities of toxones calculated by means

200 B
or the rormula z - - -~ as given above.

a + j8

For example, in one instance the calculated toxone value was

170
z = 33-33, and he found that at ^AA all the animals remained

160
alive, while they died at ~r, so that the toxone value must thus

have been between 30 and 40.

It is also evident that a complete transformation of certain parts of the
tritotoxine most nearly allied to the toxones must increase the toxone
zone, for it is impossible to recognise tritotoxoids, even when they occur in
a state of purity, in a place where it is no longer possible to detect such
a small amount of toxine. This is due to their possessing only a slight
affinity similar to that of the toxones, and thus escaping detection in the
determination of the L+ value, and also to their being confused with the
toxones in making the diagrams. In this way we can explain an apparent
increase in the amount of toxones compared with their quantity in the
fresh poison, as has been found to be the case by MADSEN (loc. cit., p. 819),
while it is denied by EHRLICH.

It is not possible to fix an absolutely sharp limit between the
different zones. There appear, rather, to be reactions taking
place both between the toxones and toxines, and the toxines and
protoxoids, which we must regard as being capable of having
some slight influence upon the conditions of quantitative com-
bination, unless we agree with MADSEN in attributing such
uncertain reactions to the influence of varying conditions of
concentration and temperature.

Toxines and Antitoxines in the Light of Physical Chemistry.—
Modern physical chemistry, which has effected so profound
a change in our views of the nature of chemical reactions, has
of late also turned its attention to the study of physiological
chemical processes. I have already shown elsewhere 1 how great
an influence these conceptions and methods have had upon the
theory of ferments. Quite recently attempts have been made to
investigate also the processes that underlie the action of toxines
and antitoxines, by means of the kinetic modes of representation
of physical chemistry. And there is reason to hope that in this
way it may be found possible to express mathematically, in
numerical values, the conceptions that the genius of EHRLICH
has given to us. As yet, however, these attempts are still in
their earliest infancy.

1 Oppenheimer, Ferments and their Actions. English edition. Griffin &
Co., 1902.

4

50 TOXINES AND ANTJTOXINES.

The processes that take place during the combination of the
toxines with the receptors of the living cells can obviously never be
investigated by this method of examination. Hence, naturally",
it has only been employed to throw light upon those processes in
which we can watch their progress and result in vitro — e.g., in
haemolysis. Here again we have to thank EHRLICH that we
possess exact methods of measurement for these processes.

Thus, the first important research in this new field deals with
hcemolysis under the influence of simple blood poisons, on the one
hand, and of specific blood-solvent haptines, such as tetanolysine,
on the other.

ARRHENIUS and MADSEN 1 investigated the course of haemo-
lysis under the influence of ammonia, sodium hydroxide, and
tetanolysine.

The material used in the tests for haemolysis was a 2-5 per
cent, emulsion of the corpuscles of horse's blood washed
thoroughly free from serum, and suspended according to re-
quirement in physiological solutions of sodium chloride or cane
sugar. The amount of haemolysis was measured colorimetrically
by comparison with solutions of horse's blood, solutions of the
strength of 2-5 c.c. in 100 c.c. of distilled water being taken to
represent 100, and a colour scale prepared by corresponding
dilution.

For comparative determinations with constant amounts of
blood (invariably 10 c.c. of the above emulsion), the only suitable
interval is that relatively small one beneath whose lower limit
haemolysis just begins, and above whose upper limit it is
complete.

The first result ascertained by this method is that haemolysis
increases very rapidly with the rise in the amount of toxine
added,2 so that, as a rough approximation, it is proportional to
the square of concentration of the toxine.

Now the " concentration " of the solvent agent does not corre-
spond absolutely with the amount added. Thus, in the case of
ammonia and sodium hydroxide, a certain proportion enters into
combination with the blood-corpuscles, and does not contribute
to the concentration. With tetanolysine, however, this com-

1 Arrhenius and Madsen, "Anwendg. d. physik. Ch. auf d. Stud, der
Toxine u. Antitoxine," Z. physik. Ch., xliv., 1, 1903.

2 Arrhenius and Madsen give the collective name ' ' toxine " to these
blood-solvent agents, while they term the specific haptine "lysine." Un-
fortunately, this use of the term "toxine" is liable to create confusion,
since it is not in accordance with the present happy limitation of the word
to haptines.

RELATION OF TOXINES TO ANTITOXINES. 51

bination is so weak that the numbers obtained do not need
correction.

This explains why a lower limit is found in which hcemolysis is
entirely absent with ammonia and sodium hydroxide, but not with
tetanolysine.

Series of experiments were next instituted, in which variations
were made in the percentage of blood-corpuscles on the one hand,
and the amount of " toxine " (using the word in the above-men-
tioned sense) on the other hand. The simplest case is that
in which the toxine is present in such excess that complete
haemolysis immediately ensues. If, in this case, we plot a curve,
and place percentage amounts of the blood on the abscissae, and
the degrees of haemolysis on the ordinates, the latter will natur-
ally rise in a continuous line. But if the amount of toxine is
smaller, only the initial part of the curve is a straight line —
i.e., so long as the quantity of blood is still so small that the
haemolysis is complete.

But when the concentration of the blood rises still further,
there occurs after a short rise a point where the whole of the
" toxine " (ammonia later than sodium hydroxide) has entered
into combination, and the curve falls again. The shape of this
maximum is sometimes a sharp point and sometimes a horizontal
line, according to the difference in the stability of the compounds
with the blood-corpuscles. These details of the action of different
inorganic simple " toxines " do not concern us here.

It is, however, extremely important that this maximum is
almost entirely absent in the case of tetanolysine — at least it
cannot be identified with certainty. On the other hand, even
this faintly indicated maximum occurs with a much smaller con-
centration of the blood than in the case of ammonia, &c. From
this it follows that the combination of tetanolysine with the blood-
corpuscles is much weaker than in the case of inorganic agents,
but that, on the other hand, it takes place so slowly that part of
the substance that may subsequently be fixed still shows activity,
whereas the combinations with ammonia, &c., take place so
rapidly that only the actual excess of " toxine " takes part in the
action.

The proof of the fact that tetanolysine possesses so weak an
affinity for the receptors of the erythrocytes might be employed
as an argument against the general theory of specific combination.

We shall return to this question of weak combination in dis-
cussing the behaviour of tetanolysine towards antitetanolysine.

Speed of Reaction of Haemolysis. — The measurement of the
velocity of reaction is one of the most important means of

52 TOXINES AND ANTITOXINES.

obtaining a closer insight into the nature of chemical reactions.
ARRHENIUS and MADSEN measure the time within which the
haemolysis has reached a definite point.

For this purpose they cause an excess of " toxine " to act for a
fixed time upon equal amounts of blood (which are taken as
exactly 100). The haemolysis is interrupted by cooling the
mixture, which is then separated in a centrifugal machine, and
finally the extent of haemolysis is determined. Since the
quantity of blood-corpuscles dissolved in the unit of time is the
reciprocal of those remaining undissolved (100 - x), the follow-
ing equation is obtained : —

(1) J^ = K
whence

It has been shown that K is not a constant, but shows a rapid
increase in the course of the experiment. This is due to the fact
that the membranes of the blood-corpuscles at first offer resist-
ance to the action of the toxine, but that this resistance becomes
continually weaker with the destruction of the membrane.

At first no blood-corpuscles at all are attacked, and it is not
until the weakest membranes give way that haemolytic action is
apparent. This power of resistance thus leads to the necessity
of an "induction period" for haemolysis, the explanation of which
is evident in these cases.

Hence this method did not yield reliable results. Its authors,
therefore, tried whether twice the amount of " toxine " in half
the time had the same effect as half the amount in twice the
time. It was found that, after making the necessary corrections
for the alteration in volume, there was this approximate ratio,
viz., that the velocity of the reaction was proportional to the concen-
tration of the toxine. This held good in the case of ammonia,
sodium hydroxide, and tetanolysine.

The quantity of unaltered blood-corpuscles can be expressed
by the following equation (in which a represents the amount of
toxine), at all events with low proportions (i.e., where x is
small) : —

d x

after integration

*/~x = 2 K a t,

RELATION OF TOXINES TO ANTITOXINES. 53

which, expressed in words, means that the quantity of blood
haemolysed is not only proportional to the square of the time of
the reaction, but also to the square of the amount of toxine — a
fact already experimentally determined, as mentioned above.

The acceleration in the velocity of the reaction with each rise
in temperature of 10° C. amounts to 2'76 : 1 with ammonia and
sodium hydroxide, and 3 -04 : 1 with tetanolysine. The relative
velocity of the reaction with ammonia compared with sodium
hydroxide is 2 '24 : 1. It is thus absolutely independent of the
concentration of the OH-ions. Hence the OH-ions are not the
real active agents in the hcemolysis.

Neutral salts check the action of their corresponding bases.
The effect of the salt is approximately proportional to the cube
root of its quantity. Ammonium salts, in particular, have a
strong restrictive action. In the case of tetanolysine salts
(though in larger quantities) have a stimulating influence.
Normal blood serum and egg albumin check the action of
" toxines," and particularly that of tetanolysine. Thus we have
here a restriction of the effect of active substances by normal
blood serum, which plays so great a part in the action of ferments.
Here, too, without doubt, we have to deal with the occurrence of
normal receptors as anti-bodies.

Relations between Toxine and Antitoxine. — The experiments
described here were made by ARRHENIUS and MADSEN with
tetanolysine, the method being an extension of Ehrlich's method
of incomplete saturation described above. They treated a
constant amount of toxine (2 c.c. of a 2 per cent, solution of
tetanolysine) with increasing amounts of antitoxine (in a 0*0025
per cent, solution), and determined the toxicity of the mixtures
— i.e., the quantity that caused a definite amount of hydrolysis
when added to 10 c.c. of a 2-5 per cent, emulsion of horse's blood.
Allowance being made for the relative volumes, the toxicity can
be expressed by the equation —

1 10 + x
= x ' '~W~'

where x represents the observed amount of toxine and G the
poison strength.

It was found by experiment that the required amount, x,
constantly increased with the quantity of antitoxine added, and
that G showed a corresponding decrease. There is thus no
reason for the assumption that the " poison spectrum" (vide supra]
of tetanolysine assumes a step-shaped form. On the contrary, the

54 TOXINES AND ANTITOXINES.

ratios between toxine and antitoxine follow a definite curve
closely resembling that which represents the relation between the
decomposed part of a substance and the products of such decom-
position. There is thus a condition of equilibrium between the
free toxine and antitoxine, on the one hand, and the compound
of the two, on the other, so that all three components are present.
This deviates from EHRLICH'S fundamental view in just the same
way as we pointed out above in the case of the compound of
lysine with the cell. For EHRLICH assumes that, in the case of
diphtheria virus, there is a firm combination between toxine and
antitoxine, so that only the actual excess of one component is
active. We shall return to this point presently.

ARRHENIUS and MADSEN have strengthened the results of these observa-
tions by theoretical calculations of the values of G and x. G is obtained
from the equation : —

Free Toxine Free Antitoxine _ „ / Toxine - Antitoxine V2

Vol. Vol. \ Vol. ;•

The amount of free and combined toxine can be calculated in a complicated
way. The amount of toxine present in 1 c. c. of a 1 per cent, solution is
taken as the unit of measurement. Now, the original mixture of toxine
with 10 c.c. of blood (without antitoxine) contained 0'23 : 10'23 units
per c.c. Suppose that, in an experiment with antitoxine, it is necessary to
add x c. c. in order to obtain the same shade of colour — i. e. , to ensure that
the same amount of free lysine is present. Then the amount of toxine
that has combined with the antitoxine is equal to the difference between
the amounts of added and free toxine = x : (10 + x) - 0"23 : 10'23, and
obviously as great as the combined quantity of antitoxine. The quantity
of antitoxine added (n) is distributed over 4 c.c. of the lysine solution, and

N
hence each unit of toxine corresponds to — c.c. of antitoxine. If, now,

the ratio of antitoxine to toxine in c.c. be represented by p — i.e., 1 c.c. of
antitoxine solution saturating p c.c. of a 1 per cent, solution of lysine —
it follows that

— . --pr . p = the amount of antitoxine per c.c.

n 10 + x J

From this must be deducted the known quantity of antitoxine in order to
obtain the actual amount of free toxine: Hence, by interpolation of these
values into the first equation, there results : —

0-23
(ll-} RF23

K and p can be calculated approximately from the mean results of the
saturation experiments, and the values thus obtained in 12 determinations
being

K = 0-115 p = 14-55,

i.e., 1 c.c. of the 0'0025 per cent, antitoxine solution employed neutralises

( x 0-23\"| _ v( x _0'23\2
VlO+JB 10-22/J VlO + tf 10'23/ '

RELATION OF TOXINES TO ANTITOXINES. 55

14'55 c.c. of the arbitrary toxic unit, or this unit corresponds to 0'069 c.c.
of antitoxin.

The values of G and x calculated from this equation are in very close
agreement with the observed results.

When large quantities of antitoxine are used, a state of equi-
librium between the toxine and antitoxine is only reached very
slowly. At the same time the toxine becomes weaker, so that
x becomes greater.

An analogous result was obtained in an experiment in which
the toxine was replaced by ammonia, and the antitoxine by boric
acid. Here, too, the neutralising effect of the boric acid was
exerted to such an extent that the law held good : —

(Free Ammonia) (Free Boric Acid) = K (Combined Boric Acid)2.

It follows, then, from these experiments that the same laws
apply to tetanolysine and its saturation by its antitoxine as
govern the saturation of bases by means of weak acids. Parts of
the free components are invariably left uncombined.

This was now made the starting point for attacks on Ehrlich's
theory of "poison-spectra."

If we plot the curve of the neutralisation of ammonia by
means of boric acid, we find that the first amount added neutral-
ises 50 per cent, of ammonia, the next only an additional 16-7
per cent., and the third 8-3 per cent, more, and from this we
could draw the conclusion that the first aliquot portion of the
neutralising solution meets with a much more toxic ammonia
than the later portions — i.e., that ammonia consists of different
toxic parts, which enter into combination with different degrees
of affinity, these standing in simple relationship towards each
other. This is obviously opposed to the views of EHRLICH on
proto-, deutero-toxines, &c.

Hence, according to ARRHENIUS and MADSEN, we must not
employ as analogies for tetanolysine the conditions that govern
the neutralisation of strong bases and acids, as EHRLICH has
done in the case of diphtheria virus. In the case of strong acids
and bases the amount oifree components is extraordinarily small,
and, practically, we have only to take into account the compound
and the excess of one of the components. Measured by this
standard, then, the neutralisation curve of diphtheria virus
deviates, as EHRLICH has shown, very considerably from the
simple curve of the neutralisation curve of acids and bases, and
lends support to the view of the existence of different poisonous
constituents. But if we take as the standard the conditions of
equilibrium that occur in the case of weak affinity, conditions

56 TOXINES AND ANTITOXINES.

such as we find with tetanolysine, these apparent differences of
affinity can be explained by the law of mass action, without
assuming the existence of different toxine zones. The presence
of one toxine and one antitoxine, which by their reciprocal
neutralisation produce different conditions of equilibrium, and
can thus account for the numerical conditions of combination,
are quite sufficient to explain why the quantity of antitoxine
does not invariably correspond with the same amount of antitoxic
energy.

If these considerations applied to other poisons, as well as to
tetanolysine, they would modify a part, although not a very
weighty part, of EHRLICH'S views ; but, at the same time, they
would give us a further insight, based upon exact scientific
methods, into these extremely important processes. Almost
simultaneously with these attacks upon EHRLICH'S views other
voices were raised against practically the same parts of his theory.

Mention must be made, in particular, of the work of BoRDET,1
who, on theoretical grounds, has come to the conclusion that
there is a kind of equilibrium between toxine and antitoxine.

Certainly these conclusions are not based upon exact physico-
chemical measurements, but are purely speculative. According
to them, the relations between toxine and antitoxine either pre-
sent certain analogies with the processes of dyeing (insorption,
&c.), which BORDET does not more closely characterise — for
throughout he avoids proofs of identity — or complex compounds
are formed containing one toxine molecule with several anti-
toxine molecules. The antitoxine is distributed over the whole
quantity of toxine in such a way that one part of each toxine
molecule appears to be neutralised and deprived of its toxicity.
Regarded as a whole these speculations, which are not based
upon new facts, are just as difficult, if not more so, to work
into a theory as EHRLICH'S " spectra," which at least explain all
the facts on the assumption of a single although complicated
hypothesis.

Moreover, BORDET'S assumption of a "partially neutralised"
toxine is absolutely incapable of being put clearly without
falling back upon the old, now happily discarded, notion of the
" destruction " of the toxicity.

Hence, BORDET'S attack is surely not capable of overthrow-
ing EHRLICH'S theory of a plurality of poisons. Still less
can EHRLICH'S position be shaken by the different attacks of
GRUBER, although their object was nothing less than the over-

1 Bordet, " Sur le mode de 1'action des antitoxines sur les toxines,"
Ann. Past., xvii., 161, 1903 (reprint).

RELATION OF TOXINES TO ANTITOXINES. 57

throw of the side-chain theory as a whole. These have been so
completely answered and demolished by EHRLICH l that we can
refrain from discussing them individually here.

At first sight the objections of ARRHENIUS and MADSEN to
EHRLICH'S theory of the manifold nature of diphtheria virus
appear to be much more important, since they are based on the
unquestionable results of experiments. This is another instance,
however, that in dealing with the theory of toxines there is
nothing against which we must guard more than too hasty
generalisations.

EnRLiCH2 maintains his position throughout in his reply to
the criticisms of ARRHENIUS and MADSEN. He, of course, admits
at once the correctness of the experimental proofs, though only
in the case of tetanolysine, whose instability and slow combining
power he himself had already recognised years ago. He refers
to one experiment in which the antitoxine activity after two
hours was forty times as great as immediately after the mix-
ture had been made. But EHRLICH contends that the facts
established in the case of this unstable, slowly -combining
poison ought not to be transferred to the extremely active
diphtheria virus, which combines rapidly (within a few minutes)
with the antitoxine, and for which alone his "spectra" have
been made.

In his reply he once more repeats, very forcibly, the reasons
that led to the development of his conception of the complex
structure of diphtheria virus, and cites in support of it numerous
separate facts, some of which were previously unpublished.

It would take too long to deal with all his reasons again, since
we should have to repeat nearly the whole of what we have
said about toxoids and toxones. Hence we will only mention a
few of them : —

The conclusion that there are toxoids of different degrees of
affinity follows inevitably, from the fact that the toxicity shows
a gradual decrease while the amount of antitoxine required for
neutralisation remains constant. EHRLICH cites the simple
example of the neutralisation of two distinct alkaloids — e.g.,
quinine and codeine, which have a different affinity for hydro-
chloric acid — as evidence that we are here dealing with neutral-
isation limits, quite analogous to those which ARRHENIUS and
MADSEN regarded as states of equilibrium between weak acids

1 Ehrlich, "Toxin und Antitoxin," Munch, med. Woch., 1903, No. 33-4
(reprint).

2 Ehrlich, " Ueber d. Giftcoraponenten des Diphtherietoxins," BerL klin.
Woch., 1903, No. 35 (reprint).

58 TOXINES AND ANT1TOXINES.

and bases. If the alkaloids be mixed in the right proportion,
the curve can assume an exactly similar form.

Even ARRHENIUS and MADSEN will admit the existence of
protoxoids, but EHRLICH shows mathematically that there must
also be other toxoids.

He further succeeded in demonstrating, by calculation of the
L+ value of a particular poison, that in this case the neutralisation
of the toxine by the antitoxine exactly corresponded with the
neutralisation of a strong base by a strong acid — i.e., that its
course had to be represented by a straight line. It was also
possible to prove, in the case of this same poison, that the forma-
tion of tritotoxoids must have taken place in those parts of the
toxine possessing the weakest affinity.

Again, it can be easily demonstrated that there can be no
change in the degree of affinity during the conversion of toxine
into toxoid. But since, now, a poison that has been allowed to
stand does show differences in its affinity, it follows that these
must have already been present in the fresh poison, in the toxine
condition, and at the same time it is evident that this proves
that there is a plurality of poisons.

EHRLICH thus firmly maintains his opinion that even in fresh
diphtheria virus there are varieties of poison with different degrees
of affinity, which subsequently undergo a partial conversion into
toxoids with different degrees of affinity.

EHRLICH also firmly maintains the existence of toxones as
primary decomposition products of the activity of the diph-
theria bacillus, upon which doubt had been thrown by ARRHE-
NIUS and MADSEN. In particular, the fact of the existence of
a poison without toxones shows that we cannot here be dealing
with a " residue of unneutralised poison" but with individual
substances possessing only a slight affinity for antitoxine which
are usually present. For if it were a question of conditions
of equilibrium these would occur in the case of all poisons.
Apart from this important extreme case without toxones, addi-
tional evidence against this view is furnished by the enormous
variation in the relative amounts of toxones, which may range
from 0 to 300 per cent.

A further argument in favour of definite existence of toxones
is their frequent diminution, "formation of toxonoids," as well
as their absolutely different physiological action (vide supra).

EHRLICH concludes from his arguments that diphtheria virus
consists of at least three varieties of poison : —

RELATION OF TOXINES TO ANTITOXINES. 59

1. The toxine.

2. The toxone, which kills rabbits suddenly, and guinea-pigs

after symptoms of illness.

3. Toxonoids, causing illness in rabbits, but harmless to

guinea-pigs.

He also maintains that diphtheria toxine has so great affinity
for its antitoxine that the curve of neutralisation of the pure toxine
would form a straight line ; the deviations can only be explained
by the assumption that parts of it possess different degrees of
affinity. These fractions of varying affinity are present in the
original poison.

The theoretical deductions of EiSENBERG1 are very similar to
the conclusions drawn by ARRHENIUS and MADSEN from their
experimental results. He, too, is inclined to attribute the neu-
tralisation of the poison and antidote to the production of a state
of equilibrium, in which an excess of each of the two active
components is present, in addition to the firmly-combined neutral
compound. He claims, by means of this theory, to be able to
obviate the difficulties in the way of accepting the view that only
the compound and the excess of one component are present. His
argument, however, also rests upon conditions that occur with
other poisons than diphtheria, such as those of tetanus poison,
hsemolytic complements, and, above all, agglutinines.

It is extremely probable that in the case of these substances
loose combinations may occur with dissociated states of equili-
brium. But in the case of diphtheria virus the conditions of
combination appear to be the only important factors, which
EISENBERG, too, admits to be possible, and that the quantity
of components liberated is very slight, as it should be in the
case of firmly '-combined compounds. Thus EHRLICH'S arguments
are also a sufficient answer to these generalisations, which in fact
do not apply to diphtheria virus.

Thus we cannot predict beforehand what the conditions will
be in the case of other poisons. The question as to the influence
of the several derivatives on the quantitative relation between
toxine and antitoxine must be specially determined for each
separate poison. Probably a whole scale of affinities will be
found to exist, ranging from loosely combined, readily dissociated
compounds, such as appear to be formed by tetanolysine to
diphtheria virus. In this connection it may be mentioned that
conditions of equilibrium also appear to occur in the case of
ricine, in which when neutralisation is nearly complete free

1 M. Eisenberg, " D. Bindungsverh. zwischen Toxin und Antitoxin,"
Centralbl.f. Bakt., xxxiv., 259, 1903.

60 TOXINES AND ANTITOXINES.

toxine and free antitoxine are both present. DANYSZJ has
described this, and has based upon it absolutely untenable
speculations similar to those of BORDET. We shall return to
the facts when we are dealing with ricine. But we must never
forget, as EHRLICH is fully justified in repeatedly insisting, that
in these weak combinations a very important part is played by
the time of reciprocal contact as well as by the concentration.
Thus if too short a time be allowed for the reaction it is possible
to draw an erroneous conclusion of a too high degree of concen-
tration. Hence, when EISENBERG brings forward as a proof the
separation of snake poison from its antitoxine at higher tem-
peratures, it must not be forgotten that, according to MARTIN and
CHERRY, this separation, which involves the secondary breaking
down of the antitoxine, is only possible for a very short time
after the admixture.

Lastly, a caution must be given against over-estimation of the
influence of mass action as regards other haptines. Contrary to
the views of EISENBERG, who concludes that conditions of
unstable equilibrium are the rule with agglutinines and pre-
cipitines, v. DuNGERN2 shows that such conditions are quite
exceptional, and that, in general, the combination of the pre-
cipitine with the precipitable substance is complete and stable.
In explanation of the quantitative ratios he concludes that there
is a plurality of precipitines, which is quite analogous to the
views of EHRLICH.

Heat Manifestation of the Action of Toxine upon Antitoxine. —
With the aid of VAN T' HOFF'S formula

d log nat K W

dt ~ 1-99T2

ARRHENIUS and MADSEN were able to calculate from the altera-
tion in the dissociation constant, K, the amount of heat liberated
in the combination of 1 grm. molecule of toxine with 1 grm.
molecule of antitoxine.

This value is equal to 6,600 cals. (with a possible error of
600 cals.). The evolution of heat is almost half as great as that
liberated on neutralising a strong base with a strong acid.

Vigorous discussion as to the bearing of the law of mass action upon the
relations between the toxine and antitoxine still continues. The more

1 Danysz, " Contrib. & I'e'tude des propr. des melanges des toxines avec
leur antitox.," Ann. Past., xvi., 331, 1902.

2v. Dungern, " Bindungsverh. bei d. Pracipitinreaktion," Centralbl. f.
Sakt.f xxxiv., 355, 1903.

RELATION OF TOXINES AND ANTITOXINES. 61

thoroughly the question has been studied the more it has been found that
ARRHENIUS and MADSEN considerably over -rated the influence of mass
action.

It was shown by VON DUNGERN in the case of diphtheria toxine and by
HANS SACHS in that of tetanolysine that the fundamental condition
theoretically essential for the view of ARRHENIUS and MADSEN — viz., the
combination between the toxine and antitoxine should be reversible — was
lacking, and that on the contrary the combination gradually became firm
and irreversible. At the same time VON DUNGERN brought direct proof
of the plurality of diphtheria poison by showing that a mixture of anti-
toxine and toxine could be non-poisonous after a single addition of the
exactly sufficient quantity of toxine, but that when the same amount was
added in two portions with an interval of twenty-four hours between them
it was poisonous. In the latter case the antitoxine being present in excess
is distributed between the toxine and the toxone, and when the second
portion is added after twenty-four hours the combination between the
toxone and antitoxine has become so stable that the toxine can no longer
find sufficient free antitoxine.

The stability of the combination between toxine and antitoxine was
shown by WASSERMANN and BRUCK in another way. They found that
when .they injected a physiologically neutral mixture of tetanus toxine
and antitoxine, together with some adrenaline into an animal, the contrac-
tion of the vessels caused by the alkaloid led to a delay in the resorption
of the antitoxine, whereas the toxine which is resorted by the nerves
(vide Tetanus) was immediately taken up, and poisoning resulted. But
if they allowed the mixture to stand for two hours before the injection,
the toxine had combined so firmly with the antitoxine that no resorption
of the former took place and the animal remained well.

Lastly, it was found by MADSEN himself that it was only possible in the
case of absolutely fresh mixtures of diphtheria toxine and antitoxine to
effect a separation of the two components by means of their different rates
of diffusion through gelatin, and that after the lapse of even a short
time such separation could no longer be effected. This was also proved
indirectly by an experiment of MORGENROTH, who found that fresh
mixtures of diphtheria toxine and antitoxine were poisonous when injected
intravenously into rabbits, but that after some time they were inert.

However, even if the complicated relationships that exist in the case of
bacterial poisons cause the curve of saturation to take such a form that it
can be shown to have some resemblance, externally at least, to the curve
of equilibrium between substances of weak affinity, yet in the case _ of
Cobra venom the conditions are such that there is apparently a pure toxine
there. At all events, according to KYES, the curve of saturation between
the toxine and antitoxine here forms a straight line, like that formed
between a strong acid and a strong base.

The discussion has had the further very important result of raising the
fundamental question whether there is any justification at all for applying
the law of mass action to the reactions between colloidal substances.
NERNST and others have given a vigorous denial to this. This has led in
turn to a closer investigation of the properties of colloids, which has
already established many important ratios between the reactions of toxine
and antitoxine and those of the colloids, the consideration of which here
would lead too far.

L. Michaelis, " tJb. die Giiltigkeit des Massenwirkungsgesetzes bei der
Reaction zwischen T. u. A.," Biochem. CentralbL, iii. [1], 1904.

62 TOXINES AND ANTITOXINES.

v. Dungern, "Beitr. z. Kenntn. der Bindungsverh. bei der Verein. v.

Di.-Gift u. A.," Deutsch. med. Woch., 1904, Nos. 8-9.
Wassermann and Brack, " t)b. d. Wirkg. der A.," Deutsch. med. Woch.,

1904, No 21.

sen, "Constit. der poison dipht.," Centralbl. f. Bakt., xxxiv., 630,
^ 1903.
Morgenroth, " t)b. die Bindung von Di.-T. u. A.," Berl. klin. Woch.,

1904, 20; Zeit.fiirHyg., xlviii., 177, 1904.

H. Sachs, " Konstitution des Tetanusgiftes," Berl. klin. Woch., 1904, 16.
Kyes, "Kobragift u. Antitoxin," Berl. klin. Woch., 1904, 19.
Nernst, "Ueb. die Anwendbarkeit d. Gesetze d. chem. Gleichgewicht auf

Gemische von T. u. A.," Z. f. Electrochemie, x., 1904.
H. Aron, "Ueb. organische Kolloide, Biochem. Centralbl., iii. [15 and 16],

1905.

Endotoxines and Bacterial Proteins. — While the production
and mode of action of true toxines, such as are formed by the
bacilli of diphtheria and tetanus are fairly well known, the case
is essentially different and more complicated with a large number
of pathogenic organisms, of which we may take the bacteria of
cholera and of typhus and B. pyocyaneus as the chief representa-
tives. If a cholera cultivation of only a few days' growth be
filtered through a bacterial filter, the filtrate is only toxic to
a slight extent. Several c.c. are necessary to kill an animal
by intraperitoneal injection, and even in these quantities the
filtrates are not deadly to all animals. But if we take the
residue of the filtration — i.e., the filtered-off bacterial cells — and
destroy these by means of weak disinfecting agents, such as
chloroform, it will be found that these dead cells are highly
toxic. A few milligrammes of them are sufficient when injected
into the peritoneum to kill an animal almost instantaneously
with severe symptoms of collapse. In this case, then, the
conditions are the reverse of those that obtained with diphtheria
bacilli, since at first but little of the poison passes into solution,
whereas the dead cells of the bacteria are extremely toxic. If,
instead of fresh bouillon cultivations, old ones that have stood
for several weeks in an incubating oven are taken, it will be
found that there is a considerable increase in the toxicity of the
germ-free filtrates. Much smaller doses are sufficient to kill
the animals used in the experiment. Yet even under these
conditions the toxicity of these filtrates never attains the same
degree, as is found with diphtheria and tetanus poison, of which
even fractions of a milligramme may cause fatal results.

The explanation of these experimental results presents no diffi-
culty. Obviously cholera bacilli, &c., secrete the greatest part of
their poison within their vital substance. This part is only
liberated when the bacteria undergo a process of destruction,

ENDOTOXINES AND PROTEINS. 63

such as occurs when they are dissolved by the fluids in the bodies
of animals, or such as take place spontaneously in old cultiva-
tions, where part of the cells of the bacteria are brought into
solution by the alkaline and other products present in such culti-
vations ; and to this extraction process must be attributed the
fact that the nitrates from old cultivations are much more toxic
than those from fresh cultivations.

If, now, we ask what position these poisons occupy as regards
our definition : whether they are true toxines, against whose
action the organism forms antitoxines, we have the following
data to go upon : — Attempts to prepare a true antitoxine to the
poisons contained in the cells of bacteria, the endotoxines, have as
yet been unsuccessful. Hence, in the absence of further know-
ledge, we must assign an exceptional position to these poisons.
Their sole distinguishing characteristic is their extreme toxic
nature in experiments on animals. But as yet no convincing
proof has been brought that they belong to the true toxines, and
possess distinct haptophore and toxophore groups.

The case is different with the poisons that pass into the filtrate
on filtration. A true antitoxine serum for these has been pre-
pared— e.g., by RANSOM and by Roux and METSCHNIKOFF for
cholera, and by A. WASSERMANN for the poison of B. pyocyaneus
(q.v.). This was done in the well-known usual manner, by pre-
viously treating the animals with increasing doses of the
poisonous filtrates. By this means sera were obtained which
were capable of neutralising with certainty many times the
amount of the lethal dose of the poisonous filtrates.

Hence, according to the results of all experiments that have
been made up to the present, the conditions in the case of these
species of bacteria are such that the chief part of the poisonous
substance adheres firmly to the cells of the micro-organisms, and
does not pass into solution. Such substances are termed the
endotoxines, and are comparable with the endoenzymes of yeast
and of bacteria themselves.

Traces of a true toxine which passes into the filtrate also occur.

The further question now arises, whether we are justified in
concluding from these experiments that the production of poison
follows the same course under natural conditions — i.e., in the
organism.

It seems fairly certain that the answer is in the negative.

On the contrary, it is extremely probable that the slight traces
of poison which we find in these cultivations, and which, as we
have seen, increase somewhat in quantity on further extraction
of old cultivations, are not those of the primary poison of the

64 TOXINES AND ANTITOX1NES.

micro-organisms — that which we see exerting its activity in the
pathology of these infectious diseases in man. An extraction
process, such as occurs spontaneously in old cultivations, is by no
means one that has little effect upon the constitution of these
unstable bodies. In such old cultivations great changes of re-
action from acid to strongly alkaline occur abruptly ; ammonium
compounds and other chemical substances are produced, which
we know tend to change and destroy bacterial poisons. Hence
we must assume that even these traces of cholera virus, &c., that
pass into solution, and to which an antitoxine can be prepared, do
not represent the primary poisons of these micro-organisms, the
poisons that they undoubtedly produce in the human organism,
but are rather a secondary and more stable modification ; and
this conclusion is based on the fact that while, as WASSERMANN
found in the case of the poison of B. pyocyaneus, we are certainly
able to prepare an antitoxine to this dissolved poison, yet this
antitoxine behaves quite differently to the antitoxines in diph-
theria. For the latter neutralise the corresponding quantities
of toxine in any given multiples of the dose, provided that their
own quantities are increased in the same proportion. Thus, if
10 doses of diphtheria antitoxine neutralise 10 doses of toxine,
1,000 doses will neutralise 1,000 doses of toxine. But in the
case of pyocyaneus poison this " law of multiples " only holds
good within very narrow limits — up to about eight or ten times
the lethal dose. Beyond that there is no neutralisation, and
animals die in spite of large doses of antitoxine.

We must, therefore, conclude that it is altogether doubtful
whether we have ever had the primary true toxine of cholera,
&c., in our hands when using the culture media at our present
disposal. It may be a question of suitable nutrient media, and
further systematic research may be required before we get nearer
to this important goal. The great influence of a suitable culture
fluid upon the production of the true poison in artificial nutrient
media is shown by the case of diphtheria virus, of which in the
earlier experiments of Boux and YERSIN from 30 to 36 c.c. were
required to kill an animal with the typical acute symptoms,
whereas a close systematic study of culture media and the choice
of suitable cultivations have had the result of reducing the
amount now required to 1 to 2 mgrms. Hence we must regard
the question of cholera virus and similar poisons as still unsettled
in many respects.

At the same time, we must not omit to call attention to the
fact that so experienced a worker in practical research into the
nature of cholera infection as B. PFEIFFER holds the view that

ENDOTOXINES AND PROTEINS. 65

the striking symptoms of poisoning that appear in the spontaneous
infection of man with cholera are also due to the absorption of
poisons, endotoxines, set free through the solution of the cholera
vibrio. In his opinion, endotoxines are thus the main effective
poison in cholera and other infectious diseases, such as typhus,
&c., that behave in an analogous manner. In support of this
view, that in these infectious diseases we have only to deal with
the bacteria as such, and the poison contained in their cells, and
not, as in the case of diphtheria, &c., with a soluble poison that
can be separated from the cells, but of whose presence in the
human subject we cannot speak with certainty, as pointed out
above ; in support of this view we have the phenomena that
occur in the disappearance of these diseases and in immunisation
against these bacteria, and the substances that are then found in
the serum. Thus we have seen that in cholera, &c., only bacteri-
cidal substances are present; but such exclusively bactericidal
substances only occur, as WASSERMANN was able to show in the
case of B. pyocyaneus, when there has been absorption of consti-
tuents of the cell-substance of the bacteria, while toxines invariably
cause a simultaneous formation in the serum of antitoxic sub-
stances, and substances possessing specific bactericidal powers.

Bacterial Proteins. — When bacteria that produce soluble
poisons are freed as completely as possible from those poisons,
there will still remain substances that belong to the cell
material. These substances, too, have a physiological action,
producing inflammation, aseptic abscesses, and necroses at the
point of application, as well as slight general symptoms, such as
fever, faintness, headache, &c.

The same effects are also produced by the albuminous sub-
stances prepared by chemical methods from the cells of the
bacteria, and termed bacterial proteins, as first proposed by
BUCHNER. They are obtained by different methods, of which
those chiefly employed are extraction with superheated water in
autoclaves, simple boiling with water, and extraction with dilute
alkalies. To these have recently been added those methods, due
to KOCH and BUCHNER, in which the bacteria are first triturated
either in the moist or dry condition, with, in some cases, the aid
of hydraulic pressure, in order to obtain their contents.

In this way a long series of bacterial proteins has been
obtained, which, although differing in some particulars, produce
practically similar effects. We shall frequently meet with them
in the special part, where, too, will be found references to the
most important researches on this subject.

It is unnecessary to describe these proteins in detail here, for

5

66 TOXINES AND ANTITOXINES.

it has been proved beyond doubt by the researches of HOMER,
BUCHNER, SCHATTENFROH, KLEMPERER,1 and many others, that
at least the albuminous substances isolated by long-continued
extraction from unruptured micro-organisms are absolutely
without specific action, and thus can be neglected in discussing
the causes of disease sui generis.

But this only applies to the albuminous substances isolated
from the cells in a state of ideal purity. To separate them,
however, in such a state is only possible in the very rare cases
when the bacteria produce only free soluble specific poisons,
from which their cells can be completely separated, as has
been done by KOSSEL in the case of diphtheria bacilli. There
then remain behind proteins devoid of specific activity, exactly
similar to those that can be obtained from the most innocuous
bacteria, and, like other foreign albuminous substances, producing
sterile abscesses, &c.

But usually it is impossible to effect a radical separation of
these proteins, in the narrower sense of the word, from the
poisons. The protein preparations from most bacteria are still
contaminated with residual particles of the specific poisonous
substances or their secondary products, especially endotoxines
and their derivatives, so that even these protein preparations
still produce characteristic symptoms of poisoning, as is the
case with cholera, typhus, and tuberculosis (see Special Part).
Here, then, the action of the pure protein cannot be demon-
strated, but only theoretically inferred.

The actions of these poisons must be kept quite distinct from
the immunising processes that are brought about by the cell
materials of the bacteria, either in the uninjured cells or in
chemical preparations — the problems of bactericidal immunity
which have been completely elucidated by the researches of
PFEIPFER and WASSERMANN as regards cholera, of PFEIFFER
and KOLLE in the case of typhus, and of KOCH in the case of
tuberculosis.

These processes have absolutely no connection with the toxic
action of the cell proteins. Here we have to deal with the
introduction of suitable receptors, which rouse into activity
bactericidal protective forces — ly sines, precipitines, and agglu-
tinines ; with processes that depend upon the complex stereo-
chemical configuration of the protein molecule.

At present we can only assume with certainty the presence
of such receptors in unaltered bacterial cells, which, like the

For the bibliography see Klemperer, "Die Beziehg. v
Immunitat u. Heilung," Z.f. klin. Med., xx., 165, 1892.

SUMMARY. 67

cholera vibrio, pneumococcus, &c., bring about those destructive
processes in toto. On the other hand, it is fairly safe to affirm
that vigorous extraction, involving the formation of chemical
protein preparations, usually changes that sensitive atomic
grouping to such an extent that injection is followed by no
bactericidal reaction, or only a very faint one, but rather that
these albuminous substances produce only the same reactions
as every albuminous substance that is foreign to the system —
i.e., the formation of precipitines1 which are certainly very
closely allied to the agglutinines. Yet it is very probable
that, in the case of somewhat more scantily represented
examples, such as Koch's tuberculin (q.v.) and Buchner's plas-
mines — e.g., of the cholera vibrio and tubercle bacillus — specific
receptors are present, so that these preparations bring about
bactericidal immunising processes.

SUMMARY.

1. A group of bacteria produces true toxines in the form of
free secretions. After extraction of these soluble poisons there
remains a residue of pure non-specific bacterial proteins. Type :
Diphtheria.

2. Another large group appears to form only endotoxines :
true toxines which are more or less firmly retained by the
living cell, so that they are only secreted to a very small
extent in an unaltered condition, and possibly not at all outside
the body. When the cells die the toxines are partly liberated
and partly retained, or are converted into secondary poisonous
modifications no longer possessing the characteristics of toxines.
Thus, in the case of this group the dead cells cannot be com-
pletely freed from other poisons ; we cannot regard the results
as being produced by pure protein alone. With this reserva-
tion, however, it is possible to detect the activity of proteins.
Type : Cholera, typhus, pneumococcus.

3. A third group possibly forms no true toxines, even within
the plasma. The cell plasma contains poisons of another kind,
which obscure the effects of the action of the protein. Type :
Anthrax, tuberculosis. It is possible that as our knowledge
increases groups 2 and 3 may be united.

4. The pyogenic action of their proteins is common to all
bacteria, and depends, in the main, upon their effects as albu-

1 For further particulars about precipitines see Michaelis and Oppen-
heimer, "Immunitat gegen Eiweisstoffe," Engelmanns Arch., 1902, suppl.
vol. .

68 TOXINES AND ANTITOXINES.

minous substances foreign to the system, — effects that can also
be produced in an exactly similar manner by foreign albuminous
substances of non-bacterial origin.

The formation of the specific precipitation ferments, the pre-
cipitines, after the introduction of any foreign albuminous
substance into an organism, is a proof that every such foreign
proteid is an injurious intruder which the organism endeavours
to overcome. Just as, according to EHRLICH'S views, all food
substances must find receptors in order to be assimilated, so
must those receptors be made adaptable to foreign proteins
when abnormally introduced — i.e., by subcutaneous or intra-
venous injection — so as to seize upon them and render them
innocuous.

In this process general reactions, fever, &c., are of frequent
occurrence. And here, too, we have a means of elucidating
the action of bacterial proteins apart from that of any specific
toxic impurities that may be present.

SPECIAL PART.

I.— THE TRUE TOXINES.

DIPHTHERIA TOXINE.

DIPHTHERIA toxine is the most important of all the bacterial
poisons, both as regards its theoretical importance and especially
on account of its relations towards artificial immunity and
serurn- therapy. It represents for us the exact fundamental
type of the true toxine; it has been used in most of the
investigations that have thrown light on the mode of action
of toxines, their relations towards disease, and the formation
of antitoxines. Diphtheria toxine is strictly a specific poison,
producing in animals almost exactly the same symptoms of
disease as are produced by infection with living diphtheria
bacilli. The knowledge of the fact that the diphtheria bacilli
themselves are not found distributed throughout the body, but
only in the false membranes or point of inoculation long ago
suggested the notion that the general symptoms were caused by
soluble poison. Even in his first work, LOFFLER 1 indicated the
existence of such a specific poison, and subsequently he isolated
it by precipitation with alcohol. Then R-oux and YERSIN 2
succeeded in establishing the fact beyond dispute.

Roux and YERSIN found that a cultivation of the diphtheria
bacillus in calf's bouillon of seven days' growth when filtered
through porcelain and proved to be completely clear and sterile,
produced typical symptoms of poisoning, especially when injected
into the peritoneum. The toxicity increased with the age of

, "Unters. tib. d. Bedeutg. der Mikroorg. f. d. Entstehg. d.
Diphth.," Mitt. Kais. Ges.-Amt.,iL, 1884. Id., "Der gegenw. Stand der
Frage nach der Entsteh. d. Diphth.," Deutsch. med. Wocli., 1890, 81.

2 Roux and Yersin, "Contribution & l'4tude de la diphtheria," Ann.
Past., iii., 273, 1889; iv., 385, 1890.

70 TOXINES AND ANTITOXINES.

the cultivation. A culture forty-two days old killed a rabbit in
five to six hours with toxic symptoms analogous to those of a
severe attack of diphtheria, whilst six days was necessary with
an equal dose (35 c.c.) of the same cultivation of seven days'
growth. The poison was also characterised by the same pheno-
mena and the same degree of infectiousness as in the case of
inoculation with living cultures. The conclusions of Roux and
YERSIN were fully confirmed by KOLISKO and PALTAUF/ who
produced the symptoms of poisoning by means of the filtrates of
a bouillon cultivation of fourteen days' growth.

Roux had already noticed that the poison was destroyed by
heat, and concluded that it was akin to the enzymes. As such
it was, in his opinion, a secretion product of the diphtheria
bacilli excreted by them into the surrounding media. Yet this
view was apparently quite out of keeping with the fact that
young and vitally active cultivations produced relatively little
toxine, whilst older cultivations gave richer yields.

Hence, GAMALEIA2 concluded that diphtheria virus was not
a secretion product of the bacilli but a constituent of their cell
contents, which was not discharged by healthy bacilli, and that
it was only when the cultivation became old and many of the
bacilli died that the poison was extracted from their decomposed
cells. This view, however, has frequently been convincingly
opposed. H. KossEL,3 in particular, proved that the toxine was
a secretion product and not a decomposition product, by the fact
that, on the one hand, he observed a very plentiful formation
of poison, even in quite young cultivations (two days old) when
grown by suitable methods, and that these began to grow weaker
after as little as five days, whilst, on the other hand, he showed
that but little toxine was present in the cells of the bacilli
themselves.

He cultivated diphtheria bacilli upon as large a surface as possible,
making the inoculations from the surface skin of a cultivation of one day's
growth. After some days the bacilli formed a tough skin from which he
poured off the bouillon. The cells of the bacilli were then repeatedly
washed with distilled water and subjected to centrifugal force until the
washings no longer gave the biuret reaction, after which they were
extracted with water rendered faintly alkaline. This extract was only

1Kolisko and Paltauf, "Zum Wesen des Croup und der Diphtheric,"
Wien. klin. Woch., 1889, No. 2.

2Gamaleia, "Les poisons bacte"riens, " Arch, de M6d. Exper., 1892.

3H. Kossel, "Zur Kenntnis d. Diphtheriegiftes," CentralU. f. Bakt.,
xix., 977, 1898.

DIPHTHERIA TOXINE. 71

slightly poisonous. AKONSON,1 however, by trituration of the bacilli cells
and extraction with ether-alcohol (4:1) followed by extraction with a
O'l per cent, solution of ethylene diamine, obtained poisonous extracts
from which he could precipitate the poison by means of alcohol. Again,
according to BRIEGER and BoER,2 the extracted cells have certainly a
considerable degree of toxicity, but this manifests itself in quite a different
manner to the toxine, producing no immunity, &c.

Diphtheria toxine is thus not a constituent of the cell contents,
but a metabolic product. As regards the preparation of diphtheria
poison, two things must be sharply differentiated — viz., on the
one hand, the selection of methods to obtain the largest possible
yield of poison, and, on the other hand, the experiments that
have been made to isolate specific substances from the toxic
cultivations and to investigate their nature, of course with the
practical object also in view of obtaining as pure a poison as
possible in a dry condition for the purposes of immunisation and
serum-therapy. In preparing liquids rich in poison the methods
must aim at the two cardinal points of, first, obtaining a solution
containing as much poison as possible, and secondly, obtaining
this poison as completely as possible in germ-free bouillon, either
by sterilisation or separation from the bacilli.

As regards the production of the toxine, the choice of the
nutrient medium is of primary importance. Koux and YERSIN
(loc. cit.) and LOFPLER (loc. cit.) cultivated their bacilli on an
ordinary meat broth, and so naturally only obtained relatively
weak solutions of the poison, especially in the early stages of
growth. Subsequently, the addition of 2 per cent, of peptone
became the general practice. According to H. KOSSEL (loc. cit.),
the nature of this peptone is not a matter of indifference. He
had employed with great advantage Dr. Aschmann's peptone
from the Luxemburg Hygienic Laboratory. Chapoteaut's
peptone has also been found an excellent material. It is
also usual to add 0-5 per cent, of ordinary salt, von DuNGERN3
has found it advantageous to add ascitic fluid or to cultivate the
bacilli on the fluid itself.

GUINOCHET 4 asserts that diphtheria toxine is produced even
in culture media devoid of albuminous substances. He was

1 Aronson, "Zur Biologie und Chemie der Diphtheriegiftes," Arch. f.
Kinderheilkunde, xxx., 23, 1900 (reprint).

2Brieger and Boer, "Ueb. d. Toxine d. Diphtheric, &c.," Deutsch. med.
Woch., 1896, 783.

3 von Dungern, " Steigerung d. Giftproduktion d. Di.-B.," Centralbl. f.
Bakt., xix., 137, 1896.

4 Guinochet, "Contribution a l'e"tude de la toxine des bacilles de la
diphtheric," Arch, de Med. Exper., 1892, 487.

1T2 TOXINES AND ANTITOXINS^

successful in producing it in urine rendered faintly alkaline, and
his results were confirmed by USCHINSKY ; to this we shall
return shortly.

The choice of the culture medium is of great importance,
particularly as regards the reaction. In ordinary bouillon the
reaction is at first faintly alkaline, then becomes faintly acid, and
then, finally, alkaline again. During the period of the acid
reaction the production of poison is considerably checked or
entirely suspended. Hence, means were sought to restrain this
injurious acidification. Even the above-mentioned addition of
2 per cent, of peptone effects some improvement, but, by itself, is
insufficient.

PARK and WILLIAMS l adopted the simple device of increasing
the alkalinity of the 2 to 4 per cent, peptone bouillon by the
addition of a measured quantity of soda solution. They exactly
neutralised their fresh bouillon (with turmeric paper as indi-
cator), and then added to each litre 7 c.c. of normal soda solution,
the mixture still showing an acid reaction with phenol-phthalein.
They found, in agreement with SpRONCK2 and VAN TuiiENHOUT,3
that the acidification was due to the presence of glucose and
glycerin in flesh that was too fresh, and therefore concluded that
horseflesh (which is rich in glycogen) should not be used ; whilst
SPRONCK advocated the employment of slightly decomposed flesh.
By the artificial addition of glucose he was able to restrict the
production of the toxine. This was in accord with the results
obtained by BLUMENTHAL,4 who also found that no toxine was
formed in culture media containing sugar when the proportion
of the latter reached more than 1 per cent. Moreover, he could
obtain no toxine from solutions of pure egg albumin or pure
peptone. On the other hand, SMITH5 regards the presence of
a slight amount of glucose (0'2 per cent.) as essential, and con-
siders that it is only necessary to prevent the injurious acidifica-
tion, in which respect he supports KUETE,G who effects this

1 Park and Williams, "The production of diphtheria toxine," J. Exper.
Med., i., 164, 1896.

2 Spronck, "Sur les conditions, dont depend la production d. poisons
dans les cultures diphte"riques," Ann. Past., ix., 759, 1895.

3 Van Turenhout, "Over de bereiding van diphtheriegif," Utrecht,
Abst., Centralbl.f. BaU., xviii., 295, 1895.

4 Blumenthal, " Ueb. d. moglichk. d. Bildg. von Diphtherietoxin aus
Eiweisskorpern und auf Zucker enthaltenden Nahrboden," Deutsch. med.
Woch., 1897, No. 24.

5 Smith, "The Relation of Dextrose to the Production of Toxine," /. of
Exper. Med., iv., 373, 1899.

6 Ruete, "Ueb. Herstellg. d. Di.-Heilserums," MiiJich. med. Woch.,
1897, 213.

by
N

DIPHTHERIA tOXINE. 73

the addition of fragments of marble to the cultivation.

oLLE,1 differing from SPRONCK, has recommended the use of
quite fresh flesh. MARTIN2 has obtained successful results by
the use of a mixture in equal parts of extract of beef, or, better,
veal, and of extract of pig's stomach.

He gives the following recipes : —

(i.) Extract of Pig's Stomach. — Five pigs' stomachs are finely minced or
crushed, and the whole (mucosa and muscidaris) mixed with 10 grms. of
pure hydrochloric acid (20 per cent.) and 1,000 grms. of water to each 200
grms., and allowed to stand for twenty-four hours at 50° C. The mixture
is then boiled and filtered through loosely-packed cotton wool, and the
filtrate mixed with 0'2 per cent, of acetic acid, neutralised while still hot,
filtered through paper, and heated to 120° C. in an autoclave. The last
process can also be replaced by repeated heating to 100° C., careful
removal of the albuminous scum and fat, and, finally, clarification by the
addition of calcium chloride and sodium phosphate (formation of a pre-
cipitate of calcium phosphate which simultaneously carries down the
albuminous turbidity).

(ii.) Meat Extract. — Perfectly fresh veal is allowed to stand for twenty
hours at 35° C., after which it is treated with 1,000 grms. of water for each
500 grms. of material, and again left for twenty hours at the same tem-
perature. It is then expressed, and the resulting liquid mixed with
5 grms. of sodium chloride and 20 grms. of peptone, neutralised, and, after
the addition of 7 c.c. of normal soda solution, sterilised at 120° C.

A mixture of equal parts of the two extracts, heated to 70° C. and
sterilised by filtration through Chamberland filters, forms an excellent
medium which does not become acid and in which very active toxines are
speedily formed, O'l c.c. being required for a lethal dose after thirty hours,
and 0'002 c.c. after five to seven days.

MADSEN's3 method is to leave finely-minced lean veal, two to three days'
old, in contact with twice its weight of water for fifteen hours, after which
it is boiled and strained. The liquid is then mixed with 1 per cent, of
Witte's peptone and 1 per cent, of sodium chloride, rendered faintly
alkaline, boiled for forty-five minutes, filtered, and the filtrate transferred
to Erlenmeyer flasks holding a litre, and sterilised for fifteen minutes at
120° C. in autoclaves. He determines the reaction with litmus paper, and
also the "titer" of the bouillon — i.e., the amount of normal soda solution
required to give a distinct pink coloration with phenol-phthalein as indi-
cator. The ratio of the two measurements is such that neutral solutions
showing a "titer" of over 20 c.c. are unmistakably acid to litmus, while
with 16 c.c. they are amphoteric, and below 10 c.c. are alkaline. Accord-
ing to his experience the age of the flesh has no material influence. The
addition of calcium carbonate recommended by SPRONCK (loc. cit. ) and VAN
TURENHOUT (loc. cit.) keeps the reaction permanently alkaline, but at the
same time does not materially promote the formation of toxine.

1 Nicolle, "Preparation de la toxine diphte'rique," Ann. Past., x., 333,
1896.

>2 Martin, "Production de la toxine diphte'r," ibid., xii., 26, 1898.

3 Madsen, "Zur Biologic des Diphtheric bacillus," Zeit. f. Hyg., xxvi.,
157, 1897.

74 TOXINES AND ANTITOXINfiS.

SpRONCK1 subsequently abandoned his meat extract and
cultivated his bacilli in a decoction of yeast, whereby he obtained
very active solutions of poison. (After forty-eight hours the
lethal dose was 0*05 c.c., and after five to six days 0*005 c.c.)

He boiled 1 part of commercial yeast with 20 parts of water for twenty
minutes with continual stirring. He then allowed the mixture to settle,
decanted the supernatant liquid, added ordinary salt and Witte's peptone
(obtained directly from Rostock), neutralised it, and added normal soda
solution in the proportion of 7 c.c. per litre. Lastly, he heated and filtered
the liquid, and sterilised the filtrate at 120° C.

It is evident from these experiments that the production of
toxines depends upon the alkalinity of the medium, and, as
Roux and YERSIN had already found, it increases with the in-
crease in the alkalinity. MADSEN, however, found (loc. cit.) that
the amount of toxine was not invariably proportional to the
alkalinity.

A second method of increasing the yield of toxine is the intro-
duction of air. Roux and YERSIN found that a current of air
promoted the production of poison, although, according to
MARTIN (loc. cit.) and PARK,2 this stimulation is not very pro-
nounced, provided culture media, good in other respects, are
employed ; whilst MADSEN considers it directly injurious in such
a case, on the ground that when there is a plentiful production
of toxine the air destroys a greater amount of poison than is
formed under its influence. SPRONCK, too, is inclined to believe
that while the growth of the bacilli, and therefore the absolute
production of toxine, is promoted by the admission of air, yet
that the relative production of toxine by a given number of
bacilli is thereby diminished rather than increased. VAN
TURENHOUT (loc. cit.) considers that the admission of air
accelerates the oxidation of the nitrogenous substances, arid with
it the cessation of the acid reaction, but that it also destroys
poison. ARONSON 3 has very successfully replaced the introduc-
tion of air by growing the bacteria in surface cultivations, so that
they offered as large an area as possible to the air. SCHIERBECK 4
recommends treatment with carbon dioxide.

1 Spronck, "Preparation de la toxine diphterique," Ann. Past., xii.,
701, 1898.

2 Park, " The Preparation of Diphtheria Antitoxine," Med. Record,
xlvii., 484, 1895.

3 Aronson, " Immunisierungs- und Heilversuche bei der Diphtheric,"
Wien. med. Woch., 1894, 1956.

4 Schierbeck, " Ueb. d. Einfluss der C02 auf das Wachstum der Diphth.-
B.," Arch. f. Hyg., xxvii., 339, 1896.

DIPHTHERIA TOXINE. 75

Too much air and pure oxygen have a very injurious effect
upon the poison, especially at the incubation temperature (Roux
and YERSIN, MARTIN, MADSEN).

When cultivations are made in this way with every possible
precaution to promote energetic toxic activity, very active toxines
are usually obtained in a short time. Even after thirty to forty-
eight hours they are unmistakably present. In seven to about
thirty days the toxicity reaches its maximum, and then begins
very slowly to decline, this being due to a simultaneous decrease
of production in the cultivation as it grows old and to the com-
mencement of the formation of toxoids. The transplantation of
cultivations again restores their toxigenic power (Roux and
YERSIN). As a general rule, then, in testing primary poisons,
cultivations are used that are not more than three weeks' old
(MADSEN, loc. cit.). For immunising purposes, however, much
older poisons can frequently be used with advantage, since the
toxoids they contain have also an immunising action.

The development of a bouillon of the greatest possible toxicity
does not invariably correspond with that of a particularly viru-
lent living cultivation ; sometimes bacilli that are only slightly
virulent may produce extremely active toxines (MARTIN, loc. cit.).

On the other hand, there are also varieties of diphtheria bacilli
that possess neither virulence nor toxigenic power. LUBOWSKI J
has described an instance of this kind. A bacillus of human
origin produced no active toxines whatsoever, whereas the serum
of the child was strongly antitoxic. The inference is that the
bacillus had been very toxigenic, but that it had completely lost
its powers.

In general, the production of poison shows extraordinary
variations, and its amount cannot be estimated beforehand, even
in the case of a similar cultivation in the same nutrient medium.
MADSEN has vainly endeavoured to explain these variations • the
kind of flasks and the sterilisation of the nutrient medium have
no influence on the results.

We have reached such a stage in the preparation of the poison
of diphtheria bacilli that for immunising purposes we now use
only poisons 0-02 c.c. of which will kill with acute symptoms
guinea-pigs of 250 grms. The essential conditions for the pro-
duction of the virus are : —

1. A suitable cultivation which grows upon the surface — i.e.,
develops on the bouillon in the form of a membrane ;

2. Alkalinity of the bouillon ;

1 Lubowski, " Ueb. einen atoxischen- und avirulenten Diphtheriestamm,"
Zeit.f. Hyg., xxxv., 87, 1900.

76 TOXINES AND ANTITOXINES.

3. Presence of a suitable peptone, preferably Chapoteaut's, in
the proportion of 2 per cent. ;

4. A sufficient supply of air, for which reason the flask should
only be filled to about a third of its capacity ;

5. Not too short and not too long a period of growth at 37° C.
This depends upon the culture, and tests must be made with each
different variety to determine when the maximum production of
poison has been attained. As a rule, it takes from about ten
days to three weeks.

Lastly, means have also been sought to obtain a solution of
poison which, while effecting rapid immunisation, would yet be
but little poisonous, and so would not endanger the lives of
the animals used for the experiments even when given in large
doses. BRIEGER, KITASATO, and WASSERMANN l have found that
diphtheria cultivations grown upon thymus bouillon lose their
toxigenic capacity whilst their immunising power is but little
affected. Here there is presumably a formation of immunising
toxoids. Then, still more recently, MADSEN has succeeded in
effecting immunisation by means of the toxones of diphtheria,
and we shall have more to say about this when we deal with
them later on.

When liquids rich in poisons have thus been obtained, all that
is necessary is to sterilise them. The methods employed for this
purpose are the ordinary ones in general use : heat, addition of
antiseptics, and nitration through bacterial niters.

Roux and YERSIN found that solutions of the poison were rendered non-
poisonous after a few minutes at 58° C. , but that the dry poison could be
heated for more than an hour at 98° C.

FRANKEL 2 endeavoured to sterilise cultivations by heating them for an
hour at 65° to 70° C., whilst BRIEGER and FRANKEL3 found that small
quantities could be sterilised with certainty when heated from three to
five hours at 50° C., but that the poison was speedily destroyed above
60° C.

BEHRING and WERNICKE 4 added calcium chloride to obtain a precipitate
of calcium phosphate in the cultivations, and sterilised the dried precipi-
tate by heating it at 77° C., and this did not materially injure the toxine
simultaneously carried down.

1 Brieger, Kitsato, and Wassermann, " Ueb. Immunitat. u. Giftfesti-
gung," Zeit.f. Hyg., xii., 137, 1892.

2 irankel, " Immunisierung Versuche bei Diphth.," Berl. klin. Woch.,
1890, 1133.

3 Brieger and Frankel, " Ueber Bakteriengifte," ibid., 1890, 240.

4 Behring and Wernicke, "Ueb. Immunisier. u. Heilung von Versuchs-
tierenb. d^ Diphth.," Zeit.f. Hyg., xii., 10, 1892.

DIPHTHERIA TOXINE. 77

Of the chemical agents that destroy the living bacilli, and
simultaneously have a preservative influence upon the poison
solution, and are also added for the latter purpose alone after
filtration, the following are employed : —

Iodine trichloride, IC13, by BEHKING and WERNICKE.

Trier esol, in a O3 per cent, solution, by ARONSON (loc. cit.).

Phenol, also in a 0*3 per cent, solution (SPRONCK, loo. cit.).

Toluene is used by EHRLICH, who keeps his poisons at 15° C.
in the dark under a layer of that substance. According to
ABBA,1 diphtheria toxine can be preserved for two years if kept
under toluene in the dark and in the cold.

The best and most suitable means of preparing and preserving
diphtheria virus for practical immunising purposes is the fol-
lowing method, described by EHRLICH and WASSERMANN : — The
bouillon cultures prepared in the manner described above are
filtered through double filter paper, so that the coarse bacterial
membranes are separated. The filtrate is then covered with a
layer of toluene of the depth of about two fingers'-breadth, and
the whole thoroughly shaken at frequent intervals during two
days. All living micro-organisms will then be destroyed and
the liquid will contain only the poison. When required for use
the poison is invariably withdrawn by means of a pipette, inserted
beneath the preservative layer of toluene. The poison must be
kept in a cool place, and in particular protected from light by
being placed in black flasks or flasks covered with paper. This
is the method of preparing and preserving the poisons in use
in the Prussian Control Station.

For more delicate biological and chemical investigations this
rough filtration is insufficient, and in such cases the liquid must
be completely freed from the cells of the bacilli by filtration
through a proper bacterial filter, and then sterilised.

Attempts to Prepare Pure Diphtheria Toxine. — Attempts to
isolate diphtheria toxine in a pure or, at least, concentrated
condition were made even in the earliest period of the investi-
gation of diphtheria bacilli.

LOFFLER 2 found that evaporation or shaking with ether did
not yield active poisons ; but, on the other hand, by extracting
inoculated meat broth with glycerin he obtained a slightly toxic
substance which could be precipitated by means of alcohol. He
termed this, even in this condition, an "enzyme."

1 Abba, " Ueb. d. Dauer des toxisch. Vermogens beim D.-T. u. Antit.,"
Centralbl.f. Bakt., xxiii., 934, 1898.

2 Loffler, "Der gegenw. Stand d. Frage n. d. Entsteh. d. Diphtherie,"
Deutsch. med. Woch., 1890, Nos. 5 and 6.

78 TOXINES AND ANTITOXINES.

Subsequently Roux and YERSIN, MADSEN, and others obtained,
by means of precipitation with alcohol, saturation with ammonium
sulphate, and precipitation with calcium phosphate, active dry
preparations, which, of course, made no pretensions to purity.
The fundamental idea underlying these experiments is that the
diphtheria poison is carried down mechanically by the voluminous
precipitate — e.g., of calcium phosphate — produced in its solution.

BEIEGER was the first to undertake these investigations in a
systematic and thorough manner. At first, in collaboration with
C. FRANKEL/ he looked for ptomaines in the diphtheria cultures,
but soon had to admit that there were absolutely no volatile bases
present. On this account BRIEGER has considerably modified his
views on the significance of ptomaines. He now obtained from
the cultivations, not only of diphtheria bacilli, but also of many
other bacteria, poisonous proteid substances which he termed
toxalbumins.

The method by which BRIEGER and FRANKEL isolated their
diphtheria toxalbumins is as follows : —

The globulins are first separated from the bouillon by saturating it with
magnesium sulphate at 30° C. This gives a slight precipitate which is
completely non-poisonous.

The poisonous proteids are next precipitated from the bouillon thus
treated, or, equally well, from the fresh bouillon, by means of ammonium
sulphate or sodium sulphate, or also by the addition of a large excess of
alcohol. Since the ammonium sulphate precipitate must be redissolved
and dialysed (not without loss), it is best to employ precipitation with
alcohol.

The bouillon is evaporated to a third of its volume at 30° C. , and treated
with ten times its quantity of absolute alcohol, preferably with the addition
of a few drops of acetic acid. After standing for twelve hours in an ice -chest
the liquid is filtered, and the precipitate taken up with water and again
precipitated with alcohol, this treatment being repeated six or eight times
until the substances dissolve to a completely clear solution in water.
Finally, the preparation is dialysed and dried in vacuo at 40° C.

This treatment yields a snow-white powder, the aqueous
solution of which does not coagulate on boiling. It gives no
precipitate with sodium sulphate, sodium chloride, magnesium
sulphate, nitric acid, or lead acetate, but is precipitated by
saturated solutions of carbon dioxide and by all the usual
reagents for proteids. The substance is thus allied to the
albumoses; it yields a benzoyl derivative, but not a phenyl-
hydrazine compound. In the dry state it can be heated to
70° 0. without injury. It is poisonous, a dose of 2 '5 mgrms.

1 Brieger and Frankel, "Ueber Bakteriengifte," Berl. klin. Woch., 1890,
241.

DIPHTHERIA TOXINE. 79

per kilo, causing certain death. Hence, it is not very poisonous,
a part of the toxine apparently having been destroyed.

WASSERMANN and PROSKAUER l have modified this method. They evapo-
rate the bouillon (which has been neutralised by the addition of 10 to
12 c.c. of normal soda solution, and sterilised by filtration through
KITASATO'S niters, at a temperature of 27° to 30° C. in vacuo to a tenth
of its volume. It is then dialysed in running water at a lower tempera-
ture, by which means the salts and peptones are removed, and the globulins
(which are not poisonous) partially precipitated. The contents of the
dialyser are filtered until clear — an important point — and then poured
into ten times their volume of 60 to 70 per cent, alcohol acidified by the
addition of a few drops of acetic acid, and allowed to stand for twenty-four
hours. The resulting precipitate is separated by filtration and the filtrate
allowed to fall drop by drop into absolute alcohol. The new precipitate
that is formed is also filtered off, and both dissolved in a small quantity of
water and precipitated by the addition of twice the amount of a saturated
solution of ammonium sulphate. This precipitate is again dissolved, and
the solution dialysed until it no longer gives the sulphate reaction, after
which it is again introduced into absolute alcohol and the process repeated
until the aqueous solution is perfectly clear. The preparations are then
dried in vacuo at 37° C. They give all the reactions of albumoses. Only
preparations precipitated by alcohol of, at least, 60 per cent, strength are
poisonous — not the others, so that in this way it is possible to effect a
further separation from the bulk of proteids simultaneously precipitated.
These products were also obtained in the same manner from extracts made
with glycerin and ordinary salt from the organs of human beings infected
with diphtheria. All these preparations, however, were only very slightly
poisonous. Ten mgrms. were required to kill a rabbit in three to four
days, and with 3 mgrms. death did not take place until after eight weeks.

Products of greater toxicity were obtained by WASSERMANN
and PROSKAUER by extracting the organs of poisoned rabbits
with glycerin. This process yielded a , white powder which,
when injected in a dose of 0'2 mgrm. into the veins, killed a
rabbit in six to fourteen days.

The separation of the true toxine from the associated proteid
impurities, and therewith our knowledge that the active principle
is not a protein is also due to BRIEGER,2 who, in collaboration
with BOER, has prepared the toxine in a fairly pure condition.

For this purpose, BRIEGER and BOER made use of the method of
precipitating the toxine in the form of a double salt of zinc. The
precipitation is practically quantitative, at all events in the case of
diphtheria poison, but the difficulty of separating the toxine from the
zinc is rendered the greater by the fact that hydrogen sulphide, which

1 Wassermann and Proskauer, " Ueb. die von d. Diphtheriebacillen
erzeugten Toxalbumine, " Deutsch. med. Woch., 1891, 585.

2Brieger and Boer, "Ueb. d. Toxine d. Diphth. u. d. Tetanus," Deutsch.
med. Woch., 1896, 783 j Zeit.f. Hyg., xxi., 259, 1896.

