EXPERIMENTAL RESEARCHES

ON

SPECIFIC THERAPEUTICS

UNIVERSITY

OF

&i£QR**&

EXPE RIMENTAL RESEARCHES

SPECIFIC THERAPEUTICS

BY

PROF. PAUL EHRLICH, M.D., D.Sc.OxoN.
»«

Director of the Konigliches Institut fur Experimentelle Therapit,
Frankfort.

Ifcrlnris for 1907

OF
THE ROYAL INSTITUTE OF PUBLIC HEALTH

PAUL B. HOEBER

69 EAST 5QTH STREET

NEW YORK

Ess,

[Printed in England],

PREFACE.

IT was with very sincere pleasure that I
accepted the invitation of the Council of the
Royal Institute of Public Health with which
they honored me, to deliver the Harben Lec-
tures in 1907, for many relationships both of a
scientific and social character have always con-
tributed to render a visit to England one of
stimulus and recuperation.

The subjects which I chose for my lectures
are closely related with those questions, with
the study of which I have been connected during
the past ten years, viz., the theories of Immunity
and Atrepsy, which latter is connected with my
researches on Cancer, and the theory of
Chemico-therapeutics which I have enunciated
in connection with my work on the Trypano-
somes.

206460

vi PREFACE.

I am publishing these lectures with a feeling
of sincere gratitude for the kind and hospitable
reception which I received in London from my
friends and colleagues, and especially am I
anxious to express my indebtedness to Professor
William R. Smith for all his kindness, which
rendered my visit so enjoyable. I would also
desire to thank Dr. Carl Prausnitz for his very
valuable help, and Professor R. Tanner Hewlett
for kindly revising the English translation of
the Lectures.

P. EHRLICH.

FRANKFORT,

April 22, 1908.

EXPERIMENTAL RESEARCHES ON
SPECIFIC THERAPEUTICS.

LECTURE I.

ON IMMUNITY WITH ESPECIAL REFERENCE TO THE RELA-
TIONS EXISTING BETWEEN THE DISTRIBUTION AND
THE ACTION OF ANTIGENS.

THERE can be no doubt that the three great
fields of knowledge, Pharmacology, Toxicology
and Therapeutics, in their theoretical and prac-
tical aspects form the most important branches
of medicine. It is matter, therefore, for no
surprise that in the study of the various sub-
stances with which these sciences are concerned,
the mode of action and the reasons for such
call for much consideration, and theory and
speculation necessarily form a great part of our
study.

2 EXPERIMENTAL RESEARCHES

Besides, pharmacology has but just emerged
from the stage of pure observation and descrip-
tion. One was content formerly with describing
the physiological effects and the secondary action
of substances which act pharmaco-dynamically,
as well as the morphological changes which
they bring about in the organs and tissues
of the body. Observations made on an em-
pirical basis such as this, formed a mass of
most needful knowledge, and even to-day we
have no hesitation in admitting that the study
of the symptomatology of drugs is still a work
of absolute necessity and must yield very fruitful
results. Indeed, by such means we learn not
only how to make use of known drugs in a
purposeful manner, but also how to avoid their
undesired secondary actions. But merely to
increase the contents of our pharmacopoeia is
not to add to our resources in this desirable
direction : for such an increase may depend on
accidents, which, in their turn, may be the
outcome of empiricism. It is to the great
influence which chemistry exerts on medical
science that we owe the change in this state of
affairs ; for it is especially necessary to have
clear ideas of the relations between chemical
constitution and pharmacological action.

ON SPECIFIC THERAPEUTICS. 3

About the middle of the last century the
influence of these inquiries- made itself especi-
ally felt, and this influence is chiefly evident in
the mass of drugs with which the united efforts
of synthetic chemistry and pharmacology have
enriched us. But observers were content with
an advance in this direction, based on rational
grounds. They recognised a limited number of
atom-groupings, which were of importance either
for their therapeutic or their toxic action ; but
the drugs used were directed not against the
causes of disease but against the symptoms to
which these gave rise ; it was not the causes but
their effects which were combatted. Thera-
peutics were chiefly symptomatic, and so it is in
many cases to-day. Since the search after the
seat and cause of disease has, from the time of
Morgagni, and especially under the leadership
of Virchow's genius, influenced our entire field
of thought, the effect of these considerations has
become more evident in our treatment. The
features of an aetiological treatment, directed
against the causes and the seat of disease, were
not satisfactorily brought out by merely insisting
on the relationship which existed between the
constitution of drugs and their action ; the fact
was overlooked that between chemical consti-

B 2

4 EXPERIMENTAL RESEARCHES

tution and pharmacological action another and
important bond of union exists, which influences
the relations between the pharmaco-dynamic
agent and the substance on which it is intended
to act. This bond of union is the mode of
distribution, and represents the sum of the
peculiarities of the cells and tissues and of the
drug. In this we have to do with a principle
so obvious that it should at once be accepted as
an axiom, but even when accepted as an axiom
it is scarcely ever applied to the study of
practical questions. The reason for this, I
think, lies partly in a certain disinclination to
attempt to master the difficulties of the problem,
but mainly in the fact that, in view of the
triumphs of synthetic chemistry, per se, the
biological factor of the pharmaco-dynamic
action is somewhat lost sight of. I may
mention that from the beginning, as the result
of my studies on dyes, I have endeavoured to
point out the necessity of the study of localisa-
tion ; in a word, to give pharmacology an
aetiologically therapeutic tendency; regarding
the details of this I shall speak further in my
second lecture. Excepting the case of dyes,
which, by reason of their easily appreciable
properties allow their distribution in the organism

ON SPECIFIC THERAPEUTICS. 5

to be followed, the study of the laws which govern
distribution is exceedingly difficult. Besides,
owing to the great number of chemicals at our
disposal, a large amount of empirical work must
be carried out before we are able to find those
substances which in any given case will give the
desired mode of distribution. All the greater,
then, must be our admiration for the powers of
nature, in view of the fact that the living
organism, when it takes upon itself the pro-
duction of curative agents, does this in such a
manner as to form ideal aetiological remedies.
The protective substances of the blood, with
which this lecture is concerned, completely fulfil
the requirements of the case, and the study of
antigens and antibodies may form the basis of the
relationships which must exist between con-
stitution, distribution, and action, in order that
our treatment may be successful.

Since the conditions of substances treated of
in the study of immunity are specially clear and
matters of common knowledge, I shall begin the
discussion of my views regarding distribution
and localisation with the consideration of this
class of phenomena of pharmaco-dynamic and
toxicological action.

The discovery of antitoxins by von Behring,

6 EXPERIMENTAL RESEARCHES

fundamental in itself, has opened to pharmaco-
logy and therapeutics this new field in which the
principle of distribution is exemplified in an
ideal manner : for antitoxins and antibacterial
substances are, so to speak, charmed bullets
which strike only those objects for whose
destruction they have been produced by the
organism. I call these substances monotropic ;
the monotropism of these antibodies is charac-
terised by the fact that they are bacteriotropic or
generally speaking aetiotropic, i.e., they are di-
rected against bacteria or against those products
of their metabolism (toxins) which cause disease.
The definition of monotropism is, therefore, here
overlapped by that of specificity, which, in the
language of the study of immunity, is the cha-
racteristic of monotropic action. As the cause
of this specificity we must note, in my opinion,
only the effect of chemical relations which exist
between the agents of infection, or their products,
and the antibodies. From the very beginning
my standpoint has been, that all those sub-
stances which have the power to bring about
the creation of antibodies, I mean the antigens,
must be distinguished, as a matter of princi-
ple, from the other pharmaco-dynamical and
poisonous substances. That distinction I con-

ON SPECIFIC THERAPEUTICS. 7

sider to be of the greatest importance, and this
is further borne out by the fact that in spite of
the most strenuous endeavours we have been
unable to find any antigen of a known chemical
constitution.

I believe that the absorption of antigen-like
substances by the body is a phenomenon
which bears a close resemblance to the assimi-
lation of nutritive substances. In the case of
the other poisonous substances we find more
simple phenomena as causes of localisation, but,
in my opinion, the powers which influence the
distribution of toxins and kindred substances,
belong to the domain of chemical synthesis.
There has been a tendency of late to bring the
colloid nature of immune substances into the
foreground, and thus the impression is conveyed
that the whole subject of these phenomena
might be explained on the ground of the sub-
stances being colloid. Against this view it
seems to me to be necessary to insist upon the
fact that colloid nature and chemical reactive
power do not exclude one another ; for colloids
possess, just as other substances do, certain
groupings of atoms which render them capable
of reactions of a synthetic nature. Thus, one
may introduce into certain aromatic nuclei of

8 EXPERIMENTAL RESEARCHES

the protein molecule chemical atom-groups, e.g.,
the nitro-, the amido-, etc., or, on the other
hand, the reactive power of any such groups as
may be present can be inhibited, as by the
removal of the amid. I would here remind you
of the researches of Obermayer and Pick, which
show that by purely chemical substitution of
this kind one may also profoundly alter the
antigenic character of albuminous substances,
so that in place of specificity for a kind
(" Artspecifitat ") we have a specificity of
chemical constitution. I would remind you
also of the behaviour of certain colloid dyes —
for example, chromazon-red, which possesses all
the properties of the azo-dyes, but differs from
them in possessing an aldehyde group ; it is
thus able to react with hydrazin, and by union
therewith to be transformed into a hydrazon of
blue colour. Besides, I do not believe that for
the explanation of such reactions, especially of
their specific nature, a few analogies drawn from
certain phenomena of the hitherto obscure chem-
istry of colloids are sufficient ; and as a matter
of fact it seems to me that the ever-increasing
endeavour to build causal relationships on the
ground of purely formal analogies, is but little
fitted to advance our knowledge. The condi-

ON SPECIFIC THERAPEUTICS. 9

tion therefore necessary for such action is the
presence of two groups, whose chemical rela-
tionship is of the closest, and whose interaction
is therefore the condition of their union. This
axiom as to union is the basis of my side-chain
theory.

Let us consider in the first place the action of
comparatively simple toxins, which differ in the
phenomena of intoxication to which they give
rise ; the action of the diphtheria-toxin, for
example, is absolutely different from that of the
tetanus-toxin. In the case of toxins, too, we are
justified in assuming a connection between their
chemical constitution and action, just as in the
case of poisons of known chemical constitution ;
this connection can in many cases be proved.
As an instance of this latter I would note the
fact that in the cocaine series it is the residue
of benzoyl which causes their anaesthesiophoric
character ; that the soporific action of certain
disulphons is entirely due to the presence in
them of ethyl groups ; and that the dulcific
character of certain sweet substances, e.g.,
phenetidin-urea, is due to a like group. In the
case of the toxins there is a difficulty in our
way, due to the fact that, up till now, they have
not lent themselves to chemical analysis ; but

10 EXPERIMENTAL RESEARCHES

in their case, too, it has been found to be a
fruitful heuristic principle to formulate for them
similar relations between their constitution and
their action. Thus I designate that group of
the toxin molecule which is the cause of its
peculiar poisonous action as its toxophoric
group ; but the presence of this toxophoric
group is not of itself sufficient to bring about
the poisonous action — for whilst the guinea-pig
is exceedingly sensitive to the tetanus toxin,
the rabbit possesses a relative immunity. The
cause of such difference we must therefore
attribute to the distribution or localisation of
the toxin.

When the poison and the organs sensitive to
it do not come in contact, or when sensitiveness
of the organs does not exist, the action remains
absent.

If we assume those peculiarities of the toxins
which cause their distribution to be localised in
a special group of the toxin molecule, and the
power of the organs and tissues to react with
the toxin to be localised in special atom groups
of the protoplasm, we arrive at the basis of my
side-chain theory. The distributive group of
the toxins I call the " haptophoric group," and
the corresponding chemical organs of the proto-

ON SPECIFIC THERAPEUTICS. 11

plasm the " receptors." The relations between
the receptors and the haptophoric groups repre-
sent the conditions under which distribution takes
place. The toxic action can occur only when
receptors fitted to anchor the toxins are present.

The existence of receptors can be proved
experimentally. For, if to a solution of toxin
there be added receptors suitable for anchoring
the toxin, the solution becomes non-poisonous ;
this " binding experiment" plays a great part
in the study of immunity, since Wassermann,
working on the basis of the conclusions to be
drawn from the side-chain theory, first showed
that those cells of the central nervous system,
which were known to be affected by the action
of tetanus-toxin, anchored the toxin.

