®Ije J. ^ pU pbrarg

QH5TI

NORTH CAROLINA STATE UNIVERSITY LIBRARIES

S00599068 .

/

MEDICAL CONTRIBUTIONS TO THE
STUDY OF EVOLUTION

MEDICAL CONTRIBUTIONS

TO THE

STUDY OF EVOLUTION

BY

J. G. ADAMI

M.D., F.R.S., F.R.C.P.

LONDON
DUCKWORTH AND CO.

3 HENRIETTA STREET, COVENT GARDEN, W.C.

1918

All rights reserved

TO

SIR T. CLIFFORD ALLBUTT

K.C.B., F.R.S., F.R.C.P.

&c.

PHYSICIAN, PHILOSOPHER, FRIEND

WHO AS REGIUS PROFESSOR

PRESIDED OVER THE DELIVERY OF THE EARLIEST

OF THESE STUDIES

IT AND ITS OFFSPRING

A3

.«.\:

PEEFAOE

When at the beginning of the Great War the invitation was
tendered to me to deliver the time-honoured Croonian Lectures
before the Royal College of Physicians of London, it was long
months before I could determine my subject. During these
months the burden of administrative work in connection with the
Canadian Army Medical Service became progressively greater.
Research in the laboratory was out of the question. My official
duties at that period dealt largely with returns and statistics
of invalidism, but soon it became apparent that months,
if not years, must elapse before the vast mass of material
being collected by the Medical Research Committee at
the British Museum would be fully available. Inevitably,
therefore, I was led to fall back upon previous studies,
and, when still debating, a chance discussion with a leading
British biologist convinced me that the time was ripe to
bring together and sum up the conclusions regarding Adapta-
tion which as a student of pathology I had reached gradually
in the years preceding the War. I judged from the discussion
above referred to that that earlier work was not known to
biologists in general. Varied as is his reading and brilliant his
memory, this distinguished biologist was evidently wholly ignorant
concerning it. It seemed also that it would be serviceable to
present the conclusions reached, not so much from the point
of view of their medical bearing, as from that of their biological
significance, in order that both morphologist and physician might
observe the direction in which medical research is surely leading

vii

viii THE STUDY OF EVOLUTION

us with reference to matters that form the basis of general
biology. Hence the Croonian Lectures upon Adaptation and
Disease, delivered in June 1917, which form the first part of
this volume.

Elsewhere I have dealt with this subject of adaptation for
the benefit of the student of medicine ; 1 in dealing with the
same subject more from the point of view of the biologist, I
had of necessity to refer to and repeat data and deductions
employed in my earlier writings. The views here enunciated
have been arrived at in orderly sequence from the year 1891
onwards, and it has been difficult to review that sequence with-
out employing the facts and arguments by which the successive
steps were attained.

While making this confession I have at the same time found
myself confronted with the difficulty that, in compressing my
treatment of the subject into four lectures, much that should
have been dealt with in fuller detail, and with greater wealth
of examples, has owing to the exigencies of space and time
been unavoidably passed over very rapidly. To remedy this,
and at the same time to demonstrate that no newly developed
doctrine is here taught, but one arrived at and published many
years ago, it has seemed advisable to reprint in their entirety 2
certain earlier papers and addresses, all bearing upon adapta-
tion and tissue modification, which establish and amplify the
successive stages of my argument. These are useful as documents
in the case. Personally, re-reading them after a lapse of many
years, and so with some sense of detachment, they have struck
me as to a certain extent entertaining. May their other readers
have the same opinion, and obtain a like satisfaction !

It may be held that an apology is due to not a few well-known

1 Notably in the two editions of my Principles of Pathology in 1908 and
1910, but those are works of some thousands of pages ; it is perhaps natural
that biologists in general have been deterred from consulting them.

2 Save for slight structural alterations, advisable in converting delivered
addresses from the spoken to the printed form. These alterations have been
duly noted whenever it has seemed possible that there has been introduced
thereby the slightest departure from the sense of the original.

PREFACE ix

British biologists, many of them for long years my personal
friends, on the ground that my Croonian Lectures implied that
I regard them all as " academic." This, let me hasten to assure
them, is very far from being the case. If these lectures be read
with reasonable care it will be seen that no such charge is made.
I was concerned not to cover the whole ground of modern in-
vestigations upon the problems of evolution, but to show the
bearing of medical research upon matters pertaining thereto.
I realize full well that it would have strengthened my case to
refer to the work and views of Walter Heape, Professors E. W.
MacBride, A. Dendy, and M. Hartog — to mention only those
known to me from old association — as confirming my earlier
conclusions. I realize this after the event more fully than
before, when I had in mind others who from their position and
reputation have, rightly or wrongly, been regarded popularly
as yet more leading representatives of British biology.

Nor had I in mind American biologists. To have referred
to them would have necessitated entrance at length into the
Mendelian controversy, for all the morphologists in the States,
I am inclined to think, are at the present moment keenly engaged
over this matter. Possibly in the future, I shall have a word
to say upon the important studies of Castle, Osborne, Whitman,
Morgan, Jennings, and others, and the interpretation of the same
in the terms of the biophoric concept. In the meantime may
I assure my American confreres that I cannot call to mind one
of them whom I would accuse of being " academic." On the
contrary, never has there been a period when American biologists
have manifested so great an individual independence, so notable
a diversity of opinion, so pronounced a desire to arrive individu-
ally at the truth, irrespective of the schools and their teachings.

The time indeed has passed when any biologist should justly
be labelled as either a Lamarckian or a Darwinian. With
abundant material presented to him and freedom of individual
judgment, it is scarce possible that the student of to-day should
accept unreservedly the teaching of either Lamarck or Darwin.

x THE STUDY OF EVOLUTION

He who is concerned at arriving at the truth is impatient of such
labels : it is the truth he seeks, not the confirmation of the views
of any one predecessor, however great be the admiration for his
work and however large the debt he owes to him.

Yet as regards the ancient controversy, this perhaps should
be said, that in so far as between Darwin and Lamarck the
essence of the teaching of the latter is that variation is an active
process, a reaction on the part of living matter to its environment,
the conclusions reached in these pages undoubtedly favour the
Lamarckian view. Nevertheless, to accept them does not mean
that the principle of natural selection is thereby excluded, or that
the two principles are mutually antagonistic, but only that the
influence of external forces is the primary process in the produc-
tion of variation, and that natural selection is secondary, culling
out those grades and forms of variation which are least economical
and represent the less perfect adaptation on the part of individuals
to the conditions in which the family or species finds itself for
the time being. Seen thus, evolution, whether what we regard
as progressive or as regressive, is the outcome of an active
process of continuous adjustment between organisms and their
environment.

The survival of the fittest, it will be seen, does not depend
upon chance variation. A given environment leads to variation
in a particular direction, provided that the change in sur-
roundings is not so great as to be beyond the adaptive powers
of the organism. Where " chance " enters is in the nature of
the new environment to which the individual, and the race,
may be exposed. To the extent that the individual is unable
to control his surroundings, to that extent is the race exposed
to chance. It does not appear to have been sufficiently realized
hitherto that here essentially it is that chance is operative.
Conjugation and amphimixis, it is true, are a cause of individual
variation, but from the point of view of the race are distinctly
conservative processes, tending to maintain the mean.

When the first of the series of Croonian Lectures was pub-

PREFACE xi

lished in the pages of the British Medical Journal, together with
an abstract of the second, the biologist whose attitude with
reference to adaptation was the origin of those Lectures, saw fit
to criticize them more suo. That my readers, with the Lectures
before them, may reach their own conclusions regarding the
validity of the criticism, it has been deemed proper to publish
the correspondence as an Appendix.

It is my pleasant duty to acknowledge my indebtedness to
Messrs. Macmillan & Co. for their permission to reprint here
the greater portion of a chapter from an article published origin-
ally in Sir T. Clifford Allbutt's System of Medicine, and republished
subsequently in book form under the title Inflammation ; to
Messrs. Lea Brothers & Co. (now Messrs. Lea & Febiger) of
Philadelphia, for permission similarly to publish an extract
from an article on " Inheritance and Disease " contributed to
Sir William Osier's Modern Medicine, to the Harvey Society
of New York, and the Editors of International Clinics, British
Medical Journal, the Lancet, New York Medical Journal, Mon-
treal Medical Journal and its successor the Journal of the
Canadian Medical Association, the Journal of Pathology and
Bacteriology, Medical Chronicle, and Clinical Journal, to reprint
articles contributed to their pages. To my friend, Captain
R. G. Mathews, C.A.M.C., I am indebted for kindly co-operation
in the preparation of the illustrations.

London, August 1917.

CONTENTS

PAET I
ADAPTATION AND DISEASE

BEING THE CROONIAN LECTURES DELIVERED TO THE ROYAL
COLLEGE OF PHYSICIANS, LONDON, JUNE 1917

CHAPTER I

PAGE

The Basal Problem of Evolution 3

CHAPTER II

Adaptation in the Bacteria and the Evolution op the

Infectious Diseases 15

CHAPTER III

Adaptation in the Bacteria {continued) . . . . 28

CHAPTER IV

Adaptation to Disease-producing Agencies in the Higher

Animals 45

CHAPTER V

The Inheritance of Acquired Conditions in the Higher

Animals .60

xiii

aiv THE STUDY OF EVOLUTION

CHAPTER VI

PAGE

The Physico- Chemical Basis of Immunity and of Evolution.

The Biophoric Concept 70

CHAPTER VII

Review of Certain Conclusions resulting from the Bio-
phoric Concept ........ 83

PART II
HEREDITY AND ADAPTATION

CHAPTER I

On the Variability of the Bacteria and the Develop-
ment of Races (1892) 103

CHAPTER II

On Theories of Inheritance with Special Reference to

the Inheritance of Acquired Conditions in Man (1901) 132

CHAPTER III

Adaptation and Inflammation (1905) 161

CHAPTER IV

The Myelins and Potential Fluid Crystalline Bodies of

the Organism (1906) 168

CHAPTER V

The Dominance of the Nucleus (1906) . . . . 188

CHAPTER VI

The Reductio ad Absurdum of Weismannism (1907) . . 210

CONTENTS

xv

CHAPTER VII

PAGE

A Lecture on Life (1909) 216

CHAPTER VIII

On Habits, Symptoms, and Disease (1911) .... 235

CHAPTER IX

Parenteral Digestion and Immunity (1914) . . . 250

PART III
ON GROWTH AND OVERGROWTH

CHAPTER I

On Growth and Overgrowth, and on the Relationship
between Cell Differentiation and Proliferative
Capacity (1900) 263

CHAPTER II

On the Causation of Cancerous and other New Growths

(1901) 276

CHAPTER III

On the Classification of Tumours (1902) .... 305

CHAPTER IV

Syncytioma (Deciduoma) Malignum : its Bearing upon the

Essential Nature of Malignancy (1902) . . . 326

CHAPTER V

Unipotentiality, Pluripotentiality, and Totipotentiality
of Cells : a Note upon the Classification of Tumours
(1907) 334

xvi THE STUDY OF EVOLUTION

CHAPTER VI

PAGE

On the Eelationship between Tumours Proper (Blastomas)

and Hyperblastosis (1913) 340

APPENDIX I

The Experimental Production of Specific Variation in
Typhoid Bacilli, by F. B. Bowman, M.B. (Tor.), Major,
C.A.M.C 351

APPENDIX II

Sir E. Ray Lankester rebukes Rudeness : a Correspond-
ence in the Pages of the British Medical Journal . 354

Index 363

ILLUSTRATIONS

PLATES

PLATE T0 FACE pAGE

I. Fossil and recent blue-green Algae (after Walcott) . . 18

II. The Myelins : doubly refracting fatty globules from the adrenal

cortex and liver 168

III. The Myelins : Myelin forms from soap ; Ductile Crystals

(after Lehmann) . . . . . . . .176

IV. The Myelins : Distortion of fond crystals and aberrant forms

of myelin globules . . . . . . .186

V. Schema of the stages of fertilization of the ovum (after

. Boveri) 192

VI. Schema of tissue relationships . . . . . .312

VII. Diagram of a lepidic and a hylic tissue . . . .316

FIGURES IN THE TEXT

FIG. PAGE

1. Diagram of chemical constitution of an amino-acid nucleus of
a protein ........

2. Diagram of constitution of a protein ring

3. Diagram of the affinities of a compound of two carbon atoms

(after Bayliss) .......

4. Diagram of cell to show nuclear skein with chromomeres .

5. Diagram of mode of growth and deduplication of the biophoric

molecule ........

6. Diagrammatic representation of the nature of enzyme action

7. Diagram of simple dissociation of peptone molecule

8. Diagram of interaction of substratum and receptor through inter

mediation of enzyme ......

xvii

75
76

77
78

81
86

87

87

XV111

THE STUDY OF EVOLUTION

FIG. PAGE

9. Diagram of alteration in constitution of bioplioric molecule . 89

10. Graphic formula of the Pyrimidine ring . . . .89

11. Diagram of production of multiple side-chain elements . . 90

12. Diagram of interchange of allelomorphic side-chains between

biophores of paternal and maternal origin . . .94

13. The different forms of chromosomes in the human spermato-

cyte (after Moore and Arnold) . . . . .192

14. Cells of the salivary gland to show production and discharge

of " plasmosomes " (after Maximow) . . . .199

15. Aberrant mitoses (after v. Hansemann and Pianese) . . 201

16. Diagram of constitution of protein ring .... 207

17. Diagram of relationship between nuclear biophores and cyto-

plasmic proteins . . . . . . . .208

18. Chromosomes, chromomeres and " ids " . . . . .214

19. Diagrammatic representation of constitution of protein or

peptone nucleus . . . . . . . .222

20. Diagram of growth of biophoric molecule .... 228

PART I
ADAPTATION AND DISEASE

bv thelact tP-NTRIBUTI^OF MEDICAL RESEARCH TO

THE STUDY OF EVOLUTION

_ 4JU

CHAPTER I

INTRODUCTORY : THE BASAL PROBLEM OF EVOLUTION

I have selected this subject of Adaptation aud Disease, not
merely because of its importance to physicians, but also, and
to an even greater extent, because of its broad biological signifi-
cance. The time is ripe, and more than ripe, for attention to
be directed to the bearing of the investigations of the bacterio-
logist on the one hand, and of the student of immunity on the
other, upon what are some of the most important problems of
general biology. For some little time I have been impressed
by the fact that the latter-day investigations in medical science
are of the very highest significance to the general biologist,1
and that with singularly rare exceptions the professional biologist
— be he zoologist or botanist — has been superbly indifferent to
them and to their bearing upon the basal problems of heredity
and variation, and this notwithstanding the fact that investiga-
tions into heredity and variation are, and must always be, his
greatest concern.

For this indifference and neglect there are, or may be, several
extenuating circumstances. We ourselves are largely to blame
in that we are more concerned with the bearing of our results
upon our own medical work than with their wider biological
significance. This wider significance, if it is referred to at all,
finds incidental note. The titles of our articles, that is, are
not such as attract the attention of the biologist, nor do we
publish in the journals to which he is addicted. Not to mention
foreign publications, the Journal of Pathology and Bacterio-
logy, the Journals of Experimental Medicine, of Hygiene, In-

1 Five years ago this was the subject of my opening address as President of
the Biological Section of the Royal Society of Canada. Trans. Roy. Soc.
Canada, 3rd series, 1912, vol. vi. section 4, p. 1.

3

■-.•<;'■ V'

K C. St®

4 ADAPTATION AND DISEASE

fectious Diseases, of Medical Research are as unknown territory
to him as the Quarterly Journal of Microscopical Science, the
Journal of Morphology, or the Proceedings of the Linnean Society
are to us. We would sooner think of writing a letter to the
Times than to that academic go-between of men of science,
Nature.

And then, again, just as, in the old unhappy far-off days,
a Bart's man disbelieved that any good surgery could come
out of Thomas's or George's, and the Thomas's or George's
surgeon returned the compliment ; just as the British medical
man disbelieved such articles as appeared in the American
medical journals, or, to come nearer to our own times, just as
the Germans of late years, with rare and distinguished excep-
tions, regarded themselves as the sole recipients of pathological
truth, treating British, French, and American pathologists in
general as very Nazarenes, so, it has to be confessed, there is
a tendency for the academic biologist to be indifferent to, if
not actually to resent, and throw discredit upon the work of
those who, not belonging to his particular class, are therefore
to be regarded as of the irature of outsiders. This is not a
revelation of the spirit of pure science, but a comforting demon-
stration that men of science are, nevertheless, pure human beings.
Underlying this spirit is a natural and in many respects wise
caution in accepting the observations of workers with whose
quality and standing the individual is unfamiliar. But this
hesitancy may be carried too far.

If I speak with a little feeling, it is because I still cannot
forget the reception accorded by zoological confreres to the
most original, and at the same time most sound physiologist
of his period, my old teacher and friend Walter Gaskell, when
he was led by his studies upon the functions of the nervous
system — studies which have so profoundly influenced modern
medicine — to trace the development of the same, and doing this,
after long years of close study of its comparative morphology,
to reach conclusions regarding the origin of the vertebrate
which were not in harmony with the doctrines of descent then
currently accepted. So far as I, an outsider, could determine,
each link in the chain of Gaskell's reasoning was supported by
appeal to observed facts, and by microscopical studies of singular
interest ; so far as I, a pathologist, could test his conclusions,

WALTER GASKELL 5

I found that incidentally they explained, as no other or earlier
work had explained, the inter-relationship between the sym-
pathetic system and the endocrine system, between the pituitary,
thyroid, adrenals, and genital organs. Nor, so far as I can weigh
evidence, can I find that any essential link in the chain has
been shown to be out of place. Yet the attitude of the morpho-
logists as a body was that of the Levite of the parable. His
brother physiologists could not take up the cudgels on Gaskell's
behalf : it was for the morphologists to determine the value of
the morphological evidence upon which his conclusions were
based, and the morphologists in general declined to notice it,
but, as though they regarded it as presumptuous for him, a
physiologist, to enter their territory, they passed by on the
other side. And the years followed the years, and Gaskell
died feeling sore that the most sustained piece of work of his
life had been side-tracked by those whom it should most have
interested.

Do not misunderstand me, and think that I am making a
specific attack upon zoologists and botanists. As I have pointed
out, we of the medical profession are tarred with the same brush.
All I would urge is that just as we who are most interested in
the advance of medical science accept gladly the results and
the discoveries of workers in all branches of science, applying
them to the elucidation and treatment of disease, so, in return,
when investigators into the problems of medicine make notable
advances, these be accepted willingly and utilized by the workers
in other branches. It is not a matter of what we owe to those
other branches in the first place ; that is freely admitted. The
renascence of medicine in our generation is due to the labours
of men like Ferdinand Kohn the botanist, Pasteur the physical
chemist, and Metchnikofi the zoologist, but if the dwarf, perched
on the shoulders of the giant,1 sees further and sees more than
does the giant, it is not well to neglect his observations on the
ground that he is a dwarf.

1 As might be expected, this metaphor was not original with Burton in the
Anatomy of Melancholy. I rind that Firmin-Didot (Aide Manuce, Paris, 1875,
p. 343) cites Haureau (Histoire de la philosophie scholastique, Paris, 1872) as
deriving it, according to Joannes Sarisberiensis (Metalogicus, vol. iii. cap. 4)
from Bernard of Chartres in the twelfth century.

6 ADAPTATION AND DISEASE

The Nature of Variation

The supreme biological problem of our times has been that
of the ways and means of evolution. The fact of evolution
all thinking minds accept. But as to how evolution has been,
and is being, brought about is a very different matter. There
is still as much debate as there was in the year following the
publication of the Origin of Species. Upon consideration
it will be seen that the fight truly centres upon the cause or
causes of variation — whether the tendency to vary is something
inherent in living matter, numerous variations being produced
through this inherent tendency, of which those that are best
fitted for their environment alone survive and are perpetuated :
or whether variation is primarily and essentially brought about
by the influence of forces acting from without upon a relatively
labile living matter ; whether, that is, variation 1 is primarily
inherent, proceeding from within, or primarily acquired, proceeding
from without.

This, I would emphasize, is the basal problem of evolution,
but oddly enough it has been largely neglected, the fight through
all the years waging around what, after all, is a secondary problem,
that, namely, as to whether properties acquired by the parent
are capable of being transmitted to and reproduced in the off-
spring. Long years prior to Darwin, you will remember, Lamarck
propounded that they were, as did Erasmus Darwin and Lord
Monboddo ; Darwin wanted to believe that this was possible,
but could obtain no clear evidence, and brought in finally a
verdict of " non proven." Herbert Spencer made this trans-
mission one of his " principles of biology," but Weismann
violently opposed the doctrine, carrying with him the bulk of
latter-day biologists, until to-day Bateson, replete with his
studies upon Mendelian properties, reaches the antipodal sugges-
tion that when a new property manifests itself in any individual
of any species it is not new, not due to addition, but to subtrac-
tion and loss of properties already possessed : there is nothing
new under the sun, and, in his opinion, evolution — like the squid —
progresses backwards ; what appears to be a new property is, on

1 By an oversight this was delivered as " variability " at the Royal College
of Physicians and so printed in the pages of the British Medical Journal. Pace
Sir Ray Lankester (for whose criticism consult the correspondence at the end
of this volume), my meaning was as obvious as was the oversight.

/

WILLIAM BATESON 7

the contrary, primaeval, is the epiphany of a property possessed
by the forerunners of the species which all along has lain latent,
present but unable to manifest itself in consequence of the co-
existence of inhibitory factors.

That I may not be thought to exaggerate or misstate, let me
quote, in the first place, from his address on Heredity delivered
to the International Medical Congress here in London in 1913 :

" Perverse as such a suggestion may appear, I do not think
we should close our minds to the possibility that these dominants
arise by a process of loss of some inhibitory factor. . . . Let me
call your attention also to the inference which this suggestion
would have on the conception of evolution. We might extend
the same reasoning to all cases of genetic evolution, and thus
conceive all alike as due to loss of elements present in the original
complex."

That clearly this was not a passing fancy is shown by the
address delivered by him at Melbourne in the summer of 1914,
as President of the British Association for the Advancement
of Science.

" We are even more sceptical," said Mr. Bateson, " as to
the validity of an appeal to changes in the conditions of life
as direct causes of modification, upon which, latterly at all
events, Darwin laid much emphasis. ..."

" Abandoning the attempt to show that positive features can
be added to the original stock (the italics are mine), we have further
to confess that we cannot often actually prove variation by
loss of factors to be a real phenomenon." Nevertheless this
must be so, and he quoted the case of the " Coral King " Primula
given off from the " Crimson King," concluding that here " the
salmon (pigment) must have been concealed as a recessive from
the first origin of the variety," and continued : " Variation both
by loss of factors and fractionation of factors is a genuine
phenomenon of contemporary nature. If we have to dispense,
as seems likely, with any addition from without, we must begin
seriously to consider whether the course of evolution can at all
seriously be represented as an unpacking of an original complex
which contains within itself the whole range of diversity which
living things present." And further :

" At first it may seem rank absurdity to suppose that the
primordial form or forms of protoplasm could have contained

8 ADAPTATION AND DISEASE

complexity enough to produce the divers types of life. But
is it easier to imagine that these powers could have been con-
veyed by extrinsic additions ? Of what nature could these
additions be ? Additions of material cannot surely be in question "
(the italics are mine). . . . And he winds up :

" In spite of seeming perversity, therefore, we have to admit
that there is no evolutionary change which in the present state
of our knowledge we can positively declare to be not due to
loss."

Now Professor Bateson is scarce a believer in the multiple
origin of animal and plant forms : we know, indeed, that years
ago he traced man and all vertebrate forms from the inverte-
brates through an out-of-the-way animal, Balanoglossus, and
incidentally in his Melbourne address he mentions this solution
of the problem only to reject it ; wherefore, pushed to its logical
conclusion, the Batesonian doctrine means this, that the primal
unit, or units, of protoplasm from which all living animal and
plant forms have descended possessed within them in a latent
form the " Anlagen," or, not to be beholden to our enemies,
the originals, of every organ and distinctive portion of an organ
or part, even down to the conformation and coloration of in-
dividual hairs and scales, and feathers and leaves, and petals
and stamens, of all the manifold forms of life subsequently
derived therefrom : that which was to outward seeming the
most simple form of life was, verily, in constitution the most
marvellously complex — and that actually what we regard as
the higher forms of life are the lower, owing their development
not to progressive accretions of properties, but to the reverse,
so that reversion, instead of being a degenerative manifestation,
a loss of properties acquired by the species, is, on the contrary,
a recovery of higher and completer powers. Did ever any
exercise of mediaeval scholasticism lead to more perverse con-
clusion?

The truth seems to be that Professor Bateson and the Men-
delians, so far as regards the problems of evolution, are working
in a cul-de-sac. Valuable and fascinating as are their observa-
tions for the establishment and amplification of the law dis-
covered by Gregor Mendel of cross-breeding of members of a
species, that law only deals with the interplay of allelomorphs,
id est, with the combinations and permutations of what for

WILLIAM BATESON 9

simplicity sake may be termed positive and negative unit pro-
perties possessed by the species : it only establishes the extent
of the variation possible within the boundary of the species,
and granting the existence of a definite number of allelomorphs,
the number of possible strains obtainable within the limits of
the species. But here it stops — save that of late workers have
recognized the possibilities of fractionation of allelomorphs, and
so of increase in the number of permutations and combinations.
Accepting Mendelian data (and let me say I accept them whole-
heartedly), I fail to see how any amount of interplay between
properties already possessed by the species will result in the
production of individuals which are outside the species. At
most we produce different strains which, by cross-fertilization
with other strains, produce individuals reverting to the usual
type or types. If you invest in a kaleidoscope at a toy-shop,
with the mirrors set at an angle of 60°, no amount of
rotation will produce other than a six-sided pattern, or increase
the number of colours in the pieces comprising the pattern.
Only from without can new elements of other colour be added,
thereby producing patterns of a new order : only from without
can the angle of the two mirrors be altered so as to produce,
say, a four- or a twelve-sided pattern. The interplay of allelo-
morphs is not evolution, nor is it capable of throwing light
upon the progressive development of new species. When
Professor Bateson, from the vantage ground of his studies of
the last fifteen years or so, begins to lay down the law regarding
evolution, I cannot but help being reminded of Bombus, the
bumble-bee — I would not say " in vacuo bombinans," for that
was said of a zoological monstrosity, and Professor Bateson is
no chimsera,1 but — blundering out of the fields and hedgerows
into a greenhouse, and bumping its head noisily again and again
against the glass because of its incapacity to drive into that
head the fact that transparency and penetrability are not neces-
sarily associated phenomena.

Nor is he alone. It so happens that some eighteen months
ago, at the request of a mutual friend, a Fellow of the Royal
College of Physicians, I found myself drawn into a lively dis-
cussion with the Nestor of British biologists. I had not sought

1 Rabelais, Pantagruel, bk. ii. cap. 7. I owe the reference to my friend
Mr. Louis Taylor.

10 ADAPTATION AND DISEASE

the fight, but, once in, confess that I enjoyed the oppor-
tunity of measuring my lance against so doughty a knight.
The discussion arose over a statement by the Fellow in
question that the biologists of to-day had much to learn by
turning their attention to the results being gained in medical
laboratories. And, sure enough, after protesting with not a
little vigour that the boot was on the other foot — that we medical
men in our ignorance of biological progress were rediscovering
and regarding as our own what had been gained by zoologists
and botanists a generation or so previously — Sir Ray Lankester
proceeded to justify absolutely my friend's original contention
by urging that one fallacy in all Lamarckian doctrine 1 is that
adopted by Herbert Spencer, namely, what he called " direct
adaptation." There is really, he laid down, no such thing.
The supposed mysterious, and as it were miraculous, property
of direct adaptation is always due to survival by selection of
organisms which varied in many directions — the production of
corneous epithelium, of increased hairiness, etc., being favourable
variations, which hence have become inherent in tissues of all
animals.2

1 I confess that I do not like being dubbed a Lamarckian, and that because,
as commonly accepted, Lamarckianism is supposed to deal purely with the
direct acquirement of alteration in structure through use and environment,
and Herbert Spencer, by using the term " direct equilibration," is largely
responsible for this vulgar error. The phenomena we pathologists deal with
present modification of structure merely as a secondary change : our pheno-
mena underlie structural alteration. But in justice to Lamarck it deserves
note that he expressly lays down " whatever the environment may do, it does
not work any direct modification in the shape and organization of animals.
But great alterations in the environment of animals lead to great alterations in
their needs, and these alterations in their needs necessarily lead to others in
their activities. Now if the needs become permanent, the animals then adopt
new habits, which last as long as the needs that evoked them " — and it is the new
habits, he points out, which induce the structural alteration. (I quote from
Hugh Elliot's excellent translation of the Philosophie zoologique — Lamarck,
Zoological Philosophy, Macmillan, 1914, cap. 7.) Granting this, I hold that the
physicochemical explanation which will be put forward in the course of these
lectures is something more precise and more limited in its scope than the
" habits " of Lamarck.

2 As Sir Ray Lankester has accused me of garbling his statement (see
Appendix II.), and has not seen fit to withdraw the charge, I have no hesitation
in quoting his exact words : " One fallacy in all Lamarckian doctrine is that
adopted by H. Spencer, viz. what he called ' direct adaptation.' There is
really no such thing. The supposed mysterious, as it were miraculous, property
of direct adaptation is always due to survival by selection of organisms which
varied in many directions — the production of corneous epithelium, of increased
hairiness, etc., being favourable variations, and hence have become inherent in
tissues of all animals. But no more ! "

RAY LANKESTER 11

Here then we see the two foremost British biologists of our
day, the one in doubt whether change in environment can be
a direct cause of modification, and, filled with these doubts,
willing to accept as a postulate that positive features cannot
be added to the original stock, whereby he is led to an utterly
perverse hypothesis ; the other equally denying that there can
be external influences of such a nature that specific variation —
i.e. variation in particular directions — may be induced, and
taking the stand that variation is multitudinous, the favourable
variation alone having the opportunity to be propagated and
reproduced.