80 TOXINES AND ANTITOXINES.

would otherwise have been a very suitable agent, destroys the toxine.
After long and careful experiments they arrived at the following method
of separation : —

The poison-bouillon (blood serum was used with good results as the
culture medium) is treated with twice its volume of a 1 per cent, solution
of zinc chloride, and the resulting precipitate thoroughly washed with
water, and then vigorously shaken with a 3 to 6 per cent, solution of
ammonium carbonate. Ammonium phosphate solution is next added until
the whole of the precipitate passes into solution, and there is only a slight
turbidity due to separated zinc phosphate. This is allowed to subside and
then collected on a toughened filter and thoroughly washed with water.
The filtrate is saturated with solid ammonium sulphate, the precipitate
redissolved in water, and the solution precipitated with solid sodium
sulphate, by which means the peptones are left in solution.

As thus prepared the toxine no longer gives the reaction of
proteids. The zinc compound also shows no proteid reactions,
and is optically inactive, but turns red when boiled with a
solution of iron chloride.

Alcohol, ether, acetone, acids, and weak oxidising agents
rapidly destroy the poison, while weak alkalies and reducing
agents do not affect it. Diphtheria toxine can also be obtained,
although, of course, only in very small quantities, from dialysed
urine — i.e., proteid-free culture medium (GUINOCHET, loc. cit.) —
and from other similar nutrient liquids (UscniNSKY1). The culture
medium used by USCHINSKY had the following composition : —
Glycerin, 40 to 50 parts; sodium chloride, 5 to 7 parts; am-
monium lactate, 10 parts; calcium chloride, O'l part; magnesium
sulphate, 0'2 part ; and potassium hydrogen phosphate, 1 part ;
in 1,000 parts of water. The poison did not give the usual
proteid reactions.

Properties of Diphtheria Toxine. — The poison is not known in
a state of chemical purity. Hence, all that can be stated about
it relates to the preparations that contain it in admixture with
other substances. Its most important properties have already
been described by Koux and YERSIN.

It is probably not a proteid^ since the purest preparations
(vide supra) do not give the proteid reactions. An attempt
made by ARRHENIUS and MADSEN 2 to determine the molecular
weight from the speed of diffusion into gelatin has so far only
led to the conclusion that toxines have at all events a much
smaller molecular weight than antitoxines.

1 Uschinsky, "Ueber Diphtherieculturen auf eiweissfreier Nahrlosung,
Centralbl. f. BaU., xxi., 146, 1897. Id., " Les Poisons de la Diphtheric et
du Cholera," Arch, de Med. Exptr., 1893, 293.

2Arrhenius and Madsen, "On the molecular weight of Di.-T.," Fest-
schrift des Statens Serum Institut, Copenhagen, 1902.

DIPHTHERIA TOXINE. 81

It is very sensitive to external influences. The true toxine is
very rapidly destroyed by boiling and fairly rapidly at about
60° C., but the heated solution still retains a certain toxic power,
and when injected into animals produces emaciation and paralysis
as a secondary affection, and frequently causes death. The
poison appears to undergo a similar change within the organism.
Roux and YERSIN have prepared poisonous substances with
quite analogous properties from the organs and urine of children
who had suffered from severe attacks of diphtheria. These
produced toxic effects resembling those caused by toxones (vide
infra).

The simultaneous action of oxygen and light injure diphtheria
toxine very rapidly, but either of the factors alone has much less
effect.

According to PIAZZA/ diffused daylight acts very slowly ; a perceptible
decrease can be observed after twenty-three days, but it does not become
considerable until after ninety- six days. Direct sunlight acts very ener-
getically when oxygen is admitted. Only the ends of the spectrum, the
heat rays, and the actinic rays (red and violet) have any action, the centre
(yellow) being completely inactive.

The toxine is also rendered inactive by acids. Roux and
YERSIN state that acidification of a solution of the poison with
lactic or tartaric acids renders it nearly harmless ; phenol, boric
acid, and borax have a less injurious effect. Neutralisation par-
tially restores the activity.

According to BRIEGER and BOER, it is extraordinarily sensi-
tive to the action of oxidising agents, whereas reducing agents in
slightly alkaline solution have but little effect. This explains
why the slightly alkaline reducing fluids of the body form its
best medium. DELEARDE2 asserts that it is rendered innocuous
by antipyrine. It is also destroyed by other antiseptics when
more concentrated, and by salicylaldehyde (SALKOWSKI 3). Yeast
weakens the toxine (NOBECOURT 4).

It is extremely probable that, in general, the nature of the
surrounding medium has an influence upon the condition and the
action of the toxine, but it is far too sweeping, and is contra-

1 Piazza, "Influenza della luce solare sullatossina difter.," Ann. d'Igiene
Sperim. [New Series], v., 521, 1895; Abst. Centralbl. f. Bah., xix., 914,
1896.

2 De"14arde, " Rech. expe"r. sur les proprie'te's, &c., de 1'antipyrine,"
Arch. de. Med. Exper., 1897, 786.

3 Salkowski, " Ueb. d. Wirkg. d. Antiseptica auf Toxine," Berl. Uin.
Woch., 1898, 545.

4 Nobe'court, "Action des levures, &c,," Soc. BioL, Hi., 753, 1900.

6

82 TOXINES AND ANTITOXINES.

dieted by the undoubted changes that can be detected in the
poison itself, to attempt, like DANYSZ,* to attribute every altera-
tion in the action to changes in the surrounding media. The
other external characteristics are quite analogous to those
belonging to all enzymes.

Diffusion through parchment can be plainly perceived. Thus,
when 5 c.c. of the poisonous bouillon were diffused into 12 c.c. of
water, there passed through, on the average, in twenty-four hours
a lethal dose for guinea-pigs (Roux and YERSIN). On the other
hand, it does not diffuse through membranes formed from the
organs of animals — e.g., the oesophagus, large intestine, the
gall bladder, and the small intestine (CnASSiN and Moussu2).
Nor does it pass through collodion (RoDET and GUECHOFP 3).

It is completely insoluble in pure alcohol, which, however,
slowly destroys it (WASSERMANN and PROSKAUER, loc. cit.), as
also does acetone and ether (BRIEGER and BOER, loc. cit.). It is
likewise destroyed by the digestive ferments in the stomach and
intestine (PALTSCHiKOWSKi4).

Like the enzymes, it is carried down from its solutions by
falling precipitates. On fractional precipitation with calcium
chloride the resulting calcium phosphate precipitate simul-
taneously removes a large proportion of the toxine, the second
precipitate being the most poisonous. The precipitation, how-
ever, is never quite complete, still less so in the case of aluminium
phosphate. In particular, the characteristic poisons that pro-
duce the secondary paralysis remain for the most part in solution.
This calcium phosphate precipitate can be heated to 70° C.
without injury, and is not perceptibly damaged by being kept
for twenty minutes at 100° C. When subcutaneously injected it
slowly parts with its poison, besides producing inflammatory
processes (separation of fibrin, false membranes), so that it
appears to act exactly like the living bacilli. It is also pre-
cipitated by nucleo-histone and nucleic acid (FREUND and
GROSZ5). Toxines almost free from ash can be obtained by
dialysing the aqueous extract of this precipitate, or by precipita-
tion with alcohol.

1 Danysz, "Constitution des toxines," Ann. Past., xiii., 581, 1899.

2 Chassin and Moussu, "Influence de la dialyse, &c.," Soc. Biol., lii.,
694, 1900.

3 Rodet and Gue'choff, Soc. Biol, lii., 965, 1900.

4 Paltschikowski, " Ueb. d. Veranderungen der diphtheritischen Toxine
in dem Nahrungswegen," Abst. Centralbl. f. Bakt., xxv., 1899. See also
the General Part.

5 Freund and Grosz, " Ueb. d. Bez. zw. Gerinnung u. d. Wirkg. d. Anti-
toxine," Centralbl. f. inn. Med., 1895, 613, 637,

DIPHTHERIA TOXINE. 83

Special attention was given some time ago to the behaviour of
solutions of diphtheria virus under the influence of electric currents.
In the first place, SMIRNOW1 and KnuGER2 allowed continuous
currents of low intensity to act upon the toxines, and found that
when the current was cautiously applied there was a slight de-
velopment of acidity at the anode. When the degree of acidity
was such that 1 c.c. of the liquid neutralised about 1-2 c.c. of
normal soda solution the current was discontinued, and it was
then found that the toxic value had been considerably lowered,
but not the immunisation value. Perfectly analogous results
were obtained by D'ARSONVAL and CHARRIN3 in their experiments
with high-tension alternating currents of great power. On these
results were based theoretically untenable and very far-reaching
speculations on the formation of " artificial therapeutic serum "
without the aid of an animal, &c. ; in addition to which the
fantastic idea occurred to D'ARSONVAL and CHARRIN of treating
infection in the body itself by means of such high-tension
currents, which, as is well known, have no action at all upon
man.

All these somewhat mystical conceptions, however, about the
influence of electricity have been explained by MARMIER.* He
has shown that during the electrolysis with continuous currents
oxidising substances are formed, notably hypochlorite and free
chlorine from sodium chloride invariably present, and that these
have a secondary action on the toxophore group of the poison.

As regards alternating currents, MARMIEU was able to demon-
strate that, in spite of cooling with ice, such intense heating
occurred that this alone was sufficient to explain the weakening
of the poison. When he excluded this factor by means of in-
genious devices, there was no perceptible action on the toxine
either of diphtheria or tetanus.

Hence we have here to deal with secondary alterations under
the influence of the electric current, and these evidently tend to
a more rapid formation of toxoid, with accompanying destruction
of the toxophore group.

Physiological Action of Diphtheria Toxine. — Diphtheria toxine
is extremely poisonous to many animals. Roux and YERSIN
estimate the single lethal dose for a guinea-pig at about 0*05

1 Smirnow, "Ueb. d. Behandlungd. Diphtherie," Berl. klin. Woch., 1894,
683; 1895, 645, 675.

2 Kritger, "Ueb. die chem. Wirkg. d. Electriz. auf toxische und
immunis. Bakteriensubstanzen," Deutsche med. Woch., 1895, 331.

3 d'Arsonval and Charrin, "Action des Courants a" haute frequence sur les
toxines bacteriennes," Comptes Rend., cxxii., 280, 1896.

4 Marraier, " Les toxines et l'e"lectricite," Ann. Past., 1896, 469,

84 TOXIN ES AND ANTITOXINES.

mgrm., and for a rabbit about (H mgrm. of the organic ash-free
substance, containing only a fraction, and possibly only a very
small one, of the pure toxine.

CALABRESE and ZAGAEi1 assert that it has absolutely no action
upon cold-blooded animals, but COURMONT, DOYON, and PAVIOT 2
observed paralysis and emaciation in the case of warmed frogs.
The action of the toxine corresponds in all essential points with
the general effects produced by the living bacilli. Hence it is
unnecessary to give an exact description of these symptoms here.
We will content ourselves rather with a quite superficial survey,
since the symptoms of diphtheria are described in detail in
numerous text-books.

As regards the general action of diphtheria virus, the following
points have been recorded : —

One of the most striking symptoms is the considerable en-
largement of the blood-vessels, which occurs after a certain
period of incubation (vide infra).

The temperature first rises and then sinks below normal (to
25° C.), this being due, according to ARLOING and L/AULANiE,3
to the decrease in the vital energy and the oxidising processes.
With very large doses the hypothermia may persist (COURMONT
and DOYON 4).

The whole metabolism of animals after poisoning with diph-
theria toxine has been studied, with reference to the amounts
of chlorine and nitrogen, by PACE5 and others. One of the
phenomena is an extensive decomposition of albumin.

Special interest attaches to the question whether diphtheria
toxine has a directly injurious effect upon the heart's action or
not. It is true that collapse of the heart eventually occurs, and
that the heart remains arrested in diastole ; but it is not yet
decided whether this is not the secondary result of a primary
paralysis of the vasomotors.

KRASSNOWS asserts that the main action of diphtheria toxine

1 Calabrese and Zagari, " Ricerche sulla tossina ed antitossina difter. ,"
Giorn. Internaz. di Scienze Mediche, 1895, No. 4, 19-21; Baumgarten's Jb. ,
1895, 215.

2 Courmont, Doyon, and Paviot, "Action de la toxine diphte'rique sur le
Systems nerveux de la grenouille, &c.," Soc. BioL, xlvii., 210, 1895.

3 Arloing and Laulanie", " Et. expe'r. sur les troubles imprimes . . . par
les toxines diphth.," Soc. Biol, xlvii., 433, 1895.

4 Courmont and Doyon, " Marche de la temperature dans I'intoxication
diphth.," Arch, de Physiol., xxvii., 252, 1895.

5 Pace, "Influenza della tossine difter.," // Polidinico, 7; Baumgarteii's
Jb., 1900, 180.

6 Krassnow, " Zur Pharmakol des Di.-T.," Wratsche-bnaja Gazetta, 1904,
23; Biochem. CentralbL, iii., NO. 304, 1904.

DIPHTHERIA TOXINE. 85

is upon the respiratory centres, and that only in the final stages
is the heart itself also affected.

ENRIQUEZ and HALLION 1 found that the pressure of the blood
did not fall until after some time ; this was confirmed by BECK
and SLAPA,2 who only observed a sinking in the pressure of the
blood shortly before death.

Comprehensive investigations into the behaviour of the heart's
action in cases of poisoning with diphtheria toxine have been
made by ROMBERG, PASSLER, BRUHNS, and MiJLLER,3 who con-
clude from the results that there is not a primary effect upon
the heart's action, but that the injury to the vasomotors is the
only direct action. The heart is only injured in consequence of
a deficient supply of blood. STEJSKAL* has severely criticised
these results and claims, on the strength of his assumed more
accurate experiments, that diphtheria virus does cause direct
injury to the heart.

He concludes that after a very short acceleration of the heart's
action there occurs a short period of continuous weak action,
which is followed by a period of acceleration, the final result
being a decided decrease in the \york done by the heart, attribut-
able to injury of the heart rather than of the vasomotors.

These conclusions are supported by the researches of FENY-
VESSY 5 and SHARP,6 who were also able to prove the direct effects
of diphtheria virus upon the isolated heart of a frog.

MOLLARD and REGAUX7 observed myocarditic alterations in
the heart after death, which possibly are also to be regarded as
evidence of direct injury.

EppiNGER8 also observed serious changes, which he termed
myolysis, in the muscle of the heart. From this he concluded

1 Enriquez and Hallion, "Sur les effets physiol. de la toxine diphth.,"
Arch, de Physiol., xxvii., 515, 1895.

2 Beck and Slapa, "Ueb. den Einfl. des D.-G. auf den Kreislauf," Wien.
klin. Woch., 1895, 333.

3 Romberg, Passler, Bruhns, a.nd Miiller, " Unters. lib. d. allg. Pathol.
der Kreislaufstorung bei akuten Infektionskrankh.," Arch. f. klin. Med.,
Ixiv., 652, 1899.

4 Stejskal, " Kritisch-experim. Unt. lib. d. Herztod infolge von Diphth.-
Toxin.," Z.f. klin. Med., xliv., 367, 1902.

5 Fenyvessy, "Ueb. d. Wirkg. d. D.-T. u. Antit. auf das Froschherz.,"
Jahrb.f. Xinderh., new series, xliii., 216, 1896.

6 Sharp, "The action of the products of the organism of diphtheria on
the heart of a frog," Journ. of Anat. and Physiol., xxxi., 199, 1897.

7 Mollard and Regaux, "Le'sions du myocarde dans 1'intox. aigiie par la
t. d.," Ann. Past., xi., 97, 1897.

8 Eppinger, "Die toxische Myolyse des Herzens bei Di.," Deutsch. med.
Woch., 1903, Nos. 15, 16.

86 TOXINES AND ANTITOXINES.

that the toxine attacked the muscle directly, combined with it,
and thus injured it.

As regards individual organs, the following facts are on
record : —

Intestine. — COURMONT, DOYON, PAViOT1 observed exudations
and enteritis membranacea in the small intestine of dogs.

Liver. — COURMONT, DOYON, and PAVIOT 2 found a hepatitis
parenchymatosa, with hypercemia and internal hemorrhage, in
dogs after intravenous injection. BALDASSARI 3 observed thick
swellings and similar changes, such as occur in poisoning by
phosphorus and arsenic. He records the occurrence of epithelial
changes in the kidneys.

Local lesions of the eye similar to those that occur in diph-
theria have been observed by MORAX and ELMASSIAN.* GATTI 5
has also found that direct contact of diphtheria toxine with the
retina produced injuries which led to swelling of the tissue, with
conservation of the nuclei and ganglionic cells.

But, above all, it also causes serious disturbances in the central
nervous system, which we cannot describe in detail here. Investiga-
tions on this point have been made by ENRIQUEZ and HALLION,G
CROCQ fils,7 THOMAS,8 MuRAWJEW,9 and others.

It is very questionable, however, whether these lesions of the
nervous system are to be directly attributed to diphtheria toxine
as such. So far as these changes have to do with the diph-
theritic secondary paralysis, they must probably be attributed
to the toxones. Whether the pure toxine can also produce
secondary paralysis, or whether this is only a function of the
toxones, cannot be asserted with certainty as yet, although the
latter alternative is the more probable. The symptoms of

1 Courmont, Doyon, and Paviot, "Lesions intestinales dans 1'intoxic-
diphth.," Arch, de Phys., xxvii., 484, 1895.

* Courmont, Doyon, and Paviot, ' ' Lesions hepatiques engendres par la
t. d.," Arch, de Phys., xxvii., 687, 1895.

3Baldassari, "Ueb. d. Wirkg. d. D.-T. auf den Zellkern," C. f. allg.
Pathol.,vn., 625, 1896.

4 Morax and Elmassian, "Action de la toxine d. sur les muqueuses,"
Ann. Past., xii., 210, 1898.

5 Gatti, "L'azione di alcune tossine batteriche sopra gli elementi della
retina," abst. by the author in Biochem. CentralbL, L, No. 775, 1903.

6 Enriquez and Hallion, "My elite exper. par t. d.,"Soc. BioL, xlvi., 312,
1894.

7 Crocq, "S. 1'alterat. du. Syst. nerveux dans 1. paralys. diphth.," Arch,
med. exper., vii., 503, 1895.

8 Thomas, Boston Med. and Surg. J., 1898, No. 4, et seq.

9 Murawjew, "Des D.-T. u. Antit. in ihrer Wechselwirkg. auf das
Nervensystem d. Meerschw.," Fortschr. d. Med., 1898, 93.

DIPHTHERIA TOXINE. 87

general poisoning are completely analogous to those produced
by the living bacilli, so that we are justified in assuming that
the latter cause such deleterious effects solely by means of the
toxine that they produce.

Guinea-pigs after poisoning with diphtheria toxine show, on
post-mortem dissection, the typical appearance of the disease
produced by the bacilli.

FLEXNER! has given a most careful description of the appear-
ance on section after poisoning with diphtheria toxine : oedema,
swollen glands, congestion and haemorrhage of the suprarenal
bodies ; otherwise there is little that is characteristic even under
the microscope. COUNCILMAN, MALLOKY, and PEARCE* have also
made a careful investigation of these symptoms.

Roux and YERSIN were able to produce a typical pseudo mem-
brane in guinea-pigs by inoculation of the trachea and vagina
with diphtheria virus.

The guinea-pig is the most susceptible animal, but, according to
EHRLICH, different varieties have different degrees of suscepti-
bility. Horses, goats, and sheep are very susceptible. Rabbits
are less susceptible, and mice still less, being almost immune.

After subcutaneous or intravenous inoculation the poison dis-
appears from the circulation with extraordinary rapidity, being
concentrated and fixed by the receptors of the tissue. This can
be recognised by the fact that doses of an ti toxine introduced
after the injection of the poison lose all power of action with
extreme rapidity as the difference in time increases.

DdNiTZ3 found that a quantity of an ti toxine just sufficient to
neutralise seven times the lethal dose could no longer save the
animal fifteen minutes after inoculation, and that after one and
a half hours even very large doses were insufficient.

On the other hand, the toxic power of the blood after injec-
tion of diphtheria toxine has also been directly investigated.
BoMSTEiN4 inoculated rabbits with a dose double that required
to kill a guinea-pig for each c.c. of the blood of the rabbit.
After an hour there was still O5 thereof present, after three
hours 0-25, and after twelve hours 0-12. The rate of decrease

1 Flexner, " The pathology of toxalbumin intoxication," Johns Hopkins
Hosp. Record, vi., 259, 1897 (reprint).

2 Councilman, Mallory, and Pearce, "Diphtheria: A Study of Bac-
teriology, &c.," 1901. Quoted in detail by Vaughan and Novy, The
Cellular Toxines, 1902, p. 75, et seq.

3 Db'nitz, " Ueb. d. Grenzen d. Wirksamkeit d. Diphtherieheilserums,"
Arch, infernat. d. Pharmacodyn. , v., 425, 1899.

4Bomstein, "Ueb. d. Schicksal der Diphth.-T. in Tierorganismus,"
CentraM.f. Bakt., xxii., 785, 1898.

88 TOXINES AND ANTITOXINKS.

is thus continually smaller, and the last traces seem to disappear
very slowly.

Similar results were obtained by CnoLY.1 He found the
toxine still unchanged after five minutes, and that half, at most,
had disappeared after two hours. The poison is then, also, not
to be detected in the organs of the body, nor is it excreted by the
urine or intestinal fluid (BOMSTEIN), although BftUNNER2 claims
to have detected it in the juices of the muscle, and SALTERS in
sweat (?).

It does not act by way of the stomach or the intestines. It is
destroyed, like all other toxines, by the digestive fluids, notably
by the liquid of a pancreatic fistula ; less readily by the gastric
juice. In very large doses it shows a certain amount of activity
(NENCKI, SIEBER, and ScnouMANOwsKi4), as also when the
mucous membrane is artificially injured (see also the General
Part).

CoNSiGLio5 claims to have observed a characteristic property of diph-
theria toxine. He found that in small doses it had a stimulating effect
upon fermentative processes, but a restrictive one in larger doses ; whilst,
on the other hand, it invariably had a very unfavourable influence upon
the germination processes of seeds.

Toxoids and Toxones. — Nothing whatever is known of the
properties of the toxoids that occur in old cultures (see General
Part). We have not the slightest idea of the chemical process
which induces the supposed inactivity of the toxophore group.
Nor can we state whether they are physiologically indifferent in
the free state, or whether they still possess a slight toxic capacity,
although, according to numerical results — sharply differentiated
values obtained during the alteration of the poison — the latter
alternative is not very probable.

The only proof that other substances are present in weakened cultures is
the fact recorded by BRIEGER and FRANKEL (loc. cit.) that they found in
such cultivations a non-poisonous substance, which was somewhat soluble
in dilute alcohol, and could be distinguished from the toxine by chemical
tests — e.g., by the formation of a phenyl-hydrazine compound. This sub-

1 Croly, "Sur 1. disparition de la tox. diphth. injectee dans le sang.,"
Arch, intemat. de Pharmacodynamie, iii.

2 Brunner, "Unt. lib. die Wirk. von Bakterien- u. Pflanzengiften," Abst.
Centralbl.f. Bakt., xxiv., 184, 1898.

3 Salter, "The Elimination of Bacterial Toxines," Lancet, 1898, i., 152 :
cf. Walsh, ibid., 362.

4 Nencki, Sieber, and Schoumanowski, "Die Entgiftung der Toxine,"
Centralbl.f. Bakt., xxiii., 840, 1898.

5 Consiglio, " Azione di alcune tossine, &c.," Arch, di Farm., vi., No. 3,
1898; MalysJb., 1898, 634.

DIPHTHERIA TOXINE. 89

stance was also found in abundance by WASSERMANN and PBOSKAUER (loc.
cit. ) in weakened cultivations. It must be left an open question whether
it has anything to do with the toxoids.

On the other hand, we know that free toxones, which form a
second primary decomposition group of products, and can be
studied beyond doubt in the transitional stage, do exert quite dis-
tinct toxic effects. EHRLicn1 has observed oedema, and secondary
paralytic affections. MADSENS found that they produced cedema,
but not necroses or alopecice. The secondary paralysis occurred
in thirteen to thirty-three days, and, in the case of large doses,
invariably ended in death. He also observed differences in the
behaviour of different cultures. With very much weakened
poisons, in particular (i.e., those in which a large amount of
toxones was present in the transition from L0 to L+), there was
a more frequent occurrence of cases in which death was delayed.
On the other hand, it is very remarkable that rabbits are killed
with very acute symptoms by toxones (EHRLICH).

But the question of toxones is also of great theoretical interest.
If toxones are really true haptines — i.e., substances which possess
haptophore groups identical with, or closely allied to, those of the
true toxines — they must also be able to throw off receptors — i.e.,
to produce immunity, if EHRLICH'S side-chain is correct.

This very interesting question has been experimentally investi-
gated by MADSEN and DREYER,3and answered in the affirmative.
In the case of many poisons which were so far neutralised by
antitoxine (vide supra) as to contain only toxones still in the
free state, they succeeded in producing a fairly high degree of
immunity in guinea-pigs, rabbits, goats, and horses. The serum
of the animals thus treated contained fairly considerable, though
very fluctuating, quantities of antitoxine, reaching nearly as much
as 400 I.E. per c.c. in the case of a horse. They were then also
immune against the action of large doses of the fully poisonous
toxine, from which the conclusion may be drawn that there is a
close relation between the haptophore groups of the toxones and
those of the true toxine.

The rabbit occupied an exceptional position. At first the
animal proved extremely susceptible. Mixtures which only
produced the effect of toxones upon guinea-pigs still produced

1 Ehrlich, " Zur Wertbemessung des Diphtherieheilserums," Klin.
Jahrb., vi.

2Madsen, "Constit. du Poison diphthe"rique," Ann. Past., xiii., 568,
1899.

3 Madsen and Dreyer, " Ueb. Immunisierung mit d. Toxonen d. Diph-
theriegiftes," Zeit. f. Hyg., xxxvii., 249, 1901.

$0 TOXlNfiS AND ANTITOXlNES.

acute symptoms in the rabbit. On the other hand, most of
the animals experimented upon with such mixtures, which
undoubtedly contained nothing but toxones still in the free
state, perished later with secondary paralytic affections. One
animal, however, remained alive and was able to receive without
injury very large amounts of toxone, eventually even several
multiples of a lethal dose of the toxine ; but its serum never
showed even the slightest trace of an antitoxine constituent.

No explanation of this remarkable phenomenon can yet be
given, unless we are willing to accept the view that although the
active immunisation had stimulated the formation of protective
substances, these had not been thrown off in a free state into
the circulation (" sessile receptors " in EHRLICH'S terminology).

Owing to the weak affinity of diphtheria toxone it is fixed
much more slowly by the receptors of the organism — i.e., it does
not disappear so rapidly from the circulatory system as the
toxine. We have shown above that even fifteen minutes after
the injection of seven times the lethal dose DONITZ was no
longer able to save the animal by means of a quantity of
antitoxine exactly neutralising the poison in vitro.

DREYER1 has made corresponding experiments with the
toxones.

After two hours the equivalent amount of antitoxine was
invariably able to render completely harmless a dose of toxone
which would otherwise have been undoubtedly fatal after twelve
to eighteen days, both being introduced into the rabbits by
intravenous injection. With an interval of five hours between
the injections the first signs of paresis appeared ; with ten hours'
interval they invariably occurred, though not until after twenty
to twenty-two days ; with an interval of sixteen to twenty-four
hours the antitoxine was completely powerless.

Similar results were obtained in the therapeutic experiments
made by DREYER with larger quantities of antitoxine, both
substances being subcutaneously injected into guinea-pigs. A
five-fold dose introduced twenty -four hours after the injection of
a dose of toxone which would be undoubtedly fatal after twelve
to eighteen days, was sufficient to prevent death, though not
invariably to prevent paresis.

Of seven animals that received after forty-eight hours a dose
of antitoxine neutralising 5,000 to 10,000 times the amount of
toxone injected, one had no paresis, while the others had only
slight attacks ending in a cure after eighteen to twenty-five days.

1 Dreyer, "Ueber die Grenzen d. Wirkung d. Diphtherieheilserums
gegeniiber d. Toxonen," Zeit.f. Hyg., xxxvii., 267, 1901.

DIPHTHERIA TOXINE. 91

After an interval of 4 x 24 hours, and with a dose of anti-
toxine neutralising 21,000 times the amount of toxone, there
were two fatal cases of paresis, while one recovered after eighteen
to twenty-one days. With an interval of 5 x 24 hours and the
same dose of antitoxine (neutralising 21,000 times the amount
of toxone) there were three cases of paresis, one of which ended
in death after sixteen to twenty-nine days, while the other two
recovered. The toxone is thus fixed relatively very slowly, and
can still be influenced for a long time by the antitoxine, and
probably also detached much more readily than the toxiiie from
an already formed combination with the receptor.

Diphtheria Antitoxine. — Diphtheria toxine produces in the
body of an animal the corresponding antidote, the antitoxine.
The antitoxine, which is thus a normal, but complementary,
product of the body, occurs therefore in the fluids of the tissues
and in the secretions of individuals that have recovered from
diphtheria, or in artificially immunised animals. We cannot
deal here with the mode of production of antitoxine and its
relationship to artificial immunity and serum therapy; only
what is known of diphtheria antitoxine, as such, can be described
here.

The two main sources for the artificial preparation of anti-
toxine are blood serum and the milk of immunised animals.
The more highly the animal is actively immunised, the richer
these fluids in antitoxine, and the greater the prospect of
obtaining it from them in a highly concentrated, if not pure,
condition.

The antitoxine is present in considerable quantities only in
the serum of artificially immunised animals, although, naturally,
it has been found to a less extent in the blood of those con-
valescent from diphtheria.

It is, however, particularly interesting that even the normal
serum of many species of animals contains diphtheria antitoxine,
and that thus haptophore side-chains are normally present in
the circulatory system. A. WASSERMANN 1 was the first to detect
antitoxine in 85 per cent, of the adults and 60 per cent, of the
children examined. Antitoxine has also been frequently found
in horse serum — e.g., by DiEUDONNE2 and CoBBETT.3

1 A. Wassermann, "Ueb. d. pers. Prophylaxe gegen Diphth.," Zeit. f.
Hyg., xix., 408, 1895.

2Dieudonne, "Ueb. Diphtheriegift neutralis. Wirkung d. Serum-
globuline," Arb. a. d. Kaiserl. Ges.-Amt., xiii., 293, 1897.

3Cobbett, "Enthalt das norraale Pferdeserum Antitoxin ?" Centralbl.f.
Bakt., xxvi., 458, 1899. See also Lancet, 1899, ii., 332.

92 TOXINES AND ANTITOXINES.

As an explanation of the fact that children at the breast are
seldom attacked by diphtheria great importance attaches to the
discovery of FISCHL and v. WUNSCHHEIM l supplementing the
work of WASSERMANN, that the serum of healthy suckling babies
already contains fairly considerable quantities of diphtheria
antitoxine ; they detected it in 83 per cent, of all the children
examined. The correspondence with the numbers observed by
A. WASSERMANN indicates that here there must be a transference
of antitoxine from the mother to the baby by means of the
placental blood and the milk.

As in the case of the toxine we must here distinguish between
two aims — those that have in view only the preservation and
concentration of the antitoxine for practical purposes with or
without the production of a solid product, and, on the other
hand, the attempts that are made in careful investigations to
obtain an insight into the constitution of the antitoxine itself,
by determining its ratio to the different proteids in its mother
liquids and by isolating as completely as possible the active
principle.

Preservation of the Serum. — The first necessity is to preserve
the serum so that it retains its full amount of antitoxine, while
decomposition is prevented. The serum will keep for a fairly
long time if protected from the action of air, light, or bacterial
contamination. Since it is a good nutrient medium for bacteria
it must be prepared and kept with strict aseptic precautions. In
addition to this an antiseptic agent, such as phenol in solutions
up to 0*5 per cent, in strength, is usually added.

According to DI MARTINI 2 sterilisation by nitration is un-
suitable, since a considerable proportion of the antitoxine is kept
back by the Chamberland filter. This is categorically denied by
DziERZGOWSKi.3 CoBBETT4 has thoroughly investigated this
point, so extremely important from a practical point of view, and
asserts that such retention frequently occurs, especially with the
Berkefeld or Martin filter, and may be very considerable, par-
ticularly when the pores of the filter begin to be clogged.
Hence the filtration ought not to be forced. When it does not
take place easily without the application of high pressure from

1 Fischl and v. Wunschheim, " Ueb. Schutzkorper im Blute der
Neugeborenen," Prager med. Woch., 1895, No. 45, et seq.

2 di Martini, " Sul comportamento del siero antidifterico filtrate," Rif.
Med., 1896, No. 266 ; Abst. Centralbl. f. Bakt., xxiv., 861, 1898.

3 Dzierzgowski, " Z. Frage : Ueb. die Verluste des Diphth.-Heils. bei der
Filtration," Centralbl. f. Bakt., xxi., 233, 1897.

4 Cobbett, "Der Einfluss der Filtration auf das Diphth. -Antitoxin,"
Centralbl. f. Bakt., xxiv., 386, 1898.

DIPHTHERIA TOXINE. 93

outside it is very scanty. The antitoxine is retained by a
gelatine filter (BRODIE l).

The antitoxine is rapidly destroyed when heated to the boiling
point. Even a temperature of 60° to 70° C. is injurious (VAN
DE YELDE 2). The dry antitoxine, however, can resist a tempera-
ture of 110° C. for half an hour, and of 140° C. for fifteen
minutes (CAMUS3). It can resist lower temperatures to some
extent. SPRONCK* has heated it to 59° C. for sterilisation
without materially weakening its antitoxine power. Temper-
atures up to 36° C. have no influence upon it (PALMIRSKI
and ORLOWSKi5), although, according to MULLER,G they are
very injurious when continued for longer periods (two
months).

It appears to be unaffected by very low temperatures
(BuJWiD7). Direct sunlight and plentiful introduction of air
(long continued shaking) are injurious (PALMIRSKI and OR-
LOWSKI, loc. cit.). MULLER (loc. cit.) found that yellow and red
light were harmless, even after being allowed to act for months,
but that blue and green light were very injurious, as was also
daylight after long continued action, whilst MARENGHI 8 found it
had no effect if the exposure were short. According to MULLER
all gases have a very injurious action, and hence he recommends
that the tubes should be filled as full as possible so as to exclude
any considerable amount of gas. It is thus best to keep them in
the dark, upon ice, and closed in such a way as to exclude air
and to protect them from contamination. Solid preparations of
antitoxine should also be protected from moisture by means of
phosphoric anhydride in vacua, as recommended by EHRLICH ;
under such conditions they remain unaltered for years.

1 Brodie, /. ofPathoL, 1897, 460, quoted by Martin and Cherry, "The
Antagonism between Toxines and Antitoxines," Proc. Roy. Soc., Ixiii,,
420, 1898.

2 Van de Velde, "Beitr. z. Kenntnis der antitox. Kraft d. antidiphth.
Serums," CentralU. f. Bakt., xxii., 527.

3 Camus, "Resistance aux temp, e'leve'es des vaccins desse'che'es," Soc.
Biol, 1., 235, 1898.

4 Spronck, "Chauffage du serum antidiphth6rique," Ann. Past., xii.,
695, 1898.

5 Palmirski and Orlowski, Medycyna, xxiii.; Abst. in Gentralbl. f. Bakt.,
xix., 916.

6 Miiller, "Ueber d. Resistenz d. Diphtherieheilserums gegeniiber
verschiedenen Einfliissen," CentralU. f. Bakt., xxiv., 251, 1898.

7 Bujwid, " Ueb. e. Methode d. Concentrat. d. diphth. Heils. mittelst
Ausfrieren," CentralU. f. Bakt., xxii., 287, 1897.

8 Marenghi, "Ueb. die gegenseitige Wirkung des antidiphth. Serums
imd des Diphth. -Toxins," CentralU. f. Bakt., xxii., 520, 1897.

94 TOXINES AND ANTITOXINES.

SPRONCK1 asserts that antidiphtheritic serum that has been
kept for twenty minutes at a temperature of 58° C. no longer
possesses the property of producing the unpleasant erythema
resembling urticaria, which is frequently the case with serum
that has not been thus heated.

It appears to be destroyed in the digestive tract. DZIERZ-
GOWSKi2 found that immunisation could not be effected by
introduction of the antitoxine per os. Only in the case of rabbits
could a slight absorption be observed after its introduction into
the empty stomach.

Hydrochloric acid has a particularly injurious action, while
neutralised pepsin and the pancreas and gall are fairly innocuous
to it. Yet it is not absorbed by way of the intestine.

There is no doubt, however, but that it must be gradually
destroyed under the influence of the digestive ferments, for its
resistance to the action of trypsin is not absolute. At all events,
PICK (loc. cit.) found that the antitoxine was destroyed to a large
extent (almost two-thirds), even after nine days' exposure to the
digestive action of trypsin.

It also disappears rapidly from the blood after subcutaneous
injection, as was shown by PASSiNi,3 whose explanation is that a
partial combination takes place within the tissues.

BOMSTEIN* has studied the question of the survival of the
antitoxine. After a single introduction of a quantity of the
antidiphtheritic serum it rapidly disappears from the blood.

It is not known what becomes of it, but it appears to be
destroyed, for it is only during the first days that slight traces
are present in the urine ; nor can it be detected with certainty
in the organs.

An interesting observation recorded by MuRAWJEW5 is that the anti-
serum is far from being inert, and that it produces somewhat serious
injuries in the cells of the spinal cord of guinea-pigs. He, therefore, gives
a warning against the use of too large doses at once.

This, however, is doubtless to be attributed to the serum of the foreign
species of animal, and not to the antitoxine, since normal serum also pro-

1 Spronck, " Chauffage du Se"rum antidiphth.," Ann. Past., xii. , 697, 1898.

2 Dzierzgowski, "Die Bezieh. d. Verdauungsfermente zum Anti-
diphtherieserum," Arch. d. Sciences Biol, vii., 337 ; Maly's Jb., 1899, 957.

3 Passini, " Vers. fiber die Dauer d. antidiphth. Schutzimpfung," Wlen.
Uin. Woch., 1896, 1111.

4 Bomstein, " Z. Frage der passiven Immunitat. bei Diphtheric,"
CentraiU.f. BaU., xxii., 587, 1897.

5 Murawjew, "Das Diphtherietoxin u. Antitoxin in ihrer Wechsel-
wirkung auf das Nervensvstem d. Meerschweinchens," Fortschr. d. Med..
1898, 93.

DIPHTHERIA TOXINE. 95

duces similar injuries, although the results of my experiments contradict
the conclusion of LINOSSIER and LEMOiNE1 that even minute quantities of
normal horse-serum produce serious illness (persistent albuminuria) in
rabbits.

Concentration of the Antitoxine. — Soon after the first active
antitoxine sera had been prepared, experiments were begun with
the object of isolating the antitoxine.

At first these were concerned with the purely practical idea of
preparing solid substances which had not lost the specific property
of the antitoxine. For this purpose the usual method employed
was simply to precipitate the antitoxine and the proteids of the
serum simultaneously — e.g., by means of concentrated ammonium
sulphate, and to dry these precipitates at a low temperature in
vacua. In this way fairly active, dry residual preparations of the
antitoxine were obtained, and these, when shaken with very weak
alkalies, yielded solutions of the antitoxine ranging up to several
hundred times the strength of the original serum.

ARONSON,2 for example, proceeded as follows : — One hundred c.c. of the
serum were diluted with 100 c.c. of water, and treated with 70 c.c of a 10
per cent, solution of aluminium sulphate. Ammonia was then added in
such quantity that the solution still remained slightly acid, and the result-
ing precipitate was washed with 150 to 200 c. c. of cold water. It contained
as much as 95 per cent, of antitoxine. Similar precipitates were produced
by zinc sulphate and potassium ferrocyanide, and by iron chloride and
ammonia. These precipitates were repeatedly shaken for long periods in a
shaking apparatus with a hundredth part (calculated on the original serum)
of very dilute alkali solutions, which just turned red litmus blue.

BuJWiD3 freezes the serum, with the result that the water first
separates as ice, leaving behind a turbid solution very rich in
toxine. The crystals can then either be separated by "centri-
fuging" or slowly thawed, in which process two layers are
formed, so that the upper layer, containing pure water, can be
withdrawn. Results quite analogous to these were simultan-
eously obtained by ERNST, COOLIDGE, and CooK.4

BRIEGER and BOERS found that precipitates of calcium phos-

1 Linossier and Lemoine, "Action nephrotoxique des inj. d. se'rums
normaux," Soc. BioL, lv., 515, 1903.

2 Aronson, "Weit. Unters. iiber Diphtherie," Berl. klin. Woch., 1894,
425.

3 Bujwid, " Ueb. e. Methode z. Concentr. d. Diphtherieheils. mittelst
Ausfrieren," Centralbl. f. Bakt., xxii., 287.

4 Ernst, Coolidge, and Cook, "The Effect of Freezing upon the Anti-
diphtheritic Serum," J. Boston Med. Soc., if., 166 ; Baumgarten's Jb., 1898,
242

5 Brieger and Boer, " Ueber Antitoxine und Toxine," Zeit. f. Hyg., xxi.,
259,

96 TOXINES AND ANTITOXINES.

phate or of the hydroxides of the heavy metals could not be
used as means of isolating diphtheria antitoxine. Moreover, the
method employed by TISSONI for the isolation of his tetanus anti-
toxine (q.v.) — viz., salting out with solid magnesium sulphate at
30° C. — gave a yield of at most 50 per cent. On the other hand,
BRIEGER and BOER have obtained quantitative yields by the
following method : —

Ten c.c. of the immune serum are mixed with 10 c.c. of distilled water,
and the mixture treated with 4 grms. of dry potassium chloride, and 4 to 5
grms. of finely-powdered sodium chloride, and left for eighteen to twenty
hours in an incubating oven, after which the precipitate is dissolved in
water and dialysed. Frequently the precipitate agglomerates into nearly
a solid mass on stirring with water, and no antitoxine is dissolved. In
such cases it can only be extracted with weak alkalies. A more exact ex-
planation of this anomaly cannot be given. After the residue has been
dissolved, the solution is treated with an equal volume of finely-divided
magnesium sulphate, and again left for two to three hours in the incubating
oven, the result being that the antitoxine is quantitatively precipitated.
In this way they obtained 0*2 grm. of active dry substance from 10 c.c. of
the serum.

BRIEGER and BOER have also endeavoured to concentrate the
antitoxine by precipitation with the salts of heavy metals. Zinc
salts proved to be the most suitable.

The serum was diluted with five times its volume of water and treated
with twice the quantity of a 1 per cent, solution of zinc sulphate or zinc
chloride, and the precipitate washed with water. ^ It was then dissolved in
a ,^ N. solution of alkali hydroxide, and the zinc precipitated with
carbon dioxide. These experiments showed that in the precipitation with
zinc sulphate the antitoxine was precipitated with the zinc, while the zinc
chloride remained in solution. The portion containing the antitoxine was
dried in a desiccator, and it was found that the zinc albuminate was
somewhat soluble in water, while the zinc antitoxine was insoluble. The
latter was, therefore, again dissolved in alkali solution and the zinc once
more precipitated with carbonic acid. It was not possible to effect by
this means a complete separation of the antitoxine from the zinc.

BRIEGER and EnRLicn1 were the first to obtain active dry
preparations from the milk of immunised animals by means of
ammonium sulphate. They obtained one preparation which
contained 14 per cent, of ammonium sulphate and was 400 to
600 times as active as the original milk.

WASSERMANN 2 has modified this method to some extent,
proceeding as follows : —

1 Brieger and Boer, " Beitr. z Kenntn. d. Milch immuner Tiere," Zeit. f.
Hyg., xiii., 336, 1893.

aWassermann, "Ueb. Concentrierung v. Antitoxin aus Milch, Zeit. f.
Hyg., xviii., 236, 1894,

DIPHTHERIA TOXINE. 97

The milk collected with aseptic precautions is treated with about 20 c.c.
of N. -hydrochloric acid per litre, coagulated as rapidly as possible by
means of rennet enzyme, and filtered from the separated paracasein. The
whey is then thoroughly shaken with chloroform and allowed to stand,
after which the supernatant liquid is decanted from the layer of chloroform
charged with fat and bacterial cells. Each 5 litres of the whey is next
treated with a calculated amount (determined by a preliminary experi-
ment) of 30 to 33 per cent, ammonium sulphate solution, and the pre-
cipitate dried on porous earthenware at 35° C. in vacuo, separated as
completely as possible from solid ammonium sulphate and dissolved in
water to form a solution of the desired concentration.

The experiments to obtain antitoxine preparations from the
liquid serum have hitherto proved nearly worthless from the
practical point of view, since at present it is preferred to
immunise animals to such an extent that the serum of the
animal itself contains so many antitoxic units per c.c. as to
render concentration superfluous.

On the other hand, the scientific side of the question has been
followed with equal interest, and attempts have been made to
isolate the antitoxine as such as completely as possible and to
determine whether it is to be considered as an individual sub-
stance, and with which proteid of the serum it combines. In
addition to this, experiments have been made to discover points
of difference between ordinary serum and that containing anti-
toxine.

SZONTAGH and WELLMANN* made a comparative examination
of normal and antitoxic horse-serum. In neither did they detect
any nucleo-albumin, while the proportions of globulin and of ash
and chlorine, and the specific gravity were the same in each.
The only difference observed was that the antitoxic serum
contained, on the average, about 0-25 per cent, more nitrogen
than the normal serum ; they rightly attribute, however, no
special importance to this difference. On the other hand, they
claim to have found that the therapeutic serum shows a decrease
in the lowering of freezing point and in the electrical conduc-
tivity, which are apparently proportional to the amount of
antitoxine.

FREUND and STERNBERG 2 have endeavoured to isolate the anti-
toxine in as pure a state as possible by the following method : —

The therapeutic serum was treated with alum and dialysed. No
precipitate was produced in the dialysate by zinc salts. Nor was the

1 Szontagh and Wellmann, " Vergleichende chem. Unters. iib. d. normale
Pferdeserum und das Diphtherieheilserum," Deutsch. med. Woch., 1898, 421.

2Freund and Sternberg, " Ueber Darstellung des Heilkorpers aus d.
Diphtherieheilserum," Ze.it. f. Hyg., xxxi., 429, 1899.

98 TOXINES AND ANTITOXINES.

antitoxine precipitated together with the zinc on the addition of alkali
carbonate or phosphate, though it was carried down with the zinc
hydroxide on the addition of an alkali hydroxide. From this precipitate
it could be extracted by means of an excess of very dilute alkali, of only
sufficient strength to just colour phenolphthalein. A solution of baryta
could not be used since it destroyed the antitoxine. Unsatisfactory
results were also obtained with iron salts. The solution was then
saturated at the ordinary temperature with magnesium sulphate, which
precipitated the antitoxine quantitatively. After further extractions with
alkali and dialysis the final product was dried in vacua.

Still better results were obtained by the following method : — The serum
was first treated with a third of its volume of a 5 per cent, solution of
alum. This precipitated the albumins, leaving the whole of the antitoxine
in solution. The solution was then filtered and dialysed. The globulins,
and with them the antitoxine, were precipitated from the dialysate by
semi-saturation with ammonium sulphate, and the precipitate washed
with a semi-saturated solution of ammonium sulphate. The precipitate
was then redissolved in water and the solution dialysed, concentrated
in vacuo, and filtered.

In this way they obtained from 500 c.c. of serum 9 grms. of a dry
preparation in the form of a reddish -brown gelatinous mass. This dis-
solved slowly in water (4'7 grms. in 16 c.c.) yielding a syrupy reddish-
brown fluid, which readily became turbid and could only be filtered slowly.
The filtrate contained the same proportion of antitoxine as the unfiltered
preparation. The addition of phenol was not injurious. The preparation
behaved like a globulin.

ASTROS and RiETSCH1 claim to have separated the antitoxine
from the serum almost quantitatively.

They diluted the therapeutic serum with four parts of water, and then
added sufficient sodium chloride and potassium chloride to give a 20 per
cent, solution, after which 0'5 per cent, of phenol was added, and the
liquid left for twenty-four hours at 33° C.

The theoretically important experiments of PICK may pos-
sibly be also of practical importance. He showed that by
saturating the liquid with ammonium sulphate, to the extent
of about a third, a part of the globulins without any thera-
peutic value was precipitated, whilst the active anti-bodies
were separated when the degree of saturation reached 38 to
46 per cent. By this means he effected a tenfold to fifteenfold
concentration. We will deal presently with the significance of
his work.

Equally important, from the practical point of view, are the
experiments of PROSCHER,2 who claims to have freed the anti-
toxines from proteid admixtures by digestion with trypsin.

1 Astros and Rietsch, "Essais d'Extraction de 1'antitox. diphth. " Soc.
BioL, lii., 337, 1900.

2 Proscher, German Patent, F. 13,756, June 6, 1902.

DIPHTHERIA TOXINE. 99

Especial interest attaches to the question whether the anti-
toxine circulates as such in the blood, and, if so, whether it is
a proteid or not, or whether it enters into combination with a
definite proteid of the blood. This also involves the question
whether its precipitation together with proteid substances is
simply a mechanical carrying down, or whether the precipitation
is due to chemical reactions of a specific nature.

It is undoubtedly very probable, a priori, that the antitoxine
is a proteid, since according to the theory it is a constituent of
the cell protoplasm. There is also absolutely no ground, as in
the case of toxines, for regarding the antitoxine as an agency
of enzymic character. On the contrary, it has absolutely no
action by itself, either toxic or fermentative, and only enters
into combination with the toxic ferment of bacteria, and it is
a quite unwarranted generalisation to transfer the enzymic
character of toxines to their antitoxines without further proof.
The antitoxine lacks, to retain the graphic mode of representa-
tion, every specific group with the exception of the haptophore
group, and thus has neither a toxophore nor a zymophore group;
it is indeed a haptine, but one that possesses only one, the
haptophore, group. In point of fact, there are many reasons
against the antitoxine being of an enzymic character. The
only real argument in support of that view is the fact that it
becomes inactive at 100° C., and that is surely not a sufficient
reason. Against this, and in support of the view that it is a
specific proteid, must be set the fact that it is not simply carried
down mechanically, like the enzymes, but that it shows well-
marked precipitation reactions. As has already been shown by
BRIEGER and BOER (loc. cit.), the antitoxine is only carried
down, together with zinc carbonate, when it has previously
been precipitated by means of zinc sulphate, and not when
zinc chloride has been used for the purpose. In like manner
FREUND and STERNBERG were able to demonstrate that the
antitoxine was not precipitated simultaneously with the slight
precipitates produced by alum in the serum, although these
carried down all albumins.

It would appear, then, that the antitoxine either actually
enters into combination with one of the proteids of the blood,
or we must assume that the particular proteid of which the
antitoxine consists approximates so closely in its reactions to
a proteid of the serum that it is very difficult to distinguish
"between them.

Quite recently, however, PROSCHER (loc. cit.) has stated that
he has prepared a diphtheria antitoxine which no longer shows

100 TOXINES AND ANTITOXINES.

any proteid reactions at all. The further development of this
question will be awaited with the greatest interest. The other
question as to which of the serum proteids the antitoxine is
combined with has been decided in favour of the globulins.

BELFANTI and CARBONE x were the first to discover that the antitoxine
was carried down with the globulins when they were precipitated by means
of ammonium or magnesium sulphate, but not by acetic acid. The anti-
toxine was also found by SniRNOW2 in the globulin precipitate thrown
down by magnesium sulphate.

On the other hand, DIEUDONNE 3 found that the globulins which he had
precipitated by means of acetic acid and carbon dioxide were completely
nactive. On dialysis, the separated globulins exhibited but little pro-
tective power and that was mainly due to the filtrate. If, however,
magnesium sulphate were used for the precipitation the whole of the
antitoxine remained in the precipitate. A very similar result was ob-
tained in the case of the globulins of normal serum, which also had a
slight antitoxic power.

Again, Hiss and ATKINSON,* and also ATKINSON,S found that the anti-
toxine could be quantitatively precipitated by magnesium sulphate, and
that the immune serum yielded a more copious precipitate than normal
serum.

IDE and LEMAIRE 6 assert that the antitoxine is precipitated by saturating
its solution with ammonium sulphate to the extent of 28 to 44 per cent.

These apparent contradictions as to the role of the globulins
have been satisfactorily explained by SENG.7 He was able to
show (as was also done simultaneously by MARCUS 8) that there
were two kinds of globulins in the therapeutic serum — viz., in-
soluble globulins, which could be precipitated by acetic acid,
carbon dioxide, dilution with water, and dialysis, and a second
category, the soluble globulins, which could only be precipitated

1 Belfanti and Carbone, " Contributo alia Conoscenza dell'antitossina
difterica," Arch, per le Scienze Med., xxii., No. 2; abst. in Centralbl. f.
BaU., xxiii., 1898.

2 Smirnow, "Note sur la de*termin. du pouvoir antitoxique du serum
antidiphthe'rique," Arch. d. Science Biolog. de St. Petersbourg, iv., No. 3,
1895.

3 Dieudonne, " Ueb. Diphtheriegift neutralisier. Wirkg. der Serum-
globuline," Arb. a. d. Kaiserl. Ges.-Amt., xiii., 293, 1897.

4 Hiss and Atkinson, " Serum globulin and Diphtheria Antitoxine," J. of
Exper. Med., v., 47, 1901.

5 Atkinson, "The fractional precipitation of the globulins and albumins
of normal horse serum and diphtheric antitoxic serum," /. of Exper. Med.,
v., 67, 1901.

6Ide and Lemaire, "Et. s. 1. repartition de 1'antitoxine diphth., &c.,"
Arch. Internat. d. Pharmacodyn., vi., 477.

7 Seng, " Ueb. d. qual. u. quant. Verhaltnisse d. Eiweisskorper im Diph-
therieheilserum," Ze.it. f. Hyg., xxxi., 513, 1899.

8 Marcus, " Ueb. in Wasser losl. Serum globulin," Z. f. phys. Ch,, xxviii.,
559, 1899.

DIPHTHERIA TOXINE. 101

by the other reagents for globulins, especially anatrionium and
magnesium sulphates. The antitoxine combined exclusively
with the latter class, and thus remained in th£, filtrate tl &nng
the precipitation by dialysis, carbon dioxide;-£cV ' !£his- explain*;
the contradictory results in the work of BELFANTI and CAKBONE
and of DIEUDONNE.

SENG proceeded as follows in the isolation of his "soluble
globulins " : —

The albumins (which do not combine with any antitoxine) are first
separated by means of a 5 per cent, solution of alum. The filtrate is
dialysed, and this causes the precipitation of slight quantities of the
insoluble globulins, which are also free from antitoxine and amount to
TT to -£% of the total quantity of globulins present. This precipitate is
collected and washed with water, and the filtrate ought not to become
turbid on the dilution with a large volume of water, or give any reaction
for sulphuric acid. The globulin is then precipitated by magnesium sul-
phate at 30° C., or by semi-saturation with ammonium sulphate, and the
precipitate redissolved and subjected to further treatment by the usual
methods (vide supra). It contains somewhat more ash, notably aluminium
salts, than the globulins precipitated directly from the serum by means of
ammonium sulphate.

The quantity of these soluble globulins, as compared with those
of the normal serum, appears to have been somewhat increased
at the cost of the insoluble globulins, but this cannot be demon-
strated with certainty.

SENG has also studied the very important question whether chemical
differences can be observed between the soluble globulins of the therapeutic
serum, as thus obtained, and those of normal serum separated in the same
way. As was to be expected, he has not succeeded in coming to a definite
conclusion, but it appears as though at least two important constants —
viz. , the specific rotation and the temperature of coagulation— are higher in
the case of the antitoxine globulin than in that of normal globulin. SENG
himself remarks, very justifiably, that the physiological bearing of all these
factors is far too great to allow of distinct differences being observed be-
tween the sera of different animals ; it would be necessary to examine the
serum of one animal before, during, and after immunisation. Some definite
result might then possibly be obtained.

Meanwhile this question has received further thorough inves-
tigation in a series of researches carried out by Hofmeister's
pupils.

It was found that three distinct kinds of globulins could be
separated from the blood serum by fractional precipitation with
ammonium sulphate. In the first place, jibrinoglobulin is precipi-
tated when the saturation with ammonium sulphate reached 21-5
per cent, by volume ; then, according to FULD and SpiRO,1 Hof-

1 Fuld and Spiro, " Ueb. labende u. labhemmende Wirkung d. Blutes,"
Z.f. phys. Ch., xxxi., 132, 1900.

102 TOXINES AND ANTITOXINES.

xneister's so-called euglobulins are precipitated when the degree of
saturation reaches 28 to 36 per cent. ; whilst the pseudoglobulins
are deposited at a saturation of 36 to 44 per cent, by volume.
The -pseudoglobuiinS) which dissolve in water forming a clear
solution, correspond to the "soluble globulins" of MARCUS and
SENG.

PiCK1 was able to prove that neither the fibrinoglobulin, nor those pro-
teids which were precipitated on the further addition of a saturated solution
of ammonium sulphate (up to 36 per cent. ), possessed any antitoxic capacity
when separated from the antidiphtheric serum of a horse.

Only that portion which is precipitated at a greater degree of
saturation (from 38 per cent, upwards) contains antitoxine.

The aqueous solution of this substance gave only a slight turbidity when
treated with a saturated solution of ammonium sulphate to the extent of
36 per cent., whereas at 38 per cent, there was a dense deposit containing
a third to a fourth of the total quantity of antitoxine ; at 42 per cent,
the precipitate contained as much as five-sixths ; and at 46 per cent, the
whole amount of antitoxine. The filtrate, which now contained only
serum albumin, had absolutely no protective power.

Hence, in the case of the immune serum of the horse, the
antitoxine is combined with the pseudoylobulin. On the other
hand, in the immune serum of the goat, it is combined with the
euglobulin, as is also the case with goats' milk in which the pro-
tective substances are precipitated when the degree of saturation
with ammonium sulphate reaches 27 to 30 per cent.

We may again point out here that the euglobvlins are insoluble in water,
and are therefore precipitated on dialysis ; this explains the great losses
experienced by WASSERMANN and BRIEGER and COHN in their attempts to
concentrate the antitoxine in goats' milk by precipitation with ammonium
sulphate and subsequent dialysis.

TETANUS TOXINE.

The specific spasm-producing poison formed in cultivations by
Nicolaier's tetanus bacilli is not the only toxine of these micro-
organisms. They produce, in addition to tetanus poison in the
narrower sense, also another true toxine, tetanolysine, which has
a solvent action on the blood-corpuscles, and with which we
will deal presently, and possibly also other poisons not of the
nature of toxines. Here our interest centres chiefly about the
characteristic convulsion-producing poison, tetanospasmine. It is
unquestionably a true toxine with haptophore and toxophore

1 Pick, "Z. Kenntnis d. Immunkbrper, " Hofm. Beitr. z. chem. Phys.,
i. , 1902 (reprint).

TETANUS TOXINE. 103

groups, and is closely related to diphtheria toxine in its main
characteristics, and, like the latter, is very valuable for our
theoretical conceptions.

The specific convulsion-producing poison of tetanus was isolated
at an early period from the cultivations by the same methods as
were employed for diphtheria poison. Almost simultaneously
KITASATO and WEYL1 and BRIEGER and FRANKEL (loc. cit.)
separated from cultivations their toxalbumin which they recog-
nised as the specific poison, soluble in water. FABER 2 was the
first to obtain, by filtration of not quite pure cultivations, an
active poison, which he found even at that stage to possess the
essential properties of tetanus toxine.

The results of further investigations into tetanus toxine were
then published by TIZZONI and CATTANi,3 KITASATO,* and by
VAILLARD 5 and his co-workers.

The bacteria, when grown anaerobically on ordinary bouillon
and on blood serum, produce very poisonous toxines. Particu-
larly active solutions of poison were obtained by VAILLARD and
VINCENT, when they first allowed a cultivation to grow for
eighteen days on a culture medium and then filtered it. Fresh
inoculation of this culture medium was fruitless, but by adding
a certain amount of fresh nutrient liquid, a new growth of very
active toxines was obtained. VAILLARD subsequently employed
a slightly alkaline bouillon containing some peptone ; KITASATO
recommended that the culture medium should be freshly prepared
on each occasion. FERMI and PERNOSSIG found that agar
cultivations were the most poisonous, especially so when grown
in an atmosphere of nitrogen.

BRIEGER and COHN 7 found that the toxicity of the cultivations
could be raised by adding to the culture media precipitates
obtained by means of alcohol from old typhus cultivations or

1 Kitasato and Weyl, "Zur Kenntnis der Anaerobien," Zeit. f. Hyg.,
viii., 404, 1890.

2 Faber, "Die Pathogenese des Tetanus," Berl. Uin. Woch., 1890, 717.

3 Tizzoni and Cattani, "Sur le poison du tetanos," Arch. Ital. d. Biol.,
xiv., 101, 1890.

4 Kitasato, "Exper. Unters. iiber das Tetanusgift," Zeit. f. Hyg., x.,
287, 1891.

5 Vaillard, " Sur 1'immunite contre le t6tanos," Soc. Biol., xliii., 147,
1891 ; Vaillard and Vincent, " Contrib. a 1'gtude du tgtanos," Ann. Past.,
v., 1, 1891 ; Vaillard and Rouget, do., ibid., 385 ; Vaillard, " Sur quelques
points concernant 1'immunite contre le te'tanos," Ann. Past., vi., 224, 1892.

6 Fermi and Pernossi, "Ueber das Tetanusgift," Zeit. f. Hyg., xvi.,
385, 1894.

7 Brieger and Cohn, "Unters. iib. d. Tetanusgift," Zeit.f. Hyg., xv., 1,
1893.

104 TOXINES AND ANTITOXINES.

from decomposed flesh. WLADIMIROFF 1 cultivated the bacilli in
Erlenmeyer flasks in an atmosphere of hydrogen, and after seven
days added to the culture medium 0*5 per cent, of phenol. He
introduced the hydrogen through tubes, which passed through
two openings in an india-rubber stopper, and the ends of these
tubes were subsequently closed by fusion. DEBRAND2 asserts
that tetanus bacilli grow very well in symbiosis with B. subtilis,
and produce powerful toxines. In the case of these bacilli also
it has been found that there is not invariably a parallelism
between the amount of growth and the toxicity.

RUPPELL and RANSOM 3 found that the freezing point of the
bouillon was somewhat lowered on the production of toxine,
which indicates a fresh formation of molecules; and, as the
cultivation becomes weaker, the freezing point ought to be
slightly raised again.

USCHINSKY 4 cultivated tetanus bacilli on a proteid-free medium of the
following composition : —

Water, lOOO'O

Glycerin,

Sodium chloride,

Calcium , ,

Magnesium sulphate,

Dipotassium hydrogen phosphate,

Ammonium lactate,

Sodium aspartate, .

30 to 40

5 „ 7

O'l

0-2 to 0-4
2 „ 2-5

6 „ 7

3-4

preferably with the addition of 1 to 2 per cent, of grape sugar. The air
was excluded by means of liquid paraffin.

The usual means of sterilisation are employed, especially nitra-
tion through a Chamberland filter, and heating to 60° C.

Tetanus toxine is extraordinarily sensitive to physical and
chemical influences. According to BEHRING and KNORR5 it
frequently loses very rapidly a considerable proportion of its
toxic power on keeping ; its toxicity is frequently reduced to
a hundredth part of the original amount in a few days.

The atmospheric oxygen, in particular, has a very injurious

1 Wladimiroff, " Antitoxinerzeug. d. Tet. -Giftes," Zeit.f. Hyg., xv., 405,
1893.

2 Debrand, "Sur un nouveau precede de culture du bacille du te'tanos,"
Ann. Past., xiv., 757, 1900.

3 Ruppel and Ransom, "Ueb. Molekularverhaltnisse von Tetanusgift-
losungen," Z. f. phys. Ch., xxvii., 109, 1899.

4 Uschinsky, "Ueber eine eiweissfreie Nahrlosung f. pathog. Bakt.,"
CentralU.f. Bakt., xiv., 316, 1893.

5 Behring and Knorr, " Ueb. den Immunisierungswert des Tetanus-
heilserums," Zeit.f. Hyg., xiii., 407, 1893.

TETANUS TOXINE. 105

effect upon the toxine, so that when slowly filtered, for example,
it is very rapidly weakened ; and this is especially the case with
an alkaline solution and in the presence of light, which, by itself,
is not very injurious. KITASATO found that sunlight destroyed
the poison in fifteen to eighteen hours. FERMI and PEENOSSI
give the time as eight hours, and assert that in the dry condition
or in benzene it is not affected by sunlight.

A current of 0'5 amp. is stated by FERMI and PERNOSSI to
destroy it in two hours, but alternating currents of high tension
have no influence if heating be prevented (MARMIER 1).

It is very sensitive to the action of heat. According to
VAILLARD it is destroyed, for the most part, even at 65° C., but
not completely at 80° C. KITASATO asserts that it is destroyed
at 60° C. in twenty minutes, at 55° C. in one and a half hours,
and gradually destroyed at 35° to 37° C. In particular, the
addition of sodium chloride in the proportion of more than 5 per
cent, causes the temperature of incubation to be rapidly injurious.
In the dry condition the toxine is destroyed in thirty minutes at
150° C. ; in amyl alcohol and benzene it can be heated for an
hour at 80° C. (FERMI and PERNOSSI). According to MORAX and
MARIE2 it resists a temperature of 154° C. for about fifteen
minutes. Alcohol destroys it (TizzoNi and CATTANI). Phenol
in the proportion of 0*6 per cent, is not injurious, and chloroform
has little or no influence.

Gases, such as carbon dioxide, carbon monoxide, hydrogen,
and hydrogen sulphide, were found by FERMI and PERNOSSI to
be without influence.

According to Roux and VAILLARD 3 oxidising substances,
such as dilute potassium permanganate solution, are specially
injurious, as is also carbon dioxide under pressure.

Other acids are also injurious, tartaric acid to a very slight
extent, whilst dilute lactic acid has a favourable influence
(BRIEGER 4). A comprehensive research on the influence of the
most diverse substances has been made by FERMI and PERNOSSI
(loc. cit.).

Iodine trichloride has also an extremely injurious influence.
The action of a dilute (1 : 500) solution of iodine is very char-
acteristic. Even when only present in small quantities it

1 Marmier, "Les toxines et Telectricite," Ann. Past., x., 469, 1896.

2 Morax and Marie, "Action de la Chaleur seche sur la tox. let.," Ann.
Past., xvi., 418, 1902.

3 Roux and Vaillard, " Contrib. & 1'etude du tetanos " Ann. Past., vii.,
65, 1893.

4Brieger, "Weitere Erfahrungen iiber Bakteriengifte," Zeit. f. Hyg.,
xix., 101, 1895.

106 TOXINES AND ANTITOXINES.

speedily renders the toxine non-poisonous, but leaves its im-
munising properties uninjured. Very similar results were
obtained by EHRLICH in experiments with carbon bisulphide.
It can hardly be doubted that there is here a rapid destruction
of the toxophore group with a survival of the haptophore group,
or in other words a, formation oftoxoid. Thymus extract appears
to have a similar result upon the growth, according to BRIEGER,
KITASATO, and WASSERMANN (loc. cit.).

It appears to offer great resistance to putrefaction ; at all
events, SYMANSKI l found tetanus poison still present in decom-
posing cadaveric remains after forty-eight days. There is,
however, some reason for doubting whether this was true
tetanus poison.

Concentration of the Toxine. — Experiments with the object
of isolating the active principle from tetanus cultivations were
made at an early period.

BRIEGER and FRANKEL prepared a "toxalbumin" by the method de-
scribed by them. VAILLARD obtained by evaporation of the solution of
poison in vacua a brown residue, which was insoluble in alcohol and could
be slowly dialysed.

TIZZONI and CATTANI either simply dried the cultivations and
then dialysed them, or treated them with ammonium sulphate,
extracted the precipitate with water, and then used dialysis.
Subsequent evaporation in vacua then left solid toxines.

BRIEGER and COHN 2 first treated the cultures with ammonium
sulphate. The precipitate was dissolved in water and part of
the proteids separated by means of very small quantities of basic
lead acetate and ammonia. The liquid was filtered from the
lead precipitate and salts, amino-acids and peptones removed
by dialysis. In this way they obtained a slightly Isevorotatory
solution of the toxine, which was free from sulphur and gave no
proteid reactions. They also prepared it from a proteid-free
culture medium similar to that of USCHINSKY (vide supra).

Then BRIEGER 3 found that the poison could not be precipitated
by means of ammonium sulphate from very toxic solutions which
no longer contained albumoses. He endeavoured to effect a
further purification by precipitation with uranium acetate and
decomposition with metaphosphoric acid or lead acetate and
obtained preparations that no longer gave the biuret reaction.

iSymanski, " Sitzungsbericht," Deutsch. med. Woch., 1901, 318.

2Brieger and Cohn, "Unters. lib. d. Tetanusgift," Zeit. /. Hyg.t xv.,
1, 1893.

3Brieger, "Weit. Erfahrung. lib. Bakteriengifte," Zeit. f. Hyg., xix.,
101, 1895.

TETANUS TOXINE. 107

BRIEGER and BOER (loc. cit.) also made further experiments to
obtain purer preparations by means of the zinc method described
in the section on diphtheria toxine, or they used ammonium
sulphate for the precipitation, dissolved the precipitate and
reprecipitated the toxine with an equal volume of a O5 per
cent, solution of mercuric chloride. They then collected the
precipitate on a filter, thoroughly washed it, and treated it with
ammonium carbonate as previously described. HAYASHI 1 by a
somewhat modified method (precipitation first with ammonium
sulphate and then with zinc chloride) has obtained preparations
which, in his opinion, no longer contained non-poisonous albu-
moses. Hence, he concludes that since these preparations are
invariably precipitated by ammonium sulphate and give the
biuret and Millon's reactions, tetanus poison itself must be an
albumose.

Owing to the very poisonous effects which tetanus toxine,
even in the smallest doses, has upon certain species of animals,
special stress has been laid upon the analogy that exists between
that toxine and the ferments. In point of fact there are many
reasons in support of that view. VAILLARD definitely terms it
a ferment, and other authors, such as TIZZONI, BRIEGER, &c.,
incline to this view, which as a matter of fact has only been
opposed, and that not very strongly, by FERMI in his various
published researches. In my opinion we are justified in con-
cluding that there is a considerable analogy between the toxines
and also some other haptines and the ferments, and I have
advocated this view in several places.2

There is no doubt that rennet is a haptine, but, as regards the
other ferments, there are no data for settling their precise affini-
ties. But there is absolutely no justification for the assumption
that tetanus toxine acts like a ferment in the sense of affecting
the enzymic decomposition of substances, for tetanus toxine is
only akin to a ferment in the mode, and not in the scope, of its
action. It does not effect the changes in starch, cane sugar, &c.,
brought about by enzymes. It is true that VAILLARD found a
gelatin-liquefying enzyme in virulent cultivations, but many
micro-organisms, pathogenic as well as non-pathogenic, produce
similar proteolytic and other enzymes.