In order that the poisonous action may take
place, we must presume not only the presence
of receptors, but that they should be present in
positions favourable to the toxin action being
brought about. When both the receptors and
the organs sensitive to the action of the poison
are present, the conditions for infection are
naturally most favourable ; then distribution
takes place at once, the poison circulating until
it reaches the sensitive cell. The action of the
tetanus-toxin, in the case of the guinea-pig, is

12 EXPERIMENTAL RESEARCHES

monotropic, for receptors for the tetanus spasmin
are present only in the central nervous system ;
on the other hand, in the rabbit, suitable
receptors are present not only in the central
nervous system, but also in other organs and
in the connective tissue, and the type of distri-
bution in this animal is more complex, and
depends upon the point of entrance of the
infection or injection. Thus the varying sensi-
tiveness of different species to the same toxin
may be explained. The localisation of the
receptors is, then, of great moment for the
distribution of toxin in the organism, and con-
sequently for its sensitiveness to poisonous
action. One may therefore, in general, dis-
tinguish four different types of distribution :

(1) In which fitting receptors are not present,
the animal possessing natural immunity (the
formation of antibodies cannot take place).

(2) In which receptors are present, but only
in those organs on which the poison does not
act, or in organs of lesser importance ; here,
again, the animal possesses natural immunity
(antibodies may be formed, and immunisation
is easily carried out).

(3) In which the receptors are distributed
over various parts of the organism, and are

ON SPECIFIC THERAPEUTICS. 13

present in the organs which are sensitive to the
action of the poison. Here a relative immunity
exists, and the possibility of infection depends
chiefly upon the manner in which the poison is
introduced (the conditions for antibody forma-
tion are present, and those for immunisation
are more or less favourable).

(4) In which the receptors are present only
in the organs which are sensitive to the poison ;
in this case the organism is exceedingly sensitive
to the poison (antibody formation is possible ;
but immunisation is difficult to carry out, and
must be begun with small doses, or with
weakened poisons).

I think that this systematic division clearly
shows the possibilities of distribution and
localisation, as well as the conditions under
which a poisonous action can take place, and
what must be the aim of antitoxic treatment,
viz., the alteration of the natural conditions of
distribution and the interference with the action
of the poison.

In my view, which is based on extensive
experience in experimental research, the anti-
toxins are purely and simply receptors fitted for
union with the poisons which have entered the
circulation. When, for therapeutic or prophy-

14 EXPERIMENTAL RESEARCHES

lactic purposes we inject an antitoxic serum,
the number of receptors in the organism fitted
for union with the poison is increased, the
quantity of receptors introduced being charac-
terised by the fact that, as antitoxins, they
represent dissolved cell contituents, which, by
their union with the toxin, can do no harm, but
by their presence in considerable quantity bring
about a marked change in the conditions of
distribution in the organism. The receptors of
the organs which are sensitive to the action
of the poison are reinforced by a large number
of free antitoxin receptors, which, by reason of
the factor of distribution, and also because they
take up the toxin in the juices of the organism,
bring about this condition of affairs. The toxin
is thus either kept away from the organs which
it threatens to attack, or it comes in contact
with them only in an inconsiderable amount,
and so the disease runs a favourable course.
Of course the toxin, which has already become
united to the cell, and has exerted a deleterious
influence upon it, cannot be rendered innocuous ;
for we know that in all reactions of a similar
nature observed in the study of immunity,
primary union is followed by a stage of
secondary consolidation, a stage in which, even

ON SPECIFIC THERAPEUTICS. 15

by the addition of a large number of receptors
of the needed description, we cannot release
the toxin from the union. We see, then, the
limitations of antitoxic treatment, which is, in
the true sense of the term, a distributive treat-
ment ; the antitoxins are really specific drugs
by using which we run no risk of harming the
component parts of the organism, but which,
thanks to their specific monotropism, influence
the toxic agent alone. The relations which exist
between constitution and action are therefore in
no sense influenced by the action of the anti-
toxins. The toxin molecule, which is anchored
by the antitoxin, still possesses a toxophoric
group, which, if it could but obtain a suitable
localisation, would immediately develop its
characteristic action.

I have always held the view that the anti-
toxins do not in any way destroy the toxin, but
that they merely limit its sphere of action by
combining with it. If further proof of the
correctness of this view be required, we have it
in the researches carried out by Morgenroth,
who showed that in suitable cases the toxin
may be entirely regained from a perfectly neu-
tral union of toxin and antitoxin, just as glu-
cosides may by suitable treatment be resolved

16 EXPERIMENTAL RESEARCHES

into their two components. The antitoxin,
then, exercises its curative influence merely by
anchoring the distributive group of the toxins.

When we speak of monotropism or specificity
in the case of antigens and antibodies, we of
course mean only the chemical relations between
haptophoric groups and receptors. For example,
toxins may be specific and yet act upon the cells
of all species of animals, if the toxin-anchoring
receptors be widespread amongst them. The
other extreme we find in those cases in which
the organism itself reacts to elements intro-
duced into it and forms substances, antibodies,
which act on the corresponding antigens ; we
are here dealing with antibodies which are
specifically monotropic, just as the antitoxins
are, and which affect only those substances to
which they owe their origin. The only difference
is that the antitoxins act by localisation alone —
when they have anchored the toxins their work
is done. The other antibodies, it is true, act at
first in a similar way on the substances sensitive
to them ; but they have a further action on the
anchored substances, an action which is either
direct, as in those cases (agglutinins and pre-
cipitins) in which, like the toxins, they have
special ergophoric groups, or indirect, in that

ON SPECIFIC THERAPEUTICS. 17

the union is merely a preliminary to their further
action on their prey. Thus one class of these
antibodies has the power of rendering the cells
assimilable by phagocytes (opsonins, bacterio-
tropic substances) ; another class (amboceptors)
has the power of rendering the cells liable to
the action of toxin-like constituents of the blood
serum (complements). In the latter case by
the simultaneous action of two substances a
destructive effect is produced. These sub-
stances are also called cytotoxins. For the
study of these the way has been prepared by
the work of Pfeiffer, Metchnikoff, and Bordet,
and by the discovery of Metchnikoff and
Bordet that haemolysins are produced by im-
munisation against blood corpuscles.

These haemolysins are of special importance in
considering the question of the relations between
constitution, distribution, and action, because,
in their case, the haptophoric and toxophoric
groups are distinct, distribution and toxic
effect being dependent upon two different sub-
stances, the more stable amboceptor controlling
the distribution, and the labile complement
the toxic effect. The complement has no
direct relations with the cell, on which it acts
only through the medium of the amboceptor,

c

18 EXPERIMENTAL RESEARCHES

It is a normal constituent of the blood-serum,
and its quantity undergoes no change as a re-
sult of the process of immunisation. On the
other hand the amboceptor is a new formation
brought about by immunisation, and, on the
ground of the principle already admitted, it
must appear probable that the amboceptor, like
the antitoxins, possesses a marked monotropism
for the corresponding antigen. It was there-
fore not the result of accident, but of logical
sequence, that my first researches, carried out
with Morgenroth, on the mechanism of haemo-
lysis led us to the fundamental conclusion
that the amboceptor alone stands in direct
relation to the cell, and is quantitatively
anchored by it. This anchoring of the ambo-
ceptor takes place with a maximum of chemical
energy — it occurs even at o° C.

The union of the amboceptor and the cell has
no harmful results on the latter. On the other
hand, the amboceptor-laden cell is exposed to
the action of the complement, which by itself is
harmless. As regards complements, what I
have said about toxins holds true ; they may be
regarded as toxin-like substances which possess
a haptophoric group, and a toxophoric group
which I call the " zymotoxic " group. That

ON SPECIFIC THERAPEUTICS. 19

these groups are independent of each other, is
well seen in the case of modified complement—
"complementoid," as numerous test-tube experi-
ments have proved. Of special interest are the
conditions of their distribution. The real state
of affairs, which has been thoroughly investi-
gated, especially in the case of haemolysins, is
this : the intact erythrocytes do not unite with
the complement, which, however, is anchored
by the complex of erythrocyte and ambo-
ceptor. A closer acquaintance with the condi-
tions which govern distribution in this matter,
cannot be arrived at by the hypothesis that
the erythrocyte is sensitised by the amboceptor
in such manner that an action of the com-
plement is rendered possible. If one accepts
the theory of Bordet, which really consists of
the denial of the existence of direct relations
between amboceptor and complement, one enters
the realm of pure speculation ; for one must
then presume new affinities between the ery-
throcyte and complement to arise under the in-
fluence of the amboceptor. For this assump-
tion we have no grounds. Bordet's method of
proof must, therefore, limit itself to indirect
conclusions, and consists merely of objections
to the view held by Morgenroth and myself,

c 2

20 EXPERIMENTAL RESEARCHES

that the amboceptor and the complement stand
in direct relationship to one another. As a
matter of fact, from all sides proofs of the
existence of this direct relationship have been
advanced, and I think that the great majority
of my colleagues to-day accept my view, which
is known as the amboceptor theory. It is
true, as we have from the beginning insisted,
that the distributive relation of the complement
in the presence of the amboceptor is not that of
maximum chemical affinity ; indeed, we have,
on the contrary, as a rule an exceedingly loose
relationship which perhaps corresponds to a
reversible reaction. To show that this relation
is purposeful, we need only the following proof
— amboceptors are already present in large
quantity and of various kinds in the blood-
serum of normal animals. What, then, would
happen if the entire mass of normal ambo-
ceptors reacted with pronounced avidity with
the complement ? Obviously the entire mass
of complement would be anchored by the com-
plementophile groups of the amboceptors, and
there would be no free complement present in
the living body. The grave results of such a
state of affairs are evident ; as soon as the
necessity for the action of complements with a

ON SPECIFIC THERAPEUTICS. 21

special kind of amboceptor arose, there would be
no complement available, all having previously
been used up for the action of indifferent ambo-
ceptors. It is thus owing to the fact that the
complement is free or only loosely united to the
amboceptors in the circulation, that at a given
moment it is ready for use. The maximum
stimulus to action is rendered possible by the
anchoring of the amboceptor to the erythrocyte,
the avidity of the former toward the complement
being thus increased. This increase of avidity
which consists in the chemical affinity of the
complementophile group for the complement
being carried to its maximum, represents the
gist of our knowledge of the action of the
amboceptor.

The amboceptor, therefore, exercises the
important function of bringing about a specific
modification of those conditions existing in the
organism which determine the distribution of
complements, and which otherwise are not very
evident. It causes the complements to become
monotropic by its union with the given sub-
stance. The complements are thus localised
by amboceptors which have previously become
united to the substance — cell or otherwise. At
the same time this action represents a purposive

22 EXPERIMENTAL RESEARCHES

saving ; if the complement were already a con-
stituent part of the amboceptor, then — as the
complement is easily destroyed — there would
often be no action. The purposive saving is
evident from the fact that only in case of need
the amboceptor becomes able to combine with
the complement, and from the fact that com-
plementoids which, by reason of having lost the
zymotoxic group, become incapable of action,
have at the same time lost some of their avidity
for the amboceptor. By this change in the dis-
tributive quality of the complement that is
associated with the formation of complementoid,
the sphere of action of the amboceptor is con-
siderably extended. Of other possible influences
which may govern distribution and action we
have an indication given by the work of Ferrata.
Under Morgenroth's direction he carried out a
research, from which it appears that a comple-
ment is not a single body, one and indivisible ;
for in a salt-free medium it is split up into two
components, which are capable of action only
when they work in concert in a salt solution.
As to the intimate relations of these two com-
ponents, the researches which were carried out
at the suggestion of Sachs by Dr. Brand appear
to indicate that in the blood serum they are
always united.

ON SPECIFIC THERAPEUTICS. 23

The increase of avidity, which forms the
basis of the mode of action of the amboceptor,
not only governs the phenomena which occur in
the organism, but, since it lends itself to test-
tube experiments, also opens up a wide field for
serum diagnosis. If we remember that the
amboceptors, by their union with the sensitive
substances — cells or dissolved bodies — exert
such a marked localising influence on the
complements, it is evident that in a mixture
of an amboceptor and its corresponding antigen,
if the presence of the one constituent of the
mixture be known, that of the other can be
proved by the occurrence of the phenomenon of
complement-fixation.