Now if there be one fact that is constantly being impressed
upon the student of immunity and the worker in pathogenic
bacteriology, it is that " direct adaptation " (i.e. specific modifica-
tion in response to a specific alteration in environment within
limits which will presently be laid down) is one of the basal
phenomena of living matter. Our studies make it impossible
for us to be blind to the fact that environment is capable of
exerting a profound influence upon living beings, bringing about
modifications of function and even of structure in particular
directions. But evidently our experience and the diverse obser-
vations upon which that experience is based are unknown to the
academic biologists, although some biologists are beginning to see
light. Thus it delighted me to run across the following admission
in a valuable work published within the last few weeks — Pro-
fessor D'Arcy Thompson's Growth and Form i1 " So long and
so far as ' fortuitous variations ' and the ' survival of the fittest '
remain engraved as fundamental and satisfactory hypotheses
in the philosophy of biology, so long will these ' satisfactory
and specious ' causes tend to stay ' severe and diligent enquiry '
to the great arrest and prejudice of future discovery. " With
this strong opposition on the one hand, and encouragement on
the other, it has seemed to me a useful task to bring together
and marshal in order the data bearing upon these matters as
they present themselves to us, workers in medical science.

1 Growth and Form, by D'Arcy W. Thompson, F.R.S., Camb. Univ. Press,
1917, p. 6.

12 ADAPTATION AND DISEASE

The Method of Attack

There are two ways by which problems of this nature are
a priori most likely to be solved, namely, by experiments upon
the very simplest and, again, upon the most complex forms of
life. I see that Dr. Bayliss as a physiologist casts doubts upon
the value of the former as compared with the latter,1 pointing
out that their very simplicity is in the majority of cases a dis-
advantage, and quoting Claude Bernard to the effect that the
lower (unicellular) forms of life possess all the essential pro-
perties which exist in the (multicellular) forms higher in the
scale, but possess them in a confused state, distributed, as it
were, throughout the organism. The one organic cell fulfils
a variety of purposes which in the higher organisms are relegated
to distinct groups of cells. While freely admitting this conten-
tion as regards the study of f miction, it has to be pointed out
that for problems of adaptation and heredity the unicellular
organisms possess the supreme advantages of rapid reproduction,
coupled in the very lowest forms (according to our present
knowledge, or want of knowledge 2) with a complete absence
of the disturbing influence of sex and conjugation. There is,
that is, a greater likelihood of obtaining results, and the experi-
ment becomes simpler, where we can in the course of a few hours
subject one hundred generations to a particular alteration of
environment, than when weeks, months, or years elapse before
one new generation shows itself. In the latter case, to obtain
results, either the alteration of environment must be made
intensive to a degree that is likely to interfere with various
vital functions, or, exhibited in a less intensive form, must act
over unduly long periods. I know that certain biologists are
unwilling to regard the products of asexual binary division of
the bacteria, or the torulae of yeast (the results of budding),
as true generations, and deny the right to regard the individual
bacillus as an individual. One very distinguished biologist
went so far as to declare to me that a long cultivation of a
bacterial growth is " one continuous individual," in other words,

1 W. M. Bayliss, Principles of Physiology, 1915, p. 291.

2 For myself, with sex so widespread an attribute of living beings, I confess
that I am wholly prepared to find that the schizomycetes, or some of them,
exhibit a sexual, or conjugation, phase.

THE METHOD OF ATTACK 13

that not merely do all the millions of bacilli in a single colony,
say of the Bacillus tuberculosis, constitute one individual, but
that one individual Bacillus tuberculosis is spread all over the
habitable world, and that tuberculosis and not the tubercle
bacillus is the entity. This is an impossible position : it formu-
lates that the " divisa " are " indi visum." The very idea of
individual connotes independent existence, or, if we take a wide
survey and include forms like the compound myxomycetes,
hydrozoa and polyzoa, potentially independent existence. When
a bacillus grows and divides and each half floats away, there
is no longer one individual, and whether, as among the bacteria,
the division is binary, or, as in man, one of the many billions
of cells that constitute the body midergoes a similar binary
division, and one of the products floats away and becomes
fertilized, in both cases we deal with the development of a new
generation. By the same process of reasoning, basing our-
selves upon the doctrine of the continuity of the germ plasm,
we might with equal logic declare that all living beings constitute
one continuous individual !

And as regards the advantage in researches upon the bacteria
of being freed from the perturbing factor of sex, let me interject
that starting with Weismann and his doctrine of amphimixis,
but more particularly during the last fifteen years under the
influence of the Mendelians with their studies upon cross-breeding,
we have been inclined to lay far too much stress upon the part
played by sexual conjugation in the production of variation.
In nature, in general, the tendency, if not the function, of sexual
conjugation is to preserve the mean, not to induce the extreme,
is to perpetuate the species rather than favour the variety :
sexual conjugation only preserves and intensifies a variation
when circumstances favour the segregation of individuals of
the two sexes each possessing the variation, or, I would add,
when those circumstances actually lead to the appearance of
the variation in more than one individual in a particular locality.
Even if a property be dominant, with indiscriminate mating
and without segregation, such new property is apt to show itself
proportionately in fewer and fewer individuals in successive
generations. In simpler language, turn a highly-bred mastiff
or terrier or beadle loose in a canine population, and his de-
scendants descend to plain " yellow dog." It follows, therefore,

14 ADAPTATION AND DISEASE

that on the whole we are more likely to obtain results where
this levelling action of conjugation is wanting.

I propose, therefore, to take up first the evidence of adapta-
tion as affecting the pathogenic bacteria, next, of adaptation as
it affects man and the higher animals, and, lastly, to discuss the
application of the data brought forward to our conception of
disease and disease processes on the one hand, of the evolutionary
process on the other.

CHAPTER II
adaptation in the bacteria

The Evolution of the Infectious Diseases1

It is absurd in these days to imagine that the infections have
always been with us : absurd to expand the Batesonian hypo-
thesis and imagine that when the woman gave the man the
fruit of the tree — or whenever man became man — with that
most becoming knowledge of his nakedness he acquired the
germs of all bodily ills : that while lues may already have in-
fected his ancestors and their relatives the higher apes, the
germs of typhoid, cholera, gonorrhoea, and other purely human
ailments were already there, only waiting for his appearance
to enter into his body, as into a house newly swept and garnished.

The Antiquity of Zymotic Diseases

Admittedly some diseases have been with us from the remotest
historical times, aye, and we have evidence, from prehistoric
times. The earliest Greek medical writings afford us an un-
mistakable picture of tuberculosis. If we may accept Bishop
Ussher's chronology, the fifth and sixth chapters of the First
Book of Samuel show — as I believe I was the first to point out,
in the light of our modern knowledge of the natural history of this
disease 2 — that the Oriental plague was active over three thousand
years ago, presenting the same striking characters as it manifests
to-day. At a still earlier historical era, in fact in the earliest

1 This chapter follows the lines (though with considerable modification) of
an address given by me before the Medical Society of the State of Vermont at
its annual meeting at Rutland, Vt., in Feb. 1915, and pubb'shed in the 1914
volume of the Transactions of that Society.

2 Montreal Medical Journal, xxiv., 1896, 995.

15

16 ADAPTATION AND DISEASE

of all known writings, namely in the Ebers papyrus, descrip-
tions are given of leprosy, and the Ebers papyrus is said to be
of the twelfth dynasty, so that leprosy has been known for
some four thousand years.

It is through my old friend, Sir Armand Ruffer, whose recent
death we deplore, a man who accomplished much for Egypt,
and through his studies in what he described as paleopathology,
that we owe the knowledge of prehistoric disease in man.
Mummies of yet earlier dynasties — and pre-dynastic — said to
date back as far as four thousand years B.C. or six thousand years
ago, demonstrate the most respectable antiquity of Bilharziosis,
not to mention such everyday disturbances as Rheumatoid
Arthritis and Pyorrhoea alveolaris. The characteristic eggs of
Bilharzia are still distinguishable in the dried-up tissues of
pre-dynastic mummies, so that anaemia and haematuria con-
stituted a plague of Egypt before the Pharaohs as they do to-day
when the Sultans and their suzerainty have passed away.1
Tuberculosis, too, was there already. He demonstrated the
bacilli in tissues in mummies from the Herst collection at Cairo
belonging to the twentieth dynasty or so.

Diseases which are not peculiar to man have indeed been
diagnosed in fossil remains. Thus, according to Moodie,2 caries
has been noted in Permian fishes, pyorrhoea in the jaw-bone of
an early tertiary three-toed horse, and arthritis and osteomye-
litis in the remains of cave bears.

The Antiquity of the Bacteria

This, after all, is only what might be expected : these lower
forms of life preceded mammals and man, and bacteria must
have been among the very early forms of particulate life. So

1 Just as the earliest known written document is medical, so my late col-
league in Anatomy at McGill University, who, while this is passing through the
press, from being Director of Recruiting and Brigadier-General has become
Minister of National Service and Sir Auckland Geddes, has impressed upon me
that the earliest known human fossil is pathological, pointing out that the
remarkable thickness of the Piltdown skull, coupled with the characteristic
outline of the temporal ridge, can only find their explanation by a diagnosis of
Acromegaly, and suggesting that it is thanks to this disease and its results that
we owe the survival of these remains through the ages. We do not, however,
claim Acromegaly as a result of zymotic disease.

2 Trans. Chicago Pathological Society, x., 1916, 84.

ANTIQUITY OF THE BACTERIA 17

long ago as 1879, Van Tieghem called attention to the evidence
of bacterial existence, if not the actual presence of bacteria, in
the silicified vegetable remains from the coal measures of St.
iStienne. Since then, in a series of admirable memoirs, Renault 1
has confirmed and extended those observations. That marine
Algae play an important part in certain geological horizons,
notably in connexion with oolitic and dolomite structures, was
laid down by Professor Garwood 2 in 1913.

But even before this date a brilliant investigator, Drew,
who died all too young, had in 1911 studied the calcareous
muds now being deposited in the lagoons on the coast of Florida,
and had shown that in the warm tropical waters of the Gulf
the commonest living form is a denitrifying bacillus which, by
the removal of the nitrogen from the water, leads to a combina-
tion between calcium and dissolved carbon dioxide, with, as
result, the precipitation of the insoluble calcium carbonate.
He isolated and cultivated this form and obtained in vitro the
deposit of calcium carbonate from the sea water. In the follow-
ing year he found the same Bacterium calcis extraordinarily
abundant in the chalky mud of the Great Bahama reef, as many
as 160 million bacilli being present in 1 c.cm. of the surface ooze.3

These observations indicate, therefore, that bacteria, even
to a greater extent than the larger marine Algae, have been
responsible for the deposit of the vast beds of chalk and lime-
stone in which no coralline or other fossils are to be detected.
And now, in the oldest of all stratified rocks, in the pre-Cambrian
or Algonkian deposits of America, Dr. Walcott, the head of the
Smithsonian Institute, has discovered fossilized chains and
clusters of cells which from their appearance and his description
belong to the Cyanophyceae, the blue-green Algae closely related
to the bacteria, and with these other clusters of coccus-like
cells which may, or may not, be bacteria.4

He likewise concludes that "it is quite probable that the
bacteria were the most important factor in the deposition of

1 Renault, B., " Sur quelques micro -organismes des combustibles fossiles,"
Bull, de la Societe de V Industrie minirale, St-Mienne, vols. vii. and xiv., 1899,
1900 (with folio atlas of 20 plates of untouched photographs).

2 Geological Magazine, x., 1913, pp. 440, 490, and 545.

3 Papers from Tortugas Laboratory, Carnegie Institute of Washington, v.,
1914, 44.

4 If bacteria then they are much larger than the micrococci of to-day.

O

18 ADAPTATION AND DISEASE

the Algonkian limestones/' 1 and that thus bacteria are to be
regarded as largely responsible for the oldest of all known sedi-
mentary rocks.

The Geological Parallel

Bacteria have been there, almost ab initio, but this does not,
therefore, mean that all orders of pathogenic bacteria have
always been with us, or that while certain infections are of an
eminently respectable antiquity, this is necessarily true of all
zymotic disease. Recognizing the unity of natural phenomena
we must, I take it, hold that zymotic phenomena run parallel
with geological. I mean this, that certain species, and indeed
certain genera, have existed unchanged through countless ages
to the present day. To-day we find the brachiopod Lingula
living buried in the sand between the tide marks in the Tropics :
we find identical fossil remains in the Cambrian rocks, usually
(though wrongly) described as the earliest geological formations
in which fossil remains have surely been discovered. As I have
just pointed out, algae and bacteria are now known to antedate
them by a long period. Here, then, is an animal which has per-
sisted unchanged, over countless millions of years. The pearly
Nautilus, with its exquisitely chambered shell, is still found in
certain bays in the south-west Pacific : as Sir Archibald Geikie
declared : " This is a genus which has persisted through the
greater part of geological time," for identical chambered shells
are found as fossils reaching back to the Silurian epoch. Limulus,
the King crab, which is no crab but an arthropod, regarded
by Gaskell as in the line of ancestry of the vertebrates and
of man, is closely related to the Devonian Eurypterids, the
fossil remains of that period differing in but secondary characters
from the shells of Limulus of to-day.

And so it is with shrimps like Anaspides 2 and yet other
crustaceans, and with fishes such as Ceratodus, the King fish
of Australia, which have remained apparently unaltered over
extraordinarily long periods of geological time. But while this
is so, we know equally well that during the same period other

1 Walcott, C. D., " Precambrian Algonkian Algal Flora," Smithsonian
Miscellaneous Collections, vol. Jxiv. No. 2, 1914.

2 I owe these latter examples to my colleague at McGill University, Professor
A. Willey, F.R.S.

PLATE I

A

o

0

*

*

c.

)

p

■''#>/-#

c

X

Fossil Blue-green Algae (Walcott).

A. Camasia spongiosa x 350. A blue-green alga from the Algonkian Limestone. Chain of

cells as photographed and partly outlined.

B. The same, higher magnification x 1400. Outline sketch.

C. A group of seven supposedly round cells from the Algonkian Limestone as they occur

free in the matrix x 350. U.S. National Museum, Catalogue No. 60722.

D. The same, touched and outlined.

E. A blue-green alga (Schizothrix) from a recent calcareous deposit at Green Lake,

Onondaga Co., New York, x 350. For comparison.

F. Photograph of same after treatment with hydrochloric acid.

G. Outline drawing of another chain x ca. 1200 from a recent calcareous deposit, after

treatment with hydrochloric acid. For comparison.

THE GEOLOGICAL PARALLEL 19

forms have undergone wide and progressive variations. I
know no more convincing demonstration of the truth of evolu-
tion than the wonderful series of fossils collected by Marsh at
the National History Museum in New York, showing the pro-
gressive stages from the four-toed Eohippus no bigger than a
mastiff, to the one-toed horse of historical time. Of late, I
believe, an equally instructive series of fossils, starting from a
small tapir-like ancestor leading up to the modern elephant,
has been obtained from Egypt, and has been housed at South
Kensington. Here is the point which I would emphasize : the
vast majority of fossils are remains of species and genera which
have passed away.

Now what is true in respect to these fossils would seem to
obtain in respect to zymotic diseases and their causative agents.
Some extend back unaltered in their characters to the very
earliest periods : others appear and cause wide devastation for
a few generations, and then are known no more. It is difficult
to ascribe to faulty observation only the fact that many of the
plagues and epidemics of classical times are unrecognizable to-day :
the simpler and more sensible view to take is that many of them
were diseases which have died out. Nor are we without modern
instances to this effect. Of these the tritest is the " Sweating
sickness " (Sudor anglicus) which, first noted in England in the
autumn of 1485, led to a high mortality, affecting not so much
the poor and ill -fed, as those in comfortable circumstances.
It was an extraordinarily acute visitation, death ensuing often
in the course of a few hours : some are described as passing into
their death-agony while walking in the street, without time
being afforded for them to be shrived. It came in epidemic
form with heavy mortality, and spread rapidly. The attacks,
characterized by high fever, profuse sweating and little pain,
are described as lasting for twenty-four hours. Of this disease
there were outbreaks in 1485, 1507, 1517, 1528, and finally in
1551. Creighton, it is true, suggests that it was introduced
from the Continent by the rabble of mercenaries who came
over with Henry VII., that, in fact, it is the same disease as the
Picardy sweat. Unfortunately for this hypothesis there is no
evidence of the prior existence of such a disease in Picardy or
elsewhere on the Continent, and the Picardy sweat was not heard
of until two hundred and thirty years after the " Sudor anglicus '

20 ADAPTATION AND DISEASE

was first recorded, and when it did appear it was an endemic,
and not a violently epidemic, disorder.

Two diseases in this connexion inevitably come to mind —
diseases so characteristic in their manifestations and natural
histories, that had they presented themselves in Europe in the
Greek and Roman eras, they could not have escaped the notice
of the Aristotles and Pliny s, let alone the purely medical writers.
I refer to Syphilis and Cholera.

Regarding Syphilis the problem is difficult to solve, nor can
it be said that an agreement has been reached as to the date of
its first appearance. This would, however, appear to stand out
clearly — (1) that not a trace of the disease was found in 10,000
skeletons examined bv Elliot Smith 1 from burial-grounds of
ancient Egyptians ; (2) that as pointed out by Dr. Norman Moore
in the course of the discussion at the Royal Society of Medicine
in 1912, Galen with his keen observation and excellent descrip-
tion of the nervous conditions known to him is absolutely silent
as to the eminently characteristic symptoms of locomotor ataxy
and general paralysis of the insane. There could be no more
authoritative evidence that the disease as we know it did not
exist in ancient Rome. Syphilis, therefore, originated or reached
Europe at some later period. The discussion now centres as
to when that was, and there are the two parties, the one favour-
ing the American origin, through the returned sailors of Columbus,
the other denying this, and taking a generally agnostic position.
Opposed to the American theory is that excellent anthropologist
Dr. Hrdlicka of Washington, who in his extensive studies of
hundreds of pre-Columbian human remains has not come across
a trace of syphilis, while post-Columbian bones show that the
disease wrought fearful havoc among the American Indians ;
and now comes SudhofT 2 with strong documentary evidence
from civic ordinances in Germany in 1495 — the year of the fall
of Naples — in favour of the view that the disease was already
well known and causing alarm in Germany at a time when it
was just being recognized in Italy.

It would be out of place here to enter deeply into the value
of these respective contentions and to discuss whether Haiti or

1 Vide Sir H. Morris, " Discussion on Syphilis," Proc. R. Soc. Med. S., 1912.
See also Elliot Smith, Lancet, 1908, ii. 521, and 1909, ii. 1596.

2 Graphische und typographische Erstlinge der syphilitische Litteratur, Munich,
1912. See also his Aus der Fruhgeschichte der Syphilis, Leipzig, A. Barth, 1912.

THE GEOGRAPHICAL PARALLEL 21

Germany is to be regarded as the original home of syphilis.
Whichever view be accepted it is clear that we deal with an
infection which is not coeval with the appearance of man. Either
we deal with a disease which made its first appearance within
historical time, or a disease which appeared in America long
after the New World became separated from the Old — and there,
it may be, appeared in Haiti and the West Indies after these
were separated from the mainland. If the former view be
correct we have a further example of the phenomenon imme-
diately under review — if the latter, another of the many examples
of diseases peculiar to one district, which may or may not eventu-
ally spread to other parts of the world. As Major Strong and
the Harvard Commission have recently shown, there are human
diseases peculiar to South America. There is a " blastomycosis '
peculiar to the San Joaquin Valley in California. We know
that Measles was unknown to the Pacific Islands prior to the
eighteenth century, that cholera, endemic in the delta of the
Ganges, similarly only spread along the trade routes to the
rest of the world early in last century.

The Parallel of Geographical Distribution

This last group of cases — and it might be added to exten-
sively— is exactly parallel to what has been ascertained regarding
the geographical distribution of animals. To afford an easily
grasped illustration I may note what we may term the epizootic
which followed the introduction of the European rabbit into
Australia, or the English sparrow into North America. Now
rabbit and sparrow are relatively old species : the reason why
they had not hitherto been present in areas eminently adapted
for their existence, is that they made their appearance in the
Old World at some period subsequent to the separation of the
New World and Australia from the Old. There are, I know,
other considerations to be taken into account : I might, that
is, have selected more convincing examples from among the
abundant data regarding the geographical distribution of animals,
but these will serve as familiar examples.

The very fact, therefore, that certain highly infectious
diseases may be traced to a previous or existing limited and
localized incidence, is a proof that their origin is to be dated

22 ADAPTATION AND DISEASE

to a period subsequent to the dispersal of peoples over the
habitable globe.

to'

The New Diseases of the Great War

But to return to the main point, that of the evolution of
new infections. No one can fail to be impressed with the fact
that while among the millions of our troops engaged, three of
the terrible plagues of former campaigns have been virtually
suppressed — small-pox, enteric, or typhoid, and typhus — other
and less fatal epidemics have manifested themselves, and some
of these are hitherto undescribed and unrecognized conditions.
Either they are new diseases, or old diseases assuming a new
form under changed conditions of environment. There has
been some doubt as to whether all the cases called Trench fever
are of the same nature, whether under the name we deal with
one or more than one disorder, but this has during the last few
months been dissipated, as the symptomatology of the disease
has become clearly denned ; to the best of my knowledge here
is a disease, or group of diseases, that is new to us. The same
is true of " Trench shin."

The New Diseases op Animals

Nor is our evidence confined to man and human infections.
It is, for example, difficult to believe that a disease leading to
so great a mortality, and with symptoms and lesions so character-
istic as are those of Hog Cholera, could have occurred in previous
generations without being observed and described. But so far
as I can discover this disease was unknown until the 'sixties,
when it made its appearance in the United States, and thence
it has spread over the civilized — and hog-eating — world.

How, then, are we to picture to ourselves the evolution of an
infective disease ? Are there any data which permit us to
deduce reasonably sure conclusions as to the nature of the
process ? I believe that there are, and that the observations
of the last few years, and more particularly certain observations
made here in London, have given the key.

NEW DISEASES : THE NATURE OF INFECTIONS 23

The Nature op the Infections and of Pathogenic Microbes

We must take as our first postulate that every infectious
disease is brought about by the growth within the body, and
diffusion throughout the system, of toxines * of one or other
particular species of micro-organism. This being so, it is deserv-
ing of note that these pathogenic microbes do not form an order
by themselves, but, on the contrary, are singularly diverse in
their affiliations. Still influenced by the work of the last quarter
of the nineteenth century, we are apt to forget this, and classing
the whole subject under the heading of pathogenic bacteriology,
to overlook this notable diversity. How few remember that a
generation before pure cultures of any bacteria were obtained
and tested, Schoenlein demonstrated the relationship and the
constant presence of a mould in Favus. How few again recall
Goodsir's work of the 'forties and his association of a Sarcina
with gastric atony. Flagellate and ciliate protozoa, sporozoa
and yet other genera of protozoa, forms like the spirochaetes,
which certainly are not bacteria, but still are not surely classified,
seeming to be midway between protozoa and protophytes :
filterable viruses which, if protophytic, are evidently of more
than one order and, again, are not bacteria as generally accepted,
even if, according to Hort, some of them represent a phase in
the life - cycle of certain forms included among the bacteria.
With scarce an exception every genus of micro-organism has
its representative or representatives among the pathogenic
microbes, or, put otherwise, every pathogenic microbe has closely
related forms or species differing from it in little beyond the fact
that the one is virulent, the other non-virulent.

On Allied Species of Virulent and Non- virulent Microbes

Next it is to be noted that these allied species are found
suggestively growing in the cavities or on the mucous surfaces
of the body, in the same habitat as the virulent forms, or again
in water and food-stuffs which are taken up by the individual.

1 There is to my knowledge only one clear-cut example of an organism
which, growing upon one of the surfaces of the body without penetration into
the tissues, sets up disease by the diffusion inwards of its toxines — the Bacillus
botulinus, which is responsible for one form of meat-poisoning. Strictly
speaking, this organism induces an intoxication and not an infection.

24 ADAPTATION AND DISEASE

Non-virulent diphtheroid bacilli are so common in the throat
as to be the curse of Army bacteriologists : the same may be
said regarding harmless diplococci which can only be differen-
tiated from meningococci by a relatively elaborate technique :
only within the last few weeks the frequent presence of these
" parameningococci," affording the " group agglutins " as dis-
tinct from the specific agglutins of the meningococci proper,
in one military district was on the verge of causing serious
trouble ; as Andrewes and others have pointed out, the majority
of streptococci obtainable from the throat are non- virulent,
and yet with them, as I shall point out later, there appear to
be an extraordinary number of strains of streptococci obtain-
able from the throat of varying grades of virulence ; non- virulent
spirilla flourish in water just as do the closely-related but virulent
cholera spirilla : from the alimentary canal there may be isolated
non-pathogenic members of the typhoid-coli group, only dis-
tinguishable by physiological or bio-chemical tests from typhoid
bacilli, certainly not by any morphological test : harmless gram-
negative diplococci morphologically indistinguishable from gono-
cocci are obtainable from the vulva : acid-fast bacilli closely
related to the tubercle bacillus of cattle are found in hay, butter,
etc. ; only after some years of study have the non-pathogenic
amoebae of the intestine been clearly differentiated from the
Entamoeba histolytica of dysentery — and the list might be
greatly extended.

This state of affairs in itself leads to the conclusion that patho-
genic microbes at some period, or periods, have originated from
the microbes saprophytic upon the body surfaces, or existing
commonly in the water and food-stuffs ; that they have origin-
ated by adaptation of these forms to growth, not merely on,
but within the tissues.

The Doctrine of Fixity of Bacterial Species

This is the simplest, most rational conclusion, but I must
point out to you that for long years the dominant German
school of bacteriologists upheld vigorously the doctrine of fixity
of bacterial species, closing their eyes to the steadily growing
volume of data which tended to establish the potential varia-
bility of the bacteria. The great leader of German bacterio-

FIXITY OF BACTERIAL SPECIES 25

logists, Koch, in particular supported the doctrine. For twenty-
years or so he laid down that the tubercle bacillus which he had
discovered was identical in its properties from whatever animal
it was isolated, whether from man, the ox, or bird, opposing
Straus and others when they called attention to the marked
distinction in the properties of the human and avian types.
You will remember how suddenly and dramatically he turned
round at the great Tuberculosis Congress in London, announc-
ing that the human and bovine forms of bacillus were absolutely
distinct species — incidentally without any acknowledgment of
the work done by American workers, Theobald Smith, Dinwiddie
and Frothingham, in establishing these differences between the
two types, and incidentally, also, he being a Government servant,
to the apparent material advantage of the agrarian party in
Germany. Many of my readers will recall the consternation
of the hygienists and of the various governmental Agricultural
Departments when he pointed out that, being distinct species,
there was little or no danger of the disease being conveyed from
cattle to man by the milk. It has cost Great Britain alone
thousands of pounds and years of work of a Royal Commission
to demonstrate Koch's error, and show that notwithstanding
recognizable and characteristic differences between the typus
humanus and typus bovinus, nevertheless intermediate forms
exist, and that the typus bovinus can obtain a lodgment and
develop lesions, more particularly in susceptible children during
the milk-drinking period. Koch would never have perpetrated
his successive blunders but for his firm conviction that environ-
ment is incapable of altering the properties of bacteria.

Koch's attitude in this matter has affected me from the
fact that from the time of my thesis for the M.D. at Cambridge
I have taught the variability of the bacteria,1 as again from
the fact that in 1899, many months before Koch's notorious
pronouncement, I had, in an address delivered to the Canadian
Medical Association at Toronto,2 brought together all the data
bearing upon the difference between human and bovine tuber-
culosis and their causative agents, and at a request which reached

1 Vide the article upon the "Variability of the Bacteria" reprinted as
Chapter I. of Part II. of this volume.

2 On the significance of Bovine Tuberculosis and its eradication and pre-
vention in Canada (read before the Canadian Medical Association, Toronto,
August 30, 1899), Philadelphia Medical Journal, December 30, 1899.

26 ADAPTATION AND DISEASE

me through the German Consul in Montreal, had forwarded
copies of my address to Berlin. Listening to Koch's speech, I
could not but feel that he had utilized and perverted the data
I had brought together. Basing myself upon those data, and
upon experiments which showed that inoculations of the human
bacillus would not set up tuberculosis in cattle, I had urged the
Canadian Government not, as Koch suggested, to cease the
endeavour to eradicate tuberculosis among cattle, but on the
contrary to set aside a restricted area, such as Prince Edward's
Island, kill or remove all cattle affording the tuberculin reaction,
and then to proceed to raise pedigree and other herds in these
protected areas, guaranteed free from Tuberculosis, wherewith
to supply the rest of Canada, and if need be the world, with
sound stock, the danger of these cattle becoming infected from
tuberculous human attendants being, as I pointed out, practic-
ally nil. And as the Canadian authorities, pending the findings
of the British Royal' Commission, took no action, I venture to
repeat the recommendation in England as regards the Channel
Islands and the Isle of Man. As it is, I believe that the former
are already relatively free from bovine tuberculosis. It would
not be a difficult matter, therefore, to ensure that the whole
original stock of Jersey and Guernsey cattle become absolutely
free from this devastating disease, and that the value of the
herds become markedly augmented.