Toxoids and Toxones, — The question of the presence of such

1 Hayashi, " Ueber die chemische Natur des Tetanustoxin," Arch. f.
exp. PathoL, xlvii., 9, and Chem. CentralbL, i., 411, 1901.

2 Oppenheimer, "Toxine u. Schutzstoife," Biol. CentralbL, 1899, 799;
id., Ferments and their Actions, London, 1901; id., " Zur Theorie der
Fermentprozesse," Munch, med. Woch., 1901.

108 TOXINES AND ANTITOXINES.

non-poisonous haptines has as yet hardly been systematically
investigated. The haptophore group appears to undergo a
simultaneous change during the extraordinarily rapid decrease
in strength that occurs in fresh poisons ; but at the same time it
has not yet been proved beyond doubt that there can be a diminu-
tion in the toxic activity while the neutralisation value remains
constant. All observers, however, have recorded differences
between the direct and the indirect toxic value (i.e., that required
to neutralise a corresponding quantity of antitoxine), which point
to the existence of non-poisonous haptines.

BEHRING,1 in particular, has called attention to this fact. He
found that in the case of fresh poisons the direct toxic value was
the same as the indirect value (expressed in terms of T^Vo of an
antitoxine unit). Hence he named these "equivalent poisons."
On the other hand, in older cultivations he invariably found the
direct toxic value was smaller — i.e., toxoids had obviously been
formed, and had neutralised part of the antitoxine, so that it was
necessary to add fewer toxic units to reach L0 than in the case of
fresh poisons containing only toxine. Such poisons containing
toxoids are well adapted for immunising purposes, as has been
stated by BRIEGER, and is easily explicable. The toxoids appear
to have different toxic effects upon different animals ; at all
events, the poison scale approximates very closely to that of the
equivalent poisons (vide supra) in the case of certain species.

The existence of toxones appears probable in the light of the
experiments of KNORR,2 who found, for example, that on ap-
proximate neutralisation with antitoxine (and therefore in the
" differential zones ") an added quantity of poison had a consider-
ably smaller toxic value than its corresponding direct toxic value,
just as in the case of diphtheria poison far more than a lethal
dose is required to convert the value L0 into L+. He himself,
however, appears to reject this conclusion, and to account for the
fact in a different way.

Physiological Action of Tetanus Toxine. — Tetanospasmine
varies very greatly in its action upon different species of
animals, especially with subcutaneous injection.

In the case of very susceptible animals, its toxic effect is
astonishingly great when it is introduced subcutaneously in the
usual way. VAILLARD prepared a solution, O'OOl c.c. of which
killed a guinea-pig. This dose contained about 0*000025 grin, of

1 Behring, " Ueber Tetanusgiftmodifikationen," Fortschr. d. Med., xvii.,
501,1899.

2 Knorr, " Die Entstehung des Tetanus- Antitoxins," Fortschr. d. Med.,
1897, 657.

TETANUS TOXINE. 109

organic substance — i.e., only a very small trace of pure toxine.
The organic matter required for a mouse was only 0*000000025
grm. According to BRIEGER and COHN (loc. cit.}, the lethal dose
of a far from pure toxine was about 0-0000005 grm. for a mouse,
and 0 '00023 grm. for a man, but occasionally still more poisonous
preparations have been obtained. The individual variations are
also very great. According to BEHRING and KNORR, the certain
lethal dose for a mouse is about six times as great as that which
is just below that required to kill. Still greater, however, are
the variations in the toxicity due to the susceptibility of the
different species of animals.

KNORR,1 in his paper on the subject, gives a scale of suscepti-
bility to tetanus poison. He finds the horse to be the most
susceptible animal. If we take the dose that kills 1 grm. of
horse as unity, the amounts of poison required by the following
animals are as follows : —

1 grm
1 ,

1 ',

1

of guinea-pig,
goat, .
mouse,
rabbit,
hen, .

2 times as much.
4 ,,

13

2,000
200,000

BEHRING2 gives the following scale for fresh poisons, which
show no difference between the direct and indirect toxic value
(equivalent, poisons) : —

One lethal dose for 1 grm. of mouse ( + Ms) kills

12 grms. of horse.
6 „ guinea-pig.
0'2 grm. of goat.
Titf » rabbit.

The numbers thus agree fairly well. In the case of old
poisons, in which, according to BEHRING, the direct toxic value
( + Ms) is much smaller than the indirect value ( + ms) — i.e., the
relative value for the quantity which, after neutralisation of T^OQ-
an ti toxine unit, is still lethal (xoW an ti toxine units = 40,000
Ms), are quite irregular. In like manner the ratios between the
smallest dose that causes illness and the smallest fatal dose vary
very greatly with the species of animal.

1 Knorr, " Das Tetanusgift und seine Beziehg. zum tier. Organismus,"
Munch, med. Woch., 1898, 321, 362.

2 Behring, "Ueber Tetanusgiftmodifikationen," Fortschr. d. Med.,
xvii., 501, 1899.

110 TOXINES AND ANTITOXINES.

The toxic effect is generally considerably increased when the
poison is introduced by subdural or intercerebral injection.
Thus, for example, the otherwise comparatively refractory hen
can be poisoned fairly readily by the intercerebral injection of
tetanus poison. We will deal with the probable cause of this on
a later page.

The phenomena on intravenous injection are the normal ones,
but the incubation period is somewhat longer. When introduced
into the stomach tetanus poison is practically non-poisonous.
RANSOM 1 concluded that it passed unchanged through the in-
testines, and claimed to have found it again in the faeces.
CARRIERS, however, and numerous other investigators were
unable to detect any toxine in the excreta after the introduction
of large doses of toxine per os. CARRIERS 2 accordingly made
experiments to determine where tetanus toxine becomes in-
nocuous. He found that the poison was attacked even by the
saliva diastase, that pepsin was less injurious, that trypsin had
a considerable action upon it, and that bile in large quantity
completely destroyed it. He could not detect any influence of
the intestinal mucous membrane and the intestinal bacteria upon
the poison, although FERMI and PERNOSSI had found it to be
very pronounced. NENCKI and SCHOUMOW-SIMANOWSKI 3 assert
that the digestive fluids render it completely innocuous, and
notably the bile in conjunction with the fluid from a pancreatic
fistula, whilst trypsin alone has less effect, being less injurious
than pepsin. YiNCENZi,4 on the other hand, states that normal
bile has hardly any destructive influence, but that the bile of
animals infected with tetanus has a slight effect under certain
conditions ; thus, for example, death inevitably follows in three
to four days when the bile is active. On the other hand, the
poison is somewhat weakened by the oxydase of the leucocytes,
and, according to SiEBER,5 is completely destroyed by the
oxydase of the spleen.

It is absorbed with great rapidity on subcutaneous injection.

1 Ransom, "Das Schicksal d. Tetanusgiftes nach seiner intestinal
Einverleibung," Deutsch. med. Woch., 1898, 117.

2 Carriere, "Toxines et digestion," Ann. Past., xiii., 435, 1899 (gives a
bibliography of the subject) ; cf. the General Part.

3 Nencki and Schoumow-Simanowski, "Ueber die Entgiftung der Toxine
durch die Verdauungsafte," Centralbl. f. Bakt., xxiv., 84 ; cf. Dzierzgowski
and Sieber, Archiv. des Sciences Biol. de St. Petersb., viii.

4 Vincenzi, " Ueb. antitoxische Eigenschaften der Galle tetanisierter
Tiere," Deutsch. med. Woch., 1898, 534.

5 Sieber, "Ueb. d. Entgiftung der Toxine durch die Superoxyde, &c.,"
Z. f. phys. Ch., xxxii., 573, 1901.

TETANUS TOXINE. Ill

It has been established by numerous experiments that a rat into
whose tail the toxine has been injected can no longer be saved
by amputation of the tail even after two to three hours.

At the same time tetanus poison shows an unmistakable
period of incubation. COURMONT and DoYON1 state that, in the
case of guinea-pigs, the symptoms appear at the earliest after
twelve hours, and in mice after six to eight hours. With
medium doses the latent period for mice is two to three days ;
for guinea-pigs, two days ; rabbits, two to four days ; the ass,
four days ; and for the horse five days. In the case of man it
fluctuates between one day and sixty days. The action is much
more rapid after subdural injection (BLUMENTHAL and JACOB 2),
or intercerebral poisoning (Roux and BORREL 3), as well as after
direct injection into the spinal cord (MEYER and RANSOM 4).

According to a table given by MEYER and RANSOM the periods
of incubation after subcutaneous injection are as follows : —

Mouse, . . . . . . 8 to 12 hours.

Guinea-pig, 13 ,, 18 ,,

Rabbit 18 „ 36 „

Cat, 28 „ 70 „

Dog, 36 „ 48 „

Man, ...... 4 days.

Ass, 4 ,,

Horse, 5 ,,

Hence it increases with the size of the body, and thus stands
in relationship to the slowness with which the poison is diffused
(vide infra).

The incubation period becomes smaller with the increase of
the dose, but not proportionally so, and still occurs even after
the largest doses. In the case of mice it never falls below eight
hours. Thus, in one series of experiments, the incubation
periods for mice were as follows : —

Lethal Doses

Hours.

13, . . .

36

100, . . .

24

333,

20

1,300, . . .

14

3,600,

12

1 Courmont and Doyon, vide supra.

2 Blumenthal and Jacob, "Zur Serumtherapie des Tetanus," Berl. klin.
Woch., 1898, 1079.

3 Roux and Borrel, "Tetanos cerebral," Ann. Past., xii., 1898.

4 Meyer and Ransom, " Unters. lib. d. Tetanus," Arch. exp. Path.,
xlvi., 369, 1903.

112 TOXINES AND ANTITOXINES.

Tetanus poison is therefore rapidly absorbed without at once
manifesting its activity. Yet, even a short time after inocula-
tion, free toxine can no longer be detected in the organism of
susceptible animals. On the contrary, it has, for the most part,
entered into combination with the receptors. Only in the blood,
and, to a less extent, in the lymph, can toxine be detected after
the injection (RANSOM 1), and in the case of the lymph for an
hour afterwards at most.

With these exceptions it cannot be found in any of the organs
or secretions (MAKIE2). BRUSSCHETTINI 3 alone claims to have
detected it in the kidneys, whence, according to him, it is ex-
creted. In like manner it has been found by various observers
in the urine. BRUNNER* detected it in the urine of animals
used in experiments, but not in that of infected men, and his
results were confirmed by BEHRiNG.5 KARTULis6 could only
find it in the urine when very large doses had been given. It is
usually vainly sought in the urine of men suffering from tetanus,
though VULPINS 7 found the urine left in the bladder after death
so poisonous that 2 c.c. rapidly infected a guinea-pig.

In the case of refractory animals the toxine circulates for
a long time in the blood, as we have described in the General
Part, without entering into combination with the organs, until
eventually it slowly disappears.

In addition to its effect upon the central nervous system,
which is the absolutely dominating symptom of tetanus poison-
ing, and with which we deal at length below, tetanospasmine
also produces certain general symptoms, which can only be briefly
mentioned, since it can be no part of our task to give a complete
description of the pathology of tetanus.

The pressure of the blood remains unchanged, as has been
found by HANS MEYER and HALSEY,S contrary to the results
of former investigators, and in this respect the symptoms differ
from those of strychnine poisoning which is otherwise so similar.

1 Ransom, " Die Lymphe nach intravenoser Inj. von T.-T.," Z.f. phys-
Ch., xxix., 349, 1900.

2 Marie, "Rech. s. 1. toxine tetanique," Ann. Past., xi., 591, 1897.

3 Brusschettini, quoting from Brunner, loc. cit.

4Brunner, "Klin, und exp. Stud. lib. Tet.," Beitr. z. klin. Chirurg., ix.
to xii., 1892-4.

5 Behring, " Die Blutserumtherapie," iii., 54, Leipzig, 1892.

6 Kartulis," Unters. iiber das Verhalten des Tet.-Giftes im KSrper,"
Diss., Berlin, 1892 (Dec.)

7 Vulpins, "Ueb. einen Fall von Wundstarrkrampf m. Tiervers.,"
Deutsch. med. Woch., 1893, 992.

8 Hans Meyer and Halsey, " Tetanusstudien," Festschr. f. Jajft, Bruns-
wick, 1901.

TETANUS TOXINE. 113

Eventually, infected animals show a marked reduction in tem-
perature and severe inanition before death (HARNACK and
HOCHHEIM 1).

Effect upon the Central Nervous System. — With the exception
of these few pronounced general symptoms, tetanus and also
the completely identical poisoning with tetanospasmine are
characterised in the case of most animals by the predominance
of symptoms due to an affection of the central nervous system.
It would appear then that the receptors which combine with the
poison and thus initiate the disease only occur in the nervous
system. There are only a few facts that have been urged against
the view that this is the only point of attack for the poison.
Thus VAILLARD, using very small doses of toxine, claims to have
observed local poisoning of the neighbouring muscular tissue.
The rabbit also occupies an exceptional position, since under
certain conditions the poison can be fixed in other ways to an
overwhelming extent, so that the animal dies of " tetanus sine
tetano" (DoNiTz). This is a particularly striking instance of
the observation that has frequently been made, that the poison
can combine with receptors situated in organs of little vital
importance, so that the fixation of the poison does not lead to
any predominating symptoms of illness. We shall return to
this phenomenon presently.

The direct symptoms, the combination of the poison with the
substance of the central nervous system, were first observed by
SHAKESPEARE2 and YERHOOGEN and BAERT,S who poisoned
animals by the subdural injection of material from the central
nervous system of tetanised animals.

BESEEDKA 4 in repeating this experiment left the normal brain
substance of a guinea-pig in contact with a large quantity of
toxine for a considerable time (up to six days) in an ice chest,
and then removed the excess of free toxine by careful washing.
He was able to produce tetanus in mice by means of this
substance.

PASQUINI 5 also found the central nervous system to be
poisonous.

1Harnack and Hochheim, " Ueb. Wirkg. d. Briegerschen Tet.-G.,"
Z.f. Jclin. Med., xxv., 46, 1894.

2 Shakespeare, "Preliminary report on exper. researches concerning the
infectious nature of traumatic tetanus," Centralbl. f. Bakt., ii., 541, 1887.

3 Verhoogen and Baert, "Premieres recherches sur la nature du te"tanos,"
Baumg. Jahresber., 1890, 198.

4Besredka, "De la fixation de la tox. te*t. par le cerveau," Ann. Past.,
xvii., 138, 1903.

5 Pasquini, " Sulla presenza del veneno tetanico negli organi," Rif, Med.,
1902, 22, 23 ; Abst. in Centralbl. f. Bakt,, xxxi,, 117.

8

114 TOXINES AND ANTITOXINES.

GOLDSCHEIDER and FLATAU,1 JouKOWSKY,2 and others have
succeeded in demonstrating directly under the microscope the
action of tetanus toxine upon the ganglionic cells, especially
those of the anterior cornua. Particularly vigorous discussion
has taken place on the question whether the nervous system is
the only point of attack for the poison ; and also whether it is
only the central nervous system that is attacked.

The question appears to have been decided with certainty by
the investigations of GuMPRECHT3 in favour of the view that
the central nervous system, and above all the spinal cord, is
exclusively attacked by the poison, and that the symptoms else-
where are to be attributed to these primary ones. COURMONT
and DoYON4 conclude that the whole of the sensory neuron is
liable to the primary attack.

The statement that the poison attacks primarily only the
central nervous system is apparently not easy to reconcile with
the fact of "local tetanus" The two can only be brought into
agreement by the assumption that there is a direct transmission
of the poison in the nerves from the point of infection to the
spinal cord. This conclusion was first drawn by BRUSSCHETTINIS
who was himself able to detect the poison in the nervous system,
whereas other blood-free organs, including the muscles at the
point of infection did not contain it.

This view has recently received support from the particularly
comprehensive experiments of HANS MEYER and RANSOM.° Their
results were as follows : — After subcutaneous inoculation with
tetanus toxine the poison could be detected in the nerve. This
most important result was simultaneously confirmed by MARIE
and MORAX * who found the poison in the sciatic nerve of one
infected leg, and also in the other after the occurrence of general
tetanus ; it was also found, in particular, in the masseteric nerve
but not in the optic nerve, which is indeed an integral portion
of the brain.

1 Goldscheider and Flatau, "Ueber die Ziele der modernen Nervenzel-
lenforschung," Deutsch. med. Woch., 1898, 165.

2 Joukowsky, "De 1'influence de la toxine te"tan. sur le system nerveux,"
Ann. Past., xiv., 464, 1900.

3Gumprecht, "Zur pathol. d. Tetanus," Deutsch. med. Woch., 1894,
546; Id., "Vers. iiber d. physiol. Wirkg. des Tetanusgiftes," Pflugers
Arch., lix., 105 (Bibliography), 1895.

4 Courmont and Doyon, "Le T(5tanos," Paris, BaiUiere, 1899.

5 Brusschettini, Rif. Med., 1892, quoted from Brunner.

6 Hans Meyer and'Ransom, "Unters. lib. d. Tetanus," Arch. exp. Path.,
xlix., 369, 1903.

7 Marie and Morax, "Rech. sur 1'absorption de la tox. t6t.f" Ann. Past.,
xvi., 818, 1902; xvii., 335, 1903.

TETANUS TOXINE. 115

The motor, sensory, and sympathetic nerves are equally suit-
able for the attack, according to MORAX and MARIE, although
according to MEYER and RANSOM the normal axis cylinder is the
exclusive conveyor of the poison, which reaches it only by way
of the nerve end plates on the muscle and not through the blood.
The poison is only distributed in a centripetal direction. MEYER
and RANSOM also found that, in addition to severance of the
nerves, injection of antitoxins into them prevented the poison
from reaching the spinal cord. This happened with certainty in
local (subcutaneous) poisoning, and sometimes also after intra-
venous injection. The centres belonging to the protected nerves
remain free, while the animal suffers from general tetanus.
Even in the spinal cord itself severance can prevent the further
distribution of the poison. In the case of animals thus treated
no poisoning of the brain occurred ; they lay for three weeks in
a continuous tetanic state and eventually died apparently from
exhaustion. The direct transmission of the poison by way of
the nerves appears the more probable from the fact that when
directly injected into the nerves its action is much more intense
(about ten times as much). In these experiments any quantities
of poison which might possibly have passed into the lymph
glands or blood on injection, were in each case neutralised by
large quantities of antitoxine. The period of incubation is
considerably shortened, and if the injection is made directly
into the lumbar cord it is reduced to a few hours, as had already
been shown by the experiments of Roux and BoRREL1 on
tetanus in the head. The variations in the length of the
incubation periods (vide supra) are explained by the slow trans-
mission through the nerves.

It follows from these experiments that the poison is first
transmitted exclusively by the nerves themselves to the centres
of the spinal cord, and that the specific symptoms are produced
by these. If the poison attacks only the sensory centres, there
results a simple tetanus dolorosus without convulsions. It also
results from this strict differentiation that the poison does not
reach the spinal cord by way of the blood or lymph channels.
Under normal conditions the poison only reaches the motor
ganglia by way of the motor neuron, and produces there a
condition of over-excitability towards the usually latent stimuli
which proceed from the sensory neurons. This is not the place
to go further into the theory of tetanus itself.

The usual point of attack of tetanus poison, and so of the
disease, is thus the spinal cord. But it is not exclusively the

'Roux and Borrel, " T<5tanos Cerebral," Ann. Past., xii., 225, 1898.

116 TOXINES AND ANTITOX1NES.

spinal cord that offers suitable haptophore groups for tetanus
toxine. Roux and BORREL (loc. cit.'), for example, were able to
prove that, on subdural introduction of the poison, the poison
combined exclusively with the brain cells and produced cerebral
tetanus. And, lastly, the experiments of DONITZ * on rabbits
and of MiYAMOTO2 show that, under certain conditions, the
poison can be fixed so rapidly by haptophore cells, other than
those of the central nervous system, that the animal dies
without spasms from a "tetanus sine tetano" In the case of
rabbits this fixation of the poison in the less vitally important
organs, which withdraw it from the central nervous system,
undoubtedly partially accounts for their smaller susceptibility.
At any rate, Roux and BORREL were able to poison rabbits
much more readily by intercerebral injection, in which the
poison comes into direct contact with the central nervous
system, than by subcutaneous injection, which was not the
case with guinea-pigs. Even the hen, so little susceptible in
other respects, can be poisoned by intercerebral injection.
Under certain conditions, which are still imperfectly known,
many poisons (MIYAMOTO employed a very old preparation)
appear to lose their predominating affinity for the central
nervous system. It is surely not unjustifiable to assume, pro-
visionally, that in this case substances have been formed from
the original toxine which are undoubtedly still poisonous, but
have lost their characteristic action upon the central nervous
system, there being possibly a formation of toxoids of a special
kind. On the other hand, in the case of refractory animals —
e.g., alligators — and also in that of hens, which are not completely
refractory, METSCHNIKOFF (vide infra) observed a slight forma-
tion of antitoxine without preliminary symptoms of illness,
which surely points to a combination with distributed indi-
vidual receptors.

The conclusion drawn from the study of the symptoms of
the disease — viz., that the tetanus poison combines with the
substance of the central nervous system, could also be confirmed
experimentally.

It was also shown, simultaneously, by WASSERMANN and
TAKAKI 3 and by RANSOM,* that an emulsion of the fresh brain

1 Donitz, "Ueber das Tetarmsantitoxin," Deutsch. med. Woch., 1897, 428.

2 Miyamoto, "Beitrage zur Tetanusvergif tung, " Deutsch. med. Woch.,
1900, 479.

3 Wassermann and Takaki, "Ueber tetanusantitoxische Eigenschaften
des Centralnervensystems," Berl. klin. Woch., 1898, 5; Wassermann,
" Weiterere Mitt, iiber Seitenkettenimmunitat," ibid., 209.

4 Ransom, quoted by Behring, Deutsch. med. Woch., 1898, 68.

TETANUS TOXINE. 117

of a guinea-pig could combine with and render non-poisonous
a certain quantity of tetanus toxine. It is true that this power
of combination is not very considerable ; according to PALTAUFI
1 c.c. of the emulsion can combine with 100 lethal doses at
most — a point which must not be overlooked in attempting to
come to a correct conclusion about this question.

It is the cells of the brain that enter into combination, as was
shown by BLUMENTHAL2 and MiLCHNER,3 who found that the
liquid was free from toxine after being subjected to centrifugal
force.

RANSOM,* too, was able to show that on injection of tetanus
poison into the subarachnoid space the nerve substance combines
with more poison than can be accounted for by the blood that
it contains, but that the combination is not absolute, for a part
of the toxine passes into the circulatory system even in this
method of injection. On the other hand, MEYER, like BLUMEN-
THAL and JACOB5 before him and ScuuLZE,6 found the cerebro-
spinal fluid to be/ree from toxine.

WASSERMANN regarded the results of this experiment as a
confirmation of the side-chain theory. The receptors that are
capable of combining with the poison in the brain substance,
and also in the living brain, are the same that circulate in the
blood as free receptors, so that WASSERMANN is right in forming
his conception of side-chain immunity from these experiments.

Objections have been raised on various sides, and notably by
the school of METSCHNIKOFF r and by BEHRING and KITASHIMA,S
against the conclusions drawn from WASSERMANN'S experiments.
It would take us too far from our theme to deal thoroughly with
the whole subject in dispute ; and so I will content myself with

1 Paltauf, "Diskussion zu dem Vortrag von Grueber," Wien. klin. Woch.,
1901, 51.

2 Blumenthal, " Ueber die Verandergungen des Tetanusgiftes im Tier-
korper," Deutsch. med. Woch., 1898, 185.

3 Milchner, "Nachweis der chemischen Bindung von T. -G. durch Nerven-
substanz," BerL klin. Woch., 1898, 369.

4 Ransom, "Die Injektion von Tetanustoxin bezw. Antitoxin in den
subarachnoidalen Raum," Z. f. phys. Chem., xxxi., 282, 1900-01.

5 Blumenthal and Jacob, " Zur Serumtherapie des Tetanus," Berl. klin.
Woch., 1898, 1079.

6 Schultze, " Spinalpunktion u. Einspritz. v. Antitoxinserum bei Tet.
traumaticus," Mitt. Grenzgeb. d. Med, u. Chir., v., 169, 1900.

7 Metschnikoff, Immunitdt, German ed. by Meyer, Jena, 1902. Also,
Metschnikoff, "Influence de I'organisme sur les toxines," Ann. Past., xi.,
801, 1897; xii., 81, 1898. Marie, "Propriete antite"tanique des centres
nerveux.," ibid., xii., 91, 1898. Besredka, "De la fixation de la toxine
tet.,"ibid., xvii., 139, 1903.

8 Behring, Allg. Ther. d. Infekt.-Kr., i., 1033.

118 TOXINES AND ANTITOXINES.

calling attention to the work of MARX,1 who has resurveyed the
whole ground.

MARX has found, by over 200 experiments on mice, that the
antitoxic action of the brain is of such a nature that it is merely
supplemented by the addition of serum, and that the action of
the brain does not, as was found by KITASHIMA (loc. cit.) in ex-
periments with very large amounts of toxine, destroy the action
of the antitoxine.

MARX thus confirms W ASS ERM ANN'S view that the combined
receptors of the brain act qualitatively and quantitatively in a
completely analogous manner to the free receptors that are
present in the antitoxine serum.

And this affords a fresh support of EHR LICK'S view that
the poison can only combine with suitable receptors, and that
these receptors, when broken off, have a neutralising effect upon
the poison, although they promote its action when acting in
conjunction with living cells.

It is impossible to see the force of the objection first brought
against these experiments — viz., that the brain of animals, such
as hens, &c., that are only slightly susceptible, has no protective
action — for it is self-evident to anyone who bases his conclusions
on the fundamental axioms of Ehrlich's theory, that only such
nerve substance as actually combines with the tetanus toxine —
i.e., possesses suitable haptophore groups — can exert a protective
influence ; and such haptophore groups are evidently only present
in the brains, &c., of these animals, which indeed are not refrac-
tory, not naturally immune to tetanus. It is thus one of the
most striking results of Ehrlich's theory that it at once throws
a clear light upon the formerly so inscrutable problem of natural
immunity. Both infection and the relative formation of anti-
bodies can only take place where there is a reciprocal combination
of haptophore groups. Where these are lacking there can be
absolutely no infection or neutralisation of the poison.

Indeed, the central nervous system is not the exclusive place
where the antitoxine can be produced. As we have mentioned
above, receptors that can combine with the poison also occur in
other places where the fixation of the poison does not lead to
any injurious effects — e.g., in the connective tissue, &c. But
wherever receptors are present, there, too, can the formation of
antitoxine take place. The production of antitoxine is a function
of the haptophore group, and has absolutely nothing to do with
the action of the poison as such. Thus, according to EHRLICH

1 Marx, "Die Tet.-G. neutralis. Eigensch. d. Gehirns," Zeit. f. Hyg.,
xl, 231, 1902.

TETANUS TOXINE. 119

and WASSERMANN, the central nervous system has a protective
influence, not because it offers a point of attack for the toxophore
group, but because it contains receptors, which may also be
present in other tissues.

Against this view METSCHNIKOFP raised the objection that
even the brain of frogs has absolutely no protective influence.
Although frogs are absolutely insusceptible to tetanus in the
cold, yet when warmed above 20° C. they are extremely sus-
ceptible, as is also the case to a less extent with other amphibia
and reptiles, and also with marmots (BiL LINGER 1), which do not
die during hibernation (i.e., when their temperature is lower),
but only after they wake from their sleep. Bats, too, when kept
cold, exhibit considerable resistance so long as they sleep (MEYER
and HALSEY, loc. cit.). Now the brain of the frog does not
combine with, or show any protective influence against, tetanus
toxine, and no antitoxine ever appears in the body of the frog
when poisoned with tetanus. Although this assertion is doubt-
less correct, and although the process of the breaking off and free
movement of the haptophore groups has as yet been but little
explained, yet the interpretation given by METSCHNIKOFF of this
phenomenon — viz., that the central nervous system of the frog
does not combine in the slightest degree with tetanus toxine —
is certainly not correct. MoRGENROTH,2 in a very interesting
research, was able to demonstrate that tetanus poison, even in
the cold, does unquestionably combine, although slowly, with the
central nervous system of the frog, but that the toxophore group
is inactive. On raising the temperature, however, it immediately
begins to act.

The combination of the toxine with the central nervous system
takes place very rapidly after its injection into the circulatory
system. DECROLY and RoNSSE3 showed that in the case of
rabbits the blood was not poisonous even one minute after the
intravenous injection of a lethal dose of tetanus poison. Such is
the rapidity with which the toxine disappears from the blood.4

1 Billinger, " Winterschlaf und Infektion," Wien. Uin. Woch., 1896, 769.

2 Morgenroth, "Zur Kenntnis des Tetanus des Frosches," Arch. Inter-
nal, d. Pharmacodynamie, viii., 255, 1900 (reprint).

3 Decroly and Ronsse, "Pouvoir toxique et antitoxique du Sang, &c.,"
Arch. Internat. de Pharmacodyn. , vi., 211, 1899.

4 Donitz (vide supra), of course, assumes that in the case of rabbits
receptors are also present in other organs, and that this accounts for the
extraordinarily rapid disappearance of the toxine. In the case of guinea-
pigs and mice it circulates longer in the blood. I have already had
frequent opportunities of calling attention to the importance of such
distributed receptors.

120 TOXINES AND ANTITOXINES.

Now, very important consequences follow from this extraordin-
arily great affinity of the toxine for the nervous system. Thus
it is easy to render tetanus poison harmless by means of anti-
bodies before it enters the nerve tracks, but much more difficult
to break up the compound when once formed, to separate the
toxine from the receptor, to cure the tetanus. With every hour
the combination becomes more stable, and the activity of the
serum less. It has been shown by DONITZ l how rapidly this in-
jurious result takes place. The same quantity of an ti toxine,
which, when simultaneously injected, protected a rabbit from
a quadruple lethal dose, was absolutely unavailing when injected
four minutes after the toxine (intravenous injection being used
in each case). After one hour as much as 40 times the amount
of antitoxine was necessary. After five hours even huge doses,
such as 600 times the single effective dose, were useless. MOR-
GENROTH, to his disappointment, found that the serum failed
in its action in an exactly analogous manner in the case of frogs,
although more slowly. In the therapeutic treatment of men it
has frequently been observed that after the occurrence of the
characteristic symptoms of tetanus — i.e., after the lapse of the in-
cubation period — even huge doses of immune serum are usually
of no avail.

I have only dealt with this question to this extent, and the
further discussion of it does not concern us here, because, on
these facts, taken in conjunction with the occurrence of a longer
or shorter incubation period (vide supra), attempts have been
made to base conclusions as to the alterations in tetanus poison
even in the human system.

Thus certain authors (e.g., COURMONT and DoYON2) have
assumed that the poison that eventually produces the outbreak
of tetanus is not the primary toxine of the cultivations. The
latter is asserted to act only as a ferment, which, under suitable
conditions, decomposes the protoplasm of the attacked cell with
the formation of the true tetanus poison, which can then produce
its poisonous effects without a period of incubation. In reply to
the arguments urged by these authors in support of their view,
the following remarks may be made : — As regards the period of

1 Donitz, "Ueber das Tetanusantitoxin," Deutsch. med. Woch., 1897,
428.

2 Courraont and Doyon, inter olios. — (a) "Mecanisme de production des
contractures du te"tanos," Arch, de Phys., 1893, 64; (b) "La substance
toxique qui engendre le te"tanos," Sem. Med., 1893, 122 ; (c) " Du t6t de la
grenouille, ibid., 1893, 302; (d) "De la produc. d. t. chez la poule," ibid.,
1893, 486. See also their papers already quoted, and their work Le
Tetanos, Paris, 1899. (For other investigations see Blumenthal, loc. cit.)

TETANUS TOXIN E. 121

incubation it is assumed that the true poison is produced during
this period, and that until then there is no action. This compli-
cated view is not necessary, for the mode of action of the poison
is at least as satisfactorily accounted for by the slow action of
the toxophore groups and by the different degrees of slowness in
the distribution of the poison in the nerve tracks as found by
MEYER and RANSOM (vide supra}. A further point against their
view is that the period of incubation becomes shorter in pro-
portion to the amount of toxine introduced into the animal,
although, as is the case with every toxine, there is a limit to this
reduction of the incubation period by larger doses.

Again, the complicated behaviour of the incubation period in
the case of the frog, which has been brought forward as a special
argument in support of this view can be explained more simply.
H.. was able to prove that on warming the animal

.

there was at first only a very rapid and firm combination of the
toxine, but no poisoning. If the frog, after being kept at 32° C.
for twenty-four hours, was then again placed in the ice chest,
it remained permanently well, although if again exposed to a
temperature of 32° C. it died after an incubation period that was
shortened by twenty-four hours, notwithstanding the injection of
enormous quantities of antitoxine. If, as COURMONT and DOYON
assume, a secondary change in the poison were produced by the
warming, the frog would not remain permanently well even in
the ice chest. The heat has here only promoted the formation
of an extremely stable combination, and this requires further
warming before it causes death by the action of the toxophore
group. The fact that after combination has taken place the
antitoxine is more or less completely ineffective has been put
forward as of special significance in support of the theory of
the secondary poison, but in my opinion this is far from being
the case. The predominance of the antitoxine circulating in the
blood over the toxine, which it prevents from combining with
the central nervous system is solely due to the fact that it does
so circulate in the blood and thus immediately seizes upon the
toxine when introduced into the circulatory system and mechani-
cally wards it off from the threatened tissues, and not to its
possessing any greater affinity for the toxine. It has but little
power of breaking up the combination between the nerve cell
and the toxine when once formed, and it does so only by mass
action when very large doses of antitoxine have been used.
It is thus obvious that the antitoxine must be ineffective if
introduced some time after the poisoning, and much more so
when only applied after the appearance of the symptoms, and

122 TOXINES AND ANTITOXINES.

there is absolutely no need for us to assume the existence of a
new poison without affinity for the antitoxine. Nor can the
theory derive any greater support from the fact that the brain
of animals that have died of tetanus can still combine with the
poison.

But if these considerations do not altogether explain the want
of influence of the antitoxine upon the toxine after the lapse of
a certain time, the discovery by MEYER and RANSOM (loc. cit.)
of the fact that tetanus antitoxine cannot follow the poison in its
passage along the nerve tracks affords all that is still needed in
explanation of this phenomenon. This fully accounts for the
want of activity of the antitoxine, and there is no longer any
need to call to our aid the hypothesis of a secondary poison.
This hypothesis has not, however, as yet been thereby definitely
disproved.

An experimental proof brought forward by COURMONT and
DOYON in support of their view was the fact that the transfu-
sion of blood from a dog suffering from tetanus into another
immediately produced symptoms of tetanus in the latter. KRAUS/
too, succeeded in rapidly poisoning other mice with the blood
serum from a mouse infected with tetanus, as had previously
been done by Nissen.2 Here then, in their opinion, there
was manifested the activity of the secondary poison. It was
next claimed that this poison had been detected in animal
organs.

BLUMENTHALS prepared from the organs of an animal that
had died of tetanus, a poison which, in a dose of 0*35 c.c., caused
death with tetanic symptoms in seventeen minutes without an
incubation period, and was not rendered ineffective by the
antitoxine. BUSCHKE and OERGEL* have also succeeded in
preparing a poison that acted instantaneously from the liver,
spleen, and spinal cord of an animal that had died of tetanus,
whilst TAUBER 5 obtained a similar preparation in smaller quan-
tity from the spinal cord, brain, and liver. Such reputed dis-

1 Kraus, "Beitrasr zur Klinik des Tetanus," Z. f. klin. Med., xxxvii.,
247, 1899.

'2 Nissen, " Ueb. den Nachweis von Toxin im Blut, &c.," Deutsch. med.
Woch., 1891, 775.

3 Blumenthal, "Weit. Beitr. z. Kenntn. des Tetanusgiftes," Z. f. klin.
Med., xxxii., 325, 1897; Id., "Ueber die Veranderungen des T.-G. im
Tierkorper," Deutsch. med. Woch., 1893, 149.

4Buschke and Oergel, "Beitrag zur Kenntnis des Tetanus," Deutsch.
med. Woch., 1893, 149.

5Tauber, "Bin Beitr. z. Kenntnis d. Tetanus," Wien. klin. Woch.,
1898, 747.

TETANUS TOXINE. 123

coveries of poison in the organs of dead animals must be received
with great scepticism ; it is impossible to know with absolute
certainty what has actually been extracted and what kind of
poisonous substances may have been formed during the process
of disease with its alterations in the protoplasm — substances that
may have absolutely no connection with tetanus toxine, and
against which the antitoxine is powerless. Even BLUMENTHAL
himself, as also more recently COURMONT and DOYON, now cite
these discoveries with great caution, and leave it an open question
how far this poison from the organs corresponds with the actual
tetanus poison. BLUMENTHAL regards it as a combination of the
true poison with the cell substance. This cannot be proved,
however, by such means. To recapitulate, COURMONT'S fermen-
tation theory of tetanus toxine has not as yet been proved, and
is as yet superfluous. Whether, notwithstanding, it is not
correct has yet to be decided. But it is advisable not to obscure
this extremely difficult field by additional hypotheses for which
definite support is lacking.

Tetanus Antitoxine. — The antitoxine of tetanus behaves
towards its toxine in a manner essentially analogous to that of
diphtheria — i.e., its antitoxic power can be shown to stand in
certain definite relationships to the toxine. Both enter into
combination with their toxines, and subsequent separation is
impossible. The combination does not take place so rapidly,
however, in the case of tetanus toxine, and hence, according to
DoNiTZ,1 the test should not be applied until after forty-five
minutes. Moreover, the degree of saturation depends upon the
concentration ; the greater this is, the more rapid and complete
the saturation, so that solutions of as nearly as possible equal
concentration must be taken for the comparison. As regards the
quantitative relationships the conditions are approximately the
same as in the case of diphtheria. In this case, however, they
have not been so fully elucidated, which is to be attributed, in
the main, to the extraordinary instability of the toxine. The
compound also appears to be much less stable, and sooner to
attain conditions of equilibrium (on this point see the General
Part). The serum can be preserved for a long time by the
addition of 1 per cent, of chloroform or O6 per cent, of phenol if
kept cold and in the dark. Alcohol and distilled water are
harmless to it (BEHRING 2).

Roux and VAILLARD (loo. cit.) were able to dry the cow's

1 Donitz, Sericht ub. d. Thatigkeit des Icgl. Institute f. Serumforsch. , &c.
Reprint from the Klin. Jahrb., 1899, vii.
'2 Behring, Die Blutserumtherapie, iii., Leipzig, 1892.

124 TOXINES AND ANTITOXINES.

serum in a vacuum without loss, and when it was required for
use dissolved the residue in six times its quantity of water.

The milk of immunised animals also contains antitoxine.
According to BBIEGEB and OOHN 1 it can be concentrated from
the milk in the following manner : —

The milk is first coagulated by means of rennet in an analogous manner
to that used in the preparation of diphtheria antitoxine. The filtered
whey is shaken with chloroform, allowed to stand, decanted, and then
saturated with ammonium sulphate to the extent of 32 per cent. The
precipitate is redissolved, a small amount of basic lead acetate added, and
the new precipitate washed with water rendered very slightly alkaline.
The filtrate and washings are now saturated with ammonium sulphate, and
the resulting precipitate mechanically freed from the excess of solid
ammonium sulphate by stirring it with pure chloroform. The salts fall
to the bottom, while the light compound of the anti-bodies rises to the
surface and is removed. In this way it was possible to concentrate the
antitoxine to the extent of 300 to 400 times the strength of that originally
present in the milk. A still further purification of the antitoxine can be
effected by treating the filtrate, after removal of lead, not with ammonium
sulphate, but first with sodium chloride, and then with sodium phosphate.
Practically none of the antitoxine is precipitated with the sodium chloride
precipitate, whereas it is nearly all carried down by the sodium phosphate
precipitate.

The properties of the antitoxine have naturally received much
attention. It does not differ materially from diphtheria anti-
toxine. It is partially destroyed at 68° C., though not completely
so even at 80° C. Acids (hydrochloric acid, in the proportion of
1 part to 15 of antitoxine, and lactic acid) have also a destructive
influence.

Very weak alkalies do not injure it, but, when concentrated,
rapidly destroy it. It does not readily putrefy (BEHRING).

It is not dialysable, and is partially retained by Chamberland
porcelain filters. TIZZONI and CATTANI 2 attribute to it the
nature of a ferment, for which, in my opinion, there is no
justification (see under Diphtheria). They draw this conclusion
from the fact that it can be precipitated by alcohol, and can be
extracted, though slowly, from the precipitates by means of
glycerin. Doubtless, like diphtheria antitoxine, it is also a
substance closely allied to the globulins.

1 Brieger and Cohn, "Beitr. z. Concentr. geg. Wundstarrkrampf
schiitzenden Substanz.," Zeit. f. Hyg., xv., 439, 1893.

2 Tizzoni and Cattani, " Sur les propriete's de 1'antitoxine du tetanos,"
Arch. ital. d. BioL, xvi., 394 (abstract), 1891; "Ueber d. Eigenschaften
des Tetanus- Antitoxins," CentralU. /. Bdkt.t ix., 685, 1891; "Fernere
Unters. ub. das Tetanus Antitoxin," ibid., x., 33, 1891 ; Tizzoni, "Ueb. d.
experim. Immunitat gegen Tetanus," Festschr. fur Virchow, Berlin,
iii., 29, 1892.

TETANUS TOXIN E. 125

It had already been found by TIZZONI and CATTANI that it
was precipitated by magnesium sulphate. They had also observed
that the globulins that were precipitated by means of weak
acids (acetic acid, carbonic acid) or those separated by means of
dialysis did not carry down the antitoxine, but only those
globulins that were precipitated by means of solid magnesium
sulphate at 30° C.

PICK (loc. cit.) was able to confirm these statements by means
of his more delicate methods. According to his results tetanus
antitoxine is distributed in exactly the same manner as diphtheria
antitoxine; thus, in horse serum, he found it to be combined
exclusively with the pseudoglobulin.

The value of tetanus antitoxine has been calculated by
BEHRiNG1 in the following manner : — A "single" serum is taken
to be that which will protect 1 grm. of an animal against the
action of a certain lethal dose. Thus 1 c.c. of serum of the
strength of 1 in a million will protect 50,000 mice of 20 grms.
each ; hence a mouse requires ^^Jo^ c.c. or 0-00002 c.c., a sheep
of 50 kilos. 0-05 c.c., and a horse of 400 kilos. 04 c.c. Recently,
however, the serum has been tested as accurately as diphtheria
serum in the Royal Institute for the Investigation and Testing
of Serum. Test poisons are standardised upon a test serum that
has been kept unaltered, and are then used for the valuation of
the sera under examination.

A fact of the greatest importance for the estimation of the
action of antitoxine in the organism was that established by
HANS MEYER and RANSOM (loc. cit.) — viz., that the antitoxine,
unlike the toxine, is not capable of penetrating the axis cylinder.
Hence the antitoxine is absolutely powerless against the poison
when once the latter is in the nerves. Nor can it penetrate even
by way of the blood and lymph tracts into the centres. Thus it
can only neutralise the excess of poison in the tissues. And this
explains why even highly-immuned animals can be poisoned when
the poison is introduced directly into their nerves. MEYER and
RANSOM hoped to be able to effect a cure by the direct injection of
antitoxine into the nerves.

1 The values given by BEHRING (Die Elutserumtherapie, ii., 20) do not
exactly tally. If 1 c.c. protects 50,000 mice, one mouse requires not
0-00005 c.c., but 0-00002 c.c., and a horse of 400 kilos, not 0 -25, but 0'4 c.c.

126 TOXINES AND ANTITOXINES.

BOTULISM TOXINE.

A third true toxine is the active principle in many cases of
flesh poisoning.

Considerable light was thrown upon the fairly obscure etiology
of botulism by the discovery of VAN ERMENGEM,* who isolated
from a poisonous ham a saprophytic bacillus, B. botulinus, which
even then he concluded to be the producer of a specific, extremely
active toxine. According to KEMPNER, it can also be detected in
the faeces of swine, and ScHNEiDEMUHL2 regards it as the cause
of the so-called birth-paralysis of cattle. Although substances
had already been isolated at an earlier period, and asserted to be
the active agents in the poisoning — e.g., that isolated by v. ANREPS
from sturgeon's flesh — yet botulism toxine was the first that was
proved to be the specific poison of flesh poisoning, and a true
toxine.

VAN ERMENGEM isolated it by filtration of cultivations of his
Bacillus botulinns. It is extremely poisonous, the lethal dose
for man being as little as 0-035 mgm. This in itself points to
its being a true toxine. A further proof is the specific nature of
its action, the symptoms of illness produced corresponding
exactly with those of botulism.

It produces the same symptoms in the eyes, aphonia, constipa-
tion, and retention of urine. Fever does not occur. Finally,
symptoms resembling those of bulbar paralysis are manifested,
and end in death.

Its action does not begin until after a certain period of
incubation.

According to FoRSSMAN,4 the mode of introduction has an
influence upon the results. Thus, intercerebral injection did not
produce the same characteristic form or intensity of illness as
followed subcutaneous injection ; there was a greater difference
with intraperitoneal, and the greatest with intrapulmonary in-
jection. Violent dyspnoea is then the predominating symptom
of the poisoning. Moreover, after interpleural injection the
toxine is five to nine times as poisonous, although the period of
incubation after a single lethal dose is longer. On the other

1 Van Ermengem, " Ueber einen neuen anaeroben Bacillus u. s. Bezieh.
zum Botulismus," Zeit. f. Hyg., xxvi., 1, 1897.

2 Schneidemiihl, " Ueb. Botulismus beim Menschen und die sog. Geburts-
paralyse bei Rindern," Centralbl.f. Bakt., xxiv., 619, 1898.

3 v. Anrep, "Intoxication par les ptomaines," Arch. Slaves de Biol., i.,
341, 1886, quoted by v. Ermengem, loc. cit.

4 Forssman, "Beitr. z. Kenntn. d. Bakt. d. Botulismus," Author's
abstract in Centralbl.f. Bakt., xxix., 541, 1901.

BOTULISM TOXINE. 127

hand, the minimum period of incubation (with large doses) is
six hours after subcutaneous injection, and four hours after intra-
pulmonary injection.

Like all true toxines, it is very sensitive to external influences.
It is rapidly weakened by air and light, and also by raising the
temperature even to 58° C. for three hours. Alcohol, ether, and
oxidising substances also destroy it rapidly, while reducing agents
are relatively but little injurious. On the other hand, it is a
remarkable fact that it can act by way of the intestinal canal, as
was found by VAN ERMENGEM, and confirmed by FORSSMAN. The
fluids of the stomach and small intestine do not injure it, but it
is speedily destroyed by the contents of the large intestine.

BRIEGER and KEMPNER * have prepared the toxine in a concen-
trated form by Brieger's method : —

The germ-free filtrate is partially neutralised with ammonia, and treated
with twice its volume of a 3 per cent, solution of zinc chloride. The preci-
pitate is carefully washed and cautiously treated with a 1 per cent, solution
of ammonium bicarbonate until the mixture just shows an alkaline reaction,
after which it is decomposed with ammonium phosphate, the liquid filtered
from the precipitated zinc phosphate, and the toxine precipitated with
ammonium sulphate.

In this way they obtained very small yields of solid toxine in
quantitative experiments.

Botulism toxine is a specific nerve poison. KEMPNER and
POLLACK 2 and MARINESCOS have simultaneously studied its
action upon the anatomy, and especially the alterations produced
in the cells of the anterior cornua, and the destruction, chromato-
lysis, and decomposition of Nissl's granule, with which I cannot
deal here.

It is, however, very important that, owing to this strong
affinity for the nerve substance, botulism toxine is fixed and
rendered harmless by it in an analogous manner to tetanus
poison.

KEMPNER and SCHEPILEWSKI 4 found that the brain and spinal
cord were able to combine with considerable quantities of poison,
and that that property could be utilised both in the prophylactic
injection of brain substance, as well as for the neutralisation of

1 Brieger and Kempner, " Beitrag z. Lehre v. d. Fleishvergiftung,"
Deutsch. med. Woch., 1897, 521.

2 Kempner and Pollack, "Die Wirkung des Botulismus toxine auf die
Nervenzellen," Deutsch. med. Woch., 1897, 505.

3 Marinesco, " Lesions des Centres nerveux produites par la toxine du
B. botulinus," Soc. Biol., xlviii., 31, 1896; Sem. Med., 1896, 488.

4 Kempner and Schepilewski, "Ueb. antitoxische Substanzen geg. d.
Botulismusgift.," Zeit.f. Hyg., xxvij., 213, 1898,

128 TOXINES AND ANTITOXINES.

toxine already in the system for a period of twelve hours after
its introduction. These properties distinguish the action of the
substance of the central nervous system from that of certain
simpler chemical substances, which, while equally combining
with the botulism poison in vitro, completely lack its immunising
and curative capacity. Examples of such substances are lecithin
and cholesterin, but not cerebrin. This property of the brain
substance is destroyed by boiling.

According to CHARRIN and BARDiER,1 it also acts as a heart
poison. It retards the action of the heart, and acts more rapidly
than diphtheria poison in this respect.

KEMPNERS subsequently prepared an antitoxic serum against
botulo-toxine by the immunisation of goats. The antitoxine acts
in accordance with the law of multiples. It can, however, only
have any curative effect when used within twelve hours ; the
dyspnceic form in particular is asserted by FORSSMAN to lie in-
capable of cure. The practicability of the treatment is therefore
as doubtful here as in the case of tetanus.

PYOCYANEUS TOXINE.

Bacillus pyocyaneus, which is very pathogenic for many
animals, also produces a true toxine. The poisonous action of
this toxine has frequently been investigated, as have also the
phenomena of immunity that result on its introduction into
the bodies of animals. There are also numerous researches on
the " toxicity " of B. pyocyaneus in scientific literature, in which
naturally no distinction was made between the poisonous effects
of the cells and of the filtrate.3 Essentially, however, immunity
against B. pyocyaneus is bactericidal ; no antidote to the poison
is produced in immune bodies, but the bacilli themselves perish
under the influence of a specific agent directed against them.
Here, then, we meet with conditions similar to those that occur
in typhus and cholera, with which we shall deal in a later page.

The conditions differ, however, from those of cholera in one
essential particular. Whilst in that disease the supposed toxine
can only be separated very sparingly from the vibriones, and, in

1 Charrin and Bardier, " Action cardiaque, propri^te speciale de la
botuline," Soc. BioL, 1., 60, 1898.

2 Kempner, " Weit. Beitr. zur Lehre von der Fleischvergiftung," Zeit.f.
ffyg., xx vi., 481, 1897.

3 The most important literature on the subject is given by Breymann,
"Ueb. Stoffwechselprod. des B. pyocyaneus," Centralbl. f. Bah., xxxi.,
841, 1902.

PYOCYANEUS TOXINE. 129

the main, remains attached to the cells themselves, it can be
separated almost completely from the cells in the case of pyocy-
aneus, so that we must assume that secretory processes occur
here similar to those that produce diphtheria toxine.

B. pyocyaneus has a very characteristic behaviour, as was shown
by WASSERMANN.1 Frequently it develops in the body of an
animal, and is thus infectious, and then the bactericidal protective
forces come pre-eminently into action.

On the other hand, however, it develops in its cultivations
a true toxine, which can be separated from the cell substance,
and produces in the organism a true antitoxine.

B. pyocyaneus thus occupies a very interesting intermediate
position between the purely toxic diphtheria bacillus, on the one
hand, and the bacteria of the cholera type, on the other hand, in
the case of which immunity is also pre-eminently bactericidal,
not antitoxic, while the supposed endotoxines have not as yet
been isolated in the free state.

WASSERMANN proved that this poison could be* separated
almost completely from the bacilli, so that the cells, just as in
the case of diphtheria, contained practically no more toxine;
but, above all, he showed that by injection of the living bacilli in
small, but increasing, doses an immunising process results, which
was directed exclusively against the bacilli themselves, and had
absolutely no influence upon the poison.

But, on the other hand, he succeeded in producing a true
antitoxic solution by means of the soluble poison. In this case
the serum had absolutely no bactericidal power in vitro ; yet, at
the same time, an animal rendered proof against the poison was
also immune against the living bacilli, just as in the case of
diphtheria ; for the bacilli, deprived of their keenest weapon,
their toxic function, are merely harmless intruders in the
organism rendered proof against the poison, and speedily perish
without inflicting any injury.

WASSERMANN thus proved that B. pyocyaneus produced a true
toxine, which in this respect was undoubtedly akin to diphtheria
and tetanus toxine. It is somewhat more stable than these,
especially as regards its behaviour towards heat. It is not
completely destroyed even by boiling.

In consequence of this a fact of theoretical importance was
established in the case of pyocyaneus poison.

It was found that in a neutral mixture of pyocyaneus toxine
and antitoxine the latter could be completely eliminated by heat,

1 Wassermann, "Unters. iib. einige theoret. Punkte d. Immunitatslehre,"
Zeit.f. Hyg.y xxii., 263, 1896.

130 TOXINES AND ANTITOXINES.

so that the previously neutral serum became poisonous again.
This proves beyond doubt that what CALMKTTK had previously
shown to be the case with snake poisons also applies to bacterial
toxines — viz., that the antitoxine does not effect any destruction
of the poison, but that there is a simple form of combination
which renders the toxine incapable of attaching itself by means
of its haptophore group to the cell and poisoning it.

WASSKRMANN prepared his toxine by cultivating B. pyocyaneus
on beef bouillon containing 2 per cent, of peptone, and subse-
quently sterilising it by means of toluene. Further investigation
of the poison and its antitoxine are still required. In particular
an answer is still needed to the question whether the toxic
principle is identical with pyocyanolysine, with which we shall
deal presently, or whether the B. pyocyaneus, as is probably the
case, resembles the tetanus bacillus in producing two active
substances, a toxine and a lysine.

It can be isolated by filtration through Chamberland filters.
Its toxicity cannot be compared, at all events as yet, with that
of tetanus toxine, for example. WASSERMANN found the lethal
dose for guinea-pigs to be 0*5 c.c.

The quantitative ratios between pyocyaneus toxine and anti-
toxine show a very important peculiarity.

In this case the law of multiples only holds good up to about
10 times the lethal dose. From that point upwards even great
doses of antitoxine no longer afford protection.

Undoubtedly WASSERMANN is right in concluding from the
relatively slight toxicity of the poison, as well as from this
restricted formation of antitoxine, that pyocyaneus toxine con-
tains secondary poisons of non-haptoid nature derived from the
toxine, such as have hitherto only been obtained outside the
body in the case of cholera and typhus. In this respect, too, it
would seem that B. pyocyaneus occupies an intermediate position
between the diphtheria bacillus, which produces relatively stable
true toxines, and the bacteria of cholera and typhus.

THE TOXINE OF SYMPTOMATIC ANTHRAX.

A weak immunising poison which did not cause death, and
was only partially destroyed at 115° C., was obtained by Roux 1
by filtration of cultures and from flesh juices of infected animals.

DUENSCHMANN 2 cultivated the bacilli of symptomatic anthrax

1 Roux, "Immunite centre le charbon symptomatique," Ann. Past., ii.,
49, 1888.

2 Duenschmann, " Etude experim. sur le charbon symptomatique,"
Ann. Past., viii., 403, 1894.

THE TOXINE OF SYMPTOMATIC ANTHRAX. 131

anaerobically on macerated flesh or ox serum, and found in the
filtrate after seven days a poison with a specific action, which
killed guinea-pigs in doses of 5 to 6 c.c. It had no protective
action against living bacilli.

ARLOiNG,1 in his latest researches on immunisation against
symptomatic anthrax, makes no mention of any toxine being
formed by the bacteria.

Our knowledge of this otherwise practically unknown poison
has been greatly enlarged by the recent monograph of GRASS-
BERGER and ScHATTENFROH.2 Their results show that a true
toxine is produced by the bacillus of symptomatic anthrax which
had not been discovered by the researches of previous workers.
They find that the formation of the toxine by the bacillus only
takes place under certain definite conditions, notably when the
micro-organism shows itself as a typical producer of butyric acid.
For this purpose the presence of fermentable sugar, or still
better of calcium lactate, in the culture medium is a primary
necessity. It very frequently happens that there is no formation
of toxine at all ; the " denaturalised " micro-organisms are no
longer able to ferment the lactic acid and produce no trace of
toxine. In such cases the power of sporulation may be either
restricted or retained in full vigour. On the other hand, there
may be an energetic formation of toxine in a quiet after-
fermentation in which chiefly the lactic acid is decomposed,
while none was formed in the first vigorous fermentation, and
this may also occur in cultivations which are free from sugar but
contain lactic acid. A further point of the utmost importance
is the purity of the cultures, since other bacteria apparently
injure the very unstable toxine. A temperature of about 37° C.
is also necessary for the production of the toxine. The formation
of toxine by the bacillus of symptomatic anthrax is a true free
secretion. Filtration through infusorial earth was found to be
the best means of separating the toxine from micro-organisms,
since more compact filters absorbed too much of the poison.

The effects of the toxine on guinea-pigs are similar to those
caused by infection with symptomatic anthrax — viz., oedema,
areas of haemorrhage, lowering of the temperature, oedema of
the lungs — in fact all the general symptoms of toxine poisoning.
The period of incubation is only a few hours, while the disease
lasts for two to four days or, after very large doses, six to seven
hours.

1 Arloing, " Serotherapie du charbon symptomatique," Comptes Rend.,
cxxx., 548, 1900; cxxxi., 319, 1900.

2 Grassberger and Schattenfroh, Ueber das Rauschbrand, 1904.

132 TOXINES AND ANTITOXINES.

The unit of measurement chosen was a solution of the poison,
O01 c.c. of which killed a guinea-pig of 200 to 300 grms. on
subcutaneous injection. In the case of rabbits death did not
take place until after the lapse of an hour after the intravenous
injection of even 1,000 lethal doses ; the lethal dose for them on
subcutaneous injection was O'l to 0-2 c.c. of the normal poison.
Similar amounts per kilo, of body weight were also fatal to
monkeys, dogs, hedgehogs, mice, hens, pigeons, sheep, and oxen.
The lethal dose for a young ox was 40 c.c. of the normal poison,
while sheep required about 2 c.c. Frogs were refractory, but
retained the poison within their body.

The poison passes very slowly through a porcelain filter and
is hardly dialysable. It resists freezing and thawing, and light
has but little injurious effect upon it, but it is affected when
heated even to 30° C. On the other hand, it can be dried in a
vacuum in the cold. It is almost completely destroyed in an
hour when exposed to a temperature of 50° C., so that very large
doses (7 to 10 c.c.) then produce only local swelling. It becomes
rapidly weaker even when kept in an air-tight vessel and still
more rapidly when exposed to the air.

It is destroyed by permanganate added in the proportion of
0*15 per cent., by phenol (1 per cent.), and formaldehyde (0-1 per
cent.), while "salting out" with ammonium sulphate and pre-
cipitation with alcohol and ether have a considerable injurious
effect. Chloroform is absolutely without influence upon it.

The poison of symptomatic anthrax is also shown to be a true
toxine by the fact that it is possible to produce an antitoxine to
it. Guinea-pigs cannot be used for this purpose since they are
too susceptible, but it can readily be done with rabbits and oxen.
It is very easy to immunise the latter, and they yield sera of
high antitoxic value (up to 400-fold), whereas even highly
immune rabbits give sera which are only weak in antitoxine.
The ratios between the toxine and antitoxine show numerical
proportions exactly analogous to those observed, e.g., in the case
of diphtheria toxine (marked variations in the value of D,
swellings caused by the injection of mixtures in the "differential
zone," &c.), and thus point to the presence of toxones. On the
other hand, there does not appear to be any formation of toxoids,
since the decrease in the toxicity keeps parallel with the decrease
in the combining power. Moreover, combination appears to
take place very slowly.

The antitoxine appears to be very stable (it can be kept for
two years). It is not dialysable ; it can be dried and can resist
a temperature of 60° to 65° C. for an hour.

TETANOLYSINE. 133

Over-saturated mixtures of toxine and antitoxine are fre-
quently still poisonous to guinea-pigs, though other animals can
be immunised by their means, so that here too there appear to
exist conditions of equilibrium about the neutral point, which
need further investigation.

BACTERIAL HJEMOLYSINES.

Closely allied to the true toxines are those bacterial substances
which exert a specific activity on the red corpuscles of the blood,
altering their plasma in such fashion that the haemoglobin
exudes, the blood being thus "laked." They differ essentially,
however, from blood poisons of the ordinary kind, such as, e.g.,
phenylhydrazine, &c., in acting physiologically as true toxines —
i.e., producing anti-bodies, antilysines, in the organism. They
thus approximate on the one hand to the true toxines, and on
the other hand to other hcemolytic liaptines, such as ricine, &c.,
as well as to the specific hcemolysines which are formed on the
introduction of erythrocytes foreign to a body and also in normal
sera. Whether these lysines are simple haptines or whether
they do not rather consist of amboceptor and complement has
not yet been definitely decided ; yet in the case of staphylolysine,
at all events, all the arguments up to the present support the
view that they are simple haptines (BoiiDET, EHRLICH, and
MORGENROTH).

Of these bacterial hsemolysines only two are as yet definitely
known — tetanolysine and staphylolysine. Other bacteria also
exhibit hsemolytic activity, but it is not yet quite certain
whether this is to be attributed to specific lysines, although
it is true that in the case of coli-lysine, for example, anti-bodies
are known. But the chief argument against their being of the
nature of toxines is the fact that they can resist a temperature
of 120° C.

A further interesting point is the fact that there is an extra-
ordinary difference in the degree of resistance offered by the
blood-corpuscles of different species to the various lysines, and
that the erythrocytes of certain species are naturally more or
less completely immune against each of them.

TETANOLYSINE.

Tetanolysine was discovered by EnRLicn1 in cultivations of
the tetanus bacillus.

1Ehrlich, Ges. d. Charittdrzte, [3], ii., 1898; B&rl. Uin. Woch.t 1898,
No. 12.

134 TOXINES AND ANTITOXINES.

The reasons that led EHRLICH to conclude that it was a different
poison to the true tetanus poison, the convulsion-producing tetano-
spasmine, are as follows : —

The ratio between the amounts of tetanolysine and tetano-
spasmine in the cultivations and preparations derived from them
is not constant. Some solutions of the poison are rich in the
former and relatively poor in the latter, and vice versd.

Tetanolysine is more sensitive to external influences than
tetanospasmine.

Tetanolysine combines with the erythrocytes, while tetanus
toxine is left by them in solution.

In proportion as the tetanus solution varies in the amounts
of the two poisons, so also the anti-serum prepared by means of
this solution contains varying relative quantities of the corre-
sponding anti-bodies, so that it is sometimes more antitoxic and
sometimes more antilytic in its action.

Tetanolysine was thoroughly investigated by MADSEN1 in
EHRLICH'S Institute, and his results were confirmed by KRAUS

and CLAIRMONT.2

MADSEN obtained, by means of precipitation with ammonium
sulphate from a bouillon culture of tetanus, a preparation, of
which O000001 grin, was the amount required for a lethal dose
for a mouse.

This poison dissolves the blood-corpuscles of many animals ;
rabbits' blood, being particularly sensitive, is used for the experi-
ments, in the form of a 5 per cent, emulsion in a physiological
solution of sodium chloride.

The amount of solution is determined colorimetrically by com-
parison with a standard solution of blood. It depends, ceteris
paribus, on the amount of poison added. There is a difference,
however, in the sensitiveness of individual blood-corpuscles ;
moreover, tetanolysine has much less action in the cold than at
the incubation temperature, whereas other hsemolytic poisons do
not possess this property.

Tetanolysine is extraordinarily sensitive to external influences.
Even at the ordinary temperature it becomes considerably weaker
in less than an hour, especially in dilute solutions, but concen-
trated solutions also soon lose part of their hsemolytic power.
Higher temperatures, even 50° C., have a very injurious effect.
The lysine can be kept on ice without undergoing decomposition
for twenty-four hours, and in the dry condition is absolutely stable.

1 Madsen, "Ueber Tetanolysin," Zeit. f. Hyg., xxxii., 214, 1899.

2 Kraus and Clairmont, "Ueber Hamoly sin und Antihamoly sin," Wien.
klin. Woch., 1900, 49.

TETANOLYSINE. 135

It is particularly interesting to note that MADSEN was able to
show that this weakening was due to the formation of toxoids.

In his experiments he followed exactly the same methods as
were used by EHRLICH in elucidating the constitution of diph-
theria poison — i.e., he determined the conditions of combination
with the specific anti-body of tetanolysine. This antilysine is
present in the preparations of antitoxine against tetanus. In
the first place, MADSEN established a unit of toxic activity and
a unit of the antilytic power of the anti-body, in accordance
with Ehrlich's methods, though here obviously the experiment
on an animal had to be replaced by a determination of the
blood-solvent power in a test tube. Then, on investigating the
conditions of the partial neutralisation of the poison with anti-
toxine, he found that they were quite analogous to those in
the case of diphtheria poison, for the neutralisation did not
take place regularly throughout the whole quantity of poison,
but zones of different combining power with regard to the anti-
lysine could be detected.

An addition of only one-thirteenth of the total amount of
antitoxine required to neutralise the toxic unit was sufficient
to reduce the hsemolytic power by a half ; an addition of a fifth
neutralised as much as nine-tenths of the poison ; whilst one-half
neutralised ninety-nine hundredths.

From this it follows that that part of the poison which has
the greatest affinity for the antilysine is also endowed with the
main proportion of the activity, that then come a second and
third zone with less affinity and also smaller solvent capacities,
and finally the " spectrum " is completed by a zone with slight
affinity and little toxic activity.

We have here, then (to employ the terminology adopted in
the case of diphtheria poison), a zone of highly active prototoxine,
with then a broad zone of less active deuterotoxine (hemitoxine *?),
followed by the zone of tritotoxine, and, lastly, the toxones, which
only act upon certain particularly susceptible erythrocytes, and
also enter much more slowly and feebly into combination.

In general, only prototoxine and deuterotoxine act in the cold
(MADSKN'). " If a poison is so far neutralised with antilysine that
these two groups remain inactive, the still remaining tritotoxine
has absolutely no solvent action at 8° C., even when present in
the largest quantities.

This, as MADSEN has shown, is due to the fact that at that
temperature the toxophore group of the tritotoxine is inactive,
for it combines with the erythrocytes even at that temperature,
1 Madsen, quoted by Dreyer, Zeit. /. Hyg., xxxvii., 274, 1901.

136 TOXIN ES AND ANTITOXINES.

so that solution takes place after separating the mixture by
centrifugal force and applying heat. MORGENROTH (p. 121) was
able to demonstrate the occurrence of similar phenomena in the
case of tetanus in the frog,

A further analogy with diphtheria poison is seen in the mode
of the formation of toxoids. Thus, tetanolysine decreases in
strength very rapidly, and mainly, as in the case of diphtheria
poison, at the cost of the prototoxine zone. The deuterotoxine
zone is relatively more stable.

Like tetanus poison, tetanolysine differs from diphtheria poison
in requiring a certain amount of time, even several hours, to
combine with the antitoxine.

Its action upon the erythrocytes also does not begin at once,
but only after a certain period of incubation, which decreases
with the increase in the amount of poison.

ARRHENIUS and MADS EN (loc. cit.) have made a closer examina-
tion of this incubation period, and attribute it to the restrictive
effect of the membrane (see General Part).

According to ARRHENIUS and MADSEN it is not absolutely
indispensable to assume the existence of poison spectra, as
described above, in the case of tetanolysine, and, in their opinion,
the quantitative ratios can be better explained by the hypothesis
of dissociated conditions of equilibrium such as we described at
length in the General Part. I have, however, cited MADSEX'S
investigations for their bearing on EHRLICH'S " spectra " and the
conclusions originally drawn from them, because it is not
definitely decided that there is a plurality of poisons in this case.

The results obtained by TIZZONI and CENTANNI l show that the
toxoids can also have an immunising effect, for they found that
an anti-body to the lysine could be prepared by means of tetano-
spasmine which apparently contained no lysine.

Further experiments made by MADSEN 2 upon tetanolysine are
very interesting, and have also an important bearing upon the
question of the therapeutic action of antitoxines — i.e., upon their
power of again liberating the lysine from its combination with
the cell. It was found that, by the addition of antilysine, lysine
in combination with the blood-corpuscles could be liberated
again, and that the blood platelet could be "healed" even after
it had been attacked. The necessary dose, however, increased
very rapidly, just as in the case of diphtheria poison and tetano-

1 Tizzoni and Centanni, Real Accad. Bologna, 1900, quoted by Neisser
and Wechsberg, loc. cit.

2 Madsen, "Ueber Heilversuche im Reagenzglas " Zeit.f. Hyg., xxxii.,
239, 1899.

PYOCYANOLYSINE. 137

spasmine. Even after five minutes twice the dose was required;
after fifteen minutes, three times ; and after thirty minutes, five
times the amount of a single protective dose. Beyond that time
accurate measurements were not possible, since, prior to the
addition of the antilysine, the amount of solution was so great
that the shade of colour could no longer be accurately matched.

PYOCYANOLYSINE.

A similar bacterial hsemolysine has been obtained from culti-
vations of B. pyocyaneus by BULLOCH and HUNTER. l

They isolated it from eight different cultivations, the " results
being practically the same."

The hsemolytic power was tested upon the blood-corpuscles of
the most different species of animals, the amounts usually em-
ployed being 0-5 c.c. of the unfiltered culture, or 1-5 to 2 c.c.
of that filtered through a Chamberland filter. Rabbit's blood
offered some resistance.

There is very little pyocyanolysine present in young cultures,
the filtrate being practically devoid of it. Cultivations three to
four weeks' old yield a filtrate which also contains lysine, though
invariably in a very small proportion as compared with that in
the main cultivation.