This method, which was elaborated by Neisser
and Sachs on the basis of the work of Bordet,
Gengou and Moreschi, for the medico-legal
test of the source of blood by " complement
deviation/' depends solely on the principle of
this increase of avidity. And, thanks to the
genius of Wassermann, we possess a similar
method of sero-diagnosis for some infectious
diseases, the cause of which is at present
unknown or invisible — a method which has
yielded valuable results and gives great promise
for the future.

24 EXPERIMENTAL RESEARCHES

In practice, too, we have a further advantage
in the fact that the increase of avidity affects
not one complement alone, but, as a rule, all the
complements circulating in the blood. For the
complement-deviation test we may then choose
at will a complement which exercises any special
function, and, merely on practical grounds, we
choose hsemolytic complements. The peculiar
power of the amboceptor to fix a large number
of complements, is not to be wondered at
in view of the biological function of the ambo-
ceptor. This I hold to be, under physiological
conditions, that of seizing upon and elaborating
nutritive substances. By its cytophile group
the amboceptor is enabled to combine with
substances of the most varied kinds, provided
that they possess fitting receptors. We have,
then, merely an increase of purposive function
in the fact that the amboceptor is furnished
with a host of complementophile groups which
enable the most varied kinds of complement to
act — it may be simultaneously. This is the
consequence of the multiceptive nature of the
amboceptor.

As regards increase of avidity, the fact that
the anchoring of the amboceptor to the cell
causes the avidity for the complement to be

ON SPECIFIC THERAPEUTICS. 25

increased, in no way precludes an increase of
the avidity of the amboceptor for the cell taking
place in consequence of the anchoring of the
complement. Besides a few normal amboceptors
of the blood-serum in which this is observed,
we find that it occurs in a marked degree in the
case of those poisons which act by the united
action of two components, the part of the comple-
ment being played by lecithin. Thus, the power
of forming lecithids has been studied by Kyes,
who proved its existence in the case of snake
and scorpion poisons, and by Morgenroth and
Carpi, who studied bee poison. The action of the
venom in these cases corresponds to that of an
amboceptor, and the conditions for more careful
analysis are very favourable ; for snake poison
and lecithin are both stable substances, and the
latter belongs to a class whose chemical com-
position is known. Thus Kyes was enabled to
isolate and analyse the substance produced by
the reaction of snake poison with lecithin.
Cobra-lecithid was found to be a mono-stearyl
lecithin, and represents the result of a typical
series of processes which certainly take place in
cell life and in many phases of immunity. These
processes occur in consequence of the union
of albuminous substances, possessing special

28 EXPERIMENTAL RESEARCHES

actions, with lecithin or other fatty substances,
the resulting compound showing special charac-
teristics. Thus, cobra-lecithid is, like cobra
poison, soluble in water, and like lecithin,
soluble in fat solvents; in its physical properties,
therefore, it is a link between albumins and
lipoid substances, and is therefore especially
suited for distribution.

It is a special characteristic of the formation
of lecithids that out of two absolutely non-
poisonous substances a powerful poison is
formed, which, in its biological and physico-
chemical reactions, differs markedly from the
two substances from which it sprang, and does
not contain lecithin as such, one fatty-acid
residue being removed during the process of
formation. I believe that this possibility of the
transformation of certain substances, inactive
in themselves, from albuminoids to lipoids of
powerful action — a transformation which has
been proved to occur in the case of snake
poisons — plays a prominent part in the pro-
cesses of the living body, and is of importance
for distribution and action. I would again in-
sist upon the fact that in lecithid-formation there
results, from the reaction of two non-poisonous
substances upon one another, a product which

ON SPECIFIC THERAPEUTICS. 27

behaves absolutely differently from these. It
is not unnecessary to lay special stress upon
this point, for repeated attempts have been
made recently by Michaelis and Rona to draw
a parallel between the process of lecithid-for-
mation and certain phenomena of colloid-
absorption ; however, the tertium comparationis
of such conclusions consists solely of analogies
of the most formal kind, and the above-sketched
important fundamental fact is absolutely un-
touched. The study of the lecithids from the
chemical point of view has shown that in them
we have a field in which the phenomena of
immunity and those of pure chemistry meet,
and the hope can therefore be cherished that
we may thus obtain a nearer insight into the
chemical peculiarities of antigens and anti-
bodies. Already the large mass of facts col-
lected shows how fruitful the investigation into
chemical structure, and the questions which
arise therefrom as to distribution, are for the
analysis of immune substances and their re-
actions.

As to the biological factor in the formation of
antibodies, it has been shown, by the recent
investigations of Pfeiffer and Wassermann, that
a part of the process takes place in the obscure

28 EXPERIMENTAL RESEARCHES

province of the physiology of stimulation.
Light, however, has been thrown on the sub-
ject of the specificity of reactions by the view I
have expressed that the receptors of the cell-
protoplasm are the seat of the process, and that
their regeneration and elimination are the con-
sequences of this. Supposing these processes
to occur in the normal organism, and to be
merely intensified in the case of active immuni-
sation, it is possible to understand that antibodies
of the most varied kinds may exist in the serum
of the normal organism, and also to comprehend
the processes of immunity by viewing them from
the standpoint of the physiology of nutrition and
metabolism. Thus I am glad to be able, by my
conception of the problem, to come into entire
agreement with Metchnikoff. The immense
number of exceedingly various substances which
possess haptine characters in the blood serum,
substances of whose existence at one time one
did not even dream, is thus to be explained as
an expression of a many sided and differentiated
action of the most varied organs. One may
already by simple means differentiate the mul-
titude of serum substances into antitoxins,
amboceptors, agglutinins, precipitins, opsonins,
complements, ferments, antiferments, etc., but a

ON SPECIFIC THERAPEUTICS. 29

deeper study of the subject shows that each of
these divisions, in its turn, consists of a multi-
tude of functionally different components, and
thus a pluralistic view of the observed pheno-
mena is the only justifiable one. Although in
some quarters the endeavour is still made to
reduce everything to the most simple form, I
believe that such a rudimentary way of looking
at things is not justified by the appreciably
complex character of natural phenomena and
vital processes ; for we see, in the investigation
of immunity, that when earlier opinions, based on
experiment, must be replaced by newer ones, the
process takes place always by the substitution
of a complex conception in place of a simple
one; I would remind you of Buchner's idea
regarding alexin, which had to give way to the
proved fact that all cytotoxins have a complex
nature ; and also I would remind you of the
anti-complements, which we had supposed to be
simple bodies, while it is now known that anti-
complementary action is, as a rule, the result of
the concerted action of two substances. In the
case of the complements, too, in the light of
Ferrata's researches, we must assume that the
conception of these as simple substances is
wrong. The more readily, then, may we

30 EXPERIMENTAL RESEARCHES

assume the existence of a multitude of substances
as the cause of the various actions which are
exerted by one and the same blood serum, a
multitude whose existence has in numerous
cases been proved. Objections which have been
raised by several observers, especially by Bordet,
against our method of proof, and which consist
in stating that in every experiment the substance
(whose unity is assumed) has been injuriously
affected, and in attributing any difference to the
varying degree of sensitiveness of the various
test-substances alone, cannot be justified ; for,
if one takes the trouble to work — as I have
always urged that one should do — quantitatively,
such sources of error are immediately excluded
from our conclusions. And even in spite of
these, especially in the case of the proof of the
multitude of complements, in many cases one
has been able, by employing means of the most
varied kind, to obtain either the loss of a certain
function, or an absolutely disproportionate
change of degree in isolated functions. In the
plurality of haptines present in the serum we
have a wide field open for more profound
observation of the mechanism of receptor-
metabolism, of the laws governing variations,
and the influences which bring these about, a

ON SPECIFIC THERAPEUTICS. 31

field from which new light will be shed on
human pathology and clinical medicine. Suc-
cessful work in this field must proceed on a
broad basis. One would have first to make
exact observations regarding a large number of
functions of human blood serum, and then one
would have to investigate systematically in
cases of all sorts of diseases, anomalies of
nutrition, etc., the causes of departure from the
normal, as to whether these result from the
failure of certain functions, or from the existence
of new functions acting under pathological con-
ditions. Thus, without doubt, would be detected
differences in the sum of the functions of the
cell, and this section of the physiology and
pathology of the blood might be named the
blood canon.

I firmly believe that by extensive research we
shall find that there exist great differences, the
result of biological laws, and these will permit
us to come to correct conclusions as to the
origin of certain substances in the cell, and to
apply these conclusions to diagnosis and thera-
peutics. Of course, the united action of many
observers in many institutes, and the closest
relations between clinical and laboratory work,
are needful in order that progress may be made

32 EXPERIMENTAL RESEARCHES

in the direction indicated. The recent work of
Wassermann, who by means of complement-
fixation, was able to prove the existence in
the blood serum of anti-bodies for certain
nutritive substances (glycogen, albumoses,
peptones, etc.), and to obtain an increased con-
centration of these by increased doses of the
nutritive substances, appears to me to be very
promising. In the case of pathogenetic or patho-
gnomonic questions it does not appear to be
of use to seek for those haptines which may be
present in the blood as the result of immunisa-
tion ; for in their case one would either find
differences that are but slightly marked, or, if
one found sufficiently-marked differences, one
would have to be very careful in drawing con-
clusions from them. It would be better, there-
fore, to avoid using as test objects those sub-
stances which are already present in the normal
body, or gain entrance to it by infection. In
order to be able to draw absolutely correct con-
clusions as to the relations existing between
certain substances in the serum and the normal
or pathological activity of organs, one ought to
choose cells or other elements, regarding which
one may assume that they never come into
relation with the human body in a natural way.

ON SPECIFIC THERAPEUTICS. 33

We have a small beginning in this direction,
I think, in the research which I carried out
with Wechsberg ; we compared the behaviour
of human blood-serum towards the trypanoso-
mata in the case of healthy and diseased indi-
viduals, and we found that in cases of liver
disease the amount of trypanosomicidal sub-
stances in the serum is markedly decreased, as
will be seen from the subsequent table.

After Laveran had by his researches shown
that the serum of man and of a few species of
monkeys had a trypanosomicidal influence which
was not possessed by the sera of other animals,
it appeared worth while to pursue these re-
searches further, and to study this action quan-
titatively, and when influenced by various
conditions. For this purpose we employed a
strain of the trypanosoma of mat de caderas, two
or three control animals being also employed in
each case, which succumbed at the latest on
the fifth, and generally on the fourth day. The
animals were injected with equal quantities in
a similar manner, and as soon as the parasites
appeared in the blood (on the second day) the
curative serum was injected. By this means
we obtained a titration of a number of human
sera, regarding which I shall only give the

D

34 EXPERIMENTAL RESEARCHES

following particulars as to the results obtained
with small doses.

As Laveran had shown, the normal serum
caused the parasites to disappear, but, after
a varying period, they again made their appear-
ance and generally soon afterwards caused
death. In cases in which ^ to T^- c.c. of serum
was injected, the results in the majority of sera
tested were extraordinarily constant, in the
other cases death occurred between the tenth
and fourteenth day.

The cases were as follows : —

Normal Serum I., death on the twelfth day.

Normal Serum II., death on the twelfth day.

Diabetes mellitus, death on the twelfth day.

Acromegaly, presumably between the tenth
and eleventh day (0*0625 c.c. caused death on
the ninth day).

Concretio cordis, death on the eleventh day.

Colica saturnina, death on the eleventh day.

Polycythasmia rubra, death on the twelfth
day.

Carcinoma of stomach with metastases in the
liver, death on the thirteenth day.

Peritoneal tuberculosis, death on the thir-
teenth day.

Leucaemia myelogenica, death on the four-
teenth day.

ON SPECIFIC THERAPEUTICS. 35

In the cases of liver affections, on the other
hand, the following results were obtained : —

(1) Carcinoma of the bile ducts with complete
bile stasis: 0*5 — ro c.c. serum, death on the
fifth day ; 2*0 c.c. serum, death on the eleventh
day.

(2) Carcinoma of the stomach, with jaun-
dice, 0*1 — 0*25 c.c. serum, death on the fifth
day.

(3) Alcoholic cirrhosis of the liver, 0*125 c.c.
serum, death on the sixth day.

(4) Biliary cirrhosis with jaundice, 0*125 c.c.
serum, death on the seventh day.

(5) Alcoholic cirrhosis of the liver, 0*125 c.c.
serum, death on the eighth day.

(6) Biliary cirrhosis with jaundice, 0*125 c.c.
serum, death on the thirteenth day.