Saying this it would seem that I myself support the doctrine
of the fixity of bacterial species, and even of bacterial strains,
that I believe not in the adaptive qualities of the bacteria, but
in their conservatism. The truth is that I believe in both.
As regards these two strains, there are some very interesting
data the relationship of which has still to be worked out. Thus,
as regards other animals such as the rabbit, the typus bovinus
is the more virulent, and while the typus humanus cannot be
conveyed to cattle save by the exhibition of unduly massive
doses in young and little resistant animals, the typus bovinus
can be conveyed to human beings in a more natural course.
But here again it is the young and less resistant individuals
that are susceptible. Nevertheless if, as Krumweide of New
York and others have shown, some 25 per cent of tuberculous
children under five years afford the bovine type, how is it that
this type is almost unknown in adults ? The evidence at our

TYPUS HUMANUS AND TYPUS BOVINUS 27

disposal indicates that quite a large proportion of cases of active
tuberculosis are but the recrudescence of tuberculosis gained
in childhood. May there not be a fallacy, therefore, in the
argument that cases affording bacilli of the human type have been
infected by other human beings % The relative rarity of bacilli
of the bovine type in adults demands either that (a) the bovine
infection acquired in childhood is peculiarly fatal, so that all
affected die in their early years (and this is wholly contrary to
the evidence at our disposal) ; or (6) that on the contrary, in a
large number of cases, it gradually dies out and is replaced at
a later date by infection from another human being with the
typus humanus (and of this again there is no clear evidence) ;
or (c) that gradually, through long residence in a human environ-
ment, the bovine form takes on the character of the human strain.
The fact that we from time to time encounter intermediate
strains seems to favour this third view. But to my knowledge
no decisive observations have so far been made.

CHAPTER III

THE NATURE OF THE ADAPTATIVE PROCESS IN THE BACTERIA

In fact, studying the bacteria, two apparently opposite and
contradictory facts are to be made out : on the one band tbey
are extraordinarily conservative to the extent that, given the
same environment, the same food-stuffs, habit of life, temperature
conditions, etc., they manifest the same characters, and even,
within certain limits when the environment is altered, they
retain their original characters with a certain obstinacy. Typhoid
bacilli from cases of Enteric show scarce recognizable differences,1
whether occurring in England, Australia, North America, or
Japan. And on the other hand, alter their environment beyond
certain limits — their food-stuffs, temperature of growth, etc. —
and their characters are liable to alter. Grow the diphtheria
bacillus, for example, in peptone broth, in milk, or upon peptone
broth agar, and this at the same temperature, and at the end
of twenty-four hours the differences are recognizable under
the microscope. Nothing, in fact, is more easy to demonstrate
than this capacity on the part of bacteria as a class to vary
according to alteration in environment.

Fluctuations, Mutations, and Modifications

Winslow divides these variations exhibited by bacteria into
three orders, (1) fluctuations, (2) mutations, and (3) impressed
variations. Some term these last, modifications. Personally,

1 Several years ago Ehrlich pointed out to me the difference between the
results of his diazo reaction in our Montreal cases of typhoid as compared with
his results in Germany, and suggested that we might be dealing with distinct
strains of the B. typhosus. Sir William Osier tells me that similarly he is
impressed by the difference in the bodily reactions to tertian malaria as acquired
in the Salonica area, and as he was familiar with it in Maryland.

28

FLUCTUATIONS: MUTATIONS: MODIFICATIONS 29

despite its heresy, my opinion is that the majority — if not all —
of the last two are substantially of the same order.

As fluctuations are designated those familiar variations in
one or other quality which distinguish the individuals of the
same sept or clan from one another. The various members of
the families to which each of us belong, while human beings, are
not identical, but vary within certain limits in height, breadth,
weight, length of arm, and so on. The various descendants
of a single individual, though they may approximate in these
particulars to the parents, vary from that parent in a very obvious
manner. Taken together the members of a tall family or strain
may, as a group, average above the normal, or of a stumpy
strain below the normal, but, with this, there is individual varia-
tion in height. And so it is with bacteria. With these fluctua-
tions we need not concern ourselves — they are in the strictest
sense individual : descendants placed in the same environment
need not exhibit them : a man 5 ft. 5 in. in height, for example,
marrying a woman of 5 ft. 1 in. may have children all of them
over 5 ft. 5 in. in height.

By mutation we mean a very different phenomenon. From
causes we cannot as yet wholly fathom, a Shakespeare may be
the offspring of a country butcher : in a field of clover a plant
may show itself with quadripartite in place of tripartite leaves.
The same phenomenon is to be observed among bacteria, but
studying these forms devoid of sexual conjugation, the question
is whether we deal with one or two orders of events. Let me
give an example to the point. It has been pointed out by Barber 1
that while the ordinary culture of Bacillus coli shows separate
short, stumpy bacilli, occasionally while examining smears from
such we encounter filamentous chains of cells. He points out
that all the bacilli have grown floating in the fluid medium :
all appear to have been subjected to the same influences, and
yet here and there a bacillus when it divides has its progeny
remaining united instead of floating away to lead an independent
existence. He found that by isolating a filamentous form and
breeding from it he could obtain a new race constantly showing
a filamentous arrangement, with cultural characters distinguish-

1 Kansas University Science Bulletin, iv., 1907, No. 1. Only recently in the
British Medical Journal of June 16, 191 7, Mr. Ainley Walker has called attention
to the same phenomenon.

30 ADAPTATION AND DISEASE

ing it from the ordinary B. coli and with fermentative powers
more active than normal. If we accept his argument, here is
a typical example of a true mutant, for the development of
which the only satisfactory explanation is, that there has been
some accidental protoplasmic inequality in some one cell division,
and that one of the two resultant cells has by this inequality
gained in excess — or lost — some one protoplasmic constituent ;
thereby a new equilibrium has had to be established between
the organism and its surroundings, this adaptation showing
itself in some alteration of properties.

I fancy that we all look askance at any invocation of
" accident " to explain natural phenomena, regarding this as
a negation of natural law, as something reprehensible, evasive,
and almost dishonest, in that the more we study nature the
more it is impressed upon us that there is a cause for everything.
In this very instance it may be pointed out that while all organ-
isms floating in a fluid medium appear to be subjected to the
same influences, as a matter of fact thev are not. In such
medium organisms at the surface are exposed to an environ-
ment which differs widely from that of organisms totally sub-
merged, and the analogy may be drawn between what occurs
here and the tendency exhibited by many yeasts to exhibit
septal, filamentous forms on surface growths, and isolated torula
forms when submerged.

That in the Bacteria Mutations are Impressed Variations

So far as I can determine the majority of instances of bacterial
mutation depend or follow upon a preliminary alteration of
environment, and these alterations must be regarded as the
controlling factors, resulting in impressed variations. Numerous
cases are on record in which, by growing bacteria in media
containing certain sugars, eventually they have gained the
power to split up and ferment these sugars, or in media, like
milk, containing fats, have developed lipolytic powers ; have,
that is, adapted themselves to feed upon the new food-stuff.
Close upon thirty years ago the late Sir Lauder Brunton and
Macfadyean x reported experiments of this nature, showing that
by growth in the presence of particular sugars certain species

1 Proceedings of the Royal Society, xlvi., 1889, 542.

NATURE OF MUTATIONS 31

of bacteria acquired the power of fermenting them. In 1897
Miss Peckham 1 succeeded in modifying the typhoid bacillus
so that it would produce indol, one of the diagnostic features
of this micro-organism being that it is a non-producer of indol.
In 1906 my late colleague Dr. Klotz,2 now Professor of Pathology
at Pittsburgh, studied the assumption of these sugar-fermenting
powers, and the loss of the same, by a member of the typho-coli
group, the B. perturbans. The following year Twort 3 demon-
strated that he could cultivate paratyphoid bacilli so that they
will ultimately ferment saccharose, typhoid bacilli to ferment
lactose and dulcite, dysentery bacilli to ferment saccharose.
Interestingly enough these observations date back to the founder
of the science of bacteriology, to Pasteur and his classical observa-
tions upon fermentation. You will remember how he discovered
that ordinary tartaric acid is optically inactive because it is
a combination of equal amounts of the two, dextro- and laevo-
rotatory, isomers, and that he obtained the laevo-rotatory form
in a pure state by the action of moulds, which used up the dextro-
rotatory acid leaving the other isomer untouched. But doing
this, he noted that if he continued the experiment the rotation
to the left reached a certain point and then steadily diminished ;
in other words, that after using up the food of election the mould
now turned itself to assimilate and use up the laevo-rotatory
food. Now, as a general rule, dextro-rotatory organic substances
alone are produced and utilized by living animals.

The steps by which bacteria have been found to gain these
powers of adapting themselves to and utilizing unaccustomed
food-stuffs are most instructive, and are best shown by the
observations of Massini,4 Penfold,5 and other observers of the
last few years. These observers employed a solid medium, a
peptone broth agar medium (containing, as usually made, a little
muscle sugar), upon which the bacteria studied can be grown
readily, and to this they added in the process of preparation the
particular foreign sugar or glucoside, and made surface growths of
the bacteria. Where this is done it is found that the surface

1 Journal of Experimental Medicine, ii., 1897, 549.

2 Journal of Infectious Diseases, ii. (Supplement), 1906, 35.

3 Proc. Royal Society, Biol, lxxix., 1907, 329.

4 Arch. f. Hygiene, ixi., 1907, 250.

6 Journ. of Hygiene, xii., 1912, 195. See also Miiller, R., Centrlb. f. Bakt.
xlii.3 1908, 157, and ibid. Abt. 1 Orig. lviii., 1911, 97.

32 ADAPTATION AND DISEASE

colonies at first show no change from the usual type of growth,
but after several days little papillae appear on the surface of
the colonies. These have a more opaque white appearance,
and, if litmus has been added to the medium as an indicator,
the papillae take it up and assume a reddish colour, this indicat-
ing the production of acid in the papillae, whereas it is not being
produced elsewhere in the colonies. The Typhoid bacillus, for
instance, does not normally ferment isodulcite, but if the Typhoid
bacillus be planted upon isodulcite agar, in five days these
papillae appear — and in these papillae the bacilli are " mutants " ;
they have gained the power of fermenting this sugar. Test the
bacilli from the non-papillated portions of the colonies, and they
have gained no such power. If bacilli be carefully removed from
one of the papillae and be plated on isodulcite agar, it is found
that a race of Typhoid bacilli has been gained, all the members
of which have this new power of immediately fermenting iso-
dulcite. The bacilli constituting the non-papillated areas are
unchanged ; they cause no dissociation of the isodulcite, although
if plated similarly on isodulcite agar, their colonies will repeat
the phenomenon : in five days scattered papillae will show them-
selves which present the new property.

These, again, at first appear to be excellent examples of
what may be termed true or " chance " mutations. Obviously
not all the bacilli in the colonies take on the power of utilizing
isodulcite, dulcite, or other sugar, as a food-stuff, only certain
individuals. But on further study the further facts are to be
noted — first, that the phenomenon is constant — take any typhoid
bacillus, treat in this way, and in five days or so the dulcite-
fermenting papillae will show themselves ; and, secondly, these
dulcite -fermenting individuals present themselves constantly
in a particular relationship and superficial position. Bacilli
which are at or near the surface of the agar and obtain from it
abundant food have had no need to adapt themselves to the
new food-stuff. It is those furthest away, separated from the
agar by several layers of bacilli, that manifest this new property,
those which from their position receive only the diffused food-
stuffs which have passed by their more fortunately placed
comrades lower down. It looks very much as though, in the
struggle for existence, these bacilli that have found themselves
deprived of their accustomed food-stuffs have taken to absorbing

NATUKE OF MUTATIONS 33

and dissociating the unusual food-stuff faute de mieux, that, in
short, we have here a repetition of Pasteur's original observa-
tion that when the accustomed food-stuff is used up, the un-
accustomed will be attacked, just as we to-day are making a
virtue of necessity and are thriving upon potato — and wheat —
substitutes. As a matter of fact, if the experiment be modified
so that all the multiplying individuals are subjected to the same
environment — if, that is, instead of employing a solid medium
a fluid medium containing the new glucoside be taken and the
bacilli be grown on this, then the conditions of experiment can
be so arranged that not some but all the bacilli acquire the
property of fermenting the unusual glucoside. All, that is,
acquire the new property.

On Impressed as distinct from Chance Variations

If I remember aright, this was first demonstrated in 1913 or
1914. Unfortunately I am away from my library and notes,
and have been unable to find the reference. In my dilemma —
for I did not wish to make the vague statement that this was
so without affording a reference — my colleague Major F. B.
Bowman, Officer in Command of No. 2 Canadian Mobile Labora-
tory x at Folkestone, with great kindness came to my rescue and
willingly conducted the very simple experiment — an experiment
so simple that it can easily be repeated. The underlying principle
of it is that the bacteria must not be supplied with an abundance
of alternative food-stuff. They are accustomed to obtain their
carbon from amines (proteins) when carbohydrates are not present.
Do not, therefore, use a peptone broth of the usual concentration.
Some years ago I pointed out that with members of the typho-
coli group 10 ccm. of peptone broth added to a Winchester quart
of sterilized water affords upon inoculation a pronounced tur-
bidity in twenty-four hours, in other words, an active growth
of the bacilli.

Major Bowman, therefore, made up such a diluted peptone
bouillon — or " Soupe maigre " — with 1 per cent added of the
unusual glucoside in this case Isodulcite, or Rhamnose,2 along

1 Now No. 1 Canadian Central Laboratory.

2 For which I am indebted to Professor Rosenheim of King's College and
(through Captain Tulloch) to Professor Mackenzie of Aberdeen.

D

34 ADAPTATION AND DISEASE

with Azolitmin as an indicator. He inoculated a relatively
large quantity of this medium with a pure culture of B. typhosus
obtained from the Army Medical College at Millbank, having
previously tested the strain and having found that it was
absolutely typical in its fermentation reactions (producing acid,
but no gas, with Glucose and Mannite, while Lactose, Saccharose,
Dulcite and Isodulcite gave negative results).

Each day he made plates from the main culture upon
azohtmin-isodulcite-peptone-agar, with the following results :
The main culture in 100 ccm. of 1 per cent azolitmin dilute
broth showed no change until the fifth day, when the blue solu-
tion took on a slight change, showing that some acid was being
produced. At the same time the agar transplants of the fourth
day exhibited a pinkish tint of the medium showing beginning
fermentation, while the agar transplants of the fifth day at the
end of twenty-four hours showed a distinct pink halo around
every colony — as did also the plate of the sixth day. (I give
Major Bowman's note as Appendix II., explaining that up to the
last moment I hunted for the lost reference, and thus only gave
him time to perform the experimentum crucis. Nor indeed,
knowing how fully occupied he was with routine military work,
could I ask for or could he offer more.)

And here is the point : add typhoid bacilli to distilled water
containing 1 per cent isodulcite, and they are not killed off,
that is to say, they undergo no greater reduction in number than
if added to pure distilled water. If we dealt with chance varia-
tion and the survival of the fittest, some of the colonies on the
agar at the end of four or five days would be unchanged, others
would show surrounding reddening of the azolitmin medium.
But all show the reddening.

Here then, contrary to Bateson, we have evidence of positive
acquirement from without, and, contrary to the Lankesterian
dogma, we can so arrange our experiment as to obtain, not
evidence of variation in many directions, the favourable variants
alone surviving, but evidence that organisms placed in a given
environment all vary in one identical direction with clockwork
regularity. It is no matter of haphazard : we can of a certainty
induce the particular variation in a particular bacterial species.

Or if Professor Bateson desires a perfect example of varia-
tion by loss of factors, we can afford him an equally telling

EXALTATION OF VIRULENCE 35

example which, by the constancy of the results obtained, is
infinitely superior to his sportive production of a salmon-coloured
" Coral king " Primula. It is the classical example adduced long
years ago by Pasteur and his lieutenants, Roux and Chamber-
land — a bacteriological commonplace, since it is the basis of
Pasteur's method of vaccination against anthrax, which, com-
mercially speaking, has saved France hundreds of thousands
of sheep and cattle and many millions of francs. By exposing
cultures of anthrax bacilli to a heat a degree or two below that
which will kill them, or again by the action of minute quanti-
ties of certain antiseptics added to the medium of growth, races
of anthrax bacilli are obtainable which have wholly lost the
power of spore production. These races may be grown for
months and years — for thousands and, indeed, hundreds of
thousands of generations — upon the ordinary media of the
laboratory without regaining this striking and characteristic
property. Were a culture of this strain given to a systematic
botanist without information as to its origin, he most assuredly
could classify it as a distinct species. Here again is no matter
of chance variation in many directions — all the individuals of
the colony subjected to the particular temperature become
asporogenous.

On Exaltation of Virulence

Now if these things be true regarding other properties, they
must be largely true regarding the virulence and pathogenic
powers of bacteria. We know in the first place that virulence,
when it is the property of any bacterial species, is capable of
great modification. This, we observe, is not a matter of chance.
We know methods by which we can surely exalt or depress
that virulence. We know that by " passage " of a virus through
a succession of animals of one species we can rapidly intensify
the virulence for animals of that species, while simultaneously
we may by this procedure reduce its virulence for animals of
another species. We can, with Marmorek, take a culture of
streptococci so weak that only the most susceptible animals are
influenced by it, and then only by the exhibition of relatively
enormous amounts of the culture to the new-born (which have
least powers of resistance) ; but recovering the cocci from these

36 ADAPTATION AND DISEASE

we can by careful passage through selected animals so augment
the virulence that eventually the hundredth or the thousandth
of a drop of a twelve-hour culture, or even much less than this,
may cause the death of strong adults in six hours or less.

It is quite possible that these streptococci may eventually
afford us the best material for demonstrating the development
of pathogenic from harmless bacteria. They are singularly
widespread in nature : they are common on the skin, where
the temperature is lower than it is in the passages of the body ;
they are constantly present in the mouth, and, as pointed out
by Andrewes, the majority of cultures obtained from the mouths
of healthy individuals are harmless when inoculated into lower
animals. Certain it is, also, that after long years of work and
the investigations of numerous trained observers, we seem to
be almost as far off as ever from establishing the dividing lines
between different species — if species proper they are — of strepto-
cocci. Of late, it is true, we are accustoming ourselves to recog-
nize and attribute particular properties to Streptococcus faecalis,
Streptococcus equinus, haemolytic streptococci, etc. ; we have
no doubt regarding the existence, and, what is more, the utility
of recognizing these " strains." But if we depend on fermenta-
tion tests (and it is only by biochemical and not by morpho-
logical properties that we can distinguish between the forms),
then, as shown by Ainley Walker, continued growth in the
laboratory results in changes of the fermentation tests, so that
there is as much evidence that we deal with the different strains
of one species which exhibit differences in consequence of their
previous habitat, environment, and history, as there is that
they constitute distinct species. For a time it looked as though
Rosenow, of Chicago, had definitely settled this point, and by
passage through animals had converted non-haemolytic into
haemolytic strains, and even streptococci into pneumococci
(which have many properties in common). But Holman, of
Pittsburgh, has brought forward such forcible criticism of Rose-
now's results, calling attention to the frequency of combined
growths of more than one strain or species of streptococci, and
the need for pure cultures gained from isolated cocci, that before
Rosenow's work can be accepted in its entirety, this laborious
confirmatory work must be undertaken. Saying this, I firmly
believe that Rosenow is along the right fines — and pragmatically

EXPEKIMENTAL PRODUCTION OF PATHOGENICITY 37

speaking, his observations and hypotheses gain confirmation
from the clinical observations of Billings 1 and the Chicago
school. I like to flatter myself with the thought that an address
which I delivered in Chicago in 1899 2 was the starting-point
for these activities.

The Experimental Conversion of non-Pathogenic into

Pathogenic Microbes

Years ago Vincent, one of the most distinguished of French
bacteriologists, to whom we owe the recognition of the relation-
ship between the Bacillus fusiformis and Vincent's angina, or
better, Vincent's disease,3 described how by placing broth cultures
of a perfectly harmless, non-pathogenic microbe from the soil,
the Bacillus megatherium, in hermetically sealed celloidin
capsules, and inserting the capsules within the tissues of lower
animals, he was able to accustom, or adapt, them to growth
within the organism of warm-blooded animals.4 By simple
subcutaneous or intramuscular injection of a few ccm. of a
broth culture of this bacillus into the rabbit, for example, no
harmful results ensue : the bacilli are rapidly destroyed. Placed
in delicate celloidin capsules the bacilli gain nourishment from
the diffused lymph, and are at the same time protected from
the phagocytic action of the leucocytes. By this means he an-
nounced that the bacilli gradually gained pathogenic properties.
That was close upon twenty years ago, and unfortunately I
cannot hear that this experiment of Vincent's has been repeated
and confirmed. When some years ago, in my laboratory at McGill,
Dr. Charles Higgins employed this method with tubercle bacilli
of the typus humanus, inserting celloidin capsules containing a
glycerine broth culture of these into the tissues of the calf, even
after some months he obtained no clear evidence of modification.
The capsules became surrounded by dense fibrous tissue, and
where the cultures did not die out they showed no assumption

1 Local Infection, Appleton, 1916.

2 " Latent Infection and Subinfection," American Medical Assoc. Journal,
Dec. 1899.

3 Better, because, as Vincent himself pointed out, the organism may be
associated not merely with a tonsillitis (angina), but also with a special form of
stomatitis and gingivitis, while also the conjunctiva may be affected, and the
organism may grow upon the glans penis, setting up a balanitis.

4 Ann. de Vlnst. Pasteur, xii., 1898, 785.

38 ADAPTATION AND DISEASE

of the characters of the typus bovinus. Vincent is so careful
an observer that I do not like to reject his work. At the same
time I confess I should feel more satisfied had it been confirmed
by other workers.1

It is here in London, at University College, that what appears
to be the crucial demonstration has been afforded by a pair of
well-known and capable workers, and that by the employment
of a most ingenious method. Every medical man nowadays
is familiar with the name at least of anaphylaxis, even if he is
unable to grasp the nature of the paradoxical phenomenon —
the phenomenon, that is, of inducing not immunity, but the
reverse condition of increased susceptibility to foreign proteins,
by the injection of the same into the tissues. If minute doses
of any protein foreign to the organism — even if it be so harm-
less a one as egg white — be introduced parenterally, i.e. into
the tissues as contrasted with alimentary canal, and some eight
days or more later a second and larger dose be similarly intro-
duced, instead of tolerance having been established, the very
reverse condition is found to have been set up. Grave con-
stitutional symptoms supervene, and in animals like the guinea-
pig death may ensue with startling rapidity in the course of a
minute or two. I believe that I was the first to point out,2
and that with Professor Vaughan's entire concurrence, that
these phenomena are explained by that observer's most remark-
able studies upon the dissociation products of proteins. As he
has abundantly demonstrated,3 treated with alcohol rendered
alkaline by the addition of caustic soda, all true proteins, includ-
ing the bodies of bacteria (or their protein constituents), become
dissociated into an intensely toxic and a non-toxic moiety.
The explanation, therefore, of anaphylaxis — and this is confirmed
by Abderhalden's later work — is that it is the first stage in the
production of immunity : the organism has reached the stage
of producing enzymes which dissociate the proteins into a toxic
and a non-toxic moiety, but has not attained to the later stage
in which specific enzymes are developed, dissociating the toxic

1 It deserves note that the bacteriology of soil bacteria has been worked out
very imperfectly, and it is evident that more than one species has been described
as B. megatherium. It may therefore be that the B. megatherium of other
workers is not the same organism as that employed by Vincent.

2 Principles of Pathology, 1st ed. i., 1908, p. 509.

3 See Vaughan, Protein Split Products. Philadelphia, Lea and Febiger, 1913.

THIELE AND EMBLETON'S EXPERIMENT 39

moiety into harmless derivatives. And these anaphylactic
phenomena are obtainable, let me repeat, by the introduction
of bacterial proteins into the tissues, and even of living
bacteria.

So long ago as 1904, W. V. Shaw,1 and later Schloss and
Foster 2 in 1913, called attention to the fact that in monkeys
arthritis only occurs after a second intravenous injection of
streptococci, the first setting up no obvious changes. In 1911
my late colleague in Montreal, Dr. Duval,3 now Professor of
Pathology at Tulane University, New Orleans, when studying
the bacteriology of leprosy made the curious observation that
while by a single inoculation of the bacilli which he had isolated
and cultivated he was unable to produce lesions in monkeys
(Macacus rhesus), lepra nodules containing the bacilli could be
obtained as the result of inoculations repeated at short intervals.
He and Couret conclude (p. 303) : " That the infection is more
likely to follow where sensitization is first established is definitely
proved by the specific experiments that we have carried out upon
a variety of laboratory animals. The first injection, we assume,
sensitizes the animal, and may consist of either killed or viable
lepra bacilli."

Now, starting from these phenomena of anaphylaxis, Thiele
and Embleton,4 working at University College, were led to the
happy idea of introducing harmless bacteria into the tissues,
and then in a week or ten days later giving a second injection
of the same bacteria. Those first injected being non-pathogenic
became disintegrated, and liberating their proteins were found
so to influence their host that upon a second injection into the
now susceptible animal, the bacteria, instead of being destroyed,
took on active growth, and what is more, the anaphylactic
state, by permitting the active growth within the weakened
organism, was found by them to lead to an associated adapta-
tion to the unaccustomed environment.

The Bacillus mycoides is an absolutely harmless micro-
organism common in garden soil, and growing in nature only
at a low temperature. It is in fact destroyed at body heat.

1 Journ. Path, and Bad. ix., 1904, 158.

2 Journ. Med. Research, xxix., 1913-14.

3 Journ. of Exper. Med. xiii., 1911, 374, and Duval and Couret, ibid, xv.,
1912, 292.

4 Zeitschr. f. Immunitatsforschung u. exper. Therapie, xix., 1913, 643.

40 ADAPTATION AND DISEASE

Following the classical experiment on adaptation of Dallinger1
in the 'seventies, they first gradually accustomed the bacilli
to grow at body temperature by cautiously increasing the tem-
perature of growth, degree by degree, over a considerable period,
until they would multiply instead of being killed at 37*5° C. Next
they sensitized animals by injecting into them a fair dose of
the killed bacilli. Ten or fifteen days later they gave a second
injection of the five bacilli, recovered the bacilli from the tissues
of the animals, and repeated the process with another animal
of the same species.

Doing this they found that the bacillus which had been
motile lost its flagella, acquired a capsule, became stumpier and
thicker, and, to their astonishment, acquired all the characters
of the anthrax bacillus, both morphologically and in the
lesions it produced. The harmless, naturally non - pathogenic
bacillus became in fact so virulent that a small amount of
the culture introduced into guinea-pigs killed them in twelve
hours.

Nor is this the only non-pathogenic organism that these
observers have induced to become pathogenic by this method
of preliminary anaphylaxis. They have obtained parallel results
with other non-pathogenic microbes (B. phlei, B. smegmatis, etc.).

Their work has been confirmed independently by Kniest
Faber,2 who noted that employing virulent streptococci he
obtained arthritis in rabbits upon a first injection, but with a
weakly pathogenic form such as Streptococcus mitis var. viri-
dans he found that in order to set up the lesions of arthritis
he had to give rabbits two sensitizing doses ; that these doses
had to be narrowly specific ; he had, that is, to employ the
particular strain of streptococcus. Preliminary intravenous in-
jections of the S. mitis set up no arthritis, but if he injected it
direct into the knee-joint, so setting up a localized arthritis
of the one joint, allowed this to subside, and now gave an
intravenous injection of the same organism, there followed
immediately an acute arthritis of the treated joint. This form

1 Proc. Roy. Soc. xxvii., 1878, 332. Dallinger, it may be recalled, gradually,
over long months, raised the temperature at which the flagellate infusorian,
Dallingeria drysdalei (Kent), would grow and multiply, from 60° F. to 158° F.
( = 70° C.) — i.e. some 20° C. higher than that causing the death within a few
minutes of ordinary (untreated) vegetative infusoria.

2 Journ. of Exper. Med. xxii., 1915, 615.

THE EVOLUTION OF INFECTIONS 41

of sensitization throws an interesting light npon the relapses
so frequently noted in streptococcal infections.

Observe what this signifies. It signifies that there is a means
whereby organisms that are saprophytic and non-pathogenic,
living upon the surface of the body, whether on the skin or
upon the moist mucous membranes, can become pathogenic,
invading and living within the tissues. What is necessary is
that these organisms first become habituated to growth at the
body temperature and then gain entrance through some solution
of continuity, and that, gaining entrance, they in the first place
induce the anaphylactic stage, so that when some week or more
later organisms of the same species gain entrance into the
supersusceptible tissue, whether through a lesion or through
conveyance by phagocytic leucocytes, they now are able to
multiply within the tissues, and multiplying acquire pathogenic
properties ; in other words, adapt themselves to their environ-
ment, gain the power of assimilating and dissociating the proteins
of the body, in which process they bring about the production
of disintegration substances, of toxic proteins or amino-acids,
in other words of toxin es.

But, it may be asked, if the process is so simple as this why
do we not observe time and again the evolution of new diseases ?
In answer to this question I would urge that the matter is not
so simple as it seems. In the first place it will be observed that
a particular concatenation of circumstances is requisite, namely,
a first entrance of bacteria into the tissues through some lesion,
and then, when the anaphylactic stage is at its height, a second
entrance of the bacteria in much larger numbers, for anaphy-
lactic phenomena only show themselves when the second exhibi-
tion of the protein (in this case of the bacteria) is larger than the
first, and when a definite interval of at least seven days intervenes
before that second exhibition. As is well shown in Pasteur's
method of inoculation against rabies, when a virus is inoculated at
successive daily intervals no anaphylactic phenomena are mani-
fested. And in the second place it must be remembered that
ordinarily the habitual growth of bacteria upon the surface of the
body favours the production, not of anaphylaxis, but of immunity.
It is by now well established that from the mucous surfaces of
the body bacteria are being constantly introduced into the
tissues, in driblets as it were, through the agency of leucocytes

42 ADAPTATION AND DISEASE

which wander out on to the mucous surfaces, take up bacteria
of various orders, and wander back again into the tissues — a
process which leads to the gradual habituation of the organism
to the dissociation products of the bacteria thus incepted, or
in other words, to a progressively increased capacity to dis-
sociate and digest the bacteria. Now, where immunity is already
established, anaphylaxis (the preliminary stage, as I have pointed
out, in the process) cannot be redeveloped.