BULLOCH and HUNTER conclude from this that the lysine is
combined with cells of the bacilli, and is not liberated until the
cultivations grow older. In consequence of this combination
with the cells it is somewhat protected from the destructive
effect of heat, so that the unfiltered cultivation can resist a
temperature of 100° C. for a short period (fifteen minutes),
whereas the poison in the filtrate is rapidly destroyed by boiling.
The antilysine has not yet been detected.

These results were, in the main, confirmed immediately after-
wards by WfiiNGEROFF,2 who, however, also obtained the lysine
by filtration of the cultures. Subsequently MARG. BREYMANN 3
found the lysine only in the filtrates, even in the case of young
cultivations.

WEINGEROFF was able to give a direct proof that the lysine
combined with the blood-corpuscles, while the toxine present in

1 Bulloch and Hunter, " Ueber Pyocyanolysin," CentralbL f. Bakt.,
xxviii., 865, 1900.

2 Weingeroff, " Zur Kenntnis des Hamolysins des B. pyocyaneus,"
CentralbL f. Bakt., xxix., No. 20, 1901.

3 M. Breymann, " Ueb. Stoffwechselprod. d. B. pyocyaneus," CentralbL
/. Bakt., xxxi., 481, 1902.

138 TOXINES AND ANTITOXINES.

the same solution remained at liberty. This showed that the
lysine and toxine were discrete.

LuBENAU,1 too, found that an old pyocyaneus cultivation
(twenty-one months) had a very pronounced hsemolytic action \
it was strongly alkaline ; neutralisation perceptibly weakened
the hsemolytic function, but did not destroy it.

LOEW and KozAi2 found that the admission of air and the
addition of sugar promoted the formation of pyocyanolysine.

According to BREYMANN the lysine, resists the action of heat.

COLILYSINE.

A principle that attacked the corpuscles of the blood, and was
not destroyed by heat, was discovered by KAYSER 3 in cultivations
of B. coli.

It is produced when the reaction of the liquid is very faintly
acid. Its action is most pronounced upon the blood of the dog,
and then upon that of the horse, ox, and rabbit, while it has
little or no effect upon the blood of man, the guinea-pig, sheep,
goose, and pigeon. The order of the degree of activity upon the
different kinds of blood differs from that of sfcaphylolysine.

The poison is present in filtered cultivations of three days'
growth, not in the cells of the bacteria. The haemolysis is not
preceded by agglutination.

The lysine can resist a temperature of 120° C. for thirty
minutes without injury. It combines with the erythrocytes in
the cold like a true toxine, and, on warming, solution takes place.
Its stability on keeping varies very considerably.

On subcutaneous injection there is formed in the organism an
antilysine which is stable at 56° C. Moreover, " healing " — i.e.,
restriction of haemolysis when once commenced — can be effected
by means of antilysine (cf. Tetanolysine). Normal serum,
especially that of the horse, also contains the antilysine.

STAPHYLOLYSINE.

The history of the investigation of the staphylococci poisons
and their role is, in the main, a description of the same labyr-
ynthine paths as have usually been followed in the examination

1 Lubenau, " Hamolyt. Fahigkeiten einzelner pathog. Schizomyceten,"
Centralbl.f. Bakt., xxx., 356.

2 Loew and Kozai, " Ueb. d. Bild. des Pyocyanolysins," Malys Jb.,
xxxi., 912, 1901.

3 Kayser, "Ueb. Bakterienhamolysine, bes. d. Colilysin," Ze.it. f. Hyg.,
xlii., 118, 1903.

STAPHYLOLYSINE. 139

of bacterial toxines. Fortunately in this case, unlike that of the
streptococci, for example, definite results have been yielded by
the latest researches.

As regards the earlier investigations only three are really of importance.
RODET and COURMONT 1 discovered in cultivations of staphylococci an im-
munising substance precipitable by alcohol. REiCHEL,'2 who made a closer
study of the problem, was able, by filtration of cultivations of Staphylo-
coccus pyogenes aureits, to isolate a specific poison, which was, however, not
very toxic, and against the action of which immunity could be obtained.

MOSNY and MARCANO3 found that staphylococci were slightly poisonous,
and, on injection, produced an antitoxine.

We know now that Staphylococcus pyogenes aureus produces
two specific poisons, one of which, leucocidine, has a poisonous
action on the leucocytes, whilst the other, a lysine, affects the red
corpuscles of the blood.

The first to indicate briefly the existence of a staphylotoxine
acting on the red corpuscles, as a substance distinct from leuco-
cidine, was VAN DE VELDE,* and then KKAUS and CLAiRMONT.5
This lysine was first systematically studied, however, in Ehrlich's
Institute by N KISSER and WECHSBERG.6

Staphylococcus pyogenes aureus produces a blood - solvent
toxine, which can be detected in bouillon cultures after three to
four days' growth, the optimum period for its formation being
ten to fourteen days. The best condition for its production is in
the still faintly acid broth, to which has been added from a third
to a half of the quantity of normal alkali required for its
complete neutralisation. It can be separated by filtration.

The capacity for producing lysine varies very greatly with the
different species of staphylococci, and apparently, as is also the
case with the true toxines, stands in no direct ratio with regard
to its virulence for man. The true pyogenic species, albus and
aureus, invariably produce it, but there are numerous other non-
pathogenic species that produce no lysine. The quantity of

1 Rodet and Courmont, " De 1'existence . . . dans des cultures du
staphylocoque d'une substance vaccinante," Comptes Rend., cxiii., 432, 1891.

2 Reichel, " Ueb. Immunitat gegen das Virus von Eiterkokken," Arch.
/. Uin. Chirurg., xlii., 237, 1891.

3 Mosny and Marcano, " De Faction de la toxine du staphyl. pyog.,"
Sent. Med., 1894, 544^

4 Van de Velde, ' ' Etude s. 1. mecanisme de virulence du staphylocoque
pyogene," La Cellule, x. ; id., "Contribution a 1'immunite des lapins centre
le Staphylocoque," Ann. Past., x., 580, 1896.

5 Kraus and Clairmont, " Ueber Hamolysine u. Antihamolysine." Wien.
Tdin. Woch., 1890, 49.

6 Neisser and Wechsberg, "Ueber das Staphylotoxin," Zeit. /. Hyg.,
xxxvi., 299, 1901.

140 TOXINES AND ANTITOXINES.

lysine formed varies with the different species, as does also
the period of time at which the maximum production occurs.
According to LUBENAU l this fluctuates considerably within
a few hours. The addition of glucose to the cultivations reduces
the yield of lysine (KAYSER2).

Staphylolysine answers the requirements of the true toxines
as regards the influence of external factors. While it can
generally be kept for a long period unaltered in an ice chest
(with the addition of phenol) it is destroyed in twenty minutes
at 56° C., is injured at 48° C., and even when kept in an
incubating chamber loses its activity within a few weeks.

It can resist the action of considerable quantities of alkalies,
acids, and ordinary salt ; but strong alkalinity, especially at the
incubation temperature, is injurious.

The blood-corpuscles of different animals vary considerably
with regard to their susceptibility towards the same lysine ; the
erythrocytes of the rabbit appear to be the most susceptible,
while those of man, of the goat, and, above all, of the goose, offer
much greater resistance. A still further complication is intro-
duced into the conditions by the fact that the normal serum of
most species of animals exercises a more or less pronounced
protective action against the lysine, so that in order to obtain
comparable results, it is necessary to use washed blood-corpuscles.
Rabbit's blood, apart from its special susceptibility, is also the
best medium for the test, owing to the fact that normal rabbit's
serum has only a very insignificant protective capacity. Even
with the same blood it is possible, as in the case of tetanolysine,
to demonstrate differences in the susceptibility of the erythro-
cytes, so that even weak solutions of lysine dissolve certain blood-
corpuscles, whereas considerably stronger solutions must be
employed to obtain complete solution.

ANTISTAPHYLOLYSINE.

The protective force exerted by certain normal sera in varying
degree against the action of staphylolysine must be attributed to
the presence of a specific anti-body whose activity is exclusively
directed against staphylolysine. Normal horse serum, in par-
ticular, is sometimes so rich in it that as little as O'Ol c.c. will
afford protection against a dose of staphylolysine that would
otherwise dissolve the corpuscles en masse.

1 Lubenau, " Hamolytische Fahigkeit einiger pathogener Schizomyceten,"
Centralbl.f. Sakt., xxx., 356, 1901.

2 Kayser, "Einw. d. Traubenzuckers auf Staphylococcus," Zeit. f. Hyg.t
xl., 21, 1902.

ANTISTAPHYLOLYSINE. 141

Normal horse serum also frequently affords protection against tetano-
lysine. From this KRAUS and CIAIRMONT have drawn conclusions as. to
the identity of the two lysines and their anti-bodies. NEISSER and
WECHSBERG, however, were able to show that, on the one hand, immune
sera that afforded protection against staphylolysine had no action at all
upon tetanolysine, and that, on the other hand, tetanus sera, which acted
energetically against tetanolysine, afforded under certain conditions much
less protection than normal serum against staphylolysine.

Staphylolysine thus shows itself to be a true toxine in so far
as it possesses a specific anti-body, which is present even in many
normal sera including human serum, in varying amounts, and
is invariably formed whenever an animal is rendered immune
against the lysine.

This is attained by subcutaneous or intravenous, but not
peritoneal, injection of small doses twice or thrice repeated into
goats or rabbits.

A very interesting point is the protection afforded to the
lysine by combining with the anti-body. Although the anti-
body can resist temperatures up to 68° C., which speedily destroy
the lysine, it is not possible to eliminate the toxine f rom a normal
mixture by means of heat, and to leave the antitoxine free.

All these lysines, whatever their origin, produce the same anti-
lysine effective against them all, so that it would seem that we
have here to deal with a simple haptine. The evidence is in-
sufficient for us to decide whether we have here products of the
activity of staphylococci or normal side-chains in the stricter
sense of the word. Yet anti-bodies of the most diverse kind
occur in normal serum with such relative frequency as to render
it probable that normal receptors may actually have an affinity
for the poisons in question, which in this case may have the form
of free amboceptors resembling the receptors of the blood-
corpuscles.

The Constitution of Staphylolysine.— Staphylolysine has many
analogies with tetanolysine in its constitution. It is not con-
structed on the type of Bordet and Ehrlich's hsemoly sines — i.e.,
it does not, like them, consist of two parts, the "amboceptor"
and the "complement" for after being heated to 56° C., at which
temperature it becomes inactive, it can not be rendered active
again either by normal serum or any other means. Nor does
this lysine, after being rendered inactive, still produce an anti-
body, as do the intermediate bodies that withstand the heat in
the case of the haemolysines. Its structure is thus that of the
simple toxine ; the haptophore and toxophore groups are attached
to one nucleus. As in the case of the simple toxines, the hapto-
phore group is able to enter into combination even in the cold,

142 TOXINES AND ANT1TOXTNES.

while solution does not take place owing to the toxophore, group
remaining inactive. If the erythrocytes of rabbits' blood be
treated for some hours with staphylolysine at 0° C., no solution
occurs ; but on carefully washing the erythrocytes, and warming
them to 37° C. solution, it immediately takes place. Perfectly
analogous conditions of combination and activity have been
observed by MORGENROTH (loc. cit.) in tetanus in the frog, and
by MADSEN in the case of tetanolysine. Staphylolysine is thus
closely allied to the true toxines.

This analogy goes still further. Adopting Ehrlich's method of
determining the constitution of a toxine, NEISSER and WECHS-
BERG have treated a definite amount of toxine with increasing
fractions of the antitoxic unit, and have thus obtained " spectra,"
which present many analogies with those of diphtheria toxine
and tetanolysine. There is no need to go more fully into these
details here.

Hsemolysines are also produced by other bacteria, as has been
described — e.g., by LUBENAU (loc. cit.), and KRAUS and CLAIRMONT
(loc. cit.). Agents with a solvent action upon the blood are
formed by cholera micro-organisms and other similar vibriones.
LEVY 1 isolated from typhus cultivations a lysine which acted
most upon dogs' blood.

By means of immunisation with typhoid cultures he succeeded
in preparing an antilytic serum. Typhoid lysine also resists the
action of heat. A streptolysine that was fairly stable when
heated has been described by BESREDKA.2

It only occurs in young cultivations, and varies in its proper-
ties with the nature of the culture medium. It is only destroyed
after two hours' heating at 70° C., and is not dialysable.

It does not form an anti-body under any condition, and thus
is apparently not a haptine at all. For this reason I give no
further particulars about it here.

A hsemolysine isolated from cultures of the pneumococcus has
been described by CASAGRANDI.S It is characteristic of this that
only the non-pathogenic varieties of this diplococcus should form
a lysine. Its constitution is similar to that of the other toxines.
and it forms an anti-body. Certain species are also stated to
produce a specific leucocidine in addition to the lysine.

1 E. and P. Levy, "Ueber die Hamolysine des Ty.-B.," Centralbl. f.
Bakt., xxx., 405, 1901.

2Besredka, " De 1'h^molysine streptococcique," Ann. Past., xv., 880,
1901.

3 Casagrandi, " L'Emolisina e la Leucolisiua Diplococcica," Bull. Soc.
Lands. Rom., xxvii., 2; Biochem. Centralbl., i., 402, 1903.

THE LEUCOCIDINE OF THE STAPHYLOCOCCI. 143

Lastly, specific hcemagglutinines obtained from the filtrates
from bacteria — e.y.t from staphylococcus and different vibriones —
have recently been described by KRAUS and LuowiG.1 They
are destroyed at 58° C. and form specific anti-bodies. They
must be kept quite distinct from the specific hsemolysines.

A lysine of the plague bacillus has also been described by

IJRIARTE.2

THE LEUCOCIDINE OF THE STAPHYLOCOCCI.

Simultaneously with the production of- lysine there is formed
in the case of the typical species of staphylococci a second soluble
toxine, leucocidine, which was independently discovered by VAN
DE YELDE (loc. cit.) and by BAIL,S and was more closely studied
by NEISSER and WECHSBERG.4 Its specific activity is directed
mainly against the leucocytes, which it kills and dissolves, and
also against certain other cells, such as hsematoblasts, ganglionic
cells, &c. Owing to the fact that it appears to destroy the
leucocytes in the living organism also, it produces infarct and
other alterations in the kidneys. It does not appear, however,
to have a specific action upon the epithelium of the kidneys.
NEISSER and WECHSBERG tested its activity by means of their
" bioscopic method" in which they took the reducing power of
the leucocytes as the measure of their vitality, a dilute solution
of methylene blue being used as the indicator. If the leucoci-
dine was active no decolorisation of the methylene blue took
place. In such determinations it is obvious that the quantity of
leucocytes present must be taken into account; this was done
by determining beforehand the single reducing dose, Lr, for each
exudation from the living cells, and taking this result as the
basis of the calculation.

Leucocidine is produced in bouillon cultivations in about four
days, and reaches its maximum in about eight days. As regards
the influence of the alkalinity of the culture fluid almost the
same conditions appear to hold good for leucocidine as for
staphylolysine.

Leucocidine invariably occurs in association with the lysine.

1Krausand Ludwig, "Ueb. Bakteriohamagglutinine," Wien. Idin. Woch.,
1902, 120.

2 Uriarte, " H^molysine du bac. pesteux.," Soc. BioL, Ivii., 254, 1904.

3 Bail, "Ueber leukocide Substanzen in den Stoffwechselprodukten d.
Staph. pyog. aureus," Arch. f. Hyg., xxxii., 133, 1898.

4Neisser and Wechsberg, "Ueber eine neue einfache Methode z. Beob.
von Schadigungen leb. Zellen und Organismen," Munch, med. Woch.,
1902, 1261.

144 TOXINES AND ANTITOXINES.

The same species of staphylococci that produce the lysine also
produce leucocidine, although far from invariably in correspond-
ing quantity. As in the case of the lysine, passage through an
animal increases the production of poison. It is, nevertheless,
a specific poison unmistakably distinct from, the lysine^ and
possesses its own haptophore and toxophore groups. The lysine
does not combine with the leucocytes. Leucocidine passes into
the sterilised filtrates and is thus a soluble poison.

It is somewhat less stable than the lysine. At 50° C. it is
destroyed in twenty minutes, and at 58° C. in ten minutes.
When kept, preserved with phenol, in an ice chest it very
rapidly becomes weaker, its activity being from twenty-five to
sixty times less after sixteen days. Eventually it becomes quite
inactive.

Its action is fairly slow, so that it is essential to extend the
period of observation to two hours. It is shown to be a true
toxine by the fact that it is capable of forming an antitoxine.
DENYS and VAN DE VELDE x were the first to prepare an anti-
leucocidine by means of the injections of the filtrates from
cultivations. Then NEISSER and WECHSBERG discovered anti-
leucocidine in normal horse serum and human serum, but not in
that of the rabbit ; in like manner they invariably obtained an
antileucocidine of uniform character by immunising rabbits and
goats.

We have thus in the leucocidine of staphylococci a true toxine,
which has specific action upon leucocytes — characteristic activity
which has hitherto only had an analogy in the pneumococcus
leucocidine discovered by CASAGRANDI (loc. cit.).

1 Denys and Van de Velde, " Sur la production d'une antileucocidine,
&c.," La Cellule, xi., quoted by Neisser and Wechsberg.

145

II. —THE ENDOTOXINES.

CHOLERA VIRUS.

THE question of the nature of cholera virus, and of its position
with regard to the true toxines, is still far from being definitely
settled. Apart from the unfortunate fact that we meet with
direct contradictions in the experimental results, there is also
one very important circumstance that increases the difficulty
of making a survey in this and many very similar cases, such
as the poisons of typhus, pneumococcus, &c. The earlier
investigators of immunity were not at that time able to draw
the fundamental distinction between antitoxic and bacteri-
cidal immunity, and thus at first no systematic research was
made to determine the existence of a cholera toxine and anti-
toxine. But even at the present day the difficulty of investi-
gation is greatly increased by the fact that immunity against
cholera is undoubtedly, in the main, bactericidal, and that anti-
toxic immunity, even if it exists at all, occupies an absolutely
secondary position. Even inoculation with the dead cells
apparently produces a purely antibacterial immunity.

The history of cholera virus begins with R. Kocn,1 who had
long regarded cholera as a disease due to toxine poisoning,
although it was only after tedious experiments that he suc-
ceeded in causing rapid poisoning, and then only by means
of living bacteria. On the other hand, NICATI and RiETSCH2
obtained poisonous filtrates devoid of specific activity, as was
also done by YAN ERMENGEM.S Then followed the usual inves-
tigations of the soluble crystalloid substances of a ptomaine
character which had been isolated from the cultures, but were
soon recognised as not responsible for the toxic action of the
vibriones. The toxalbumins, too, which BRIEGER and FRANKEL
isolated by their method from cholera cultivations, were found

1 R. Koch, " Vort. uber die Cholera," Berl Idin. Woch., 1884, 498; id.,
"Zweite Conferenz z. Erort. d. Cholerafrage," ibid., 1895 [37a], 8.

2 Nicati and Rietsch, "Effets toxiques des produits, &c.," Comptes
Rend., xcix., 929, 1884.

3 Van Ermengem, "Sur 1'inoculation des produits de culture du bacille
virgulse," Bull. Acad. Med. Be/g., [3], xviii., 1221, 1884.

10

146 TOXINES AND ANTITOXINES.

by them to be substances having only a slight toxic action, and
not producing any specific effects. Their cholera toxalbumin, in
particular, was insoluble in water and not poisonous to rabbits.

The first successful attempt to isolate a poisonous and approxi-
mately specifically active product from Koch's vibrio was made
by PETRi.1 He discovered in peptone cultivations a soluble
poison, which killed guinea-pigs with hypothermia and other
symptoms of cholera poisoning, although fairly large doses
(2 c.c.) were required. His " toxopeptone," it is true, differed
widely from the true toxines in that it could resist boiling.
He also confirmed the far-reaching observation first made by
CANTANi,2 that the cells after being killed invariably still con-
tained an abundance of virus, so that the filtered cultivations
were never as poisonous as the original sterilised cultures.
KLEMPERERS also found that the dead cells were still poisonous
and could act per os when the acidity of the stomach was
partially neutralised and the intestine in a state of rest.

Next come the experiments of HuEPPE4 and SCROLL 5 to obtain a
"toxine" by means of anaerobic cultivations in eggs and precipitation
with alcohol, but their results were contradicted by GRUBER and WIENER/
WESBROOK,7 and DoNiTZ,8 who were unable to prepare active specific
poisons by these means.

GAMALEiA9 concluded that there were two cholera poisons.
He cultivated the vibriones for fifteen days in calf's-foot bouillon
and sterilised the culture at 120° C. The poison thus obtained
caused death preceded by a marked fall of temperature and
paralysis, as well as by hypersemia of the abdominal organs.
There was absolutely no acclimatisation to this poison, which

1 Petri, "Unters. iib. die d. d. Wachstum der Cholerabakt. entstehenden
chemischen Umsetzungen," Arb. Kais. Ges.-Amt., vi., 374, 1890.

2 Cantani, "Giftigkeit der Cholerabazillen," Deutsch. med. Woch., 1886,
789.

3 Klemperer, "Ueb. kiinstlichen Impfsehutz gegen Choleraintoxikation,"
Berl. Tdin. Woch., 1892, 789.

4 Hueppe, "Ueb. d. Aetiologie u. Toxikologie der Cholera asiatica,"
Deutsch. med. Woch., 1891, 417.

6 Scholl, "Unters. iib. giftige Eiweisskb'rper bei Cholera asiatica," Arch,
f. Hyg., 1892, 172.

6 Gruber and Wiener, "Ueb. d. intraperiton. Cholerainfektion," Wien.
Tdin. Woch., 1892, 543.

7 Wesbrook, " Contrib. a 1'etude d. toxines du Cholera," Ann. Past.,
viii., 318, 1894.

8 Donitz, "Ueb. d. Verhalten d. Chol.-Vibr. im Huhnerei," Zeit.f. Hyg.,
1895, 20, 31.

9 Gamalei'a, "Recherches exper. sur les poisons du Cholera," A rch. de
Med. Exper., 1892, 173.

CHOLERA VIRUS. 147

originated from the cells of the vibriones and was of a nucleic
character. In addition to this, however, the cultivations con-
tained a poison, a nucleo-albumin, which was unstable when
heated; it occurred in the cultures that had been sterilised at
58° C., while the filtrates were only very slightly poisonous. It
produced very violent diarrhoea and other choleraic symptoms.

WASSERMANN1 found that the dead cells when used in the
proportion of eight to ten times the quantity of the living
vibriones produced the typical symptoms, ending in death, in
a guinea-pig. By evaporation of the cultures and precipitation
with alcohol he obtained a poison which, in a dose of 0*02 grm.,
killed guinea-pigs but produced no antitoxic immunity. PFEIFFER
and WASSERMANN2 and ISSAEFFS assert that the serum of im-
mune animals does not confer antitoxine immunity. KLEMPERER*
found filtered cultivations to be slightly poisonous.

The question was then further investigated by WESBROOK.
He cultivated the vibriones on alkali albuminate which gave
no biuret reaction. After three weeks there was an unmis-
takable biuret reaction. The poison, when filtered through
porcelain, was fatal to guinea-pigs in doses of 0'5 to T5 c.c.
and possessed immunising capacity.

He attempted to isolate it by neutralising the alkali albu-
minate with hydrochloric acid so that it was precipitated,
concentrating the filtrate in vacuo at 40° C., and dialysing it
after the addition of alcohol. Both the precipitated albuminate
and the albumose left in the filtrate were poisonous.

He also obtained from a rich growth of the bacteria in proteid-
free culture-media (like those employed by USCHINSKY), to which
a certain proportion of sodium hydroxide had been added, a
poison which, after imperfect purification, yielded a brown sub-
stance that gave no biuret reaction and only a faint xantho-
proteid reaction. It had a poisonous action and immunising
capacity. From these results he concluded that cholera virus
was only in combination with proteids, but was not a proteid
itself.

This poison was invariably only sparingly produced in the
cultivations, and was but very slightly poisonous in comparison

1 Wassermann, " Unt. iib. Immun. geg. Choi, asiatica," Zeit. f. Hyg.,
xiv., 35, 1893.

2 Pfeiffer and Wassermann, " Unters. iib. das Wesen d. Choleraimmuni-
tat," Ze.it. f. Hyg., xiv., 46, 1893.

3Issaeff, "Unt. lib. d. kiinstl. Imm. geg. Cholera," Zeit. f. ffyg.,xv.,
287, 1894.

4 Klemperer, " Schutzimpf . d. Menschen geg. asiat. Cholera," Berl. Uin.
Woch., 1892, 970.

148 TOXINES AND ANTITOXINES.

with diphtheria toxine and tetanus toxine, and hence PFEIFFER1
asserted that cholera virus possessed a totally different nature
and mode of production to these true toxines. He put forward
the view that cholera virus was not an excretion product of the,
vibriones, but a substance firmly retained within the cell in the
normal condition, an " endotoxine," which only left the bacteria
after the death of the latter. This would account for the rela-
tively small toxic action of the filtered cultivations compared
with the highly poisonous effects of the living cultures, and, as
was shown by PFEIFFER, of the dead bacilli.

PFEIFFKK advanced, in support of his view, the following facts
which he had observed : — The germ-free filtrate of bouillon cul-
tures is only about half as poisonous as the original cultivation
before boiling. If boiled cultivations are filtered the germ-free
filtrate is more poisonous than that from the unboiled culture.
But if, on the other hand, the filtrate from the untouched culti-
vations is boiled, it loses its toxic property.

If the vibriones are killed by means of chloroform or thymol
the poisonous property is retained, whereas the addition of
alcohol and precipitation with ammonium sulphate have an
injurious effect upon it. When the vibriones are allowed to
dry slowly, so that they die, they retain their toxic power.
If they are then heated with water and the liquid filtered
through a Chamberland filter, the filtrate is non-poisonous ;
nor is glycerin any more effective in extracting the poison
from the dead cells. PFEIFFER 2 concludes that there is a
primary very unstable poison which is converted by heat into
the secondary poison, prepared by SCROLL and others, which while
acting in the same manner is ten to twenty times weaker.

PFEIFFER3 still firmly maintained his theory even after
RANSOM 4 had claimed to have prepared, from cultivations, an
active cholera toxine possessing true immunising powers, and
after SoBERNHEiM5 had, at an earlier period, obtained by filtra-
tion of old cultivations in a rapid state of decomposition a poison
that acted per os, and also produced immunity. RANSOM himself

1 Pfeiffer, " Unters. lib. d. Choleragift," Zeit.f. Hyg., xi., 393, 1892.

2 See also Pfeiffer, " Studien zur Cholera- Aetiologie," Zeit. f. Hyg., xv.,
268, 1894.

3 Pfeiffer, " Ueber die spezifische.n Antikorper der Cholera," Zeit. f.
Hyg., xx., 217, 1895; id., "Ein neues Grundgesetz der Immunitat,"
Deutsch. med. Woch., 1896, Nos. 7 and 8.

4 Ransom, "Choleragift u. Choleraantitoxin," Deutsch. med. Woch.,
1895, 457.

5 Sobernheim, " Experim. Unters. iib. Choleragift und Cholerschutz,"
Zeit.f. Hyg., xiv., 485, 1893.

CHOLERA. VIRUS. 149

gave no further details as to the preparation of his poison, but
BEHRiNG1 subsequently stated that Ransom's poison was obtained
from cultivations of five to ten days' growth, by means of a short
heating at 100° C., filtration through a Pukall's filter, and pre-
cipitation with alcohol.

This solid poison is stated to produce violent choleraic symp-
toms and death when tried upon animals. It causes in the
organism of the animal a slight formation of antitoxine (protec-
tion against four to six times the lethal dose), and in this respect
should be regarded as a true toxine. Against this, however,
must be placed its stability when heated, which, according to
RANSOM, is one of its characteristics, and also its relatively
small toxic action (0-07 grm. of the solid poison is the lethal
dose for guinea-pigs). PFEIFFER, therefore, regards Ransom's
poison as a secondary product, and is of opinion that its
reputed antitoxic power is not greater than that of the normal
serum.

METSCHNIKOFF, Roux, and TAURELLI-SALIMBENI 2 brought
forward the following interesting experiments in opposition to
Pfeiffer's theory : — They placed 3 to 4 c.c. of a peptone solution,
part of which contained living vibriones and in part of which
the micro-organisms had been killed by means of chloroform, in
sterilised collodion capsules, which were then hermetically closed
and placed in the abdominal cavity of guinea-pigs. The animal
treated in this way with the dead vibriones became slightly ill,
whereas that poisoned with the living micro-organisms perished
in three to five days with typical choleraic symptoms. Post-
mortem section showed the usual alterations found in cholera,
but in none of the organs were there any vibriones, although
these were alive in the capsule.

Some of the animals survived and then showed an increased
resistance.

The inevitable deduction to be drawn from this experiment is
as follows : —

In the culture medium contained in this closed capsule which
thus represented a model intestine, there was formed a poison
which could diffuse through the collodion membrane whilst the
vibriones developed vigorously inside the capsule. On the other
hand, but little poison was liberated from the dead cells — only
sufficient to produce slight illness.

1 Behring, ' ' Untersuch. Ransoms iib. die Agglutination der Cholera-
vibriones," Deutsch. med. Woch., 1898, 294.

2 Metschnikoff, Roux, and Taurelli-Salimbeni, ' ' Toxine et antitoxine
cholerique," Ann. Past., x., 257, 1896.

150 TOXINES AND ANTITOXINES.

At first the vibriones showed vigorous increase within the capsule, but
soon they assumed other forms and then no longer increased ; they did not
die, however, but could still be cultivated months afterwards, on nutrient
media. Passed through the body of an animal after intraperitoneal injec-
tion they again yielded cultures of a high degree of virulence.

They then endeavoured to obtain this poison also from culti-
vations of the vibriones thus made virulent. Old cultures have
relatively little toxic action, and they, therefore, used quite
young cultivations of two to twenty-four hours to three or four
days' growth.

These when filtered had an average toxicity of O'S c.c. per 100 grms. of
animal. The addition of serum to nutrient liquids increased the degree
of toxicity and also the development of a torula form in the cultivation.

The poison thus obtained was also stable when boiled, but was
soon rendered inactive by air and light. When hermetically
sealed up it kept for a long time. Alcohol and ammonium
sulphate did not precipitate it. It behaved in a manner
analogous to Ransom's poison. By injection of increasing doses
it was possible to produce an antitoxic immunity, which although
greater than that of normal serum, was yet exceedingly insignifi-
cant compared with that of diphtheria and tetanus antitoxines.
In the most favourable case 1 c.c. neutralised six times the
lethal dose ! In this respect its behaviour is the very opposite
of Pfeiffer's bactericidal antiserum, obtained by means of im-
munisation with dead vibrio cells, which is devoid of any
neutralising effect upon the toxine. This is an argument against
the view that the dead cells contain any appreciable amount of
immunising toxine, since otherwise this, when liberated, would
necessarily show its power of forming an ti toxine.

COURMONT and Do YON 1 also obtained a soluble poison that
could be filtered ; it had an exceedingly small toxic effect, the
lethal dose for a rabbit being 4 c.c. (!). It produced hypothermia
and hsemorrhagia, also paralysis and peripheral neuritis. It
was extremely sensitive to the influence of light and air. Culti-
vations sterilised at 50° C. proved somewhat more poisonous.

HAHN,2 on the other hand, employing Buchner's method of trituration
and expression at a pressure of 4 to 5 atmospheres, isolated from quite
young cultivations, a cholera plasmine, in the form of a yellowish-brown
liquid, which when injected in doses of 0'5 to 0*6 c.c. protected guinea-pigs
after eight days against ten times the lethal dose. Particularly interesting

1 Courmont and Doyon, " Effets de la toxine choler.," Arch, de Phys.,
xxviii., 785, 1896.

2Hahn, ' ' Immunisierungs- u. Heilversuche mit den plasmat. Zellsaften,
&c.," Munch, med. Woch., 1897, 1344.

CHOLERA VIRUS. 151

points in connection with these plasmines are that they are only very
slightly poisonous, and that the immunity that they confer is apparently
a true bactericidal one ; it is not as yet possible to prove that the active
substances expressed are those that cause the formation of the bactericidal
intermediate products. This question is of great theoretical importance
and will be specially dealt with when we come to consider Tuberculine
(q.v.). Buchner's plasmines cannot help us in the matter of the poisons
and antitoxines of cholera.

It is not altogether easy to form a definite conception of the
nature and mode of action of cholera virus from these apparently
contradictory experimental results.

For this purpose the question must be formulated in a more
precise manner in accordance with our theoretical conceptions
concerning bacterial toxines.

The question is then no longer: " Do the vibriones produce
a soluble poison that gives rise to choleraic symptoms'?" but
it is necessary to ask whether they produce a toxine — i.e., a
poison which enters into combination in a specific manner, and
eventually produces specific antitoxic immunity.

When we put the question in this form the following conclu-
sions may apparently be drawn with some degree of probability
from the experimental results.

It appears to be the fact that the vibriones produce a true
toxine, which is very sensitive to all external influences, and
even through the ageing of the cultivations undergoes a change
in a manner that has yet to be further investigated. This toxine
differs materially, however, from those of the diphtheria type, in
that it is not, like such toxines, secreted in the free state and
almost without residue by the vibriones, but appears to be
energetically retained by them intra vitam. We here meet with
peculiarities similar to those that occur in the case of true
ferments. While, for example, yeast excretes only one enzyme,
yeast diastase, it also contains others — e.g.,invertase and maltase —
which are not given up by the living yeast cell, and can only be
extracted after the cell has been killed, or after its membrane
has been ruptured by means of glass powder.

In addition fco these the yeast cell also contains E. Buchner's
zymase, which can only be set at liberty by the very drastic
means of trituration and expression under high pressure.

Perfectly analogous conditions are found in the case of the
inverting enzyme of Monilia Candida, which is also active after
the addition of toluene, but has not yet been isolated from the
cell of the mould-fungus by any of these methods.1

1 For further details of these ' ' Endoenzymes " see my Ferments and
their Actions, Griffin & Co., London.

152 TOXINES AND ANTITOXINES.

And the cholera vibrio retains such ferment within itself in
firm combination, but capable of isolation by drastic means, as
was shown by GERET and HAHN when they isolated a proteolytic
enzyme. In like manner, then, the toxine also appears to be
firmly combined with the living cell, to be an endotoxine analo-
gous to the endoenzymes. After the death of the cell, however,
it passes, partially at all events, into the culture medium, just
like yeast invertase, so that, as PFEIFFER has shown, cultures
killed by thymol, &c., or by drying, give indications of the poison
they contain. But that the toxine, after the death of the
vibriones, is really free, dissolved in the medium, and diffusible,
is shown, again, by the experiments of METSCHNIKOFF, Roux,
and TAURELLI-SALIMBENI, in which it was found that living
vibriones enclosed in a collodion capsule exerted their toxic
activity outside the membrane. For, without doubt, in an ex-
periment on these lines, of luxuriant growth on a restricted
culture medium, a process of death and decomposition would
take place simultaneously with the development of new vibriones.
And since in this case the toxine can, immediately after its
production, exert its activity upon the organism, while itself
protected from every injurious influence, we may conclude with
some degree of probability that here the primary true cholera
toxine is the active agent. Probably this presumably true toxine
is also active in cases of real cholera in the living organism.
And this toxine ought, therefore, also to produce an antitoxic
reaction in the organism. That this reaction is so slight ought
not to excite surprise, for, in the first place, the amount of toxine
thus set free can only be very trifling, so that a high degree of
immunisation is not to be expected ; and, further, this degree
of antitoxic activity is lacking in a complete poison that has
undergone secondary alterations, and, as we shall see below,
probably contains numerous toxoids, the result being that the
antitoxic power of the serum must appear too small. Thus the
cholera vibriones, after death and decomposition, give up part of
their endotoxine to the culture media ; at the same time it is
highly probable that they still retain the larger proportion of the
toxine, just as zymase is retained by yeast. Possibly the method
of aseptic autolysis tried by CONRAUI l upon typhoid bacilli may
enable us to prepare the primary cholera poison.

This primary poison, a true toxine, is extraordinarily sensitive
to external influences, in which respect it resembles zymase, and
its full activity can therefore only be demonstrated under con-

1 Conradi, "Ueber losliche, durch Autolyse erhaltene Giftstoffe,"
Deutsch. med. Woch., 1903, No. 2.

CHOLERA ANTITOXINE. 153

ditions as favourable as those of the above-described experiment
of METSCHNIKOFF and his collaborators.

A further conclusion to be drawn is that cultivations that
have been killed contain only a very small amount of the toxine,
and this accounts for the want of success in the attempts to
isolate it as a chemical entity, and also the but trifling toxic
action of the collodion capsules containing the dead cells.

As the vitality of the cells grows weaker the proportion of
true toxine present also decreases very rapidly, so that in old
cultivations or those that have been destroyed there is little
poison present except that secondary heat-resisting product that
has been described by various observers. This is certainly no
longer a true toxine, since it resists boiling and is relatively but
slightly poisonous. This poison, too, is only secreted to a small
extent, but is, at all events, sufficiently stable to be isolated as
a definite substance distinct from the body cells of the vibriones.
It has, moreover, a slight power of forming an antitoxine ; it
thus shows all the characteristics that belong to Ehrlich's
toxoids — viz., greatly reduced toxic capacity, greater stability, and
the power of forming an antitoxine, as well as of entering into
combination.

Hence, from these considerations, we can conclude with some
degree of probability that the cholera vibrio produces a true toxine,
an endotoxine, comparable with yeast invertase, which is only
separated scantily, if at all, from the living cell, and that this
toxine is extremely unstable, and is very readily transformed
into a secondary mixture of poisons rich in toxoids.

At the same time, it is highly probable that, in addition to
this poison so scantily secreted, the cholera vibrio, presumably,
like all bacteria, also contains in its protoplasm a simple non-
specific bacterial protein, which produces symptoms of inflam-
mation, as, for example, that of the diphtheria bacillus. Only in
this case it is not possible to separate the enzymic poisons so
quantitatively as in the case of diphtheria, so that it is difficult
here to obtain objective proof whether any such poisonous protein
is present, in addition to the specific poisons.

CHOLERA ANTITOXINE.

We have shown above that it is probable that there is a cholera
toxine which produces antitoxine to a small extent. No decisive
unquestionable proof, however, that the serum of the animals in
the experiments contains an active antitoxic substance has been
given ; and so long as cholera antitoxine has not been actually

154 TOXINES AND ANTITOXINES.

detected in the fluids of the body, the whole question whether
KOCH'S vibrio produces a true toxine must be regarded as an
open one.

The question whether or no the serum of the animals used in
the experiments confers antitoxic immunity does not necessarily
coincide with that which is alone important, from the theoretical
point of view — viz., Does it contain anlitoxine ? This antitoxine
may possibly be very unstable, or, owing to its small quantity
and its great dilution when introduced into the animal, it may
become inactive. This would explain the negative results
obtained by PFEIFFER, KOLLE, WASSERMANN, ISAEFF, and others
in experiments on animals, and of LAZARUS* on convalescent
cholera patients, assuming that there really exists a cholera
antitoxine, proof of which has certainly not yet been brought.

TYPHOID VIRUS.

Almost the same description might be given of typhoid as of
cholera virus, only that in this case the experimental material is
far more scanty, and the question of the existence of a toxine
and antitoxine has hardly passed the first stages of discussion.

All that is certain is that typhoid bacilli can, under certain
conditions, produce true toxine poisoning, and that the bacteria
must therefore produce an active poison. On the other hand, it
is equally certain that immunity against typhoid bacilli is essen-
tially, as in the case of cholera, not antitoxic, but bactericidal,
and antitoxic immunity, if it occurs at all, is only of slight
importance.

Here, too, the history of the investigations of the poison begins with the
researches of BmEGER,'2 who first isolated typhotoxine, and subsequently
his toxalbumin. These and similar preparations are not the specific
typhoid poison.

On the other hand, as was found by BEUMER and PEIPER,S
and subsequently by CHANTEMESSE and WiDAL,4 cultures steri-
lised at 100° to 120° C. in autoclaves are poisonous, the dose
being from five to six times that of the living cultures ; the older
they are the more toxic they become.

1 Lazarus, " Ueb. antitoxische Wirksamkeit des Blutserums," Berl. klin.
Woch., 1892, Nos. 43-44.

2 Brieger, Weiteres fiber Ptomaine, Berlin, 1885.

3 Beumer and Peiper, "Bakt. Stud. lib. Typhusbaz.," Zeit. /. Hyg.,
ii., 110, 1887.

4 Chantemesse and Widal, " L'immunite centre le virus de la fievre
typhoide," Ann. Past., ii., 54, 1888. Id., " £t. experim., &c., de 1'infec-
tion typhique," ibid., vi., 755, 1892.

TYPHOID VIRUS. - 155

Soluble poisons were prepared by BRIEGER, KITASATO, and
WASSERMANN (loc. cit.) by heating the liquid cultures to 80° to
90° C. and precipitating with alcohol, and also by concentrating
them at 37° C. before the addition of the alcohol. These had
slight toxic and immunising properties, while the nitrates were
absolutely inactive. SiROTiNiN1 obtained a poison possessing
specific^) activity by filtration, but PFEIFFER and KoLLE's2
results were a direct contradiction of this. BITTER'S 8 prepara-
tion was obtained by extraction with concentrated glycerin and
evaporation in vacuo at 36° C.

SANARELLI* cultivated very poisonous typhoid cultures on
glycerin bouillon for a month at 36° C., and then sterilised them ;
these cultures were frequently rendered more poisonous by intra-
peritoneal injection into mice. From them he isolated, by means
of maceration for several days at 60° C., a very weak soluble
poison which produced specific symptoms in the mucous mem-
brane of guinea-pigs and monkeys, especially that of the intestine.
The lethal dose for rabbits amounted to 10 c.c. per kilo. There
was nothing specific, however, in the symptoms in the case of
this animal.

RoDET5 found that filtered typhoid cultivations were slightly
toxic, causing elevation of temperature and local signs of necrosis ;
the residual cells had hardly any poisonous action. Special
importance attaches to the results obtained by CnANTEMESSE,6
who obtained a very active typhoid toxine speedily decomposing
in the air, by cultivation of the bacilli on an extract of spleen
which had been digested with pepsin and again neutralised.
Here, too, we find the same behaviour as in the case of cholera,
the poison being not completely destroyed at 100° C. — i.e.,
doubtless being converted into a less poisonous, but more stable,
modification. It is possible that in this case also we have to
deal with toxoids.

It acts upon susceptible animals, producing drowsiness, para-

1 Sirotinin, "Die Uebertrag. von Typhusbazillen auf Versuchstiere,"
Zeit.f. Hyg., i., 465, 1886.

2 Pfeiffer and Kolle, "Ueb. d. spez. Reaktion d. Typhusbazillen," Ze.it.
f. Hyg., xxi., 203, 1896.

3 Bitter, ' ' Ueb. Festig. v. Versuchstieren geg. d. Intoxikation durch
Typhusbazillen,';^^./. Hyg., xii., 298, 1892.

4 Sanarelli, "Etudes sur la fievre typhoide expe"rim." Ann. Past., viii.,
193, 1894.

5 Rodet, " Sur les propriete's toxiques des cultures des bacilles d'Eberth,"
Soc. Bid., 1., 774, 1898.

6 Chantemesse, "Toxine typhoide soluble," Prog. Mtd., 1898, 245; id.
" Losliches Typhustoxin," Wien. med. Blatter, 1898, 18 et seq.

156 TOXINES AND ANTITOXINES.

lysis, and also, especially on intravenous injection, acceleration
of the pulse and lowering of the blood-pressure. Introduced
per os it is harmless.

VINCENT 1 found that filtered cultures of EBERTH'S bacillus
were very poisonous when injected into the cranium, and that
even animals rendered immune to typhoid were not protected
against its action.

Hens and pigeons are almost refractory, while other animals
show varying degrees of susceptibility. Rabbits are three times
as susceptible as guinea-pigs.

Soluble poisons were also found by MARTIN 2 in filtered
typhoid cultivations ; he concluded that the poison retained in
the cells was identical with this. He cultivated the bacilli in a
bouillon partially prepared from extract of spleen and containing
alkali albuminate.

The collodion capsule method has been tried by RODET and
GuECHOFF,3 but with little success.

CONRADI 4 obtained typhoid poison by aseptic autolysis of the
bacilli, mixing them with a 0*8 per cent, solution of sodium
chloride and leaving them for at most forty-eight hours in the
incubating chamber. In this way he prepared a solution of a
cell-free poison, which in a dose of 2 c.c. killed guinea-pigs in
twenty-four hours.

This method and also that of MACFADYEN and ROWLAND 5 are
particularly promising for the elucidation of this question in the
future. MACFADYEN triturated the bacteria at the temperature
of liquid air and thus obtained a poison that caused acute
symptoms, and also, according to his preliminary experiments,
produced antitoxic immunity.

This typhoid endotoxine has been described more fully by
MACFADYEN 6 in a detailed investigation recently published. It
was found to be very poisonous. In immunisation experiments

1 Vincent, "Inoculation intercranielle du bacille Eberth, &c.," Soc. BioL,
lv., 1214, 1903.

2 Martin, "Die chem. Prod. path. Bakt.," Wien. med. Blatter, 1898,
No. 25 et seq.

3Rodet, A. and G. Guechoff, "Versuche die Methode der Kollodium-
sackchen auf die Kenntnis der toxischen Produkte des Eberthschen Bacillus
und des B. coli anzuwenden," Soc. BioL, lii., 962, 965, 1900.

4Conradi, " Ueber losliche, durch aseptische Autolyse erhaltene Gift-
stoffe," Deutsch. med. Woch., 1903, No. 2.

5Macfadyen and Rowland, "An intracellular Toxine of the Typhoid
Bacillus," Proc. Roy. Soc., Ixxi., 77, 1902, also Centralbl. f. Bakt., xxxiv.,
618, 1903.

6Macfadyen, " t)ber das Vorkommen intracellularer Toxine," Z. fur
allgem. Phys., iii., 303, 1904.

DYSENTERY. 157

it had a protective action against both the toxine and infection
with living bacilli.

Whether these soluble poisons are true toxines or toxoids — i.e.,
whether or no they produce antitoxines in the body — is still more
doubtful than in the case of cholera. PFEIFFER and KOLLE
absolutely deny that there is any excretion of a free soluble
poison or formation of antitoxine, whilst BITTER concludes that
there is a slight production of antitoxine. On the other hand,
CHANTEMESSE observed an unmistakable and energetic formation
of antitoxine, after the injection of his typhoid toxine, especially
in the case of the horse.

We have thus, as in the case of cholera, a poison firmly
retained in the cells of the bacilli, and only to be separated from
them with difficulty. Our theoretical conclusions must, there-
fore, with all reserve, be very similar to those we have drawn
in the case of cholera virus — viz., that there possibly exists a
typhoid toxine, but that it is hardly excreted in the free state,
and that the poisons that have been prepared are greatly altered
secondary products.

BACILLUS COLI COMMUNIS.

The only statements that the coll bacillus produces a toxine
are those of BARBA-MORRHY l (of whose results only a short
abstract giving no particulars of the nature of the poison was
accessible to me), and that of RODET, who discovered them at
the same time in typhoid cultures (q.v.). According to MARTIN'S
results (loc. cit.) the coli bacillus and also Gaertner's B. enteritidis
appear to behave in quite an analogous manner to the typhoid
bacillus as regards the production of their poison.

VAUGHAN 2 by heating the cells with dilute (1 per cent.)
sulphuric acid, obtained a poisonous substance which was cer-
tainly not a toxine. With regard to colilysine, vide supra.

DYSENTERY.

CONRADI (loc. cit.) has succeeded, by means of the method of
aseptic autolysis, in preparing a soluble dysentery poison.

After eighteen hours' autolysis he obtained poisons which
killed large rabbits after the intravenous injection of 0-1 c.c.
The symptoms were violent diarrhoea, paralysis, decrease of
temperature, &c. In cases of chronic poisoning with smaller
doses there were also intestinal ulcers and all the pathological
and anatomical appearances of dysentery.

1 Barba-Morrhy, Baumgartens Jb., 1897, 403.

2Vaughan, "The intracellular Toxines of some Pathogenic Bacteria,"
J. Amer. Med. Assoc., 1903, 828 ; Biochem. CentralbL, i., No. 1,056.

158 TOXINES AND ANTITOXINES.

PLAGUE TOXINE.

The plague bacillus also produces soluble poisons which pro-
bably belong to the true toxines, although the question is still
unsettled.

The nitrates of quite young cultivations have no poisonous
action, as was shown by the concordant results of the German
Commission,1 WERNiCKE,2 ALBRECHT and GHON.S

On the other hand, toxic properties are developed in old cul-
tivations (even after five days), and increase with their age.
The cultures cause death, preceded by emaciation, degenerative
changes in the liver, and weakness of the heart's action. The
plague toxine has been most thoroughly studied by MARKL.

MARKL4 found, first of all, that the cells of bacilli that had
been killed by chloroform were very poisonous, but subsequently
made use of the filtrates from bouillon cultures instead, which
proved fully as poisonous.

The toxic power was particularly great in the case of old
cultures that had been grown at a lower temperature (about
20° C.), and it increased up to about the second month, after
which it became stationary, and then decreased until eventually
the toxine disappeared. Plentiful admission of air to the cultures
was most essential. The temperature of incubation had an in-
jurious effect upon the toxines.

The lethal dose of the most active poisons amounted to 0-005
to O'Ol c.c. for mice and to about O'l c.c. for rats.

In the case of these animals the poisoning proceeded rapidly,
with symptoms of collapse, but without any anatomical altera-
tion except fatty degeneration of the liver; with rabbits and
guinea-pigs, however, it only acted in this way when very large
doses had been given, and otherwise its action was more pro-
tracted, extending over several weeks. Occasionally a splenic
tumour and pigmentary atrophy of the liver were produced. In
the case of a cat he observed, in addition to marasmus, loss of
hair and wide-spread necroses of the skin. In guinea-pigs it pro-
duced rigor and extreme lowering of the temperature (25° C.).

1<<Bericht der dentschen Pest-Comm.," Arb. Kaiserl. Ges. Ami., xvi.,
1899.

2 Wernicke, " Ueb. Immun. Vers. b. d. Beulenpest," Centralbl. f. Bah.,
xxiv., 1894.

3Albrecht and Ghon, "Bakt. Unters. itb. d. Pestbac.," Wiener A kad.,
Ixvi., 1898.

4 Markl, "Beitrag z. Kenntnis der Pesttoxine," Centralbl. f. Bakt.,
xxiv., Nos. 18-20 (Bibliography), 1898; id., " Weit. Unters. iib. Pest-
toxine," Zeit.f. ffyg., xxxvii.,401, 1901.

PNEUMOTOXINE. 159

The poison is very sensitive to external influences. Even at
the ordinary temperature the toxic power of the solutions rapidly
decreases, and this loss takes place more rapidly at high summer
temperatures (25° C.) and immediately at 70° C. It is true that
nitrates thus heated are still poisonous to guinea-pigs and rabbits
when introduced in large doses, but in quite a different way,
possibly on account of the formation of toxoids.

MARKL has also endeavoured to purify plague toxine, but his
experiments (precipitation with alcohol) are little more than
preliminary. He has found that it is intimately associated,
like all true toxines, with albuminous substances.

It is probable that he was dealing, in the main, with
secondary poisons which are formed in such old cultures.
They cause a formation of antitoxine, even if only to a slight
extent.

KOSSEL and OvERBECK/too, succeeded in producing immunity
by means of filtered cultivations that had been heated to 60° C.

PNEUMOTOXINE.

The state of affairs with regard to pneumococcus, the cause
of croupous pneumonia, appears to be very similar to that found
in the case of cholera and typhoid.

The first important experiments throwing light upon the poison
of this diplococcus were those carried out by the brothers KLEM-

PERER.2

They attempted to produce immunity by means of cultures
which, although weakened or inhibited, still contained the cells
of the cocci ; but they also employed germ-free solutions for the
purpose.

They tried not only the simple filtrates from cultivations, but
also a purulent pleural exudation of which the pneumococcus was
the exciting cause, and which had been proved experimentally
to be free from bacteria, in addition to the heated sputum of
pneumonia patients and glycerin extracts of agar cultivations
filtered from the cells of the bacteria.

In this way they obtained fairly weak toxic substances pro-
ducing no specific effects ; by heating these products to 60° C.

1 Kossel and Overbeck, " Bakt. Unters. uber Pest," Arb. Kaiserl. Ges.-
Amt., xviii., 1901.

2 G. and F. Klemperer, "Vers. iiber Immunisierung u. Heilung bei der
Pneumokokkeninfektion," Berl. Ein. Woch., 1891, Nos. 34, 35; G. Klem-
perer, "Die Bezieh. versch. Bakteriengifte z. Immunitat u. Heilung,"
Zeit.f. klin. Med., xx., 165, 1892.

160 TOXINES AND ANTITOXINES.

they succeeded in eliminating their toxic power almost com-
pletely without destroying the immunising capacity. The serum
of the animals used in the experiments contained a specific
antitoxine which neutralised the dissolved poison in vitro and
after previous injection. At the same time, both the toxic
capacity and formation of antitoxine were very limited in
extent.

Almost identical results were obtained about the same time
by FOA and CARBONE] and ScABiA,2 who also succeeded in
producing antitoxic immunity by means of germ-free filtrates
and glycerin extracts of blood from animals infected with
pneumonia.

BELFANTi,3 too, produced immunity, though only to a very
slight extent, by means of filtered germ-free sputum.

FOA also succeeded in separating the immunising principle by
precipitation with alcohol or ammonium sulphate.

PANE4 found filtered cultivations to be very slightly poisonous.
WASHBOURNE5 states that he, too, was able to produce immunity
with filtered cultures.

IsAEFF6 found that pneumococci produced only slightly toxic
products in the ordinary culture media, but that their toxic
power could be considerably increased by frequent, at least
twelve, passages through rabbits. From the blood of the heart
of such rabbits he obtained, by the addition of 1 per cent, of
glycerin and some sodium carbonate, followed by filtration
through a Chamberland filter, a poison which killed rabbits when
introduced in the proportion of 1 per cent, of their body weight,
and was greatly weakened when heated at 70° C., and destroyed
at 100° C. The fluid from the peritoneum of such animals was
very poisonous, but not fatal, after being sterilised by heat.
Although he, too, produced a certain degree of immunity by
means of germ-free filtrates, he yet asserted that he was abso-
lutely unable to observe any antitoxic immunity, and that it was
only antibacterial.

1 Foa and Carbone, "Sulla immunita verso il diplococco pneumonico, "
Gazz. med. di Torino, 1891, 1 ( Centralbl. f. Bah., x., 768).

2 Foa and Scabia, "Sulla immunita della pulmonite," ibid., 1892, 13-15
(Centralbl. f. Bakt., xi., 615).

3 Belfanti, "Sulla immunisazione per mezzo di filtrati di sputo pneu-
monico," Rtf. Med., 1892, 126 (Centralbl. f. Bakt., xii., 401).

4 Pane, " Ueber d. Heilkraft d. antipneumon. Serums," Centralbl. f.
Bah., xxi., 664, 1897.

5 Washbourne, ' ' Experiments with the Pneumococcus, " Journ. of Path. ,
iii., 142; Baumgartens Jb., 1895, 62.

6 Isaeff, " L'immunit^ centre le pneumocoque,'Mnn. Past., vii., 259, 1893.

GONOCOCCAL POISON. 161

obtained toxines which, in large doses, caused death,
preceded by fever, diarrhoea, and loss of weight ; he claims to
have produced antitoxic immunity by means of innoculation
with cultivations weakened at 50° C.; the serum is stated to
neutralise the toxine in vitro.

CARNOT 2 succeeded in producing typical croupous pneumonia
by intrapulmonary injection of 2 to 6 drops of pneumotoxine.

CARNOT and FOURNIERS were able to keep pneumococci alive
and virulent for a long time, when they used as culture media,
blood, serum, or, still better, fresh cerebral substance, while
sterilised culture-media prepared from these were less satis-
factory. They separated without delay, by means of dialysis,
the toxines that were formed, and in this way obtained poisonous
dialy sates, which they concentrated by evaporation in vacuo, or
by precipitation with nascent calcium phosphate. The symptoms
produced by them were similar to those caused by poisoning
with the living cocci. The same investigators 4 found that their
poison had an intense action upon the muscular tissue of the
heart and blood-vessels ; even after the introduction of small doses
very acute inflammation and perforations were produced.

The general conclusion to be drawn from all these observations
is that the existence of a secreted true pneumotoxine has not
been established beyond doubt.

GONOCOCCAL POISON.

The gonococcus also produces a poison which appears to belong
to the group of cholera-typhoid poisons inasmuch as it only
passes to a slight extent into the culture fluids, and rather is
mainly retained in the bacteria, and is probably only given
up in a slight degree to the liquid media after the decomposition
of the cells. The first investigations into this poison were made

by A. WASSERMANN.5

He succeeded in cultivating the gonococci on a nutrient
medium which was prepared from pig's serum, to which had been

1 Mennes, "Das Antipneumokokkenserum," Zeit.f. Hyg., xxv., 413, 1897.

2 Carnot, "Reprod. experim. de la pneumonic fibrineuse," Soc. Biol.,
li., 927, 1899.

3 Carnot and Fournier, " Sur le pneumocoque et ses toxines," Arch.
Med. Exper., 1900, 357.

4 Carnot and Fournier, "Lesions Cardiaques et musculaire par la toxine
pneumon.," Soc. Biol., Hi., 143, 1900.

6 A. Wassermann, ' ' Gonokokkenkultur und Gonokokkengift," Berl.
klin. Woch., 1897, 685; "Weitere Mitteilungcn iiber do.," Zeit. /. Hyg.,
xxvii., 298, 1898.

11

162 TOXINES AND ANTITOXINES.

added 2 per cent, of nutrose, and contained peptone and serum
albumin as necessary ingredients.

While he found that the living gonococci distributed in a
physiological solution of salt had absolutely no toxic effect upon
animals, the sterilised culture, on the other hand, had a fairly
energetic poisonous influence, and killed the animals by periton-
itis on intraperitoneal injection, though it also produced, e.g.,
keratitis with hypopyon in the eyes, &c.

The nitrate had but a slight poisonous action. Hence it would
seem that the poison, just as in the case of cholera, is retained
within the cells. In its resistance to heat, even at the tempera-
ture of boiling, it also resembles the so-called cholera poisons.

Very similar results were obtained, independently of WASSER-
MANN, by NicoLAYSEN,1 who, by means of sterilised, non-filtered
cultivations, produced purulent gonorrhoea in animals.

He found the lethal dose for a mouse to be 0-3 c.c. He, too,
is of opinion that the poison is firmly retained within the cells,
for he was unable to separate it either by extraction with dis-
tilled water or with dilute soda solution.

Almost simultaneously ScHAFFER2 succeeded in obtaining from
a cultivation of four days' growth an aqueous decoction of ascites
and flesh, a poison which, after filtration through a porcelain
filter, produced on the urethral mucous membrane acute sup-
puration which rapidly disappeared. At about the same time
the question of gonotoxine was studied by CHRISTMAS.3 He
obtained, by means of filtration through porcelain, from cultures
of ten to fifteen days' growth, in media of ascites and bouillon,
a poison which he was able to concentrate by precipitation with
alcohol. He also prepared a poison by evaporating the cultures
with glycerin at 50° C. This was stable at 50° C. to 70° C., and
could be kept for six months in the dark. It had a very virulent
effect upon animals, producing, in addition to local inflammation,
cachexia, ending in death. It only acted upon the mucous
membrane of the urethra in the case of man.

CHRISTMAS4 subsequently made a new investigation of the
subject. He found that the living gonococci contained poison,
but that, on the other hand, the dead cells yielded none when
macerated at 20° C. He cultivated the gonococci on a nutrient

1 Nicolaysen, " Zur Pathogenitat und Giftigkeit der Gonokokken,"
Centralbl.f. Balct., xxii., 305, 1897.

2 Schaffer, "Beitr. z. Frage d. Gonokokkentoxine," Fortschr. d. Afed.,
1897, 813.

3 Christmas, " Le Gonocoque et ses toxines," Ann. Past., xi., 609, 1897.

4 Christmas, "Contrib. a 1' etude du Gonococcus," Ann. Past., xiv., 331,
1900.

STREPTOTOXINE. 163

medium consisting of 75 per cent, of ascites and 25 per cent, of
bouillon, and filtered the cultures through talc on the filters or
through kieselguhr, since the porcelain filter retained the poison.
The maximum of toxic power was reached on the twentieth day.
He then precipitated the poison with ammonium sulphate or
alcohol. If subsequent dialysis was employed the symptoms of
poisoning, it is true, were different, but the lethal dose remained
unaltered.

The poison was completely destroyed at 75° to 80° C., but
could resist a temperature of 60° C.

Its action, on subcutaneous injection, was very weak, the
lethal dose for guinea-pigs being 5 to 10 c.c. On the other hand,
it acted very energetically after intercerebral injection, producing
convulsions and dyspnoea, and killing a guinea-pig, even in a dose
of O002 c.c., within six hours.

It is stated to possess strong immunising properties, affording
protection up to one hundred times the dose of poison after
intercerebral injection, but to a much smaller extent after sub-
cutaneous injection. This antitoxic immunity has not been
observed in the smallest degree by other investigators.

MOLTSCHANOFF 1 prepared a toxine from a culture in hydrocele
fluid and yeast peptone bouillon. He made use of unfiltered
cultures, which he sterilised at 70° C. He made a special study
of the action of the poison upon the central nervous system, but
I cannot give the details of his results here.

GROSZ and KRAUSS and ScHoi/rz3 absolutely deny the specific
nature of gonotoxine, and simply regard it as a non-specific, pus-
producing bacterial protein; they point out that SCHAFFEB'S
urethritis, at all events, can also be produced by the living or
dead cells of other bacteria (pyocyaneus, coli, &c.).

"We have, therefore, again to draw the same conclusions about
gonococci poison as in previous instances. The existence of
a specific antitoxine-producing gonotoxine has not yet been
proved.

STREPTOTOXINE.

In considering the metabolic products of the streptococci we
meet with the same obscurity as in the case of gonotoxine. The
existence of a true anti toxin e-forming toxine is still open to
question.

1 Moltschanoff, " Ueber das Gonokokkentoxin und seine Wirkung auf
das Centralnervensystem," Munch, med. Woch., 1899, 1013.

2 Grosz and Kraus, " Bakteriol. Studien iiber den Gonococcus," Arch.f,
DermatoL, xlix., 3, 1899.

3 Scholtz, "Beitr. z. Biologic d. Gonococcus," ibid.

164 TOXINES AND ANTITOXINES.

Certain investigators categorically deny that streptococci
produce any specific poison at all.

ARONSON l not only found the sterilised filtrates from cultures
of streptococci to be devoid of toxic power, but was also unable
to detect any poison in the sediment of cells of the cocci that
had been killed by chloroform ; local infiltrations were the only
results of his inoculations. The existence of a streptococci
poison is also denied by DE GIAXA and PANE.2

At the same time, so many investigators have found poisons,
at all events, in streptococci cultivations, that further proof is
required before this radical conclusion can be accepted.

One of the first poisons to be demonstrated in Jittered cultures
was found by MAN FRED I and TRAVERSA 3 in cultivation of
erysipelas streptococci, grown preferably at 28° to 30° C. They
described it as a readily oxidisable poison, which produced
convulsions and paralysis, and soon became inactive on ex-
posure to the air. In like manner poisonous metabolic products
of streptococci were found in filtered cultures by ROGER, 4

MARMOREK,5 HOMEN,6 FRIEDRICH,7 LAITINEN,8 ClAUDE,9 PARAS-

CANDALO,10 and SCHENK."

As regards the nature of the poison little or nothing is known.
ROGER precipitated his poison from the cultivation by the
addition of ten times the volume of alcohol, redissolved the
precipitate in a solution of common salt, and endeavoured to

1 Aronson, "Ueber Antistreptokokkenserum," BerL klin. Woch., 1896,
717.

2 de Giaxa and Pane, "Contribute alia cognizione sulla immunis. centre
la infez. da streptococco, " Riform. Med., xii. [4], 5; Baumgartens Jb.,
1896, 23.

3 Manfred! and Traversa, " SulPazione del prodotti di cultura dello
streptococco," Giorn. Inter, delle Scienze Mediche, 1888.

4 Roger, "Action des produits solubles streptocoque d'erysipe'le," Soc.
BioL, xliii., 538, 1891.

5 Marmorek, "Le streptocoque et le se'rum antistreptococcique," Ann.
Past., ix., 593, 1895.

6 Home'n, De 1'action du streptocoque et de ses toxines sur les nerfs,"
Sem. Med., 1896, 211 (Soc. BioL, xxiii., [5], 1896).

7 Friedrich, "Beobacht. iib. d. Wirkung von subkutan einverleibten
Streptokokken-Toxinen, &c.," BerL klin. Woch., 1895, 1065.

8 Laitinen, "Das Streptokokkentoxin u. seine Wirkung auf das Nerven-
system," C. f. allg. PathoL, 1896, 358.

9 Claude "Myelite ai'gue par toxines streptocoq," Soc. BioL, xlviii., 122,
1896.

10 Parascandalo, "E. neue Versuchsreihe iib. de Serotherapie bei Infek-
tionen mit pyogenen Mikroorg," Wiener klin. Woch., 1897, 937.

"Schenk, " Ueb. Streptok. -Serum und Streptok.-Toxine," Wien. klin.
Woch., 1897, 937.

STREPTOTOX IN E. 165

obtain by means of this preparation, insoluble in alcohol, the
typical general symptoms produced by the living cultivations.
His "poison" had, however, only a slight toxic activity (13 to
20 c.c. per kilo, of body weight) ; it was weakened, but not
destroyed, at 104° C.

Most of these investigators, however, simply contented them-
selves with investigating the toxic action of filtered cultures, or
of cultures after sterilisation at 65° to 70° C., or by the addition
of O5 per cent, of phenol.

SCHENK alone (loc. cit.) endeavoured to obtain a purer prepara-
tion by means of precipitation with zinc chloride and subsequent
treatment by Brieger's method.

MARMOREK found that the toxine was weakened even at 58° C.
But the most important question, whether the poison thus
obtained does or does not produce antitoxic immunity against
the streptococcus, has not yet been definitely settled.

ROGER (loc. cit.) found that his poison not only failed to pro-
duce any immunity, but that it even increased the susceptibility
to infection.

SiEBER-ScHOUMOWA1 was unable to effect immunisation by
means of filtered cultures.

On the other hand, LAITINEN (loc. cit.) cultivated streptococci
on a 5 per cent, bouillon of peptone containing 2 per cent, of
glycerin and 0'3 per cent, of ordinary salt, and having an alka-
linity of 0'2 to 2 per cent., and precipitated from these cultures
by means of ammonium sulphate or amyl alcohol toxine prepara-
tions, which, when injected into the peritoneum in the proportion
of 0*1 to 0*4: c.c., or even of O'Ol c.c. on direct injection into the
nerves, killed large rabbits, while they also had the power of
conferring a certain degree of immunity against infection.

PARASCANDALO (loc. cit.) laid special stress on the point that he
was best able to produce immunity against streptococci by means
of soluble toxines.

He cultivated them on sugar bouillon, and, after sterilising them by
means of 0'5 per cent, of phenol, filtered the cultivations after twenty-four
hours through paper. He had found that simple filtration through porce-
lain did not remove all the living micro-organisms, which was also the case
on heating them to 60° or 70° C., although a higher temperature injured
the toxine. By making the first inoculations with the filtrates of young and
only slightly toxic cultures, he was able to immunise animals gradually
against large doses of very poisonous cultures, and in this way obtained a
true therapeutic serum.

1 Sieber-Schoumowa, "Les serums therapeutiques anticocciques," Arch,
des Sciences Biol., iv., 415, 1896.

166 TOXINES AND ANTITOXINES.

Yet even his results are open to question, inasmuch as it has
been found by SCHKNK (loc. cit.) that O5 per cent, of phenol is
also insufficient to kill all the micro-organisms. He found that,
although the cultures were apparently sterile, there were yet
some living micro-organisms in the blood of the heart of the
poisoned animals, so that it is quite possible that PARASCANDALO
produced immunity by means of isolated, weakened, but yet
living streptococci. SiMON,1 too, was only able to detect the
production of toxine after the decomposition of the cells. Thus
proof has yet to be brought of the existence of streptotoxine, and
the question is as open as that of gonotoxine.

THE POISON OF THE TUBERCLE BACILLUS.

The poisonous substances produced by the bacillus of tubercu-
losis require separate treatment for the following reasons : — In
the case of other poison-producing bacteria the true toxines, or
rather those specific poisons that it is possible to describe with
more or less certainty as of the nature of toxines, can be
separately described and differentiated from the so-called bac-
terial proteins, in the wider sense, that remain behind in the cells
of the bacilli themselves after the extraction of these poisons
which the micro-organisms have produced. This sharp differen-
tiation is not possible on historical grounds in the case of tuber-
culosis, for since the time when the poisons of the tubercle
bacillus were first seriously studied, it has been the almost
invariable practice to investigate both kinds of poisons without
separating them.

The first preparations of tubercle poison, and notably R. Koch's
tuberculine, were thus a mixture of all the specific and non-specific
poisonous products of this bacillus, and no investigation was made
to determine the question of the existence of a true toxine. At
that time, too, it was not possible to formulate the question in
such precise terms, since the definition of a true toxine, and of
the true antitoxic immunity that it involved, had not then been
given with sufficient clearness.

These toxic preparations, therefore, consisted of the more or
less altered substances of the cell itself. In addition to these,
however, they also undoubtedly contained other toxic products,
some of which possibly we must regard as primary secretion
products, and eventually as true toxines, and others as secondary
products of a simpler nature, with which we shall also deal later.

1 Simon, "Uber die Gifte der Streptococcus," CentralU.f. BaTct., xxxv.,
318, 1904.

THE POISON OF THE TUBERCLE BACILLUS. 167

Here we must first determine whether there is a true tuber-
culosis toxine.

Until quite recently the only preparations actually tested with
regard to this point were those of MARAGLiANO.1 By means of
filtration through bacterial filters he obtained from fresh cultures
grown at the ordinary temperature a poisonous bouillon, which
in large doses caused death, preceded by hypothermia and sweat-
ing. The poison was destroyed at 100° C. Similar insignificant
results were obtained by BERNHEiM2 with cultures filtered
through a Kitasato filter.