Thus there was observed in every instance,
with the exception of case 6, a distinct, though
varying, deficiency in the trypanosomicidal pro-
perties of the serum, and that this deficiency
was not merely an apparent one, produced by
the presence of inhibiting substances, e.g., the
products of bile stasis, was shown by experi-
ments in which a mixture of equal parts of the
serum from case i and from the acromegaly
case were injected : 0*5 c.c. of the mixture,

D2

86 EXPERIMENTAL RESEARCHES

death on the sixteenth day ; 0-25 c.c. of the
mixture, death on the tenth day.

Naturally one can only expect to find such
marked differences in the case of those sub-
stances which owe their origin to a certain
organ or combination of organs. If the place
of origin of other haptines is widely diffused
throughout the connective tissue of the organ-
ism, one can scarcely hope in such a case to
achieve much of value for diagnostic purposes.
An extensive organisation, which does not con-
fine itself to investigation of a single haptine-
substance, or indeed to a few such, is therefore
necessary, in order that more exact analyses
may be made, and results of a practical value
be obtained.

Looking back upon what I have said, you will
see that it is the principle of distribution which
governs the processes resulting in active im-
munisation. The antibodies — the protective
substances in the serum — all possess the power
of reacting, with maximum chemical energy,
with their corresponding antigen, as, e.g., in the
anchoring of bacterial cells. This anchoring is
a necessary preliminary for further reaction,
which may be of the most varied kind.
We are acquainted with a number of such

ON SPECIFIC THERAPEUTICS. 37

haptines or antibodies, of which the following
are examples : —

(1) The agglutinins, which cause clumping of
the cells.

(2) The amboceptors, which play the part of
carriers in the action of the complement.

(3) The opsonins, which render bacteria liable
to be seized by the phagocytes ; also those hap-
tines which are directed against the contents,
or metabolic products, of bacteria.

(4) The antitoxins.

(5) The anti-endotoxins, which are directed
against the endotoxins, to our knowledge of
which Macfadyen, whose early death the scien-
tific world deplores, has contributed so much.

(6) The precipitins.

(7) The antiferments, which are directed
against certain ferments of the bacterial cells,
e.g., pyocyanase.

There is no doubt, however, that many more
haptines exist. In order to obtain an idea of the
extreme diversity of the phenomena which
cause immunity, one must look for other
substances whose functions are those which
characterise the haptines — the substances, for
example, which prevent cell division, or those
which combat the biological adaptation of

38 EXPERIMENTAL RESEARCHES

bacteria. The possibilities which I have here
indicated appear to be limitless, and the need
for study in this direction is evident.

I would here express my dissent from a pre-
judice which often makes itself felt, to the effect
that in this matter there exists a profound con-
tradiction between humoral and cellular im-
munity. As a matter of fact, to assume that
the action of antibodies is merely a process
of humoral pathology, is to put an artificial
construction on the facts observed : for the
side-chain theory is founded upon the view that
the antibodies are purely and simply the pro-
duct of cellular secretion, and that with their
appearance in the blood there are associated
changes in the cells which correspond to the
phenomenon of serum immunity. That the
action of antibodies takes place in the juices of
the organism is an incontestable fact, which,
however, in view of the cellular processes which
give rise to it, cannot with justice be claimed to
be evidence in proof of the correctness of
humoral pathology, otherwise we must consider
the action of ferments to be one of humoral
physiology.

In the Protean forms of the phenomena of im-
munity, of course, the action of haptines by no

ON SPECIFIC THERAPEUTICS. 39

means excludes phagocytosis ; destruction of
the bacteria outside the cells and their assimi-
lation by the phagocytes are processes which
may take place alongside each other, and, by
their simultaneous action, increase the protec-
tive power. A special proof of the importance
of the study of haptines appears to me to be the
fact that — as the opsonin theory, which we owe
to Sir Almroth Wright, has made more evident,
specific haptine reactions form the basis also of
phagocytosis, which Metchnikoff has studied in
so masterly a manner. The opsonins and cyto-
tropic substances render the bacteria liable to
attack by the phagocytes, and here we have a
field in which humoral and cellular processes
meet. One cannot, however, say that the
possible causes of immunity are confined to
haptine action and phagocytosis. Perhaps the
atreptic view, by which differences of degree in
avidity on the one side or on the other are
presumed, is correct in many cases in which
other influences are at work. This view of the
case I shall treat more fully when I come to
speak of carcinoma.

The immunity of an animal is, therefore, ex-
plained as being due to the great energy of the
cells of its body, which are able to appropriate

40 EXPERIMENTAL RESEARCHES

nutritious substances for themselves, and in so
doing to deprive parasites of them. The oppo-
site condition must be due to a certain dis-
posing influence, and immunity of the parasites
must be a condition of the cause of infectivity.
The bacterial cells may in the same way be
immune against haptine substances, and may
withstand the action of the serum.

Thus there exist unstable relations between
immunity and infection, and between parasite
and host, relations which may depend on the
most varying influences, and which lead up to
the phenomena of reversible action, which calls
for further study.

One cannot, therefore, go to work in a one-
sided way when analysing and judging the
various forms of phenomena, but must carefully
consider together all the factors in question.
The study of every possibility will bear fruit
and make for an understanding of the processes
of infection and immunity. I believe, however,
that I have shown the influence exerted by
the haptines upon the cause of infection is of
great importance, not only when these are
viewed as destroyers of the cause of infection
outside the cells, but also when viewed in
connection with the results of their anchoring

ON SPECIFIC THERAPEUTICS. 41

power, chief among which at present stands
phagocytosis.

It is our task to advance by a more accurate
and more extensive study of all the haptines
and their actions, and in the first place we must
gain a knowledge of the influences exerted upon
the causes of infection by the distribution and
action of dissolved substances whose action is
cytotropic, so that we may obtain a nearer
insight into the manifold secondary phenomena
which arise from them.

42 EXPERIMENTAL RESEARCHES

LECTURE II.

ON THE ATREPTIC FUNCTION.

IN my first lecture I dealt with the anchoring
phenomena exhibited by the different kinds of
parasitotropic substances, which thus aid the
body in the process of recovery, and I chiefly
discussed, in their historical sequence, the three
therapeutically important agents, the antitoxins,
the bacteriolysins and the opsonins of Wright.
I, however indicated, that with these substances
the possibilities were not yet exhausted. To-
day I shall speak of a different series of processes
which may aid the organism to fight its adver-
saries, processes of a more passive nature, in
which a secretion of active attacking substances
does not take place.

You will remember that in the early days
of immunity the exhaustion theory played an
important role. Pasteur thought that when
the body overcomes an infection, there are
removed from it certain substances necessary

ON SPECIFIC THERAPEUTICS. 43

for the development of that particular bacterial
species ; should the same kind of bacterium again
penetrate into an organism thus modified, it
would not find those substances necessary for
its growth, and thus a new infection could
not occur. This theory has of late years, and
especially since the discovery of the different
kinds of antibodies, ceased to be of importance
in practical medicine. In point of fact, it is
hard to imagine that in the living body, as a
result of some infective disease, which need
not have produced any serious disturbances,
there should have occurred a complete and per-
manent disappearance of certain chemical sub-
stances. That during disease the internal
secretions should become disturbed, and that
possibly at this time certain cellular products
should be temporarily removed by excessive
consumption and insufficient nutrition, is pos-
sible and even probable, but it is quite unin-
telligible how, after complete recovery, such a
state of things could still persist. Yet this old
theory seems to me to contain a nucleus of truth,
as is so often the case. The chief point of
Pasteur's theory is that in the modified body
the bacteria should not be able to assimilate
certain food substances. Now, Pasteur's idea

44 EXPERIMENTAL RESEARCHES

is that this could only take place if the substances
in question were absent ; this is obviously incor-
rect. But this hypothesis is not necessary in
order to explain such insufficient nutrition of the
bacteria ; it suffices to assume that those sub-
stances may still be present, but that the
parasitic agents in question are incapable of
absorbing them ; in other words, that the
substances have ceased to be at the disposal of
the bacteria.

This phenomenon, at the time of my can-
cer studies, I embodied in the definition of
"Atrepsy." I will give you a few examples
from my researches which will make clear what
I mean by this definition, and I will begin with
a fairly simple case, viz., that of the cobra
venom.

This poison produces in the body a num-
ber of very different injuries, e.g., it affects
the nervous centres, the subcutaneous tissue,
the red blood corpuscles, the endothelia, etc.
But whilst the manifold pathological condi-
tions produced in the body, e.g., by corrosive
sublimate or any other well-defined chemi-
cal substance, are in every case the effect
of one substance on different organs — as, for
instance, nephritis, inflammation of the salivary

ON SPECIFIC THERAPEUTICS. 45

glands and necrosis of the intestinal mucosa
are the result of the one poison, corrosive
sublimate, — the case is entirely different with
snake venom, for there each separate and indi-
vidual effect is the result of a different individual
poison. Thus, one can prove with absolute
certainty that the effect on the nervous tissues
is produced by the neurotoxin, that on the
blood corpuscles by a hsemolysin, and the
inflammatory changes by a special endothelio-
toxin. It has been shown that those components
of snake venom which attack the red blood cor-
puscles are present in such a form that by
themselves they do not suffice to destroy them.

By following up the researches of Flexner
and Calmette, my former assistant and friend,
Dr. Kyes, has succeeded in throwing some light
on these difficult subjects. He proved that the
haemolytic component of the cobra venom is an
amboceptor, which, however, differs from the
bacteriolytic amboceptors in not combining with
the complements of the blood, but which enters
into union with a chemically well-defined sub-
stance, lecithin.

By shaking cobra venom solutions with a
solution of lecithin in chloroform, one can show
that the whole haemolytic toxin passes quanti-

46 EXPERIMENTAL RESEARCHES

tatively into the chloroform, whilst the watery
portion has retained the whole of the true
poisonous principle, viz., the neurotoxin. From
the chloroform solution one can isolate the
active hsemolytic principle by precipitation with
ether ; it can thus be obtained in the form of
a white powder, which, on chemical analysis,
gives figures which almost exactly fit the for-
mula of a mono-stearyl-lecithide. In this
process there has, therefore, been split off one
of the two molecules of fatty acid ; this can be
recovered quantitatively from the ether.

The substance thus obtained is produced by
the union of a small quantity of the haemo-
lytically active part of cobra venom with a
relatively large number of lecithin molecules.
It is easy to prove that in this we really have to
do with a chemical compound, and not the
result of an absorption similar to that recently
described by Michaelis between rennet and
mastix ; for it is impossible to split up the
substance by solvents into two components,
viz., mono-stearyl-lecithin and the cobra venom
amboceptor, which would be possible if it were
only such a mixture. On the other hand, in
conformity with its altered chemical character,
the substance has obtained new properties not

ON SPECIFIC THERAPEUTICS. 47

found in cobra venom. For, in the first place,
it is distinguished from the cobra amboceptor
by having become so completely thermostable,
that solutions of this cobra-lecithide can be
boiled for hours without losing a fraction of
their efficiency, whilst under similar conditions
the original venom would be rapidly destroyed. A
further important property is shown by the fact
that cobra-lecithide, even in small quantity, in-
stantaneously and without any incubation period
dissolves the erythrocytes, whilst snake venom,
even in the presence of lecithin, only produces
a similar effect after several hours. The long
incubation period observed in experiments with
the native poison is explained by the fact that
the real poisonous body has first to be formed
by a synthetic process, which, in dilute solutions
and at ordinary temperatures, may naturally
require some time for its completion. As you
see, I believe with Kyes, that the haemolysis of
the red blood corpuscles is the result of a similar
lecithide formation.

It is interesting to note the action of cobra
venom towards different kinds of red blood cor-
puscles. It has been found that the corpuscles
of the guinea-pig, man and rabbit, are at once
dissolved by solutions of the poison, but that

48 EXPERIMENTAL RESEARCHES

this is not the case with sheep or ox corpuscles.
These latter, however, are also dissolved if
a small amount of lecithin is added to the
mixture.

How is one to explain the fact that cobra
venom cannot by itself dissolve certain kinds of
blood ?