Accepting these very striking studies of Thiele and Embleton
— and as I have pointed out they gain confirmation from the
independent observations of several observers — we see how from
time to time infectious or epidemic diseases may arise de novo
through the evolution of more specialized pathogenic forms
from the more widely diffused saprophytic micro-organisms
growing upon the surfaces of the body. We can understand,
for example, how it was that Diphtheria came into prominence
only at the beginning of last century, when Bretonneau of Tours,
Napoleon's great army doctor, afforded his classical description.
It is impossible to believe that a disease with such obvious
clinical manifestations and such sharply-cut features should have
been present infecting man from the beginning of history with-
out a single physician in all the centuries recognizing and describ-
ing the throat condition, or the rapid devastation set up by the
disease. Diphtheria gains its simplest explanation as being due
to the acquirement within recent times of virulent properties by
some previously harmless diphtheroid bacillus growing in the
throat and upper respiratory passages. And so it has been with
Syphilis, with " Trench shin," Trench fever, Sweating sickness,
and, if with these, then with infections in general.

I admit that there may be contributory causes, that particular
environments may favour the origin or reappearance of particular
diseases, as again that local tissue environment may determine
particular selective development of virulence. As regards the
former we have the remarkable history of Epidemic War
Nephritis, so well brought out at a recent meeting of the Medical
and Therapeutic section of the Royal Society of Medicine, with
its first known appearance in North America as an epidemic
beginning in the spring when, during the War of the Rebellion,
the campaign in the central region became a matter largely of
trench warfare ; the absolute disappearance of this particular

THE EVOLUTION OF INFECTIONS 43

type of acute nephritis for over fifty years, to make its reappear-
ance in 1915 in the trench warfare along the western front.
This is a very remarkable history, but until the causative agent
is determined we can do no more than suggest that here is a
disease which shows itself when man finds himself in a particular
environment, dying out when that environment is departed
from. As Professor Welch of Johns Hopkins pointed out some
years ago in his Huxley lecture,1 there is evidence that if man
can adapt himself to altered environment, so can microbes ;
that if man can protect himself against the toxic action of the
bacteria with which he is surrounded, so on their part bacteria
can protect themselves against man by increasing their virulence
and altering their properties according to circumstances.

It is in this connexion that the recent series of studies by
Professor Rosenow upon the elective localization of streptococci
are of extraordinary interest.2 Briefly, by employing deep tubes
of fluid culture medium, which permit streptococci to grow at the
oxygen tension most favourable to the particular strain, and
by inoculating young growths which still retain the final quali-
ties impressed upon them by development in particular tissues
and localities, he has been able to show that strains obtained
from the gall-bladder in cases of cholecystitis, when inoculated
intravenously into animals of the laboratory, are peculiarly apt
to localize themselves in the walls of the gall-bladder and there
set up cholecystitis ; isolated from the joints in man, to set
up arthritis, from duodenal ulcers to set up gastric and duodenal
ulcers, from multiple neuritis in man to be recovered from the
nerve-trunks in the inoculated animal, in myalgia from the
muscles. Growth, that is, in a particular tissue in man, endows
the streptococcus with an elective affinity for the same tissue
in other animals. "With abundant material to work upon, first
at the Memorial Institute, Chicago, and later at the Mayo Founda-
tion, Rochester, Illinois, injecting various strains of streptococci
into animals of the laboratory, and making routine cultures from
various tissues, he found that the average incidence of the cocci
in or obtained from muscle was 27 ; with cultures from cases of

1 British Med. Journ., 1902, ii. 1105. In the same year Ainley Walker had
made the same suggestion (Journ. of Pathology, viii., 1902, 34).

2 See more particularly Journal of the American Medical Association, Ixv.,
1915, p. 1687, and lxvii., 1916, 662; Journal of Infectious Diseases, xix., 1916,
Nos. 3 and 4 (September and October); and Journal of Immunology, i., 1916, 363.

44 ADAPTATION AND DISEASE

myalgia this rose to 93 per cent. Cultures of cocci are rarely
gained from the spinal cord of the inoculated animals (4 per
cent) ; thirty-six animals inoculated with cultures from a case
of sporadic anterior poliomyelitis yielded 78 per cent of positive
results, i.e. of growths of streptococci obtainable from the spinal
cord.

With these examples before us is it possible for medical men
not to believe in " direct adaptation " ? According to the
environment so do bacteria assume special properties.

CHAPTER IV

ADAPTATION TO DISEASE-PRODUCING AGENCIES IN THE

HIGHER ANIMALS

Before passing to the other end of the scale of living forms,
and studying the phenomena of infectious disease in the terms
of adaptation, and other adaptations to disease-producing
agencies as observed in man and the higher animals, let us pause
for a moment and endeavour to summarize the conclusions
which may reasonably be drawn from the data brought forward
up to this point.

1. The evidence is abundant and conclusive that bacteria

are capable of being modified by alterations of environ-
ment of certain orders, chemical and physical.

2. The modifications in question that have been brought

forward conform with Herbert Spencer's " direct equili-
brations " : a particular alteration of environment of
certain orders leading inevitably to a particular modifica-
tion whether of function or of form.

3. By employing appropriate methods it can be shown that

not some but all the microbes subjected to particular
orders of alteration of environment exhibit the particular
modification : the hypothesis of " chance variation " in
a particular direction with survival of the fittest is in-
capable of explaining the phenomena.
Yet other conclusions are to be drawn, but for the time being
these suffice for my immediate purpose.

There are two interesting and, I think, basal observations
upon adaptation which, affecting protozoal forms higher in the
scale than the bacteria, deserve note before passing to consider
adaptation in multicellular organisms. The one we owe to the

45

46 ADAPTATION AND DISEASE

botanist Stahl,1 who a third of a century ago made the pioneer
observations upon what we now term chemiotaxis. He noted
that the plasmodium of the myxomycete Fuligo, which at first
moves away from a 2 per cent solution of common salt, will
after a time (more especially if it has suffered from lack of water)
adapt itself to the solution, advancing its pseudopodia into it.
In other words, we have here a definite example of direct adapta-
tion during the lifetime of the individual, of change of habit in
these lower forms of life based upon physical changes in the
surface layers of the individual. The other consists of certain
observations of Musgrave and Clegg 2 upon intestinal amoebae.
They noted that these in the large intestine ingest selectively,
feeding upon only one of the many surrounding species of bacteria,
and that they could succeed in growing the amoebae on the
surface of agar plates if they effected a combined growth of the
amoebae and this one species of bacillus. By the gradual
addition of pure cultures of another bacterial species, with which
at first growth may not be possible — indeed, with this form alone
afforded in the first place the amoebae died out — the amoebae
at first became accustomed to the foreign forms, taking up
occasional individuals, until eventually they could be grown
in association with the second form alone, to the total exclusion
of the first. Here again we observe accustomance and progress-
ive utilization, and in no sense a chance variation. The same
observers note that inoculating amoebae plus bacilli into the
livers of rabbits they induced abscesses which yielded both
amoebae and bacilli, but inoculating the pus from the first
abscess into a second rabbit they now obtained abscesses con-
taining amoebae alone : in other words, the amoebae had now
adapted themselves to growth in and upon the tissues of the
rabbit — an observation which throws fight upon the frequency
with which entamoebae are obtained in a pure state from hepatic
abscesses in man.

It will be seen that so far I have laid comparatively very
little stress upon the inheritance of these acquired characters
and the development of races or strains endowed with special
properties, the result of acquirement. I have done this pur-
posely, being convinced that the proper method of attack upon

1 Bolanische Zeitung, 1884, Nos 10 and 12. and Flora, lxxvi, 1892. 247.

2 Reports of the Bureau of Government Laboratories, Manila, 1904, No. 18.

IMMUNITY AN ADAPTATION 47

the problems before us is to study first the phenomena of in-
dividual acquirement of properties. Only when we have gained
some knowledge of the factors operative in cases of individual
acquirement can we with any security proceed to discuss the
inheritance of acquired characters.

Now it is in respect to these new acquirements in the higher
animals that we obtain the deepest insight into the process
involved, and that through the abundant, not to say overwhelm-
ing, studies of the last quarter of a century upon immunity.
There is a vast literature upon this subject. Some of the most
notable achievements in science, both of the last and of the
present century, have been in this province of immunity : the
names of the great workers in this domain are known to all
men — of Pasteur, Koch, Behring, Roux, Metchnikoff, Ehrlich,
Almroth Wright. Tuberculin, diphtheria antitoxin, and the
arrest of typhoid fever by inoculation are matters about which
every intelligent being is supposed to know something ; they
are even common subjects of conversation at the dinner-table,
be it in London or Glasgow, in Washington or Seattle, in Johannes-
burg, Dawson City or Adelaide. Yet I do not know of a single
general biologist who has dwelt adequately upon their significance
in this connexion.

For consider the following facts and think what they signify.
There is a certain bean, known as the Jequirity or prayer bean,
the seed of Ahrus precatorius, from which a highly active and
highly poisonous principle can be extracted, abrin, which has
been resolved into two proteins, an albumose and a globulin,
both, when injected into animals, producing symptoms of like
nature. A closely allied active principle, ricin, is obtainable
from the castor oil plant, and this has been carefully studied by
Professor Cushny,1 who has determined that in this case the
toxic substance is a globulin, and this so powerful that one gram
is adequate to kill one million and a half guinea-pigs, with necrosis
of the tissues at the site of injection, acute congestion, inflamma-
tion and oedema of various tissues. It is significant that the
action is not immediate : the animals appear to be in good
health for four or five days ; then symptoms supervene suddenly
and there is death in a very short period. There is in short a

1 Arch. f. exp. Pathol. x\i., 1898, 439 ; see also in conjunction Osborne and
Mendel, Am. Journ. of Physiol, x., 1904, 36.

48 ADAPTATION AND DISEASE

curious parallelism with the period of incubation followed by
acute disease as seen in bacterial injections.

We owe to Ehrlich a series of observations which further
enforce the parallelism. Feed mice or other small animals
upon small non-lethal doses of ricin, and slowly increase the dose,
and eventually the animals can be given with impunity one
hundred times the fatal dose. Or again, with great care, inoculate
the animals subcutaneously with minute but progressively
increasing doses, taking care to allow any signs of disturbance
to pass away before a new dose is given, and by this means
the little animals can be made to stand five thousand times the
lethal dose.

Now the castor oil plant and its poison are not objects which
the ordinary European rabbit or mouse is accustomed to
encounter. If we take the blood or other body fluids of control,
normal, animals we find that they are absolutely powerless to
neutralize the poisonous principle. Take on the other hand the
blood serum of the treated and immunized animal and mix that
blood serum with the lethal dose of ricin, or ten, or a hundred,
or a thousand times the lethal dose (according to the grade of
immunity induced), and the poison is rendered inert. That is
to say, inject the mixture into another animal of the same species
and no effects are induced. Clearly in the immunized animal the
blood now contains antitoxic substances, bodies which combine
with the toxin or break it down, rendering it inert and harmless.

And these substances have not been produced in the blood,
or from the toxin, by some form of chemical activity undergone
in the blood itself. Once an animal has been immunized it
may be bled and re-bled until many times the normal amount
of circulating blood has been removed from it — and notwith-
standing the blood continues to be antitoxic. Obviously, in
the first place, the antitoxin is produced by the cells of the body,
the tissues, or certain of them. Obviously the amount of anti-
toxin produced is not in direct proportion to the amount of
toxin injected ; in other words, once the cells, stimulated by
minute doses of ricin, have acquired the property of elaborating
an antiricin, they continue to produce it for weeks and months.
The cells have acquired a new power, that of elaborating anti-
ricin, and they continue to elaborate and discharge this body
long after the original stimulus has disappeared.

IMMUNITY AN ADAPTATION 49

Here then is a perfectly clear cut example of direct individual
adaptation in the higher animals fulfilling wholly the essential
conditions that —

1. We deal with the acquirement of a new property ; the

acquirement cannot possibly be regarded as the calling
into activity of a property previously possessed, either
by the individual or by the species : this power to
neutralize ricin is something absolutely new.

2. The acquirement is something positive, something addi-

tional ; there can here be no alternative hypothesis of
loss of inhibiting factors.

3. There can here be no possibility of ascribing the new

property to the persistence of a chance variation : the
power to develop antiricin and discharge it into the
blood can be produced in any mouse or rabbit with
absolute certainty.

4. There is here no alternative explanation of the survival

of the fittest.
I give this as a first example because in it we have no addi-
tional factor of prolif erative activities on the part of the microbes
which in the case of the development of immunity against
bacteria and their products may, to those not familiar with
bacteriology, be apt to confuse the picture. But what is true
with regard to these phytotoxins may be exactly paralleled in
the reaction that can be induced in the higher animals against
bacterial poisons — bacteriotoxins. We do not in nature find
guinea-pigs, for example, subject to diphtheria : by cutaneous
inoculation of pure cultures of the diphtheria bacillus these
animals can be given a fatal form of the disease, and what is
more, the ectotoxins, the substances given off by these bacilli
when they are grown in fluid media, are intensely poisonous for
these little animals : Roux and Yersin found that 0*0002 mg. of
the dried — and impure — toxin, obtained by salting it out from
the fluid of growth, would surely kill a guinea-pig of 250 grams
within three days, or otherwise one gram would kill five million
guinea-pigs of like weight. Now exactly as with ricin the exhibi-
tion of minute, progressively increasing doses leads to the pro-
duction of a high order of immunity, and with this the blood
serum is found to contain the antitoxin in relatively very large
quantities. In other words, the tissues have elaborated, and

E

50 ADAPTATION AND DISEASE

elaborated in excess, a body substance which neutralizes the
toxin, and what is more, once started they continue for weeks
and months to elaborate the " antibody." 1

As to how these antitoxins are produced : it used to be
thought that there was a direct conversion of toxin into anti-
toxin. But this cannot be. Kroon 2 has shown that in the
horse a single toxin unit (of tetanus) can lead in the process of
immunization to the production of 1,000,000 antitoxin units.3
Toxins, it is true, may have some such function in the first place,
but later it is the cells themselves which assimilate the necessary
constituents and build up and discharge the antitoxins.

For this also has been clearly determined — namely, that it
is the cells that take up and fix the toxins which produce the
antitoxins. Ehrlich it was who drew the important distinction
between those substances which simply diffuse into the cell
and those which diffusing become fixed in the cell, calling atten-
tion to the fact that antibodies are not formed against members
of the former group, against, for example, alkaloids. The
members of this group disappear from the blood, but by alcohol
and other extractives they can be removed from the tissues.
This is not true with regard to the second group, or, more accur-
ately, substances of this second order may be extracted from
certain tissues but not from others, and it is the latter apparently
which, fixing the toxin, produce the antitoxin. The conjunctiva
ordinarily is most sensitive to abrin, which sets up a violent
conjunctivitis, but if this be administered locally in minute
but progressively increasing amounts, the conjunctiva can be
rendered insensitive or immune to this phytotoxin. Romer 4
rendered the conjunctiva of one eye in the rabbit insensitive to
abrin, killed the animal, and now triturated the conjunctiva of
each eye with a poisonous dose of abrin. The animal inoculated
with the mixture from the untreated eye died, that inoculated
with the preparation of the immunized eye survived, or other-
wise there was clearly a local production of antitoxin by the

1 Let me admit that this is a barbarous hybrid. There is, however, a certain
felicity about it, considering that it is a provisional word which should dis-
appear when we know, as we do not know at present, the exact chemical or
physical nature of this class of substances.

2 Munchener med. Wochenschr., 1898, 321 and 362.

3 See also McFarland, Textbook of Bacteriology, 3rd ed.

4 Archivf. Ophihahnologie, lii., 1901, 72.

ANTITOXIC AND BACTERIOLYTIC IMMUNITY 51

cells of the treated eye which neutralized the toxin and rendered
it inert. And with tetanotoxin the same has been abundantly
proved. This toxin in most animals combines specifically with
nerve substance. Take a toxic dose and emulsify it with tri-
turated brain substance, and the mixture is inert.1 As Metch-
nikofr" pointed out,2 the tortoise is not affected by tetanus toxin :
after inoculation none is taken up and fixed by the brain or other
organs, and no amount of injection of the tetanotoxin will induce
the production of antitoxins. Ford 3 also has shown by other
methods that antitoxin production only occurs in individuals
whose cells have the power of binding the toxin. The full
significance of these observations I hope to discuss in the con-
cluding lecture of this series.

But ectotoxin production is a property possessed by a minority
of the pathogenic bacteria. The whole group of typho-coli
organisms — the causative agents of typhoid and paratyphoid
fevers and bacillary dysentery and allied intestinal disorders, —
the tubercle and glanders bacilli, cholera spirilla, and the whole
group of pathogenic micrococci, whether Gram-positive, like
the streptococci and pus - producing cocci, or Gram-negative,
like the gonococcus and meningococcus, and the list could be
greatly extended : all these, when grown outside the body, do
not excrete toxic substances. Yet the body can be immunized
against these also, though here the immunity is of a different
order. It is not antitoxic, but bacteriolytic. The fluids of the
body gain the power of dissolving and digesting the bacteria.
The development or acquirement of this power has been gained
by all the millions of soldiers in the present war who have been
given antityphoid inoculations or, as has been the regulation
for more than a year, T.A.B. inoculations — inoculations of mixed
vaccines of the Typhoid, Paratyphoid A. and Paratyphoid B.
micro-organisms. All this, of course, is perfectly familiar to
every one in these days of the great war. By subcutaneous
inoculation of carefully measured doses of so many million bodies
of dead bacilli, the tissues and fluids of the individual acquire
the habit of immediately digesting living bacilli of the particular
species employed, and rendering their dissociation products
harmless.

1 See Wassermann and Takaki, Berliner Min. Wochenschr. xxxv., 1895, 5.
2 Traite de VImmuniie, Paris, 1901. 3 Zeitschr. f. Hygiene, xl 1902, 363.

52 ADAPTATION AND DISEASE

And we can easily demonstrate the fact, either within the
body itself or in the test-tube. Take the peritoneal fluid of
a normal untreated guinea-pig and introduce cholera spirilla
into it, and they will immediately begin to multiply. Take
another guinea-pig and inoculate it subcutaneously with minute
doses of dead cholera spirilla at intervals of a few days, gradually
increasing the amount inoculated, and follow this by repeated
inoculations of live cholera spirilla ; and now either (a) withdraw
some of the peritoneal fluid of the highly immunized guinea-pig
and add to a drop of this a suspension of living cholera spirilla,
or (6) inject living spirilla direct into the peritoneal cavity, and
with a pipette withdraw from time to time a drop of the peritoneal
fluid and examine under the microscope, or even (c) take a normal
guinea-pig, inject into it five or ten times the fatal dose of cholera
spirilla and follow this by injecting, also intraperitoneally, 1
or 2 c.cm. of the blood serum of the immunized guinea-pig and
from time to time remove a drop of fluid from the peritoneal
cavity and examine under the microscope, preferably in each
case upon the warm stage. Whichever means of experiment is
employed, it is seen that the spirilla lose their motility, swell
up, become rounded, then become progressively smaller, and, as
Pfeifler (to whom we owe the reaction) expresses it, they undergo
solution " like sugar in water." Any member of this group of
pathogenic microbes may be selected with like results.
Radziewski,1 for example, has observed the phenomenon with
B. typhosus, B. pyocyaneus, B. pneumoniae, Streptococcus pyo-
genes, and Bacillus anthracis. Clearly in the process of im-
munization the tissues of the body, or certain of them, have
gained the property of elaborating bacteriolytic ferments, or
substances which digest and dissolve the bacterial bodies — sub-
stances not present in the untreated or control animals, and
coincidently the treated animal becomes immunized, becomes
capable of withstanding and destroying, with little or no general
reaction or bodily disturbance, many times what otherwise
would be a surely fatal dose. And here is the striking fact :
this reaction is in general narrowly specific, so narrowly specific
that it is employed by the German School to distinguish between,
for example, closely related species of spirilla. If a guinea-pig
be inoculated with the cholera spirilla from man, its body fluids

1 ZeiUchr. f. Hygiene, xxxvii., 1901, 1.

PHAGOCYTOSIS AND ADAPTATION 53

have no effect upon the Spirillum metchnikovi which sets up
an intense diarrhoea in fowls, or with other non-pathogenic
spirilla obtained from water. Only rarely in animals immunized
against other species of bacteria do we note the existence of a
group reaction — a certain amount of digestion exercised by the
concentrated serum upon allied forms, though when the serum
is diluted the reaction becomes narrowly specific. For present
purposes I need not enter into the evidence afforded that for
this bacteriolytic action the coexistence of two bodies is essential,
namely the complement present in the body fluids of the normal
(and also of the immunized) animal and the immune body, or
amboceptor elaborated by the treated animal in the process of
immunization, nor into Metchnikofl and Bordet's demonstra-
tion of the presence of these two factors. All that is necessary
is to call attention to one of the commonplaces of bacteriology
that immunization is adaptation, and that while in some cases it
is of the nature of exaltation of properties already possessed,
in others it is clearly the acquirement of a new property.

Phagocytosis

I need scarcely say that by a different means of approach,
namely through the intimate study of cell function, the admirable
studies of Metchnikoff, extending from the early 'eighties of
the last until the beginning of this century,1 teach the same
lesson. Those studies brought together in his two works upon
the Comparative Pathology of Inflammation and upon Immunity 2
demonstrated in a very beautiful manner that in the repair of
injury, the destruction of microbes and the production of im-
munity, the leucocytes play a foremost part, and what is more,
that the development of immunity is intimately associated with
the accustomance, education, or adaptation of certain orders of
leucocytes and other cells. Cells which at first are not attracted
towards microbic foci eventually become actively attracted, cells
which at first show little or no signs of incepting bacteria eventu-

1 His later studies upon senility and the prolongation of life, while highly
suggestive, were not, in my opinion, of the same high scientific value.

a La Pathologie comparee de V inflammation, Paris, Masson, 1892. English
translation by Starling, 1893. Ulmmunite dans Us maladies infectieuses,
Paris, Masson, 1901. English translation by Binnie, Cambridge University
Press, 1905.

54 ADAPTATION AND DISEASE

ally take these up freely and digest them. As I have pointed out
elsewhere,1 while MetchnikofT, perhaps naturally, minimized the
part played by other factors in the inflammatory process, the
fact remains that phagocytosis and the acquirement of phago-
cytic powers play a highly important part in the successful
resistance to infection. As shown by Sir William Irishman's
simple and beautiful experiment (which formed the basis of Sir
Almroth Wright's opsonic technique), there is not one pathogenic
microbe which cannot be shown to be taken up and digested
sooner or later by the polymorphonuclear leucocytes of the
human blood.

Into the more intimate mechanism of this and other acquire-
ments I purpose to enter in the concluding lecture of this series.

In the next place, to carry on my argument, it is to be observed
that this is not a temporary acquirement by the individual.
The length of time during which the organism continues to
elaborate actively these new " antibodies " varies, it is true,
according to the micro-organism, but in all cases the amount
elaborated is over and above the amount of " antigen " involved,
and the production continues long after the antigen, be it toxin
or be it bacterial bodies, is used up. For months after a man
has been given one or two doses of dead typhoid bacilli his blood
serum either contains specific antibodies — agglutinins — or has
a different physical constitution ; and in some cases where there
is not artificially but naturally acquired immunity, that is to
say where a man has gone through a severe attack of typhoid
fever, agglutinating power has been noted for five years and
more after recovery. In fact, this retention of properties by
the inoculated or vaccinated men has necessitated the develop-

1 Inflammation, an Introduction to the Study of Pathology, being the
roprint (revised and enlarged) of an article in Sir T. Clifford Allbutt's System of
Medicine, 4th ed., Macmillan & Co., Ltd., 1909. This gives an epitome of
Metchnikoff's observations. I had the good fortune to be working in Metchni-
koff 's laboratory at the Institut Pasteur in the winter of 1890, at the time when
he gave his first comprehensive review of his studies upon Phagocytosis and
their bearing upon Inflammation, Infection, and Immunity. With his per-
mission I translated the lecture and published it in the British Medical Journal,
1891. Following the publication of the lecture I remember that I had a
correspondence with Sir J. Burdon Sanderson in the pages of that journal.
It is curious to-day to recall that he could not accept the phagocytosis
theory of immunity, on the ground that the education and adaptation of
the leucocyte demanded an individual intelligence on the part of the individual
cell.

THE LAW OF HABIT 55

ment of a completely new technique in our Army Service now
that all soldiers are inoculated against typhoid and the two
paratyphoid fevers. As Major Dreyer has more particularly
emphasized, the mere fact of agglutination is no longer of any
diagnostic value. The whole army agglutinates the specific
bacilli of all three diseases. Smallpox, naturally acquired,
usually confers a lifelong immunity ; vaccination, as conferring
a mild and modified disease with lessened general reaction,
confers immunity for five years or so, although with each repeti-
tion of the vaccination the indications are that the period of
immunity is materially lengthened.

Now we know that the leucocytes (which evidently are the
cells most concerned in the production of " antibodies ") have
but a short period of active existence : once they pass into the
blood they do not multiply there. We must therefore con-
clude that the toxins in the first place act upon the mother or
germ cells of the lymph nodules and bone marrow, that these
gain the new powers and pass them on to successive generations
of daughter cells. The same principle must obtain for other
active cells of the body.

The Law of Habit

Here we observe the working of a law which to my know-
ledge, if recognized, has not been dealt with adequately by the
biologists in general, a law which, it may be, is coming into the
ken of the physical chemists in their studies upon the effects of
recurrent stresses upon the crystalline structure of steel and
other metals. In a notable address delivered in 1896 Weigert,1
the Frankfort pathologist, laid down the law of inertia, the law
that once a cell is stimulated to perform a certain act it con-
tinues to perform that act for some time after the stimulus has
ceased to act. Fraser Harris,2 now Professor of Physiology at
Halifax, Nova Scotia, in 1900 took up the subject more fully,
calling attention to the inertia of rest, as exemplified by the
latent period of stimulated muscle, as well as the inertia of
active function. For myself there is here something beyond

1 Deutsch. med. Wochenschr., 1896, 635. See also Gesammelte Abhandlungen,
1906, 1.

2 British Medical Journal, 1900, ii. 741, and The Functional Inertia of
Living Matter, London, Churchill, 1908.

56 ADAPTATION AND DISEASE

mere momentum : the functional activity once started, at least
in the order of events which come under my ken, continues too
long to be comparable with physical momentum. I think I
see the setting in motion of a serial or cyclic process of inter-
cellular reactions and counteractions, the one reaction starting
the other — a process not contemplated by the physicist when
he speaks of either inertia or momentum. I prefer therefore
to employ for the present a non-committal term, and to speak
of the " law of habit."

Once again, we see this law in action from the bottom to the
top of the tree of life. Take, for instance, a culture of a pig-
ment - producing microbe, such as the B. prodigiosus, which
grows normally at room temperature, producing on moist bread
and other farinaceous foods colonies and plaques that are blood-
red : subject the culture for a few minutes to a temperature
just below that which will destroy the bacilli, or to a temperature
which kept up for another five minutes would kill all the members
of the colony. Now from this tube make cultures upon potato
flour or starch paste — in fact, upon the most favourable medium
— and these first cultures are absolutely colourless. According
to the temperature chosen and the time of exposure so, it may
be that in three or four days the colonies assume a faint pinkish
tinge, or it may be that repeated cultures and weeks of growth
on the optimum medium and at optimum temperature have
to happen before the pigment production is re-established,
thousands of colourless generations having elapsed. More is
needed than merely optimum environment. It is as though
the biophoric molecules of the bacilli have been converted to a
simpler constitution by the heat, or certain specific side-chains
destroyed, or it may be certain enzymes in the bacterial body,
so that, while the living molecules are still capable of multiplying,
the habit of pigment production is lost, and although all the
components are present in the foodstuffs presented, only very
gradually are they built together to produce the pigment. And
the same is equally true, as I have already indicated, with regard
to positive acquirements : it takes days and sometimes weeks
before bacteria take on the active dissociation of a foreign sugar,
or glucoside, or alcohol, or fat, but once this power is acquired
it is apt equally to be retained for generations in the absence of
that particular fermentescible substance, just, to recall, as the

HABIT AND SYMPTOMS 57

bacillus tuberculosis, typus humanus, may exist for months
in the body of the calf without losing its specific or typical
properties.

But here, also, we see the action of another general law,
which I owe to my old teacher James Ross the neurologist, of
Manchester, the law namely that properties which have been
most recently acquired are those which are most easily lost,
with its corollary that the longer a given environment acts upon
the individual and the race, the more firmly fixed become the
properties acquired in consequence of that environment. Ross
used to point out that it is the muscles which distinguish anthro-
poids from other mammals — those of the thenar and hypothenar
eminences — which first atrophy in systemic nervous disease.
I have pointed out elsewhere x that this law is constantly in
evidence when we study cases of reversionary (phylogenetic
reversion) and familial degenerations. What are the limits
within which this law is effective I hope to touch upon later.