MARAGLIANO and his pupils also prepared therapeutic sera and
antitoxines against tuberculosis, but he himself concluded that
the antitoxine did not have a direct neutralising effect upon the
"toxine," but only acted as a stimulus to other therapeutic
forces ; hence it loses all interest for us.

These experiments appeared of great importance, and their
results were very thoroughly tested as regards their thera-
peutical application, though without any material success. But
this did not do much to advance the knowledge of the toxine
question itself.

LEDOUX and LEBARD 3 also found in cultivations that had been filtered
through a Chamberland filter substances which had, it is true, a pyrogenic
and toxic effect on intraperitoneal injection, but their action was so slight
that they were inclined to attribute it exclusively to the culture medium
itself.

More recently the results of an investigation into tuberculosis
toxine have been published by FRENKEL and BRONSTKIN.* They
cultivated tubercle bacilli on a glycerin bouillon of 5 per cent-
strength, and thus obtained solutions of poison which, after
being freed from the bacilli by filtration, killed a guinea-pig in
a few days when injected in a dose of 1-5 to 2 c.c.

The poison was rapidly weakened by light and air. They
prepared it in a concentrated form by means of precipitation
with alcohol.

They claim to have produced immunity by means of inocula-

1 Maragliano, "Heilungd. Lungentuberkulose durch Heilserumtherapie,"
BerL klin. Woch., 1895, 689; id., " Ueber das tuberkulose Heilserum u.
seine Antitoxine," ibid., 1896, 773; id., " Ueb. d. Tuberkelantitoxin,"
Mali/s Jb., 1900, 1044.

2 Bernheim, "Immun. tuberc. et serumtherapie," Soc. BioL, xlviii., 291,
1896.

3 Ledoux and Lebard, " De Faction sur la tempe'r. du bouillon des
cultures tuberc.," Arch. d. Med. JExptr., x., 601, 1898.

4 Frenkel and Bronstein, " Ueber Tuberkulosetoxin und Antitoxine,"
BerL klin. Woch., 1901, 861.

168 TOXTNES AND ANTITOXINES.

tion with this poison, not only against all the tubercle poisons,
but also against the protein.

The chief point that makes one sceptical of the results of these
investigations is that the nitrates of the cultures had so slight
a toxic action. The authors had practically nothing more than
tuberculine in their hands, and this is fairly poisonous when
employed in the same manner. None of the investigators deter-
mined whether he had obtained a specific toxine which produced
in the organism an antitoxine acting only against the toxine
itself. An antitoxic immunity against the proteins is a priori
hardly conceivable, but, in any case, could not have been pro-
duced by the supposed true toxine, and it is just this omnipotence
of the "anti-bodies" of these therapeutic sera that lays these
statements open to the gravest suspicion. It is quite certain
that the existence of a true haptophore toxine of the tubercle
bacillus has not been proved by these experiments.

On the contrary, it appears more and more probable that the
tubercle bacillus does not produce any true toxine in the sense
of our definition. Hence, notwithstanding its intrinsically very
great importance, we can only deal briefly here with the subject
of the tubercle poisons.

TUBERCULINE.

Under this name are included preparations of different kinds,
having in common the characteristic that they are obtained from
cultures of tubercle bacilli, and contain their bodies in a nearly
unchanged condition.

. These substances are probably, for the most part, not true
toxines. They are, as is shown by their mode of preparation,
proteids resisting heat, closely allied to the albumoses, and
possibly possessing a certain amount of specific activity. Their
chemical nature and relationship to the albumoses has been
thoroughly studied by KiJHNE.1 They have but little toxic
action upon a healthy organism, but they produce considerable
reactions in tuberculous subjects.

The original tuberculine of ROBERT KOCH 2 was prepared from
a culture of tubercle bacilli grown at 38° C. on the surface of
a 4 per cent, glycerin bouillon to which as much air as possible
was admitted. After six to eight weeks the cultivations were
evaporated to a tenth of their volume, and filtered through a

1 Kiihne, " Erfahrungen iiber Albumosen u. Peptone," Z. f. BioL, xxix.,
24, 1892; xxx., 220, 1894.

2 R. Koch, "Mitt, iiber ein Heilmittel gegen Tuberk.," Deutsch. med.
Woch., 1891, 101, 1189.

TUBERCULINE. 169

porcelain filter. The extract thus obtained, containing 40 per
cent, of glycerin, was very stable.

After many fruitless attempts KOCH abandoned the idea of further
purification. His method has been frequently modified. NOCARD 1 first
heated the cultures to 110° C. before treating them in the same way ; he
contented himself with filtration through paper, since his preparations
were obviously already sterile. MABAGLiANO,'2 and, subsequently, FRENKEL
and BRONSTEIN (loc. cit.), used a method of boiling, eventually in auto-
claves, in the preparation of products very similar to tuberculine.

BE BRING3 extracted from tuberculine a substance akin to
mucin by means of soda solution, and fatty substances by means
of ether, and in this way claimed to have obtained a preparation
about twenty times as poisonous as before.

NiEMANN,4 too, has endeavoured, though without any particular success,
to obtain strong tuberculine preparations by precipitation with alcohol.

As regards the toxic value of these preparations the results
obtained by these investigators have been supplemented by
GRAMATSCHiKOFF,5 who proved that tuberculine was a blood
poison, and by v. LINGELSHEIM,G who proposed to determine a
normal unit of toxic activity by means of intercerebral injections.

CARRiERE7 found that fairly considerable alterations in the
liver and kidneys followed the injection of tuberculine.

PEIPER 8 found that in non-tuberculous subjects it had a con-
siderable pyrogenic effect, but was only slightly toxic. Even
doses of O'Ol grm. produced only insignificant symptoms, such as
pains in the head and limbs, &c.

All these experiments were put upon a new basis by the fact
that the attempts to extract the poisonous substances in an un-
altered form from the tubercle bacilli were largely abandoned
in favour of experiments with the bacilli themselves. Kocn9

1 Nocard, "Des injections revelatrices de la tuberculine," Rec. de Med.
Vet., Ixxii., 369; Baumg. Jb., 1895, 705.

2 Maragliano, "Extrait aqueux des bacilles de la tuberc.," Soc. Biol.,
l.,94, 1898.

3 Behring, "BekampfungderTuberkulose," Munch.med. Woch., 1898, 580.

4 Niemann, "Ueber Tuberkuloseheilserum," Munch, med. Woch., 1897,59.

5 Gramatschikoff, " Ueber einige physiolog. Wirk. des kochschen Tuber-
kulins," Arb. pathol. List. Tubingen, i., 287; Baumg. Jahrb., 1897, 548.

6 v. Lingelsheim, "Ueber die Wertbestimmung der Tuberk. Giftpra-
parate," Deutsch. med. Woch., 1898, 583.

7 Carriere, "Etude exper. des alterations histolog. du foie, &c.," Arch.
Med. Exper., ix., 65, 1891.

8 Peiper, "Ueber die Wirkung des Kochschen Mittels auf Gesunde,"
Deutch. med. Woch., 1891, 160.

9 R. Koch, "Ueb. neue Tuberkulinpraparate," Deutsch. med. Woch.,
1897, 209.

170 TOXINES AND ANTITOXINES.

started from the observation that dead tubercle bacilli remained
undecomposed in the body for a very long period, and were
therefore not absorbed. But, according to him, so long as they
retain their form, immunity against the bacilli cannot be pro-
duced ; and it is only after their decomposition, but too late
for any curative action, that a slight bactericidal immunity is
attained. KOCH, therefore, endeavoured to get nearer to his
goal by breaking up the bacilli and using an emulsion of their
cell constituents for the inoculation. For this purpose he first
dried the bacilli rapidly, and then triturated them mechanically.
The masses were treated with distilled water, and repeatedly
subjected to centrifugal force. The first portion separated was
soluble in water, but was not only non-poisonous, but also devoid
of any immunising power (T.O.). It was only the second and
third " centrifugates " that contained the substance with specific
action. These preparations were termed "T.R." by KOCH.
They were found to possess immunising and therapeutic powers
without any particular toxic action, and notably without abscess
formation or inflammation. It was found best to use young
cultivations dried in vacuo, and to protect the preparations from
light as far as possible.

H. BuCHNER,1 working alone and also in collaboration with
HAHN,2 isolated from the bacilli by means of E. Buchner's
method of applying enormous pressure, expressed fluids "plas-
mines" with which he believed that he obtained specific effects,
although he himself regarded the question as still unsettled.

BEHRING 3 described a poison obtained by means of glycerin
water at 150° C. from the finely-divided bacteria after extraction
of the fat, and stated that 1 grm. of this was sufficient to kill
1,250 mice. It was neutralised by the serum of a cow cured of
tuberculosis (antitoxine formation ]).

LANDMANN 4 claimed to have discovered an active tubercle poison in his
"tuberhdol" and asserted that it destroyed a guinea-pig of 250 grms. when
injected in a dose of O'l grm.

He macerated tubercle bacilli for a long time at 40° C. with a physio-
logical solution of salt, distilled water and glycerin ; then decanted the

1H. Buchner, " Zur Koch's Mitt, iiber neue Tuberkulinpniparate,"
Berl. Uin. Woch., 1897, 322; id., "Die Bedeutung d. aktiven losl.
Zellprodukte, &c.," Munch, med. Woch., 1897, 12.

2Hahn, "Immun. u. Heilungvers. mit den plasmat. Zellsaften," Munch,
med. Woch., 1897, 1344.

3Behring, "Autoreferat iiber den Vortrag in Madrid," Deutsch. med.
Woch., 1898, 293.

,4 Landmann, " Ueb. eine neue Meth. der Tuberkulosetoxin-Behandlung,"
Hyg. Rundschau, x., 168; Centralbl.f. Bakt., xxvii., 280, 1900.

TUBERCULINE. 171

liquid, and treated the residual bacilli in the same way successively at 50°,
60°, &c., up to 100° C. The different extracts were then united and mixed
with culture fluid, and the whole evaporated at 37° C. in vacuo and filtered
through porcelain. In this way he obtained a mixture of all the products
of the tubercle bacilli, including both the toxines destroyed by heat as
well as the hot-water extracts, with the proteins. He claimed to have
produced therapeutic results by means of this mixture. It is obvious that
such mixtures as this cannot advance our theoretical knowledge.

Subsequently KocH,1 in a later research, published a new
method of making tuberculine preparations.

He triturated O'l grm. of the absolutely dry tubercle bacilli in an agate
mortar with a solution of 0*5 grm. of phenol and 0*85 grm. of sodium
chloride in 100 c.c. of water, a few drops being used at first, and then
more and more of the liquid until the total amounted to 100 c.c. The
mixture was then whirled for six minutes in a centrifugal machine,
decanted from the deposit, and diluted with ten times its volume of the
solution, so that eventually the liquid contained in a litre the extract from
0*1 grm. of tubercle bacilli.

This liquid was rendered turbid by the addition of agglutinating serum.
By itself it also served the purpose of diagnosis.

The question of the nature of the poisons of the tubercle
bacillus was thoroughly studied by RUPPEL.2 He found, first
of all, that the nitrates were absolutely non-specific, and, with
the exception of albumoses, chiefly deuteroalbumose, was unable
to discover anything toxic in them. Nor was he more successful
in his attempts to isolate any specific poison from the bacilli by
extraction. This really gives a definite answer in the negative
to the question whether there exists a soluble specific toxine of
the nature of diphtheria toxine.

On the other hand, he isolated from the crushed bacilli two
poisonous substances — viz., a nucleic acid containing 9*42 per
cent, of phosphorus, tuberculinic acid, and also a protamine,
which he recognised as such by its being precipitated with
picric acid, and termed tuber cidosamine. The proportion of the
finely-divided bacilli dissolved after treatment in a centrifugal
machine amounted to 50 per cent. The solution gave no proteid
reactions and probably contained only a compound of the nucleic
acid with the protamine. According to NEUFELD,S however,
the action of tuberculosamine is absolutely non-specific and v.
Lingelsheim's method of valuation (vide supra) cannot be used.

1Koch, "Ueb. d. Agglut. d. Tuberkelbacillen, " Deutsch. med. Woch.,
1901, 829.

2Ruppel, "Zur Chemie der Tuberkelbazillen," Z. f. phys. Chem., xxvi.,
218, 1898.

3Neufeld, "Zur Werthbestimmung der Tuberkulosepraparate," Deutsch.
med. Woch., 1899, 13.

172 TOXINES AND ANTITOXINES.

RUPPEL and KiTisniMA1 made a further investigation of
tuberculinic acid. It was from three and a half to four times as
poisonous as dry old tuberculine. It was much more poisonous
on intercerebral injection, especially in the case of tuberculous
guinea-pigs (1 grm. killed 40,000 kilos.). Other nucleic acids
were less poisonous. Tuberculinic acid prepared by the older
somewhat drastic method of KOSSEL was found to be five times
less poisonous than the new preparation. RUPPEL and KITI-
SHIMA then prepared from tuberculinic acid by Kossel's method,
a substance similar to thymic acid, tuberculothymic acid, " 1 c.c.
of which contained as much poison specific for tuberculous
individuals as 20 c.c. of Koch's tuberculine" (BEHRING), and
also a still more simple poisonous substance of as yet unknown
nature, tuberculosine, 1 grm. of which was equivalent to 25 to
30 c.c. of Koch's tuberculine.

According to BEHRING this represents the poison-nucleus,
around which the other substances are grouped in some way or
other, and without which there can exist no poison with the
specific action of tuberculine.

Important as these researches are with respect to the toxic
action of the tubercle bacilli, they have yet little claim upon our
attention here, since there can be no question of these substances
obtained by drastic chemical methods being true toxinvs.

A further objection to drawing comparisons between these
poisons and diphtheria poison is that the former are relatively
only slightly poisonous at all events to the healthy organism.

There is never any formation ofantitoxine in the case of these
poisons.

Tuberculine is invariably but slightly poisonous to a healthy
organism, and its toxic action can be still further reduced
by processes of purification without injuring its specific
activity.

On the other hand, it has frequently been shown by RoMER,2
BucHNER,3 and others that the action of tuberculine, as regards
its toxic effects, is non-specific. The same pyrogenic and local
inflammatory symptoms could also be produced, in the same
degree, by the proteins of other micro-organisms (pyocyaneus,

^Behring, Die Diphtherie, Berlin, 1901, 91 ; see also Behring, " Ueber
die spez. giftigen Eigensch. d. Tuberkulinsaure," Berl. Uin. Woch., 1899,
537.

2 Romer, " Ueb. d. formativen Reiz d. Proteine Buchners," Berl. klin.
Woch., 1891,886; " Tuberkulinreaktion d. Bazillenextrakte," Wien. klin.
Woch., 1891, 835.

3 Buchner, " Tuberkulinreakt. d. Proteine nicht spezif. Bakt.," Munch,
med. Woch., 1891, 841.

TUBERCULINE. 173

prodigiosusy pneumobacillus) that had been prepared in an
analogous manner.

It is thus hardly conceivable that the old tuberculine prepara-
tions should possess any specific toxic activity.

Moreover, their specific protective activity is scarcely to be
attributed to an antitoxic process. There appears, rather, to
be a liberation of specific bactericidal protective forces, by means
of these specific proteids, whether it be that by this treatment
similar substances are set free and can be utilised, just as specific
bacteriolytic processes are also brought about in the case of
cholera, &c., by such substances — viz., the receptors of the bac-
teria, which concentrate their activity upon the intermediate
products as defined by EHRLICH; or whether we have here to
deal with specific proteids concentrating their activity upon the
precipitines or agylutinines, which act upon the bacilli and are
able to bring about specific reactions of a protective or curative
nature in the diseased organism. The fact that specific agglu-
tinines were present in the serum of tuberculous subjects was
made known by ARLOING and COURMONT, and more recently
confirmed by KOCH in another way (vide supra). It would
thus seem conceivable that the substances extracted from the
bacilli were in this case able to bring about bacteriolytic
processes similar to those produced by the whole bacteria in
the case of cholera, typhus, rinderpest, &c. There has, as yet,
been no sufficient reason for concluding that in these immunising
processes a separate position is occupied by inoculation with
unbroken bacteria, since these delicate organisms are readily
absorbed, and when decomposed within the body bring about
the curative immunising processes. And the only attempts to
produce immunity with "plasmines," in the case of cholera
(HARN, loc. cit.), also support the view that in those cases
bactericidal, not antitoxic, phenomena play the chief part. On
the other hand, the tough tubercle bacilli protected by a thick
membrane of fat-like substance (cf. RUPPEL, loc. cit.) are, as
KOCH has shown, attacked either not at all or only with great
difficulty in the organism, and are thus incapable of bringing
about any bactericidal immunising processes. It has, however,
been shown by KOCH that slight phenomena of bactericidal
immunity appear during their ultimate decomposition. It
would appear, then, that by drastic means, especially mechanical
trituration of the bacilli and injection of the extracts obtained
from them, substances are liberated which lead more rapidly to
the production of antibacterial immunity. Still more effective
is the crushing of the bacilli and expression of their cell content,

174 TOXINES AND ANTITOXINES.

which causes a still more plentiful liberation of those specific
receptors, which can then develop their specific reactions in the
organism. These intermediate bodies, however, withstand heat
and can thus resist high temperatures at which a true toxine, which
produces antitoxic immunity, would certainly be destroyed.

On this assumption, the specific tuberculine reaction in tuber-
culous men, as a reaction of a bactericidal nature, would have no
connection with the toxic pyrogenic effects in healthy men ;
these poisonous principles would be only interfering impurities,
and indeed we have seen that, as a matter of fact, KOCH and
others have made successful attempts to prepare an almost non-
toxic and yet specifically active tuberculine. Moreover, the latest
researches of KOCH (loc. cit.\ from which it appears that the
serum of tuberculous subjects gives a precipitate with his new
tuberculine preparation, appear to support this assumption of
specifically active proteids and their precipitines. This hypo-
thesis, that the action of tuberculine in its most recent form is
due to the liberation of specific amboceptors, materially elucidates
and simplifies the problems. Tuberculine occupies a different
position, according to BUCHNER, from the series of specific bac-
terial poisons, and yet has claims to be regarded as a specific
therapeutic and immunising agent, just as much as those dead
cells that produce antibacterial immunity in cholera, &c.

Nevertheless, the question of tubercle poisons needs further
study. This hypothesis of the action of tuberculine does not
by any means do away with the necessity for such research ; for,
as in the case of other bacteria, there may in addition to this
be a true toxine which can also bring about specific antitoxic
immunity.

That such a toxine has not yet been discovered may possibly
be due to the methods; but, on the other hand, it is not im-
probable, as was suggested above, that tubercle bacilli do not
produce any true toxine at all. Possibly they form poisons of
another kind, which, although not of the nature of toxines, have
yet a specific character.

The results of BEHRING and his pupils in the preparation of
very toxic tubercle poisons also point in the same direction,
and their discovery of tuberculosamine and its combination
with nucleic acid must also be regarded as valuable evidence
in support of this theory, although these were undoubtedly
not the specific poisons of the tubercle bacillus. For RUPPEL
himself concludes that in these compounds we have to deal
with derivatives of the cell nucleus. These are undoubtedly
not the specific bacterial poisons, which we must certainly

MALLEINE. 175

regard as protoplasmatic or paraplasmatic products. It is not
improbable that perfectly analogous substances might be isolated
from cell nuclei or cells of other bacteria.

On the other hand, we cannot recognise as the specific poison the pro-
ducts obtained by AucLAiR,1 who claims to have isolated, by extraction
of the cultures with ether, a poison which produces caseous degeneration
of the lungs on intratracheal injection ; and, by extraction with chloro-
form, a second poison which, he asserts, produces fibrous pneumonia.

We have therefore, unfortunately, to own that the question
of a specific tubercle poison has not yet been decided, even if
we exclude the hypothetical action of tuberculine, as described
above, entirely from the discussion.

MALLEINE.

Malleine contains the cell constituents of the bacilli of glanders
in much the same way that the old tuberculine contained those
of the tubercle bacilli. The subject of malleine has been in-
vestigated almost exclusively by veterinary surgeons from the
practical point of view, so that scientifically little or nothing is
known about its nature. In the absence of further knowledge
we shall do well to apply to malleine the considerations that we
have taken into account in the case of the old tuberculine.

HELMAN,2 in 1890, was the first to prepare an extract of
glanders bacilli, and he was followed by KALNING, who boiled
pure cultivations repeatedly with water and filtered them
through Chamberland filters. PREUSSE, and subsequently
PREISZ, employed glycerin extract of potato cultures.

JOHNE and PEARSON were the first to use bouillon as the
culture medium; they cultivated the bacilli for fourteen days
at 37° C. and sterilised the filtrates, or first concentrated the
cultures at 80° C. and then filtered them (PEARSON).

Roux sterilised very virulent cultivations at 110° C. and
filtered them after evaporation.

FOTH (loc. cit.) proceeded in a similar manner. He passed
the glanders bacilli several times through animals to render
them as virulent as possible, and then cultivated them for
twenty days as a surface cultivation on Loffler's meat broth of
45 per cent, strength. The cultures were evaporated to a tenth

1 Auclair, "La Scl^rose pulmonaire, &c." Arch, de M6d. Exper., 1900,
189 ; see also Baumgartens Jb., 1898, 475, 476.

2 I have taken all these references, which were not readily accessible to
me in the original, from Foth's paper, "Das Mallein, &c.," Fortschr. d.
Med., 1895, 637.

176 TOXINES AND ANTITOXINES.

of their volume at 80° C., filtered, and treated with thirty times
their quantity of alcohol. The white precipitate thus obtained
was left to dry over calcium chloride in vacuo, and furnished a
non-hygroscopic white mass.

GUINARD l made a closer study of the action of malleine. He
found that the action of the heart was first stimulated and then
lowered, and that a similar stimulation and subsequent depres-
sion of the nervous organs also occurred. Sweating was also
observed.

ScHATTENFROH,2 on the other hand, regards malleine as abso-
lutely non-specific in its action.

The question whether or no malleine has a specific toxic action
is as little decided as in the case of tuberculine. It is possible
that here, too, we have non-specific symptoms similar to those
invariably produced by proteins foreign to the body.

ANTHRAX POISON.

We are quite ignorant whether or no there exists a true
anthrax toxine. As in the generality of cases, the answer to
this question has been rendered extraordinarily difficult by the
fact that many investigators made their experiments with the
sole object of producing immunity against the anthrax bacillus,
by any possible means, and that consequently they made use of
living bacilli or the dead cells without troubling themselves
about the possible existence of a specific antitoxine-producing
poison. At the earliest period it was hardly to be expected
that the problem could be stated in such precise terms, but
even many later investigators have given almost exclusive
attention to the practical question of immunisation. They
have endeavoured, as, for instance, has been done by SOBERN-
HEIM in his researches into anthrax immunisation, to obtain
in every possible way the desired immunity by inoculation,
without investigating whether this protection was due to a
true antitoxic immunity. We meet with this difficulty in all
other investigations into less known poisons ; yet nowhere has
the toxine problem remained in such obscurity or the results
been so contradictory as here. I must, therefore, content my-
self with mentioning the most important researches that have
dealt more or less expressly with the hypothetical toxine, and

"Effets physiolog. du Malleine," Journ. Med. Vet., xlvi.,
454; Baumgartens Jb., 1895, 311.

2 Schattenfroh, "Ueber die Wirkung von Bakterien-Proteinen," Zeit.f.
Hyg.t xviii., 456, 1894.

ANTHRAX POISON. 177

for this purpose the unusually careful bibliography compiled by
CONRADI * has naturally been of the greatest service to me.

The first attempts to isolate poisons from anthrax bacilli by
filtration were made by PASTEUR, 2 but with negative results.
The filtrates were absolutely inert, while those obtained by W.
KOCH 3 produced only elevation of temperature and dyspnoea.

The first published account about a substance obtained from anthrax
cultures, and stated to have an immunising action, was that of WOOL-
BRIDGE.4 He cultivated anthrax bacilli for two to three days at 37° C.
upon a bouillon of thymus or scrotum extract. The liquid thus obtained
had an immunising non-toxic effect after filtration. He also made very
similar experiments 5 with a solution of his tissue fibrinogen prepared from
scrota, &c., and sterilised by means of heat, care being taken that it had
only a slightly alkaline reaction ; for vigorous growths of the micro-
organisms in strongly alkaline media yielded no poison or immunising
substance to the solution. But the filtrates of cultures grown on weakly
alkaline solutions of tissue fibrinogen undoubtedly contained an immunising
non-toxic substance, which only afforded protection on intravenous injec-
tion. Yet he asserts that he obtained almost analogous results by simple
injection of his solution into animals without previous cultivation of anthrax
bacilli, and claims that the tissue fibrinogen by itself affords protection
against anthrax. This completely destroys the value of his work as regards
the discovery of a possible toxine.

Nor does any greater importance attach to the theoretical proof of
CHAUVEAU 6 that anthrax poison must exist in the soluble form, since the
placenta, being impermeable to bacteria, allows none of the bacilli to pass,
and yet the embryo of a sheep infected with anthrax is immune ; rather
its importance is less, since in rare instances anthrax bacilli can pass
through the placenta.

HANKIN attempted in a series of investigations7 to prove
that his anthrax albumoses were the chief toxic principle. He
cultivated the bacilli on a bouillon containing 0-1 per cent, of
meat extract, and a large proportion of fibrin (10 to 50 per cent.).

1 Conradi, "Z. Frage d. Toxinbild. bei d. Milzbrandbakt.,"^. /. Hyg.,
xxxvii., 287, 1899.

2 Pasteur and Joubert, ' ' Etude sur la maladie charbonneuse, " Comptes
Rend., Ixxxiv., 900, 1877.

3 W. Koch, "Milzbrand und Kauschbrand," Stuttgart, 1886, quoted by
Conradi (loc. cit.).

4 Wooldridge, "Note on the Protection in Anthrax," Proc. Roy. Soc.,
xlii., 312, 1887.

5 Wooldridge, " Vers. ii. Schutzimpf. auf chemischem Wege," Du Bois
Arch., 1888, 527.

6 Chauveau, " Sur le mecanisme de 1'immun.," Ann. Past., iii., 66, 1888.

7 Hankin, "On Immunity produced by an Albumose isolated from
Anthrax Cultures," Brit. Med. J., 1S89, ii., 65; id., "On the Conflict
between the Organism and the Microbe," ibid., 1890, ii., 810; Hankin and
Wesbrook, "Albumoses et toxalbumines du bacille charbonneux," Ann.
Past.,vi., 633, 1892,

12

178 TOXINES AND ANTITOXINES.

The sterilised liquid was inoculated and allowed to stand for
eight days, after which it was filtered, and the filtrate treated
with ammonium sulphate. The precipitate was dialysed at
.42° C., and the liquid evaporated in vacuo, or treated with
alcohol. This yielded an albumose, which weakened bacteria did
not give. Yery small doses (one part of a 1 :1 million solution
per kilo, of body weight in the case of mice) immediately pro-
duced immunity against anthrax, but this was not the case with
large doses, which had only a toxic, but not immunising action.

Similar toxic albumoses isolated almost at the same time from anthrax
cultures and the organs of animals by MARTIN x and BRIEGER and FRANKEL,'2
caused symptoms somewhat resembling those produced by the living bacilli.
Yet the definite results of all these investigations of toxalbumins 3 were very
insignificant, as was eventually the case with all the toxalbumins. Martin's
albumose, for example, had to be given in doses of 0'3 grm. per mouse (!)
to produce death.

According to the views now accepted, there can certainly be no
toxine activity here. The anthrax bacilli have naturally decom-
posed the proteids of the culture medium, and either traces of
poison have been retained by the resulting albumoses, or the
proteids foreign to the body that have been produced have in
themselves a slight toxic action, as is also the case with other
proteids.

In any case, PETERMANN,* who repeated HANKIN'S work, was
unable to detect any trace of toxic or immunising activity in his
albumose, though HANKIN and WESBROOK (loc. cit.), observing
under more accurately specified conditions, and, in particular,
avoiding temperatures above 20° C., still maintained their posi-
tion, and also again obtained frequently, though not invariably,
transient immunisation.

An important point in PETERMANN'S investigation is the proof
that anthrax cultures when simply filtered have a slight and
transient, but unmistakable, immunising effect ; this would seem
to point to the presence of a real toxine, which, however, can
only be present in very slight quantity.

1 Martin, "The Chemical Products of the Growth of the Bacillus
Anthracis," Proc. Roy. Soc., xxii. [5], 1890.

2 Brieger and Frankel, " Untersuch. iiber Bakteriengifte," Bert. Tdin.
Woch., 1890 [11-12].

3 Similar investigations with equally scanty results were also made about
the same time by, e.g., Balp and Carbone (1891), Landi (1891), G. Klem-
perer (1891), Maltzew (1891) ; for particulars about these the reader must
be referred to Conradi (loc. cit. ).

4 Petermann, " Recherches sur Fimmun. contre le charbon au moyen des
albumoses extraites des cultures," Ann. Past., vi., 32, 1892,

ANTHRAX POISON. 179

Quite in agreement with this are the results obtained by
ARLOING.1 He merely allowed bouillon cultures of anthrax to
stand, and then decanted the liquid at frequent intervals from
the deposited bacteria, thus obtaining a germ-free fluid, in which
obviously there was no possibility of the very considerable loss
of poison retained by the filters usually employed. On the other
hand, the toxine might very easily undergo decomposition during
the long standing. The fluid thus obtained had an immunising
effect when given in large doses.

There are also the results obtained by SCLAVO 2 and MARcnoux,3 who
produced immunity by means of the living bacilli ; but these have ap-
parently nothing to do with the question of a toxine. At all events,
MARCHOUX asserts that his serum acts by means of phagocytosis.

The next serious attempts to discover the toxine were those
made by MARMiEK,4 who, from cultivations on peptone-glycerin
(1 litre of water, 40 grms. of peptone, 15 grms. NaCl, 40 grms.
of glycerin, 0-5 grm. of sodium bisulphate, and 0-2 grm. of
potassium bisulphate), preferably grown at 20° C., obtained a
substance, which, when separated by precipitation with am-
monium sulphate and dialysis, or by extraction with glycerin
and precipitation with alcohol, formed a pulverulent brown mass
giving neither proteid nor alkaloidal reactions, and showing no
enzymic activity.

It was toxic to the extent of 0*08 grm. per kilo, of animal
(though not constantly so), and caused elevation of temperature,
diarrhoea, emaciation, convulsions, slowed respiration, and suffo-
cation. A dose of 0*2 grm. per kilo, was invariably fatal.
According to MARMIER the symptoms here were sufficiently
similar to those caused by infection with living bacteria as to
justify the conclusion that they were due to a specific anthrax
toxine.

There are two reasons, however, against the view that there
was a true toxine here, apart from the fact that the lethal dose
was surely a little too high for a true toxine.

In the first place, MARMIER stated that the poison could be
boiled for an hour without alteration, and could even resist being-
heated for five minutes at 120° C., and hence it could not have

1 Arloing, " Sur la presence de la substance phylacogene dans les liquides
(lu bac. anthr.," Bull MecL, 1892, 1038, quoted from Centralbl. fur RaU.,
xiii., 561, 1892.

2 Sclavo, "Ueber die Bereitung des Serums gegen den Milzbrand,"
Centralbl. f. Bakt., xviii., 744, 1895.

3 Marchoux, "S4rum anticharbonneux," Ann. Past., ix., 9, 1895.

4 Marmier, "Sur la toxine charbonneuse," Ann. Paxt., ix., 533, 1895.

180 TOXINES AND ANTTTOXINES.

been a true toxine. The poison was weakened by gold chloride,
platinum chloride, and calcium chloride.

And, in the second place, it apparently produced only a slight
immunity, and that, too, not invariably. Hence, in any case, it
could not have been a weakened toxine, since then the reduction
in toxic power must have been accompanied by the more ener-
getic immunising effect.

CONRADI (loc. cit.) has recently once more investigated the
question whether anthrax bacilli form a soluble diffusible poison,
and has, if we may anticipate, arrived at an absolutely negative
result. He started from the point of view that it was the multi-
plicity of the nutrient substrata used by previous investigators
that was frequently responsible for the diametrically opposite
results. Hence he decided to search for the anthrax poison in
the body of the animal. For this purpose he used peritoneal
exudations and extracts of organs (liver and spleen) of infected
animals, and found both, after nitration through a Kitasato or
Chamberland filter, to be invariably devoid of any toxic effect.

Moreover, he introduced virulent bouillon cultures into the
abdominal cavity of guinea-pigs, which was then enclosed in
a bacteria-proof bag made with the inner membranes of the reed
(Phragmites communis), and in this case, too, there was no toxic
action.

Two objections can be offered to CONRADI'S results.

The reed experiments only prove that there does not exist any
soluble and diffusible anthrax poison, but do not exclude the
possibility of there being a non-diffusible toxine, just as other
haptines — e.g., certain ferments — are also indiffusible. And,
indeed, the results obtained by ARLOING (vide supra), who found
a poison in unfiltered cultures, whereas the majority of investi-
gators found nothing in filtered cultivations, render it not
improbable that anthrax toxine may be readily retained by
filters and membranes.

The second objection appears to me more important. Even
granting the existence of a true haptophore toxine of anthrax, it
is possible that under certain conditions it may not be found
again in the organs and fluids of the infected bodies of susceptible
animals, owing to its entering into stable combination with the
receptors ; the poisons of tetanus and diphtheria can, even under
normal conditions, only be found again in the blood and organs
of infected animals after the introduction of large doses.

It is true that living micro-organisms were present in these
exudations, but, at the same time, these could not have been able
to have used their toxine-producing powers for any length of

ANTHRAX POISON. 181

time ; and, moreover, the toxines produced in these exudations
would surely have undergone renewed metabolism, and have
been fixed by the receptors, so that only very slight traces could
have been left behind. In addition to this the exudations in
this case would doubtless contain antitoxines, which would render
a correct deduction still more difficult. At the same time, it is
possible that such exudations may be used as suitable natural
culture-media, in which, eventually, toxines may be detected.

Be this as it may, it is impossible to draw any conclusion from
these experiments by CONRADI as to the non-existence of a true
toxine. Speaking generally, such experiments are not suitable
for determining the question of the production of poison by
a bacterium ; for other poisons of simpler nature may also be so
altered or destroyed after producing injurious effects, or even by
their physiological action, that no trace that can be detected
remains in the fluids of the tissues. Even many alkaloids, such
as cytisine, disappear in the organism without leaving a trace.
It is not safe, in view of the manifold attacks to which a poison
is exposed in the body, to draw negative conclusions as to the
non-existence of this poison ; only positive results would be
conclusive here. Thus CON R ADI'S experiments only contribute
negative evidence towards a decision on this point ; once more
a means of discovering the hypothetical poison has proved un-
workable; the question of the existence of the poison still
remains open. And, in this connection, it must not be lost sight
of that the existence of an anthrax poison is really an etiological
postulate ; for there are many cases of anthrax with fatal results
where general bacterial infection is out of the question, and
where it has been very difficult or impossible to detect living
bacilli in the organism. Surely such cases can only be explained
as cases of poisoning. Hence, notwithstanding Conradi's results,
further search must be made for the poison.

On the other hand, no objection can be offered to the experi-
ments made by CONRADI to determine whether the anthrax
bacillus produced a specific endotoxine, which is only liberated
after the death and decomposition of the cell in an analogous
manner to yeast, invertase, and zymase. He killed the bacilli
by means of toluene, or by freezing, or disintegrated them by
Buchner's method. In every instance the result was absolutely
negative. Nor could he isolate toxalbumins from the organs,
either by means of the method of Brieger and Frankel or by
that of Marmier.

Hence, the present state of affairs is as follows : —

Convincing proof of the existence of a true anthrax toxine,

182 TOXINES AND ANTITOXINES.

either in a state of free secretion or as an endotoxine retained by
the cells, has not yet been adduced, and the existence of an
endotoxine is practically disproved. With the exception of
substances causing slight pyrogenic effects due to the bacterial
proteins of the cell itself and similar to those produced by all
bacteria, whether pathogenic or harmless, no anthrax poisons are
known. At the same time we must assume that poisons that
cause the symptoms of the disease are formed ; yet these appear
to be of such a subtle character and to be produced so sparingly
during the growth of the bacilli in the attacked organism as to
escape detection. COXRADI made no experiments with regard
to the immunising power of his sterile nitrates ; and it is
conceivable that they contained toxoids which had immunising
capacity but were no longer toxic, and that this is a possible
explanation of HANKIN'S results. These obvious possibilities
still remain unsettled.

OTHER SOLUBLE BACTERIAL POISONS POSSIBLY OF
THE NATURE OF TOXINES.

When we consider that the investigation of the specific poisons
of some of the most important pathogenic micro-organisms is for
the most part still in its infancy, it is not surprising that, in the
case of less important bacteria, we have very scanty results, to
which I will only briefly refer here. It is quite possible that on
closer investigation one or more true toxines may still be dis-
covered as a product of these bacteria, but as yet there are no
certain indications of this.

As regards the majority of those infectious bacteria that grow
within the organism, we can come to the same conclusion as in
the case of cholera, streptococci, <fcc. — viz., that if the cells secrete
any true toxines they do so only in minute quantities, that these
are extremely unstable, and that the poisons must be regarded
as consisting in the main of firmly retained endotoxines. Im-
munity, as in the case of all infectious micro-organisms, is
brought about almost exclusively by bacteriolytic processes,
any eventual formation of antitoxine being insignificant and
occupying an altogether inferior position, both from the practical
and theoretical point of view.

HOG CHOLERA.

METSCHNIKOFF l and also SELANDER2 found the sterilised

1 Metschnikoff, "Zur Immunitatslehre," Congr. /. inn. Med., 1892, 282;
"Etudes sur 1'immunite," Ann. Past., v. ; vi., 2S9, 1892.

2Selander, "La maladie infectieuse des pores.," Ann. Past., iv., 545,
1890.

METSCHNIKOFF'S VIBRIO. 183

blood of diseased animals to be very poisonous. It is extra-
ordinarily easy to produce immunity against the bacteria, but
this does not increase the power of resisting the sterile poison.
The micrococci grow in immunised serum, but with reduction of
their virulence, which is restored, however, by inoculating fresh
bouillon. Hence, the serum would seem to contain antitoxine,
which neutralises the production of toxine by the growing
bacteria.

MALIGNANT (EDEMA.

Roux and CHAMBERLAND l found that cultures of the septic
vibrio freed from the micro-organisms were slightly poisonous,
while sterilised peritoneal fluid was very poisonous and had an
immunising power. But they also obtained "poisons" which
still had a very weak toxic action and some immunising power
after exposure to a temperature of 110° to 120° C. BESSON,2
too, has prepared weak poisons (5 to 10 c.c. being required to
kill a guinea-pig) by filtering cultivations and in particular those
grown on flesh.

SWINE PLAGUE.

SiLBERSCHMiDT3 discovered a weak poison with a specific
action in filtered cultivations. It was not completely destroyed
at 120° C., but was gradually weakened even at 60° C. The
filtrates produced immunity.

SELBERG* found poisonous substances with a slight specific
action attached to the cells of the bacilli of swine plague.

METSCHNIKOFFS VIBRIO.

A very interesting report on this vibrio was published by
SANARELLi.5 It grows vigorously on immunised serum after
first apparently dying, but is then almost completely innocuous,
though it again becomes toxic when grown in fresh bouillon.
In like manner the vibriones from an immunised animal can be

1 Roux and Chamberland, "Immunit§ centre la septice'mie," Ann. Past.,
ii., 561, 1887.

2Besson, "Contrib. a 1' etude du vibrion septique," Ann. Past., ix.,
179, 1895.

3 Silberschmidt, "Contrib. a 1'etude de la swine plague," A nn. Past., ix.,
65, 1895.

4 Selberg, "Beitr. z. Kenntn. d. Giftwirkung d. Schweineseuchen-
bakterien, &c.," Dies., Berlin, 1896.

5 Sanarelli, "Defense de 1'organisme contre les microbes," Ann. Past.,
vi., 225, 1892,

184 TOXINES AND ANTITOXINES.

cultivated. If the serum of an infected animal be separated
from the vibriones that it contains the latter are far more
virulent than when the serum is also present.

From these results it appears that the vibrio is capable of
increasing in the body of the animal, and that it produces
toxines there which are injurious. But in an immune animal
or in immunised serum, and also in the serum of the diseased
animal, there are antitoxines which combine with the poison
and render the vibrio harmless, although it continues to grow
in the immune animal. If the antitoxine be removed the poison
again becomes active, as is also the case when the vibrio is culti-
vated upon new culture media.

Unfortunately, SANARELLI did not attempt to isolate, by means
of filtration, the toxine that was probably present. He only
found cultures killed at 60° C. and 120° C. to be poisonous, and
naturally was unable to produce any antitoxic immunity against
the secondary poison which he obtained at 120° C. The latter
is certainly not a toxine, although it would seem that a true
toxine is produced by these bacteria and may possibly be demon-
strated by careful work.

DYSENTERY TOXINE.

A toxine supposed to be that of dysentery was found by
RosENTHAL1 in the nitrates of cultures three weeks old. The
nitrates, in doses of 0-1 to 0-2 c.c., had a fatal effect upon rabbits.
The poison is stated to be fairly resistant to heat and to form an
antitoxine.

1 Rosenthal, "Das Dysenterietoxin," Deutsch. med. Woch., 1904, No. 167.

185

III.— THE VEGETABLE TOXINES
(PHYTOTOXINES).

RICINE.

The seeds of the castor-oil plant, Ricinus communis. produce
an extremely virulent poison, which is in every way closely
allied to the bacterial toxines. It is present in the embryo
endosperm but not in the husk (WERNER). Poisons with
completely analogous action are also present, according to
SxiLLMARK,1 in ten other kinds of Ricinus — viz., R. sanguineus,
africanus, guyanensis nanus, altissimus, communis major, philip-
pinensis, brasiliensis, borboniensis arboreus, spectabilis, and jam-
aicensis. The seeds of Jatropha curcas (Curcas purgans), a West
Indian Euphorbiaceous plant, popularly termed the "purging
nut" or "Barbadoes seed," also yield a toxic product which is
possibly identical with ricine (STILLMARK).

Ricine was first investigated by DixoN.2 He prepared a
poisonous product either by extracting the seeds with hydro-
chloric acid and precipitation with a solution of soda, or by
precipitating the substance from an aqueous extract by means
of alcohol. It was obtained in a purer state by treating the
aqueous extract with lead acetate and ammonia, suspending
the precipitate in water and removing the lead by means of
hydrogen sulphide, and then precipitating the toxine with
alcohol. In addition to the toxine a non-poisonous glucoside
was also detected in the seeds.

Ricine was more thoroughly studied by SxiLLMARK,1 working
under the direction of RUDOLF KOBERT. He extracted the seed
with a 10 per cent, solution of ordinary salt, treated the extract
with sodium or magnesium sulphate, and removed the salts by
means of dialysis. The yield amounted to 2-8 per cent., calcu-
lated on the decorticated air-dried seeds.

CRUZS washed the crushed seeds with chloroform and alcohol,

1 Stillmark, "Ueber Eicin," Arb. pharm. Inst. Dorpat, iii., 59, 1889
^ contains a complete bibliography of the earlier literature on the castor-oil
plant).

2 Dixon, Australian Med. Gaz., 1887, 156, quoted by Cushny (vide infra).

3 Cruz, "La ricine," Ann. d'Hyg. PubL, xl., 344, 1898.

186 TOXINES AND ANTITOXINES.

dried them, extracted them with water, and treated the extract
with alcohol.

Great theoretical interest attaches to ricine owing to the fact
that EHRLICH l has made it the starting point of his fundamental
investigations of antitoxic immunity. He, too, separated it from
the seeds by extraction with a 10 per cent, solution of sodium
chloride, and purified it in exactly the same way as STILLMARK.
Merck's preparation is precipitated from the sodium chloride
solution by means of ammonium sulphate.

Chemical Nature of Ricine. — While the older investigators
regarded ricine as a proteid, the progress of discovery appears
to follow the same course here as was described in the case of
the bacterial toxines, and, with the increase in the accuracy
of the investigations, the belief continues to gain ground that
this toxine, although a body of high molecular weight, is not a
proteid in the narrower sense of the term.

STILLMARK regarded it as a globulin. CUSHNYS made many
tedious experiments to prove that ricine was really a proteid,
or to free it from proteid impurities, but did not obtain any
definite results. It is quite impossible to effect a separation of
the active principle from the proteids by the usual methods,
either because the coefficients of precipitation and re-solution
are the same, or, more probably, because ricine, like so many
colloidal substances, is mechanically carried down by many
precipitates as they fall, and especially by coagulated albuminous
substances, &c.

Eventually the resistance offered by ricine to trypsin afforded
a means of preparing a substance which, in all probability at
least, was not of a proteid nature. This property, which was
denied by STILLMARK, but demonstrated by CUSHNY and
MiJLLER,3 formed the starting point for JACOBY'S* striking
experiments.

He first showed that, even after being digested for a week
with trypsin, the toxic power of ricine remained unaltered, the
same result being also obtained with papdin. But he did not, of
course, draw any definite conclusion from this alone, since it was
possible that the proteids might also be able to resist the action

1 Ehrlich, "Exper. Unters. uber Imimmitat," Deutsch. med. Woch.,
1891, 976, 1218; " Zur Kenntnis d. Antitoxinwirkg.," Fortschr. d. Med.,
1897, 41.

2 Cushny, " Ueb. das Ricinusgift," Arch, exper. Path., xli., 439, 1898.

3 Miiller, "Beitrage z. Toxikol. des Ricins," Arch, exper. Path., xlii.,
302, 1899.

4 Jacoby, "Ueb. d. chem. Natur des Ricins," Arch, exper. Path., xlvi.,
28 (reprint).

RICINE. 187

of trypsin. Accordingly he modified his experiment in a very
ingenious fashion.

Bicine is precipitated by ammonium sulphate on saturating
the solution to the extent of 60 per cent., whereas the active
principle of trypsin is not separated until the solution has been
completely saturated with ammonium sulphate. JACOBY, there-
fore, separated from an extract of pancreas by means of fractional
precipitation with ammonium sulphate the proteids that were
precipitated at a saturation of 60 per cent., and used this purified
trypsin preparation for the digestion of ricine that had been
separated by saturating its solution to the extent of less than
60 per cent.

Hence this mixture, before the digestion, would have the
following composition : —

Precipitated at 60 per cent., . Ricine, and proteids com-
bined with it.
- 100 „ . Trypsin.

After the action of the digestive enzyme, however, the composi-
tion would be different; the proteid impurities in the ricine
would have been digested by the trypsin, and would only be
precipitated with more difficulty by the trypsin, and possibly not
at all ; and the composition of the mixture would then be : —

Precipitated at 60 per cent., . Ricine.

„ 100 „ . Trypsin and part of the

digested products.

Under these conditions, then, the precipitation at a saturation of
60 per cent, must yield a pure ricine.

As a matter of fact, JACOBY obtained only a very slight
precipitate by saturating the mixture to the extent of 60 per
cent, after a five weeks' digestion, and he purified this substance
by re-precipitation.

The proteid reactions were no longer obtained, while the toxic
properties were quantitatively retained. Here, then, we have
proof that pure ricine is not a proteid. And this result also
disposes of one of the remaining " toxalbumins," and hence this
term may well be regarded as now only possessing historical
value. It has played an important part in the investigations of
this subject, especially since it led to the proof that these poisons
possessed quite different properties to the crystalloid poisons;
but surely the time has arrived for us to discard it, since now it
can only create confusion. In its place let us use the general

188 TOXINES AND ANTITOXINES.

term " toxine," in the precise connotation that we have described,
which covers all the properties of this characteristic class of
bodies.

Properties of Ricine. — Ricine, like the bacterial poisons,
presents many close analogies with the enzymes, so much so that
it was classified with them by STILLMARK, although it does not
bring about any true fermentative processes, such as diastatic
action, <fec. Among these analogies is its property, mentioned
above, of being carried down by all kinds of precipitates ;
and this mechanical precipitation takes place with special
readiness in the case of subsiding proteid precipitates. Its
precipitation by nucleic acid must also be attributed to a
mechanical process of carrying down, and not to a characteristic
reaction of the toxine itself, such as TICHOMIROFF x claimed
to have observed both in ricine and in other toxines. It
also resembles the enzymes in its sensitiveness to physical
and chemical influences.

The temperature of boiling water is extremely injurious to it,
although, according to JACOBY, it does not entirely destroy its
toxic property, and the purified poison offers greater resistance
than that containing proteid impurities. JACOBY considers it
not impossible that, on boiling, a different poison is produced,
possibly a toxoid. The toxine is not destroyed by dry heat,
110° C. (STILLMARK), and is almost unaffected by pepsin-
hydrochloric acid (MULLER).

We have already described its behaviour towards trypsin, and
it is particularly interesting to note that JACOBY found that his
pure ricine was rapidly destroyed by trypsin, but that this did
not take place when fresh quantities of trypsin were added to
digested mixtures containing the unpurified ricine. Hence, it
would seem, the decomposition products of the proteids exert
a protective influence.

Hydrogen peroxide has an energetic action upon pure ricine,
but is only slightly injurious to the impure substance.

Alcohol does not dissolve ricine and does not injure it.

Ricine appears to be absolutely indiffusible, as was first shown
by STILLMARK.

Action of Ricine. — Ricine has two of the most important pro-
perties of toxines — extraordinarily great activity and a period
of incubation; but, on the other hand, does not show their
strictly specific action. Hitherto no animal has been found that
is completely immune against ricine, but the susceptibility is not

1 Tichomiroff, " Ueber die Fallg. v. Toxalbuminen durch Nuclei'nsaure,"
Zeit.f. phyaiol. Chem., xxi., 90, 1895.

RICINE. 189

the same with all animals, though the differences are no greater
than can be observed in the case of the crystalloid poisons.

Its toxic action is enormous. According to EHRLICH it is
fatal when introduced subcutaneously in the proportion of 0'03
Higrm. per kilo., while 0*18 grm. is a lethal dose per os for a full-
grown man. Subcutaneous injection of 1 grm. would kill 1*
millions of guinea-pigs, though the minimum lethal dose shows
some variation. Mice are less susceptible ; rabbits somewhat
more so. In measuring the toxic power EHRLICH invariably
uses an injection of 1 c.c. for each 20 grms. of body substance ;
on that basis a dilution of 1 : 200,000 would be undoubtedly
fatal to mice.

In like manner CUSHNY found the lethal dose for rabbits to be
0-04 mgrm. per kilo., while JACOBY found that 0'5 mgrm. of
Merck's ricine per kilo, of body weight was required to kill
rabbits.

According to OSBORNE and MANDEL 1 the lethal dose per kilo,
for rabbits of their purified ricine was 0-002 mgrm.

It also shares with toxines the property of acting much more
weakly by way of the digestive tract. Yet its action is always
plainly perceptible even when introduced in this way, though
doses a hundred times as great are required (STILLMARK, p 185).
This property depends on the much more considerable resistance
offered by ricine than by the bacterial toxines to the digestive
enzymes.

The effects produced by ricine can be classified in four groups
— viz., the local action at the point of application, the general
effects, the action on the conjunctiva, and the action on the
blood-corpuscles.

Ricine, like many bacterial toxines, frequently produces severe
indurations, inflammation, abscesses and necroses at the point
of inoculation. Whether these pathological changes are due to
the ricine itself or to impurities has not yet been definitely
determined. There is, however, some support for the view that
these symptoms are possibly only produced, as is frequently
observed in other cases, through the simultaneous introduction
of proteids foreign to the body.

The general - changes that take place in ricine poisoning had
already been studied by the older investigators, and subsequently
by FLEXNER 2 and FRANZ MULLER (loc. cit.) in particular.

The first symptom is an increase in the temperature and a

1 Osborne and Mandel, "Ricine," Amer. Journ. PhysioL, x., 36, 1903.
2Flexner, "The Pathology of Toxalbumin Intoxication," Johns Hopkins
Hospital Record, 1897 (reprint).

190 TOXINES AND ANTITOXINES.

rapid decrease in the weight of the body, while, with these
exceptions, no abnormal phenomena are noticed during the first
twenty-four hours.

The decrease in the weight of the body is much greater than
can be accounted for merely by a condition of hunger. Rabbits
in a state of hunger lose only about T*T of their total weight in
twenty-four hours, |- after forty-eight hours, and |- after seventy-
two hours, whereas rabbits poisoned by ricine lose ^ to ^ of their
weight even after twenty-four hours. Nor can this decrease be
explained solely by increased decomposition of proteids.

According to MULLER the conditions are similar to those
observed in animals suffering from fever. The stools contain
blood and albumen, and frequently blood passes into the urine.

After twenty-four to thirty hours the fatal symptoms of
poisoning suddenly appear. They begin with clonic convulsions
and weakening of the reflexes. Then follows paralysis. After
fifteen minutes the convulsions are repeated, and dyspnoea and
feeble inspiration precede death, which results within about half
an hour after the first attack.

Sometimes the convulsions do not occur. An increase in the
dose does not alter the symptoms, but shortens the period of
incubation.

At the last there are serious central disturbances of the
medulla oblongata : paralysis of the vasomotors and eventually
of the respiration. The pressure of the blood does not fall until
quite late in the attack ; and death is then near. Ricine has no
effect upon the heart.

Post-mortem dissection shows very characteristic results.
Swelling and reddening of the subcutaneous lymph glands, and
great stasis in the region of the blood-vessels in the abdomen ;
great enlargement of and red flecks in the mesentery lymph
glands and of Peyer's patches, and numerous ecchymoses in the
intestine, but no ulcerations. The spleen is much swollen and
soft. Histologically there are characteristic changes in the
blood, and notably severe leucocytosis, decomposition of cells
in the spinal marrow, and necrotic foci in numerous organs,
notably the liver. Thromboses cannot be observed. As a rule,
the muscle of the heart shows fatty degeneration.

CRUZ (loc. cit.) observed, in particular, serious alterations in
the kidneys, but STEPANOFF was unable to detect the toxine in
the urine. Haemorrhage of the suprarenal bodies is also a
characteristic symptom.

Death from ricine is thus primarily due to central paralyses,
but in addition to this the local irritant effects of the poison are

RICINE. 191

shown by the necroses of the cells. The intestinal ecchymoses,
&c., can also be accounted for in the same way, since STEPANOFF l
has shown that ricine is excreted into the intestine, and GUSH NY
(loc. cit.) has found that in frogs ricine causes a discharge of
blood containing the poison into the stomach.

Ricine has no action upon the isolated heart of the frog or
upon the nerves, while it has a slight paralysing effect upon the
muscle (STILLMARK).

The action of ricine upon the connective tissue of the eye is
characteristic. It produces severe conjunctivitis similar to that
caused by abrine (vide infra), which also frequently leads to
permanent derangement of the cornea. Panophthalmitis is also
not uncommon.

The Action of Ricine upon the Blood, — Ricine has a very
characteristic action upon the red blood-corpuscles as was first
noticed by ROBERT and STILLMARK (loc. cit.) • and this can be
observed both in fresh blood and in the case of erythrocytes
that have been repeatedly washed. The erythrocytes agglutinate
under the influence of ricine, and fall to the bottom as a coherent
flocculent mass resembling clotted blood, so that the supernatant
serum is clear.

According to KoBERT,2 a combination takes place between the
agglutinine and the arterine of the blood, and the clumping
together is due to the fact that this compound is viscous. The
action is very intense, and can be recognised even in a dilution
of 1 : 600,000. This agglutination has absolutely nothing to do
with the haemoglobin, since it also occurs in the case of the
dissolved blood-corpuscles — i.e., with the stroma by itself. The
serum appears to contain substances that have a restrictive
influence upon the action of the ricine ; at least it acts more
energetically in diluted blood.

This ricine coagulation is quite distinct from the true coagu-
lation of the blood. Any possibility of its being a simple
coagulation is indeed excluded by the fact that it can be
observed in defibrinated blood ; but apart from this, the process,
as such, can be distinguished from coagulation by the fact that
it is not influenced by sodium chloride, potassium nitrite, and
potassium chlorate. Fibrin coagulation is even retarded by ricine.
Moreover, the so-called ricine fibrin — i.e., the colourless washed
ricine precipitate — is absolutely different from true fibrin.

1 Stepanoff, "Etudes sur la ricine et 1'antiricine," Ann. Past., x., 663 >
1896.

2 Robert, "Ueber vegetabilische Blutagglutinine, " Sitzungsber. d. naturf.
Oes.t Rostock, 1900, xxv. [5] (reprint).

192 TOXINES AND ANTITOXINES.

With large doses of ricine the agglutination is followed by
diffusion of the colouring matter of the blood, so that ricine
has thus also a hcemolytic action. According to JACOBY this
haemolysis is only an intensified form of agglutination and not
a separate process.

The agglutinating action is not restricted to the blood. LAU
found that ricine also caused the agglutination of pus cells and
the cells of organs, that it coagulated milk, and produced coagu-
lations in solutions of white of egg and of plasmon, but did not
affect solutions of myosin and human serum ; while STILLMARK
obtained precipitates by means of ricine in the sera of the dog,
cat, ox, and hen.

The blood-corpuscles of different species of animals vary as
regards their susceptibility, just as they do in the case of
bacterial lysines.

The blood of highly immune animals — e.g., goats — is yet
completely susceptible to ricine poisoning, so that immunity
through lack of receptors, comparable with natural immunity,
does not occur.

LAU * concluded that the blood-corpuscles of fish were abso-
lutely proof against ricine, but FRANKEL 2 showed that this was
only relative and that the introduction of larger doses of ricine
caused agglutination of the blood of even the barbel (Barbus
fluviatilis). This resistance is due to the presence of a normal
anti-body in fish serum. Ricine antitoxine from goats' blood
also possesses this protective power. Yet, on the other hand,
normal barbel serum has no antitoxic effect upon the action of
ricine in cats' blood.

Great interest attaches to the question of how far this action
upon the blood in vitro has a bearing upon the occurrence
of general ricine poisoning. The older investigators, notably
STILLMARK, were inclined to believe that the haemorrhage and
necroses could be explained by a similar process of coagulation
within the lumen of the vessels. Yet, on the one hand,
thromboses have never been found ; and, on the other hand,
as was shown above, these symptoms must be regarded as
irritant effects of the poison in loco. Nor is any stoppage of
the brain arteries suggested by the central symptoms.

But, above all, this theory that the action of ricine upon the
blood might be made responsible as the cause of death is
disproved by the fact that MULLER was able to show that this

^au, " Ueber vegetabil. Blutagglutinine," Dissert., Rostock, 1901.
2Frankel, " Ueb. d. Wirkg. d. Ricins auf Fischblut," ffofm. Beitr., iv.,
224, 1903.

RICINE. 193

agglutination never occurred at all in the living blood. He
also discovered that pepsin-hydrochloric acid checked this action
upon the blood without affecting the toxic power — a result
which, it is true, was denied by LAU, but was conclusively
confirmed by JACOBY (vide infra).

This question is particularly important because it should
furnish an argument for deciding whether ricine is an individual
substance, or whether its action upon the blood ought to be
regarded as quite distinct from its toxic power, in the sense that
there are two separate constituents of the castor seed, exerting
their activity in a different way — a point of view put forward,
e.g., by CUSHNY and MULLER.

It is, however, extremely probable that the question in this
form is not sufficiently precise. It is quite in accord with
Ehrlich's theory that ricine should be a receptor of the second
class of a somewhat complex constitution, possessing only one
haptophore group but two ergophore groups — viz., a toxophore
group which causes the poisoning, and another group producing
the agglutination, and for which the term " agglutinophore "
group has been suggested.

This theory can be readily supplemented by the further
hypothesis that these two ergophore groups differ in their degree
of stability, a difference of susceptibility to digestion with
pepsin-hydrochloric acid being in fact capable of detection, so
that even to explain this phenomenon it is unnecessary to
assume the existence of two substances.

On the other hand, however, it is not as yet possible to demonstrate
more fully the manifold nature of ricine poisons. As FRANKEL has shown,
the antitoxine of normal barbel serum is absolutely devoid of protective
power against the action of ricine upon cats' blood on the one hand, and,
on the other hand, also against its toxic action upon rabbits and even upon
the barbel itself, so that here there is evidence in support of the view that
there is a distinct fish blood agglutinine and fish poison.

We cannot discuss this question more fully until we are able
to consider the action of ricine in the light of the side-chain
theory. As in the case of all true toxines there are four points
that justify a poison being included among the hap tines.

We have already touched upon the small lethal dose, the
incubation period, and the variable toxicological behaviour.
But the most important point of all is the formation of anti-
toxine in the organism, the production of active and passive
acquired immunity.

In this respect, also, the behaviour of ricine exactly corresponds
with that of the bacterial toxines. It was, in fact, experiments

13

194 TOXINES AND ANTITOXINES.

with ricine that led EHRLICH to his epoch-making researches
upon antitoxic immunity.

Immunity against Ricine, — EHRLICH l succeeded in rendering
white mice and rabbits immune against ricine by first introducing
small doses per os (and, in the case of rabbits, also by way of the
conjunctival sac), and then, after attaining a certain degree of
immunity, he was able by cautiously increasing the doses in sub-
cutaneous injection to reach a fairly high state of immunity, the
process taking about four months. Mice previously treated in
this way were capable, after eight weeks, of resisting the lethal
dose for a man. While a dose of 1 c.c. in a dilution of 1 : 200,000
per 20 grms. of body weight was absolutely fatal to the animals
used in control experiments, Ehrlich was able to give doses of
1 c.c. of solutions of 1 : 500, or even 1 : 250, to the immunised
animals, so that from 400 to 800 times the original immunity
was obtained. It was no longer possible to produce pano-
phthalmitis in immune animals even by the introduction of
large doses, although there was still a frequent occurrence of
the local necroses.

Now, the serum of these immunised animals contains an anti-
ricine, which, just like the bacterial antitoxines, is capable of
combining with the ricine in vitro in such wise that both its
toxic and agglutinating action are prevented in accordance with
definite numerical laws. A special point of importance in this
connection is the fact, also established by EHRLICH, that the
same quantity of serum can influence both actions in the same
way. A further proof that there is here a simple combination
between the toxine and antitoxine has been brought by DANYSz,2
who found that on treating a neutral mixture of ricine and anti-
ricine with proteolytic enzymes the antitoxine could be destroyed,
so that the toxic power again appeared ; animals that had been
given an exactly neutral mixture per os died of typical ricine
poisoning.

EHRLICH'S fundamental experiments have since been confirmed
and extended by M. JACOBY 3 in a very exact and theoretically
far-reaching investigation.

JACOBY agrees with EHRLICH'S conclusion that there is a true
quantitative combination in the action of antiricine upon ricine,

1 Ehrlich, " Exper. Unters. iib. Immunitat," Deutsch. med. Woch., 1891,
976, 1218; "Zur Kenntnis d. antitox. Wirkg.," Fortschr. d. Med., 1897,

2 Danysz, "Melanges des toxines avec les antitox.," Ann. Past., xvi., 331,
1902.

3 Jacoby, "Ueber Ricinimmunitat " Hofm. Beitr. z. chem. Physiol. u.
PathoL, i., 51, 1901.

RICINE. 195

and that antiricine consists of broken-off normal receptors. He
has observed a very interesting deviation from the bacterial
toxines. If an active immune serum be brought into contact
with solutions of ricine a distinct precipitate is formed, which is
also the case when solutions of antiricine are used (vide infra).
On the other hand, there is no precipitate produced when active
ricine is treated with normal serum or with destroyed antiricine,
or when ricine rendered inactive by boiling is brought into con-
tact with antiricine.

DANYSZ (loc. cit.) simultaneously investigated this precipitate, and found
that there was invariably an optimum of admixture at which the strongest
precipitation occurred. We shall return presently to the conclusions that
he drew from this result.

This combination of ricine with antiricine is, as JACOBY rightly
concludes, a primary one. The resulting neutral compound is,
however, in this case only sparingly soluble, and thus forms a
precipitate which also carries down, as usual, other proteids
of the serum ; for this is the only way of explaining the amount
of the precipitate. But there can be no question of any precipita-
tion of proteid by some preeipitine that has been formed, and of
this precipitate then carrying down with it the mixture of ricine
and antiricine ; for even ricine that is free from albumin produces
in the organism a serum that gives the albumin reactions. More-
over, the strictly quantitative combination that is formed is
opposed to the view that there might be any removal of the
toxic power of the ricine by adsorption during a simultaneous
precipitation of proteids • for the proteid precipitates containing
ricine that have been formed in another manner — e.g., by nucleic
acid (vide supra) — are poisonous, whereas this precipitate is not.
But as soon as there is any excess of ricine present, the filtrate
remains poisonous in exactly the degree corresponding to that
excess. A removal of toxic power so quantitatively regulated
cannot be accounted for by adsorption.

An apparently paradoxical phenomenon can be simply ex-
plained by the theory. The washed erythrocytes of highly-
immunised animals are at least as susceptible, and apparently
still more susceptible than the normal erythrocytes. It is
theoretically conceivable that the power of forming receptors
may become temporarily spent, so that the erythrocytes may
contain few or none at all, with the result that their suscepti-
bility may be very insignificant, or nil ; this appears to happen
frequently in the case of eels' blood (q.v.).

Conversely, the theory can also assume the possibility of the
erythrocytes being endowed with the power of a very vigorous

196 TOXINES AND ANTITOXINES.

formation of erythrocytes, so that they thus contain more than
the normal quantity, and that their susceptibility is therefore
unmistakably increased.

On the other hand, it is obvious, and has been confirmed
by the facts, that the erythrocytes in the natural serum of
immune animals receive very considerable protection from
the presence of anti-bodies in their serum. In one experiment
JACOBY found such blood to require ten times as much ricine
to produce the maximum agglutination as in the case of normal
blood.

This phenomenon of the haptines combining with the free side
chains in preference to those in combination has been observed
in the case of all toxines. The free receptors appear almost
absolutely to have a greater degree of affinity than those attached
to the cells.

This fact forms the basis of every " cure " in cases of toxine
poisoning — i.e., the disruption by means of the an ti toxine of
a combination already formed with the cell. We have already
shown in their respective places that this curative action is only
possible for a very short time after the combination in the case
of diphtheria and tetanus ; a very similar state of things is also
found with the blood poisons, for MADSEN was able to detect in
the case of tetanolysine (q.v.), and JACOBY, in the case of ricine,
an arrest of agglutination through the subsequent addition of
antiricine.

But, in spite of its slight affinity, the haptophore group of
ricine combines in the absence of free receptors just as quanti-
tatively with those attached to the erythrocytes ; and just as
tetanus poison becomes fixed to the central nervous system so is
ricine quantitatively attached to the erythrocytes, so that the
mixture has no action upon a fresh quantity of blood.

The agglutinating action is thus completely paralysed through
the quantitative combination of its substratum with the erythro-
cytes. We ought, therefore, to conclude a priori that if ricine is
an individual substance the toxic power of such mixtures should
also be destroyed. This, however, is not the case. MULLER
found that the filtrates from the precipitates produced by ricine
were devoid of agglutinating power, but yet were poisonous,
although to a considerably reduced extent, and without producing
the typical appearances on post-mortem dissection.

These experiments were not regarded as quite conclusive by
JACOBY. In his opinion it was possible that there might be still
some ricine mechanically carried down in the precipitate formed
in the blood, and that this ricine might again be liberated during

RICINE. 197

the slow filtration.1 But even JACOBY'S own experiments, made
with unfiltered blood rendered non-coagulable, gave the same
results. The mixture had the same degree of toxic power. He
himself, however, raises the very important objection that
although the poison in this way reaches the cells of the tissues
in a combined form, these cells may very well possess the power
of decomposing this combination and of attracting the poison to
themselves to their own injury. For the receptors of the tissues,
like the free receptors, may possess greater affinity for the toxine
than do the erythrocytes, although, again, it must be less than
that of the free receptors ; otherwise no immunity against the
toxic action could be produced.

This view is also supported by an argument which, to my surprise, was
overlooked by JACOBY. If we assume that the tissue receptors combine
more readily with the haptophore group of the ricine than does the latter
with the receptors of the blood-corpuscles, we have an explanation why in
cases of the poisoning of living animals with ricine the symptoms produced
in the blood are so slight in comparison with the general symptoms, a fact
for which it is otherwise difficult to account.

These experiments, too, do not decide the question whether or
no ricine consists of two separate substances.

JACOBY, in a further research,2 made use of Ehrlich's method
of separating haemolysines, by subjecting mixtures of ricine with
blood-corpuscles to centrifugal force. The agglutinating power
was invariably found to have disappeared from the liquid,
whereas the amount of toxine varied from 25 to 90 per cent, of
the original amount. The poison had never combined quanti-
tatively with the receptors. Its action was also qualitatively
unchanged.

JACOBY next rendered animals immune by means of this
poison freed from the agglutinine. The immune serum thus
obtained showed not only antitoxic, but also anti-agglutinating
power, although the plasma poison freed from agglutinine
required less antitoxine than ordinary ricine.

Exactly similar results were obtained with ricine that had
been previously treated with pepsin-hydrochloric acid. This
agglutinine-free poison also produced an antitoxine which had an
influence upon both functions of the ricine. The action of
pepsin-hydrochloric acid upon ricine had already been studied by
JACOBY in his earlier investigation, and he had been able to
confirm MULLER'S assertion that the agglutinating power was
considerably reduced, even to -^ of its original amount.

1 Loc. cit., p. 68. That is, if I have correctly understood the drift of
the argument, which is somewhat too briefly expressed.

2 Jacoby, "Ueber Ricinimmunitat," Hofm. Beitr., ii., 535, 1902.

198 TOXINES AND ANTITOXINES.

But the name amount of antitoxins was required after the
experiment as before to neutralise the greatly reduced agglutinat-
ing power, as well as the unaltered toxic activity. One c.c. of
antitoxine was required to neutralise the extraordinarily small
agglutinating power of 5 c.c. of pepsin-ricine ; but the same
quantity was also sufficient to destroy the toxic activity of the
same quantity corresponding to at least 15 mgrms. of ricine (30
lethal doses). But, more than that, it was possible still to add
enormous quantities of poison (up to 8 c.c. of pepsin-ricine)
without killing the animals, though they certainly became
emaciated.

One is strongly reminded by these facts of the conditions
observed in the case of bacterial toxines, especially those of
tetanus and diphtheria. There, too, EHKLICH found that more
than a single lethal dose had to be added to a completely neutral
mixture (L0) in order to obtain L+ (see the value D in the
General Part).