It has been shown that in all bloods lecithin
is present in fairly large quantities. If in spite
of that only certain kinds of blood are affected
by cobra venom, this must be explained by the
assumption that in every blood lecithin is not
present in the free state, but that it is combined
with certain substances of the erythrocyte's
stroma, and that the firmness of this union
differs considerably in the different kinds of
blood. That even in the resistant erythrocytes
— e.g., those of the sheep — lecithin is present as
the activating substance, can easily be proved
by breaking down the compound of proteid and
lecithin. This can be done by extracting with
alcohol. The alcoholic extract of sheep's blood
corpuscles can be shown to activate a mixture
of snake venom and sheep's corpuscles.

The results are different if, instead of cobra
venom, other poisons are employed. Thus, the
poison of Bothrops lanceolatiis is absolutely unable

ON SPECIFIC THERAPEUTICS. 49

to dissolve any of the above-mentioned bloods
without the addition of lecithin, whilst the poison
of Trimesurus anamalensis dissolves only the cor-
puscles of the guinea-pig, that of Bungarus
fasciatus only the corpuscles of man and guinea-
pigs, but not others. From these researches
we may deduce two facts, viz. : (i) That the
haemolytic amboceptors of snake poisons possess
different affinities for lecithin, this affinity being
lowest in bothrops venom, which dissolves none
of the before-mentioned corpuscles, and highest
in cobra venom, which dissolves three of the
five kinds ; (2) that the firmness of union of the
lecithin present in the different blood species
varies, being highest in the sheep and the ox,
lower in the rabbit, still lower in man, and
lowest of all in the guinea-pig. You see here
very clearly one of the conditions which are of
importance for the definition of atrepsy, namely,
that of indisposability. Lecithin has an affinity
for certain constituents of the cell and it also
has an affinity for the constituents of the poison;
the difference between these two affinities
decides whether cobra lecithide can be formed
and haemolysis can thus be produced. If the
affinity of the constituents of the stromata is
higher, haemolysis does not take place, i.e., the

E

50 EXPERIMENTAL RESEARCHES

lecithin is not at the disposal of the cobra ambo-
ceptor.

I will now discuss a further similar case,
namely, that of certain phenomena observed by
me in the course of my studies on trypanosomes.
As some of you may know, I have, together
with my assistants, Dr. Rohl and Dr. Browning,
proved that trypanosomes can be rendered
resistant towards all those agents which are
employed for curative purposes. If, e.g., a
mouse infected with trypanosomes is treated
with fuchsin, the parasites disappear, but
return after a longer or shorter interval. If
the treatment is then repeated, a similar result
is observed. Thus this alternation of relapse
and treatment can be repeated several times,
though not indefinitely, for in course of time the
effect of the fuchsin becomes more and more
unsatisfactory and finally disappears altogether.
If now the trypanosomes are transferred from
such an animal to normal mice, these parasites
are now, even in the normal mouse, found to be
no longer capable of being influenced by fuchsin.
There has thus been formed a fuchsin-resistant
or " fuchsin-fast " strain. I have chosen this
expression because I found that such a resis-
tance, once acquired, appears to remain un-

ON SPECIFIC THERAPEUTICS. 51

altered even after as many as a hundred succes-
sive passages. We have up to now produced
similar strains resistant to arsenical preparations,
to trypan-red and trypan-blue as well as to
fuchsin. In my next lecture I shall enter more
fully into the importance of these strains.

I will here discuss one point only which is
connected with to-day's subject, viz., the
question as to the origin of the resistant strains.
I have made the interesting observation on an
arsenic-fast strain which was obtained by several
years' treatment with atoxyl, and which was
then made still more resistant by means of a
very active arsenical preparation (No. 379),
which I shall briefly call " Trypocid." This
strain is distinguished from normal trypanosomes
in that it is no longer influenced by this highly
active arsenical preparation. One would have
expected that also in test-tube experiments, it
would have shown a high resistance towards
trypocid. Of course, such an examination could
only be carried out by comparing it with other
strains of trypanosomes. For obvious reasons
I employed the original strain from which this
resistant one had been derived, and which had
been cultivated in the laboratory. I chose two
animals containing about the same number of

£ 2

52 EXPERIMENTAL RESEARCHES

parasites and mixed their bloods with solutions
of trypocid of different degrees of concentration.
I now observed the very unexpected phenome-
non that in these mixtures the immune strain was
far less resistant to higher doses of trypocid than
the normal one. Thus whilst concentrations of
i in 500 to i in 1,000 almost instantaneously
killed the immune strain, the control strain
retained its motiiity unimpaired in these con-
centrations for not less than five minutes.

We have here, therefore, one of those striking
phenomena which have already been met with
in the study of immunity and which consist in
the simultaneous occurrence of immunity and
hypersensibility in one and the same organism ;
but the case here is somewhat more simple,
since we are working with unicellular organisms,
and may therefore hope to obtain a more accurate
idea of the case.

Allow me at this juncture to suggest to you a
few fundamental principles. In order that a
given poison, e.g., the arsenical preparation,
may act upon the trypanosomes, it must contain
certain chemical groups which seize upon the
latter. These groups I will distinguish from
the ordinary receptors, which play an important
part in the theory of immunity, by the name of

ON SPECIFIC THERAPEUTICS. 53

"chemo-receptors." In like manner the bodies
of the higher animals must possess such chemo-
receptors in certain organs, as they are also
injured by the poison. If now a mouse infected
with trypanosomes is injected with the arsenical
medicament, this substance will be distributed
between the parasites and the organism of the
mouse. If the receptivity of the parasites is the
stronger, they will be killed in the organism ; in
the alternative case, they will not be destroyed.
Consequently the curative result obtained when
experimenting under normal circumstances re-
presents also the differential between two
avidities ; i.e., if a mouse infected with the
normal strain of trypanosomes is injected with
the arsenical preparation, the " trypanotropic "
force of the drug is stronger than the " organo-
tropic " force ; we therefore obtain a curative
effect with doses which are not injurious to the
organism of the mouse.

In the resistant strains this is not the case,
since the same doses show no trace of action on
the parasites. One might imagine that this
result was due to the trypanosomes having lost
their chemo-receptors : but it is not necessary
to make this assumption, and Nature has in its
wonderfully complex mechanism found another

54 EXPERIMENTAL RESEARCHES

and far more economical expedient than the
loss of a perhaps necessary chemical structure.
It is quite evident that the same result could be
obtained if the chemo-receptors of the trypano-
somes were so far reduced in their affinity that
the proportion of distribution was altered in
favour of the chemo-receptors of the organism.
That, indeed, is what takes place. Had the
chemo-receptors quite disappeared, the immune
strains would also in the test-tube experiment
be more resistant than the normal ones ; as a
matter of fact, with certain preparations the
contrary occurs.* The phenomenon is most
easily explained by the supposition that the
affinity of the chemo-receptors has become
adjusted in such manner to the counterbalancing
affinity of the mouse organism, that in the mouse
no more arsenic remains at the disposal of the

* The fact that these resistant strains are, in the test-
tube experiment, affected even more severely than the
normal ones, is the result of the co-existence of resistance
and hyper-sensitivity in the parasite's protoplasm. Such
a combination of immunity and hyper-sensitivity has
repeatedly been observed in the course of the immunisa-
tion of higher animals, and this is now assuming increasing
importance. It is the more interesting to note that the
same phenomenon occurs also in uni-cellular organisms.
The explanation of this observation will be given later.

ON SPECIFIC THERAPEUTICS. 55

trypanosomes. If, however, this counterbalancing
effect of the mouse's organism is removed, the
poisonous agent can with full force seize the
trypanosomes. I would even go so far as to say
that we have to do in this case with a biological
adjustment of the greatest delicacy, and that
Nature limits herself to the most strictly neces-
sary gradations and never allows a reduction of
the affinity to its minimum.

As a proof of this I may mention a similar
biological observation. I sent my resistant
strain to the Liverpool School of Tropical
Medicine, where it was found that the atoxyl-
resistant strain showed the resistance as
described by me only in mice but not in rats.
This result was, of course, very surprising, but
you will easily understand it on the basis of the
previous results, by the supposition that the
avidity of the rat's organism for trypocid is
inferior to that of the mouse. In the rat,
therefore, the trypanocidal effect is preponderant,
since the trypanosomes are still able to seize
upon the poison.

Let me now pass on to a very different field
of research. I should be guilty of carrying
owls to Athens, if I were here to speak in detail
of the nature and importance of Jenner's great

56 EXPERIMENTAL RESEARCHES

discovery, which represents an epoch in our
campaign against disease and the beginning
of the era of immunisation. I shall therefore
enter only into one question, which throws a
little light upon that modification of the variola
virus which is produced by its passage through
the cow. As you know, animals are also affected
with small-pox-like diseases which are closely
related to human variola. Such diseases are
the so-called " sheep- pox " and " bird-pox."
The latter affection has recently been studied
at my Institute by Dr. Sticker and Staff-Surgeon
Marx. It is easily transferable by cutaneous
inoculation to fowls and pigeons, and produces
extensive small-pox-like hypertrophy of the
epithelium. In the diseased epithelial cells
highly staining spherical cell-inclusions are
found, which must be identified with the cor-
puscles of Guarnieri that are found in epithelial
cells infected with small-pox or vaccine. Ani-
mals which have passed through the disease
have thus acquired a complete immunity against
subsequent infections.

As was shown by my assistants, the virus of
bird-pox can be filtered through filter-candles.
It can be transferred to fowls and pigeons. In
fowls the virus retains its pathogenic properties

ON SPECIFIC THERAPEUTICS. 57

unaltered for any number of inoculations. But
the results are very different if one tries to
inoculate with the diseased products occurring
in pigeons. For although the inoculations can
also be carried out in pigeons for any number of
times, yet in such series the virulence for fowls
has completely disappeared. You see, there-
fore, that we are here dealing with a process
closely related to the modifications which the
variola virus suffers in the organism of the cow.
The only difference exists in the fact that in the
one case the virulence for the original host is
only reduced, whilst in the case of bird-pox it is
completely destroyed. In my view this can be
explained only in one way, for it is obviously
impossible to suppose that by one passage
through the pigeon's organism the parasites in
question should obtain a new set of receptors
capable of seizing upon antagonistic bodies
present in the normal fowl's serum. It is far
more natural to suppose that we have to do
here with a form of atrepsy. It is safe to
assume that the chemical composition of fowls'
ano! pigeons' tissues is not identical, and that
therefore the parasites, in their passage through
the pigeon, must assimilate substances different
to those assimilated in their passage through

58 EXPERIMENTAL RESEARCHES

fowls. Therefore, that part of the receptors
which deals with the nutritive substances in
the fowl's organism is not in use during the
passage through the pigeon. In view of their
great instability it is possible that this portion
of the receptors may become atrophied, whilst
the receptors specific for pigeon's substances
become correspondingly increased. Thus the
micro-organism would lose its power of assimi-
lating certain substances of the fowl's organism.
If such a parasite were transferred back to the
fowl — supposing one of the specific constituents
of fowls to be necessary for its proliferation— it
would no more be able to grow in the fowl.

This is, therefore, a case of the loss of certain
receptors which are absolutely necessary for
nutrition. That such a change, having been
once acquired, should be permanently trans-
mitted from one generation to another, is
hardly surprising in view of my recent obser-
vations on the modifications of arsenic-resistant
trypanosomata. Obviously, the total loss of the
chemical structures in question must render
such modifications irreparable. Such charac-
teristic instances as those in which a single pas-
sage produces a permanent loss of virulence for
a certain animal species, are not very frequent.

ON SPECIFIC THERAPEUTICS. 59

Correspondingly we may presume, that the
variola virus, when passing through the cow,
permanently loses certain atom-groups which
cannot be regained ; I believe, however, that in
cow-pox this loss is of a lesser degree than in
bird-pox.

On the other hand, in bacteriology we fre-
quently find that such alterations produced by
certain animal passages are not of a permanent
nature. Thus one can, by successively passing
streptococci through animals of one species,
obtain a maximum of virulence for that species.
It is possible, therefore, at will to produce a
rabbit-strain, a mouse-strain, etc. But one can,
by repeatedly passing such a strain through
animals of a different species, transform it into a
different strain, and thus in course of time
obtain from a rabbit-strain a mouse-strain.
This is therefore a case of atrepsy of the
receptoral apparatus of the bacteria for the
animal species in question, but this atrepsy
differs from that of the variations of small-pox
by not possessing a permanent character.

Probably the majority of so-called non-patho-
genic micro-organisms, if introduced into an
animal's body, perish by this mechanism. It is
not neccessary to assume the presence of special

60 EXPERIMENTAL RESEARCHES

poisons in the body, it suffices to suppose that
the bacteria in question do not find the needful
means of existence in the body and therefore
cannot multiply. This being the case, they
cannot for any length of time remain alive in
the body, for then the latter's defensive forces,
its phagocytes, come into action and destroy
the invaders in a non-specific manner.