Habit and Symptoms

We see these laws in action in the highest animals. There
can be no more striking examples of the action of the Law of
Habit than the abundant observations upon the production of
" antibodies " by the system once the cells have been stimulated
to elaborate them. But, as I pointed out some years ago,2
we encounter examples in all directions. We find that after
an acute inflammation, be it of the lungs or the sinuses of the
nose, we encounter cases in which the discharge continues weeks
and months after the causative agent has disappeared and the
discharges have become sterile. To us as physicians it is im-
portant to keep this in mind, for the recognition of this class of
cases materially aids prognosis : it has to be remembered that
not infrequently what is commonly regarded as symptomatic
of functional, not to say anatomical, disturbance may after all
be one or other manifestation of habit, and therefore wholly
curable, 'provided the habit can be interrupted. May it not be
that an agitation that is progressing to-day in the daily and

1 Principles of Pathology, vol. i. 2nd ed. p. 177.

2 See the address upon " Habit, Symptoms, and Disease," Chapter VIII.
of Part II. of this volume.

58 ADAPTATION AND DISEASE

Sunday press, and in certain exalted circles which have their
centre in Park Lane, would never have arisen if we, physicians
and surgeons, had a fuller grasp of this truth ? In other words,
I am inclined to believe that not a few cases of apparent organic
joint disease are habit conditions. I have pointed out that
mucous colitis has the earmarks of a habit condition, while
as to tics and other convulsive movements of nervous origin,
and hysterical manifestations, if not based upon habit then
due for their continuance to habit, their name, like that of
the Gadarene devils, is legion. As to these nervous cases, it
has been recognized from Lamarck 1 onwards that it is easier
for nervous stimuli of different orders to travel along well-worn
paths, and that thus there becomes established a condition in
which a minimal afferent impulse travelling without interrup-
tion along a particular path suffices to set in action first one and
then another of a group of co-ordinated nerve centres, so that
a minimal stimulus may eventually produce a maximal result.
On the other hand, the act of inhibition, or arrest of the passage
of a given impulse, if repeated, would seem to oppose so strong
an obstruction to the passage of impulses along a particular
path that soon a maximal stimulus will produce no result.
Here there is no anatomical change, only a matter of habit.
Put the patient under an anaesthetic or otherwise remove the
inhibitory block, and now stimulus is followed by the normal
reflex muscular act. Also, the longer the individual has been
the prey of a habit, the more difficult it is to bring about the
return to the normal. Here again our war experiences with
" shell shock " and the " fear neuroses " (as distinct from shell
concussion) are proving this point up to the hilt in the striking
difference obtained by immediate treatment behind the front,
as contrasted with treatment of the same order in special hospitals
in England, initiated weeks, and it may be months, later.2

Some of my readers may recall that I have applied this law
of habit to explain the metajrtasias, the conversion under altered
environment of tissues of one order into those of another — of
normal connective tissue into bone or cartilage, of stratified

1 Zoological Philosophy (translation by Hugh Elliot), Macmillan & Co.,
Ltd., 1914, pp. 349, 350.

2 Vide Sir John Collie's lecture at the Royal Institute of Public Health,
reported in the Times of June 14, 1917.

THE HABIT OF GROWTH 59

into columnar epithelium, and vice versa, as again to explain
the neoplasias or true tumours,1 here calling attention to the
partial antagonism between functional activities of the cells,
in which the food-stuffs absorbed are utilized in the production
of energy in the exercise of function, and the vegetative activities,
in which the food-stuffs absorbed are employed in the building
up of increased cell substance and utilized for growth and pro-
liferation. The time at my disposal forbids that I should enter
at length into the data supporting the " habit of growth " theory
of neoplasia, but as bearing upon cell inheritance of acquired
properties I cannot forbear calling attention to a form of cancer
of which I have come across at least two well-marked examples,
and of which there are several other cases by Van Hansemann
and others on record. I refer to primary squamous epithelioma
of the gall-bladder.2 Normally there is no squamous epithelium,
but only columnar epithelium in the gall-bladder. There is no
squamous epithelium anywhere in the neighbourhood. But it
is a familiar fact that 90 per cent or over of cases of cancer of
the gall-bladder are associated with the presence of gall-stones.
It has also been noted by several observers that the continuous
irritation set up by the presence of gall-stones leads to the meta-
plastic conversion of areas or plaques of the columnar-celled
mucosa, more particularly of the fundus of the gall-bladder, into
a stratified squamous epithelium. And here is the interesting
point : we encounter cancers of the gall-bladder, not of the pure
adenomatous or cylindrical-celled type, but containing definitely
epitheliomatous areas. In other words, in these cases the meta-
plastic squamous epithelium, originating from a columnar-celled
mucosa, has had its new character so firmly impressed upon it
that when it is stimulated to new growth it does not revert to
an adeno-carcinomatous type, but retains the impressed varia-
tion and gives rise to epitheliomatous squamous-celled down-
growths : an acquired variation becomes hereditarily transmis-
sible among these cells, just as must be the case when the
production of antibodies lasts for months and years.

1 See the address on " The Causation of Cancerous and other New Growths "
reprinted as Chapter II. of Part III. of this volume.

2 For the different forms of new growth, their classification and histogenesis,
see the address on " The Classification of Tumours " in Part III. of this volume.

CHAPTER V

THE INHERITANCE OF ACQUIRED CONDITIONS IN THE

HIGHER ANIMALS

Direct adaptation, or direct equilibration in the Spencerian
sense, can be proved to occur in a certain order of cases, both at
the lower end of the scale and at the upper, and this adaptation
may manifest itself as an acquirement of additional properties,
as well as by loss of properties already possessed by the individual.
But now, to advance further, it is necessary to bring forward the
evidence obtained from medical research that acquirements,
whether of defect or of excess, if they may be so termed, are
passed on to the next generation.

There can be no question that they are in the case of the
bacteria, but here so intimate and so direct is the relationship
between the organism and the environment that while, as already
stated, the longer the environment has acted on a given species
of microbe the longer the microbe retains the impressed property,
I cannot state dogmatically that there is any biochemical property
that is specifically fixed, still less, therefore, that any acquire-
ment remains fixed.

With reference to the higher animals we recognize full well
that the problem to be answered is girt with numerous restric-
tions, that conjugation and amphimixis in general act as a factor
tending to neutralize and dissipate the evidence of such heredity,
that in mammals we have narrowly to distinguish between
conditions of germinal origin (which alone are truly inheritances)
and those acquired during the uterine phase of existence ; again
between conditions of germinal origin and those impressed upon
the offspring during extra-uterine existence. Take the question
of the inheritance by the children of the tuberculous of a tuber-
culous diathesis : how are we to assure ourselves that, when the

60

INHERITANCE OF ACQUIREMENTS 61

mother is affected, the children have not been imperfectly
nourished in utero and thus rendered more susceptible to tuber-
culosis and to all other forms of infection, or in the case of alcohol-
ism that the lowered vitality of the children is not due to the
lack of care and general home misery which so often accompanies
maternal alcoholism, and becomes pronounced when both parents
are alcoholics ? For absolute evidence we are reduced to the
study of conditions in which the soma of the male parent has
been subjected to particular influences, and have to observe,
first, whether those influences have, judged by the condition
of the progeny, had influences of any order upon his germ cells,
and secondly, whether under any conditions the influences
which act upon the soma, or more accurately the tissues in
general of the male parent, produce identical disturbances in
the children. Clearly there are the two conditions of what may
be termed indirect and direct or identical inheritance.

Now, in the first place, we have clear evidence that the germ
cells are not, after the Weismannian conception (as usually
accepted), so sacrosanct that they are insusceptible to influences
which affect the body at large. Even though their growth is
restricted still they have to grow, and they have to maintain
existence, and growing they must absorb and assimilate material
brought to them by the blood and diffusing from the blood into
the lymph. If that lymph contains soluble toxic substances,
the germ cells are not precluded from absorbing them as do the
other cells of the body — and, like those other cells, from being
influenced by them. I have brought these examples forward
on previous occasions, and they have never been rebutted ; on
the contrary, the evidence of this order is steadily increasing.

Take first the oldest carefully studied example, that afforded
by Constantin Paul 1 in the 'fifties, in his studies upon the
effect upon workers exposed to lead : of 32 pregnancies in which
the husband alone was exposed to lead, there resulted 12 abortions ;
and of the 20 children born alive, 8 did not survive the first year,
4 died during the second year, and 5 during the third year.
And the lead appears to have a particular influence upon the
nervous system, for numerous other observers have called
attention to the frequency of epilepsy, idiocy, and imbecility
in the children of workers in lead. Nor need I recall that lead

1 Arch. gen. de medecine, xv., I860, 513.

62

ADAPTATION AND DISEASE

acts directly on the nervous system of the adult, producing
lead encephalopathy, retinal disturbances, etc.

Next we have Lize's parallel observations upon the workers
exposed to the fumes of nitrate of mercury in the manufacture
of looking-glasses,1 etc. : 12 pregnancies in which the father alone
was exposed, resulting in 4 abortions and 8 children born alive,
of whom 3 died before the fourth year and 1 alone was vigorous.
What can these figures mean other than that the metallic salts
absorbed into the system have affected the germ cells of the
father I

Carriere 2 inoculated a series of guinea-pigs, male and female,
with the soluble products of the tubercle bacillus. I have placed
his results in a tabular form : they are very instructive.

Effect of Inoculation of Parental Guinea-Pigs with

Tuberculin — Carriere

Births

Total
No.

Still-born.

Dying before
tenth day.

Surviving.

No.

Per cent.

No.

Per cent.

No.

Per cent.

Both parents

treated
Mother alone

treated
Father alone

treated

25

26
30

13

7
5

520
269
166

7
9
3

28-0
34-6
100

5
10

22

200
38-4
73-3

The results here are of the same order as those obtained by
Paul and Lize : there is the heaviest mortality, the greatest
number of still-births, the smallest number of those surviving
infancy where both parents have been subjected to the particular
soluble poison prior to conception. Next in order is the morbidity
and mortality when the mother alone is inoculated, but where
the father alone has been inoculated there is still evidence that
a high percentage — over 25 per cent — of the offspring are affected.
Nor do these observations by Carriere stand alone : Lustig,3

1 Union midicale, 1862, 106.

2 Arch, de mid. exper. xii., 1900, 782.

3 Cenlralblatt f. Pathologie, xv., 1904, 210 and 756.

INTOXICATION OF THE GERM CELLS 63

employing chickens, and Watson,1 working under Mott in London
and employing guinea-pigs, obtained results of the same order
by successive inoculations of the phy to toxin, abrin, to which
I have already referred. Both observers noted the resultant
diminished fertility, increase in the number of still-births, and
lowered vitality of those of the offspring that survived birth,
and both, like Carriere in the case of tuberculin, note particularly
that the offspring, instead of being immunized to the individual
toxin, are on the contrary distinctly more susceptible. Their
observations are on a par with the frequent clinical finding that
the children of those suffering from advancing tuberculosis are
more liable to succumb to tuberculosis than are those of healthy
individuals. We have here the explanation of the tuberculous
diathesis.2

The most conclusive observations, however, are those of
Professor Stockard 3 of Cornell Medical College, New York ; and
this in a field in which, although generations of medical men
have been convinced that they saw definite effects of a drug
descending upon the offspring, nevertheless social surroundings
have so complicated the subject that it seemed almost impossible
to arrive at a sure conclusion. It is indeed significant as to
the value of the pure statistical method and of Biometrics as
applied to the problems of evolution that Professor Karl Pearson 4
deduced from his Edinburgh statistics that the children of
alcoholics in the Scotch capital are, if anything, superior in
capacity to those of abstainers.

Any one who has had dealings with a guinea-pig knows how
small and well guarded is its mouth, so that forced and exact
feeding of these animals is out of the question. By a very

1 Watson, British Medical Journal, 1905, ii. 1091.

2 Let me confess that I am prepared to find that the cautious administration
of smaller doses of toxin over longer periods will have the contrary effect, that
of developing increased resistance in the offspring.

3 Arch. f. EntwicTc.-Mechanih, xxxv., 1912 ; Archives of Internal Medicine,
x., 1912; and American Naturalist, xlvii., 1913.

4 See correspondence between Professor Pearson and the late Sir Victor
Horsley in the British Medical Journal, 1911. All the same let me admit that
I am not a little attracted by Dr. Archdall Reid's contention that alcoholism,
by causing the further degeneration and eventual extinction of weakling strains,
if ruinous to the individual of these strains, is of ultimate benefit to the race.
But this does not signify that per contra alcohol is of benefit to the sound stock,
or that, as Professor Pearson's Edinburgh figures might infer, the more tem-
perate inhabitants of the Scottish capital are a weaklier lot than those who
partake freely of the national beverage.

64 ADAPTATION AND DISEASE

ingenious method Professor Stockard overcame the difficulty.
He constructed a special box-chamber in which he could subject
these animals to the fumes of alcohol, and he induced chronic
alcoholism by subjecting them for six days per week to these
fumes until they showed signs of intoxication. Treated males
were paired with treated females, normal males with treated
females, and treated males with normal females. The experi-
ment was very complete. Altogether, in his first series of 40
matings 25 gave no result, or the mothers aborted early and
ate their young, The remaining 15 matings gave 25 young
(in place of about 60). Of the 25, 8 were still-born, 7 lived for
a few days after birth and then died in convulsions, 4 were in
utero when the mother was killed (and of these one was deformed).
Six only survived.

Professor Stockard has supplied me with the following figures
regarding the results of mating of alcoholized fathers with normal
mothers. There were 24 such matings :

14 were either negative or gave early abortions ;

5 gave still-born litters (in all, 8 young) ;

5 only gave living litters (in all, 12 young).
Of the 12 living offspring from these 24 matings :

7 died in convulsions soon after birth ;

5 survived, but when two months old were only half the size
of control normal guinea-pigs of the same age.
Thus, under the influence of paternal alcoholism 24 matings
only produced as many surviving young as might be expected
from a single pairing of healthy guinea-pigs.

But this is not all. Professor Stockard has continued these
observations, and, calling attention more particularly to the
stigmata of degeneration observed in the offspring of this alcoholic
parentage, has demonstrated that if members of the second
generation which themselves have never been subjected to the
fumes of alcohol be now mated together, their offspring, members
of the third generation, exhibit the same stigmata to a still
more pronounced degree.1

He observed that those of the offspring of alcoholized males
and normal females which reached maturity were in general
undersized and nervous, and here is the point : if these second-
generation animals were brought up under normal conditions,

1 Proc. Soc. for Exper. Biol, and Med. xi., 1914, 136.

INHEEITED ALCOHOLIC DEGENERATION 65

that is to say, without alcoholization, and then unrelated bucks
and does of this second generation were mated together, instead
of their offspring showing a return to or towards the normal,
on the contrary the result was worse than when they were mated
with wholly normal animals, and their condition was worse
than that of their parents. There was a marked tendency to
progressive degeneration. In this third generation gross defects
presented, and deformities ; 17 per cent of them showed, for
example, defects of vision, were either eyeless or with opaque
corneas or complete cataracts. Mating together unrelated mem-
bers of this third generation, the outcome was yet more unfavour-
able, four dying soon after birth and three being completely
eyeless. In other words, two alcoholized great-grandfathers
influenced the progeny certainly mi to the fourth generation.

We have here the clearest evidence known to me of the
inheritance of acquired defects — evidence that fits in wholly
with our routine medical experience of the dangers of marriages
in families in which there exist already stigmata of degeneration,
diatheses of, various orders, and, more particularly, neuroses,
liability to migraine, hysteria, epilepsy, and the like — cases in
which there are not necessarily gross anatomical defects, but
rather indications of want of full development of those pro-
perties of latest phylogenetic acquirement, of the higher mental
and co-ordinative powers, and of racial immunity towards the
diseases which specially affect man. We note, familiarly, that
the children of such marriages are delicate, more liable than
other children to succumb to the infections of childhood, more
irritable and liable to nervous explosions, if not to one or other
nervous disorder, manifesting itself not necessarily in childhood,
but in adolescence, when strains which have little or no influence
upon normal individuals lead to exhaustion and nervous break-
down in one or other direction, according to the nature of the
strain.

With these experiments of Stockard before us it is absurd
to postulate that these defects of resistance are atavistic, due
to properties which have always been possessed by some one or
other strain which has been introduced into the family. To-day
we must recognize that infections of one or other order and
intoxication are capable of telling upon the parental germ plasm,
and that, at some definite point in the line of descent, conditions

F

66 ADAPTATION AND DISEASE

acting from without upon the germ cells have led to the acquire-
ment of conditions of defect. To us as medical men it is a minor
point whether there is inheritance of the exact defect seen
in the parent who has been subjected to a given form of
intoxication ; the essential fact is that soluble poisons acting
upon the tissues in general of the parent can act also upon and
modify the germ cells, so that at some definite period such change
occurs in the constitution of those germ cells that the constitu-
tion of the offspring is permanently affected, and that its germ
cells in their turn are incomplete in molecular structure.

It is not a little interesting that alcohol, lead, and bacterial
toxines in general have a profound, we might almost say specific
effect upon the nervous system in the adult or adolescent in-
dividual, and that they are so peculiarly liable to influence the
nervous system of the offspring. To this aspect of the matter
I shall refer later.

Such " parallel induction " of defects in both the body cells
and germ cells has been repeatedly noted of recent years by
workers in experimental genetics, the case that has afforded the
greatest amount of discussion being that of Salamandra maculosa,
as studied over a long series of years by Kammerer and recorded
in the Archiv fur Entwickelungsmechanik. This is a salamander
which normally hatches its eggs in the water, in which they
develop into tadpoles with gills. Very gradually Kammerer
accustomed the parent forms to live entirely on land in a damp
atmosphere, and with this, in place of producing some sixty to
seventy eggs the females became viviparous, giving birth to at
most six or seven young, with the tadpole or gill-bearing stage
completely suppressed, and breathing by lungs. That here the
environment to which the parents have been subjected tells
not merely on the growing embryos, but upon the germ cells,
is evidenced by the fact that when the members of this second
generation are supplied freely with water they produce tad-
poles, it is true, but now the metamorphosis from the water-
living and gill-breathing to the land-living and lung-breathing
stage, instead of taking months, as in the case of the normal
spotted salamander, occurs in the course of a few days.

Walter,1 I observe, objects that this is not a case of inherit-

1 Genetics, an Introduction to the Study of Heredity, New York, The Mac-
millan Company, 1913, p. 90i

PARALLEL INDUCTION 67

ance of acquired character, but that the normal gill-breathing
form represents a case of arrested development, i.e. of "neotenia."
To prove this it is necessary to show that the salamander
is the degenerate representative of some form no longer
amphibian, which, after having been accustomed to a purely
terrestrial existence, has taken to laying its eggs in the water,
and that the long gill -breathing tadpole stage is a reversion.
Is there the least evidence that this is so ? He argues that
change from gill -breathing to lung-breathing is not an acquired
character, but a purely germinal character that may be
either blocked or released by changing conditions of environ-
ment.1 Surely this is contrary to all we know or imagine
regarding the Amphibia. We have always been taught that
breathing by lungs represents evolutionary advance fitting the
vertebrate for terrestrial existence, and not a retrogression.
The complete suppression of the gill-breathing stage cannot be
regarded as other than something acquired, although, as this
is of the nature of a cutting-out of a developmental stage, while
on the one hand it may be regarded as progressive acquirement,
on the other it does not strictly come under the heading of the
positive acquisitions.

Also belonging to this border-line group of inheritance of
defects due to " parallel induction " are to be cited the cases of
increased susceptibility to specific toxines, to which I have already
referred — the observations of Lustig, Watson, and Carriere with
reference to abrin and tuberculin. To these may be added
the interesting observation of Schenck 2 and others that the off-
spring of hypersensitized fathers and normal mothers are them-
selves hypersensitized and exhibit anaphylactic phenomena, for,
as I have pointed out, anaphylaxis is a stage in the development
of immunity ; and, regarding immunity as a digestive process,
as the acquirement of an added power of dissociation of specific
toxines and conversion of the same into food-stuffs, the demon-
stration is in itself a proof that we may confidently expect by
fuller studies along these fines to establish firmly, what several
observers have already reported, namely the inheritance of

1 Walter quotes in support Marie von Chauvin's demonstration that the
gill-breathing salamander, Axolotl, can, by habituating it to progressive reduc-
tion of the water in its aquarium, be converted into Amblystoma.

2 Miinchener med. Wochenschrift, 1910, 2514.

68 ADAPTATION AND DISEASE

acquired immunity, the inheritance not merely of conditions
of defect, but of positive acquirements.

Here we come to the boundary of the territory already
acquired, unless, indeed, during the last three years of war, quiet
workers in the United States, whose observations I have been
unable to follow, have carried the demonstration yet further.
Years ago 1 1 pointed out the need to distinguish in this relation-
ship between extrinsic and intrinsic influences upon the germ
cells. Toxines are instances of the first order — introduced
from without they act upon the germ cells at the same time as
they influence the tissue cells in general. As examples of the
intrinsic and indirect, I instanced the effect of influences from
without telling upon the function of one or other organ and so
disturbing general metabolism that either the absence of normal
metabolites or presence of abnormal metabolites circulating in
the blood might bring about alteration in the constitution of the
germinal biophores, with resulting alteration in the constitution of
the offspring. In this way, I pointed out, gouty and rheumatoid
states and defect and excess of internal secretions may tell upon
the germ cells. It is thus that we best explain the development
of diatheses. I pointed out in 1912 2 that " if the cells of particu-
lar organs are highly susceptible to hormones and other active
principles of the secretions of other organs, is it not likely that
the parent germ plasm, the cell matter which by its active
growth is capable of giving origin to all these organs and tissues,
should likewise be susceptible to their action, and influenced by
their excess or deficiency ? Nay, should we not expect that
bodies of this nature, developed by the cells in their activity,
should act with greater ease upon the living substance of the
germ cells than extraneous drugs are likely to act ? As a matter
of fact, alterations in such glands as the thyroid, the pituitary,
the adrenal cortex, the pineal, are found to have a profound
influence upon the sexual function/' 3 Influences acting upon
these organs of internal secretion, and conditions acquired by
these organs, may well so modify them that the offspring in

1 On the Inheritance of Acquired Conditions, 1901, reprinted as Chapter II.
of Part II. of this volume. Article " Inheritance and Disease " in Modern
Medicine by Sir William Osier and T. McCrae, vol. i., 1907, p. 109. See
also Principles of Pathology, vol. i. 2nd edit., 1910, p. 199.

2 " A Study in Eugenics," The Lancet, 1912, ii. (Nov. 2).

3 I had laid this down in 1901. (Consult Chapter I. of Part II.)

THE VIA MEDIA 69

their development show excess or deficiency in the function of
the glands implicated in the parents, and " so conditions acquired
by the parents reproduce themselves in and become inherited
by the offspring." I am glad to see that one zoologist, my old
friend Professor E. W. MacBride, has joined himself with me in
this opinion, and with me sees in this possibility or probability
the solution of a long-standing difficulty.1

" Lamarck's contention was that the identical changes caused
in the structure of an individual animal or plant, by the action
of a novel environment, . . . are transmitted by generation to
that offspring, and continue to appear in successive generations
derived from that offspring even when the cause which set up
the original modification of structure ceased to act. . . . Darwin
. . . did categorically state that he attributed the origin of
congenital variation (by the natural selection or survival of
which he held that new species originate) to the action or in-
fluence of changed conditions upon the parental body, and
through it upon the reproductive germs." 2

It will be seen that the results of medical research here
quoted support strongly Darwin's contention, though at the
same time they indicate a group of cases in which the direct
adaptation of a particular endocrine organ may, through the
action of its internal secretion, bring about an identical change
in the same organ of the offspring. There is, that is to say,
one possible group of cases for which the Lamarckian theory
holds true, and that, oddly enough, as MacBride points out, along
the lines of Darwin's discarded hypothesis of pangenesis. Only
it is not by specific corpuscular determinants or pangens but by
secretions and the chemical activity of the same that the organs
of the body, or certain of them, are capable of influencing the
germ cells and bringing about a chemical and relatively permanent
change in their constitution.

After long years of heavy fighting we have, as it were, con-
quered our Vimy or Messines Kidge ; and, although for a time
we are held up, we can, as it were, indulge in a " Pisgah view,"
and see where and how the next advance is to be made.

1 Embryology of the Invertebrates, vol. i., 1914, and Presidential Address,
Zoological Section, British Association, 1912.

2 Sir E. Kay Lankester, The Kingdom of Man, Rationalist Press Associa-
tion, cheap reprints, No. 50, 1912, p. 74.

CHAPTER VI

THE PHYSICO-CHEMICAL BASIS OF IMMUNITY AND OF EVOLUTION

If we endeavour now to gather together the conclusions which
may reasonably be reached from a consideration of the data
afforded in the previous chapters, they appear to be the following :

(i.) Instead of there being no such thing as direct adaptation
or equilibration of the individual to his surroundings in the
Spencerian sense, this surely exists and can be demonstrated.

(ii.) In the lowest forms of life, among the bacteria, it can
be demonstrated that not only is this adaptation individual,
but that acquired properties of certain orders are inherited
through numerous generations, even when the cause which set
up the original modification has ceased to act.

(iii.) It can further be demonstrated that the inheritance
is not due to the survival and perpetuation of those individuals
only which present the favourable variation. All the members
of a given strain of bacteria subjected to a given modification
of environment can be made to vary in the same direction, and
acquire and transmit the same new property.

(iv.) Immunization of the higher animals is direct adapta-
tion. There could, in fact, be no clearer demonstration of the
process of direct adaptation than has been afforded by the
abundant researches upon the production of immunity.

(v.) It is clearly proved that there are drugs and toxines
which, acting on the tissues of the body, act also on the germ
cells, causing modification of the latter, of such a nature that
the offspring resulting from these germ cells are modified.

(vi.) The clearest evidences of this action are in connexion
with drugs and toxines which lead to conditions of defect, but
acquired special susceptibility and anaphylactic reaction are

70

DIEECT EQUILIBRATION: EPITOME 71

also inherited, indications that new properties gained by the
germ cells can also be inherited.

(vii.) The generative system is embryogenetically related to
the endocrine system, and is profoundly influenced by the internal
secretions of that system. The indications are that influences
from without affecting the organs of internal secretion and
modifying their secretions may through the hormones, etc. of
the internal secretions so influence the germ cells that the off-
spring are affected in the same direction as the parent, and that
thus acquired metabolic disturbances in the parent, whether
of the nature of defects or acquirements, are capable of being
inherited.

To the academic biologist the suggestion of a direct action
of external agencies as the prime cause of variation is as shock-
ing and as improper as would be the suggestion of a risque
story to some dear and prim old maiden lady. To the biologist
of to-day 1 the conclusions above noted are heterodoxy pure
and simple. And yet it is along these lines that medical research
is surely leading us. And what is more, these conclusions
harmonize with the experiences that come to us day after day
in studying our patients and their families.

The Morphological Concept op Unit Properties

Believing that we are in the right, where is it that the
biologists have gone wrong ?

1 I admit that this is too sweeping a statement. It was written with a vivid
memory of my correspondence with Sir Ray Lankester a few months previously,
and the attitude taken by him.

I do not think I can be accused of unduly advertising my wares or of too keen
a desire to claim priority when I have allowed sixteen years to elapse before
calling the attention of biologists in general to my address of 1901 (here repub-
lished in Part II.). The conclusions then reached by me regarding metabolites,
and, as we subsequently became accustomed to term them, hormones, and their
influence on the germ cells, have since then been enunciated by several prominent
biologists, by Heape, Bourne, Cunningham, MacBride, and Dendy, although
in each case without note of my earlier contribution, which, it should be added,
appeared not in an obscure local periodical, but in a leading organ of the medical
profession under a title that had no uncertain sound. The only British biologist
who to my knowledge has referred to my views is Professor Arthur Thomson
{Heredity, London, John Murray, 1912), but he also appears to be unaware that
I preceded Professor Yves Delage by some two years in offering a physico-
chemical hypothesis in place of determinants. On p. 454 he dismisses the
subject by stating : " We find ourselves quite unable to entertain (the view) that
it is possible to dispense with any postulate of representative particles."

72 ADAPTATION AND DISEASE

I venture to suggest that, from the morphological trend of
their studies, in striving to form a mental picture of the process
of evolution, morphologists have perforce conjured up separate
individual particles or molecular structures, each of which is the
bearer of an individual property or group of properties. Their
conceptions have perforce been in the terms of specific atomies.

We see this tendency elaborated — and, in his great sanity,
cast aside — by Charles Darwin in his pangenesis hypothesis.
It was exhibited again a merveille by Weismann with his ids,
idants, determinants, and the like, pure figments of the imagina-
tion which, nevertheless, have influenced and affected all our
generation. It has shown itself and been carried forward in
full vigour by the Mendelians ; one has only to consult their
diagrams to see how obviously their conceptions and explana-
tions assume a structural form. It is because he cannot picture
in his mind an atomy or structure making its first appearance
in the germ cell, like Venus arising out of the sea-foam, and
henceforth becoming a portion of the hereditable substance of
that cell, that Professor Bateson finds it impossible to imagine the
positive acquirement of new properties by the individual, and so
is driven to the remarkable hypothesis quoted in my first lecture.

But before all, it is the Freiburg philosopher who has led our
generation of biologists into Nephelococcygia — " Cloud Cuckoo
Land." Much as we owe to him for arresting the vague generaliza-
tions upon heredity prevalent until the 'eighties, and for his
confirmation, experimental and otherwise, of Galton's doctrine
that conditions such as mutilations and use acquirements are
not inherited, I myself am inclined to think that Weismann's
teaching has, on the whole, done more harm than good. He
was a pure morphologist, and his hypothesis was purely morpho-
logical. Had he been familiar with the physical chemistry of
his day he could not have ventured to publish it ; had he been
a physiologist with any appreciation of function he must have
modified it extensively. It is, in short, an impossible hypothesis,
and in a very extraordinary manner all its main postulates are
found contrary to experience. As I pointed out ten years ago,1

1 In the chapter upon Inheritance and Disease in Osier and McCrae's System
of Medicine, part of which is reprinted ("The reductio ad absurdum of Weis-
man's Theory ") in Part II. of this volume, Chapter VI. ; see also Principles
of Pathology, 1st edit., 1908, p. 119.

NEPHELOCOCCYGIA 73

our knowledge of the size of the molecule of albuminous matter
renders the existence of thousands of individual molecular
determinants in the minuter portions of the nucleus a matter
of sheer impossibility : they could not be packed in the space.
If I remember aright it was Herbert Spencer who pointed
out that in one feather of the peacock's tail there are 480,000
regions or parts capable of independent variation, and that
therefore, so Weismannism demands, in each of the many micro-
scopic " ids " or chromomeres in the nuclear chromatin of
the peahen's ovum there must be 480,000 determinants for a
tail feather alone. How many for the rest of the body passes
computation !

It is this faulty and untenable morphous or morphological
concept of unit-properties and hereditary entities that makes
it as difficult for Professor Bateson to picture how new accretions
(whereby he images new discrete determinants) can make their
appearance in the germ cell, as it was for the late King George
of glorious memory and muddled to imagine how the apple
got into the dumpling.

The Constitution of Living Matter

Suppose now we start from known facts and known pheno-
mena, and endeavour upon these to build up our idea of the
structure and nature of the germ cell, and so of organic basis
of heredity, variation, and evolution.