Another very interesting result was shown by JACOBY'S ex-
periments— viz., that whereas 1 c.c. of antitoxine was required
to neutralise O26 c.c. of ricine solution before the treatment with
pepsin, the same quantity was sufficient for 5 c.c. after the treat-
ment. At the same time the toxic action remained unchanged,
so that any possible destruction of the ricine molecules was out
of the question.

It surely follows from these experiments that a large number
of the haptophore groups must have disappeared, which, in the
untreated ricine, had combined with the receptors of the serum,
but, subsequently, no longer had any attraction for the anti-
ricine. But these haptophore groups could not have corresponded
with the toxophore groups, since the toxic power remained un-
changed. We are thus inevitably driven to the conclusion that
non-poisonous haptophore groups must have been present in the
crude ricine. JACOBY is thus justified in his conclusion that
there are ricine toxoids which are destroyed by pepsin-hydrochloric
acid. And they must also be syn- or protoxoids^ since they are
simultaneously saturated on neutralisation.

The conditions are quite the reverse in the case of the
agglutinating function. It is true that the number of hapto-
phore groups also shows a decrease here, but to a smaller extent
than that of the ergophore groups. If, now, we assume that the
haptophore groups are the same, it is probable that there are
here formed new toxoids of a peculiar kind, which consist of only
the haptophore and toxophore group, and have lost their ayglutino-
phore group.

RICINE. 199

And in support of the view that the haptophore groups are
actually identical, and that ricine is thus an individual haptine,
though of complex structure, we have, above all, the fact that
the antitoxic and anti-agglutinating action remain just the same
after the treatment with pepsin, and the results of immunising
experiments with ricine devoid of agglutinine (vide supra). In
the case of hap tines that were undoubtedly distinct, such as, e.g.,
tetanospasmine and tetanolysine, EHRLICH was unable to observe
any trace of such a parallelism.

On the other hand, much can also be said against their
identity, so that the question cannot yet be regarded as definitely
settled. But if we grant their identity commercial ricine must
consist of the complex groupings represented in the diagram —

toxophore W^^ agglutinophore

haptophore

and also of toxoids, either without any ergophore group,1 or with
only the agglutinating group thus—

These are eliminated by pepsin, possibly with the formation
of toxoids with groupings in this form —

Fresh ricine, like diphtheria poison, appears to contain less
toxoids, for according to JACOBY, Merck's preparation gradually
decreases in its activity.

In addition to these bodies, which we are obliged to regard as
pro- or syn-toxoids, there appear to be also toxones of ricine,
though JACOBY has not touched upon this point ; at least this
conclusion can be drawn from the ratios of D (L+ - L0) given

1 This is really the more probable ; it is difficult to conceive that there
should be toxoids containing such a sensitive group as the agglutinophore
group. This question, however, is of quite secondary importance.

200 TOXINES AND ANTITOXINES.

above, which exactly recall those observed in the case of
diphtheria poison.

It is as doubtful in the case of ricine as in that of the bacterial
poisons whether there are not cases of dissociated equilibrium
here. DANYSZ (loc. cit.), in fact, observed very striking ratios.
The fact that there is an optimum point of mixture (vide supra)
for the production of the largest precipitate is only a fairly
weighty argument ; and a far more important reason is that,
according to DANYSZ, there is never an absolutely neutral mix-
ture. He found that these mixtures invariably possessed both
toxic and antitoxic action — i.e., that although they themselves
had a slight toxic action they were yet able to retard or
altogether prevent death on the addition of a whole lethal dose.
His speculations based on these facts are very similar to those
of BORDET (see General Part). It is probable that the truth is
that we have here to deal with dissociated conditions of equili-
brium, such as were found by AERHENIUS and MADSEN in the
case of tetanolysine.

ANTIRICINE.

JACOBY has also made attempts to isolate antiricine. It may
at once be mentioned that any separation of an antitoxine from
an antiagglutinine has not been found possible by any method,
and that, notwithstanding the theoretical considerations given
above, we have thus in practice to deal with an individual
substance.

Antiricine when salted out with ammonium sulphate is quan-
titatively precipitated in the fraction that falls when the degree
of saturation reaches ^- to J- of the total amount. In this way
it can be separated from a large proportion of the other colloids.

Experiments with trypsin on the same lines as described above
in the case of ricine showed that it absolutely resisted the action
of that enzyme.

Antiricine was quite unaltered after two hours' heating at
60° C., and after digestion for thirty minutes with an equal
amount of N/10 sulphuric acid or N/10 soda solution at 37° C.
Pepsin-hydrochloric acid had no effect upon antiricine after an
hour at 35° C. On the other hand, it was destroyed by acids
at 60° C.

Antiricine thus appears to be a substance resembling the
bacterial antitoxines in stability. It is a simple receptor of
the first class, but provided with only one haptophore group.

STEPANOFF (loc. cit.) was able to detect antitoxine in the blood
of rabbits, even twenty-four hours after the injection of serum

A BRINE. 201

containing antiricine. After seven days both the antitoxine and
the power of resisting the poison had disappeared. He found it
neither in the urine nor in the intestinal tract, and there is thus
no evidence of its being excreted ; presumably it was oxidised
in the system.

ABRINE,

Abrine, a toxine closely resembling ricine, is contained in the
seeds of the jequirity (Abrus precatorius\ a plant indigenous to
the East Indies, and possibly also to Brazil. It is found in
almost every part of the plant (HENSEVAL l).

It was discovered by BRUYLANTS and VENNEMANN 2 who were
the first to announce that the active principle of jequirity seeds
had nothing to do with bacteria, but was an enzyme, a toxic
proteid. The same results were obtained by WARDEN and
WADDELLS working under the direction of ROBERT KOCH in
Calcutta. SIDNEY MARTIN 4 found that abrine consisted of a
globulin and an albumose. The action of the albumose resembled
that of the globulin, but was much weaker — i.e., according to
our present views more poison attaches itself to the globulin
than to the albumose, while the true poisonous principle is
distinct from either.

Its hsemolytic capacity was discovered by ROBERT in 1889
and investigated by his pupil HELLiN.5

Its action is so similar to that of ricine that EHRLICH (loc.
cit.) considered it necessary to prove its individuality. Some
differences were observed which showed beyond doubt that
abrine, although very similar to ricine, was yet a distinct
substance.

It is much less poisonous than ricine, especially per os.
EHRLICH found that solutions of the same strength (1:100,000)
killed mice, on subcutaneous injection, in six days, whereas in
the case of ricine they caused death even after sixty hours.
On the other hand, CALMETTE (vide infra) found 0'5 mgrm.
per kilo, fatal to rabbits within forty-eight hours, while the

1 Henseval, " L' abrine," La Cellule, xvii., 139 ; Malys Jb., xxxi., 910.

'-'Bruylants and Vennemann, " Le Jequirity," Bull Acad. Med. Belg.,
[3], xviii., 147, 1884.

3 Warden and Waddell, Non-bacillar Nature of Abrus Poison, Calcutta,
1884.

4 Martin, "The Proteids of the Seeds of Abrus," Proc. Roy. Soc., xlii.,
331, 1887; Martin and Wolfenden, "Physiol. Action of the Seeds of
Abrus prec.," ibid., xlvi., 94, 1889-90; Martin, "The Toxic Action of
the Albumose from Seeds of Abrus prec.," ibid.t xlvi., 100, 1889-90.

5Hellin, " Der giftige Eiweisskorper Abrin.," Dissert., Dorpat, 1891.

202 TOXINES AND ANTITOXIN ES.

lethal dose for a mouse was 0*001 mgrm. The lethal dose for
1 grm. of mouse was found by RoMER1 to be 0'0005 mgrm.

Like ricine it produced inflammation, but necroses were seldom
observed. On the other hand, a specific action of abrine is that
it causes the hair to fall out rapidly around the point of injection,
until complete baldness results.

The appearance on post-viortem dissection is almost the same
as in the case of ricine ; an additional effect being a characteristic
hydropic degeneration of the muscle of the heart (WERHOFSKY2).
Unlike ricine it also acts in small doses on fishes' blood (LAU,
loc. cit.).

But the conclusive fact is that a state of immunity can be
produced against abrine, which does not afford protection against
ricine ; in like manner animals rendered proof against ricine are
not immune to abrine.

Action upon the Eye. — The action of abrine upon the conjunc-
tiva is much more energetic than that of ricine, injection being
followed by serious permanent lesions of the cornea or by
panophthalmitis. In consequence of its energetic irritant
action abrine is sometimes used in ophthalmic medicine, the
violent inflammation produced frequently causing the dis-
appearance of new inflammatory vessels and leucoma in the
cornea. The maximum initial therapeutic dose is given by
ROMER 3 as about O'Ol mgrm. for rabbits.

According to EHRLICH the action of abrine can be regulated
by immunising the conjunctiva itself with increasing doses. For
practical purposes ROMER has carried this immunisation to such
a pitch that it is possible to avoid all the more serious symptoms
without affecting the therapeutic action. Antitoxine is then
produced in the conjunctiva itself.

Part of the poison is also absorbed from there and produces a
general formation of antitoxine and immunity, although not to
so pronounced an extent as after subcutaneous injection. The
greatest degree of immunity attained was 500 antitoxic units,
and local immunity was produced more rapidly than general
immunity. Passive immunisation of the connective tissue with
antiabrine serum only affords protection so long as the antitoxine
itself is still present. On the other hand, the subcutaneous in-
jection of large doses of antitoxine also protects the eye. An
attack of abrine ophthalmia can be successfully treated by local

1 Romer, "Ueber Abrinimmunitat," Arch.f. Ophthalm., lii., 90, 1901.

2 Werhofsky, " Beitr. z. pathol. Anat. der Abrinvergiftung," Zeiglers
Beitr. %. pathol. Anat., xviii., 115, 1895.

3 Romer, "Ueber Abrinimmunitat," Arch.f. Ophthalm., lii., 72, 1901.

ABRINE. 203

application of therapeutic serum, even in very severe cases. The
subcutaneous introduction of large doses also has a therapeutic
action. The local immunity of the eyes lasts longer than the
general immunity.1

Properties of Abrine. — Experiments to obtain an insight into
the constitution of abrine have been made by HAUSMANN,2 who
has applied to this toxine the method used by JACOBY for ricine
(q.v.).

Abrine is precipitated by saturating its solution with am-
monium sulphate to the extent of 60 per cent., and can be
freed from part of the proteid thrown down with it by repeated
precipitation. Abrine thus purified has a very energetic toxic
action, so much so that animals frequently die before the appear-
ance of necroses and intestinal symptoms.

Abrine resembles ricine in resisting the action of trypsin.
Hence it was found possible, as in the case of ricine, to obtain
a preparation which, while possessing a high degree of toxic
power, no longer gave the biuret reaction.

On the other hand, its agglutinating power is very much more
resistant to the action of pepsin-hydrochloric acid than is that of
ricine (q.v.), and when the action of the enzyme is very energetic
the toxic and agglutinating functions eventually disappear almost
simultaneously.

With this exception there is nothing special to be said with
regard to the chemical reactions of abrine ; so far as is known
it behaves in exactly the same manner as ricine in this respect.
According to CALMETTES it is rendered inactive by iodine tincture,
gold chloride, and hypochlorites.

It is apparently but little affected by the digestive fluids.
H ELLIN, alone, has found that it is destroyed by the enzymes
of the intestine. On the other hand, according to NENCKI and
ScHOUMOW-SiMANOWSKi 4 pepsin has no influence upon abrine.

In like manner it was found by REPiN5 that the diluted
digestive fluids and also the living mucous membrane of the
stomach and intestine and the intestinal bacteria were inactive.
He is inclined to believe that the smaller toxic effects on intro-
duction of the poison per os, which, according to HENSEVAL (loc.

1 Rehns, "Immunite acquise centre 1'abrine," Soc. Biol., Ivi., 329, 1904.

'2 Hausmann, "Zur Kenntnis des Abrins," Hofm. Beitr., ii., 134, 1901.

3 Calmette and Del^arde, "Sur les toxines non-microbiennes," Ann.
Past., x., 675, 1896.

4Nencki and Schoumow-Simanowski, "Die Entgiftung d. Toxine d. d.
Verdauungskanal," Centralbl. f. Bakt., xxiii., 840, 1898.

5 Repin, "Sur 1'absorption de 1'abrine par les muqueuses " Ann. Past.,
ix., 517, 1895.

204 TOXINES AND ANTITOXINES.

cit.), are from 200 to 250 times less than in the case of sub-
cutaneous injection, are due to the fact that, on the one hand,
abrine is extremely susceptible to the action of acids and is thus
partially destroyed in the stomach ; and that, on the other hand,
it is extremely indiffusible. He found that after forty-eight
hours' dialysis the proportion that had passed into the water
did not amount to 1 : 250. Hence it remained in the intestine,
and he was able to find it again in the faeces. HENSEVAL,
however, asserts (loc. cit.) that it is also absorbed by way of
the intestine and rectum, as well as of the bladder and
peritoneum.

CALMETTE and DELEARDE investigated the excretion of abrine.
The blood from the heart of poisoned animals was found to be
toxic in large doses (10 mgrms. introduced by intravenous in-
jection) ; but the urine was absolutely free from toxine. On
the other hand, however, they found that abrine appeared again
unaltered in the intestinal tract, when injected intravenously
into rabbits. Neither the blood from the heart nor the contents
of the intestine of immunised animals were found to contain any
poison that had been introduced.

An interesting fact which CALMETTE1 was able to demonstrate with the
aid of specific antitoxines is that the poisoned wood shavings with which
the Indians, with malevolent intent, poison cattle are smeared with
particles of jequirity ; the poison on the rags impregnated with snake-
venom and inserted into the rectum of oxen so as to poison them, could
also be identified in the same way.

Some experiments made with antiabrine serum by CALMETTE
and DELEARDE may also be mentioned here.

Antiabrine loses its activity at 58° C. Calcium chloride and
gold chloride have no effect upon it.

According to HAUSMANN (loc, cit.) the agglutinating power is
considerably increased by the addition of very slight doses of
antiabrine, which is doubtless to be attributed to the elimination
of inactive proagglutinoids (protoxoids), which have a restrictive
influence.

On mixing abrine and antiabrine a copious precipitate is
formed, as in the case of ricine, even when abrine that gives no
biuret reaction is used.

1 Calmette, "Sur le serum antivenimeux," Comptes Rend., cxxii., 203,
1896.

CROTINE. 205

CROTINE.

A third toxine closely allied to ricine is found in the seeds of
Croton tiglinm, an East Indian plant belonging to the Euphor-
biacece, from which croton oil, the most powerful purgative known
to us, is extracted.

STILLMARK (loc. cit.) was the first to prepare the toxic principle
of these seeds, using methods analogous to those employed in the
preparation of ricine.

A fuller investigation of crotine was made by ELFSTRAND/ He
decorticated in the presence of alcohol and ether, and extracted
them with water, a 10 per cent, solution of sodium chloride or
glycerin, precipitated the active constituent by means of alcohol
or ammonium sulphate, and purified it by dialysis.

It presents many analogies with the other toxines. In solu-
tion it is destroyed at 70° C., while the dry seeds are rendered
non-poisonous at 110° C. Pepsin-hydrochloric acid is said to
destroy it.

Toxic Effects of Crotine. — The lethal dose for frogs was found
by ELFSTRAND to be about 0'23 grm. per kilo. The symptoms
preceding the death of the animals were progressive paralysis and
decrease in the reflex excitability and of the faradic excitability, Jv.
affecting first the brain, then the spinal cord and nerves, and -
finally the muscles. Here again hypersemia and ecchymoses of ^
the intestinal membrane were also observed. Crotine has only
a slight influence upon the heart, and the end -plates of the
motor and sensory nerves are not affected.

Pike are killed by a dose of 0*04 to 0-1 grm. per kilo., the main
symptoms being dyspnoea and paralysis.

In the case of warm-blooded animals (rabbits, dogs, cats, rats,
hens, <fec.), local inflammation and necroses are also produced by
crotine, although only to a slight extent. It is far less poisonous
than ricine or abrine (lethal dose, about 0'05 to 0-1 grm. per
kilo.), and is also much less active when introduced per os.

The general symptoms are very similar to those produced by
ricine, viz., convulsions, lowering of the pressure of the blood,
diminution of temperature, respiratory paralysis, &c. A slight
effect upon the eye can also be detected.

The appearance on post-mortem dissection is also very similar.

Action upon the Blood. — ELFSTRAND found that crotine also
had an agglutinating action upon the blood of oxen, sheep, pigs,
pike, and frogs, and a very much smaller effect upon that of cats,

1 Elfstrand, " Ueber blutkorperchenagglutinierende Eiweisse," Gtirbers-
dorfer Veroffentlichungen, edited by R. Robert, Stuttgart, 1898, 1.

206 TOXINES AND ANTITOXINES.

while it had hardly any action upon human blood, and none at
all upon the blood of dogs, guinea-pigs, rats, hens, geese, and
pigeons.

LAU (loc. cit.) supplemented these results by others, in which
he observed agglutination in the blood of the perch, but none in
that of the cat and hedgehog. Neither investigator found any
agglutination in the case of rabbits' blood, but there was a haemo-
lytic action. Crotine is stated not to possess the power of pro-
ducing a precipitate with serum. It has an equal agglutinating
action upon the washed erythrocyfces and stroma. Oxygen is
said to promote the action of crotine, while antitoxic substances
that restrict its influence are present in the serum. It has no
influence, or at least a very much slighter one than ricine, upon
pus cells and other cells, but, on the other hand, it causes milk
to coagulate. EHRLICH and MoRGENROTH,1 and also JACOBY^
have studied the mode of action of crotine hsemolysine with the
aid of the modern methods of investigating haemolysines.

MORGENROTH proved that crotine was a haptine by the fact
that he was able to produce an immune serum in goats.

JACOBY found no support for the view that crotinolysine con-
sisted of an amboceptor and complement, and was inclined to
regard it as a haptine of the first class, like ricine.

By means of his experiments with partial saturation (c/.
General Part), he has established that crotine must possess a
complex constitution like that of diphtheria toxine, but it is
noteworthy that very small doses of antitoxine increase the toxic
action to a slight extent, since they eliminate the absolutely non-
poisonous prototoxoids, which otherwise combine with part of
the cell receptors, and thus reduce the toxic power. Then very
rapidly, with the increase in the doses of antitoxine, the bulk of
the poison is neutralised, and after that comes a wide zone with
very slight affinity — i.e., the toxones — which no longer effect
complete haemolysis, but with which JACOBY was yet able to
produce immunity.

As is the case with other blood poisons (see, e.g., Arachno-
lysine), insusceptibility and lack of power to combine with the
poison also go hand-in-hand here. The blood-corpuscles of dogs
and guinea-pigs do not combine with any trace of crotine.

A thermostable anticrotine, which apparently restricted the action of
crotine in accordance with quantitative laws, was found by JACOBY in the
extract of the mucous membrane of the stomach.

iEhrlich, "Verh. Ges. Charity-Aerate, Feb. 1898," Berl klin. Woch.,
1898, No. 12.
2 Jacoby, "Ueber Crotin-Immunitat," Hofm. Beitr., iv., 212, 1903.

HAY FEVER TOXINE. 207

It was more fully investigated by LusT,1 who found that it could be pre-
cipitated by alcohol, that it was " salted out" by ammonium sulphate (60
per cent, solution), and was not attacked by pepsin-hydrochloric acid.

ROBINE.

Robine, a fourth vegetable toxine, resembling ricine, was
discovered by POWER and CAMBiER2 in 1890, and described by
them as a phytoalbumose. Its power of agglutinating the blood
was also discovered by ROBERT. It is found in the bark of the
so-called acacia, Robinia pseudacacia.

Its poisonous character had already been recognised as the
cause of accidents to men and animals, several instances of which
are cited by LAU (loc. cit.).

It acts upon the blood in a similar way to ricine, though its
action is considerably weaker, and is absolutely lacking in the
case of the blood of cats, dogs, and man.

In like manner its toxic power is disproportionately weaker
than that of ricine and abrine. A dose of as much as 1 0 grms.
per kilo, of body weight of robine freed from impurities by
precipitation with potassium ferrocyanide and purified by re-
precipitation with acetic acid (MERCK'S commercial preparation)
was required to kill a rabbit in four days. Post-mortem dissec-
tion showed nephritis, but otherwise nothing characteristic.

EHRLICH succeeded in producing immunity against robine, and
discovered the fact that highly-immunised animals were also
proof against ricine. Hence he is inclined to believe that robine
is a toxoid of ricine. JACOBY has shown, as we have seen, that
such ricine toxoids probably exist.

A closer investigation of robine is urgently needed, since this
question is of the greatest theoretical importance.

HAY FEVER TOXINE.

Poisonous substances of a proteid nature have recently been
regarded as the primary cause of hay fever and "autumnal cold."
Toxines are present, according to DUNBAR and WEICHART,S both

1 Lust, " t)b. einen Antikorper gegen Crotin im normalen Organismus,"
Hofmeisters Beitr., vi., 132, 1904.

2 Power and Cambier, Pharmac. Journ., 1890, 711; Pharm. Rdsch.,
Feb. 1890, p. 30.

3 Dunbar, "Zur Frage betreffend die Aetiologie u. specif. Therapie des
Heufiebers," Berl. klin. Woch., 1903, 24-26; id., "Z. Aetiol. des Herbst-
katarrhes," ibid., 1903, 28.

208 TOXINES AND ANTITOXINES.

in the pollen of grasses which produce hay fever in the spring,
and in that of Ambrosiacece, to which is attributed the "autumnal
cold " of America.

It is sometimes possible to produce an antitoxine by immunis-
ing horses.

Further investigations are being made.

209

IV. -THE ANIMAL TOXINES (ZOOTOXINES).

SNAKE TOXINES.

Although venomous serpents have long been an object of fear
and of interest to widely different races of man, yet the history
of the investigation of their venom is quite recent. And this,
one must admit, is remarkable, since surely nothing should have
suggested itself more naturally to the scientific investigator than
the application of recent results in toxicology, especially in con-
nection with vegetable alkaloidal poisons, to the study of these
poisons, which are as interesting to the investigator as they are
important from the point of view of public hygiene. For, in
India alone, more than 20,000 men perish annually from the bite
of the cobra, Naja tripudians. And yet this branch of research
remained almost completely untouched until the investigations
into the nature of bacterial poisons, and especially those in-
augurated by METSCHNIKOFF, Roux, and YERSIN, compelled
attention to be directed also towards these poisons, which have
similar incredible toxic power. There were also external reasons
for this neglect. The material for these investigations, at all
events in the case of the most important venomous snakes, was
hardly to be obtained in Europe. It was only when American
medical science began to develop vigorously, and when the
foundation of the modern Colonial Empires led to the study of
tropical medicine, that the investigation of tropical venomous
snakes also received special attention. We do not, of course,
mean to assert that the poisonous characteristics of venomous
snakes had not been frequently studied ; this was notably the
case with the South European vipers, and we shall give an
outline of these earlier results in their proper place. But no
systematic chemical and pharmacological investigation of the true
poisonous substances was made until a relatively late period,
when the study of bacterial poisons had opened up quite new
avenues. Of the older researches mention may be made of those
of FoNTANA,1 FAYRER and BRUNTON,2 and WALL,3 with whose
results we shall deal presently.

1 Fontana, Trattado del vde.no della vipera, 1787.

2 Fayrer and Brunton, "On the Nature of the Poison of Naja tripudians,
&c.," Proc. Roy. Soc., xxii., 68, 1874.

3 Wall, "On the Poisons of Certain Species of Indian Snakes," Proc.
Roy. Soc., xxxii., 333, 1881.

14

210 TOXINES AND ANTITOXINES.

Naturally, search was first made, as we find almost universally in the
history of toxines, for alkaloidal substances comparable with the ptomaines.
Thus GAUTIER, in 1881, isolated two alkaloidal substances, nain and
elaphin, from the venom of the naja and the Trigonocephalus (American
fer-de-lance) respectively. He himself, however, had to admit that these
substances were relatively harmless ; and hence, in this case, science
was spared the process of disillusion which otherwise followed almost
universally the original over-valuation of the ptomaines.

GAUTIER, therefore, concluded that the "true active principle
of snake poison contained nitrogen," but that it was " not of the
nature of an alkaloid."

About the same time, WEIR MITCHELL and REiciiERT,1 in
America, made a fuller investigation of snake poisons.

The venom of European vipers (Pelias berus) had already
received somewhat more attention (FONTANA, VALENTIN,2 and
others). The true stage of development of this new branch of
biology begins with these researches. Then follow the classical
investigations of CALMETTE, to which, supplemented by those
of MARTIN, FRASER and PHISALIX, and the quite recent work
of FLEXNER, KYES and SACHS, our present knowledge is, in
the main, due.

By a lucky chance, CALMETTE, 3 who was then head of the
bacteriological institute at Saigon, obtained possession of the
fresh poison glands of 22 cobras, and this formed the starting
point for his classical investigations.

The poison glands of the serpents, which are similar to the
salivary glands, are the true source of the poison; but CALMETTE 4
found that the Uood of the cobra was also fairly poisonous. The
intravenous injection of 2 c.c. killed a rabbit weighing 1,500
grms. in three minutes. On the other hand, the liver and gall
are not poisonous. Even the blood of otherwise harmless
serpents (Tropidonotus) was found to be poisonous by PHISALIX
and BERTRAND.5 This blood poison has certain peculiarities
with which we shall deal presently.

The poison gland of the cobra yields on expression about
3 grms. of a transparent viscous fluid, which when exposed to

1 Weir Mitchell and Reichert, " Researches upon the Venoms of Poison-
ous Serpents," Smithsonian Contrib., No. 647, Philadelphia, 1885; Wash-
ington, 1886 ; quoted by Flexner, loc. cit.

2 Valentin, "Einige Beobachtg. lib. d. Wirkg. des Viperngiftes," Zeit. f.
BioL, xiii., 80, 1877.

3Calmette, "Etude experimentale du venin de Naja tripudians" Ann.
Past., vi., 160, 1892.

4Calmette, "Sur la toxicite du sang de cobra," Soc. BioL, xlvi., 11, 1894.

5 Phisalix and Bertrand, "Sur le presence des glandes venimeuses chez
les couleuvres," Soc. BioL, xlvi., 8, 1899.

SNAKE TOXINES. 211

the air congeals into masses. CALMETTE1 treated these glands
with glycerin, with distilled water, and with a 10 per cent,
solution of salt, and invariably obtained extremely poisonous
extracts. The toxine acts most energetically when directly
injected into the veins, and has less effect when injected sub-
cutaneously or into the peritoneum and trachea ; it is absolutely
without action when introduced into the intestine.

The quantity of poisonous saliva secreted was found by
CALMETTE (1895) to amount, on the average, to 0-135 grm.,
corresponding to about 30 to 45 mgrms. of dry substance,
provided that from eight to fourteen days were allowed to
elapse between the separate bites. On the other hand, after
a lapse of two months each bite yielded about 0'22 grm. of
saliva. The greatest amount that he was able to extract from
the two poison glands of a dead snake was 1'136 grms. = 0'48
grm. of dry substance. Similar results were obtained in the
investigation of other venomous snakes.

Thus, there is invariably a considerable increase in the amount
excreted, and consequently in the danger from a bite when the
snake has not bitten for a considerable time. The bite of
hibernating snakes — e.g., the European vipers — is thus the most
dangerous in the spring.

Preparation of the Poisonous Principle. — Snake toxines have
not been prepared in even an approximately pure state. The
methods of concentration are exactly the same as those employed
for all toxines and enzymes.

The poisonous principle can be extracted by water, salt
solutions or glycerin, and purified by repeated precipitation,
dialysis, &c.

MARTIN succeeded in eliminating part of the inactive sub-
stances present, by means of fractional coagulation, for on
heating a solution of holocephcdus poison in a 0*9 per cent,
solution of sodium chloride, a foreign substance separated at
85° C., while the true poison remained still active at 90° C.

CALMETTE2 subsequently used the following method in the
preparation of a stable poison relatively free from proteids : —
The solution of 1 grm. of cobra poison in 100 c.c. of water was
filtered through sterilised filter paper, then heated in a hermeti-
cally closed glass tube for thirty minutes at 75° C., and then for
twenty-four hours at 80° C., after which it was filtered through
paper to remove the separated substances and finally dialysed.

1 Calmette, "Etude expe"rimentale du venin de Naja tripiidians," Ann.
Past., vi., 160, 1892.
2 Calmette, " Sur le venin des serpents, &c.," Ann. Past., xii., 214, 1897.

212 TOXINES AND ANTITOXINES.

In this way he obtained 42 mgrms. of a dry residue, which
still showed the biuret and Millon's reaction, but no other
proteid reactions. The poison will pass readily through a
Chamberland filter.

Chemical Nature of the Toxine. — After it had been found that
the active agent was not an alkaloid there followed the usual
period of toxalbumins, which to-day are practically discarded in
the case of bacterial poisons. It is probable that snake poisons
also are not proteids in the stricter sense of the word, and if
that is so the attempts that have been made to investigate the
proteids combined with them are, in the main, only interesting
from the historical point of view.

WEIR MITCHELL found albumins in the poison of Crotalus durissus
(rattlesnake), while WOLFENDEN l isolated various proteids (globulins,
albumin, and albumoses), but no peptone from the venoms of the cobra and
Daboia. KANTHACK 2 regarded the poison as a proto-albumose. MARTIN
and SMITH 3 found a non-poisonous albumin, and two poisonous albumoses,
a hetero- and protalbumose, but no peptone in the poison of Pseudechis
porphyriacus and Hoplocephalus curtus.

As regards the constitution of the toxines themselves nothing
is known.

Properties of the Toxine. — Snake poison shows all the proper-
ties characteristic of impure toxines as regards precipitability,
&c. Cobra poison, however, according to CALMETTE, forms an
exception in not being carried down by freshly precipitated cal-
cium phosphate, which is otherwise a general characteristic of all
these colloids. It is also not precipitated by magnesium sul-
phate, and thus contains no globulin.

It dialyses slowly, but appreciably. Viper venom is weakened
by passage through a porcelain filter (PmsALix 4). It is much
less affected by heat than other haptines. Cobra poison can
resist a temperature of 90° C. for an hour, and of 38° C. for a
day ; it is only slightly injured by being kept at 97° C. for half
an hour, but its activity is completely destroyed after exposure
for the same length of time at 98° C. On the other hand, the
purified poison (CALMETTE, 1890) is very susceptible to a tem-
perature of 80° C., a solution in distilled water being more
affected than in that containing salt or glycerin. This is also

1 Wolfenden, "The Venom of the Indian Cobra," Journ. of PhysioL,
vii., 327, 1886 ; id., "The Venom of the Indian Viper (Daboia)," ibid.

2 Kanthack, " The Nature of Cobra Poison," Journ. of PhysioL, xiii., 272,
1893.

3 Martin and Smith, " The Venom of the Australian Black Snake," Proc.
Roy. Soc. New South Wales, 1892, 240; Malys Jb., 1894, 404.

4 Phisalix, Soc. BioL, xlviii., 233, 656, 1896.

SNAKE TOXINES. 213

characteristic of all haptines. It is an important point that the
poison of the blood of the cobra is much more sensitive to heat
than the saliva poison, since it is rendered inactive even after
ten minutes at 68° C. (CALMETTE and DELEARDE l).

Faradic currents have no influence upon it, but continuous
currents have a destructive effect in a solution containing sodium
chloride, owing to the electrolysis and production of chlorine.
Viper venom, however, is said to be injured by currents of high
intensity (PmsALix).

The toxine appears to resist the action of dilute phenol, mer-
curic chloride (1 : 1000), copper sulphate, iodine, potassium
iodide, alcohol, ether, chloroform, and essential oils. Ammonia,
even in large doses, does not injure it until after a long time
(KANTHACK). This highly-valued remedy has thus no action, at
all events, upon the poison itself. The poison of Vipera aspis
remained active for twenty years in a specimen of the snake pre-
served in spirit (MAisoNNEUVE2).

A 1 per cent, solution of potassium permanganate, however,
had a destructive effect upon the venom, and almost invariably
saved the animal when injected into the same place immediately
after the poisoning ; but even after the lapse of a short time the
injection had no effect, as was also the case when the permangan-
ate was introduced at another place, even into the veins, or in
the immediate vicinity of the point of inoculation. Calcium
chloride, too, has an injurious influence on the poison (PmsALix
and BERTRAND3). Gold chloride has a still more pronounced
action, but platinum chloride has no effect.

A 1 per cent, solution of gold chloride destroys the activity of the poison
even when present in a very slight proportion. It also affords protection
when introduced at other places, even against fairly large doses, and also
for a short time after the poisoning. CALMETTE proposed to use this pro-
perty of gold chloride for therapeutic purposes, but these results have been
superseded by his own discoveries of an active immunisation and serum
therapy.

A series of very interesting experiments showed that snake
poison was greatly influenced by certain substances that were
quite inert in themselves, to which PmsALix,4 in particular, has

1 Calmette and Dele"arde, " Sur les toxines non microbiennes," Ann.
Past., x., 675, 1896.

2 Maisonneuve, " Longue conservation de la virulence du venin des
Serpents," Comptes Rend., cxxiii., 513, 1896.

3 Phisalix and Bertrand, Soc. BioL, xlvii., 443, 1895.

4 Phisalix, " La tyrosine vaccine chimique du venin du vipere," Comptes
Rend., cxxvi., 431; id., "Les sues de champignons contre le venin du
vipere," ibid., cxxvii., 1036, 1898.

214 TOXINES AND ANTITOXINES.

called attention in several publications. Tyrosin and cholesterin
were found to possess this property.

Extracts of fungi with chloroform water are also stated to have
a protective influence, and also when injected beforehand to
produce immunity, especially against viper poison, the protection
beginning twenty-four hours after the injection, and lasting until
the twenty-fifth day.

The property possessed by these substances of rendering
poisonous haptines inactive is due to a combination with the
active principle, and has also been observed in the case of many
other toxines, such as tetanus and botulotoxine.

Action of Extracts of Organs and Secretions. — Starting from
the fact of the resistance offered by snakes, including harmless
species, to the poison, search was made for antidotes to snake
toxines in the extracts of different organs.

The bile, in particular, is in vitro an effective antitoxine.
Snake poisons in general resemble bacterial poisons in being
attacked by the bile. Whether the latter has only a simple
destructive action — as, for example, in the case of diphtheria
toxine — or whether it contains a definite antitoxine is not yet
known with certainty ; we shall return to this point presently.
We must, however, assume that the action is purely chemical,
for, according to CALMETTE,1 the sodium salt of glycocholic acid
has the same effect, so that there is good reason for attributing
the action of the bile as a whole to that substance. The bile
retains this property even after being heated to 100° C., but
loses it at 120°C.

Having regard to the production of side-chain immunity in
the case of tetanus, MYERS 2 endeavoured to detect an antitoxic
function in the organs of the body, but only found it in the
extract of the suprarenal bodies. But even in that case there
was only an increase in the resistance in vivo, and not any
specific antitoxic activity. CALMETTE, too, found that no com-
bination took place between the toxine and the nerve substance
or extract of the liver.

FLEXNER and NOGUCHI 3 have tested the neutralisation power
of various organs upon three times the lethal dose of the poison
of the copperhead snake, which killed a guinea-pig in forty-five

1 Calmette, "Sur le me"canisme de I'lmmunisation contre les venins,"
Ann. Past., xii., 343, 1898.

2 Myers, "Cobra Poison in relation to Wassermann's New Theory of
Immunity," Lancet, 1898, ii., 23.

3 Flexner and Noguchi, "Snake Venom in relation to Haemolysis, &c.,"
J. of Exper. Med., vi., 277, 1902 (reprint).

SNAKE TOXINES. 215

minutes in a control experiment. The only organ that had an
energetic protective action was the brain. It retarded the death
of one animal for nineteen hours, while another survived twice
the amount of a lethal dose which killed an animal in five
hours in the control experiment. Extracts prepared from other
organs did no more than retard the action to some extent. The
hsemolysine did not enter into combination in the slightest
degree.

Mode of Action of Snake Venoms. — Snake venoms, as we shall
subsequently show more fully, contain, in addition to the two
agents that act specifically upon the corpuscles of the blood, two
poisonous constituents, neurotoxine and hwmorrhagine. Since
the latter component manifests its activity chiefly in crotalus
venom, and is almost entirely absent in the case of cobra venom,
the 'following description deals primarily with the neurotoxine of
the cobra and other snakes.

This poison is extraordinarily virulent. One drop of CAL-
METTE'S first glycerin extract killed rats and pigeons in less than
an hour, and hens and rabbits in a somewhat longer time.

MARTIN l found the lethal dose of the poison of Hoplouephalus
curtus (the tiger snake) for rabbits to be 0'03 mgrm. per kilo.
This poison is stated to be the most active.

VALENTIN found the lethal dose of the poison of Vipera aspis
for the frog to be 0'5 mgrm.

FLEXNEK and NOGUCHI (loc. cit.) found that guinea-pigs were
killed by a dose of 0*3 mgrm. of the venom of the copper-head
snake (Ancistrodon contortrix).

CALMETTE2 stated that in the case of a cobra that had not
taken any food for eight months the virulence of the venom had
considerably increased. While 0'7 mgrm. of the dry poison was
originally required to kill a rabbit of 1,700 grms., 0-25 mgrm.
was sufficient after two months, and 0*1 mgrm. after the death of
the snake (for a rabbit of 2,000 grms.). Similar results were
obtained in the case of another cobra during three months.

A comparative determination of the toxicity of different snake
poisons gave the following values : —

1 Martin and Cherry, " The Nature of the Antagonism between Toxines
and Antitoxines," Proc. Roy. Soc., Ixiii., 420, 1898 ; Martin, "Relation of
the Toxine and Antitoxine of Snake Venom," ibid., Ixiv., 88, 1899.

2 Calmette, " Contrib. a 1'etude des venins," Ann. Past., ix., 225, 1895.

216

TOXINES AND ANTITOXINES.

Lethal Dose for

Lethal Dose for

Rabbits of

Guinea-pigs of

1,600 to 2,000 grms. in
3 to 4 hours.

450 to 550 grms. in
3 to 4 hours.

Naja tripudians (1 to 3),

0-3 to 0-6 mgrm.

0'05 mgrm.

Naja liaje (4 to 6),

0-3 „ 0-7 „

0-07 „

Cerastes (horned viper)

1'5 ,, 2'0 mgrms.

o-i

Crotalus,

3'5 ,,

0-3 „

Trigonocephalus, .
Hoplocephalus,

2-5 „
2-5 „

0-2 „

Acanthophis,

i-o „

0-08 mgrm.

ELLIOT, SILLAR, and CARMiCHAEL1 found the lethal dose per
kilo, of body weight of the venom of Bungarus coeruleus to be
as follows: — Frogs, O5 mgrm.; rats, 1-0 mgrm.; and rabbits,
0'08 mgrm.

According to ERASER and ELLIOT 2 the lethal dose per kilo, of
body weight of the venom of the sea-snake (Enhydrina) is 0'09
mgrm. for rats, O06 mgrm. for rabbits, and 0'2 mgrm. for cats.

Guinea-pigs are thus twice as susceptible as rabbits. The dog
is still less susceptible. The pig, hedgehog, and the mongoose
(Herpestes)) a small carnivorous mammal, are almost refrac-
tory. It required not less than 8 mgrms. of cobra poison to
kill a mongoose. The hedgehog is but little affected by the
bite of, at all events, the viper. According to PHISALIX and
BERTRAND1 forty times the lethal dose for a guinea-pig is
required. The blood of the hedgehog itself is then poisonous,
but this toxicity disappears on heating.

Snakes themselves, whether poisonous or harmless, are almost
immune, though not absolutely so, as, for example, the ring-
adder, for which the lethal dose is 0*03 grm. (ERASER,3 PHISALIX

and BERTRAND4).

Fishes, lizards, and worms are also not completely immune.

Most snake poisons also produce, like other toxines, severe
local effects, such as violent inflammation, oedema, haemorrhage,
and even necroses.

These local inflammatory effects, however, do not appear to be

1 Elliot, Sillar, and Carmichael, "Action of the Venom of the Bungarus
cc&ndeus," Lancet, 1904, ii., 142.

2 Fraser and Elliot, Lancet, 1904, 141.

3 Fraser, "Immunity against Snake Poison," Brit. Med. Journ.t 1895, i.,
1309.

4 Phisalix and Bertrand, "Glandes venimeuses chez les couleuvres,"
Soc. Biol, xlvi., 8, 1894; xlvii., 639, 1895.

SNAKE TOXINES. 217

an integral part of the general action of snake toxine, which
is quite analogous to what has been observed in the case of
other toxines. Thus, according to CALMETTE (1895), the local
action is greatly weakened by heating the poison to 80° C.,
while the general toxic power is unaffected. KAUFMANN1
states that chromic acid and permanganate have an exactly
similar effect.

The local effects also vary very considerably in intensity
with snake poisons of different origin. They are only pro-
duced to a slight extent by cobra venom, whereas they are
very pronounced in the case of crotalus venom. According to
the conclusions arrived at by WEIR MITCHELL and REICHERT,
and confirmed by later researches (vide infra), these pyrogenic
substances must be sharply differentiated from the true neuro-
toxic principle. But their action is undoubtedly connected with
that of the second main constituent, the hcemorrhagine (vide
infra).

The absorption of the venom is extremely rapid. A rat inocu-
lated at the tip of the tail cannot be saved after the lapse of a
minute by amputation (CALMETTE), and dies five minutes later
than the animal used in the control experiment. The poisoning
is also exceedingly rapid in the case of man.

The bitten limb swells up, and this is followed by contraction
of the mouth, clenching of the teeth, swooning, and death in the
deepest coma.

The mortality fluctuates between 25 and 45 per cent. It largely
depends upon the amount of poison introduced. If the snake
has bitten shortly beforehand, or if the clothes have afforded
some protection, the bite is relatively free from danger; but
it is extremely dangerous when it is in a spot that contains
numerous vessels. Injection into the veins is almost invariably
fatal.

The general symptoms begin with weakness, vomiting, short-
ness of breath, and ptosis. There is loss of faradic excitability
of the muscles, and the immediate cause of death is the stoppage
of the respiration. Frogs, however, which are able to survive
the loss of pulmonary respiration for a longer period, live for
some time (up to thirty hours).

The action of the salivary poison of the vipers (Vipera Redii,
&c.) is very similar to that of the cobra. A. Mosso2 found that
the intravenous injection of 0'0077 grm. per kilo, into a dog

1 Kaufmann, " Sur le venin de la vipfcre," Soc. BioL, xlvi., 113, 1894.

2 A. Mosso, "Die giftige Wirkung des Serum des Mureniden," Arch. f.
exper. Path., xxv., Ill, 1888,

218 TOXINES AND ANTITOXINES.

produced accelerated respiration, followed in a short time (fifteen
minutes) by paralysis of the respiratory centre. There is dimin-
ution in the action of the heart, which continues to beat after
breathing has stopped. By means of artificial respiration life
can be prolonged for about two hours • spontaneous breathing
even begins again ; but eventually it stops once more, and the
animal dies quietly after slight convulsions.

According to PHISALIX and BERTRAND,* 0-3 mgrm. of the
venom of Viper a aspis will kill a guinea-pig, the symptoms
being hypothermia, dilatation of the blood-vessels, and areas of
haemorrhage.

VALENTIN observed a diminution in the amount of oxygen
absorbed.

The heart is not directly affected. As far back as 1873 it
was proved by PANCERI and GASCO2 that the isolated heart of
an axolotl continued beating as before in their preparation of
the venom of Naja eyiziana. This venom resembles cobra poison
in its action, but is weaker.

According to ELLIOT, 3 cobra venom in very dilute solutions
(1:10 millions) has a stimulative action upon the isolated heart
of the frog, while a solution of 1 : 500,000 causes paralysis. In
the case of mammals the heart is only brought to a standstill
by very large doses, and the poison acts primarily upon the
respiratory centre.

Enhydrina venom does not act upon the isolated heart and
the smallest vessels, but upon the respiratory centre and the
peripheral nerves.

ROGERS 4 found that the venom of enhydrina produced para-
lysis of the respiratory centre and of the motor nerves. It is
thus analogous to cobra venom.

The pressure of the blood does not change during artificial
respiration. Apart from this, the initial increase in the pres-
sure is, of course, followed by a decrease, as was shown by
ALBERTONi.5 KAUFMANN,6 too, observed a diminution in the
case of the venom of Pelias berus.

1 Phisalix and Bertrand, "Toxicite du sang de la vipere," Comptes
Rend., cxvii., 1099, 1893.

2 Panceri and Gasco, " Agli effetti del veleno della Naja egiziana," Atti
Acad. Heale Napoli, 1873, 73, quoted by Mosso, loc. cit.

3 Elliot, "Action of Cobra Poison," Proc. Roy. Soc., Ixxiii., 183, 1904.

4 Rogers, "On the Physiological Action of the Poison of the Hydro-
phidse," Proc. Roy. Soc., Ixxii., 305, 1903.

5 Albertoni, "SulFazione del veleno della vipera," Lo Sperimentale,
1879, quoted by Mosso.

8 Kaufmann, Soc. BioL, xlviii., 860, 1896. -

SNAKE TOXINES. 219

VALENTIN1 found that there was a manifest want of excita-
bility in the muscles and nerves of a frog after five hours.
The central nervous system was deprived of its excitability
sooner than the sciatic nerve endings.

Thus, general poisoning by snake toxines is due primarily
to their action upon the central nervous system, and, above
all, upon the motor centres of the medulla. On the other
hand, the peripheral nerves are not affected, at all events in
the case of the frog (CALMETTE). In this action the neurotoxic
component predominates. We shall see presently that this is
not the case with all snake poisons, and notably crotalus venom,
in which the chief effects are produced by the component pos-
sessing hsemorrhagic powers. The hsemolytic component in that
venom need not be taken into account here, and will be dealt
with separately.

The fact that neurotoxine is no more active on intercerebral
injection than when otherwise introduced shows that it is a
specific poison for the central nervous system, and does not
combine with cells elsewhere, being monotropic, to use Ehrlich's
terminology.2

Cobra poison which was free from hsemolysine and heemor-
rhagine was injected into the cerebellum by FLEXNER and
NOGUCHI (loc. cit.), and it was found that the lethal dose was
not less than that required in subcutaneous injection.

The behaviour of crotalus poison is quite different in this respect, for
the lethal dose was twenty times less on intercerebral injection. This
venom contains but little neurotoxine, and, in the main, it enters into
combination in another way. The great activity of crotalus poison on
intercerebral injection is, in fact, due solely to its hsemorrhagic function,
and therefore disappears on heating the solution to 75° C., at which tem-
perature the haemorrhagine is destroyed. An intermediate position is
occupied by the venoms of the mocassin snake and of the ancistrodon,
since they contain a large proportion of both components.

Fresh cobra poison has a very violent action, resembling that
of abrine, upon the conjunctiva. The poison can be deprived
of this property, however, by being heated to 90° C., which does
not materially affect its toxicity.

No absorption appears to take place thence, such as has been
observed in the case of ricine. The same result was obtained by
VALENTIN in his experiments with viper venom.

1 Valentin, "Einige Beobachtungen iiber die Wirkungen des Vipern-
giftes," Z.f. Biol, xiii., 80, 1877.

2 Ehrlich, "Ueb. d. Bezieh. von chemischer Konstitution, Verteilg. u.
pharm. Wirkg.," Festschr.f. Leyden, 1902 (reprint).

220 TOXINES AND ANTITOXINES.

The differences observed in the action of different snake
toxines can for the most part be attributed to the variations
in the proportions of the separate components, sometimes the
neurotoxine and sometimes the hsemorrhagine predominating.
Occasionally, too, the hsemolytic principle doubtless has some
toxigenic effect.

The venoms of the crotalus, trigonocephalus, and cerastes are
distinguished from cobra poison by their much greater activity,
especially as regards the local effects (oedema, necroses, &c.).

Moreover, while, according to CALMETTE, cobra poison can be
entirely deprived of its local irritant action by heat, the toxicity
of crotalus venom is stated by M TAR LAND l to be almost com-
pletely destroyed by the same treatment. Owing to the terrible
injuries produced, M TAR LAND was absolutely unable to produce
immunity by means of subcutaneous injection, and only suc-
ceeded in his purpose by the use of intravenous injections.

As was found by FLEXNER and NOGUCHI, the characteristic
poisonous constituent in the venom of crotalidce is exclusively the
hcemorrhagine, while the neurotoxic component, which predo-
minates in cobra venom, has here only a very slight share in
the effects. The venoms of the mocassin snake and copper-head
snake (Ancistrodon) contain both components.

This explains why the venom of the crotalus and of other
allied species of snakes (Pseudechis, &c.) produce areas of haemor-
rhages that are not caused by cobra venom ; this characteristic
was closely studied by WEIR MITCHELL and REICHERT. The
poisonous constituent that produces the haemorrhage is destroyed
at 75° C., while the venom simultaneously loses part of its
toxicity, so that not less than ten to twenty times the original
lethal dose is required to cause death ; the symptoms resemble
those produced by cobra venom, and must therefore be attributed
to the neurotoxine.

It would be conceivable that the general toxicity in this case
might be due to the hsemolysine.

But the hsemolysine can be eliminated, as we shall presently
show, by making it combine with susceptible erythrocytes,
without destroying the general toxic power.

Hence, it follows that the poison that causes the haemorrhage
is not identical with the true hsemolysine any more than it is
with the neurotoxine, but that a third individual poison is
present, to which FLEXNER and NOGUCHI have given the name
hcemorrhagine.

1 M'Farland, " Immunisation of Animals to Rattlesnake Venom," abstr.,
Centralbl.f. Eakt., xxix., 496, 1901.

SNAKE TOXINES. 221

This poison is also present in cobra venom though its amount
is ten times less than in the venom of the mocassin snake and
one hundred times less than in that of the rattlesnake.

Owing to the great variation in the proportions of the three
components, neurotoxine, hsemolysine, and hsemorrhagine, the
ratio between the lethal dose and the dose of hsemorrhagine, just
capable of detection, is also very variable. Thus, in the case
of cobra venom the lethal dose (O'l mgrm.) corresponds to one
hsemorrhagic dose, that of the venom of the mocassin snake
(0'2 mgrm.) to 20, that of copper-head venom to 60, and that
of rattlesnake venom (1*0 mgrm.) to 1,000 doses.

The histological changes produced by hsemorrhagine in the
blood-vessels have been more closely studied by FLEXNER and
NOGUCHI. The effect is not one of diapedesis but of rents in the
walls of the blood-vessels, in which perforations are then formed.
Stases occur in the vessels and also giant cells which block the
small vessels. The red and white blood -corpuscles equally
escape.

They attribute this perforation of the walls of the vessels to
a cytolysine having a specific action upon the endothelium of
the walls.

On the other hand, WALL (loc. cit.) observed very marked
differences between the venoms of the colubrine cobra and the
viperine Daboia Ensseli, which cannot be accounted for solely by
the difference in the proportions of the separate components.

Daboia venom very rapidly produces violent convulsions, in the course
of which death frequently ensues, and these are followed by paralysis
which does not, however, as in the case of cobra venom, specially affect
the breathing apparatus. Nor is the respiration in general acted upon so
rapidly by daboia venom. It invariably produces mydriasis, but there is
none of the salivation that is characteristic of cobra venom. Albuminuria
is invariably produced by daboia venom, but never by cobra venom. The
former is a very powerful blood poison. Hence, those infected with the
venom are still in great danger even after surviving the first stage of
convulsions and paralysis. Whereas in the case of cobra poisoning the
question of life or death is decided within a few hours, those bitten by
the daboia may die as late as the end of the second week.

In fact, LAMB l was able to show that the haemolysine of the
daboia possesses an absolutely different amboceptor from that of
cobra venom (vide infra).

Toxoids of Snake Toxine. — The existence in snake toxine of
non-poisonous or only slightly poisonous toxoids with an im-
munising power has not been definitely determined, but is
probable.

1 Quoted by Kyes, Berl. klin. Woch., 1903, No. 43.

222 TOXINES AND ANTITOXINES.

PHISALIX and BERTRAND l found that the poisonous serum of
the common viper or adder lost its toxic, but not its immunising
power when heated for fifteen minutes at 58° C., and that the
venom of Vipera aspis behaved in a similar way when heated for
some minutes at 75° to 90° C. The same investigators found
that poison weakened by high-tension currents (vide sit/pro) could
still produce immunity.

By the aid of more accurate quantitative experiments on the
lines devised by EHRLICH, MYERS 2 has detected toxoids in cobra
hsemolysine.

FLEXNER and NOGUCHI (loc. cit.) observed an unmistakable
formation of toxoids in cobra venom that had been allowed to
stand for three weeks. The lethal dose rose from 0*1 to 0*4
mgrm., while there was no appreciable decrease in the dose of
antitoxine required for neutralisation, four lethal doses being
used as the standard poison. Hence, protoxoids are formed.
The same process occurred still more rapidly in an incubating
oven, the lethal dose increasing tenfold in nineteen days,
although, in addition to the formation of toxoids, there was
also a partial decomposition. On the other hand, pepsin and
papain completely destroyed the poison without any formation
of toxoids.

The Hsemolysine of Snake Poisons.— The analogy between
snake poisons and vegetable toxines, and in particular the poison
of eels' blood, also extends to the activity in vitro of their
hsemolytic function.

Many snake poisons also act hsemolytically in vivo, for it
was observed long ago by FONTANA that the intravenous in-
jection of vipers' venom into rabbits produced coagulations,
&c.; while, on the other hand, the blood of animals that have
died from the poison becomes incoagulable, as was recorded
by FAYRER and LAUDER-BRUNTON, 3 ALBERTONI (Joe. cit.), and
others, and confirmed by Mosso (he. cit.), as regards the venom
of the viper.

In the case of the latter there is also a formation of methsemo-
globin as a secondary product due to the action of an oxydase ;
this does not occur in the case of cobra hsemolysine (PHISALIX 4).

and Bertrand, "Att&mation du venin de vipere par la
chaleur," Comptes Rend., cxviii., 288, 1894.

2 Myers, "The Interaction of Toxine and Antitoxine, " Journ. of PathoL,
vi., 415, 1900.

3 Fayrer and Lauder-Brunton, " On the Nature of the Poison of Naja
tripiidians, &c.," Proc. Roy. Soc., xxi., 371, 1873; xxii., 68, 1874.

4 Phisalix, "Action du venin de vipere," Soc. Biol., liv., No. 27, 1902.

SNAKE TOXINES. 223

The action of snake venoms upon the blood was subsequently
more closely studied by FLEXNER and NoGUCHi.1

For this purpose they used the venoms of Naja tripudians,
Crotalus adamanteus (rattlesnake), Ancistrodon piscivor.us (mo-
cassin snake), and Ancistrodon contortrix, all of which only
showed slight differences.

The action of these was tried upon the blood of the dog,
rabbit, guinea-pig, sheep, ox, pig, necturus, and frog. The blood-
corpuscles were washed before the experiment, and then only
simple agglutination without haemolysis took place. The blood
of the rabbit was the most susceptible to this influence, and then
that of the guinea-pig, dog, sheep, pig, and ox.

The amount of haemolysis to be observed in defibrinated
blood stood in no constant relationship to the agglutination of
the washed blood-corpuscles. The separate action of the
haemolysis and agglutination could be observed at 0° C., since
the latter was not affected by the temperature.

Cobra venom was the strongest hsemolytic agent, and rattle-
snake venom the weakest. Dogs' blood was the most susceptible,
and ox blood the least, leaving out of the question frogs' blood,
which was almost completely refractory.

The haemolysines are very resistant to the influence of heat.
They were not affected at all by temperatures of 70° to 80° C.,
and were only slightly injured even after fifteen minutes at
100° C. In this respect they resemble the heat-resisting bac-
terial lysines ; the agglutinines, however, are destroyed in thirty
minutes at 75° to 80° C. The haemolysine can also withstand a
temperature of 100° C. for thirty minutes. On the other hand,
the lysine is destroyed by the same chemical agents as the toxic
components (KYES and SACHS, vide infra).

The hsemolytic principle is absolutely distinct from the true
nerve poison. A solution of the poison that has been freed from
toxine by treatment with brain substance still retains all its lytic
properties, and is therefore still poisonous. But if it is then
subjected to further treatment with blood-corpuscles, it is, in the
case of cobra venom, which contains hardly any haemorrhagine,
rendered almost completely innocuous. In immunisation, how-
ever, both an ti toxine and antilysine are formed, so that anti-
snake venom serum also prevents haemolysis.

Haemolysis occurs only in the presence of fresh serum, and this
has been shown by PmsALix2 to be also the case with the

1 Flexner and Noguchi, " Snake Venom in Relation to Haemolysis,"
Journ. ofExper. Med., vi., 277, 1902 (reprint).

2Phisalix, "Action du venin de vipere, &c.," Soc. BioL, liv., No. 27, 1902.

224 TOXINES AND ANTITOXINES.

haemolysine of vipers' venom. The serum contains a complement,
and the snake venom a (heat-resisting) amboceptor. Snake
haemolysine is thus not a simple lysine, like ricine, staphylo-
toxine, &c., but a haptine of the second class.

It contains a series of different amboceptors, which, when
tested by Ehrlich's method, combine with different blood-
corpuscles, although the poison is never entirely spent in the
process. These amboceptors, again, vary in their affinity for
different normal complements, so that all kinds of combinations
are produced, some of which are fully active, while others have
only slight activity, or none at all.

If the blood be first agglutinated by means of ricine, snake venom has
still a haemolytic action upon it, although the colourless stroma remains
agglutinated.

The poison has also the property of preventing the bactericidal function
of normal sera when added in the proportion of -£$ mgrm to 1 c.c. of the
serum. Only in the case of necturus serum is this action inconstant, for it
depends upon a fixation of the complements, which in that serum do not
invariably enter into combination.

The study of snake hsemolysine was continued by FLEXNER
and NoGUCHi1 in a later investigation. Fresh snake venom
itself was found to contain no complement, and thus to cause
only agglutination and never haemolysis in washed blood-
corpuscles.

Yet snake sera themselves may, undoubtedly, sometimes contain comple-
ments capable of entering into combination. Snake venoms contain ambo-
ceptors of different kinds, with sometimes more affinity for the complement
of their own sera, and sometimes more for that of foreign sera. They are
related to, but not identical with, the amboceptors of snake sera, the latter
being inter alia invariably " iso-complementophile."

Thus, while other poisons only produce haemolysis with the
aid of the serum complements, partial haemolysis is invariably
caused by cobra venom, even after ever so thorough a washing
of the blood-corpuscles. This circumstance, coupled with the
fact that the action of cobra venom is also promoted by serum
that has been heated, and is thus free from complements, has
led to the assumption that endo-complements for cobra venom
are present in the blood-corpuscles themselves. This conclusion
was also arrived at, about the same time, by KYES2 and KYES

1 Flexner and Noguchi, "The Constitution of Snake Venom and Snake
Sera," Univ. of Pennsylv. Med. Bull., 1902, [Nor.] (reprint).

'2Kyes, "Ueb. d. Wirkungsweise des Cobragiftes," Berl klin. Woch.,
1902, Nos. 38, 39 (reprint).

SNAKE TOXINES. 225

and SACHS,1 in Erhlich's Institute, and new and extremely
interesting discoveries were also made with regard to the
relationship between snake venom and the chemical substances
in the corpuscles of the blood. In the first place, it was found
by KYES that there were two sorts of blood-corpuscles — viz.,
those dissolved by cobra venom (such as, e.g., those of the
guinea-pig, dog, man, rabbit, and horse), and others that were
only dissolved with the aid of a complement — e.g., those of the
ox, sheep, and goat. KYES next found suitable complements
for these, and was thus able to confirm FLEXNER'S conclusion
as to the complex structure of cobra lysine.

In the case of those that were soluble by themselves, KYES
was able to disprove the assumption of a simple lysine of the
type of ricine by the fact that haemolysis occurred with dilute,
but not with concentrated, solutions of the poison, which would
obviously be out of the question in the case of simple poisons.
Such a decrease in the action of poisons added in excess is only
explicable on the assumption of a diversion of the complement
by the excess of amboceptors, as was first demonstrated by NEISSER

and WECHSBERG.2

It was found that the blood-corpuscles themselves contained a
complement, which could be fixed and diverted by the excess
of amboceptors. This endo-complement passed into solution on
treatment of the blood-corpuscles with water, and then the
originally insoluble blood-corpuscles also became susceptible to
the action of the cobra poison.

This endo-complement is destroyed by exposure to a tem-
perature of 62° C. for thirty minutes. Sometimes, too, it can
be washed almost completely out of the blood-corpuscles by
means of physiological salt solution. The discovery of such
complements in red blood -corpuscles is very interesting, as
bearing upon the views of the French school that the leuco-
cytes are invariably the source of the complements.

KYES also endeavoured to explain why heated serum is still
able to bring about the action of the poison, in which case
there can be no question of complements ; even boiling the
serum for an hour does not injure this power. In his opinion,
the substance possessing this stimulative power is lecithin, which
also acts as a "complement" for cobra venom when dissolved
in methyl alcohol. The two enter into so firm a combination

1 Kyes and Sachs, "Zur Kenntn. d. Cobragift aktivierenden Subst.,"
Berl. klin. Woch., 1903, Nos. 2-4 (reprint).

2 Neisser and Wechsberg, "Ueb. d. Wirkungsart baktericider Sera,"
Munch, med. Woch., 1901, No. 18 (reprint).

15

226 TOXINES AND ANTITOXINES.

that on the addition of ether practically no lecithin is dissolved.
This compound has an intense haemolytic action, even at 0° C.
The lecithin of the sera is more or less firmly combined with
albumin, so that it is necessary to heat it for varying periods
in order to obtain active free lecithin. The lecithin, as KYES
and SACHS have fully demonstrated, has no connection with the
true complement of sera, which is destroyed by heat.

They found, for instance, that the true complement could be
destroyed by papai'n and ether ; moreover, active sera containing
the complement had a restrictive action upon lecithin. But it
would seem that the so-called endo-complements are nothing more
than lecithin, whose sensitiveness to heat in the blood-corpuscles
is a deceptive phenomenon due to its being combined with the
haemoglobin. Aqueous extracts of the stroma freed from haemo-
globin do not show this sensitiveness to heat.

Owing to their containing lecithin, bile and heated milk, and also the
similarly constituted cephalin, have this stimulative effect. These sub-
stances have in themselves only a very slight hsemolytic action. It is
probable that the fatty acid group in the lecithin is the ultimate active
hsemolytic factor.

Cholesterin has been found an antidote to this stimulative
effect of lecithin, and it has also a protective action in normal
sera, and, as was mentioned above, has an influence upon the
toxic components of snake venom (PHISALIX). It has a similar
antihaemolytic effect upon saponine (RANSOM J).

On the other hand, cholesterin has no action upon the true
complements of sera that stimulate activity.

The amboceptors of cobra venom combine with lecithin in
accordance with quantitative laws.

The fact mentioned by FLEXNER and NOGUCHI that all the washed
blood-corpuscles are not dissolved appears to be due to the removal of
the lecithin by too liberal washing.

KYES2 even succeeded in isolating these "lecithides" of cobra
amboceptors. The 1 per cent, solution of cobra venom was
shaken for two hours with a solution of the purest lecithin in
chloroform. On now separating the chloroform layer by rapid
" centrifuging," and treating it with ether, the cobra venom
lecithide was precipitated, while the excess of lecithin remained
dissolved in the ether.

1 Ransom, "Saponin und sein Gegengift," Deutsch. med. Woch., 1901,
194.

2 Kyes, " Ueber die Isolierung von Schlangengiftlecithiden, " Berl. klin.
Woch., 1903, Nos. 42, 43.

SNAKE TOXINES. 227

The hsemolytic function of the cobra venom was attached
quantitatively to this lecithide, whereas the neurotoxic function
remained absolutely unaffected by the lecithin. The lecithide
possessed only hsemolytic powers, while the neurotoxic function
was retained exclusively by the residual poison.

The lecithide is insoluble in ether and acetone, but dissolves
in chloroform, alcohol, and toluene, and is readily soluble in
water. Its properties are thus quite distinct from those of its
two components. On standing in aqueous solution it becomes
gradually insoluble without losing its hsemolytic power. It does
not separate out from a hot solution. It does not give the
biuret reaction. It dissolves all blood-corpuscles equally, and in
contradistinction to this function of the poison itself, does so
without a period of incubation. The lecithide is hardly affected
by a temperature of 100° C. Its action, like that of the fresh
venom, is prevented by cholesterin. Closely analogous lecithides
have been obtained from all the other hsemolytic snake poisons
examined, including those of Both/rops, Naja haje, Bungarus,
Trimeresurus, and Crotalus. Thus the same " lecithinophile "
group is invariably present, even though the amboceptors may
differ in other respects.

Leucocidine of Snake Venoms. — Sterile exudations containing
20 to 25 per cent, of lymphocytes were obtained by the intra-
pleural injection of the dead cells of B. megatherium (FLEXNER
and NOGUCHI). Cobra venom in solutions containing 0-002 per
cent, acted upon these, the action of the other venoms being
weaker. The movements of the leucocytes cease first; then
follows the decomposition of the cells, and, finally, that of the
lymphocytes. In the case of washed leucocytes little more than
agglutination occurred. FLEXNER and NOGUCHI concluded from
their experiments on the modes of combination that the agglutin-
ines were identical with, but the lysines different from, the
corresponding agents of the red blood-corpuscles. The leucolysine
was also of complex structure.

Summary. — We have thus in snake venoms four distinct active
principles, the proportions of which show great variations.

1. Hcemagglutinines. — These are destroyed by a 0-2 per cent,
solution of hydrochloric acid in twenty-four hours, and in a short
time by heating them to 75° C.

2. Hcemorrhagine (principally in crotalus venom). — This is
only destroyed after about two days by hydrochloric acid (2 per
cent.) and pepsin-hydrochloric acid, and can resist the temperature
of an incubating oven.

3. Hcemolysine, which is very slowly destroyed by hydrochloric

228 TOXINES AND ANTITOXINES.

acid (up to 3 per cent.), but rapidly destroyed by pepsin-hydro-
chloric acid. Exposure to an incubating temperature destroys it
to the extent of about 80 per cent.

4. Neurotoxine, — This is fairly resistant to the action of
hydrochloric acid (up to 3 per cent.) and to pepsin and papain.
It loses about 90 per cent, of its toxicity by being allowed to
stand for nineteen days.

The hsemagglutinine and hsemolysine attack the blood-cor-
puscles exclusively, while the hsemorrhagine attacks the endo-
thelium of the walls of the vessels, and the neurotoxine the cells
of the central nervous system.

IMMUNISATION AGAINST SNAKE TOXINE.
Snake Antitoxine.

The close relationship between snake toxines and true toxines
is shown, above all, by their power of producing an antitoxine.
The first attempts to produce immunity against snake venom
were those of SEWALL,1 who used crotalus venom in his
experiments.

CALMETTE succeeded in showing that, even after a single
injection of half a lethal dose, the serum of the animal treated
had an unmistakable antitoxic action in vitro. ERASER,2 too,
was able to produce antitoxine to the venoms of the cobra,
crotalus, diemenia (South Australia), and sepedon (Africa). He
obtained preparations capable of resisting as much as fifty times
the lethal dose.

Calmette's method is essentially as follows : — About ^ of the
lethal dose is first introduced, and this is followed every two or
three days by very gradually increasing doses (up to y1^ of the
lethal dose). The same result can be obtained with poisons
chemically weakened by means of gold chloride or calcium
chloride (CALMETTE3).

After four or five weeks the animals can resist twice the lethal
dose. They can then be treated every eight or ten days with
larger doses, until a very high degree of immunity can be pro-

1 Sewall, " Exper. on the Preventive Inoculation of Rattlesnake Venom,"
Journ. of PhysioL, viii., 203, 1887.

2 Fraser, "Immunity against Snake Poison," Brit. Med. Journ., i., 1309,
1895.

3 Calmette, " Proprie'te's du serum des animaux immunises centre le
venin des serpents," Comptes Rend., cxviii., 120, 1004, 1894.

SNAKE ANTITOXINE. 229

duced. CALMETTE rendered a rabbit so immune in one year that
it was able to receive eighty times the lethal dose of cobra venom
(40 mgrms.) without showing any reaction. The serum of this
animal was so rich in antitoxine that 5 drops (about 0'25 c.c.)
neutralised 1 mgrm. of cobra venom. An ass received 0*2 grm.
of cobra venom in three months, and another O16 grm. in two
months. Half a c.c. of the serum then neutralised 1 mgrm. of
the venom.

The injection of 4 c.c. of this serum four hours previously
afforded protection against twice the lethal dose. According to
PHISALIX and BERTRAND,1 however, the antitoxine to viper
poison is not effective until thirty-six to forty-eight hours after
its introduction into the body. When an absolutely fatal amount
of venom is injected, and then, after the lapse of an hour, 4 to
5 c.c. of this serum, the animal usually survives, but an hour and
a half is the maximum time if there is to be any reasonable
certainty of a cure.

This anti-cobra serum also affords protection against the neuro-
toxic components of other snake poisons, and also against scorpion
venom and eels' blood, and to some extent against abrine, but has
no effect upon ricine or upon diphtheria virus, or tetanus poison.

The serum has just the same sort of antitoxic action as the
true antitoxines, but, on the other hand, CHATENAY (quoted by
CALMETTE, loc. cit.) observed hyperleucocytosis after the intro-
duction of the poison into immunised animals, whereas hypo-
leucocytosis could be observed in the case of the control animals.

These observations were confirmed by CALMETTE and DELBARDE. They
found that animal charcoal impregnated with abrine and introduced into
the peritoneal cavity of immunised animals was absorbed to a large extent
by the leucocytes, while there was practically no absorption in the case of
the control animals. They came to the conclusion that the leucocytes
produced and stored up antitoxine.

According to PHISALIX and BERTRAND 2 normal antisera are
produced by the guinea-pig, horse, and hedgehog, as well as by
the rabbit, but not by the hen.

Passive immunity soon disappears, but in the case of active
immunity the higher its degree the longer it persists. Immunity
can be inherited.

Each of the three components of the poison (hsemolysine,
neurotoxine, and hsemorrhagine) produces its specific anti-body

1 Phisalix and Bertrand, " Sur la proprie't6 antitoxique du sang des
animaux vaccines contre le venin de vipere," Comptes Rend., cxviii., 356,
1894.