In the case of the pathogenic micro-organ-
isms it will, however, as a general rule be safer
not to attribute too great an importance to
atrepsy. But it is evident that micro-organ-
isms can only be pathogenic for a certain
animal if they find in it possibilities of nutri-
tion. Yet, to my mind, quite a number of
infections are characterised by the fact that the
micro-organism, with the exception of only a
few remnants, becomes atreptic. As an exam-
ple of such an occurrence I would mention the
fine researches of my friend, A. Neisser, who
found that monkeys injected subcutaneously or
intra-peritoneally with great quantities of syphi-
litic virus became neither infected nor im-
munised.

It does not appear possible to assume that in
the serum of monkeys, substances should be
present which destroy the causative agents of

ON SPECIFIC THERAPEUTICS. 61

syphilis. It will be more correct to imagine
that the delicate spirochaetae have lost their
power of infecting because they have found no
suitable nutritive bodies in the serum.

On the other hand we know, from the funda-
mental experiments of MetchnikofF and Roux,
that typical primary lesions can in apes be
obtained in any part of the skin, in the lower
monkeys chiefly in the eyebrows and penis.
Infection could also be produced by rubbing
diseased tissue on to the cut surface of the
testicle. On examining the inner organs of the
monkeys, Neisser found the virus to be present
only in a few organs, especially the haemato-
poietic organs, viz., the spleen, the bone mar-
row and the lymphatic glands, as well as in the
testicles, whilst all the other organs proved
sterile. For this reason Neisser thought that
the primary processes in certain defined tissues
were of importance for the occurrence of the
generalised disease. If we follow this view, it
will be most easy to explain the state of affairs
by the hypothesis that the serum of monkeys
does not contain properties sufficient for the
nutrition of the spirochaetae, whilst in certain
cells of these animals, viz., the epithelia, the
testicular cells and the leucogenic cells, sub-

62 EXPERIMENTAL RESEARCHES

stances permitting proliferation of the spiro-
chaetae must be present. Probably there exists
only a minimum of such substances, auxiliary,
but indispensable for their nutrition.

I may here remind you of the fact that the
influenza bacillus, as was shown by R. Pfeiffer,
is not able to proliferate in ordinary culture
media, but that a trace of haemoglobin is neces-
sary for its growth. In syphilis I imagine the
case to be somewhat similar. All the organs
which can be primarily infected with syphilitic
virus would thus be the carriers of this specific
auxiliary substance.

The case is more evident, macroscopically, in
a great number of very different affections, e.g.,
variola and vaccinia, bird's and sheep's pox,
foot- and mouth-disease, trachoma, rabies,
fowl's pest and scarlatina. In all these infec-
tions there are found in the epithelial cells the
above-mentioned peculiar cell-inclusions, which
in small-pox are called Guarnieri's corpuscles, in
rabies Negri's corpuscles. According to recent
investigations we are justified in supposing that
these inclusions do not themselves represent the
parasites, but that they are derived from the
cells and consist, according to von Prowazek, of
plastin and nuclease. Within these masses

ON SPECIFIC THERAPEUTICS. 63

there would lie enclosed the exceedingly minute
disease-producing agents, these agents being in
general so small as hardly to be accurately
recognisable by the microscope, with the excep-
tion of the inclusions in trachoma, where the
parasites are clearly visible and bear the stamp
of something organised. We may safely as-
sume that in all these cases the specific localisa-
tion of the parasites in certain distinct kinds of
cells, points to the fact that in these cells,
notably the epithelium cells in small-pox and
trachoma, and the ganglion cells in rabies and
fowl pest, such specific auxiliary bodies for the
parasites are present, whilst in all the remaining
parts the micro-organisms would encounter con-
ditions of atrepsy.

The atreptic immunity, which results as I
stated from certain substances not being at the
disposal of the parasites, seems to me to play
an important role in the study of tumours.

I may even say, that I first formed the idea
of atrepsy as the result of numerous experi-
mental observations upon malignant mouse
tumours which are of cardinal interest for
modern cancer research. I believe that a
number of clinical and experimental facts in
this sphere of knowledge can only be explained
by this hypothesis.

64 EXPERIMENTAL RESEARCHES

As you know, mouse cancers are tumours
showing a far-reaching analogy to human
cancer, and it would therefore appear most
justifiable to regard the experiences gained with
them as the solid foundation of an experimental
study of tumours. Besides several analogies,
however, the mouse tumours show certain
special features, chief among which is the mode
of formation of metastases. It is most remark-
able that the highly virulent carcinomata which
have been cultivated in numerous generations,
very rarely show macroscopically visible metas-
tases, whilst the slowly growing spontaneous tu-
mours relatively more often, though also not very
commonly, form large secondary nodules in the
lungs. Our interest in this peculiar behaviour
must even be increased, since Haaland has
shown that microscopical metastases are by no
means uncommon, but that they almost always
remain below the limit of macroscopical visi-
bility. I endeavoured to solve this problem by
making, so to speak, artificial metastases, viz.,
by studying in animals affected with an inocu-
lated tumour the result of a second inoculation.
I found the interesting fact, that the effect of
the second inoculation was inversely propor-
tional to the energy of growth of the primary

ON SPECIFIC THERAPEUTICS. 65

tumour. In rapidly growing sarcomata and
carcinomata the second inoculation almost
always had a negative result ; at best, the
second tumour remained far inferior in size to
the primary one. Only in the very slowly
growing chondromata was there no distinct
difference in their energy of proliferation.
These results agree with the following obser-
vation, which has been confirmed thousands of
times. If, as is regularly done in my Institute,
the inoculation is carried out by introducing
the capillary tube at the groin and pushing it
up as far as the axilla, the greater quantity of
inoculated material will of course be deposited
at the latter site, whilst only a small portion
will remain at the groin. In the case of the
rapidly growing carcinomata and sarcomata the
large principal tumours almost always develop
in the axilla, whilst a minute nodule is usually
found in the groin. In the case of the chon-
dromata, however, both tumours usually show a
fairly equal development. Thus, in the latter
case, the macroscopical appearance somewhat
resembles that of an hour-glass, whilst in the
former it is more like a balloon with its car
attached.
To my mind, the explanation of this pheno-

66 EXPERIMENTAL RESEARCHES

menon is as follows : every proliferation depends
in the first place on the avidity of the cells for
the nutritive substances. Normally, there are
certain well-defined laws of distribution, which
guarantee the proper working of the organic
functions. The avidity of the tumour cells is in-
creased, as compared with that of the body cells.
The more energetically a tumour proliferates,
the more powerfully does it attract the nutrient
substances from the blood. In the case of a
rapidly growing tumour it may therefore very
easily occur, that for such cells which are under
very unfavourable conditions of nutrition, e.g.,
cells inoculated and metastatically carried away,
there is an insufficient supply of nutrient sub-
stances, and that they therefore either perish
from atrepsy or at least are unfavourably influ-
enced in their growth. From this point of
view it is also quite evident that slowly grow-
ing tumours can attain far greater dimensions
than rapidly growing ones, since in the former
the consumption of nutrient matter, in spite
of their size, is less than in the latter, and
thus the entire organism is injured to a lesser
extent.0

* In connection with this, attention may be called to the
recent observations of Haaland (Berl. Klin. Wochenschr.,

ON SPECIFIC THERAPEUTICS. 67

I fail to understand how von Dungern can
think himself justified in doubting the correctness
of this explanation, on the ground that hitherto
the experimental proof of a permanent increase
of cellular avidity by increased nutrition has not
been adduced. I have never asserted that we
can by artificial means permanently raise the
cellular avidity. Such a power would obviously
involve the possibility of artificial tumour pro-
duction. There is therefore no contradiction,
as von Dungern imagines, between my views
and the facts that by artificial hyperaemia better
nutrition is obtained, and that, after removal of
large parts of the body, e.g., amputation at the
thigh, the hypernutrition of the individual is
evidenced, not by the proliferation of func-
tionating epithelial cells, but by increase of
the adipose tissue. From the fact that the
increased avidity causes an increased attraction
of nutritive material, it does not follow inversely
that copious nutrition must increase the
avidity. An indirectly injurious effect of more
highly avid cells on less avid ones can there-
fore only become evident, if the body does not

1907, No. 23, p. 718), which show that pregnancy very
often produces a retarding influence on the growth of such
tumours.

F2

68 EXPERIMENTAL RESEARCHES

possess sufficient nutrient matter to saturate
every avidity. The origin of the higher avidity
of the tumour cells is quite unknown to us ; for
this question means neither more nor less than
the last problem of tumour aetiology.

An entirely different significance must be
ascribed to the specific growth-stimulating
bodies or " hormones," for the existence of
which we have received irrefutable proof chiefly
by the work of Starling. As you may know, he
observed lactation in non-pregnant females as
the result of injecting emulsions of embryos.
But, of course, I should never think of asserting,
as one might be led to think by von Dungern's
explanation of my views, that as the result of
over-nutrition a typical lactation might be set
up in a virgin.

Such specific growth- stimulating bodies play
an important role also in other cases, and are
intimately connected with atreptic immunity.
Thus we know that the influenza bacillus cannot
do without haemoglobin for its growth. There-
fore it is quite easy to cultivate the bacillus
directly from the sputum, which almost always
contains small quantities of haemoglobin. But
if we do not artificially add haemoglobin to the
sub-cultures, the bacteria very soon die in them.

ON SPECIFIC THERAPEUTICS. 69

Specific auxiliary growth- substances we must
also assume in all those infective diseases
which are characterised by a marked localisa-
tion of the poison, as in bird-pox and syphilis.

The case is quite similar in that form of
atreptic immunity which follows from my
experiments of inoculating rats with cells of
mouse tumours. As you know, hitherto no one
has ever succeeded in transferring normal or
tumour tissue to animals of a different species.
The limits of transplantability coincide with
those of bastardisation. The question, how-
ever, now arose, as to whether the same barriers
held good also for our highly virulent tumour
material. To that end I employed the rat as
the animal phylogenetically most nearly related
to the mouse. In fact, if a virulent mouse
tumour is inoculated into a rat, the result is
quite different from all hitherto known true
cases of transference of tissue outside a species ;
for during the first eight days the cells, both of
carcinomata, sarcomata and chondromata, pro-
liferate in the rat just as they do in the mouse.
During this time tumours arise from the size
of an almond to that of a date, containing nume-
rous mitoses and microscopically differing in no
way from the mouse tumours. After that time,

70 EXPERIMENTAL RESEARCHES

however, the limit of their growth is reached
and there now follows gradual resorption, which
is concluded after another week or two. If the
tumour at its maximum development is trans-
ferred to another rat, it does not gain any
foothold there, whilst if it is re-inoculated into
a mouse it again proliferates luxuriantly. It is
possible to keep on carrying out this zig zag
inoculation from mouse to rat and from rat to
mouse, and again from mouse to rat, etc., for
any length of time without the slightest check
to the energy of proliferation being encountered.
The question now arises, how it is possible to
explain, on the basis of these facts, the immunity
of the rat towards mouse-tumour cells.

We can at once exclude the existence of a
natural immunity by antibodies in the rat, in
view of the marked initial proliferation of the
tumour. On the other hand it would be possible
to imagine that the resorption of the tumour
might be the result of an active immunisation of
the rat. That such immunisation ultimately
occurs may be proved by the fact that the result
of re-inoculating a rat, in which a first tumour
has been absorbed, with a second tumour, is
always negative.

I do not, however, consider it permissible to

ON SPECIFIC THERAPEUTICS. 71

identify these two manifestations of immunity.
For in the first place it is highly improbable
that, during the period of rapid proliferation of
the tumour in the rat's organism, any consider-
able absorption of tissue elements should occur.
Further, previous to that period, a production of
antibodies has not been proved. Lastly, since
the tumour, when re-inoculated into the mouse,
never shows ever so slight a reduction of viru-
lence, one would have to suppose with von
Dungern that the anti-substance formed in the
rat's organism was activated only by the com-
plements of rats but not of mice. This, again,
has up to the present not been proved.

A far simpler and more natural explanation
of all these phenomena is afforded by my hypo-
thesis of atrepsy. According to this, the mouse-
tumour cells require for their growth not only
the ordinary nutritive substances which the rat
can also supply to them in ample quantity, but,
besides that, some well-defined substance which
is present only in the mouse's organism.