What, in the first place, do we know of the constitution of
living matter ? Will this help us ? I think it will. We know
that whatever form of life we investigate, be it in animal or
plant, mammoth or microbe, whatever form we analyse, and
whatever tissue of that form we analyse, leaving out of account
water and certain common vehicular salts to which no specific
vital functions can be attributed, just one order of highly
complex organic compounds is common to and to be isolated
from all forms of living beings, and these are the proteins. This
characteristic universal presence of these highly complex organic
compounds in itself indicates that they are intimately associ-
ated with vital functions. It is true that when we isolate
them chemically they are inert, in other words in the living

74 ADAPTATION AND DISEASE

organism with its many activities they are evidently present in
another form ; living matter, therefore, as I have expressed it
elsewhere, contains as an essential constituent proteidogenous
rather than proteid matters.

The Structure of the Proteins

But the structure of these proteins, as worked out first by
Curtius and Schutzenberger, and as more fully demonstrated
by Ernil Fischer through the actual synthesis and building up
of bodies of protein nature in the laboratory, is highly significant.
They are molecules so large that they will not pass through fine
filters, giant molecules but a little below the limit of visibility
under the highest powers of the microscope. Take one of the
simpler crystallizable proteins, namely Haemoglobin, its com-
position appears to be approximately Cyi2HngoN 0 FeS ,
in fact, the average molecular weight of proteins has been
estimated at about 15,000, or, roughly, one thousand times that
of a molecule of water. And the haemoglobin from each species
of vertebrate is a different chemical substance, nay more, our
researches in immunity indicate that there may be differences
between the haemoglobins of two members of the same family.
Certainly no two analyses of separate samples of crystalline
haemoglobin obtained from the same species give identical
results. By digestion, as every board school child knows, we
break these proteins down into peptones and albumoses. By
further disintegration Kossel and others have broken down
these peptones and protones into their constituent amino-acids,
bodies belonging to the fatty acid series — aminated fatty acids.
What Fischer has done has been to manufacture these aminated
fatty acids (acetic, propionic, etc.) in the laboratory, and
then he has accomplished the reverse process of linking these
together, until with five, ten, fifteen, and more linked together
he has obtained bodies affording the chemical tests for
peptones.

Now these amino-acids have one striking property — they are
amphoteric ; they possess both acid and basic properties, acid
because of the contained COOH group or groups, basic through
the NH2 group or groups. Thus we conceive these peptones

THE PROTEINS AND THEIR STRUCTURE 75

as composed of polymerized molecules, formed in trie main of
amino-acid molecules linked together by their otherwise un-
satisfied NH and CO affinities. Thus, to modify Hofmeister's
illustration, we may represent a portion, at least, of the peptone
molecule as follows :

/ ,J ^H-CH-COT^ ^ \ ^

cu3 I __ co°*

>

**toy». ~~ARTlCi

TVROSIN. A°

Fig. 1.

Each section between the dotted fines represents an amino-
acid nucleus. I speak of these as glycocoll, leucin, etc., but
it will be seen that in the leucin nucleus, for example, an H
of the NH2 group is replaced by the CO of the glycocoll nucleus,
and the hydroxyl of the COOH group by the NH of the tyrosin
nucleus. In Fischer's terminology the nucleus is leucyl, not
leucin. The peptone molecule is thus represented as composed
of a main chain in serial repetition of glycocoll molecules,

(#)— NH-CH-CO— {OH),

(H),

in which the NH affinity on the one side and the CO on the
other have been satisfied by linkage with a like molecule, while
an H of the CH2 is substituted (in the figure) by methyl-para-
oxybenzene, acetic acid, butylamin, etc. We have, that is to
say, a main glycocoll chain with a series of free-swinging chains
capable of being replaced or modified by processes of oxidation
or the action of enzymes.1

1 So as not to introduce complications I have here purposely left out of
account a consideration of the relationship of the purin nuclei or radicles to
such chains ; it is necessary, however, to keep in mind that the latter are
important constituents of the nuclear matter, and of what later I shall refer to
as the biophores.

76

ADAPTATION AND DISEASE

The Complexity of the Proteins

But the complex protein is composed of and breaks down
into peptones. We can therefore represent it either as a chain
or a ring of linked peptone molecules, or simply as in Fig. 2,

in which the hexagons A toF re-
present each a peptone nucleus
with its ring of glycocoll nuclei,
with, however, the " swinging
side-chains," as I have termed
them, indicated as in the main
unsatisfied, save at M, M, M,
and H. This represents but
the central skeleton of the
protein molecule, a frame in
which to place one's mental
picture of its nature and com-
plexity. Imagine that com-
plexity when, in the haemo-
globin molecule, for example,
there is somewhere inserted among the 700 or so carbon
atoms just an atom of iron and two atoms of sulphur. The
nuclei or radicles of which these are essential constituents
must have one particular place among the other nuclei
composing the ring or chain. Consider also the number of
vulnerable linkages — free-swinging side-chains that may be
detached by compounds in the neighbourhood having stronger
affinities and be replaced by others, thus altering the con-
stitution of the protein molecule as a whole. In the much
simpler bodies with which the organic chemist is in the
main concerned — bodies, for example, like those of the
benzole derivatives or the carbohydrates — we know how the
transfer of a given radicle from, say, the alpha to the delta
position upon a ring brings about profound change in the chemical
and physical properties of the compounds. Bodies, for example,
of identical molecular weight, which on analysis contain the
same number of atoms of carbon, hydrogen, and oxygen, and
on analysis afford identical radicles, mav be found to differ to a
remarkable degree. When two carbon atoms are united together
there are, or may be, six free affinities, and when these are

THE BIOPHOKES 77

satisfied by six different univalent groups, twelve different
arrangements are possible.1

Even if for the time we
leave out all consideration of C ***

introduction of new orders |

of " swinging side-chains " A — O — B B — O — A

think of the number of i

permutations and combina- q q _ £ Q — O— E

tions that are possible

when, instead of two, there p ' p

are in the ordinary protein

molecule hundreds of carbon atoms.

Fig. 3.

The Biophores

If, therefore, we regard the biophores or molecules of living
and heritable matter in the germ cells as, if not proteins, at
least proteidogenous, giving rise to proteins when dead or killed
in the process of analysis, obviously it is not necessary to demand
a separate determinant, a separate molecule for each specific
property ; it is simpler to regard properties inherent in the bio-
phores as the expression of the constitution of those biophores,
of the mode of linkage of the various nuclei, together with the
number of those nuclei and the nature of their swinging side-
chains. This conception is within the bounds of physical possi-
bility, and Weismann's ids, idants, determinants, and the like,
certainly are not.

Let us therefore accept as a working hypothesis this con-
ception of the chemical constitution of the essential living matter
of the individual, namely that it is, as I express it, proteido-
genous, giving upon analysis bodies of proteid nature, and that
these biophores or molecules endowed with vital and heritable
properties are composed of rings or it may be chains of arnino-
acid nuclei in series, each nucleus composed of a central ampho-
teric glycocoll group, to which are attached varying orders of
side-chains.

We know, in the first place, that in conjugation or amphi-
mixis the one element of the two germ cells, male and female,
that is contributed in approximately equal portions is the nuclear

1 See Bayliss, Principles of Physiology, p. 268.

L. lil

0* ri

78 ADAPTATION AND DISEASE

chromatin. We cannot help being impressed by the extra-
ordinary sequence of events all tending to this end result. No
other element common to the two germ cells is partitioned out
so meticulously. There is no other element common to the
two cells that is doled out with the same equivalence. Next,
making a careful survey of heritable and inherited properties,
we are forced to the conclusion that the likelihood is equal,
that the offspring inherits either from the male or from the
female parent. The conclusion is inevitable that the biophoric
molecules are conveyed in and by the nuclear chromatin. In
the bacteria which are so lowly that they possess no central
nucleus we observe nevertheless that chromatin-like granules
are scattered through the body substance, and that in

binary division these chromatin-
like granules undergo a certain
simple accumulation or arrange-
ment, to the end that each of
the daughter cells is endowed
with half of this specific chromatin-
like material.
Pia. 4.-Celi ^ndergomg mitosis ; 0r otherwise, whereas in these

skein stage to show the duplica- lowest forms the living Cell Sub-
tion of the chromatin thread . • n e i

with beading of the same, which stance is made up ol granules
is seen in some species. Weis- 0f chromatin, each surrounded

mann regarded these heads or , o , -, i • i

chromomeres (id) as represent- by an area ol cytoplasm, higher
ing his ids. c, centrosome ; up jn the scale these granules of

cyt, cytoplasm. f . 5T _ .

biophoric matter are collected into
a central node, the nucleus, which is separated from the
circumambient external medium by the cytoplasm or cell
body. We may indeed speak of a triple protection, namely
the cell wall or membrane, the cytoplasm, and the nuclear
membrane, all of them on the whole acting as conservative
agents, tending to preserve the biophores from sudden change
from without.

Vital Processes of the Cell

But to continue, as it were, to rehearse our catechism,
while on the whole conservative, this system is exposed to con-
stant change, owing to the fact that it is not inert but is constantly

THE NATURE OF GROWTH 79

reacting with that external medium, receiving matter from the
same either (1) in bulk (phagocytosis), or (2) by diffusion, or (3)
by surface absorption, and discharging other matter into the
same (excretion). And if we study carefully these processes of
assimilation and discharge it is to be made out that the outer
cell membrane is semi-permeable, preventing the entrance of
certain substances, permitting the entrance of others, whether
directly, or after a preliminary dissociation of the same into
simpler molecules by means of its extracellular secretions. And
once these food-stuffs are taken into the cytoplasm, there again
they are, if necessary, broken down still further into yet simpler
molecules by intracellular enzyme action.

It is being more and more surely established that these
enzymes are of nuclear origin,1 resulting from the discharge of
certain constituents from the nuclear substance, but this is not
a necessary part of my argument. The essential point is that
the food-stuff's, protein, carbohydrate, and hydrocarbon, are not
utilized by the cell as such, but are dissociated, and it is these
disintegrated and dissociated molecules that are utilized by the
cell in one of two ways, either to supply energy for the per-
formance of work, or on the other hand for growth.

The Nature of Growth

I feel almost apologetic for taking up time over such
elementary matters. And yet these are matters that, while
absolutely basal, are absolutely neglected by ordinary biologists.
Take, for example, this matter of growth. With all the extra-
ordinary width of his learning, D'Arcy Thompson 2 has just
achieved a work of some 750 octavo pages on Growth and Form,
full of most interesting observations and data, and with a

1 See Pt. II. Chap. V., "The Dominance of the Nucleus."

2 Growth and Form, Cambridge Univ. Press, 1917, p. 203. [It is but just
to D'Arcy Thompson to point out that his title was meant by him to signify
— and that by a figure of speech which, while wholly permissible, may
be construed otherwise — "Form in relationship to growth"; in other
words, while he occasionally touches upon matters physiological the work is
essentially morphological, " an easy introduction to the study of organic Form,"
as he terms it in his Preface. Perhaps I may be forgiven if it was just these
occasional physiological excursions that arrested my attention rather than
the more severely mathematic and physical groundwork of his volume.]

80 ADAPTATION AND DISEASE

distinction of style that is a pure delight, but with little more
than a page devoted to the essential nature of growth of living
matter, and that page upon the text that the older naturalists laid
stress upon the fact or statement that inorganic bodies (crystals)
grow by agglutination, organic bodies by intussusception. The
living cell, he states,1 grows very much as a piece of glue swells
up in water, by "imbibition" or by interpenetration into and
throughout its entire substance. From his statement growth
might be nothing beyond the absorption of water by a semi-
fluid, colloid mass. Now one has only to consider a moment to
see that " intussusception," " imbibition," " intercalation," are
nothing more than bald and blank terms. They are inane ; they
cannot possibly explain how two molecules of living matter appear
where there was but one before, two grains of wheat where one
was put into the ground. Growth is one of the great underlying
phenomena of living matter, and zoologists and botanists have,
in a simple Topsy-like manner, been satisfied that the pheno-
menon occurs, and have been amply content to rest with the
demonstration of the stages of mitosis and cell division.

But what has happened to the cell substance that precedes
and impels mitosis ? Obviously there is an increase in the
amount of living matter, and this translated into physical terms
means a multiplication of the molecules of living matter, or, as
I term them, the biophores or biophoric molecules.2 If we
regard these complex proteidogenous molecules as a chain, or
more simply as a ring of ammo-acid radicles, there is but one
way in which this can multiply, namely by the same method as
obtains in the growth of a crystal. Let us take the simplest
case, that of the crystallization of sodium chloride out of watery
solution. We know that when the solution or dilution of the
sodium chloride reaches a certain point the salt separates into
its constituent ions of sodium and chlorine. But while separated
these maintain a certain relationship or association. It is
remarkable that the chlorine ions do not escape or volatilize
into the air. When the concentration of the solution reaches

1 Loc. cit. p. 203.

2 I employ this term, introduced by Weismann, rather than manufacture
a new term, because our fundamental conception is identical, although it is
necessary to point out that our ideas as to the constitution of the biophores do
not entirely harmonize, he regarding them as collections of molecules, I as
individual molecules.

THE NATURE OF GROWTH

81

a certain point each ion of chlorine unites with a corresponding
ion of sodium, and aided by a sharp point or inequality of the
surface of the containing vessel, the process of crystallization
begins. The very existence of one crystal is seen to cause other
junctions to occur in its immediate neighbourhood, and indeed
just that one series of junctions necessary to form sodium chloride.
Despite the presence of other salts in solution we gain pure or
almost pure crystals of the one substance. We have in this
way growth of solid sodium chloride.

Growth of living matter, if more complicated, because the
structure of the biophores is more complicated, must neverthe-
less be of the same order. Its nature is best suggested by the
following diagram.

^

V*

•7

* * ♦ V

J '

; si.

Fig. 5. — Diagram of growth of biophoric molecules.

Namely, granted that the already existing biophore finds itself
in a medium — the nuclear cell sap — containing the necessary
radicles, should that by one of its attachments attract to itself
one of its component radicles, there is then started a process by
which in orderly sequence the other radicles become attached,
until there is built up a compound molecule, identical with the
pre-existing molecule, in association with which it has become
developed.

This conception — namely, that the growth of the biophores,
or otherwise of the molecules of living matter, is essentially of
the same nature as is that of crystals — is materially aided by the
recognition that crystallization does not of necessity demand
the production of rigid rectilinear figures, that, as Lehmann 1
was the first to point out in 1904, there are in nature abundant
examples of fluid, non-rigid, crystalline forms, and, as Aschofi

1 O. Lehinana, Fliissige Crystalle, Leipzig, 1904.

G

82 ADAPTATION AND DISEASE

and I were the first to note x in 1906, the animal organism
affords abundant examples of these fluid crystals. As D'Arcy
Thompson well remarks,2 " the phenomenon of liquid crystalliza-
tion does not destroy the distinction between crystalline and
colloidal forms, but gives added unity and continuity to the
whole series of phenomena."

1 Adami and Aschoff, Proc. Royal Soc. B., 1906, 359, and Adami, " The
Myelins and Potential Fluid Crystals of the Organism," Harvey Lectures,
2nd series, 1908, 117 (here reprinted as Chapter IV. of Part II.), and Aschoff,
Verhandl. Deutsch. Pathol. Gesellsch. x., 1907, 166.

2 Growth and Form, p. 204.

CHAPTEK VII

REVIEW OF CERTAIN CONCLUSIONS

But if we accept these views regarding the existence of biophoric
molecules and the essential nature of growth as involving a
multiplication of the same by a process akin to crystallization,
we are inevitably led to certain very interesting conclusions,
which may here be rapidly passed in review. Each, it is true,
might well be expanded into a chapter, but the limitations set
by this course of lectures demands that I do little beyond indicat-
ing the headings.

(i.) The Continuity of the Germ Plasm. — Weismann conceives
his biophores and determinants as handed down direct from
parent to offspring, so that the germ plasm is potentially eternal,
and that of a thousand generations back is contained in the
germ cells of the generation of to-day and has its influence upon
the configuration of the body of the individual derived from
those germ cells. This he speaks of as the " continuity of the
germ plasm."

But, obviously, this is not a continuity of substance, as Weis-
mann implies, but merely a potential continuity of molecular
arrangement and constitution. For growth and multiplication
of the living molecules, for one conjugated ovum and spermato-
zoon to give rise to the countless millions of sperm cells of the
male of any of the higher animals,1 there must be such active
reproduction of the parental biophores that the chance of one
of the original biophores of a grandparent finding itself in the
nucleus of a sperm cell must be infinitesimal. The likelihood
is that while such, it is true, control the constitution of the new

1 According to Loeb, the average seminal ejaculation in man contains
226,000,000 sperm cells. Consult further Marshall, Physiology of Repro-
duction.

83

Si ADAPTATION AND DISEASE

biophores, they sooner or later, in the metabolic processes of
the cell, become dissociated and broken down.

(ii.) On the Functional (Katabiotic) and Vegetative (Bioplastic)
Activities of the Cell and the Individual. — In unicellular organisms
the growth of the individual presents itself merely as increase
in size : in multicellular organisms, while there may be increase
in size of the component cells to a certain extent, growth is
brought about in the main by cell proliferation and increase in
the number of the component cells. But both in protozoa and
metazoa (and protophyta and metaphyta, if the terms be per-
missible) there are limits to this process. Here I would but
rapidly refer to what constitute these limits, namely, they are
conditioned by the relationship or interaction between the
individual and its environment. Broadly speaking, the indi-
vidual continues to grow until an economical mean is established
between the amount of food-stufis afforded by the environment
and the absorptive surface that the individual can present to
that environment. It is, as I have pointed out,1 a false con-
ception to regard the multicellular organism as a colony of
individual cells which remain united for self-defence and mutual
advantage. Rather it must be recognized that where the uni-
cellular mass of protoplasm undergoes enlargement, the greater
the mass the smaller relatively becomes the surface area, and
this diminution proceeds in a rapidly increasing ratio, until
growth is arrested by the fact that the amount of food-stuffs
absorbed cannot keep pace with the needs of the organism.
Nuclear and cell division and multicellularity constitute the
simplest mechanism whereby the nuclear and cell surface areas
can be increased in a yet greater ratio, and the greatest effective
working of the individual obtained by decentralization. We may
thus regard the mass or size attained by the mature individual
of any species as representing the sum total of protoplasmic
matter of the constitution peculiar to that species which is
capable of existing as an entity under the particular conditions
of its environment, the multicellular individual acquiring its
greater size and more complex activities by means of nuclear,
followed by cell division.

From physical considerations I thus arrived at the same view
as that mentioned by Whitman in America (1893) and Sedgwick

1 Principles of Pathology, 1st ed., 1908, vol. i. p. 35.

ANALYSIS OF CELL ACTIVITIES 85

in England (1895), and best summed up by a modification of
the aphorism of the botanist De Bary quoted by D'Arcy Thomp-
son : " It is the individual that forms cells, not cells that form
the individual." This is diametrically opposed to the teaching
of Virchow in his Cellular "paihologie that " Every animal appears
as a collection of vital units, each of which bears in itself the
full character of Life."

In either case, whether in the unicellular or in the multi-
cellular organism, the individual continues to grow and exhibit
predominantly vegetative activities until it reaches this limit
of economical relationship between itself and its environment.
In the very earliest stages the vegetative activities alone are
evident : as the above-mentioned limit is approached these
become less and less, and the cells are more and more engaged
in functional activities, that is to say, they utilize the food-stuffs
received, not in building up more living matter, but for the
performance of work and liberation of energy. And to the
extent that growth necessitates a building-up process from the
food-stuffs presented to the cell, and function necessitates a
breaking down and burning up 1 of the same food-stuffs, to this
extent the vegetative or — as Weigert terms them — bioplastic,
and the functional, or katabiotic, activities are opposed processes.

(iii.) Metabolism. — It may be thought that in mentioning
these matters I am wandering far afield, and through a singularly
arid field at that, nay more, that I do but touch upon or outline
matters which to be grasped in all their bearings need a much
fuller presentation. I touch upon them purposely in order to
emphasize that these matters of adaptation and evolution have
of necessity to be approached from the aspect of function and the
dynamics of living matter, rather than from the point of view of
cell statics. " Function," it has been said, " precedes structure,"
and it is the study of cell function that must afford the key.

If next we proceed to enquire into the state of our knowledge
of the chemistry of these active processes occurring within the
cell, we reach a further stage, for the more we study the processes
of absorption of food and utilization of the same — the metabolic
processes whether, as Gaskell termed them, of anabolism, building

1 It is not (as one is apt to imagine) the dissociation of the food-stuffs that
liberates energy, but the burning up of the same, i.e. the oxidation of the
dissociated food-stuffs.

86 ADAPTATION AND DISEASE

up, or katabolism, breaking down — the more it is borne in upon
us that the greater number — some would say all — are the out-
come of enzyme action.

(iv.) On Enzyme Action. — This is not the place to discuss
the nature of enzyme action in detail. For the most authorita-
tive treatment of the subject Professor Bayliss's works should
be consulted. Suffice it to say that the simplest and now gener-
ally accepted view is that there exist in the cell, and are dis-
charged from it, intermediary bodies — katalysts — which promote
and hasten chemical dissociations and associations without them-
selves being involved in the final stage, and that the simplest
diagrammatic representation of their action is as indicated in
the diagram in which F represents the go-between, the body

endowed with enzyme action, S the
substratum or body which undergoes
dissociation, and R the receptor
or body which is aggrandized, with
which the dissociated moiety or
f j part of the substratum enters into

union. There are, I hold, always
these two phases of dissociation
and association in each instance of
ferment activity, even if R be
F 6 but a Hydrogen or Hydroxy 1 ion.

The process is comparable with the
katalytic process of manufacture of concentrated sulphuric acid
from sulphurous anhydride, through the agency of nitrous acid.
The nitrous acid HNO? takes up a molecule of 0 from the air,
becoming nitric acid, and acting as carrier gives this over to the
sulphurous anhydride H2S03, converting this into sulphuric acid
H2S04, with this becoming free itself to take up another molecule
of oxygen and repeat the process with another molecule of sul-
phuric anhydride. And the process continues until a certain
equilibrium is reached between the amount of anhydride and of
sulphuric acid present, unless the H2S04 be removed so soon
as formed, in which case eventually a trace of nitrous acid can
convert a maximum, or theoretically infinite, amount of anhydride.
And this action is under certain conditions reversible.

Now, as I have already stated, the evidence is conclusive
that none of the ordinary food-stuffs — the proteins, carbohydrates

ENZYME ACTION

87

t£X>

and fats — are utilized by the living cell as such : through enzyme
action, whether extracellular or intracellular, they are broken
down into simpler substances, into their component radicles.
The evidence is accumulating yearly more and more that, if
not the enzymes themselves, at least the zymogens, the bodies
from which the enzymes are derived, originate in, and are dis-
charged from, the nucleus, the dynamic centre of the cell.1
Already there is evidence that certain of the amino-acids, and
it may well be of groups of conjugated amino-acids, constituting
Fischer's polypeptids, manifest enzyme action. Their amphoteric
constitution in itself favours the possession of these properties.

We can visualize the pro-
pi A.

cess thus : Let A represent

a peptone or polypeptid
molecule of food-stuff small
enough to be absorbed or
diffused from the external
medium into the cytoplasm
of the cell. This may either
(1) be broken down into
simpler radicles, a', a", a'",
and those radicles utilized by
a direct dissociation process
for the growth of the pro-
teins of the cytoplasm (B, Fig. 7) or, it may be,
nuclear biophoric molecules themselves ; B, through

unsatisfied linkage (&'), exer

*J' 0!"

Fig. 7.