2 Phisalix and Bertrand, Soc. BioL, xlviii., 396, 1896.

230 TOXINES AND ANTITOXINES.

in the immunisation process. Hence the different antisera differ
very greatly in their action. It was shown by STEPHENS and
MYERS1 that there were antisera to haemolysine, or that the
ordinary antisera also contained antihsemolysine ; yet even in
this case, according to FLEXNER and NOGUCHI, the antisera vary
in their activity, since the haemolysines, again, possess ambo-
ceptors of different kinds, and the antidotes act by means of
anti-amboceptors.

The anti-cobra venom contains an antitoxine not only to cobra
neurotoxine, but also to that of the most widely differing snake
venoms (M'FARLAND2), and also an antitoxine to the hsemolysine,
but is completely lacking in the anti-body to the hwmorrhayine
of crotalus venom ; and since this is the chief active constituent
in crotalus venom (vide supra), Calmette's "antivenine" is
powerless against that poison.

In like manner, anticrotalus serum contains chiefly antihsemo-
lysine and antihsemorrhagine, but not antineurotoxine ; hence,
it neutralises the hsemolytic but not the neurotoxic function —
i.e., the general toxicity — of cobra venom.

FLEXNER and NOGUCHI 3 obtained a serviceable antiserum to
rattlesnake venom on removing the substances that produced
the necroses by treating the toxine with dilute hydrochloric acid
or iodine trichloride. The serum had no action upon the venoms
of the cobra, daboia, and mocassin snake.

In the case of those venoms which contain both the main
poisonous components, such as those of the mocassin and copper-
head snakes, the antisera naturally contain both specific anti-
bodies.

The action of anti-snake-venom serum, according to all that is
known about it, is apparently quite analogous to that of the
other antitoxines — i.e., it combines with and neutralises the
poison without destroying it. A very interesting proof in
support of this view was furnished by Calmette's experiment
(1895), of which an outline was given in the General Part. He
found that on heating a physiologically neutral mixture of
snake toxine and antitoxine to 68° C. the antitoxine could be
eliminated, so that the original toxic action again appeared.
After the heating the mixture behaved exactly like the toxine,
either if it had contained antiserum or normal serum. In like

1 Stephens and Myers, Proc. Path. Soc., Lancet, 1898, i., 644.

2 M 'Farland, "Some Investigations upon Antivenine," Journ. Amer.
Med. Assoc., Dec. 1901 ; abstract in Gentralbl f. Bate., xxxi., 792.

3Flexner and Noguchi, "Upon the Production and Properties of Anti-
crotalus Serum," J. of Med. Research, xi., 363, 1904.

SNAKE ANTITOXINE. 231

manner, the toxine could be removed from the mixture by
precipitation with calcium chloride, so that the previously neutral
serum again had a protective action (1896). These results, which
are extremely important from the theoretical point of view, have
been conditionally confirmed by MARTIN and CHERRY,3 who were
able, under certain conditions, to effect a separation of the
poison of Hoplocephalus curtus from the antitoxine by heating
the mixture to 68° C.

Thus, when the components of the mixture have only been
allowed to act upon each other for a short time, or when a
relatively large amount of poison has been used, CALMETTE'S
statement is correct ; the antitoxine is destroyed and the toxic
activity restored by heating the mixture. After fifteen minutes,
however, the combination is so firm that it can no longer be
broken up. Again, it is impossible to effect a separation by
filtration through a gelatin filter under pressure, which otherwise
allows the free toxine but not the antitoxine to pass, since no
part of the mixture passes through the filter.

The conclusion that there is a simple combination is in no way
affected by the fact that, according to MARTIN, the dose that
combines with a definite quantity of snake venom in vitro is
much less than that required for previous immunisation when
injected subcutaneous^. The quantity is considerably greater
and may even amount to a thousand times as much. On the
other hand, only about the same quantity is required if intra-
venous injection be used.

MARTIN attributed this to a much slower diffusion of the
antitoxine than of the toxine on subcutaneous introduction, the
toxine being also very rapidly distributed through the body by
way of the anastomosing lymph tracts.

It is also worthy of note that CALMETTE (1895) found that
tetanus antitoxine and antiabrine had also a certain action upon
snake toxine, so that the serum would thus seem to be not
absolutely specific. On the other hand, the sera of animals that
had been treated with strychnine, curare, and various bacteria
proved as absolutely powerless against the venom as normal
human serum.

Anti-snake-venom serum only becomes inactive when the
temperature reaches 68° C. Calcium chloride and gold chloride
do not affect its protective power. It can also be kept for a
long period without the addition of phenol.

3 Martin and Cherry, " The Nature of Antagonism between Toxines and
Antitoxines," Proc. Roy. Soc., Ixiii. ; Martin, "Relation of the Toxine
and Antitoxine of Snake Venom," ibid., Ixiv., 88, 1899.

232 TOXINES AND ANTITOXINES.

CALMETTE 1 has devised the following method for the valuation
of snake serum : —

The poison is dried and dissolved in distilled water and the
lethal dose for 1 kilo, of rabbit determined. A rabbit weighing
2 kilos, is then treated with increasing doses of the serum under
examination and the protective dose against a single lethal dose
determined. A serum, 1 c.c. of which protects 1 grm. of animal
against the single lethal dose, is taken as unity. Thus, if 1 c.c.
of the serum under examination protects 2 kilos, of animal,
the strength of the serum is two thousand-fold. According to
CALMETTE, the minimum strength must be one thousand-fold ;
sera of, at least, four thousand-fold strength are used in the
tropics.

This true antitoxic power of the immune serum has nothing
to do with the property possessed by ty rosin, cholesterin, &c.,
of rendering the toxine innocuous, for all these substances act
just as they do in the case of tetanus toxine, only combining
with the poison in vitro, and never producing immunity. Their
action is thus quite distinct from that of the true antitoxines.

The question whether or no the serum and bile of venomous
snakes contain antitoxine stands upon a somewhat different
footing. As was found by FRASER,2 not only is the bile of the
cobra particularly active, but also that of the rattlesnake, &c.,
possesses incomparably greater powers of destroying the poison
than the bile of other animals ; and even that of non-venomous
snakes has a greater protective power than theirs. This pro-
tective function is also retained by alcoholic precipitates of these
biles. On the other hand, it must not be lost sight of that, as
FRASER 3 himself found, the anti venom is not definitely specific,
but that it also acts upon bacterial toxines.

At the same time, we frequently find that the bile merely has
a destructive action upon toxines similar to that of other digestive
fluids, so that the question whether or no the bile contains a true
antitoxine has not yet been decided.

Hence, one is hardly justified in attributing the resistance of
snakes to this function alone. It is, indeed, more conceivable
that the natural immunity of these animals is due to a continual
new formation of large quantities of antitoxine, rather than to a

1 Calmette, " Sur le venin des serpents," Ann. Past., xi., 214, 1897.

2Fraser, "The Treatment of Snake Poison with Antivenene," Brit.
Med. Journ., ii., 417, 1895; " Antivenomous Properties of the Bile of
Serpents," ibid., ii., 125, 1897.

3 Eraser, "Antitoxic Qualities of the Bile of Serpents," Brit. Med.
Journ., ii., 595, 1897.

TOAD TOXINE (PHEYNOLYSINE). 233

fixed power of reacting upon any poison absorbed. But surely
this condition must depend, in the main, upon an innate want
of receptors ; for, as we have seen, even the blood of venomous
snakes is poisonous.

In the case of the hedgehog, too, which is to some extent
refractory, the want of susceptibility appears to be principally
due to its possessing few receptors; after the introduction of
the venom its blood also is poisonous. The effect of the venom
upon the hedgehog is similar to that of tetanus toxine upon
the alligator ; for the hedgehog, although but slightly suscep-
tible, yet produces fairly considerable amounts of antitoxine.
It thus appears to possess receptors, but these are, for the most
part, in the organs less vitally important. On the other hand,
CALMETTE (1895) found the serum of the pig and mongoose to
contain extremely little antitoxine, although both animals are
almost completely immune to snake venom.

Doubtless this is also the case with snakes themselves. The
main cause of their insusceptibility may be a congenital want
of receptors, or to the existing receptors being too far apart ;
but at the same time it is, of course, not improbable that they
also produce antitoxine and excrete it with the bile.

Toad Toxine (Phrynolysine).

Certain toads contain in their skin and blood, in addition to the
better-known alkaloidal poisons (bufotaline, &c.) an apparently
true toxine possessing hsemolytic powers.

PHISALIX and BERTRAND1 appear to have been the first to
discover the existence of a second poison, and its hsemolytic
action was recorded by PuGLiESE.2

"Phrynolysine" was then investigated more closely by PRO-

SCHER.3

In his experiments PROSCHER chiefly employed extracts of the
skin of the fire toad (Bombinator igneus), though the skin of the
garden toad (Eufo cinereus) was also used.

Phrynolysine has all the characteristics of toxines, and par-
ticularly their great sensitiveness to external influences. It is
non-dialysable. It becomes inactive fairly rapidly.

It produces haemolysis equally well in a neutral or slightly

1 Phisalix and Bertrand, "Recherch. s. la toxicite du sang du crapaud
commun," Arch. d. Phys., xxv., 517, 1893.

2 Pugliese, Arch. d. Farm., 1898, quoted by Proscher (loc. cit.).

3 Proscher, " Zur Kenntnis des Krotengiftes " Hofm. Beitr., i.. 575.
1901.

234 TOXINES AND ANTITOXINES.

acid solution. Sheep's blood is the most susceptible, and then
comes the blood of the goat, rabbit, dog, ox, hen, and guinea-
pig, while that of the pigeon, frog, and toad is hardly affected.
In the case of sheep's blood, about O3 mgrm. is sufficient to
dissolve a litre completely.

There is as yet no reason for concluding that phrynolysine has
a complex structure.

Normal sera do not contain any anti-body. By immunisation,
however, it is possible to produce an antilysine, which in a dose
of 0'025 c.c. affords protection against a dose of the toxine suf-
ficient to hsemolyse 1 c.c. of a 5 per cent, emulsion of sheep's
blood.

Salamander Poison.

A poisonous substance that formed an antitoxine was dis-
covered, by PmsALix,1 in the skin of the back of the Japanese
salamander (Sieboldia maxima).

The poison is soluble in water and glycerin, and possesses
little stability. It is completely destroyed by twenty minutes'
exposure to a temperature of 60° C., and also by alcohol.

It produces oedema and areas of haemorrhage in the frog, and
in warm-blooded animals necroses also. Paralysis also occurs,
and the excitability of the nerves is gradually lost; death results
from the paralysis of the respiratory system.

The poison is weakened by being heated to 50° C., but still
retains its immunising power. Animals thus treated can then
resist much larger doses, not only of this poison byut also of
viper venom and eel's-blood poison (PmsALix2), whence we may
conclude that it has a certain degree of relationship with these
toxines.

Spider Venom.

Poisonous spiders occupy an important position in the popular
imagination. Very many spiders, and notably the tarantula,
have been accused of possessing toxic properties. Little, how-
ever, was known scientifically about spider poisons until the
appearance of ROBERT'S 3 comprehensive monograph.

iPhisalix, "Act. phys. venin du Salamandre," Soc. BioL, xlix., 723,
1897.

2 Phisalix, "Propr. immunisantes du venin du Salamandre," Soc. Biol.,
xlix., 823, 1897.

3 Robert, Beitr. z. Kenntn. d. Giftspinnen, Stuttgart, 1901 (gives the
older literature from the earliest times); id., "Giebt es fiir den Menschen
gefahrliche Spinnen?" Die Med. Woche, 1902, 154.

SPIDER VENOM. 235

ROBERT was able to show that there was no specific poison in
those species of spiders most accused of containing it, and in
particular the tarantulas. In fact, real poisons were only found
in two genera — viz., Lathrodectes and Aranea diademata (Epeira
diadema, the garden spider).

The Lathrodectes are distributed over the whole globe. The
most important species are those of Italy (L. tredecimguttatus,
Malmignatte) and of South Russia (L. erebus, Karakurte), with
those of New Zealand (L. scelio and Hasseltii, Katipo) and of
South America (L. mactans). As far back as 1765 YALMONT
DE BOMBAKE gave a description of the poisonous properties of
the malmignatte, and since then they have frequently been
studied. Experiments on animals have also been made.

A whole series of reports has been collected by KOBEET with
regard to the Russian spider, and from these it appears that the
Lathrodectes not only inflicts great injury upon cattle, horses,
and camels, but even kills men. The effects are very severe :
violent pains, priapism, sleeplessness, great prostration, cold
sweating, fever, and dyspnoea. The general symptoms closely
resemble those caused by poisoning with bacterial toxines.
No striking alterations can be observed in the vicinity of
the bite. Convalescence is very tedious, but fatal cases are
very rare.

ROBERT has made experiments on his own account with
aqueous extracts of these Crimean karakurtes, some of the
extracts being prepared from the fresh animals, and others from
those that had been partially dried.

These extracts were found to be very poisonous, producing
exactly the same results as those observed in men bitten by the
spider. The intravenous injection of a few mgrms. per kilo, of
organic substance into dogs and cats produced dyspnoea, convul-
sions, and paralysis of the respiratory system and heart, speedily
ending in death. Rabbits, rats, and birds were also susceptible
to the poison, but the hedgehog offered somewhat more resistance.
Frogs and leeches could also be poisoned. There was practically
no difference in the action of extracts obtained from the front
and from the back portion of the spider. New-born spiders
were more venomous than full-grown ones, and even the eggs
were poisonous. The poison had an injurious effect upon the
isolated heart of the frog, even in the proportion of 1 : 100,000.

ROBERT also succeeded in producing immunity by cautiously
inoculating experimental animals, so that karakurte poison
appears to be a true toxine. Boiling renders the poison com-
pletely inactive, and alcohol has the same effect. Introduced

236 TOXINES AND ANTITOXINES.

per os the poison has no effect. It has also a hcemolytic action,
and promotes coagulation.

Hence the lathrodectes contains a true toxine which in many
cases has first an irritant action upon the heart and central
nervous system, but, eventually, invariably causes paralysis.

Arachnolysine.

Of the German spiders Chiracanthium nutrix appears to
contain a poison which has not yet been investigated.

ROBERT found, however, in the common garden spider, Aranea
diademata (Epeira diadema), a toxine completely analogous to
lathrodectes poison ; it, too, was very poisonous, and it was
possible to produce immunity against it. It was somewhat less
active, but more stable, than lathrodectes poison.

The extracts of other German spiders were inert. The poison
of the garden spider was also found by KOBERT to possess heemo-
lytic powers, which were subsequently more closely investigated
by HANS SACHS, 1 with the aid of Ehrlich's method.

Arachnolysine has a very rapid and intense solvent action upon
the blood-corpuscles, but there is great variation in the resistance
offered by the corpuscles of different species of animals. The
blood-corpuscles of the rat and of the rabbit are the most sus-
ceptible, 0'028 mgrm. completely dissolving 0*05 cm. of the
blood. On the other hand, the blood of the guinea-pig, horse,
sheep, and ox is absolutely refractory.

The blood-corpuscles of chickens are quite insusceptible,
according to SACHS,2 in consequence of a total lack of receptors.
It is only when these first blood-corpuscles have gradually dis-
appeared that the lysine has any action, and it is not until after
two to four weeks that the normal susceptibility of hen's blood
is attained.

The toxine is not very sensitive to the action of heat, and can
resist a temperature of 70° C. for forty minutes.

Insusceptible blood-corpuscles do not combine with the poison,
so that the behaviour of this lysine is quite in accord with that
of other haptines. A further point of agreement is that arachno-
lysine enters into combination with the stroma of susceptible
blood-corpuscles.

SACHS, by immunising guinea-pigs and rabbits, succeeded in

1 Sachs, "Zur Kenntnis des Kreuzspinnengiftes," Hofm. Beitr., ii., 125,
1902.

2 Sachs, "Ueber Differenzen der Blutbeschaffenheit in verschied. Lebens-
altern," Centralbl. f. Eakt., xxxiv., 686, 1903 (reprint).

SCORPION VENOM. 237

producing a highly active antitoxic serum, which, when mixed
with the poison, also prevented the haemolysis. Now, inasmuch
as the blood of the guinea-pig is insusceptible, and therefore
contains no receptors, the antiarachnolysine must have been
formed from other receptors. Since, however, it has also an
antilytic action, it is probable that the same conclusions may be
drawn as to its constitution as in the case of ricine (q.v.), to
which it offers many points of resemblance.

Scorpion Venom.

VALENTIN,1 in 1874, investigated the effects of the venom in
the sting of the tail of a Tunisian scorpion (Androctonus
occitanus, Glaus).

Small frogs were usually killed by the sting, but not larger
ones. It caused tetanic convulsions, and also twitchings of the
fibrillse, while the reflex excitability gradually disappeared from
behind forwards.

Very similar results were also observed later by BERT 2 and
others.

According to WILSON 3 scorpion venom causes convulsions and
death from asphyxia in the guinea-pig.

The poison resembles veratrine in acting directly upon the
muscles, while the nerves are not affected. Certain animals that
live in the desert (jerboa, gerbillus) offer great resistance to
scorpion venom, an immunity probably acquired from frequent
stings.

CALMETTE found (1895) a poison in the tail segment of Scorpio
afer, and isolated it by extracting the crushed bodies with water
and drying the extract in a vacuum.

The poison killed mice in doses of 0-05 grm., and guinea-pigs
in doses of 0-5 grm., the symptoms closely resembling those
produced by snake venom.

It also behaves in every other respect, and especially in its
relation towards the antitoxine, so exactly like snake toxine, as
to justify the conclusion that scorpion venom differs no more
from snake venoms than do these from each other, and also that,
like them, it contains the neurotoxine of the snakes in admixture
with small proportions of foreign bodies. This conclusion is also
supported by the fact that KYES (loc. cit.) was able to isolate

1 Valentin, "Ueber d. Giftw. d. nordafrik, Scorpiones," Zeit. fiir Biol.t
xii., 170, 1876.

2 Bert, Soc. Biol., xxxvii., 574, 1885.

3 Wilson, "Action of Scorpion Venom," Journ. of Physiol., 1904, 31,

238 TOXINES AND ANTITOXINES.

from scorpion venom a lecithide possessing an immediate blood-
solvent action, and exactly analogous to that obtained from
cobra venom. The sting of the scorpion is not very dangerous to
man, but only because the amount of venom injected in a single
sting is too small.

Fish Venoms.

An apparently true toxine was isolated by BmoT1 from the
poison glands of the Trachinus draco (greater weever), though
its existence had already been known. He extracted the poison
glands by means of glycerin containing chloroform, and filtered
the neutral solution.

The poison causes convulsions and paralysis in frogs, and death
results from prostration (GRESSIN, BOTTAKD). The heart is also
directly affected (PoHL 2).

It has a very similar energetic action upon guinea-pigs, especially
on intraperitoneal injection (and this was also found to be the
case by PHISALIX with Trachinus vipera), but has less effect upon
rabbits.

A characteristic symptom is the rapid paralysis of the ex-
tremity where the poison was injected. Death results almost
instantaneously after intravenous injection ; but the animal
recovers very rapidly from the effects of smaller doses. The
poison also causes severe local injuries on subcutaneous in-
jection.

The toxine is destroyed when heated for thirty minutes at
100° C., and also by calcium chloride and gold chloride.

The poison also possesses hsemolytic powers. The normal
serum of the horse contains an antihsemolysine against trachi-
nus lysine, just as it does against snake hsemolysine. This is
destroyed at 50° C., but the lysine can resist a temperature of
100° C. for a short time (twenty minutes).

Trachinus poison is quite distinct from snake venom, since
its action is different, and an anti-snake-venom serum has no
restrictive effect either upon its toxic or its lytic function.

Babbits can best be rendered immune by the cautious injec-

1 Briot, "Etudes sur le venin de la vive (Trachimisdraco)," Journ. de
Phys. et PathoL, 1903, 271 (reprint). Briot cites the following works on
poisonous fish : — Gressin, Contrib. a I'etude de I'appareil a venin chez les
poissons du genre vive, Thesis, Paris, 1884 ; Bottard, Les poissons vdmmeux,
Thesis, Paris, 1884; Phisalix, Bull, du Museum d'Histoire Natur., 1899;
Contiere, Les poissons vtnimeux, Thesis, Paris, 1899. See also Robert,
loc. cit.

2 Pohl, Prager Med. Woch., 1893, 31.

THE POISON OF EELS* BLOOD (ICHTHYOTOXINE). 239

tion of a mixture of the poison with the serum of an animal
already immune.

The serum contains a specific antitoxine, which, however, does
not invariably afford protection against the local effects.

Certain other fishes also appear to contain poisons of the nature
of toxines, which have not yet been investigated, such as, e.g.,
the lamprey (Petromyzon), Scorpcena, Pterois, Serranus, Plotosus,
tiynanceia, &c. (KoBERT1).

Further investigation is required to determine whether the
poisoning in these cases is due to true toxines.

In the case of certain other poisonous fish — e.g., the Japanese
species of Tetrodon (fugu) — poisonous protamines appear to play
the chief part. It does not come within the scope of this book
to give further details of fish poisons in general.2

The Poison of Eels' Blood (Ichthyotoxine).

An exceptional position is occupied by the poisonous sub-
stance which occurs in the serum of the eel and certain allied
fish (Murcena, Conger). By its occurrence as a normal product
of animal life it is related, on the one hand, to the snake venoms,
which are also found in the blood of poisonous snakes ; while, on
the other hand, in its hsemolytic function it recalls the agglutin-
ating toxines of the vegetable kingdom ; and yet again shows a
certain relationship, as regards this property, with the hsemo-
lysines of the normal sera of different animals, which, according
to EHRLICH, are also not simple "alexines," but receptors of the
second class provided with amboceptor and complement.

In what group, therefore, ichthyotoxine should eventually be
placed cannot be decided off-hand. It is convenient, however,
for reasons apart from its internal constitution, to describe it
provisionally after the other zootoxines.

A similar poison appears to be produced in the poison glands
of the muraena, but little is known about it.3

The toxicity of eels' blood was discovered by A. Mosso,4 who
also studied its toxic effects. He found that the blood of poisoned
animals lost its power of coagulating, and this was confirmed by

1 Robert, "Ueber Giftfischeu. Fischgifte.Vortrag i. Rostocker Fischerei-
verein," Die Med. Woche, 1902.

2 Further details are also given by Vaughan and Novy, loc. cit., 188,
et seq.

3 Anatomical and toxicological data are given by Robert, loc. cit.

4 A. Mosso, "Die giftige Wirkung des Serums der Mureniden," Arch,
f. Exper. Path., xxv., Ill, 1889.

240 TOXINES AND ANTITOXINES.

DELEZENNE,1 who compared its action with that of propeptones.
U. Mosso 2 also investigated the properties of this poison.

Then came the notable work of KossEL3 and CAMUS and
GLEY,* who discovered the solvent action of eels' blood poison
upon the blood, and showed that immunity could be produced
against it, and that it formed an antitoxine.

Preparation and Properties of the Poison. — CAMUS and GLEY
obtained the poisonous serum by taking the blood by means of
a sterilised pipette from the aorta of the eel, placing it in a
sterilised vessel, and either allowing the serum to separate spon-
taneously or with the aid of centrifugal force. They thus
obtained about 0-6 c.c. from 100 grms. of the fish.

The serum has a slight green colour, frequently of a yellowish
shade. It can be kept for a long time unaltered if protected
from light.

No attempt has yet been made to prepare the toxic principle
of the serum in a pure form.

U. Mosso showed that it possessed exactly the same physical
and chemical properties as all toxines ; that it was destroyed by
heat, acids, alkalies, &c. ; that it could resist drying in vacua ;
and that it was not dialysable. It was insoluble in alcohol of
90 per cent, strength.

Action of Eels' Blood. — The serum of different eels frequently
varies considerably in its toxic effects, the season of the year
having an influence as well as the origin of the fish, as has also
been observed in the case of snake venoms (WEHRMANN 5).

The river eel of the Baltic coast contains, according to SPRING-
FELD,6 a very much weaker toxic serum.

Different animals also vary in their susceptibility when the
poison is introduced in the same way (intravenously). Dogs

1 Delezenne, "Action du serum d'anguille sur la Coagul. du Sang,"
Arch. d. Phys., xxix., 646, 1897.

2 U. Mosso, ' ' Recherches sur le nature du venin, qui se trouve dans le
sang de 1'anguille," Arch. Ital. d. BioL, xii., 229, 1889.

3 Kossel, "Zur Kenntnis der Antitoxin wirkung," Berl. Uin. Woch.,
1898, 7.

4 Camus and Gley, notably " Recherches sur 1'action physiolog. de
scrum d'anguille, "Arch. Internal, de Pharmacodyn. , v., 247, 1898 (reprint);
also "De la toxicite du serum d'anguille pour des animaux des especes
differents," Soc. BioL, 1898, 129; "Immunis. centre Faction globulicide,
&c.," Comptes Rend., cxxvi., 428, 1898; and "Nouvelles recherches sur
1'immunite contre le serum d'anguille," Ann. Past., xiii., 779, 1899.

5 Wehrmann, "Sur les propr. toxiques du sang, £c.," Ann. Past., xi.,
810, 1897.

6 Springfeld, "Ueb. d. gift. Wirkung des Blutserums des Flussaals,"
Dissert. Greifswald, 1889.

THE POISON OF EELS* BLOOD (ICHTHYOTOXINE). 241

appear to be the most susceptible (lethal dose, according to
Mosso, 0-02 c.c. per kilo.), while the hedgehog is almost refrac-
tory. A. Mosso states that it has no action when introduced
into the stomach, though it is poisonous when injected into the
small intestine. On the other hand, a serious case of poisoning
in man after eating eels' blood is reported by PENNAVARIA.1
Subcutaneous injection produces necroses and abscesses. The
course of the poisoning takes two distinct forms, according to the
dose. Rabbits die in convulsions a few minutes after receiving
Ol c.c. per kilo, of a very active serum, while sometimes there is
also a flow of saliva and blood into the urine. Myosis is hardly
ever absent, and sometimes exophthalmus occurs.

The effects are attributed by Mosso to paralyses of the vagus,
preceded by stimulation.

With small doses or weak poisons, however, conditions of
paralysis are produced, together with fine tremors, areas of local
anaesthesia, dyspnoea, flow of saliva, cries, &c., which, after
a great loss in weight, end in death, though only after some
hours, or even days (KOSSEL).

Similar results have been observed with guinea-pigs. When
very large doses have been given death occurs so rapidly that
sometimes even the convulsions fail to appear.

In the first case, therefore, the bulbar symptoms, especially
the rapid paralysis of the respiratory centre, predominate ; in
the second case, the spinal symptoms. The peripheral nerves
of the respiratory centre do not lose their capacity for being
stimulated.

In the case of the frog A. Mosso was able to prove that the
excitability of the nerves and muscles rapidly decreased ; sensa-
tion, especially in the hind legs, disappeared before the power of
motion, probably through the destruction of the tracts from the
spinal cord to the brain. The isolated heart of a frog was not
affected.

In warm-blooded animals the pressure of the blood rises im-
mediately after the injection, and then sinks. The action of the
heart slackens and becomes irregular, but the heart still con-
tinues to beat after death (BARDIER 2). After very large doses,
however, it was found by A. Mosso that the animal died very
rapidly from paralysis of the heart, while breathing still con-
tinued for a minute. The animals (dogs) also died from paralysis

1 Pennavaria, Farmacista Italiano, xii., 328, 1888 (quoted by Robert,
loc. cit.).

2 Bardier, "Action cardiaque du serum d'anguille," Soc. Biol., 1., 548,
1898.

16

242 TOXINES AND ANTITOXINES.

of the heart when artificial respiration was employed. The
autopsy showed congestion of the intestinal tract, lungs, and
suprarenal bodies ; severe lesions of the kidney, even when death
was rapid (PETTIT l) ; hyaline degeneration ; and swelling of the
cells. Blood was present in the urine in the bladder.

Serious alterations in the nervous system, resembling those
produced by tetanus, were detected by WESTPHAL 2 by means of
Nissl's method.

Thus we see that the effects produced by the poison are closely
similar to those caused by ricine, snake venom, &c.

Heated serum (58° C.) has still a slight action in very large doses (100
times the ordinary lethal amount) upon the animals, which lose considerably
in weight.

Action upon the Blood, — Eels' blood has an energetic haemo-
lytic action upon the blood even in the body (CAMUS and GLEY).
The iris becomes tinged with red ; areas of haemorrhage occur
and exudations of blood into the peritoneum, while erythrocytes
and haemoglobin are found in the urine. The arterial blood
contains haemoglobin. As a rule, the resistance of the erythro-
cytes is so weakened by the addition of ^-^ to TTFJTnF part of
eels' serum, that they give up their haemoglobin even to a 0'7 per
cent, solution of sodium chloride, whereas the normal corpuscles
of rabbits' blood do not part with it until the dilution of the
salt solution reaches 0*48 to O5 per cent. The blood-corpuscles
of the guinea-pig behave in a similar manner, whereas the
erythrocytes of the hedgehog have been found refractory, as
is also the case with those of hens, pigeons, tortoises, frogs,
toads, and bats. It is particularly interesting that, according
to H. SACHS, new-born rabbits have a relatively high power of
resistance, and do not acquire suitable receptors until a later
period (cf. Arachnolysine). The haemolytic function of eels'
serum is not affected by cautious neutralisation with hydro-
chloric acid.

The addition of other sera has also no influence upon it, but
it is destroyed by heating it to 55° C. The solvent action does
not take place at 0° C., but is very energetic at 23° C.

According to WENDELSTADT,S the addition of small quantities
of glycogen has an influence upon the haemolysis.

1 Pettit, "Alterations renales consecutives a 1'injection du serum
d'anguille," Soc. Biol, L, 320, 1898.

2 Quoted by Kossel (loc. cit. ).

3 Wendelstadt, "Einw. v. Glykogen a. hamolyt Vorgange, Central, f.
Balct., xxxiv., 831, 1903.

THE POISON OF EELS' BLOOD (ICHTHYOTOXINE). 243

Immunisation against Eels' Blood, — The poison of eels' blood
also shows itself to be a true toxine by the fact that it is
possible by means of it to produce immunity in susceptible
animals.

According to KOSSEL, CAMUS and GLEY, and WEHRMANN and
TcHiSTOViTCH,1 rabbits are the best animals to use for the
purpose. From 0*05 to 1 c.c. is first introduced, either by
subcutaneous or intravenous injection ; the animals usually
stand this dose well, and can then readily be immunised to a
higher degree. It is very difficult to immunise guinea-pigs since
they u'sually succumb ; dogs bear the treatment well, but yield
only weak antisera. Goats, on the other hand, appear to be
suitable animals. Hens and pigeons yield only traces of anti-
toxine, and that, too, only against the blood-solvent action in
vitro. Pigeons are very susceptible to the poison, although their
erythrocytes are hardly attacked at all by eels' serum. The
formation of an ti toxine is very rapid, and even after three or
four injections a serum is obtained of about -f$ to ^V ^he
neutralising strength — i.e., from 10 to 20 c.c. of serum are
required to neutralise 1 c.c. of eels' serum.

TCHISTOVITCH determined the strength of his serum as follows:
— Five drops of eels' blood (1 : 10 of a 0*7 per cent, solution of
sodium chloride) were treated with increasing doses of antitoxine.
The reagent for testing both the hsemolytic and the toxic action,
consisted of a few c.c. of rabbits' blood diluted to 20 c.c.
Parallel tests were made in each case, and some remarkable
results were obtained. Thus, while the resistance offered by the
rabbits themselves showed a continual increase, the proportion
of antitoxine in their serum did not increase in the same manner ;
but, on the contrary, became continually less.

At the same time, the erythrocytes of these animals also
showed special characteristics. KOSSEL and CAMUS and GLEY
had simultaneously discovered that the erythrocytes of immun-
ised animals were in themselves (i.e., thoroughly freed from
serum) under certain conditions refractory to the hsemolytic
action of eels' blood. TCHISTOVITCH now discovered that the
blood-corpuscles as such dissolved very readily when the pro-
portion of antitoxine was high ; but that they were more or less
refractory when the amount of antitoxine in the serum decreased.

There is here a certain parallelism in the phenomena, which,
assuming the correctness of the facts, appears to point to a
disappearance of the receptors for eel toxine, both in the body

1 Tchistovitch, " Etudes 'sur 1'immunisation centre le s^rum d'anguille,"
Ann. Past., xiii., 406, 1899.

244 TOXINES AND ANTITOXINES.

cells and in the erythrocytes. If we assume that the toxic and
hsemolytic principle of eels' serum are identical, we must also
assume that receptors adapted to it are present in the body cells
as well as in the erythrocytes ; a cessation of the formation of
receptors under the influence of the immunisation process would
then be sufficient to account for both the immunity to the action
of the poison notwithstanding the diminished production of anti-
toxine and also the want of susceptibility of the erythrocytes.

It is, of course, open to question whether both principles are
really identical. Here we meet with the same difficulty as in
the case of ricine, in which, too, the action upon the blood can
be easily prevented without destroying its toxic power; and
there, also, the protective influence upon the erythrocytes
affords, under normal conditions, a measure of the antitoxic
power. And yet, as we have seen above, the question whether
or no ricine contains two active substances has not yet been
settled, although there is much to be said in favour of JACOBY'S
view (q.v.) that we may here be dealing with a double-branched
receptor.

In the case of eels' blood, however, the conditions are some-
what different.

Here the question is whether the blood- sol vent function of
the serum ought not to be separated entirely from the toxic
function, inasmuch as there may here be hsemolytic processes
exactly analogous to those produced by several other normal
sera acting upon foreign erythrocytes. As EHRLICH and
MORGENROTH have shown, in numerous researches, the action
in those cases is to be attributed to series of peculiar haptines
with different specific amboceptors and complements.

It has, of course, not yet been proved that there are not also
here, as JACOBY assumes for ricine, two separate cryophore groups
on one amboceptor, one of which has a hsemolytic and the other
a toxic action.

This is also not irreconcilable with the results obtained by
TCHISTOVITCH, who found that the hsemolytic function was
destroyed by heating the toxine to 55° C. (which, however, is
absolutely denied by CAMUS and GLEY), and that the serum
thus rendered partially inactive produced antitoxine just as
before. All these facts could be explained just as well by a
partial formation of toxoid as by assuming the existence of two
specific haptines with different haptophore groups.

In any case, a definite conclusion can only be obtained by
exact combination experiments, on the lines devised by EHRLICH.
An attempt must be made to determine whether or no eels'

TOXINE OF FATIGUE. 245

serum still retains its toxic power after removal by means of
specific combinations of the receptors adapted to the corpuscles
of the blood.

Toxine of Fatigue.

An apparently true toxine has been separated from the muscles
of over-fatigued mammalia by WEICHART,1 who prepared an ex-
tract from them under the strictest aseptic precautions. In
small doses the toxine has an immunising effect, but in large
doses is fatal. It produces a true antitoxine on immunisation.

1 Weichart, " t)ber Ermiidungs toxin, I. and II.," Munch, med. Woch.,
1904, No. 1, et seq.

247

BIBLIOGRAPHY

AND INDEX OF AUTHORS' NAMES.

The numbers in italics indicate pages of this book. The references marked
with an asterisk have not been seen by me in the original.

ABBA, Ueb. d. Dauer des toxisch.

Vermbgens beim D.-T. u. Antit.

Centralbl. f. Bakt., xxiii., 934

(1898). 77.
*ALBERTONI, Sull' azione del veleno

della vipera. Lo Sperimentale,

1879. 218.
ALBRECHT & GHON, Bakt. Unters.

lib. d. Pestbac. Wiener Akad.,

Ixvi. (1898). 158.
*v. ANREP, Intoxication par les

ptomaines. Arch. Slaves de Biol.,
:i., 341 (1886). 126.
ARLOING, Sur la presence de la

substance phylacogene dans les

liquides du bac. anthr. Bull.

Medic. , 1892, 1038, through

Centralbl. /. Bakt., xiii., 561

(1892). 179.
— Serothe'rapie du charbon sympto-

matique. Comptes Rend., cxxx.,

548 (1900) ; cxxxi., 319 (1900).

131.
ARLOING & LAULANIIS, Et expe'r. sur

les troubles imprimes . . . par

les toxines diphth. Soc. Biol.,

xlvii., 433 (1895). 84-
ARMAND & CHARRIN, Transformation

de la matiere organique azote",

&c. Bull. Med., 1891, 356;
U892, 957; abst. Centratbl. f.

Bakt., xi., 248 (1892). 6.
ARON, H., Ueb. organische Kolloide.

Biochem. Centralbl., iii. (1904).

[15-16]. 62.

ARONSON, Berl. klin. Woch., 1893,
625. 37.

Immunisierungs- u. Heilversuche
bei der Diphtheric. Wien. med.
Woch., 1894, 1956. 74.
Weit. Unters. iiber Diphtheric.
Berl. klin. Woch., 1894, 425. 95.
Ueber Antistreptokokkenserum.
Berl. klin. Woch., 1896, 717. 164.

— Zur Biologic und Chemie der Diph-
theriebazillen. Arch. f. Kinder-
heilkunde, xxx. , 23 ( 1 900). 71.

ARRHENIUS & MADSEN, Anwendg.
d. physik Ch. auf d. Stud, der
Toxine u. Antitoxine. Z. Physik.
C7i.,xliv., 1 (1903). 60.

D'ARSONVAL & CHARRIN, Action des
courants a haute frequence sur
les toxines bacteriennes. Comptes
Rend., cxxii., 280 (1896). 83.

ASAKAWA, Die Basis der natiirl. Im-
mun. des Huhnes gegen Tetanus.
Centralbl. f. Bakt., xxiv., 166
(1898). 18.

ASTROS & RIETSCH, Essais d'extrac-
tion de 1'antitox. diphth. Soc.
Biol., Iii., 337(1900). 98.

ATKINSON, The Fractional Precipi-
tation of the Globulins and
Albumins of normal Horse Serum
and Diphtheric Antitoxic Serum.
Journ. of Exper. Med., v., 67
(1901). 100.

AUCLAIR, La Sclerose pulmonaire,
&c. Arch, de Med. Exper., 1900,
189. See also Baumg. Jb., 1898,
475, 476. 175.

248

BIBLIOGRAPHY.

B

BAIL, Ueber leukocide Substanzen !

in den Stoffwechselprodukten d.

Staph. pyog. aureus. Arch. f.

Hyg., xxxii., 133 (1898). 143.
BALDASSARI, Ueb. d. Wirkg. d.

D.-T. auf den Zellkern. C. f.

allg. PathoL, vii., 625 (1896). 86.
BALDWIN & LEVENE, Einw. proteo-

lyt. Ferm. auf Bakterientoxine.

Journ. of Med. Research., vi.,

120 ; Malys Jb. , 1 901 , 953. 21.
BARBA-MORRIHY, Baumgartens Jb.,

1897, 403. 167.
BARDIER, Toxine et cceur. Soc.

BioL, xlix., 311 (1897). 25.

— Action cardiaque du serum d'an-
guille. Soc. BioL, 1., 548 (1898).

BECK & SLAPA, Ueb. den Einfl. d.

D.-G. auf den Kreislauf. Wien.

klin. Woch., 1895, 333. 85.
v. BEHRING, Allg. Ther. d. Infekt.-

Kr., L, 1033. 117.

— Die Blutserumtherapie, iii., Leip-
zig, 1892. 112, 123, 125.

— Untersuch. RANSOM'S lib. d. Ag-
glutination der Choleravibrionen.
Deutsch. med. Woch., 1898, 294.
149.

— Bekampfung der Tuberkulose.
Munch., med. Woch., 1898, 580.
169.

— Abstract by the author of his
lecture in Madrid. Deutsch. med.
Woch., 1898, 293. 170.

- Ueb. Tetanusgiftmodifikat. Fort-
schr. d. Med., xvii., 501 (1899).
108, 109.

— Die Diphtherie. Berlin, 1901.
172.

— Ueber die spez. giftigen Eigensch.
d. Tuberkulinsaure. Berl. klin.
Woch., 1899, 537. 172.

BEHRING & KNORR, Ueb. den Im-
munisierungswert des Tetanus-
heilserums. Zeit. f. Hyg., xiii.,
407 (1893). 104.

BEHRING & WERNICKE, Ueb. Im-
munisier. u. Heilung von Ver-
suchstieren b. d. Diphth. Zeit. f.
Hyg., xii., 10(1892). 76.

BELFANTI, Sulla immunisazione per

mezzo di filtrati di sputo pneu-
monico. Mif. Med., 1892, 126;
CentralU. f. Bakt., xii., 401.
160.

BELFANTI & CARBONE, Contribute
alia conoscenza dell' antitossina
difterica. Arch, per le Scienze
Med. , xxii. , No. 2 ; abst. Cen-
tralU. f. Bakt. , xxiii. (1898). 100.

Bericht der deutschen Pest-Comm.
Arb. Kais. Ges.-Amt., xvi. (1899).
158.

BERNHEIM, Immun. tnberc. et serum-
the'rapie. Soc. BioL, xlviii., 291
(1896). 167.

BERT, Soc. BioL, xxxvii., 574 (1885),
237.

BESREDKA, De 1'hemolysine strepto-
coccique. Ann. Past., xv., 880
(1901). 142.

— De la fixation de la tox. tet. par
le cerveau. Ann. Past., xvii.,
138 (1903). 113, 117.

BESSON, Contribut. a 1'etude du

vibrion septique. Ann. Past.,

ix., 179 (1895). 183.
BEUMER & PEIPER, Bakt. Stud, iib,

Typhusbaz. Zeit. f. Hyg., ii.,

110 (1887). 164.
BILLINGER, Winterschlaf und Infek-

tion. Wien. klin. Woch., 1896,

769. 119.
BITTER, Ueb. Festig. v. Versuch-

stieren geg. d. Intoxikation durch

Typhusbazillen. Zeit. f. Hyg.,

xii., 298 (1892). 155.
BLUM, Ueb. Antitoxinbildung bei der

Autolyse. Hofm. Beitr., v., 142

(1904). 28.
BLUMENTHAL, Ueb. d. Moglichk. d.

Bildg. von Diphtherietoxin aus

Eiweisskorpern und auf Zucker-

enthaltenden Niihrboden. Deutsch.

med. Woch., 1897, No. 24. 72.

— Weit. Beitr. z. Kenntn. des
Tet.-G. Z.f. klin. Med., xxxii.,
325 (1897). 122.

— Ueber die Veranderungen des
Tetanusgiftes im Tierkorper.
Deutsch. med. Woch., 1898, 185.
'117, 122.

BLUMENTHAL & JACOB, Zur Serum-
therapie des Tetanus. Berl. klin.
Woch., 1898, 1079. Ill, 117.

BIBLIOGRAPHY.

249

BOMSTEIN, Z. Frage der passiven Im-

munitat bei Diphtherie. CentralbL

f. Bakt., xxii., 587 (1897). 94.
- Ueb. d. Schicksal der Diphth.-T.

im Tierorganismus. CentralbL f.

Bakt., xxiii., 785 (1898). 87.
BORDET, Sur le mode de 1'action des

antitoxines sur les toxines. Ann.

Past., xvii., 161 (1903). 56.
BREYMANN, M., Ueb. Stoffwechsel-

prod. d. Bac. pyocyaneus. Cen-
tralbL f. Bakt., xxxi., 481 (1902).

128, 137.
BRIEGER, Weiteres liber Ptomaine.

Berlin, 1885. 154.
— Weitere Erfahrungen iiber Bak-

teriengifte. Zeit. f. Hyg., xix.,

101 (1895). 14, 105, 106.
BRIEGER & BOER, Ueb. d. Toxine d.

Diphtherie u. s. w. Deutsch. med.

Woch., 1896, 783. 71, 79.

— Ueber Antitoxine und Toxine.

Ze.it. f. Hyg., xxi., 259. 95.
BRIEGER & COHN, Unters iib. d.

Tetanusgift. Zeit. f. Hyg., xv.,

1 (1893). 103, 106.
- Beitr. z. Concentr. d. geg.

Wundstarrkrampf schiitzenden

Substanz. Zeit. f. Hyg. , xv. , 439

(1893). 124.
BRIEGER & EHRLICH, Beitr. z.

Kenntn. d. Milch immuner Tiere.

Zeit.f. Hyg., xiii., 336 (1893). 96.
BRIEGER & FRANKEL, Ueber Bak-

teriengifte. Berl. Uin Woch.,

1890, 241. 76, 78, 178.
BRIEGER & KEMPNER, Beitrag z.

Lehre v. d. Fleischvergiftung.

Deutsch. med. Woch., 1897, 521.

127.
BRIEGER, KITASATO & WASSERMANN,

Ueb. Immunitat u. Giftfestigung.

Zeit. f. Hyg., xii., 137 (1892).

15, 76.
BRIOT, Etudes sur le venin de la vive

(Trachinus draco). Journ. de

Phys. et PathoL, 1903, 271.

238.
BRODIE, Journ. of PathoL , 1897, 460.

Quoted by MARTIN & CHERRY,

The Antagonism between Toxines

and Antitox." Proc. Roy. Soc.,

Ixiii., 420(1898). 93.
BRUCK, Beitr. z. Theorie d. Immuni-

tat. Zeit. f. Hyg., xlvi., 176
(1904). 27.

BRUNNER, Klin. u. exp. Stud iib. Tet.
Beitr. z. klin. Chirurg., ix.-xii.
(1892-94). 112.

— Z. Kenntnis d. Tetanusgiftes.
Zeitschriftf. klin. Med., xxxi., 367
(1897). 16.

- Unt. iib. d. Wirk. von Bakterien-

u. Pflanzengiften. Abst. CentralbL

f. Bakt., xxiv., 184 (1898). 88.
*BRTTSSCHETTINI, Bif. Med., 1892,

114.
BRUYLANTS & VENNEMANN, Le

Jequirity. Butt. Acad. Med.

Belgique, iii., 18, 147 (1884). 201.
BUCHNER, Tuberkulinreakt. d. Pro-

tei'ne nicht spezifischer Bakt.

Munch, med. Woch., 1891, 841.

172.

— Bakteriengifte u. Gegengifte.
Munch, med. Woch, 1893, 449.
7, 13.

— Zu R. KOCHS Mitt, iiber neue
Tuberkulinpraparate. Berl. klin.
Woch., 1897, 322. 170. ^

— Die Bedeutung der aktiven los-
lichen Zellprodukte, &c. Munch,
med. Woch., 1897, 12. 6, 170.

BUJWID, Ueb. e. Methode d. Con-
centrat. d. diphth. Heils. mittelst
Ausfrieren. CentralbL f. Bakt.,
xxii., 287 (1897). 93, 95.

BULLOCH & HUNTER, Ueber Pyo-
cyanolysin. CentralbL f. Bakt.,
xxviii., 865 (1900). 137.

BUSCHKE & OERGEL, Beitrag zur
Kenntnis des Tetanus. Deutsch.
med. Woch., 1893, 149. 122.

CALABRESE & ZAGARI, Ricerche
sulla tossina ed antitossina
difter. (Giorn. internaz. di
Scienze Mediche, 1895, No. 4,
19-21). Baumgartens Jahrb.,
1895, 215, 84. '

CALMETTE, Etude expe"rimentale du
venin de Naja tripudians. Ann.
Past., vi., 160(1892). 210.

— Sur la toxicite du sang de cobra.
Soc. BioL, xlvi., 11 (1894). 210.

250

BIBLIOGRAPHY.

CALMETTE, Proprie'tes du serum des
animaux immunises centre le
venin des serpents. Comptes
Rend., cxviii., 120, 1004 (1894).
888.

— Contrib. & I'e'tude des venins.
Ann. Past., ix., 225 (1895). 815.

— Sur le se'rum antive'nimeux.
Comptes Rend., cxxii., 203
(1896). 204.

— Sur le venin des serpents. Ann.
Past., xi., 214 (1897). 211, 232.

— Sur le me'canisme de 1'immunisa-
tion centre les venins. Ann.
Past., xii., 343 (1898). 214.

CALMETTE & DELEARDE, Sur les
toxines non microbiennes. Ann.
Past., x., 675 (1896). 203, 213.

VON CALCAR, Ueb. d. Constit. des
Diphtheriegiftes. Berl. klin.
Woch., 1904, No. 39. 27.

CAMUS, Resistance aux temper,
elevee des vaccins desseche's.
Soc. BioL, 1., 235 (1898). 28, 93.

CAMUS & GLEY, Recherches sur
1'action physiolog. de se'rum
d'anguille. Arch. Internal, de
Pharmacodynamie, v., 247 (1898)

840.

CANO-BRUSSO, Untergang d. Tetanus-

giftes im Darm. Malys Jb., 1901

914. 21.
CANTANI, Giftigkeit der Cholera-

bazillen. Deutsch. med. Woch.,

1886, 789. 146.
CARNOT, Reprod. experim. de la pneu-

monie fibrineuse. Soc. BioL, li.,

927(1899). 161.
CARNOT & FOURNIER, Sur le pneu-

mocoque et ses toxines. Arch.

Med. Exper., 1900, 357. 161.
Lesions cardiaques et muscu-

laires par la toxine pneumon.

Soc. BioL, lii., 143 (1900). 161.
CARRIERE, Etude expe'r. des altera-
tions histolog. du foie, &c. Arch.

Med. Exptr., ix., 65 (1891). 169.

— Toxines et digestion. Ann. Past.,
xiii., 435 (1899). 20, 110.

— Du sort de la toxine te"tanique
introduit dans le tube digestif.
Soc. BioL, li., 179 (1899). 20.

CASAGRANDI, L'Emolisina e la Leuco-
lisina Diplococcica. Bull. Soc.

Lands. Horn., xxvii., 2. Bioch.
Centr., i., 402 (1903). 142.
HANTEMESSE, Toxine typhoide
soluble. Progr. Med., 1898,
245. 155.

— Losliches Typhustoxin. Wien.
med Blatter, 1898, 18, et seq. 155.
HANTEMESSE & WIDAL, L'immunite
centre le virus de la fievre typh-
oide. Ann. Past., ii., 1888, 54.
164.

Et. experim., &c., de 1'infec-

tion typhique. Ann. Past., vi.,
755 (1892). 154.
CHARRIN, Action des sues digestifs
sur les poisons microbiens. A rch.
dePhys.,!S98,67. 21.
CHARRIN & BARDIER, Action car-
diaque proprie'te* speciale de la
botuline. Soc. BioL, 1., 60
(1898). 128.
CHARRIN & CASSIN, Fonctions pro-
tectices actives de la muqueuse
intestinale. Sem. Med., 1895,
545. 20.

CHARRIN & LEFEVRE, Action de la
pepsine sur la toxine dipht. Soc.
BioL, xlix., 830 (1897); Sem.
Med., 1897, 296. 21.
CHARRIN & LBVADITI, Action du
pancreas sur la toxine dipht.
Soc. Biol.,li., 215 (1899). 21.
CHARRIN & Moussu, Influence de la
dialyse, &c. Soc. BioL, lii., 694
(1900). 13, 82.

CHAUVEAU, Sur le mecanisme de
1'immun. Ann. Past., iii., 66
(1888). 177.

CHRISTMAS, Le Gonocoque et ses
toxines. Ann. Past., xi., 609
(1897). 162.

— Contrib. a l'e"tude du Gonococcus.
Ann. Past., xiv., 331 (1900). 162.

CLAUDE, Myelite ai'gue par toxines
streptocoq. Soc. BioL, xlviii.,
122 (1896). 164.

— Deux cas d'hemorrhagie de la
vesic. biliaire, &c. Soc. BioL,
1896, 169; Sem. Med., 1896, 62.
86.

COBBETT, Der Einfluss der Filtration
auf das Diphth.- Antitoxin. Cen-
tralbl.f. Bakt., xxiv., 386 (1898).
92.

BIBLIOGRAPHY.

251

COBBETT, Enthailt das normals Pfer-
deserum Antitoxin ? Centralbl. f.
Bah., xxvi., 458 (1899) ; see also
Lancet, 1899, ii. , 332. 91.

— Excretion of Dipht. Tox. in the
Urine. Brit. Med. Journ., 1900,
i., 21. 16. •

COBBETT & KANTHACK, Ueb. das
Schicksal d. Diphtherietoxins im
Tierorganismus. Centralbl. f.
Bakt., xxiv., 129 (1898). 33.

CONRADI, Z. Frage d. Toxinbild. bei
d. Milzbrandbakt. Zeit. f. Hyg.,
xxxvii., 287 (1899). 177.

— Ueber losliche, durch Autolyse
erhaltene Giftstoffe. Deutsch.
med. Woch., 1903, No. 2. 152,
156.

CONSIGLIO, Azione di aloune tossine,
&c. Arch, di Farm., vi., No. 3
(1898). Maty a Jb., 1898, 634.
26, 88.

COUNCILMAN, MALLOBY & PEABCE,
Diphtheria. — A Study of Bac-
teriology, &c., 1901, quoted at
length by VAUGHAN & NOVY,
The Cellular Toxines, 1902, p.
75, et seq. 87.

COUBMONT & DOYON, (a) Mecanisme
de production des contractures du
te"tanos. Arch, de Phys., 1893,
64. (6) La substance toxique qui
engendre le tetanos. Sem. Med.,
1893, 122. (c) Du te"t. de la
grenouille. Ibid., 1893, 302.
(d) De la prod. d. t. chez la
poule. Ibid., 1893, 486. .120.

Marche de la tempe"rat. dans

1'intoxication diphth. Arch, de
Phys., xxvii., 252 (1895). 84.

Effets de la toxine choler.

Arch, de Phys., xxviii., 785
(1896). 150.

Le Tetanos. Paris, Bailliere,

1899. 1U.

COUBMONT, DOYON & PAVIOT, Action
de la toxine diphtherique sur le
systeme nerveux de la grenouille,
&c. Soc. Biol. , xlvii., 210 (1895).

84.

Lesions intestinales dans

1'intoxic. diphth. Arch, de Phys.,
xxvii., 484 (1895). 86.

-— Lesions hepatiques engen

dres par la t. d. Arch, de Phys.,

xxvii., 687 (1895). 86.
CBOCQ, S. 1. alt^rat. du syst. nerveux

dans. 1. paralys. diphth. Arch.

Med. Exper., vii., 503 (1895).

86.
BOLY, Sur 1. disparition de la tox.

dipht. injecte'e dans le sang.

Arch. Intemat. de Pharmacody-

namie, iii. 88.
BUZ, La ricine. Ann. d' * Hygiene

Publique, xl., 344 (1898). 185.
JUSHNY, Ueb. das Ricinusgift. Arch.

Exper. PathoL, xli., 439 (1898).

186.

DANYSZ, Constitution des toxines.

Ann. Past. , xiii. , 581 (1899). 82.
— Contr. a 1'^tude des propr. des

melanges des toxines avec leurs

antitox. Ann. Past., xvi., 331

(1902). 60, 194.
DEBBAND, Sur un nouveau precede

de culture du bacille du tetanos.

Ann. Past.,xiv., 757 (1900). 104.
DECBOLY & RONSSE, Pouvoir toxique

et antitoxique du sang, &c. Arch.

Internat. de Pharmacodyn. , vi.,

211 (1899). 119.
DELEABDE, Rech. expe"r. sur les pro-

prietes, &c., de 1'antipyrine.

Arch, de Med. Exper., 1897, 786.

81.
DELEZENNK, Action du se"rum

d'anguille sur la coagul. du sang.

Arch. d. Phys., xxix., 646 (1897).

940.
*DENYS & VAN DE VELDK, Sur la

production d'une antileucocidine,

&c. La Cellule, xi. 144.
DEUTSCH & FEISTMANTEL, Die Impf-

stoflfe und Sera. Leipzig, 1903,

p. 40. 23.
DIEUDONNE, Ueb. Diphtheriegift

neutralis. Wirkung d. Serum-

globuline. Arb. a. d. Kais. Oes.-

Amt., xiii., 293 (1897). 91,

100.
*DIXSON, Australian Med. Gaz.,

1887, 156. 185.
DONITZ, Ueb. d. Verhalten d. Choi.-

252

BIBLIOGRAPHY.

Vibr. im Hiihnerei. Zeit.f.Hyg.,
xx., 31 (1895). 146.
DONITZ, Ueber das Tetanusantitoxin.
Deutsch. med. Woch., 1897,
428. 15, 116, 119, 120.

— Ber. iib. die Thatigkeit des Kgl.
Instituts f. Serumforschung, &c.
See also Klin. Jahrbuch, vii.
(1899). 44, 123.

— Ueb. d. Grenzen der Wirksamkeit
d. Diphtherieheilserums. Arch.
Intemat. d. Pharmacodyn. , v.,
425(1899). 87.

DBEYER, Ueber die Grenzen d.
Wirkung d. Diphtherieheil-
serums gegeniiber d. Toxonen.
Zeit. f. Hyg., xxxvii., 267
(1901). 90. „

DUENSCHMANN, Etude experim. sur
le charbon symptomatique. Ann.
Past. , viii. , 403 ( 1 894). 180.

DUNBAR, Zur Frage betreffend die
Aetiologie u. specif. Therapie des
Heufiebers. Berl. Idin. Woch.,

1903, 24-26; Z. Aetiologie des
Herbstkatarrhes. Berl. Tclin.
Woch., 1903, 28. 207.

v. DUNGERN, Steigerung d. Giftpro-
duktion d. Di.-B. Centralb. f.
Bakt., xix., 137 (1896). 71.

- Bindungsverh. bei d. Pracipitin-
reaktion. CentralU. f. Bakt.,
xxxiv., 355 (1903). 60.

— Beit. z. Kenntn. der Bindungs-
verh. bei der Verein v. Di.-Gift.
u. Ant. Dentsch. med. Woch.,

1904, Nos. 8, 9. 62.
DZIERZGOWSKI, Z. Frage : Ueb. die

Verluste des Diphth. -Heils. bei
der Filtration. CentralU. f.
Bakt., xxi., 333(1897). 92.

— Zur Frage iiber die Beziehungen
zwischen dem antidiphth. Heil-
serura u. d. Diphtherietoxin.
Arch. Internal. -de Pharmacodyn.,
v., 1 (1898). 32.

- Die Bezieh. d. Verdauungsfer-
mente zum Antidiphtherieserum.
Arch. d. Sciences Biolog., vii.,
337. (Malys Jb., 1899, 957).
94.

DZIERZGOWSKY & SIEBER, Archiv.
des Sciences Biol. de St. Pttersb.,
viii. 110.

EHRLICH, Die Wertbemessung des
Diphtherieheilserums. Klin.
Jahrbuch, vi. 15, 35, 89.

— Exper. Unters. iib. Immunitat.
Deutsch. med. Woch., 1891, 976,
1218. 30, 194.

— Experimentelle Unters. iiber Im-
munitat. Deutsch. med. Woch.,

1897, 41. 30, 186.

— Zur Kenntn. d. Antitoxinwirkg.
Fortschr. d. Med., 1897, 41. 30,
31, 186, 194.

— Verh. Ges. Charite'-Aerzte, Feb.

1898. Berl. klin. Woch. 13S,
206.

— Ueb. d. Bezieh. von chemischer
Konstitution, Verteilg. und
pharm. Wirkg. Festschr. f.
Leyden, 1902. 819.

— Toxin und Antitoxin. Munch,
med. Woch., 1903, Nos. 33, 34.
9, 57.

— Ueb. d. Giftkomponenten des
Diphtherietoxins. Berl. klin.
Woch., 1903, No. 35. 57.

EISENBERG, M., D. Bindungsverh.

zwischen Toxin und Antitoxin.

CentralU. f. Bakt., xxxiv., 259

(1903). 59.
ELFSTRAND, Ueber blutkorperchen-

agglutinierende Eiweisse. Gor-

bersdorfer Veroffentlichungen.

Edited by R. ROBERT. Stutt-
gart, 1898, 1. 205.
ELLIOT, Action of Cobra Poison.

Proc. Roy. Soc., Ixxiii., 183

(1904). 218.
ELLIOT, SILLAR & CARMICHAEL,

Action of the Venom of Bungarus

caeruleus. Lancet, 1904, ii., 142.

216.
ENRIQUEZ & HALLION, Myelite ex-

pe"r. tox. dipht. Soc. Biol., xlvi.,

312(1894). 85.

— Sur les effets physiol. de la toxine
dipht. Arch, de Phys., xxvii.,
515 (1895).

EPPINGER, Die toxische Myolyse des
Herzens bei Di. Deutsch. med.
Woch., 1903, Nos. 15, 16. 85.

VAN ERMENGEM, Sur 1'inoculation
des produits de culture du bacille

BIBLIOGRAPHY.

253

virgule. Bull. Acad. Med. Belg.,

iii., 18, 1221 (1884). 145.
VAN ERMENGEM Ueber einen neuen

anaeroben Bacillus u. s. Bezieh.

zum Botulismus. Zeit. f. Hyg.,

xxvi., 1 (1897). 126.
ERNST, COOLIDGE & COOK, The Effect

of Freezing upon the Antidipht.

Serum. Journal Boston Med.

Soc., ii., 166; Baumgartens Jb.,

1898, 242. 95.

FABER, Die Pathogenese des Tetanus.

Berl. klin. Woch., 1890, 717. 103.
FAYRER & LAUDER-BRUNTON, On the

Nature of the Poison of Naja

tripudians, &c. Proc. Roy. Soc.,

xxi., 371 (1873). 222.
FAYRER & BRUNTON, On the Nature

of the Poison of Naja tripudians,

&c. Proc. Roy. Soc., xxii., 68

(1874). 209.
FENYVESSY, Ueb. d. Wirkg. d. D.-T.

u. Antit. auf das Froschherz.

Jahrb. f. Kinderh., New Series,

xliii., 216(1896). 85.
FERMI & PERNOSSI, Ueber das

Tetanusgift. Zeit. f. Hyg., xvi.,

385 (1894). 14, 103.
FISCHL & v. WUNSCHHEIM, Ueb.

Schutzkorper im Blute der Neuge-

borenen. Prager med. Woch.,

1895, 45, et seq. 92.
FLEXNER, The pathology of tox-

albumin intoxication. Johns

Hopkins Hosp. Record, vi., 259

(1897). 87, 189.

FLEXNER & NOGUCHI, The Constitu-
tion of Snake Venom and Snake

Sera. Univ. of Pennsylv. Med.

Bull., Nov. 1902. 224.
— Snake Venom in relation to

Haemolysis, &c. Journ. of Exper.

Med., vi., 277 (1902). 914, 223.

— Upon the Production and

Properties of Anti-crotalus Serum.

J. of Med. Research, 1904, xi.,

363. 230.
FOA & CARBONE, Sulla immunita

verso il diplococco pneumonico.

Oazz. Med. di Torino, 1891, 1.

(CentralU.f. Bakt., x., 768.) 160.
FOA & SCABIA, Sulla immunita della

pulmonite. Gaz. Med. d. Torino,

1892, 13, 15; CentralU.f. Bakt.,

xi., 615. 160.
*FONTANA, Trattado del veleno della

vipera, 1787. 209.
FORSSMAN, Beitr. z. Kenntn. d. Bakt.

d. Botulismus. Abst. in Centralbl.

f. Bakt., xxix., 541 (1901). 126.
FOTH, Das Mallein, &c. Fortschr. d.

Med., 1895, 637. 175.
FRANKEL, Immunisiernngsversuche

bei Diphth. Berl. klin. Woch.,

1890, 1133. 76.

— Ueb. d. Wirkg. d. Ricins auf
Fischblut. Hofm. Beitr., iv., 224
(1903). 192.

FRASER, Immunity against Snake
Poison. Brit. Med. Journ. , 1895,
i., 1309. 216, 228.

— Antitoxic Qualities of the Bile of
Serpents. Brit. Med. Journ.,
1897, ii., 595. 232.

— Antivenomous Properties of the
Bile of Serpents. Brit. Med.
Journ., 1897, ii., 125. 232.

— The Treatment of Snake Poison-
ing with Antivenene. Brit. Med.
Journ., 1895, ii., 417. 232.

FRASER & ELLIOT, Action of Sea

Snake Venoms. Lancet, 1904,

141. 216.
FRENKEL & BRONSTEIN, Ueber Tuber-

kulosetoxine und Antitoxine.

Berl. klin. Woch., 1901, 861. 167.
FREUND & GROSZ, Ueb. d. Bez. zw.

Gerinnung u. d. Wirkg. d. Anti-
toxine. Centralbl. f. inn. Med.,

1895, 613, 637. 82.
FREUND & STERNBERG, Ueber Dar-
stellung des Heilkorpers aus d.

Diphtherieheilserum. Zeit. f.

^^.,31,429(1899). 97.
FRIEDRICH, Beobacht. lib. d. Wir-

kung von subkutan einverleibten

Streptokokken - Toxinen, &c.

Berl. klin. Woch., 1895, 1065.

164.
FULD & SPIRO, Ueb. labende u. lab-

hemmende Wirkung d. Blutes.

Z. f. Phys. Ch,, xxxi., 132 (1900).

101.

254

BIBLIOGRAPHY.

GAMALEIA Les poisons bacte'riens.
Arch, de Med. Exper., 1892. 70.

— Recherches exper. sur les poisons
du cholera. Arch, de Med. Exper.,
1892,173. 146.

GATTI, L'azione di alcune tossine

batteriche sopra gli element! della

retina. Abst. Bioch. Centr., i.,

No. 775 (1903). 86.
DE GIAXA & PANE, Contribute alle

cognizione sulla immunis. contra

la infez. da streptococco. Riform.

Med., xii. [4], p. 5 (Baumgart.

Jb.t 1896, 23). 164.
GIBIER, Effets produits par 1. toxines,

&c., injectees dans le rectum.

Sem. Me-d., 1896, 202. 20.
GOLDBERG, Ueber Ausscheidung des

Tetanusgiftes durch die Nieren-

sekretion. CentralU. f. Bakt.t

xxvi., 547 (1899). 16.
GOLDSCHEIDER & FLATAU, Ueber die

Ziele der modernen Nervenzellen-

forschung. Deutsch. med. Woch.,

1898, 165. 114.
GRAMATSCHIKOFF, Ueber einige

physiolog. Wirk. des Kochschen

Tuberkulins. Arb. patholog . Inst.

Tubingen, i., 287; Baumg. Jahrb.,

1897,584. 169.
GRASSBERGER & SCHATTENFROII,

Ueber das Rauschbrand, 1904.

131.
GROSZ & KRAUS, Bakteriol. Studien

iiber den Gonococcus. Arch. f.

Dermatol., xlix., 3 (1899). 163.
GRUBER & WIENER, Ueb. d. intra-

periton. Cholerainfektion. Wien.

klin. Woch., 1892, 543. 146.
GUINARD, Effets physiolog. du mal-

lei'ne. Journ. Med. Vet., xlvi.,

454; Baumg. Jb., 1895, 311. 176.
GDINOCHET, Contrib. a 1'etude de la

toxine du bacille de la diphtheric.

Arch. d. Med. Experim., 1892,

487. 6, 71.
GUMPRECHT, Zur Pathog. d. Tetanus.

Deutsch. med. Woch., 1894, 546.

114.

— Vers. itber d. physiol. Wirk. des
Tetanusgiftes. Pflugers Arch.,
lix., 105 (1895). 114.

H

HAHN, Immunis. u. Heilungsvers.

mit den plasmat. Zellsaften.

Munch, med. Woch., 1897, 1344.

150, 170.
HANKIN, On Immunity produced

by an Albumose isolated from

Anthrax Cultures. Brit. Med.

Journ., 1889, ii., 65. 177.
— On the Conflict between the

Organism and the Microbe. Brit.

Med. Journ., 1890, ii., 810.

177.
HANKIN & WESBROOK, Albumoses et

toxalbuminoses du bacille char-

bonneux. Ann. Past., vi., 633

(1892). 177.
HARNACK & HOCHHEIM, Ueb. Wirk.

d. Briegerschen Tet.-G. Z. f.

klin. Med., xxv., 46 (1894). 113.
HAUSMANN, Zur Kenntnis des Abrins.

HAYASHI, Ueber die chemische Natur

Tetanustoxins. A re, h. f. exp.

PathoL, xlvii., 9; and Ghem.

CentralU., i., 411 (1901). 107.
HELLIN, Der giftige Eiweisskorper

Abrin. Diss. Dorpat, 1891. 201.
HENSEVAL, L'abrine. La Cellule,

xvii., 139. Malys Jb., xxxi., 190.

Ml.
HIRSCHLAFF, Antimorphinserum.

Berl. Ein. Woch., 1902. 4.
Hiss & ATKINSON, Serumglobulin

and Diphtheric Antitoxin. Journ.

of Exper. Med., v., 47 (1901).

100.
HOMEN, De Faction du streptocoque

et de ses toxines sur les nerfs.

Sem. Med., 1896, 211 ; Soc. Biol,

1896,23. 164.
HUEPPE, Ueb. d. Aetiologie u. Toxi-

kologie der Cholera asiatica.

Deutsch. Med. Woch., 1891, 417.

146.

I

IDE & LEMAIRE, Et. s. 1. repartition
de 1'antitoxine dipht., &c. Arch.
Internal, d. Pharmacodyn. , vi.,

477. 100.

BIBLIOGRAPHY.

255

ISSAEFF, I/immunite* centre le pneu-
mocoque. Ann. Past., vii., 259
(1893). 160.

- Unt. iib. d. kiinstl. Imm. gegen
Cholera. Zeit.f. Hyg., xv., 287
(1894). 147.

JACOBY, Die chemische Natur des
Ricins. Arch. exp. Path., xlvi.,
28(1901). 12,186.

— Ueber Ricinimmunitat. Hofm.
Beitr., i., 51 (1901). 18,194.

— Ueb. Ricinimmunitat. Hofm.
Beitr. z. chem. Physiol., i., 57
(1901). 24.

— Ueb. Ricinimmunitat. Hofm.
Beitr., ii., 535 (1902). 197.

- Ueb. d. chem. Natur des Ricins.
Arch, exper. PathoL, xlvi., 28.
186.

- Ueber Crotin-Immunitat. Hofm.
Beitr., iv., 212 (1903). 206.

JOUKOWSKY, De 1'influence de la
toxine te"tan. sur le systeme ner-
veux. Ann. Past., xiv., 464
(1900). 114.

K

KANTHACK, The Nature of Cobra

Poison. Journ. of Phys., xiii.,

272(1893). 212.
KARFUNKEL, Schwankungen des

Blutalkaleszenzgehaltes nach Ein-

verleibung von Toxinen, &c.