In the rat, then, these cells would only be
able to go on growing so long as there still
remained with them some of the specific growth-
stimulating substance that was introduced with
them at the time of inoculation. When it has

72 EXPERIMENTAL RESEARCHES

all been used up, growth can only be further
stimulated by giving a fresh supply of this sub-
stance, e.g., by re-inoculation into the mouse.

Just as it is impossible to explain this immunity
of rats by antibodies, one cannot thus explain
that form of immunity in mice which is evidenced
in the number of positive results following inocu-
lation with different tumour strains. Generally
speaking, the proportion of positive results for
one strain, with the exception of unavoidable
chances, varies only within narrow limits.
Thus, the strain of carcinoma which we culti-
vate in grey mice regularly grows in about 20
to 25 per cent. ; Jensen's tumour, according to
the publications, in about 40 to 60 per cent.,
our other carcinomata and sarcomata in 90 to
100 per cent., and our chondroma unexception-
ally in 100 per cent. The constancy of these
differences can never be explained by the
assumption of antibodies, but only as the
expression of a certain vitality of the tumour
cells which is constant for each strain.

The atreptic condition is most clearly shown
in the cases repeatedly mentioned by Michaelis,
Bashford and Haaland, where the transference
to other races of mice of such virulent tumours
was either quite impossible, or was only successful

OAT SPECIFIC THERAPEUTICS. 73

in a very low percentage of cases. In the latter
a gradual adaptation to the strange culture
medium could only with great difficulty be
obtained.

From my previous remarks it follows that the
increased avidity of the cells for the food sub-
stances is the most important characteristic of
the tumour cells. But this increase does not
suffice to explain all the phenomena observed.
Albrecht already insisted that for malignant
tumours one must admit not only an increase
but an alteration in the assimilation, in such
manner that the " structure materials " taken
up from the surrounding media must in some
way be bound or laid up " until they had
reached an amount sufficient for the division
of the cell." Besides this increased food-
absorption, the result of action of the receptors,
chemiotactic remote efforts must, to my mind,
play an important role.

Such a phenomenon is very well shown by
the chondroma which has been cultivated for
many years in my Institute. This tumour shows
a well-marked chemiotactic effect on the blood-
vessels, in such manner that even small inocu-
lated tumours, a week or two old, shine blue
through the skin, whilst larger ones in their

74 EXPERIMENTAL RESEARCHES

entirety present the appearance of a haemor-
rhagic tumour. This angiotactic attraction of
the blood-vessels is an essential condition for
the growth of the tumours ; for if, as is the case
in intra-peritoneal inoculation, these angiotactic
properties cannot become active, or if they have
been lost, as is the case in immunised animals,
proliferation remains very limited both as
regards its duration and its extent, and it
rapidly ceases whilst the whole tissue becomes
necrotic.

As you see, I have dealt with a number of
apparently very different subjects of biology and
pathology, which are, however, united by a
common bond. We have before us the com-
petition of the different individualities for their
nutrient materials. The majority of such
chemical compounds are organically bound
within the cell to its constituents. This is, I
think, especially true of the lipoid bodies, like
fat, lecithin and others. In this respect I may,
perhaps, remind you that normal tissue, e.g.,
the cortex of the kidney, contains fat in such a
hidden modification, since it shows fat granules
neither microscopically nor by the osmic acid
reaction. If one examines a similar organ in
the stage of fatty degeneration, it shows macro-

ON SPECIFIC THERAPEUTICS. 75

scopically a white colour, and by microscopical
examination with osmic acid shows an enormous
number of fat granules. Yet, by extracting a
normal and such a diseased kidney with ether
and by examining the ether residue, Dunham
showed that the quantities of extract thus
obtained, which is but a mixture of fats and
lecithin, is exactly the same in both cases. In
this case, therefore, the occurrence of fatty
degeneration consists only in the fact that the
organic union between the proteids and the
lipoids has ceased, and that the component
parts have been liberated, and thus rendered
easily recognisable.

These facts are completely analogous to the
observations which I have previously communi-
cated when discussing Ryes' researches on
snake venom. I even think it probable that
the sugars may, in a similar manner, be chemi-
cally firmly connected with the cells, and in
like way the complex carbohydrates, e.g., gly-
cogen, may well be present in the body in a
hidden form. That is at least made probable
by an observation of mine on finding glycocen
in the polynuclear cells, viz., that the normal
polynuclear leucocytes never show the charac-
teristic reaction with iodine, whilst it is fre-

76 EXPERIMENTAL RESEARCHES

quently observed if some injury has befallen the
body. Thus, as you are aware, in a number of
intoxications the leucocytes give a positive
glycogen reaction. Even the slightest stimuli
may cause the leucocytes to split off glycogen in
their protoplasm.

It will be necessary to assume that all these
compounds are joined to certain protoplasmic
groups, and that the decomposition of these
compounds is the result of a fermentative
action, resembling the decomposition of amyg-
dalin, under the influence of emulsin, into
glucose, hydrocyanic acid and benzaldehyde.
Further, it is very interesting to note that one
can often, by simple extracting substances like
alcohol or ether, remove from the cells fat in the
form of substances which still contain a proto-
plasmic component, although chemically they
would appear at first sight to be pure lipoid
substances. Such extracts have been prepared
from red blood corpuscles by Landsteiner, Bang
and Forssmann, who found that by injecting
such extracts into animals one can obtain
specific haemolysins. This phenomenon, which
they endeavoured to use as an argument against
my side-chain theory, is explained in the
simplest manner by Ryes' observations upon

ON SPECIFIC THERAPEUTICS. 77

the snake-venom lecithide. This lecithide would
appear to be a pure fatty substance, being
soluble in chloroform and even in toluol and
alcohol, but as a matter of fact it contains,
as I was before able to show you, a very slight
quantity of the poison in chemical combination.
And it is this portion which induces the bio-
logical reaction. The fact that in spite of this
component being of a non-fatty nature, the com-
pound apparently behaves like an ordinary fat,
is due to the presence in it of a great number of
fatty molecules which have given to the whole
substance their specific characters.

I have frequently laid stress upon the fact
that many nutrient substances are not present
in the cells in a free state, and that, therefore,
they are not at the disposal of any invader, but
that a struggle is always necessary before they
can be rendered accessible. In the case of
snake venom I have shown that we are dealing
with a simple difference between two chemical
avidities. In the case of bacteria the affair may
be rather more complicated, according to the
occurrence of either of two possibilities, viz. :
(i) A direct assimilation in consequence of a
higher degree of affinity, or (2) an indirect
action by injury to the cell. Supposing, for

78 EXPERIMENTAL RESEARCHES

example, that a microbe had penetrated into a
cell, but had not the power of directly splitting
off the fatty bodies ; then these substances
would not be immediately assimilable by it.
Yet the microbe might obtain possession of
them if it secreted a poison injurious to the cell,
which would decompose the protoplasm. In
such a case the fat would be liberated, and
could be assimilated by the micro-organism,
even though the immediate affinity between
the two were very low. At any rate, these
considerations may so far be of interest as
they touch upon an issue of biology and patho-
logy, which has, on the whole, been somewhat
pushed into the background of late years, since
one has accustomed oneself in every infective
process to think in the first place of a specific
bactericidal action through the direct toxic
substances of the body, its haptines, etc. This
has led to our neglecting somewhat the simple
possibility that certain micro-organisms may
only be able to grow if certain conditions are
given for their development, and that they must
perish if these conditions are withheld. To
my mind this possibility is of importance in
many respects, and is especially significant for
our views on cancer immunity. At any rate,

ON SPECIFIC THERAPEUTICS. 79

we are dealing in these questions with a highly
complex field of research, in which many possi-
bilities are present, the most important being
this difference of avidities.

We have previously seen that the trypano-
somes may alter their avidity for arsenic, and thus
acquire an apparent immunity. The contrary
may occur in the case where the substance in
question is not a poison, but a nutritive body.
Here we should be perfectly justified in sup-
posing the avidity of the receptor to have
become increased. The same thing may take
place with such micro-organisms which can
physiologically diminish or increase their avidi-
ties. Thus in all these cases the struggle lies
between the adaptability of the parasite and
that of the host. The one whose adaptability is
the highest will remain the victor.

Of course this struggle is to a great extent
influenced by indirect actions, consisting in the
secretion by each antagonist of dissolved sub-
stances hostile to the vitality and receptivity of
the other organism. On the part of the bac-
teria, these substances are the toxins and the
dissolved intracellular substances ; on the part
of the body, the anti-substances.

A further vole is played by actions of a pro-

80 EXPERIMENTAL RESEARCHES

tective and defensive nature. Thus we have
shown that bacteria congregate in those parts
where they find the most favourable conditions
of nutrition, whilst the organism, both by its
phagocytes and by means of encystment, endea-
vours to render the pathogenic germs harmless
and to eliminate them. You see, therefore, that
this is a war waged in different fields, but in
which to every action there corresponds a
reaction. The war is waged in a three- fold
manner — by variations of affinity, by variations
in the poison, and by localisation.

ON SPECIFIC THERAPEUTICS. 81

LECTURE III.

CHEMO-THERAPEUTIC STUDIES ON TRYPANOSOMES.

THE studies on trypanosomes, which have of
late been carried out with the greatest energy,
especially in the Institut Pasteur in Paris, the
School of Tropical Medicine in Liverpool, and
the Speyerhaus in Frankfort-on-the-Main, are
of the greatest theoretical and practical im-
portance ; for in this field of study the patient
research work of Experimental Therapeutics
has first come into full activity, and at the same
time a basis has been gained on which we may
hope successfully to undertake the practical
attempt of suppressing sleeping-sickness. The
exceedingly great difficulty of these studies is
evidenced by the fact that hundreds and thou-
sands of substances have to be examined by
animal experiments, before a few producing a
therapeutic effect can be found. I myself have
in course of time examined more than 600 such
substances. From their effect on animals, hints

G

82 EXPERIMENTAL RESEARCHES

may be obtained as to what to avoid, and in
what direction therapeutics should advance. I
will not enter here into the details of my work,
but will only give you a general resume of the
chief results.

The substances which have hitherto proved
efficient in combating trypanosome infections
can be divided into three groups, viz. : — (i) The
group of the basic triphenyl-methane dyes ; (2)
The group of the benzidine dyes ; (3) The
group of the arsenicals.

The salts of mercury may be used as adju-
vants to each of the foregoing.

Among the triphenyl-methane stains, which
were first employed by Wendelstadt, para-
fuchsin is by far the most satisfactory. It is
possible, by prophylactic feeding with this drug,
to render mice resistant to trypanosome infec-
tions for a long time, and I believe that this
innocent substance might well be employed in
the Tropics for the prophylaxis of human
sleeping-sickness.

As regards the second type of substances,
belonging to the benzidine group, trypan-red
is a cotton dye which shows the remarkable
property of staining mice an intense red, which
remains fast for months. I have also prepared

ON SPECIFIC THERAPEUTICS, 83

a number of similar dyes belonging to the ben-
zidine series, some of which possess an even
more powerful action than trypan-red. In
several cases a blue dye belonging to the same
group, which was prepared by Nicolle and
Mesnil, has proved still more efficient.

The value of arsenic acid which has for a
long time been employed in the treatment of
trypanosome diseases was first recognised by
Laveran. Later on it was discovered, at the
Liverpool School of Tropical Medicine, that
atoxyl was a far more powerful substance. It
is prepared by treating aniline with arsenic
acid. The makers believed it to be an anilide of
arsenic acid, i.e.9 a compound in which the
arsenic acid radical is supposed to be attached
to the amido group in a similar manner as the
acetic acid is present in acetanilide. Now, as
you all know, the ammonia radical, which in
aniline is very ready to react, has completely
lost this property in acetanilide. This latter
body is in fact almost an indifferent compound.
Similarly, by the view that atoxyl was the
anilide of arsenic acid, any further development
of this compound was rendered impossible. I
was, however, able to prove, together with
Dr. Bertheim, that this compound is to be

G2

84 EXPERIMENTAL RESEARCHES

regarded as something entirely different, namely,
that the amido group is quite free and that the
arsenic acid radical is attached to the benzol
ring in the para position. Atoxyl must, there-
fore, be regarded as the sodium salt of paramido-
phenyl-arsinic acid.