the

an

~~~\^ C a „«*,A,a-

cising a stronger attraction
for certain of the constituent
radicles of A, such as a', in
which case a' becomes part
and parcel of B, and a" and
a'", whether combined or
loosened, are left free in the
cytoplasm. Or (2) where the
conformation (that is to say,
constitution) of A and of B
(or better A' and B') is not
such as to permit of this direct attraction and linkage, it may
1 See the address upon " The Dominance of the Nucleus," Pt. II. Chap. V.

Fig. 8.

88 ADAPTATION AND DISEASE

happen that free radicles, C, suspended in the cytoplasm may-
possess the necessary linkage, and in that case these will act as
catalysts or enzymes, first attracting away the radicle a! from
A', and then in turn by superior attraction being impelled to give
it up to satisfy the linkage " b " of B'.1

Possibilities arising out of above View of Enzyme Action

But if this conception and visualization of enzyme action
approaches the fact, do you see where it leads us ? I am not
going to take up all the lines of thought and possibilities opened
up by this conception, but only those bearing directly on my
theme.

1. In the first place the remains of A left free in the cyto-
plasm, if of amphoteric type, may in itself act as a catalyst.

2. In the second we are given an understanding of the
mechanism whereby foreign proteins may be utilized by the
cell. Those proteins are complexes with multiple linkages :
they are composed of combinations of the various amino-acids,
and the number of amino-acids is limited. It is, therefore,
inevitable that existing enzymes, whether extracellular or intra-
cellular, will loosen and detach certain of the radicles, thus dis-
sociating the protein and affording food-stuffs capable of being
utilized by the cell. But doing this, to take the simplest case,
the relative numbers of molecules of the different amino-acids
presented to the cell — and to its nuclear biophores — will differ
from the normal, and, if the exhibition of the foreign protein

1 This conception of attraction and separation of radicles from a compound
is, I admit, difficult to accept when — as we are apt to do — in the mind's eye we
picture the molecules of the different salts, e.g. simple salts, like NaCl or KN03,
as solid, well defined substances. It must, however, be remembered that in the
cell we deal with substances in solution, that the more dilute this solution, the
greater the liability towards ionization, towards a loosening of the constituents
of a compound one from the other. Under the conditions, therefore, in which
enzyme action proceeds, we may safely predicate that a relatively slight attract-
ive force is needed to detach molecules or radicles from one compound, causing
them to become associated (also loosely) with other molecules, and to be in
turn detached from this relatively loose combination. In solution, that is, we
have a state of unstable equilibrium, and without entering into a discussion of
the relative strength of electric charges of the molecules, or discussing whether
we deal with adsorption phenomena or true chemical combination, we can, I
think, compromise by agreeing that the conditions in which enzyme action
proceeds are those which favour the occurrence of loose molecular combina-
tions.

ENZYME ACTION AND ADAPTATION 89

continues, so gradually will there be developed a tendency for
the proportion of the radicles constituting the biophoric mole-
cules to vary, those deficient in number being replaced by those
that are abundant.

Or again, the simpler complexes due to the breaking down
of the foreign protein may not be identical in constitution with
those which prior to the exhibition of the foreign protein had
been attracted to and had constituted certain particular side-
chains of the biophoric molecules. In this way again the con-
stitution of the biophores may become altered. This we may
visualize as follows (Fig. 9) :

•> SiAeCUin . N C

I I

>

c

j|ew5ideeUiv ' |!

N

Fig. 10. — Pyrimidine ring.

Fig. 9.

I would recall how a slight difference in the relative attachments
of the side-chains, say, of the Pyrimidine ring (Fig. 10) alters the
chemical properties of the compound : x in like manner what
may appear to be but a slight modification in the constitution of
a biophoric side-chain may be expected to lead to material
differences in the reactions of the whole molecules, that is, in
the case of germinal biophores, of the individual.

1 I came to these views honestly through a respectable collateral heredity.
My mother's brother, the late Dr. D. J. Leech, of Manchester, as the subject of
his Croonian Lectures at the Royal College of Physicians in 1893 discussed the
pharmacological action of the nitrates, comparing and contrasting the action

of the nitrates, - 0 - N= 0, the nitrates, - 0 - N^^, the nitro compounds,

- N^~, the nitrosamines, =N=N - 0, and the oximes, =N - 0 - H. He was

one of the first English pharmacologists to dwell upon the modification in the
action of drugs brought about by slight differences in the order of attachment
or composition of radicles. (See the British Medical Journal, 1903, and his
collected papers upon " The Pharmacological Action and Therapeutic Uses
of the Nitrates," edited by Dr. R. B. Wild, Manchester, 1902.)

90 ADAPTATION AND DISEASE

(v.) The Mechanism of Immunity. — One more postulate
needs to be laid down. You will remember how I pointed out
in my second lecture that the introduction of toxines, whether
phytotoxins like abrin, or bacterial toxins like those of diphtheria
and tetanus, lead to the abundant if not over-production of
antitoxins on the part of the cells.

The simplest explanation of this continued production is
that when the toxin (which has all the earmarks x of an enzyme)
enters the cell, it combines with and detaches certain side-chains.
Whether the cell in consequence becomes destroyed or, on the
other hand, succeeds in neutralizing the toxin, depends upon
two factors — the number of toxin molecules that gain entrance,
and the rate at which the cell can reproduce the lost side-chains.
Just as a damaged crystal if placed in a saturated solution of
the particular salt repairs itself, so we have to assume the capacity
on the part of the living molecules to build up lost parts. So
long, that is, as the toxin in the cytoplasm acts as a catalyst
detaching the particular side-chains, for so long will the molecular
complexes build up side-chains until from the law of habit this
building up exceeds the immediate need of the cell, and the
side-chains being in excess are excreted to become the antitoxins
of the blood serum. We must, that is, regard these side-chains
as now built up in series from the simpler cytoplasmic molecules

of the cell-sap, and when
so built up easily liberated.
Here parenthetically it de-
serves note that the bodies
which are antitoxins for man
are toxins for the bacteria
— all, in fact, depends upon
the point of view.

That this conception of
the mechanism of immunity and progressive adaptation 2 is
substantially correct has, I urge, been shown by the later
admirable studies of Professor Victor C. Vaughan of Ann
Arbor, Michigan, and Abderhalden of Berlin. Vaughan was

1 Corrected from " properties " in the lecture as delivered, this being the
more accurate statement.

2 A conception laid down by me in 1908 in my Principles of Pathology,
though there following naturally upon my article upon " Inheritance " published
in 1901.

Fig. 11.

THE MECHANISM OF IMMUNITY 91

the first to show that every true protein when digested
with several times its bulk of absolute alcohol in the pre-
sence of 2 per cent of sodium hydroxide, splits up into a poison-
ous and a non-poisonous moiety, the former soluble in alcohol,
the latter insoluble. The poisonous moiety, when introduced
into the system by any path other than through the alimentary
canal, is rapidly fatal in relatively minute doses, setting up
symptoms identical with those seen in the anaphylactic stage.
I have pointed out (p. 38) that anaphylaxis, or hypersuscepti-
bility, is the first phase in the production of immunity. On the
principle, therefore, that like results are produced by like sub-
stances we arrive at the conclusion that this hypersensitiveness
or anaphylaxis is essentially due to the disintegration of the intro-
duced proteins with rapid liberation of the poisonous moiety.

But the first dose of a foreign protein introduced into the
tissues sets up none of the phenomena, the second does.
Evidently in the course of the few days necessary for anaphylactic
phenomena to manifest themselves the organism has gained a
new power, that of splitting up the foreign protein. Abderhalden
demonstrated the first stages in this process. In 1910 he and
his pupils published a remarkable series of " Serological studies
by means of the optical method." Starting from the fact that
the different proteins and peptones, when in suspension, present
each in the polariscope a specific index of rotation, he demon-
strated first that enteral digestion of proteins, i.e. the introduc-
tion of the same through the intestinal tract, has absolutely no
effect upon the physical condition of the blood serum : its
rotatory power is unaltered. But parenteral injection of any of
the more common proteins, i.e. by the subcutaneous or intra-
venous route, leads to a very different state of affairs : the index
of rotation of the removed blood plasma undergoes progressive
alteration, a clear indication that the protein is being broken
down, and this is confirmed by the fact that by dialysis peptones
can be separated off from the plasma.

Within forty-eight hours, therefore, after the introduction
of a foreign protein, the system has responded either by elaborat-
ing a proteoclastic enzyme, or, it may be, by a rapid increase
in the production and discharge of already existing proteoclastic
enzyme present in certain cells. But what I would particularly
emphasize is that in this first phase the enzyme is not specific.

92 ADAPTATION AND DISEASE

It is a general or indifferent proteoclastic ferment. The blood
plasma at this stage will act not only on the particular protein
inoculated, but also on other proteins, will break down indiffer-
ently bodies such as globulin, casein, gliadin, gelatin, and their
peptones with the production of slightly toxic bodies.

It might be laid down that this indifferent ferment first pro-
duced is the cause of the anaphylactic phenomena. Against
this view is the fact that Abderhalden's ferment is recognizable
in twenty-four to forty-eight hours, whereas anaphylactic
phenomena are not obtainable until the sixth to the tenth day,
and again, the anaphylactic phenomena are specific : it is only
the protein first introduced that on subsequent injection sets
up disturbance. Abderhalden's proteolytic and peptolytic
enzymes are active in respect to any protein. We have evidence,
therefore, of the development of two stages : (1) elaboration and
discharge into the blood within forty-eight hours of indifferent
proteoclastic and peptolytic enzymes ; (2) the gradual produc-
tion within six or eight days of more high-developed specific
enzymes having the power of rapid dissociation of the particular
protein introduced, with liberation in the course of a few minutes
of sufficient toxic disintegration products to set up grave
symptoms. The rapidity with which the phenomena may show
themselves, and that upon the introduction of extraordinarily
small second doses of any particular protein, suggests that there
is a selective taking up of the protein by particular cells, in the
first place, with production and storage of the specific enzyme,
so that the moment the second dose is absorbed by these cells
there is an explosive disintegration and direct action of the
poisonous disintegration product.

But now by cautious reinjection of the foreign protein we
can pass through the anaphylactic to the succeeding stage of
immunity. Having accomplished this the introduction of the
foreign protein no longer leads to acute poisoning. Any one
who has observed Pfeiffer's phenomenon — who has seen living
typhoid bacilli swell up and dissolve " like pieces of sugar '
in the blood serum or peritoneal fluid of a highly immunized
guinea-pig — cannot but be impressed by the extraordinary and
specific digestive powers that have been acquired by the body
fluids through this process of immunization. We see here a
further evolution of the same process — a third stage — namely,

ANAPHYLAXIS AND IMMUNITY 93

the eventual development by the cells of the organism of further
enzymes, which disintegrate the poisonous moiety of the protein
disintegration products, and so carry on the degradation of the
foreign protein to a harmless stage ; which break up the associa-
tion of radicles still further, so that now they are converted into
more elementary constituents which, instead of disturbing the
cell activities, can, perchance, be utilized as food-stuffs. As
my colleague Dr. Fraser Gurd 1 has shown, in the immunized
animal the earlier enzymes are not replaced, they are still active
and abundant, but evidentlv after thev have broken down the
specific protein, with the liberation of the poisonous substance,
this is immediately acted upon by the new specific enzymes and
rendered innocuous.2

(vi.) Cell and Tissue Differentiation : Ontogeny. — With this
clear indication, therefore, before us of the way in which the
cell substance can become modified and adapt itself according
to need, it is not necessary, with Weismann, to postulate the
existence of two orders of living matter, germ plasm and somato-
plasm. It suffices to hold that there is one common biophoric
material, or at most biophoric material in part derived from one,
in part from the other parent, and to recognize that as the fertil-
ized cell divides and redivides and the resultant cells find them-
selves in different relationships, so do they become modified
and differentiated, giving origin to the various tissues, the
eventual germ cells being so developed and so placed that they
are exposed to the minimal amount of differentiation.

If embryogeny or, more exactly, ontogeny — the development
of the individual — is a recapitulation of the phylogeny — of the
evolution of the race — it is this not as an historical reminiscence
and nothing more. It is this only so far as it is a necessity,
only so far as, for the unfolding and elaboration of the cell
structure of the various tissues up to the stage in which the
adult of a particular species will find itself in equilibrium with
its surroundings, the biophores distributed to those cells must,
of necessity, in the course of their growth and multiplication,
pass through a particular succession of equilibrations, of modifica-
tions, of accretions and losses. Where a given modification can

1 Amer. Jour. Trop. Dis. and Prevent. Med. i., 1914, 776, and Jour. Med.
Res. xxxi., 1914, 205.

2 For a fuller study of these matters see the paper upon " Parenteral Diges-
tion " in Pt. II. Chap. IX.

94

ADAPTATION AND DISEASE

be attained by a short cut, there certain phases of the phylogeny
fail to be recapitulated. In short, the development of any
particular tissue may aptly be compared with the synthesis of
one of the higher organic compounds. That cannot be accom-
plished by taking so much Carbon, Hydrogen, Nitrogen, etc.,
mixing and heating them together, but demands a long process
of building up — it may be a score or more successive processes
of combinations, associations and dissociations — before the final
product is attained ; the constant endeavour of the chemist
being to simplify the process, to arrive at the same results by
finding some reaction which will cut out half a dozen or more
of the steps and lead more rapidly and more economically to
the same result.

(vii.) Amphimixis. — This mention of the contribution of the
two parental germ plasms to the fertilized germ cell, or zygote,

Fig. 12. — A, biophore of paternal, B, biophore of maternal origin ; 1, 2, and 3,
various allelomorphic side-chains ; 4, side-chains common to both, orders of
biophores.

leads to certain important considerations. If the metabolic
activities of the cell are of the order here stated, and if, as is
evident from the studies of cytologists and embryologists,
equivalent amounts of heritable material — or biophores from
the two parents — reach the zygote and, in the subsequent process
of active cell division of the embryo, are contributed to each

BIOPHORIC INTERACTION 95

cell, i.e. to each cell nucleus, again the static conception cannot
be permissible. We must realize that these two orders of bio-
phores, existing in each cell of the body and activating each
cell, are throughout life undergoing growth, building up side-
chains and radicles, and that there is a constant recurrence of
processes of association and discharge from each molecule out
of and into the nuclear sap. But if this is the case and the two
orders of active living molecules He side by side in the fluid
medium of the nucleus, it is impossible to imagine, according
to the current Mendelian doctrine, that the two do not interact
one upon the other, that the two orders of side-chain substances
and dissociation products keep severely apart. We may safely
say that there must be interaction : that where tissue cells are
undergoing active growth and metabolism and radicles and
groups of radicles built up by biophores of the one order become
dissociated from those biophores, if they are of an order common
to both paternal and maternal biophores, they may be employed
indifferently in the building up and growth of biophores of both
orders ; if of an order peculiar to, say, the paternal biophores,
may under certain conditions be attracted to and built into
the growing maternal biophores. We can imagine environ-
mental conditions favouring the predominant growth of just
the one group of biophores so that the cells of some one tissue
take on the characters of the cells in the one parent, or, more
easily, in actively working tissue-cells, such an interchange and
selection of allelomorphic radicles that eventually one common
type of modified biophoric molecule may govern the cell (Fig. 12).
And at a slower rate due to their more latent state, this same
interchange, we must predicate, is capable of occurring in the
biophores of the germ cells, so that eventually a succession of
heterozygous generations will give rise to forms without the
pure dominant gross characteristics, but showing an approxima-
tion to intermediate properties ; 1 or, again, eventually, the inter-
change of radicles may be such that the various dominant radicles
belonging to both orders of biophores be attracted to, built into
the growing molecules, and reproduced by the biophores derived
from the one parent — whereas the biophores of the other origin in

1 As in Correns's observations upon the hybridization of the two strains of
Mirabilia jalapa, alba and rosea, in which the heterozygotes of the F. genera-
tions are not of strong'dominant rose colour but of a paler dilute pink.

96 ADAPTATION AND DISEASE

their growth would show a greater affinity for the radicles whose
constitution lead to the appearance of recessive qualities.

The existence of determinants as entities, as I have demon-
strated, cannot on physical grounds be accepted, and if in their
place we accept differences in the constitution or arrangement
of individual radicles composing essential portions of the com-
bined and complex molecules of living matter, it is still possible
to interpret the facts of Mendelism and indeed interpret not a
few phenomena which by the hypothesis of determinants cannot
be explained.1

Summary

However, I realize that here I advance into regions outside
the boundaries of current medical thought. To sum up, it may
be said that, according to this hypothesis, each species must be
regarded as having for its essential living matter a distinct
organic compound, a compound as distinct as any inorganic salt,
but differing from that simpler inorganic salt in that, whereas
the central ring or chain is of relatively fixed constitution, the
radicles composing that ring or chain are capable of attracting
and then of reproducing a series of side-chains which may vary,
so that within the species there may be various strains, just as
we may speak of various strains of crystalline haemoglobin being
obtained from different samples of human blood.

Let me freely admit that the diagrams here afforded are
very crude. With the necessarily condensed presentation of
my subject they have seemed to me essential. To those un-
familiar with the advanced organic chemistry of to-day, to have
attempted illustrations of the successive stages of my argument
by elaborate graphic chemical formulae would have been worse
than useless. I shall be satisfied if I have rendered it clear that
it is possible to replace an impossible hypothesis based upon sup-
posititious independent and transposable determinants by one
based upon what we know of the composition and physical

1 Of set purpose, as being outside the scope of my earlier work and of medical
research, I did not in the course of these lectures refer to the newer studies of
the last few years upon differentiation of the individual chromosomes. Such
differentiation with localization of groups of unit characters in particular
chromosomes is, I would point out, distinct from Weismann's determinants.
I hope to take up this matter later.

THE SPENCERIAN TRAGEDY 97

structure of the main and outstanding constituent of living
matter — the proteins — a hypothesis that renders it possible for
us to understand how digestion or, more broadly, metabolism
is the keynote of the whole matter ; how through this universal
svstem of dissociation of food-stuffs into their elemental radicles
it is possible for the cell to utilize and accustom itself to new
and foreign food-stuffs ; how doing this, the constitution of
the living molecules may become altered ; and how Professor
Bateson's incapacity to comprehend the possibility of progressive
acquirements by the individual and his germ cells is due to his
static method of approach to the subject. To one who regards
life, not from the morphological point of view, in terms of form,
but from the physiological, in terms of function — who regards
life as a moving equilibrium, who regards it as in essence " a
state of persistent and incomplete recurrent satisfaction and
dissatisfaction of . . . certain proteidogenous molecules," 3 and
metabolism as the primary and basal characteristic of living
matter — for such an one there exists no such stumbling-block.

Progressive Accretion of Properties

Just as Weismann's hypothesis of heredity breaks down owing
to the sheer physical impossibility of all the determinants it
demands being packed into the microscopic nuclear chromosome,
so the Batesonian hypothesis of a backward evolution by the
progressive removal of inhibitory factors, like the baseless
fabric of a vision, fades into nothingness once it is confronted
with the demonstration that positive, direct acquirements can
surely be brought about. Both hypotheses, indeed, enter into
the limbo of the past, as examples of the Spencerian tragedy —
that of a deduction destroyed by a fact.

I imagine that it will be your experience — it certainly has
been mine — that it is those whose gentle birth is least obvious,
and genealogy most dubious, those who hang on to the border-
line of good family, who most vaunt themselves regarding their
descent. They do, net realize that in these days it is a man's
ascent that ennobles hir2, not his descent, even if, paradoxically,
ascent is not possible without good rich blood : it is the pro-
gressive accretion of properties, not the progressive loss. But,

1 Adami, Principles of pathology, 1st ed., 1908, vol. i. p. 55.

H

98 ADAPTATION AND DISEASE

says Professor Bateson, in nature it is the opposite : it is by the
progressive loss of properties that from being an amorphous mass
of protoplasm man has become man. Life, according to him,
first appeared upon earth enshrined in matter of maximum
complexity, so complex that it was without form and void, and
only as through the vast aeons of geological time this matter
fell into order and simplified itself, and successive species
developed, did we eventually in these latter days arrive at the
simplest and least complex of all creatures — man, simple man.

Does it not appear to you that this is Topsy-turvydom ?
Is it not on the face of it more probable that the reverse has
been the case, that the earliest matter that could be recognized
as living was the simplest ? It is not that the earliest living matter
possessed all the determinants of all the organized parts of all
future forms of life, but that its constitution was such that it
possessed from that constitution, and in consequence of its
metabolic activities, the potentiality to undergo progressive
modifications of that constitution, which modifications mani-
fested themselves, as an outcome, in progressive changes of
structure. The potentiality was there, not the determinants.

Conclusion

One last word. I feel that as one on active service some
apology may be required from me for having taken your time,
and, it may be said, my country's time, in dealing in the course
of these lectures with a matter so wholly foreign to the war,
to military medicine and military duties. Were what I have
placed before you wholly new, had I collected, thought out,
and elaborated the material of these lectures during the two
years since I received the invitation to deliver the course, an
apology would, I think, rightly be in place. As a matter of fact
these lectures are little beyond what I have taught and written
in the fifteen years and more preceding the war : they are a digest
and compend of those earlier writings and conclusions, brought
up to date by means of modern instances confirmatory of
or expanding that earlier work. Four hours before delivering
the last of these lectures, in order to complete my references,
I went to the Library of the Royal Society of Medicine in Wimpole
Street and took out the volume of the British Medical Journal for

CONCLUSION 99

1901, containing an address delivered by me at Brooklyn, New
York, in May of that year.1 That address I had not looked up
for a decade or so, and it was with not a little surprise that I
found laid down there the physico-chemical conception of in-
heritance here given, and the doctrine of direct inheritance of
metabolic conditions, such as gout and of disturbances of the
internal secretions. Rather, therefore, the apology should be
that I have plagiarized myself in so wholesale a manner. I shall,
however, be satisfied if it is demonstrated that the work of
medical men of this generation, of pathologists and bacteriologists
— work formed upon the observations and methods of the great
biologists of the past — is repaying the debt to biology by estab-
lishing principles which are basal for general biological advance.

1 See Pfc. II. Chap. II.

PART II

ON VARIABILITY AND ADAPTATION

101

CHAPTER I

ON THE VARIABILITY OF BACTERIA AND THE DEVELOPMENT

OP RACES 1

(1892)

From a medical as from a biological point of view, the variability
of micro-organisms is a subject of the highest interest. If it
can be clearly demonstrated that modifications of varying degrees
of permanency can be induced in sundry species of bacteria,
that " new races " are capable of being developed under the
influence of conditions that are easily recognizable and easily
controlled, then not only do we gain further knowledge of the
laws governing evolution, but — what for us is of more immediate
import — we become enabled to clear up not a few of the difficulties
constantly presenting themselves in the study of zymotic, or
mvcotic, disease. The class of difficulties to which I refer is
so familiar that here I need give but the briefest indication
thereof. It is notorious, for instance, that successive epidemics
of one disease, say, scarlatina or influenza, are characterized
by well-marked differences in symptomatology, and in the
intensity of effects upon the individual ; it is notorious also
that during the progress of one epidemic remarkable changes
are to be made out in the virulence of the malady. Often at
the onset the cases are so mild, the symptoms so vague, that
diagnosis can only be pronounced with exceeding caution.
Such doubtful early cases may be followed by a succession in
which the disease is typical and of increasing virulence, and
eventually is reached the period of decline of the epidemic, and
now again the cases assume a dubious aspect with indefinite
atypical symptoms. Or take a disease which now unfortunately

1 Reprinted from the Medical Chronicle, September 1892.

103

104 ON VARIABILITY AND ADAPTATION

has become more or less endemic, namely, diphtheria, and
observe the series of progressive cases that is to be made out
from the simple sore throat associated with the mildest con-
stitutional effects, through slight cases of croup to those terribly
malignant attacks in which the manifestation of the disease
and death are almost synchronous.

If the micro-organisms of disease have constant attributes,
if the species of bacteria are fixed, their characters unvarying,
then all these differences in the nature and intensity of the
symptoms produced by any one micro-organism at diverse
periods and in diverse individuals can only be due to subtle
meteorological changes, to differences in the power of resistance
to microbic invasions displayed by the individual, and to the
dose, if I may so term it, of infective material incepted by the
individual. Now, without doubt, all these factors are of very
great importance, but at the same time it cannot be denied that
they do but vaguely advance our knowledge. Up to the present
the study of atmospheric influences as affecting zymotic diseases,
while impressing upon us one or two main truths, has, in other
respects, involved us in a maze of contradictions, and while the
last few years have greatly extended our knowledge of the
nature of the conflict between the organism and the microbe,
clinical experience is constantly bringing before us the para-
doxical phenomenon of disease, in its severest form, attacking
individuals in whom one would hold the constitutional powers
of resistance to be at their highest. So, too, difficulties in the
way of exact determination are insuperable, clinically, with
reference to the " dosage " of infective material.

But if, while granting freely that all the above-mentioned
factors play each a part, it can be shown that modifications of
the powers of micro-organisms can be induced, both within the
organism and outside it — alterations in appearance, in the
ferments and other products of their vital activity and (among
pathogenic microbes) in virulence ; alterations, not temporary,
but continuing through generations ; alterations that are capable
of being demonstrated in the test-tube and under the micro-
scope— then we reach one stage further in actual advance in
our knowledge of infectious disease, and from this advance gain
a firmer standpoint from which to appreciate the other factors.

Largely influenced by the admirable researches of the German

FIXITY OF BACTERIAL SPECIES 105

bacteriologists, and in this country by the teaching of Dr. Klein,
the tendency of the medical world has been to regard the species
of micro-organisms, and especially of pathogenic micro-organisms,
as fixed. If in each case of a given disease a given microbe is
discovered with specific cultural characteristics : if the more
closely zymotic disease is studied the more clear becomes the
evidence of its infectious nature, of its passage either directly or
indirectly from person to person, while the more the subject is
studied the weaker becomes the evidence of the spontaneous
origin of infectious disease : then the greater is the difficulty of
arriving at any other conclusion than that pathogenic microbes
belong to species that are fixed and possess unchanging attributes.
I need not say how valuable, not to say necessary, it has been
in a young science like bacteriology to hold this view of the
fixity of species, and, speaking broadly, this view must still
be held. In the great majority of cases the micro-organisms
gained from the tissues of the patient suffering from any one
zymotic disease approximate sufficiently closely to the type for
us to say with absolute confidence that no specific alteration is
to be discovered in them. But while acknowledging this much
it must be confessed that there has been a tendency to overlook
differences in the mode of growth of pathogenic microbes — gener-
ally, it is true, minute, though not always so — and to dwell
upon differences in the resistance of the organism more than
upon differences in the pathogenic properties of the micro-
organisms whenever, in an individual case, or in an epidemic,
the symptoms have varied from the ordinary. While bacterio-
logists in general admit, to a greater or less extent, that bacteria
show variations, I know of no work in which the facts in con-
nexion with such variations have been brought together ; it may,
therefore, be useful at the present time if I gather together
some of the many researches into the variability of microbes,
and show that this variability can clearly be made out.

Undoubtedly where a number of facts are brought together,
all pointing in one direction, there is a danger that by some
these facts will be held to support an extreme theory, and one
that cannot surely be gained from them — the theory, in this
case, that non-pathogenic bacteria can easily become pathogenic,
and that zymotic disease may, as a frequent event, arise de novo.
While, possibly, further research may prove that in a certain

106 ON VARIABILITY AND ADAPTATION

limited number of cases a harmless saprophyte is capable of
developing into a pathogenic microbe, at the present moment 1
the facts are wanting to support this hypothesis. The extreme
conclusion to be reached now is that there are species of patho-
genic micro-organisms whose virulence is capable of manifesting
great modification, and that such modification is fitted to explain
some of those differences in the course of zymotic disease which
must have arrested the attention of all medical men.

On ceetain Characteristics of the Bacteria as a Whole
affecting the liability to variation

At the very onset we should expect to find the bacteria
peculiarly well adapted for the study of the most elementary
problems of evolution : their organization, as compared with
other forms of fife, is so simple, the generations — if generations
they may be called 2 — succeed each other with such extraordinary
rapidity, and, what is of yet greater importance, our study of
these forms would seem to be wholly freed from the perturbing
influence of sex. On the one hand there can, in the bacteria,
be no tendency towards the production of races through sexual
agency ; on the other the reverse tendency is equally absent,
namely, the tendency of conjugation to preserve the mean, to
lead to the development of individuals approximating to the
type.

Here, then, in studying these unicellular asexual bacteria,
we investigate the simplest and most fundamental principles of
heredity and evolution, and the influence of environment be-
comes most evident. Weismann, who in the higher forms of
life would make the development of individual differences
and the inheritance of the same almost wholly dependent upon
the phenomena which accompany the fusion of the male and
female reproductive cells, freely acknowledges that to the

1 [i.e. in 1892.]

2 The word generation as applied to the Schizomyeetes is not a little un-
satisfactory— yet I know no other word at present in use that is capable of
replacing it. Apart from the sexual significance that may attach to it, this
word embraces the idea of parentage. Now, among the bacteria there is
no true parentage, for what happens in the process of fission is that the bacillus
divides into two portions, each of which becomes a separate individual, but
neither can be regarded as the parent form ; each equally with the other
represents, and, to a certain extent, is, the earlier single individual.

THE CHARACTERISTICS OF BACTERIA 107

protozoa we must look for the origin of " hereditary individual
variability," and, dealing with them, emphasizes this influence of
environment. If, for instance (he states), a protozoon, by con-
stantly struggling against the mechanical influence of currents in
water, were to gain a somewhat denser and more resistant proto-
plasm, or were to acquire the power of adhering more strongly
than other individuals of its species, the peculiarity in question
would be directly continued on into its two descendants, for the
latter are at first nothing more than the two halves of the
former. It therefore follows that every modification which
appears in the course of its fife, every individual character, how-
ever it may have arisen, must necessarily be directly transmitted
to the two offspring of a unicellular organism.1

And in connexion with the bacteria we can modify the
environment at will to an extent that is far beyond our power
in regard to higher organisms ; we can alter the media of growth,
can make these more or less nutritious, can make them acid,
neutral, or alkaline ; can alter the superincumbent air to such
an extent that if desired we can in many cases bring about
growth in vacuo or in an inert atmosphere of hydrogen or carbonic
acid ; we can add or remove this or that normal or abnormal
food-stuff to or from the medium of growth, or can add salts
and antiseptics, and observe the effects upon the microbes
studied ; we can observe the mutual effects of concurrent growth
of two or more forms, and the effects which the product of
growth of a microbe exert upon it, or again can modify the
temperature within the widest limits. With no other forms of
life are such extraordinarily diverse modifications of environ-
ment possible.

I propose now to set forth in regular order the facts at our
disposal with regard to the production and existence of varia-
tions of microbes. To give all is, from their very extent, im-
possible in this connexion, but sufficient may be given to show
very clearly that this variation is far more widespread than is
generally supposed, and that we can safely attribute to this
variability many of the difficulties already mentioned met with
in the clinical study of infective disease.

It is unnecessary for me to enter into the earlier history of
the subject, into the controversy that raged around " poly-

1 Essays upon Heredity, Oxford, 1889, p. 278.

108 ON VARIABILITY AND ADAPTATION

morphism " and round Nageli's theory as to variability,1 for at
the time that these were most discussed the subject was not
ripe for treatment. The methods of investigation were imperfect,
pure growths could not with certainty be obtained, and con-
sequently the arguments advanced pro and contra rested upon
insecure bases. Still Lankester's observations 2 on the change
of form of the Clathrocystis, or Beggiatoa, roseo-persicina,
and later Kurth's " Observations on the Bacillus Zopfii," 3 have
stood the test of time, and may be said to have laid a solid
foundation upon which rests our present knowledge of variation
among microbes.

Transient Variability

We possess now such a mass of separate observations upon
transient variability of bacteria that it is difficult to know how
much to say on this subject here, how much to leave unsaid.
Every one who of late years has studied bacteriology must have
come across numerous cases of this nature. Every one admits
that according to the age of a growth, according to the nature
of the medium of growth, changes in form and properties are
common. They may be slight, but if looked for are almost sure