Zeit.f. Hyg., xxxii., 149 (1899).

26.
KARTULIS, Unters. iiber das Verhal-

ten des Tet.-Giftes im Korper.

Diss. Berlin, 1892 (Dec.). 112.
KAUFMANN, Sur le venin de la vipere.

Soc. Biol, xlvi., 113 (1894). 217.
- Soc. Biol., xlviii., 860 (1896).

218.
KAYSER, Einw. d. Traubenzuckers

auf Staphylococcus. Ze.it. f.

#y<7.,xl.,21(1902). IjO.
— Ueb. Bakterienhamolysine, bes.

d. Colilysin. Zeit.f. Hyg., xlii.,

118 (1903). 138.

KEMPNER, Weit. Beitr. zur Lehre

von der Fleischvergiftung. Ze.it.

f. Hyg., xxvi., 481 (1897). 128.
KEMPNER & SCHEPILEWSKI, Ueb.

antitoxische Substanzen geg. d.

Botulismusgift. Ze.it. f. Hyg.,

xxvii., 213 (1898). 127.
KEMPNER & POLLACK, Die Wirkung

des Botulismustoxins auf die

Nervenzellen. Deutsch. med.

Woch., 1897, 505. 187.
KITASATO, Exper. Unters. itber das

Tetanusgift. Zeit. f. Hyg., x.,

287 (1891). 13, 103.
KITASATO & WEYL, Zur Kenntnis der

Anaeroben. Zeit. f. Hyg., viii.,

404(1890). 103.
KLEMPERER, G., Schutzimpf. d.

Menschen geg. asiat. Cholera.

Be.rt.Wn. JTocfc. , 1892, 970. 147.

— Ueb. kiinstlichen Impfschutz

fegen Choleraintoxikation. Berl.
lin. Woch., 1892, 789. Ij6.

— Die Beziehungen verschiedener
Bakteriengifte z. Immunitat u.
Heilung. Z. f. klin. Med., xx.,
165 (1892). 66, 159.

KLEMPERER, G. & F., Vers. iiber
Immunisierung u. Heilung bei der
Pneumokokkeninfektion. Berl.
klin. Woch., 1891, Nos. 34, 35.
159.

KNORR, Die Entstehung des Tetanus-
antitoxins. Fortschr. d. med.,
1897, 657. 33, 108.

— Das Tetanusgift und seine
Beziehg. zum tier. Organismus.
Munch, med. Woch., 1898, 321,
362. 29, 109.

KOBERT, Ueber vegetabilische Blut-
agglutinine. Sitzungsb. d. naturf.
Ges., Rostock, 25, v., 1900. 191.

— Beitr. z. Kenntn. d. Giftspinnen.
Stuttgart, 1901. 234.

— Giebt es fur den Menschen gefahr-
liche Spinnen ? Die Med. Woche,
1902, 154. 234.

— Ueber Giftfische u. Fischgifte.
Vortrag i. Rostocker Fischerei-
verein, 1902. Die Med. Woche,
1902. 238, 239.

KOCH, R., Vortr. iiber die Cholera.
Berl. klin. Woch., ISS4,, ±98. 145.

— Zweite Conferenz z. Erort d.

256

BIBLIOGRAPHY.

Cholerafrage. Berl. klin. Woch.,

1895 [37a], p. 8. 145.
KOCH, R., Mitt, iiber ein Heilmittel

gegen Tuberk. Deutsch. med.

Woch., 1891, 101, 1189. 168.
— Ueber neue Tuberk ulinpraparate.

Deutsch. med. Woch., 1897, 209.

169.

- Ueber d. Agglutinat. d. Tuberk el -
bazillen. Deutsch. med. Woch.,

1901, 829. 171.

KOCH, W., Milzbrand und Rausch-
brand. Stuttgart, 1886. 177.

KOLISKO & PALTAUF, Zum Wesen
des Croup und der Diphtherie.
Wien. klin. Woch., 1889, No. 8.
70.

KOSSEL, H., Zur Kenntnis d. Diph-
theriegiftes. Centralbl. f. Bakt.,
xix., 977 (1898). 10, 70.

— Zur Kenntnis der Antitoxinwir-
kung. Berl. klin. Woch., 1898.
7. 340.

KOSSEL & OVERBECK, Bakter. Unters.
itber Pest. Arb. Kaiserl. Ges.-
Amt, xviii. (1901). 159.

KRASSNOW, Z. Pharmakol. des Di.-T.
Wratsche-bnaja, 1904, 23; Gaz.
Biochem. Cent., iv., No. 304.
84.

KRAUS, Beitrag zur Klinik des Te-
tanus. Z. f. klin. Med. , xxxvii. ,
247(1899). 122.

- Ub. ein aeut. wirkendes Bakterien-
toxine. Centralbl. f. Bakt. , xxxiv.,
488(1904). 23.

KRAUS & CLAIRMONT, Ueber Hiimo-

lysine u. Antihamolysine. Wien.

klin. Woch., 1900, 49. 134, 139.
KRAUS & LUDWIG, Ueb. Bakterio-

hamagglutinine. Wien, klin.

Woch., 1902, 120. 143.
KHUGER, Ueb. die chem. Wirkg. d.

Elektriz. auf toxische und

immunis. Bakteriensubstanzen.

Deutsch. med. Woch., 1895, 331.

83.
KUHNE, Erfahrungen iiber Albumo-

sen u. Peptone. Z.f. Biol. , xxix. ,

24(1892); xxx., 220 (1894). 168.
KYES, Ueb. d. Wirkungsweise des

Cobragiftes. Berl. klin. Woch.,

1902, Nos. 38, 39. 224.

- Ueber die Isolierung von Schlan-

ger
We

engiftlecithiden. Berl. klin.
"och., 1903, Nos. 42, 43. 226.
KYES, Kobragift u. Antitoxin. Berl.

klin. Woch., 1904, 19. 62.
KYES & SACHS, Zur Kenntn. d.
Cobragiftaktivierenden Subst.
Berl. klin. Woch., 1903, Nos. 2-4.
225.

LAITINEN, Das Streptokokkentoxin

u. seine Wirkung auf das Nerven-

system. C. f. allg. PathoL , 1 896,

358. 164.
LANDMANN, Ueb. eine neue Meth.

der Tuberkulosetoxin - Behand -

lung. Hyg. Rundschau, x., No.

8; Centralbl. f. Bakt., xxvii., 870

(1900). 170.
LATJ, Ueb. vegetabil. Blutagglutinine.

Diss. Rostock, 1901. 192.
LAZARUS, Ueb. antitoxische Wirk-

samkeit des Blutserums. Berl.

Klin. Woch., 1892, 43, 44. 154.
LEDOUX & LEBARD, De Faction sur la

temper, du bouillon des cultures

tuberc. Arch. d. Med. Exper.,

x., 601 (1898). 167.
LEVY, E. & P., Ueber die Hamo-

lysine des Ty.-B. Centralbl. f.

Bakt., xxx., 405 (1901). 142.
v. LEYDEN-BLUMENTHAL, Der Te-
tanus. Nothnagels Handb.,

Wien, 1901. ~16.
v. LINGELSHEIM, Ueber die Wert-

bestimnnmg der Tbc. - Gif tprii-

parate. Deutsch. med. Woch.,

1898, 583. 169.
LINOSSIER & LEMOINE, Actionnephro-

toxique des inj. d. serums nor-

maux. Soc. Biol. , Iv., 515 (1903).

95.
LOFFLER, Unters. lib. d. Bedeutg.

der Mikroorg. f. d. Entstehg. d.

Diphth. Mitt. Kais. Ges.-Amt,

ii., 1884. 69-.
— Der gegenw. Stand der Frage

nach der Entsteh. d. Diphth.

Deutsch. med. Woch., 1890, 81.

69, 77.
LOEW & KOZAI, Ueb. d. Bild. des

Pyocyanolysins. AIalysJb.,xxxi.,

912(1901). 138.

BIBLIOGRAPHY.

257

LUBENAU, Hamolytische Fahigkeit

einiger pathogener Schizomy-

ceten. Centralbl. f. Bakt., xxx.,

356(1901). 138,140.
LUBOWSKI, Ueber einen atoxischen

undavirulenten Diphtheriestamm.

Zeit. /. Hyg., xxxv., 87 (1900).

7, 75.
LUST, Ueber einen Antikorper gegen

Crotin im normalen Organismus.

Hofm. Beit., vi., 132 (1904). 207.

M

MACFADYEN, Ueber das Vorkommen
intracellularer Toxine. Z. f.
allgem. Physiol., iii., 303 (1904).
156.

MACFADYEN & ROWLAND, An intra-
cellular toxin of the typhoid-
bacillus. Proc. Roy. Soc., Ixxi.,
77 (1902); Centralbl. f. Bakt.,
xxxiv., 618 (1903). 156.

MCFARLAND, Immunization of Ani-
mals to Rattle-snake Venom.
Abst. Centralbl. f. Bakt., xxix.,
496 (1901). 220.

— Some Investigations upon
Antivenene. Journ. Amer. Med.
Association, Dec. 1901. Abst.
Centralbl f. Bakt., xxxi., 792.
230.

MADSEN, Zur Biologic des Diphtheric-
bacillus. Zeit. f. Hyg., xxvi.,
157 (1897). 73.

— Constitution du poison diph-
th£rique. Ann. Past., xiii., 568
(1899). 36, 44, 89.

- Ueber Tetanoly sin. Zeit. f. Hyg.,
xxxii., 214 (1899). 134.

- Ueber Heilversuche im Reagenz-
glas. Zeit. f. Hyg., xxxii., 239
(1899). 136.

— Constit. der poison dipht. Cen-
tra/bl. /. Bakt., xxxiv., 630
(1903). 62.

MADSEN & DREYER, Ueb. Immun.
mit den Toxonen d. Diphtherie-
giftes. Zeit. f. Hyg., xxxvii.,
249(1901). 27, 89.

MAISONNEUVE, Longue conservation
de la virulence du venin des
Serpents. Comptes Rend. , cxxiii. ,
513 (1896). 213.

*MANFREDI& TRAVERSA, Sull'azione
dei prodotti di cultura dello
streptococco. Giorn. Intern, delle
Scienze Mediche, 1888. - 164.

MARAGLIANO, Heilung d. Lungen-
tuberkulose durch Heilserum-
therapie. Berl. Idin. Woch.,
1895, 689. 167.

- Ueb. das tuberkulose Heilserum
und seine Antitoxine. Berl. klin.
Woch., 1896, 773. 167.

— Ueb. d. Tuberkelantitoxin. Malys
Jb., 1900, 1044. 167.

— Extrait aqueux des bacilles de la
tuberc. Soc. Biol., 1., 94 (1898).
169.

MARCIIOUX, Se"rum anticharbonneux.
Ann. Past., ix., 9 (1895). 179.

MARCUS, Ueb. in Wasser losl. Serum-
globulin. Zeit f. pJiys. Ch.,
xxviii., 559 (1899). 100.

MARENGHI, Ueb. d. gegens. Wirkg.
antidiphth. Serums u.des Diphth.-
Toxins. Centralbl. f. Bakt. , xxii. ,
520 (1897). 32, 93.

MARIE, Rech. s. 1. toxines te"tanique.
Ann. Past., xi., 591 (1897). 112.

- Propriety antit^tanique des
centres nerveux. Ann. Past.,
•xii., 91 (1898). 117.

MARIE & MORAX, Rech. zur 1' absorp-
tion de la tox. te"t., Ann. Past.,
xvi., 818 (1902); xvii., 335
(1903). 114.

MARINESCO, Lesions des centres ner-
veux produites par la toxine du
bac. botulinus. Soc. Biol., xlviii.,
31 (1896) ; Sent. Med., 1896,488.
127.

MARKL, Beitrag. z. Kenntnis der
Pesttoxine. Centralbl. f. Bakt.,
xxiv., Nos. 18-20 (1898). 158.

— Weit. Unters. lib. Pesttoxine.
Zeit.J. Hyg., xxxvii., 401 (1901).
158.

MARMIER, Sur la toxine charbon-
neuse. Ann. Past., ix., 533
(1895). 179.

— Les toxines et I'electricit6. Ann.
Past., x., 469 (1896). 13, 83,
105.

MARMOREK, Le streptocoque et le
s§rum antistreptococcique. Ann.
Past., ix., 593 (1895). 164.
17

258

BIBLIOGRAPHY.

MARTIN, The Proteids of the Seeds
of Abrus. Proc. Roy. Soc. , xlii. ,
331 (1887). 201.

— The Toxic Action of the Albumose
from Seeds of Abrus prec. Proc.
Roy. Soc., xlvi., 100 (1889-90).
201.

- The Chemical Products of the
growth of Bacillus anthracis.
Proc. Roy. Soc., xxii. [5], 1890.
178.

— Production de la toxine diphte"r.
Ann. Past., xii., 26 (1898). 78.

— Die chem. Prod. path. Bakt.
Wien. med. Blatter, 1898, No. 25,
et seq. 156.

— Relation of the Toxine and Anti-
toxine of Snake Venom. Proc.
Roy. Soc., Ixiv., 88 (1899). 215,
231.

MARTIN & CHERRY, The Nature of

the Antagonism between Toxines

and Antitoxines. Proc. Roy. Soc.,

Ixiii., 420 (1898). 215, 231.
MARTIN & SMITH, The Venom of the

Australian Black Snake. Proc.

Roy. Soc. New - South - Wales,

1892, 240. Malys Jb., 1894,

404. 212.
MARTIN & WOLFENDEN, Physiolog.

action of the Seeds of Abrus prec.

Proc. Roy. Soc., xlvi., 94 (1889-

90). 201.
DI MARTINI, Sul comportamento del

siero antidifterico filtrato. Rif.

Med., 1896, No. 266; Abst.

CentralU. f. Bakt., xxiv., 861

(1898). 92.
MARX, Die Tet.-G. neutralis.

Eigensch. d. Gehirns. Zeit. f.

Hyg.,x\., 231 (1902). 118.
MENNES, Das Antipneumokokken-

serum. Zeit. f. Hyg., xxv., 413

(1897). 161.
METSCHNIKOFF, Zur Immunitatslehre.

Congr. f. inn. Med., 1892, 282.

182.

— Etudes sur 1'immunite, Ann.
Past., v., 289 (1892). 182.

— Influence de I'organisme sur les
toxines. Ann. Past., xi., 801
(1897); xii., 81 (1898). 18, 117.

— Immunitat, German translat. by
MEYER, Jena, 1902. 18, 117.

METSCHNIKOFF, Roux & TAURELLI-

SALIMBENI, Toxine et Antitoxine

cholerique. Ann. Past., x., 257

(1896). 149.
HANS MEYER & HALSEY, Tetanus-

studien. Festschr. f. JAFFE,

Braunschweig, 1901. 112.
HANS MEYER & RANSOM, Unters.

lib. d. Tetanus. Arch. exp. Path.,

xlix., 369(1903). 111,114.
MICHAELIS, L., Ueb. die Giiltigkeit

des Massenwirkungsgesetzes bei

der Reaction zwischen T. und A.

Biochem. CentralbL, iii. [1],

(1904). 61.
MICHAELIS & OPPENHEIMER, Ueb.

Immunitat geg. Eiweisskorper.

Engelmanns Arch., 1902, Suppl.-

H. 24, 29, 67.
MILCHNER, Nachweis der chemischen

Bindung von T. -G. durch Nerven-

substanz. Berl. klin. Woch.,

1898. 117.
*WBIB MITCHELL & REICHERT,

Research upon the Venoms of

Poisonous Serpents. Smithsonian

Contrib. , No. 647. Philadelphia,

1885; Washington, 1886. 210.
MIYAMOTO, Beitrage zur Tetanusver-

giftung. Deutsch. med. Woch.,

1900, 479. 116.
MOLLARU & REGAUX, Lesions du

myocarde dans 1'intox. aigiie par

la tox. dipht. Ann. Past., xi.,

97(1897). 85.
MOLTSCHANOFF, Ueber das Gono-

kokkentoxin und seine Wirkung

auf das Centralnervensystem.

Munch, med. Woch., 1899, 1013.

163.
MORAX & ELMASSIAN, Action de la

toxine d. sur les muqueuses. Ann.

Past., xii., 210 (1898). 86.
MORAX & MARIE, Action de la

chaleur s6che sur la tox. tet.

Ann. Past., xvi., 418 (1902).

105.
MORGENROTH, Zur Kenntnis des

Tetanus des Frosches. Arch.

Internal, d. Pharmacodyn., vii.,

265 (1900). 19, 119.

— Zur Frage d. Antimorphinserums.
Berl. Idin. Woch., 1903, 21. 4-

— Ueber die Bindung, von Di. -T. u.

BIBLIOGRAPHY.

259

Ant. Berl. Uin. Woch., 1904,

20 ; Zeit. f. Hyg., xlviii., 177

(1904). 62.
MOSNY & MARCANO, De Faction de

la toxine du staphyl. pyog. Sem.

Med., 1894, 544. 139.
Mosso, A., Die giftige Wirkung des

Serum der Mureniden. Arch. f.

exper. Path., xxv., Ill (1888).

217, 239.
Mosso, U., Recherches sur la nature

du venin, qui se trouve dans le

sang de 1'anguille. Arch. Ital. d.

BioL, xii., 229 (1889). 240.
MULLER, Ueber die Resistenz d.

Diphtherieheilserums gegeniiber

verschiedenen Einfliissen. Cen-

tralU. f. Bakt., xxiv., 251 (1898).

93.
MULLER, FRANZ, Beitrage zur Toxi-

kol. des Ricins. Arch, exper.

PathoL, xlii., 302 (1899). 186.
MURAWJEW, Das. D.-T. und Antit.

in ihrer Wechselwirkg. auf das

Nervensystem d. Meerschw.

Fortschr. d. Med., 1898, 93.

86, 94.
MYERS, Cobra Poison in relation to

Wasserniann's new Theory of

Immunity. Lancet, 1898, ii.,

23. 214.
- The Interaction of Toxine and

Antitoxine. Journ. of Pathol.,

vi., 415 (1900). 222.

N

NEISSER & WECHSBERG, Ueber eine
neue einfache Methode z. Beob.
von Schadigungen leb. Zellen und
Organismen. Munch, med. Woch. ,

1900, 1261. 143.

- Ueb. d. Wirkungsart bakteri-
cider Sera. Munch, med. Woch.,

1901, No. 18. 225.

- Ueber das Staphylotoxin.
Zeit.f. Hyg., xxxvi., 299 (1901).
139.

NENCKI & SCHOUMOW-SIMANOWSKI,
Ueber die Entigiftung der Toxine
durch die Verdauungssafte. Cen-
tralbl. f. Bakt., xxiii., 840. 20,
88, 110, 203.

NERNST, Ueb. die Anwendbark. d.

Gesetze d. chem. Gleichgewicht

auf Gemische von T. u. A. Z. f.

Electrochem., x., 22 (1904). 62.
NEUFELD, Zur Wertbestimmung

der Tuberkulosegiftpr'aparate.

Deutsch. med. Woch., 1899, 13.

171.
NICATI & RIETSCH, Effets toxiques

des produits, &c. (Cholera).

Comptes Rend., xcix., 929 (1884).

145.
NICOLAYSEN, Zur Pathogenitat und

Giftigkeit der 'Gonokokken.

Centralbl. f. Bakt., xxii., 305

(1897). 162.
NICOLLE, Preparation de la toxine

diphte"rique. Ann. Past., x.,

333 (1896). 73.
NIEMANN, Ueber Tuberkuloseheil-

serum. Munch, med. Woch.,

1897, 59. 169.
NISSEN, Ueber den Nachweis von

Toxin im Blute, &c. Deutsch.

med. Woch., 1891, 775. 122.
NOBECOURT, Action des levures. &c.

Soc. BioL, Hi., 753 (1900). 81.
NOCARD, Des injections revelatrices

de la tuberculine. Recueil de

Med. Veter., Ixxii., 369. Baumg.

«/&., 1895, 705. 169.

OPPENHEIMER, Toxine u. Schutz-

stoffe. BioL Centralbl., 1899,

799. 107.
— Zur Theorie der Fermentprozesse.

Munch, med. Woch., 1901. 107.
— Die Fermente und ihre Wirkun-

gen. 2nd Ed., Leipzig, 1903.

10, 49, 107, 151.
OSBORNE & MENDEL, Ricine. Amer.

Journ. Physiol, x., 36 (1903).

189.

PACE, Influenza della tossine difter.
II Policlin., 7 ; Baumg. Jb., 1900,
180. 84.

260

BIBLIOGRAPHY.

PADOA, Ueb. d. verschied. Wirkung

des Diphtheric- u. Typhustoxins.

Riform. Med., 1899, No. 26;

Malys Jb. , 1899, 92 1 . 25.
PALMIRSKI & OKLOWSKI, Medycyna,

23. Abst. Centralbl. f. Bakt.,

xix., 916. 93.
PALTAUF, Diskussion zu dem Vortrag

von GBUBEB. Wien. klin. Woch. ,

1901, 51. 117.
PALTSCHIKOWSKI, Ueb. d. Verande-

rungen der diphtheritischen

Toxine in den Nahrungswegen.

Abst. Centralbl. f. Bakt., xxv.,

843 (1899). 82.
* PANCERI & GASCO, Agli efFetti del

veleno della naja egiziana. Atti

Acad. Reale Napoli, 1873, p. 73.

218.
PANE, Ueber d. Heilkraft d. anti-

pneumon. Serums. Centralbl. f.

Bakt., xxi., 664 (1897). 160.
PABASCANDALO, E. neue Versuchs-

reihe iib. d. Serotherapie bei

Infektionen mit pyogenen Mik-

roorg. Wien. Uin. Woch., 1897,

861. 164.
PARK, The Preparation of Diph-

theria-Antitoxine. Med. Record,

xlvii., 484(1895). 74.
PABK & WILLIAMS, The Production

of Diphtheria Toxin. Journ. oj
Exper. Med., i., 164 (1896). 72.
PASQUINI, Sulla presenza del veleno

tetanico negli organi. Rif. Med.,

1902, 22, 23. Abst. Centralbl. f.
Bakt., xxxi., 117. 113.

PASSINI, Vers. iiber die Dauer
d. antidiphth. Schutzimpfung.
Wien. klin. Woch., 1896, 1111.

94.
PASTEUB & JOUBEBT, Etude sur la

maladie charbonneuse. Comptes

Rend., Ixxxiv., 900 (1877). 177.
PEIPEB, Ueber - die Wirkung des

KOCHS chen Mittels auf Gesunde.

Deutsch. med. Woch., 1891, 160.

169.
*PENNAVARIA, Farmacista Italiano,

xii., 328 (1888). 241.
PETEBMANN, Recherchessurl'immun.

centre le charbon au moyen des

albumoses extraites des cultures.

Ann. Past., vi., 32 (1892). 178.

PETBI, Unters. iib. die d. d. Wach-
stum der Cholerabakt. entste-
henden chemischen Umsetzungen.
Arb. Kais. Ges.-Amt., vi., 374
(1890). 146.

PETTIT, Alterations renales consecu-
tives a 1'injection du serum
d'anguille. Soc. BioL, 1., 320
(1898). 242.

PFEIFFER, Unters. iib. d. Choleragift.
Zeit. f. Hyg., xi.,3Q3(l8Q2). 148.

— Studien zur Cholera-Aetiologie.
Zeit. f. Hyg., xv., 268 (1894).
148.

— Ueber die spezifischen Antikorper
der Cholera. Zeit. f. Hyg., xx.,
217 (1895). 148.

— Ein neues Grundgesetz der Im-
munitat. Deutsch. med. Woch.,
1896, Nos. 7 and 8. 148.

PFEIFFER & KOLLE, Ueb. d. spez.

Reaktion d. Typhusbazillen.

Zeit. f. Hyg., xxi., 203 (1896).

155.
PFEIFFEB & WASSERMANN, Unt. iib.

das Wesen d. Choleraimmunitat.

Zeit. f. Hyg., xiv., 46 (1893).

147.
PHISALIX, Soc. Biol., xlviii., 233,

656 (1896). 212.

— Act. phys. du venin du Sala-
mandre. Soc. BioL, xlix., 723
(1897). 234.

— Propr. immunisantes du venin du
Salamandre. Soc. BioL, xlix.,
823 (1897). 234.

— La tyrosine vaccine chimique du
venin de vipere. Comptes Rend. ,
cxxvi., 431. 213.

— Le sues de Champignons vaccinent
contre le venin de vipere. Comptes
Rend., cxxvii., 1036 (1898). 213.

— Action du venin de vipere. Soc.

BioL, liv., No. 27, 1902. 222 1

%23
PHISALIX & BEBTKAND, Toxicite du

sang de la vipere. Comptes

Rend., cxvii., 1099 (1893). 218.
Recherch. s. la toxicite" du

sang du crapaud commun. Arch.

d. Phys., xxv., 517 (1893). 233.
Attenuation du venin de

vipere par la chaleur. Comptes

Rend. , cxviii. , 288 (1 894). 222.

BIBLIOGRAPHY.

261

PHISALIX & BERTRAND, Sur la pro-
priet6 antitoxique du sang des
animaux vaccines centre le venin
de vipere. Comptes Rend., cxviii.,
356 (1894). 229.

- Soc. Biol., xlvii., 443 (1895).
£18,

— Glandes venimeuses chez les
couleuvres. Soc. Biol., xlvi., 8
(1894); xlvii., 639 (1895). 216.

Soc. Biol., xlviii., 396 (1896).

££9.

— — S.I. presence des glandes veni-
meuses chez les couleuvres. Soc.
Biol., xlvi., 8 (1899). 210.

PIAZZA, Influenza della luce solare

sulla tossina difter. Ann.

d'Igiene Sperim. [Nuova Ser. ], v. ,

521 (1895). Abst. Centralbl. f.

Bakt., xix., 914(1896). 81.
PICK, Z. Kenntnis d. Immunkorper.

Hofm. Beitr. z. chem. Phys., i.

(1902). 102.
POHL, Ueber Fischgifte. Prager

med. Woch., 1893, 31. 238.
POWER & GAMBLER, Pharmac.

Journal, 1890, 711 ; Pharm.

Rdsch., Feb. 1890, 30. 207.
PROSCHER, Zur Kenntnis des Kroten-

giftes. Hofm. Beitr., i., 575

(1901). 233.

— Germ. Patent F., 13,750 (20, vi.,
1902). 98.

PUGLIESE, Arch. d. Farm., 1898.

R

RANSOM, Choleragift u. Cholera-
antitoxin. Deutsch. med. Woch.,
1895, 457. 148.

- Das Schicksal des Tet.-Giftes
nach seiner intestinalen Einver-
leibung. Deutsch. med. Woch.,
1898, 117. 20, 110.

- Die Lymphe nach intravenoser
Inj . von T. -T. Z. f. phys. Chem. ,
xxix., 349 (1900). 112.

— Die Injektion von Tetanustoxin

bezw. Antitoxin in den subarach-

noidalen Raum. Z. f. phys. Ch.,

xxxi. , 282 ( 1 900-0 1 ). 117.

— Sapoiiin und sein Gegengift.

Deutsch. med. Woch., 1901, 194.
££6.

REHNS, Immunite acquise centre
1'abrine. Soc. Biol., Ivi., 329
(1904). 203.

REICHEL, Ueb. Immunitat gegen das
Virus von Eiterkokken. Arch,
f. Ein. 07w-ttr0.,xlii.,237 (1891).
139.

REPIN, Sur 1'absorption de 1'abrine
par les muqueuses. Ann. Past.,
ix., 517 (1895). 20, 203.

RODET, Sur les proprietes toxiques
des cultures des bacilles d'Eberth.
Soc. Biol., 1., 774 (1898). 155.

RODET & COURMONT, De 1'existence
. . . dans des cultures du
staphylocoque d'une substance
vaccinante. Comptes If end. ,
cxiii., 432 (1891). 139.

RODET, A. & G. GUECHOFF, Versuche,
die Methode der Kollodiumsack-
chen auf die Kenntnis der
toxischen Produkte des Eberth-
schen Bacillus und des B. coli
anzuwenden. Soc. Biol., Hi.,
962, 965 (1900). 13, 82, 156.

ROGER, Action des produits solubles
du streptocoque d'erysipele. Soc.
Biol., xliii., 538 (1891). 164.

ROGERS, On the Physiol. Action of
the Poison of the Hydrophidse.
Proc. Hoy. Soc., Ixxii., 305
(1903). 218.

ROMBERG, PASSLER, BRUHNS &
MULLER, Unters. iib. d. allg.
Patholog. der Kreislaufstorung
bei akuten Infektionskrankh.
Arch. f. Tdin. Med., Ixiv., 652
(1899). 85.

ROMER, Ueb. d. formativen Reiz d.
Proteine BUCHNERS. Berl. klin.
Woch., 1891, 886. 17 i.

— Tuberkulinreaktion d. Bazillen-
extrakte. Wien. Tdin. Woch.,
1891, 835. 172.

— Ueber Abrinimmunitat. Arch.
/. Ophthalm. , lii. , 90 (1901 ). 202.

ROSENTHAL, Das Dysenterietoxin.

Deutsch. med. Woch., 1904, No.

7. 184.
Roux, Immunite centre le charbon

symptomatique. Ann. Past., ii.,

49(1888). 130.

262

BIBLIOGRAPHY.

Roux & BORREL, Tetanos cerebral.

Ann. Past.,xii. (1898). Ill, 115.
Roux & CHAMBERLAND, Immunite

centre la septicemie. Ann. Past.,

ii., 561 (1887). 183.
Roux & VAILLARD, Contrib. k I'&ude

du tetanos. Ann. Past., vii., 65

(1893). 105.
Roux & YERSIN, Contribution a

1 'etude de la diph.th.erie. Ann.

Past., iii., 273 (1889); iv., 385

(1890). 6, 69.
RUPPEL, Zur Chemie der Tuberkel-

bazillen. Z. f. phys. Ch., xxvi.,

218 (1898). 171.
RUPPEL & RANSOM, Ueb. d. Mole-

kularverhaltnisse von Tetanus-

giftlosungen. Z. f. phys. Ch.}

xxvii., 109 (1899). 104.
RUETE, Ueb. Herstellg. d. Di.-

Heilserums. Munch, med. Woch.,

1897, 213. 72.

SACHS, Zur Kenntnis des Kreuzs-
pinnengiftes. Hofm. Beitr., ii.,
125 (1902). 236.

— Ueber Differenzen der Blutbe-
schaffenheit in verschied. Lebens-
altern. CentralU. f. Bakt.,
xxxiv., 686 (1903). 236.

— Constit. des Tetanusgift. JBerl.
klin. Woch., 1904, 16. 62.

SALKOWSKI, Ueber die Wirkung der
Antiseptica auf Toxine. Berl.
Uin. Woch., 1898, 545 (No. 25).
14, 81.

SALTER, The Elimination of Bac-
terial Toxines. Lancet, 1898,
i., 152. 16, 88.

SANARELLI, Defense de 1'organisme
contre les microbes. Ann. Past.,
yi., 225 (1892). 183.

— Etudes sur la fievre typhoide
experim. Ann. Past., viii., 193
(1894). 155.

SCLAVO, Ueber d. Bereitung des

Serums gegen den Milzbrand.

CentralU. J. Bakt., xviii., 744

(1895), 179.
SCHAFFER, Beitr. z. Frage d. Gono-

kokkentoxine, Fortsclir. d. Med. ,

1897, 813. 162.

SCHATTENFROH, Ueber die Wirkung

von Bakterien-Proteinen. Ze.it.

f. Hyg., xviii., 456 (1894). 176.
SCHENK, Ueb. Streptok.- Serum u.

Streptokokken - Toxine. Wien.

klin. Woch., 1897, 937. 164.
SCHIERBECK, Ueb. d. Einfluss der

C(X> auf das Wachstum der

Diphth.-B. Arch. f. Hyg.,

xxvii., 339 (1896). 74.
SCHNEIDEMUHL, Ueb. Botulismus

beim Menschen und die sog.

Geburtsparalyse bei Rindern.

CentralU. f. Bakt., xxiv., 619

(1898). 126.
SCROLL, Unters. lib. giftige Eiweiss-

korper bei Cholera asiatica. Arch.

f. Hyg., 1892, 172. 146.
SCHOLTZ, Beitr. z. Biologic d. Gono-

coccus. Arch.f. Dermatol. , xlix. ,

3(1899). 163.
SCHULTZE, Spinalpunktion u. Eins-

pritz. v. Antitoxinserum bei Tet.

traumaticus. Mitt. Grenzgeb. d.

Med. u. Chir. , v. , 169 (1900). 117.
SELANDER, La maladie infectieuse

des pores. Ann. Past., iv., 545

(1890). 182.
SELBERG, Beitr. z. Kenntn. d. Gift-

wirkung d. Schweineseuchenbak-

terien, £c. Diss., Berlin, 1896.

183.
SENG, Ueb. d. qualit. u. quantit.

Verhaltnisse der Eiweisskorper

im Diphtherieheilserum. Zeit. f.

Hyg., xxxi., 513 (1899). 100.
SEWALL, Exper. on the Preventive

Inoculation of Rattle - snake

Venom. Journ. of Phys., viii.,

203 (1887). 228.
SHAKESPEARE, Preliminary Report

of Exper. Researches concerning

the Infectious Nature of Trau-
matic Tetanus. CentralU. f.

Bakt., ii., 541 (1887). 113.
SHARP, The Action of the Products

of the Organism of Dipht. on the

Heart of the Frog. Journ. of

Anat. and Physiol., xxxi., 199

(1897). 85.
SIEBER, Ueb. d. Entgiftung der

Toxine durch die Superoxyde, &c.

Z. f. phys. Chem., xxxii., 573

(1901). 14, HO.

BIBLIOGRAPHY.

263

SIEBER - SCHOUMOWA, Les scrums

therapeutiques anticocciques.

Arch, des Sciences Biol., iv., 415

(1896). 165.
SILBERSCHMIDT, Contrib. a 1'etude de

la swine plague. Ann. Past., ix.,

65 (1895). 183.

SIMON, Ueber die Gifte der Strepto-
coccus. Centralbl. f. Bakt., xxxv.,

318(1904). 166.
SIROTININ, Die Uebertrag. von

Typhusbazillen auf Versuchstiere.

Zeit. f. Hyg., i., 465 (1886).

155.
SMIRNOW, Ueb. d. Behandlung der

Diphtherie. Berl. Idin. Woch.,

1894, 683; 1895, 645, 675.

83.

— Note sur la determin. du pouvoir
antitoxique du serum antidiph-
therique. Arch. d. Scienc. Biolog.
de St. Petersbourg, iv., No. 3
(1895). 100.

SMITH, The Relation of Dextrose to

the Production of Toxine. Journ.

of Exper. Med., iv., 373 (1899).

72.
SOBERNHEIM, Experim. Unters. ub.

Choleragift und Choleraschutz.

Zeit. f. Hyg., xiv., 485 (1893).

148.
SPRINGFELD, Ueb. d. gift. Wirkung

des Blutserums des Flussaals.

Diss., Greifswald, 1889. 240.
SPRONCK, Sur les conditions, dont

depend la production d. poisons*

dans les cultures diphtheriques.

Ann. Past., ix., 759 (1895).

70.

— Chauffage du serum antidiph-
therique. Ann. Past., xii., 695
(1898). 93, 94.

— Prepar. de la tox. dipht. Ann.

Past., xii., 701 (1898). 6,74-
STEJSKAL, Kritisch-experim. Unt.

iib. d. Herztod infolge von Diphth-

Toxin. Z. f. Idin. Med., xliv.,

367 (1902). 85.
STEPANOFF, Etudes sur la ricine et

1'antiricine. Ann. Past., x., 663

(1896). 191.
STEPHENS & MYERS, Schlangengift.

Proc. Path. Soc., Lancet, 1898,

i., 644. 230.

STILLMARK, Ueb. Ricin. Arb.
pharm. Inst. Dorpat, iii. (Stutt-
gart, Enke), p. 59 (1889). 185.

SYMANSKI, Sitzungsbericht. Deutsch.
med. Woch., 1901, 318. 106.

SZONTAGH & WELLMANN, Verglei-
chende chem. Unters. iib. d.
normale Pferdeserum und das
Diphtherieheilserum. Deutsch.
med. Woch., 1898, 421. 97.

TAUBER, Ein Beitr. z. Kenntnis d.

Tetanus. Wien. Idin. Woch.,

1898, 747. 122.
TCHISTOVITSCH, Etudes sur I'im-

munisation contre le serum d'an-

guille. Ann. Past., xiii., 406

(1899) 243.
TEISSIER & GUINARD, Effets des

toxines microbiennes. Arch. Med.

Exper., ix., 994 (1897). 25.
THOMAS, Boston Med. and Surg.

Journ., No. 4, et seq., 1898.

86.
TICHOMIROFF, Ueber die Fallg. v.

Toxalbuminen durch Nuklein-

saure. Z. f. physiol. Ch., xxi.,

90(1895). 188.
TIZZONI, Ueb. d. experim. Immunitat

gegen Tetanus. Festschrift fur

VIRCHOW. Berlin, 1892, iii., 29.

184.
TIZZONI & CATTANI, Sur le poison du

tetanos. Arch. Ital. de Biolog.,

xiv., 101 (1890). 103.
Sur les proprietes de 1'anti-

toxine du tetanos. Arch. Ital.

de Biolog., xvi., 394 (1891).

184.

— Ueber d. Eigenschaften d. Te-
tanus-Antitoxins. Centralbl. f.

Balct., ix., 685(1891). 124.
— Fernere Unters. iib. das Tetanus-
Antitoxin. Centralbl. f. Bakt.y

x., 33 (1891). 124.
*TIZZONI & CENTANNI, Real. Accad.

Bologna, 1900. 136.
VAN TOURENHOUT, Over de bereiding

van diphtheriegif . Utrecht, 1 895.

Abst. Centralbl. f. Bakt., xviii.,

295 (1895). 72.

264

BIBLIOGRAPHY.

U

URIARTE, He"morysine du bac. pest.

Soc. BioL, Mi., 254 (1904).

143.
USCHINSKY, Ueber eine eiweissfreie

Nahrlosung f. pathog. Bakt.

CentralU. f. Bakt., xiv., 316

(1893). 104.

— Les poisons de la diphtherie et
du cholera. Arch. d. Med. Exp.,
1893, 293. 6, 80.

— Ueber Diphtheriekulturen auf
eiweissfreier Nahrlosung. Cen-
tralU. f. Bakt., xxi., 146 (1897).
80.

VAILLARD, Sur I'immunit^ contre le

tetanos. Soc. BioL, xliii., 147

(1891). 108.
— Sur quelques points concernant

1'immumte contre le tetanos.

Ann. Past., vi., 224 (1892). 103.
VAILLARD & ROUGET, Ann. Past.,

vi., 385 (1892). 103.
VALENTIN, Ueb. d. Giftw. d. nordaf-

rik. Skorpiones. Z. f. BioL, xii.,

170(1876). 287.

— Einige Beobachtg. iib. d. Wirkg.
des Viperngiftes. Z. f. BioL,
xiii., 80 (1877). 210, 219.

VAUGHAN, The Intracellular Toxines
of some of the Pathogenic Bact.
Journ. Amer. Med. Assoc., 1903,
p. 828; Bioch. Centr., i., No.
1056. 157.

VAN DE VELDE, Beitr. z. Kenntnis
der antitox. Kraft d. antidiphth.
Serums. CentralU. f. Bakt., xxii.,

K97 Q&
t)t£t. ao.

— Etude s. 1. mecanisme de viru-
lence du staphylocoque pyogene.
La Cellule, x. 139.

— Contribution a 1'immunite des
lapins contre le staphylocoque.
Ann. Past., x., 580 (1896). 139.

VERHOGEN & BAERT, Premieres re-
cherches sur la nature du tetanos.
Baumgartens Jahresber., 1890,
198. 113.

VINCENZI, Ueb. antitoxische Eigen-

schaften der Galle tetanisierter

Tiere. Deutsch. med. Woch.,

1898, 534. 110.
VINCENT, Inoculation intracranielle

du bacille Eberth, &c. Soc. BioL,

lv., 1214(1903). 156.
VULPIUS, Ueb. einen Fall von

Wundstarrkrampf mit Tiervers.

Deutsch. med. Woch., 1893, 992

112.

W

WALL, On the Poisons of certain
Species of Indian Snakes. Proc.
Roy. Soc., xxxii., 333 (1881).
209.

*WARDEN & WADDELL, Non-bacillar
Nature of Abrus Poison. Cal-
cutta, 1884. 201.

WASHBOURN, Experiments with the
Pneumococcus. Journ. of Path.,
iii., 142; Baumg.Jahresb., 1895,
62. 160.

WASSERMANN, A., Unt. iib. Immun.
geg. Choi, asiatica. Zeit. f.
Hyg., xiv., 35 (1893). 147.

— Ueb. Concentrierung v. Antitoxin
aus Milch. Ztit. f. Hyg., xviii.,
236(1894). 96.

- Ueber die personliche Disposition
u. Prophylaxe gegen Diphtherie.
Zeit. f. Hyg., xix., 408 (1895).
19, 91.

— Unters. lib. einige theoret.
Punkte d. Immunitatslehre. Zeit.
f. Hyg., xxii., 263 (1896). 129.

- Weitere Mitteil. uber Seitenket-
tenimmunitat. Berl. klin. Woch.,
1898, 209. 17, 116.

— Gonokokkenkultur und Gonokok-
kengift. Berl. klin. Woch., 1897,
685. 161.

— Weitere Mitteilungen liber do.
Ze.it. f. Hyg., xxvii., 298 (1898).
161.

WASSERMANN & BRUCK, Ueber die
Wirkg. der Antit. Deutsch. med.
Woch., 1904, No. 21. 62.

WASSERMANN & PROSKAUER, Ueb.
die von d. Diphtheriebazillen
erzeugten Toxalbumine. Deutsch.
med. Woch., 1891, 585. 79.

BIBLIOGRAPHY.

265

WASSERMANN & TAKAKI, Ueber

tetanusantitoxischeEigenschaften

des Centralnervensystems. Berl.

klin. Woch., 1898, 5. 19, 116.
WEHRMANN, Sur les propr. toxiques

du sang, &c. Ann. Past., xi.,

810 (1897). 240.
WEICHARDT, Ueber Ermiidigungs-

toxine, I. and II. Munch, med.

Woch., 1904, No. 1, et seq.

245-
WEINGEROFF, Zur Kenntnis des

Hamolysins des Bac. pyocyan.

CentralU. f. Bakt., xxix., No. 20

(1901). 137.
WENDELSTADT, Einw. v. Glykogen

auf Hamolyt. Vorgange. Cen-
tralU. f. Bakt,, xxxiv., 831

(1903). 242.
WERHOFSKY, Beitr. z. pathol. Anat.

der Abrinvergiftung. Zieglers

Beitr. z. pathol. Anat., xvii., 115

(1895). 202.
WERNICKE, Ueb. Immun. - Vers. b. d.

Beulenpest. CentralU. f. Bakt.,

xxiv. (1898). 158.
WESBROOK, Contrib. a 1'etude d.

toxines du Cholera. Ann. Past.,

viii., 318 (1894). 146.
WILSON, Action of Scorpion Venom.

Journ. of Phys., xxxi. (1904).

237.
WLADIMIROFF, Antitoxinerzeug. d.

Tet.-Giftes. Zeit. f. Hyg., xv.,

405 (1893). 104.
WOLFENDEN, The Venom of the

Indian Cobra. Journ. of Phys.,

vii., 327 (1886). 212.

— The Venom of the Indian Viper
(Daboia). Journ. of Phys., vii.,
357.

WOOLDRIDGE, Note on the Protection
in Anthrax. Proc. Roy. Soc.,
xlii., 312 (1887). 177.

— Vers. it. Schutzimpfung auf
chemischem Wege. Du Bois
Arch., 1888, 527. 177.

ZINNO, Beit. z. Stud. d. Entstehung
d. Toxine. CentralU. f. Bakt.,
xxxi., 42(1902). 6.

267

INDEX.

ABRINE, 20, 24, 201-204, 229.

„ Anti-, 202, 203, 204, 231.
Abrus precatorius, 201.
Abscess, 189, 241.
"Acacia," 207.
Acetone, Influence of, 82.
Acids, Influence of, 81, 124, 200, 204.
Agglutination, 24, 192, 193, 196, 197,

203, 205, 206, 207, 223.
Agglutinines, 24, 59, 66, 173, 191,

193, 223, 227.
Agglutinophore, 193, 199.
Air, Influence of, on toxines, 127,

132, 150, 158, 167, 168.
Albumins, 84, 98, 101, 212, 226.
Albuminuria, 95.
Albumoses, 171, 178, 201, 212.
Alcohol, Influence of, 82, 105, 127,

132, 188, 213, 234.
Alexines, 239.

Alkalies, Influence of, 124, 140, 143.
Alligators, Susceptibility of, to tox-
ines, 18, 116,233.
Amboceptors, 25, 141, 221, 224, 225,

230, 239, 244.
,, Anti-, 230.

Ammonia, Influence of, 50, 51, 52,

53, 213.

Ammonium sulphate, 132.
Amphibia, Susceptibility of, 119.
Anaesthesia, 241.
Ancistrodon contortrix, 215, 223.

,, piscivorus, 222.

Androctonus occitanus, 237.
Anthrax poison, 67, 176-182.
Anti-bodies, Specific nature of, 2,

53, 133.

Antilysines, 133.
Antipyrine, 81.
Antiseptics, Influence of, 81.
Antitoxine units, 35, 36, 38.

Antitoxines, 4, 6, 15, 17, 28, 29,

30-62.

Antivenine, 230.
Aphonia, 126.
Arachnolysine, 236.

,, Anti-, 236, 237.

Aranea diademata, 235, 236.
Arterine, 191.

Ass, Susceptibility of, 111, 229.
a-toxines, 48.
Autolysis, Aseptic, 157.
Axolotl, 218.

B

BACILLUS of anthrax, 181.
,, botulinus, 126.
,, ,, Toxine of, 2.

,, coli communis, 157, 163.
,, enteritidis, 157.
,, of glanders, 175.
,, megatherium, 227.
,, of plague, 158.
,, prodigiosus, 172, 173.
,, pyocyaneus, 65, 128, 129,
130, 137, 163,
172.

Toxine of, 2.
,, subtilis, 104.
,, of symptomatic anthrax,

131.
,, tetani, 104.

tuberculosis, 166, 168, 170,

173.

,, of typhoid fever, 154.
Barbadoes seed, 185.
Barbel, Susceptibility of, 192.
Barbus fluviatilis, 192.
Bats, Immunity of, 242.

„ Susceptibility of, 119.
Bile, 226.

268

INDEX.

Bile, Action of, on toxines, 21, 110,

214, 232.

Birds, Susceptibility of, 235.
Birth-paralysis, 126.
Blood, 190, 202, 205, 206, 207, 210,
216, 223, 224, 242, 243, 244.
,, Toxic power of infected, 87,

88, 233.
Blood -corpuscles, 224, 225, 227, 228.

(red), 225.

,, ,, (red), Action of

toxines, &c., on,
133, 134, 137,
139, 140, 191,
192, 206, 236,
242, 243, 244.
,, -pressure, 85, 112, 156, 190,

205, 218, 241.

„ -vessels, 84, 161, 190,218,221.
Bombinator igneus, 233.
Borax, Influence of, 81.
Boric acid, Influence of, 55, 81.
Bothrops poison, 227.
Botulism antitoxine, 128.
toxine, 126-128.
Brain, Action of, on toxines, 116,

117, 118, 119, 127.
,, -substance, 215.
/3-toxines, 48.
Bufo cinereus, 223.
Bungarus coeruleus venom, 216.

,, poison, 227.
Butyric acid, 131.

CACHEXIA, 162.

Calcium chloride, Influence of, 213,

231, 238.

,, peroxide, Influence of, 14.
Camel, Susceptibility of, 235.
Carbon bisulphide, 106.
Cat, Susceptibility of, 111, 158, 205,

216, 235.

Cattle, Susceptibility of, 235.
Cephalin, 226.
Cerastes venom, 220.
Cerebrin, 128.
Chicken, Immunity of, 236.
Children, Susceptibility of, 92.
Chiracanthium nutrix, 236.
Chlorine, Influence of, 83, 213.

Chloroform, Influence of, 105, 132,

213.
Cholera, 66, 128, 129.

,, antitoxine, 63, 149, 150, 153,

154.

,, lysine, 142.
,, plasmine, 150, 151.
„ poison, 5, 6, 62, 63, 64, 65,

67, 145-153.
,, toxalbumin, 145.
,, vibrio, 65, 145, et seq.
Cholesterin, 128.

Influence of, 214, 226,

227.

Chromic acid, 217.
Coagulation, 191.
Cobra, 209, 210.

,, amboceptors, 226.
,, antitoxine, 228, 229.
,, haemolysine, 25, 222.
„ venom, 20, 61, 210, 211, 212,
215, 216, 217, 218, 219,
220, 221, 222, 223, 224,
225, 226, 227.
,, Anti-, 230.
Codeine, 57.
Coli antilysine, 138.
Colilysine, 133, 138.
Coma, 217.
Complement, 25, 141, 224, 225, 226,

239, 244.
Conger, 239.
Congestion, 87.
Conjunctiva, 202, 219.
Conjunctivitis, 191.
Constipation, 126.
Convulsions, 103, 163, 179, 190, 205,

218, 221, 235, 237, 238, 241.
Copper Sulphate, 213.
Copperhead snake (Ancistrodon)

poison, 214, 215, 219, 221.
Crocodiles, Susceptibility of, 18.
Crotalus adamanteus, 223.
,, antitoxine, 228.
,, durissus, 212.

venom, 215, 216, 219, 220,

221, 223, 227, 230.
Crotine, 205-207.

,, Anti-, 206.
Croton oil, 205.

,, tiglium, 205.

Culture media, 7, 71-76, 80, 91, 103,
104, 131, 146, 147, 161, 162, 163,
165, 167, 168, 175, 177, 179, 183.

INDEX.

269

Curare, 231.
Curcas purgans, 185.
Cytisine, 181.
Cytolysine, 221.

DABOIA Russeli, 221.

,, venom, 212, 221.
Deutero-albumose, 171.

,, toxines, 48, 135.
Diarrhoea, 25, 157, 161, 179.
Diastase, 10, 21, 110.
Diemenia antitoxine, 228.
Differential zones, 41.
Dilatation of blood-vessels, 84.
Diphtheria antitoxine, 28, 32, 33,

57, 61, 91-102.

„ toxine, 2, 3, 6, 13, 15, 20,
21, 23, 25, 32, 33, 35-
49, 54, 58, 59, 61, 67,
69-102, 180, 229.

Dog, Susceptibility of the, 122, 132,
138, 205, 216, 223, 225, 234, 235,
240, 241, 242.
Dysentery, 157.

,, poison, 157.
,, toxine, 184.

Dyspnoea, 126, 163, 177, 190, 205,
235.

E

EELS' blood, 32, 33, 195, 222, 234,

239-245.
Elaphin, 210.

Electricity, Influence of, 13, 105, 213.
Emaciation, 158, 161, 179, 190, 241.
Endocomplements, 224, 225.
Endothelium, 221, 228.
Endotoxines, 5, 62, 63, 65, 66, 67,

145, etseq., 182.
Enhydrina venom, 216, 218.
Epeira diadema, 235, 236.
Epitoxoids, 39.
Equivalent poisons, 108.
Ergophore groups, 24, 25, 29, 193-

199, 244.
Erythrocytes, 33, 133, 142, 191, 195,

196, 197, 206, 242, 243 (see also

Blood-corpuscles, red).

Essential oils, Influence of, 213.
Ether, Influence of, 82, 127, 132,

213, 226.
Euglobulin, 102.
Exophthalmus, 241.
Extracts of fungi, Influence of,

214.
Eye, Action of toxines on the, 86,

202, 205.

FATIGUE antitoxine, 245.
,, toxine, 245.

Fever, 25, 161, 235.

Fibrin, 191.

Fibrinoglobulin, 101.

Fire toad, 233.

Fish venoms, 238-239.

Fishes, Susceptibility of, 216.

Formaldehyde, 132.

Frogs, Immunity of, 132, 242.

„ Susceptibility of, 19, 23, 84,
119, 121, 205, 215, 216,
217, 218, 219, 223, 234,
235, 237, 238, 241.

G

GARDEN spider, 235.

toad, 233.

Gases, Influence of, 105.
Globulins, 78, 79, 97, 98, 100, 101,

210, 212.

Glycogen, Influence of, 242.
Goats, Susceptibility of, 87, 109,

140, 192, 225, 234, 242.
Gold chloride, Influence of, 213, 231,

238.
Gonococcus, 161, 162.

poison, 22, 161-163.
Gonorrhoea, 162.
Goose, Susceptibility of, 109, 138,

140.

Greater weever, 238.
Guinea-pig antiserum, 229.
,, Immunity of, 236.

270

INDEX.

Guinea-pig, Susceptibility of, 23, 36,
37, 87, 89, 90, 94, 95, 105, 108,
109, 111, 113, 116, 119, 130, 131,
132, 133, 138, 147, 149, 155, 156,
158, 159, 163, 167, 170, 172, 180,
183, 189, 214, 215, 216, 223, 225,
234, 238, 241, 242, 243.

H

BJEMAGGLUTININES, 143, 227, 228.

Hsematuria, 241, 242.

Haemoglobin, 132, 242.

Haemolysines, 32, 34, 133, et seq.,
206, 215, 220, 221,
222, 223, 227, 228,
229, 230.
Anti-, 230, 234, 238.

Haemolysis, 24, 50, 53, 133, et seq.,
192, 206, 222, 223, 224, 226, 227,

233, 234, 236, 238, 239, 240, 242,
243, 244.

Hemorrhage, 131, 150, 216, 218,

234, 242.

Hgemorrhagine, 215, 216, 219, 220,
221,227,228,229,
230.

,, Anti-, 230.

Hair, Action of toxines on the, 25,

158.

Haptines, 2, 5, 22, 23, 38, 41, 45, 50,
89, 99, 103, 133, 141, 193, 196, 199,
206, 212, 213, 214, 224, 244.
Haptophore Groups, 5, 23, 24, 25,
30, 37, 42, 44, 99, 108, 116, 118,
130, 141, 168, 180, 192, 197, 198,
199, 200.

Hay fever antitoxine, 208.
,, toxine, 207, 208.
Heart, Action of toxines on the, 25,
84, 85, 86, 128, 158, 161, 176, 190,
202, 205, 217, 218, 235, 238, 241.
Heat of combination of toxine with

antitoxine, 60, 61.
Hedgehog antiserum, 229.
,, Immunity of, 242.
,, Susceptibility of, 132, 216,

233, 241.

Hemitoxine formation, 47, 48.
Hen, 229.
, , Immunity of, 242.

Hen, Susceptibility of, to toxines,

18, 19, 22, 109, 110, 116, 118, 132,

156, 205, 215, 234, 236, 243.
Hepatitis, 86.
Hetero-albumose, 212.
Hog cholera, 182, 183.

,, ,, antitoxine, 183.
Holocephalus poison, 211.
Hoplocephalus curtus poison, 212,

215, 231.
Horse antiserum, 229, 238.

,, Immunity of, 236.

, , Susceptibility of, 87, 109, 1 1 1 ,

125, 138, 225, 235.
Hydrochloric acid, Influence of, 94,

124.
Hydrogen peroxide, Influence of, 14,

188.

Hydroxyl (OH-ions), 53.
Hypersemia, 86, 146, 205.
Hyperleucocytosis, 229.
Hypochlorite, Influence of, 83.
Hypoleucocytosis, 229.
Hypopyon, 162.
Hypothermia, 25, 84, 150, 167, 218.

ICHTHYOTOXINE, 239-244.
Immunity, 4, 16, 41, 66, 67, 89, 118,
' 119, 128, 132, 145, 147,
148, 159, 160, 161, 163,
165, 167, 170, 173, 174,
177, 178, 179, 180, 182,
183, 186, 193, 194, 197,
202, 203, 206, 207, 208,
220, 222, 228-233, 234,
235, 236, 238, 239, 243,
244.
,, units, 36, 37, 38, 42, 43,

45, 125.
Incubation period, 23, 110, 111, 115,

120, 121, 126, 127, 131, 136, 188.
Induction period, 52.
Inflammation, 162, 189, 202, 205,

216.
Intestines, Action of toxines on the,

25, 86, 242.
Invertase, 10, 151.
Iodine, Influence of, 105, 106, 213.
,, trichloride, Influence of,
105.

INDEX.

271

Iodine trichloride, Preservative
effects of, 77.

JATROPHA curcas, 185.
Jequirity, 201, 204.
Jerboa, 237.

K

KARAKURTE, 235.
Katipo, 235.
Keratitis, 162.

Kidneys, Action of the toxines on
the, 25, 86, 143, 169, 190, 242.

LACTIC acid, 131.

,, Influence of, 81, 105,

124.

Lamprey, 239.

Lathrodectes, 235.

Lecithides, 226, 227, 238.

Lecithin, 25, 128, 225, 236.

Lecithinophile group, 227.

Leeches, Susceptibility of, 235.

Lethal dose, 34, 35, 36, 37, 40, 41,
42, 45, 83, 84, 109, 111, 126, 130,
132, 134, 149, 150, 155, 158, 162,
163, 170, 172, 179, 184, 189, 194,
198, 201, 205, 207, 210, 215, 216,
219, 220, 221, 222, 232, 241.

Leucocidine, 139, 142, 143, 144, 227.
Anti-, 144.

Leucocytes, 143, 227, 229.

Leucocytosis, 190.

Leucolysines, 227.

Leucoma, 202.

Light, Influence of, 13, 44, 81, 93,
105, 127, 132, 150, 167.

Liver, Action of toxines, &c. , on the,

25, 158, 169, 190.
„ extract, 214.

Lizards, Susceptibility of, 216.

Loss of weight, see Emaciation.

Lungs, 175, 242.

Lymph glands, 28, 190.

Lymphocytes, 227.

Ly sines, 24, 50, 54, 66, 139, 142, 143.

MALIGNANT oedema, 183.

Malleine, 175, 176.

Malmignatte, 235.

Maltase, 10, 151.

Man, Susceptibility of, 109, 111, 120,
138, 140, 189, 206, 207, 217, 225,
235, 238, 241.

Marasmus, 158.

Marmots, Susceptibility of, to tox-
ines, 119.

Medulla oblongata, 190.

Mercuric chloride, 213.

Methaemoglobin, 222.

Milk, 226,

,, Antitoxine preparations with,
97, 124.

Mocassin snake poison, 219, 221.

Mongoose, 216, 233.

Monilia Candida, 11, 151.

Monkeys, Susceptibility of, 132, 155.

Mouse, Susceptibility of, to toxines,
18, 22, 87, 109, 111, 118, 119, 122,
125, 132, 134, 158, 162, 170, 178,
194, 201, 202.

Mucous membrane, 110, 162.

Muraena, 239.

Muscles, 245.

,, Influence of toxines, &c.,
on the, 113, 114,115,190,
191, 205, 217, 237, 241.

Myolysis, 85, 241.

N

NAIX, 210.

Naja egiziana venom, 218.

,, haje poison, 227.

,, tripudians, 209, 223.
Necrosis, 155, 158, 189, 190, 202,

205, 216, 220, 234, 241.
Necturus serum, 224.
Nephritis, 207.
Nerve substance, 214.
Nervous system, Influence of toxines,

&c., on the, 113, 114, 115, 116,

119, 127, 176, 205, 218, 219, 228,

241, 242.
Neurine, 3.
Neurotoxine, 215, 219, 220, 221, 227,

228, 229, 230, 237.
Anti-, 230.
Nucleic acid, 171, 174, 188, 195.

272

INDEX.

(EDEMA, 89, 131, 216, 220, 234.
Oxen, Immunity of, 236.

„ Susceptibility of, 132, 138,

205, 223, 225, 234.
Oxidising substances, Influence of,

127. '

Oxydase, Influence of, 110, 222.
Oxygen, 218.

Influence of, 14, 81, 104,
105, 206.

PANCREATIC juice, Action of, on

toxines, 21.

Panophthalmitis, 191, 194, 202.
Papaiin, 186, 222, 226.
Papayotin, 21.

Paralysis, 89, 126, 146, 150, 155, 156,
157, 190, 205, 218, 221, 234, 235,
238, 241.

Pelias berus, 210.
Pepsin, Action of, on toxines, 20,

110, 203, 222.

Perch, Susceptibility of, 206.
Peripheral neuritis, 150.
Peritonitis, 162.
Petromyzon, 238.
Peyer's patches, 190.
Phenol, 132, 165, 166, 213.

,, Preservative effects of, 77,

81, 105.

Phenylhydrazine, 133.
Phrynolysine, 233, 234.
Phytotoxines, 10, 185-308.
Pig, Susceptibility of, 205, 216, 223,

233.
Pigeon, Immunity of, 242.

,, Susceptibility of, 18, 132,

138, 156, 215, 234, 243.
Pike, Susceptibility of, 205.
Plague antitoxine, 159.
,, lysine, 143,
,, toxine, 158, 159.
Plasmines, 67, 170, 173.
Plotosus, 239.
Pneumobacillus, 172, 173.
Pneumococcus, 159, 160, 161.

,, antitoxine, 160.

,, lysine, 142.

„ poison, 67.

Pneumonia, 175.
Pneumotoxine, 159-161.
Poison glands, 210.
Potassium iodide, 213.

,, permanganate, Influence

of, 105, 132, 213, 217.
Precipitines, 29, 66, 67, 68, 173, 174.
Priapism, 235.
Protamines, 239.
Proteids, 78, 212.
Proteins, Poisonous non-specific, 3,

62, 65, 66, 67.
Proto-albumose, 212.
Protoplasm, 3, 6.
Prototoxines, 48, 135.
Protoxoids, 39, 40, 45, 47, 48, 58,

198, 204, 222.

Pseudechis porphyriacus poison, 212.
Pseudoglobulins, 102, 125.
Pterois, 239.
Ptomaines, 1, 2, 78.
Ptosis, 217.

Pulse, Acceleration of, 156.
Purging nut, 185.
Pus formers, 67, 68, 162.
Pyocyaneus antitoxine, 32, 63, 129.
poison, 29, 32, 64, 128-

130.
Pyocyanolysine, 130, 137, 138.

QUININE,

RABBIT antiserum, 229.

Rabbits, Susceptibility of, 19, 23,
61, 70, 79, 84, 87, 89, 91, 109, 111,
113, 116, 119, 120, 132, 134, 137,
138, 140, 142, 150, 155, 156, 157,
158, 159, 160, 165, 184, 189, 201,
205, 207, 210, 215, 216, 222, 223,
225, 229, 234, 235, 236, 238, 241,
242, 243.

Rat, Susceptibility of, 111, 158,205,
215, 216, 217, 235, 236.

Rattlesnake, 212.

Receptors, 17, 18, 50, 66, 68, 87, 112,
113, 116, 117, 118, 119, 120, 141,
173, 174, 192, 195, 196, 197, 198,
200, 233, 236, 239, 242, 243, 244.

INDEX.

273

Rennet, 107.

Respiration, Action of toxines, &c.,
on the, 85, 179, 205, 217, 218, 234,
235, 241.

Retention of urine, 126.
Ricine, 12, 20, 24, 30, 31, 33, 133,

185-200, 224, 229.
Anti-, 24, 31, 33, 193, 194,

195, 197, 198, 200, 201.
anti-body, 192.
Ricinus africanus, 185.
altissimus, 185.
borboniensis arboreus, 185.
brasiliensis, 185.
communis, 185.

,, major, 185.
guyanensis nanus, 185.
jamaicensis, 185.
philippinensis, 185.
sanguineus, 185.
spectabilis, 185.
Ring adder, Susceptibility of, 216.
Robine, 207.
Robinia pseudacacia, 207.

SALAMANDER poison, 234.
Salicyl aldehyde, 81.
Salivation, 241.
Saponine, 226.
Saturation units, 42, 43.
Scorpaena, 239.
Scorpio afer, 237.
Scorpion, Susceptibility of, 18.
,, venom, 229, 237-238.
Sepedon antitoxine, 228.
Serranus, 239.
Serum as a vehicle for antitoxines,

92, 97, 98.

,, Protective action of, 140.
,, Standard, 38.
Sheep, Immunity of, 236.

,, Susceptibility of, 87, 132,
138, 177, 205, 223, 225, 234.
Sieboldia maxima, 234.
Skin, Action of toxines on the, 25,

158.

Sleeplessness, 235.
Snake hsemolysine, 224.
,, sera, 224.
,, toxines, 209-228, 230.

Snake venom, 23, 24, 29, 32, 33, 60,

215, 219, 223.
,, ,, antitoxine, 28, 32, 228-

233.

Snakes, Susceptibility of, 216, 233.
Sodium glycocholate, Influence of,

214.
,, hydroxide, Influence of, 50,

51, 52, 53.
" Spectra" of poisons, 46-48, 56, 135,

136, 142

Spider venom, 234-236.
Spinal cord, Action of toxines, &c.,

on the, 114, 1J5, 127, 190.
Spleen, 190.

Staphylococcus pyogenes albus, 139.
,, ,, aureus, 139.

Staphylolysine, 133, 138-140, 141.

,, Anti-, 139, 140-141.

Streptococcus, 164.
Streptolysine, 142.
Streptotoxine, 163-166.
Strychnine, 231.
Suffocation, 179.
Suprarenal bodies, 190, 215, 242.
Sweating, 167, 176, 235.
Swine-plague poison, 183.
Symptomatic anthrax antitoxine,

132.
,, ,, toxine, 130-

133.

Synanceia, 239.
Syntoxoids, 39, 45, 47, 48, 198.

TARANTULA, 234, 235.

Tartaric acid, Influence of, 81, 105.

Temperature, Influence of, 12, 76, 81,
83, 93, 94, 105, 113, 124,
129, 131, 132, 133, 134,
138, 140, 141, 142, 143,
144, 149, 150, 155, 159,
160, 162, 163, 165, 167,
174, 179, 183, 184, 188,
200, 204, 205, 212, 214,
216, 219, 220, 222, 223,
225, 227, 230, 231, 234,
236, 238, 242, 244.
„ of body, 131, 155, 157, 158,
177, 179, 189, 205.
18

274

INDEX.

Tetanolysine, 9, 32, 50, 51, 52, 53,
55, 56, 57, 61, 102,
133-137, 140, 141.
Anti-, 135.
Tetanospasmine, 9, 14, 102, 108, 109,

112, 113, 134, 136.
Tetanus, 102-125.

antitoxin, 15, 16, 17, 28, 33,
61,96, 108, 109, 115, 116,
118, 120, 122, 123-125,
231.

toxine, 2, 3, 9, 13, 15, 16,
18, 19, 20, 21, 22, 23, 33,
61, 102-125, 180, 229.
Tetrodon, 239.

Thymus extract, Influence of, 106.
Tiger snake, 215.
Toad, 234, 242.

,, toxine, 233-234.
Toluene, Preservative effects of, 77.
Tortoise, Immunity of, 19, 242.
Toxalbumins, 1, 12, 78, 79, 154, 187,

212.

Toxic unit, 41, 42, 45, 169.
Toxines, 1, 2, 5, 6-25, 29-62, 188.
Toxoids, 33, 37, 38, 39, 40, 42, 43,
44, 57, 82, 88, 106, 108, 116, 132,
135, 136, 155, 159, 182, 188, 198,
199, 207, 221, 244.

Toxones, 38, 39, 40, 41, 42, 43-49, 58,
59, 89, 90, 91, 108, 132, 135, 199,
206.

Toxonoids, 58, 59.
Toxopeptone, 146.

Toxophore groups, 5, 23, 30, 31, 37,
44, 48, 83, 88, 119, 121, 135, 141,
193, 199.

Trachinus draco, 238.
viper, 238.

Tricresol, Preservative effects of, 77.
Trigonocephalus, 210.

,, venum, 220.

Trimeresurus poison, 227.
Tritotoxines, 48, 49, 135.
Tritotoxoids, 58.
Tropidonotus, 210.
Trypsin, 186, 187, 188, 200, 203.
,, Action of, on antitoxin es,

94.
,, ,, on toxines, 21,

110.
Tubercle poison, 166-168.

Tuberculine, 67, 166, 168-175.
Tuberculinic acid, 171, 172.
Tuberculosamine, 171, 174.
Tuberculosin, 172.
Tuberculosis, 66.

,, antitoxine, 170.

,, poison, 5, 67.

Tuberculothymic acid, 172.
Tuberkulol, 170.
Typhoid antitoxine, 157.

,, lysine, 142.

,, virus, 154-157.
Typhotoxine, 154.
Typhus, 66, 128.

,, poison, 5, 25, 65, 67.
Ty rosin, Influence of, 214.

U

ULCERS, 157.
Uniceptors, 28.
Urethritis, 162, 163.

VIBRIO, Metschnikoff's, 183, 184.

„ Naskin's, 23.
Viper hsemolysine, 224.

„ venom, 2 1 0, 211, 212, 214, 216,

217, 218, 219, 222, 234.
Vipera aspis poison, 213, 215, 218,

222.

,, redii, 217.
Vomiting, 217.

W

WEAKNESS, 217.

Worms, Susceptibility of, 216.

YEAST, 151.

Zootoxines, 209-245.
Zymase, 151.

BELL AND BAIN, LIMITED, 43 MITCHELL STREET, GLASGOW.

THIS BOOK IS DUE ON THE LAST DATE
STAMPED BELOW

AN INITIAL FINE OF 25 CENTS

WILL BE ASSESSED FOR FAILURE TO RETURN
THIS BOOK ON THE DATE DUE. THE PENALTY
WILL INCREASE TO SO CENTS ON THE FOURTH
DAY AND TO $1.OO ON THE SEVENTH DAY
OVERDUE.

0)0^0 fr^fTftv

iliiJ

/ oiaWY

f\i ' • V

MQV 5 1940

NOV 1 o iQAn

iiuf j^ y id*tu
M'iV t ^ tQ^-

AT JLO |134i
inni j j jj-f

JAN Is iybu

MAR 211951

^L«A/E./ &Sb

MffR 30 1951

- : '• - •

jiWY9% 7,185'

JUN.ZB'^gy-

MAY 1'58

^P24 58JW

LD 21-100m-8,'34

07

BfOLOGf
LIBRARY

UNIVERSITY OF CALIFORNIA LIBRARY