We have obtained a great number of sub-
stances derived from it, and have tested them
therapeutically. Thus we found that, e.g., by
the introduction of the acetyi radical into the
amido group the acetyl-amido-phenyl-arsinic
acid is produced, which is far less toxic for mice
than atoxyl. By means of this substance it is
even possible to solve one of the most difficult
therapeutic problems imaginable. For mice
that have been infected with our most virulent
strains, and which, without treatment, would
die within three days, can in two-thirds of the
cases be saved by this substance if it is given
twelve to fourteen hours before death. Such
curative results have never yet been published ;
for in previous communications there were men-
tioned almost exclusively experiments carried
out at earlier periods of infection. In our cases
acetyl-atoxyl cures mice in nearly 100 per
cent. Unfortunately, acetyl-atoxyl is decom-
posed in other animals, especially in the horse.

ON SPECIFIC THERAPEUTICS. 85

into acetic acid and aniline, and it cannot,
therefore, be nearly as effective as it is in the
mouse. Nevertheless, the results obtained in
the mouse are so excellent that they must
encourage us to energetically pursue the path
of these discoveries.

I mentioned in my last lecture that it is pos-
sible to obtain strains of trypanosomes which
are resistant to these active substances. Studies,
which I have carried out with Dr. Rohl and
Dr. Browning, have shown this resistance to be
of a very high degree. Thus, atoxyl-fast strains
show no sign of being influenced by the highest
doses of atoxyl which are applicable to mice.
I gave you the explanation for this resistance,
viz., that it is the result of a decrease in the
avidity of the trypanosomes for these trypano-
cidal substances. Further, I pointed out that,
apart from this resistance, the parasites' pro-
toplasm, although not their nuclei, had become
hyper-sensitive towards the arsenical prepara-
tion. Thus we meet with an important instance
of association, in one and the same cell, of
resistance and hyper-sensitivity. This associa-
tion, which is of late years proving to be of
increasing importance in the question of
mammalian immunity, is therefore present

86 EXPERIMENT A L RE SEA RCHES

even in so simple an organism as this proto-
zoon.

Although in the mouse atoxyl-resistance usu-
ally develops only after a fairly long period, I
have occasionally seen it after a fortnight. This
observation is very important, because it makes
it possible that the negative results observed
by Ayres Kopke, Broden, Rodhain, Todd and
van Campenhout, in cases of sleeping-sickness
which had received a long-continued treatment,
may also be due to such a resistance.

In the same way as against atoxyl, trypano-
somes can also be rendered resistant to para-
fuchsin, trypan-red and trypan-blue, and further,
it was shown in my Institute by Franke in his
work with apes, that it is even possible for try-
panosomes to become resistant to those pro-
tective substances which are formed in the ani-
mal after its recovery from a trypanosome infec-
tion. A similar discovery has recently been
made by Levaditi for the spirochaetes of relaps-
ing fever. This is therefore a general law, and
it is especially important, because this change
in the trypanosomes, having once become ac-
quired, remains an hereditary property. Thus I
have cultivated for 125 generations my atoxyl-
fast strain, without finding any decrease in its

ON SPECIFIC THERA PE UTICS. 87

power of resistance. We must further note
that every such resistance is specific. Thus
the fuchsin-fast strains are resistant only to this
substance, but not to atoxyl and trypan-red, and
vice versa. This resistance extends, however, to
chemically different substances of the same
group, so that we can in any individual case
decide to which group a given new trypano-
cidal agent belongs. Thus, e.g., the atoxyl-fast
strain is resistant also to a number of related
substances. Among hundreds of trypanocidal
bodies, it is thus possible to single out those
substances which contain arsenic.

It has further been shown that the resistance
to trypan-red and trypan-blue is a mutual one,
in spite of considerable chemical differences
between these two substances, which in fact
have nothing in common with one another
except that they contain naphthalin rings, which
are substituted in a certain position (3, 6) by
sulphuric acid radicals.

Thus we possess in this specific resistance, so
to speak, a therapeutic sieve, a cribrum thempeu-
ticum, with which we can undertake the classifi-
cation of any new chemo-therapeutic substance.
If such a substance is found to have a destruc-
tive effect on our three different resistant strains,

88 EXPERIMENTAL RESEARCHES

it necessarily belongs to a fourth chemical
group. One can, of course, materially simplify
this examination by employing strains that
have been rendered resistant to all the groups
hitherto known. Thus I have, together with
Dr. Browning, obtained a strain which is resis-
tant to atoxyl, trypan-red and fuchsin. By
using such a strain, a single animal experiment
suffices for the purpose of establishing any new
group of such chemo-therapeutic agents. For,
if one injects the substance in question to such a
trebly resistant mouse, the trypanosomes will
either go on proliferating — then the substance
must belong to one of the three known groups
— or the trypanosomes are injured ; in that case
we have before us a new type. As you see, we
are therefore enabled by means of these resis-
tant strains to separate substances of different
modes of action from one another, and thus to
solve a problem which hitherto could not even
be suggested. A way is thus shown us by
which we may enter upon the consideration of
the most intricate problems of pharmacodyna-
mics and, if I may say so, obtain a fuller know-
ledge " de sedibus et causis phannaconwn"

I also wish to lay especial stress upon my
view that the drugs, also, are attracted by and

ON SPECIFIC THERAPEUTICS. 89

bound to the protoplasm molecule by certain
atom groupings. I am inclined to look upon
this as somewhat analogous to the binding of
the toxins and of similar proteid bodies. Yet
on the other hand, there are fundamental differ-
ences between the two. For, as I have always
insisted, the mode of binding the toxins is
peculiar in so far as it is the result of a certain
kind of assimilation which obviously consists in
processes of a more or less synthetic nature.
These toxin-receptors which produce immunity
are bodies of a more independent character,
and appear to be especially destined for pur-
poses of assimilation. This high degree of
independence is evidenced by the fact that, in
conformity with my side-chain theory, these
receptors are very easily reproduced by the cell
in excessive numbers, and after being separated
from the cell, find their way into the blood.

I have now formed the opinion that in like
manner a part of the chemically defined sub-
stances is attached to the cell by groups
corresponding to these receptors ; these atom
groupings I will distinguish from the toxin-
receptors by the name of " chemo-receptors."
This view is more particularly supported by
the fact before mentioned, that the atoxyl-fast

90 EXPERIMENTAL RESEARCHES

strain is also resistant to a number of sub-
stances related to atoxyl, but otherwise showing
widely different chemical characteristics. Evi-
dently, therefore, the arsenic acid radical here
represents the point of attack which is common
to this series of substances. It is the arsenical
radical as such, which is bound by the chemo-
receptors. These chemo-receptors, must, how-
ever, be regarded as of a simpler structure than
the toxin-receptors. They do not show a similar
degree of independence ; they cannot therefore
become increased in chronic intoxications, and
since they are sessile, they cannot be thrown off
into the blood. The number of such chemo-
receptors for poisons which a trypanosome cell
possesses, represents the number of points of
attack. By means of the resistant strains we
can count off one by one these groups that are
open to attack.

The discovery of any trypanocidal substance
will always be the result of chance and experi-
ment. Every point of attack may, of course, be
liable to be attacked by a host of different bodies,
all of which have one specific group in common.
It must be the goal of our experiments to
search, by systematic chemical researches, for
the most suitable substances, and thus to dis-

ON SPECIFIC THERAPEUTICS. 91

cover the "optimum centres." This problem is
by no means an easy one, since according to
my experience of many years, which agrees
entirely with that of Mesnil, different animal
species have different optima, and therefore the
work must be specially done for each species of
animals and each kind of trypanosomes. Thus,
e.g., in the mouse, Trypanosoma gambiense is far
more easily attacked by arsenical preparations
than the other strains of animal-pathogenic
trypanosomes, whilst fuchsin and its relatives
show an exactly opposite behaviour. Still, in
spite of all these difficulties, the problem is not
insoluble, since there are only a limited num-
ber of animal species and of trypanosomes ; we
may, therefore, confidently cherish the hope
that the united forces of the observers who are
at work on this subject in various laboratories
will ultimately succeed in attaining the victory
over sleeping-sickness, which is practically the
most important.

Especially with a view to practical thera-
peutics the knowledge of the different points
of attack is necessary, since it furnishes us
with the possibility of fighting the disease by
the combination of several chemo-therapeutic
agents.

92 EXPERIMENTAL RESEARCHES

The principle of this combinatory method of
therapeutics is, to attack the enemy simul-
taneously from two or three sides, and thus to
produce a completely successful result by the
combination of several different substances,
each of which alone does not show a sufficient
effect. Thus it has been found by Laveran
and by Franke that certain infections which are
influenced neither by trypan-red nor by atoxyl
alone, can be cured if these two substances are
used in conjunction. I, too, have in course of
time repeatedly met with similar cases, and am
convinced that substances which by themselves
are not very efficient, may yet produce good re-
sults if they are used as adjuncts to those more
powerful, though not completely curative, sub-
stances. Supposing, e.g., a highly trypanocidal
remedy to kill ninety-eight among one hundred
parasites, its administration would be followed
by an instantaneous improvement, but not a
definite cure, since the two surviving trypano-
somes must in course of time produce a relapse.
But if one combined this powerful remedy with
another which, even though much weaker, yet
sufficed to kill just those two surviving but
attenuated parasites, then such a combination
would serve to bring about the otherwise un-

ON SPECIFIC THERAPEUTICS. 93

attainable cure. In this respect a publication
from the Liverpool School of Tropical Medicine
is most interesting, as it demonstrates the
successful result of the employment of supple-
mentary doses of perchloride of mercury in
the atoxyl treatment of animals infected with
sleeping-sickness.*

For man the combinatory treatment is of the
greatest importance for a further reason. The
complete sterilisation of the body, i.e., the total
destruction of every parasite, is often only to be
accomplished by the administration of doses
which come very close to the lethal dose. Such
a procedure directly endangering life may be
permissible in the animal experiment, but never
in man. One is, however, justified in hoping
that it might be possible to cure the disease
without endangering life by the simultaneous
administration of three, four or five substances,
chosen in such a manner that their actions are
concentrated on the parasites, whilst in the
organism of the vertebrate host they are dis-
tributed over several different organs.

It is only natural that modern therapy which

* The perchloride in these cases has only the effect of
an adjuvant, since it has been shown that its action on the
trypanosomes, if administered alone, is quite inadequate,

94 EXPERIMENTAL RESEARCHES

is, as a rule, brought to bear upon man under
circumstances making it very difficult to judge
correctly of the processes occurring during treat-
ment, has in the last decades worked chiefly
with simple bodies. On the other hand, I
would remind you that in the prescriptions of
the physicians of the middle of last century, a
predilection for long and complex prescriptions
was evident ; and although nowadays this predi-
lection may appear to us obsolete, yet it must
surely have sprung from a good power of obser-
vation, enhanced by rich personal experience.
Although this former practice of therapeutic
procedure may have overshot the mark, it would
yet appear to contain a nucleus of truth, and
it has certainly received the approval of science
through these recent discoveries.

The importance of combined treatment is
shown, lastly, in the phenomena of resistance
already discussed. The fact that by frequently
repeated administration of not completely
sterilising doses, there is gradually acquired a
resistance to the substance in question, makes it
especially desirable that the first attack should
be as complete as possible. This object, from
what I have previously said, may probably best
be carried out by a suitable combination of

ON SPECIFIC THERAPEUTICS. 95

substances. I would like here to call your
attention to the fact that two decades ago I
discovered a considerable efficacy of methylene
blue against certain forms of malaria. This
dye has not, however, obtained any extensive
employment, since its effect is inferior to that of
quinine. On the ground of our recent experi-
ence one would, I think, be perfectly justified
in attempting to re-inforce the attack on the
malaria parasites by employing combined doses
of quinine and methylene blue.

Having now reached the end of my lecture,
will you allow me once more to call your
attention to the fact that the importance of
these trypanosome studies is a twofold one ;
first, for theoretical reasons, because they afford
us a deep insight into the finest mechanism of
the action of the drugs ; and second, for prac-
tical reasons, since by the methods of research
employed and especially by the development of
the combined treatment, the way has been
prepared for a successful fight against trypano-
some diseases. At the same time we are also
justified in hoping that even beyond these
narrow limits the studies may prove generally
fruitful in the great campaign against infectious
diseases,

206460

LIBRARY

THE UNIVERSITY OF CALIFORNIA LIBRARY