to be found. I would proceed to mention cases in which such
changes are very marked.

(1) Change of Form. — The Bacillus Zopfii,3 for example, when
grown on various nutrient media manifests itself at first in the
form of long twisting tortuous filaments ; at a later period these
give place to chains of bacilli of fair size, and eventually in an
old growth the bacilli are replaced by what some would consider
as cocci, others as spores. We need not now concern ourselves
about this controversy. The important point is that, trans-
ferred to a fresh " soil," the bacillary form, just like the spore or
coccus form, develops into typical long convoluted threads.

Even more definite changes of shape are producible by slight

1 Nageli, Die niederen Pike, Munich, 1877.

2 Lankester, Quart. Journ. Micr. Science, xiii., 1873, p. 408. [It may in-
terest Sir Ray Lankester to observe that I have been familiar with the observa-
tions upon his peach-coloured bacterium for a quarter of a century and more,
and have given him due credit. The reason why I did not refer to these in my
Croonian Lectures is obvious : he has consistently ascribed the modifications
in form to indirect and not to direct equilibration.]

3 Kurth, Botanische Zeitung, 1883.

TKANSIENT MODIFICATIONS 109

alterations in the medium of growth. Take, for instance, the
Eberth-Gaffky bacillus of enteric fever. This is a bacillus of
fair size — three to five times as long as it is broad, with rounded
ends, and, thanks to the possession of long vibratile cilia along
its sides, it is normally endowed with great rapidity of motion.
It has, further, the property of resisting the action of fairly large
amounts of carbolic acid in the medium of growth — a property
which has been utilized by Vincent to separate it out from other
microbes, and to gain pure diagnostic cultures from the faeces
of those affected by the disease, or from contaminated water.
If beef-broth be taken containing about 1 part of carbolic acid
in 600, the Eberth - Gaff ky bacillus, almost alone among the
bacteria, will grow in it at a temperature of 42° C, rendering
the broth turbid in the course of a few hours. But if a drop of
this be examined, one finds not the characteristic motile bacillus,
but shorter forms and often diplococci, and these are non-motile.
In fact, there is an absolute want of resemblance to the type
bacillus. Yet if a culture of these be made in the ordinary beef-
broth of the laboratory, devoid of carbolic acid, there is an
almost immediate return to type.1 But there is a much simpler
method of gaining polymorphism in connexion with this microbe.
Employing a series of potatoes of diverse origin, it is found that
upon some the growths of the microbe are atypical not only in
naked-eye appearance but also under the microscope, developing
into long filaments. This atypical growth has been commented
upon by numerous observers, including Frankel and Simmonds,2
Buchner,3 Schiller,4 and Kamen.5 Buchner's observations
render it probable that this atypical growth is in close relation-
ship to the known varying acidity of the potato.

Still more striking are the results obtained by Charrin6 in the
case of the small bacillus of blue pus, which within the body
of an infected animal, or grown in beef-broth, is what may be
termed a microbacillus — a minute short bacillus, the individuals
being often attached in pairs end to end.

If /3-naphthol be added to the beef-broth in the proportion

1 Vincent, Comptes rendus de la Soc. de Biologie, 1890, No. 5.

2 Frankel and Simmonds, Zeitschrift f. Hygiene, ii., 1887, p. 138

3 Buchner, Centralblatt f. Bakteriologie, iv., 1888, p. 353.

4 Schiller, Arbeiten aus dem kais. Gesundheitsamt, v.

5 Vide Petruschky, Centralblatt f. Bakteriologie, vi., 1889, p. 657

6 Charrin, La Maladie pyocyanique, Paris, Steinheil, 1889.

110 ON VARIABILITY AND ADAPTATION

of 0-02 per cent, after forty -eight hours long straight bacilli
are to be made out, broader than the type, and from three to
five times as long, If, instead, alcohol be added (about 4 per
cent), at the end of twenty-four hours the change is even more
pronounced ; besides large bacillary forms, similar to those
last mentioned, there may be somewhat curved filaments from
ten to fifteen times as long as the type, if not longer. If O015
per cent potassium bichromate be added, then, even after fifteen
hours, one can make out long undulating filaments stretching
across the whole field of the microscope. And if the growth
be acted upon by 0*7 per cent boric acid for six days, then in
place of the microbacillus we see " comma " forms, S-shaped
forms, and close spirals ; while 0-1 per cent creasote acting for
some weeks causes the production of micrococci and diplococci.

There could be no better example than this of the poly-
morphism induced by environment. In all these cases, however,
we are not dealing with permanent variations. Return to
" ordinary " media brings about in all a return to what may
be styled " type." I say what may be styled " type," inas-
much as the broth, potatoes, and other materials employed by
the bacteriologist are the common soils in and upon which can
be grown most bacteria, but they are not of necessity their
natural habitat, and when environment evidently plays so im-
portant a part in determining the shape of bacteria it is quite
possible that what we consider as typical are, in many cases,
not the usual (or primitive) but are acquired forms.

(2) Changes in Pigment Production. — But shape is far from
being the only property of microbes that is modified by environ-
ment, and very striking results are to be gained by a study of
those microbes which, in the course of their growth, produce
pigment — the chromogenic bacteria.1 Of these a very large
number lose the faculty of pigment formation in the absence of
oxygen, many again when grown in the absence of diffuse fight,
and all under the action of the direct rays of the summer sun.
An alteration in the medium of culture has also its effect. Thus
the Bacillus ruber of Kiel (better known in this country as the
" rouge de Kiel," the name under which our attention was first
drawn to it by Laurent 2), the Micrococcus indicus, the Micro-

1 Schottelius, Kolliher's Festschrift, Leipzig, 1887.

2 Laurent, Annates de VInst. Pasteur, iv., 1890, p. 465.

TRANSIENT MODIFICATIONS 111

bacillus prodigiosus, and yet other forms which produce a red
colouring matter vary in the intensity of their colour production
according to the acidity or alkalinity of the medium of growth.
The most beautiful example of temporary modifications of
easy production by alteration of environment is supplied
in Gessard's very full study of the Bacillus pyocyaneus.1
Ordinary beef-broth sown with this bacillus soon takes on a
greenish-blue slightly fluorescent appearance, and the blue colour
becomes more marked if there be a relatively large surface
exposed to the air, or if at intervals the culture be shaken, and
the pellicle of surface growth be made to sink to the bottom of
the vessel. Evidently then there are at least two pigments
produced, and Gessard succeeded in causing the independent
development of either. For he found that upon solidified white
of egg the bacillus gives rise to the bright green colour alone,
while upon the agar-agar, to which glycerine and a fair quantity
of peptone had been added, no green is developed, the medium
rapidly becoming tinged throughout an intense deep blue. It
may be added that the absence of free oxygen causes the develop-
ment of a colourless growth. These observations, I repeat,
can be easily confirmed.2

(3) Modifications of Ferment Production. — Similarly modifica-
tions can be induced in the ferment production of microbes.
Thus many micro-organisms, which elaborate a ferment capable
of liquefying gelatine, do not manifest that power if glycerine
be added to the medium of growth, or (from a different cause)
lose that power, as Cartwright Wood has shown 3 in the case of
the Cholera spirillum, of the Bacillus prodigiosus and the Micro-
coccus indicus, if minute quantities of carbolic acid be added to
the culture medium. Again, as indicated by Lauder Brunton
and Macfadyean,4 the bacteria which form a peptonizing
enzyme on proteid soil can also produce a diastatic enzyme on
carbohydrate soil.

(4) Modifications of Pathogenicity. — With regard to the

1 Gessard, Annates de VInst. Pasteur, iv., 1891, p. 737; and v., 1891, p. 65.

2 Adami, Meeting of East Anglian Branches of British Medical Assoc.
Cambridge, Lancet, July 24, 1891.

3 CartAvright Wood, Proc. Royal Soc. Edin. xvi., 1889 ; and Laboratory
Report, R.G.P. Edin. ii., 1890, p. 253.

4 Lauder Brunton and Macfadyean, Proceedings Royal Society, xlvi., 1890
p. 542.

112 ON VARIABILITY AND ADAPTATION

temporary loss of pathogenic power — of virulence — the case is
rather different. It would seem to be most difficult to bring
about a loss of virulence that does not tend to be permanent.
That is to say, very special methods have to be employed whose
action is exerted upon more than one " generation " of the
microbes, such as passage through a series of animals less and
less susceptible to the invasion of the given microbe in order to
exalt organisms, like the pneumococcus of Talamon-Frankel,
whose virulence can only be retained outside the body by constant
removal to fresh media ; others, like the typhoid bacillus, which,
according to Chantemesse and Widal,1 are so susceptible that
a loss of virulence is almost immediate. This, after all, is only
what is to be expected if, as seems most probable, the pathogenic
property is that which has been latest acquired. A late acquisi-
tion is most easily altered, and the tendency to reacquirement
of the same is least evident. Intensification of pathogenic
properties can, however, be induced by other means than by
passage through animals. Hueppe's observations would seem
to show that the Bacillus cholerae asiaticae grown anaerobic-
ally gains in virulence.

This principle — that it is difficult to stamp a modification
firmly upon a species of bacterium, as, indeed, upon any animal
or vegetable species — is exemplified throughout the long series
of attempts that have been made to produce new races of micro-
organisms, and now, when I proceed to recount some of the more
successful of these attempts, the principle will be seen to be
constantly in evidence.

Modifications of Longer Duration

From the transient modifications which I have noted above
it is easy to pass to a series of instances in which the return to
the normal only follows after an increasing number of genera-
tions, in which numerous generations retain the acquired char-
acteristics, and the original properties only gradually reassert
themselves. Take, for example, almost any of the chromo-
genic bacteria. If a minute quantity of a high-coloured typical
growth be removed and spread over the cut surface of a sterilized
potato it will be found that the individual colonies which result

1 Chantemesse et Widal, Annates de VInst. Pasteur, ii., 1888, p. 54.

MODIFICATIONS OF LONGER DURATION 113

present differences in pigment production ; the majority re-
produce the type, but some are even deeper in colour than the
original, others paler, others colourless. If from one of these
colourless colonies fresh potato growths be made, a large number
of coloured colonies develop, but also a large number of colour-
less, and the more frequently this selective process is repeated
the greater becomes the proportion of the resulting colourless
colonies. Yet even after a very great number of growths made
in this way there is always a tendency for certain individuals
and their progeny to revert to type. Slight alterations in the
nutrition of the individuals probably induced the first differences
in pigment production, and the induced modifications tend to
reproduce themselves through many " generations." x

The same occurs if a culture of one of these non-pathogenic
forms be heated for a few minutes to a point approaching that
at which immediate death of the special microbe is brought
about. After this treatment there may be numerous " genera-
tions " produced, incapable of pigment production. Illustrating
this and the preceding order of appearances, I showed a series
of specimens at the Nottingham meeting of the British Medical
Association in July (1892). The series included cultures of the
Bacillus ruber, Plymouth ; Bacillus ruber, Kiel ; Microbacillus
prodigiosus, Bacillus indicus, and Sarcina erythromyxa. To yet
more marked effects of high temperature I shall revert at a
later period.

Again, old cultures of most micro-organisms become
attenuated, and although these attenuated forms still propagate
themselves, it may be weeks — that is to say, almost countless
" generations " — before there is a return to primitive appearance
and primitive properties. Thus old growths of Koch's cholera
spirillum, or the Finkler- Prior spirillum of Cholera nostras,2
or, again, of the Vibrio metchnikovi 3 yield colonies which
have an absent or greatly reduced power of liquefying gelatine,
and only very slowly and under favourable conditions does this
power manifest itself once more. From old cultures of the
Bacillus pyocyaneus, the Pink torula, and many more chromo-
genic microbes, I find, in common with other observers, that

1 See more particularly Charrin and Gessard, loc. cit.

2 Firtsch, Archiv jiXr Hygiene, viii., 1888, p. 369.

3 R. Pfeiffer, Zeitschr. fur Hygiene, vii., 1889, p. 347.

114 ON VARIABILITY AND ADAPTATION

colourless or faintly-coloured growths are obtainable, which for
long show no reversion to type.

It would seem that, without there being necessarily any
attenuation, there may be modifications persisting for long
periods. It has been noted by many observers that for some
considerable period after Staphylococcus pyogenes aureus has
been separated out of pus and cultivated, the colonies are colour-
less, and that only eventually may the golden-yellow pigment
become produced. During the last long vacation I was a victim
to this order of affairs, for having obtained a micrococcus from
some pus gained from a case at the Addenbrooke Hospital, I
made two successive cultures of the pure growth during the
course of a week, and the second of these I employed to distribute
to the bacteriological class as the Staphylococcus pyogenes
albus. The cultures made by the class developed the golden-
yellow pigment, and proved that we were dealing, not with the
" albus," but with the " aureus." It may be added that, save
for colour difference, the two forms are indistinguishable.

The Pkoduction of Races

From such cases as these we can pass on almost imperceptibly
to the development of what may be termed races. I will presently
discuss the qualifications that must be applied in employing this
expression. For immediate purposes it is enough to state that
a slight modification of environment acting over a sufficient
number of " generations " — that is to say, acting for a sufficiently
long period — or a powerful stimulus applied temporarily to the
individual bacteria of one " generation," may lead to modifica-
tion of the properties of any species which, so far as we can see,
are permanent so long as the microbes so acted upon are cultivated
under ordinary conditions upon the usual media of bacterio-
logical research.

To make this statement perfectly clear, I may give an example.
The Bacillus ruber of Kiel grows well upon a piece of sterilized
potato at the ordinary temperature (15-25° C), the culture
becoming a deep crimson red. If now the culture be kept at a
temperature of 37° C. for two days, and a new piece of potato
be inoculated from this and kept another two days at the same
temperature, and a series of six to ten cultures be made thus ;

PRODUCTION OF BACTERIAL RACES 115

or if, on the other hand, a potato culture be exposed to a tempera-
ture of 55° to 57° C. for a few minutes, in either case fresh cultures
made in series upon potato at the ordinary temperature may for
months remain absolutely colourless.

It is scarcely necessary to say that the classical example of
this order of phenomenon is to be found in the remarkable observa-
tions of Pasteur, Chamberland, and Roux upon the attenuation
of the anthrax bacillus.1 The anthrax bacillus may be kept
(in broth) at a temperature of 42*5° to 43° C. for a week or more
without there being any noticeable diminution of the virulence
of the growth. But after this the virulence slowly and steadily
diminishes. Made under ordinary conditions from a growth
treated thus for twelve days cultures will no longer, when in-
oculated, cause the fatal disease in sheep. After thirty-one days
the cultures are so feeble that the very susceptible guinea-pigs
survive, and mice alone are killed. In forty days or so the
bacilli subjected to the above-mentioned temperature are en-
feebled to the point of death, and no further cultures from the
original broth are possible. What is more, each successive
growth obtained daily after the eighth day from a culture sub-
jected to the Pasteur process, if periodically resown, under
ordinary conditions (in alkaline beef-broth at 35° to 37° C),
retains permanently, for months and years, the grade of
diminished virulence that had been impressed upon the original
culture.

Several other methods have been suggested for attenuating
the anthrax bacillus, all of them possessing the advantage of
being more expeditious,2 but all, save one, labour under the
disadvantage that the lowering of virulence produced is unstable,
and upon continued cultivation the successive growths vary
and are liable to regain the pathogenic property of the original
culture. The exception is Roux's method,3 in which, by the
addition of minute quantities of potassium bichromate or carbolic
acid to the culture medium, a race is, after a short time, developed
which has lost the power of spore-formation. According to the
duration of the action of either antiseptic, and the amount of

1 Pasteur, Chamberland, and Roux, Comptes rendus, xcii., 1881.

2 See Toussaint, Comptes rendus, xci., 1880 ; Chauveau, Comptes rendus,
xcvi., 1883, p. 678 ; and Wossnessensky, Comptes rendus, xcviii., 1884,
p. 314.

3 Roux, Annates de VInst. Pasteur, iv., 1890, p. 1.

116 ON VARIABILITY AND ADAPTATION

either which is present, so are the daughter cultures of particular
degrees of virulence. These asporogenous races appear to retain
their characters indefinitely under normal bacteriological
conditions.

Here again it must be pointed out that " permanent " races
are by no means necessarily attenuated races. Starting from
a culture of the Bacillus anthracis, so weak that only young
mice are affected, it is possible, by passage of the virus through
a series of animals whose resistance to the disease is in an ascend-
ing scale, to gain a series of races of gradually increasing virulence,
and as Malm has shown 1 it is possible to pass beyond the
virulence of the type and obtain a race that will kill not only
sheep but the very refractory dog.2

Another example to the same effect may be gained from
Pasteur and Thullier's researches into swine erysipelas (Rouget
des pores, or Rothlauf).3 This fatal and most infectious
disease is caused by a minute bacillus first isolated by Loffler.
Rabbits also succumb to the malady, but not so very readily.
It is found that passage through these latter animals causes the
virus steadily to augment in strength until a stage is reached
most fatal to rabbits, but cultures obtained at this stage induce
but a transient illness in pigs, which, consequently, can be in-
oculated or " vaccinated " against the disease. Here exaltation
of virulence for one animal and attenuation for the pig can be
produced by keeping cultures at 37° C. for a long period.

Equally instructive cases may be gleaned from researches
upon non-pathogenic micro-organisms. Let me take first the
Microbacillus prodigiosus. It was found by Wasserzug4 that
the addition of antiseptics to the medium of culture, in quantities
sufficient to retard growth, led not only to the loss of colour
production, but also to accompanying remarkable changes in
form : some individuals become much elongated, others develop

1 Malm, Annates de VInst. Pasteur, iv., 1890, p. 520.

2 I may here call attention to a statement in Professor C. Frankel's
Bakterienkunde (edition of 1891, p. 177) : " Es ist . . . bisher noch nicht
gegluckt, eine dauernende Verstarkung der Bakterien, eine haltbare Zunahme
ihrer naturlichen Virulenz herbeizufuhren . . . auf Tiere . . . eine raschere
und andere Wirkungsweise an den Tag zu legen, als sonst an ihnen bemerkt
wurde." While these statements, in face of French researches, were, to say the
least, already very debatable a year ago, they certainly now cannot be regarded
as other than quite incorrect.

3 Pasteur and Thullier, Comptes rendus, xcvii., 1883, p. 1113.

4 Wasserzug, Annates de VInst. Pasteur, ii., 1888, p. 75.

PRODUCTION OF BACTERIAL RACES 117

into actual spirilla. This polymorphism and its extent depend
(as I have already shown in the case of the Bacillus pyocyaneus)
upon the composition of the medium. Upon returning such
altered microbes back to the potato there is a gradual return to
the original state. Yet it is possible by slight modification of
this method to produce permanent races. Taking beef-broth
to which tartaric acid has been added, Wasserzug found that
when the culture was a day or two old there was a great develop-
ment of spirilla. Later, when through the production of tri-
methylamin the medium became alkaline, the spirilla ceased
to be developed ; there was a return to the coccus or shortened
bacillary form. If now he did not permit the production of
the alkaline trimethylamin, but made new acid growths before
the cultures were more than forty-eight hours old, he discovered
that the longer he persevered with his series of cultures the
longer the microbe took when grown upon potato to return to
the coccus form, until finally by such acid growth and subsequent
heating to 50° C. for a few minutes, all that he could obtain upon
sowing on potatoes was, not a micrococcus, but a long bacillus —
a permanent race of such.

Or let us return again to the Bacillus pyocyaneus, which,
though pathogenic, may be discussed along with other chromo-
genic forms.

By placing his modifications, to which I have already referred,
under special conditions of heat, etc., Gessard found that he
could obtain permanent races upon ordinary beef-broth, one
giving the green fluorescent pigment alone, another the blue
pyocyanin, another colourless.1 The same observer, studying
the bacillus of blue milk (Bacillus cyanogenus or syncyanus),
found that by passages through egg albumen he could obtain a
race giving a deep pigment (which becomes red with alkalies,
blue with acids), another race which is only fluorescent, gained
from cultures that are some months old, and a third colourless
race, this last either by taking extremely old cultures or by
warming a recent culture to a temperature which does not
cause death.

Similarly Laurent,2 working with the Bacillus ruber from
Kiel (Bacillus rouge de Kiel), found that exposure of a growth
to bright sunlight for three hours gave colonies which with but

1 Loc. cit. 2 Loc. cit.

118 ON VARIABILITY AND ADAPTATION

few exceptions were colourless, and though with further succes-
sive cultivations there was manifested a tendency for many of
the latter colonies to revert to type and develop red pigment,
yet by careful selection he was able to gain an absolutely de-
colourized race which remained colourless, that is, upon agar-
agar, and in broth at a temperature of 25° to 35° C, a temperature
and media at which and in which the original growth always
shows a strong reddish-violet tint. On potatoes at the same
temperature he was able to make thirty-two successive cultures
extending over a year, and in these not the least trace of colora-
tion ever appeared.

These facts, if alone they were all that we had to consider,
would fully justify the statement that it is possible to evolve
artificially species or subspecies of bacteria. But there are
other facts which must of necessity be taken into account before
we can arrive at any definite conclusion — facts which so far as
I know apply at present to micro-organisms alone, though every-
thing points to the belief that they are operative, and should be
considered, in discussing the development of new species and
subspecies, among higher organisms.

In cultivating the bacteria we are confronted with the fact
that successful culture necessitates the employment of special
media of growth. We have to sow the microbes in fluid or solid
substances, whose composition depends upon the mode of life
of the special form studied. And thus there is and can be no
common soil in which all forms propagate themselves with equal
facility. We are deprived of what I may term a common base.
We cannot, with certainty, say that such and such a form,
which we have separated out, is a distinct species, inasmuch as,
grown upon a common normal soil, it presents permanent char-
acteristics. The most that we can do is to approach, as near as
possible, to the formation of such a common soil. Thus for
many pathogenic and most other bacteria we find that slightly
alkaline beef-broth, and beef-broth that has been rendered solid
by the addition of gelatine or agar-agar, are media in which
growth occurs freely, and in which, certain elementary precau-
tions being taken, a given micro-organism retains its characters
for countless " generations," retains them so completely that
in the case of the majority of pathogenic microbes a series of
successive cultures, if made with due precautions and sufficient

THE LIMITATIONS OF PERMANENCY 119

frequency, will yield microbes which will all produce typical
symptoms when inoculated into a series of rabbits, guinea-pigs,
or other animals, the microbes of the last of the series of succes-
sive cultures possessing the same degree of virulence as the
first of the series.

Where this is the case, are we justified in stating that the
species is permanent ? Upon first consideration one is inclined
to say that undoubtedly we are, yet further thought brings in
doubt, for this is far from being all. There is another and
remarkable order of appearances to be considered, which can
be exemplified from every case that I have given of the establish-
ment of permanent races.

Take, for instance, the races of anthrax bacilli produced by
Pasteur's method. These, undoubtedly, when kept upon our
approximation to a common soil, namely, upon beef -broth,
preserve a well-defined permanency. But select any one of
these, and pass it through a series of animals of increasing re-
fractoriness to the disease, and there is evolved a race growing
equally well upon the beef-broth, whose virulence has been
greatly increased. Or take Gessard's " permanent ' modifica-
tions of the Bacillus pyocyaneus and the Bacillus cyanogenus
respectively ; grow any of the former upon the rich culture
medium formed of agar-agar, glycerine, and peptone, and all
develop equally into a race producing large quantities of an
intense blue pigment ; grow any of the latter upon broth con-
taining 2 per cent of glucose, and all give rise to the specific
blue pigment of the typical bacillus. Again, take Laurent's
colourless race of the " Bacillus rouge de Kiel," and place it
upon potatoes at a temperature of about 18° instead of 25°
to 35° C, and the coloured form reappears.

Or to epitomize. It is possible with bacteria to bring about
modifications which persist when a return is made to the ordinary
media of the bacteriologist (whether these be only our approxi-
mation to the normal — beef -broth, potatoes, etc. — or whether
they be the bodies of one or other species of animal, mice, sheep,
etc.), so that now cultivated upon these " ordinary " media, the
modification preserves its characteristics, and remains markedly
different from the initial or type growth upon the same media.
And here we have a distinction between the races of bacteria
and the races of animals under domesticity, which, returned to

120 ON VARIABILITY AND ADAPTATION

the wild state, are said to revert to type. Now with the bacteria,
where such races have been developed, reversion to type may
still be brought about by a further process, namely, by a further
alteration of environment. I know of no case quite analogous
to this among the higher animals.

It is only within strictly limited conditions of environment
that the characteristics of the usual form or of a race can be
preserved ; overstep these limits, make certain alterations in
the environment, and type and race alike are liable to vary,
although the liability is perhaps the greater for the race to
approximate towards the type than for the type to produce
well-marked races. The only cases that I can call to mind
where this law apparently does not hold good are both of modifi-
cations of the anthrax bacillus. It is possible so to attenuate
the anthrax bacillus that a race is developed which is absolutely
non-pathogenic even for the most susceptible of small rodents,
and so far no efforts have succeeded in re-establishing the
virulence of such attenuated forms. Roux's asporogenous race
also would seem thus far to remain asporogenous under all
conditions, although variations may be induced in its virulence.
Even these cases, however, will probably be found eventually
not to be exceptions to the rule.

For these reasons I have throughout spoken of types and
races rather than of species and subspecies or varieties, for by
employing the more uncommon terms in this connection I have
attempted to guard myself against appearing to hold that among
the bacteria there is anything beyond a relative permanency.
A study of the observations that have been made up to the
present time does not lead to the conclusion that new species
are readily developed ; all that it clearly indicates is that among
the bacteria our employment of this term " species " must be
elastic — the limit to which the individual may depart from the
type (and from which it may again revert to type) is very far
removed. But granting freely that not one of the cases here
described records the development of a new species, this must at
least be admitted, namely, that the longer the action of any one
factor upon the bacterial growth, the longer the time requisite
to ensure a return to the typical condition ; the stronger the
impress left upon any individual microbe, the longer, again, the
time requisite to ensure a similar return. If, therefore, within

ON NATURAL RACES 121

the relatively short period of experiment changes so marked,
changes persisting through so many generations or cell multiplica-
tions, can be brought about, surely it is not difficult to comprehend
changes in the natural world whose action is sufficiently great
and is prolonged through a sufficiently long period to produce
races of even greater permanency, races which might be termed
new species.

On Natural Races

Now apart from these results of experiments we possess
already a series of data whose explanation is easiest on the
supposition that we are dealing, not with forms that are wholly
distinct and of separate origin, but with allied forms and natural
races. Perhaps the best example is one that has been extensively
discussed. Fehleisen's Streptococcus erysipelatous, the microbe
causing erysipelas, is, in microscopic appearance and in mode
of growth, indistinguishable from the streptococcus which can
be isolated from many boils and cases of abscess formation.
Yet certainly the two micro-organisms differ in pathogenicity,
differ in the symptoms they induce when isolated. Are we to
say that here we are dealing with two distinct species ? A few
years ago all bacteriologists answered this question in the affirma-
tive, but now most would reply in the negative, although some,
among whom may be included Crookshank,1 still adhere to
the old view. The observations of Eugen Frankel would seem to
indicate that truly we are dealing with what are only races.2
Frankel obtained from a case of " universal peritonitis " a pure
growth of streptococcus. There could be here no question of
erysipelas. Taking some of a fifth culture of this, five months
after it had been isolated, he inoculated it into a rabbit's ear
and produced an exquisite bullous erysipelas, with streptococci
in the lymphatics, identical with human erysipelas. The same
inoculated into the peritoneal cavity gave rise to a fibrinous or
fibrino-purulent peritonitis. From this it is evident that the
difference between erysipelas and pyaemia, in its various forms,
depends on mode, and locality, of infection and partly on the
quantity of virus, on the individual, and lastly, on differences

1 Crookshank, International Congress of Hygiene and Demography,
August 1891.

2 Eug. Frankel, Cenlralbl. f. Bakteriologie, vi., 1889, p. 691.

122 ON VARIABILITY AND ADAPTATION

in the intensity of virulence of the races of streptococcus.1 It
is interesting to note how, in clinical practice, one observes this
conclusion borne out ; to observe the succession of cases passing
through ordinary erysipelas and acute lymphangitis,2 bullous
and phlegmonous erysipelas, to acute pyaemia. Indeed, one
is driven to the same conclusion with reference to the Strepto-
coccus pyogenes that Levy arrived at after several years' work
at pyogenic micro-organisms in general.3 Levy denies that
anything can be prognosed from the nature of a microbe causing
suppuration. " The prognosis of the process whose origin is
referable to a micro-organism depends upon the virulence of
the same, and the degree of virulence of pyogenic micro-organisms
is subject to extraordinary change."

It may be that similar changes in virulence will explain the
phenomena displayed in the case of swine erysipelas and mouse
septicaemia, cholera, and the Odessa fowl disease respectively.
The micro-organisms of the first pair of diseases are in appear-
ances, size, and method of growth scarce to be distinguished ;
but whereas the one form produces a fatal disease in pigs, the
other has no action whatsoever upon these animals ; rabbits
also are susceptible to swine erysipelas, while they are said not
to be affected by mouse septicaemia. Still it is worthy of notice
that Loffler has occasionally obtained fatal results in rabbits
with inoculations of the bacillus of the latter disease, and that
the bacilli of both diseases show themselves peculiarly fatal to
white mice, producing similar effects. Quite recently, Lorenz4
has shown that pigs inoculated with cultures of the bacillus of
mouse septicaemia are protected against swine erysipelas, and
has described a third form, causing an eruptive disease in pigs,
which he terms " Backsteinblattern." The bacillus of this, in
mode of growth and properties, is intermediate between the
other two, and of these three any one, either virulent or properly
attenuated, confers immunity against the other two.

So again with the second pair of diseases. It is impossible

1 A paper by Behring (Centralblatt f. Bakteriol. xii., 1892, p. 192), just
published, confirms this view of the identity of the pathogenic streptococci,
and shows that an animal rendered immune to one of the various races
of these (vide Lingelsheim, Zeitschrijt f. Hygiene, x., 1891, p. 331 ; gives
literature of subject) is now immune to all other pathogenic streptococci.

2 Vcrneuil and Clado, Comptes rendus, cviii., 1889, p. 7l4.

3 Levy, Arch. f. exp. Pathol, und Pharmak. xxix., 1891, Parts I. and II.

4 Lorenz, Archiv f. wiss. u. prakt. Tierheilkunde, xviii., 1892, p. 38.

NATURAL KACES OF BACTERIA 123

to distinguish with certainty a preparation of the Vibrio metch-
nikovi from one of Koch's cholera spirillum, and the differences
in the mode of growth of the two are not greater than may be
determined between two cultures of the cholera spirillum of
different origin. Yet Vibrio metchnikovi is pathogenic for
pigeons and fowls and only affects guinea-pigs (which are sus-
ceptible to inoculations of the cholera spirillum) when inoculated
in relatively large quantities. According to Gamaleia1 injec-
tions of the Vibrio metchnikovi render guinea-pigs immune
to cholera. Pfeiffer denies this, though Gamaleia has reiterated
the statement,2 and ascribes Pfeiffer's failure to gain immunity
to the fact that he employed attenuated cultures. It must,
however, be admitted that Gamaleia's statement still awaits
confirmation, and, for the present, it is better to look upon the
series of microbes which includes Koch's spirillum of Asiatic
cholera, the Finkler-Prior spirillum of Cholera nostras, Deneke's
S. tyrogenum, Miller's spirillum obtained from the mouth, and
the Vibrio metchnikovi as a group of very closely allied species.

Similarly so high an authority as Professor Hueppe 3 would
class together into one group, as the bacteria of haemorrhagic
septicaemia, a large number of bacteria which microscopically
are indistinguishable, whose ends stain more deeply than the
central region, whose growth upon bacteriological media is
similar, and which, further, are identical in the appearance of
the individual colonies. C. Frankel 4 would separate these into
two groups, one including the bacillus of ferret plague (Eberth
and Schimmelbusch), the bacillus of swine plague (Billings),
that of hog cholera (Salmon), that of Danish swine plague
(Selander), and the bacillus of the German " Schweinepest."
All these are motile and ciliated. The other group contains
forms that are identical, or almost identical, namely, Loffler's
bacillus of chicken cholera, Schutz's swine plague bacillus,
Cornil's bacillus of duck cholera, Gaffky's microbe of rabbit
septicaemia, and Kitt and Hueppe's bacterium of " Wildseuche '
(deer plague). All these only differ in regard to the animals
which they specially affect.

Here, too, I would mention a most interesting observation of

1 Gamaleia, Annates de VInst. Pasteur, iii., 1889, p. 609.

2 Ibid, iv., 1890, p. 330.

3 Hueppe, Berliner Iclin. Wochenschrift, No. 9, 1890.

4 C. Frankel, Grundriss der Bahterienkunde, Berlin, 1891, p. 401.

124 ON VARIABILITY AND ADAPTATION

Hueppe and Cartwright Wood. These observers gained from
ordinary earth a harmless saprophytic bacillus, in appearance
and mode of growth resembling closely the Bacillus anthracis.
Inoculated into mice, which are of all animals the most sus-
ceptible to anthrax, pure growths of the earth bacillus induced
immunity against this very fatal disease.1

On the observed Modifications of Pathogenic Microbes

There is, however, another perhaps more satisfactory way
of approaching this subject of the occurrence of natural races.
I refer to the actual differences in mode of growth and properties
that are to be distinguished in cultures obtained from diverse
typical cases of any given disease. This subject has not been
worked out so fully as it deserves, nevertheless certain very
interesting observations have been made. Thus the Talamon-
Frankel diplococcus of acute croupous pneumonia has been
studied by Banti.2 Banti describes four varieties of the species
Diplococcus lanceolatus capsulatus. No. 1 is the typical form de-
scribed by Frankel and Weichselbaum ; No. 2 is identical in form
and culture, but causes a " diplococcus septicaemia " in rabbits,
with small spleen and destruction of the red corpuscles ; No. 3
is similarly identical, but produces a septicaemia with moderate
enlargement of spleen and diffusion of haemoglobin ; and No. 4
only differs from the rest in that its virulence outside the body
disappears even more rapidly than is the case with the other
forms ; it produces a mild septicaemia associated with albumin-
uria, and the animals all recover. All were obtained originally
from typical cases of pneumonia, and while in the years 1886,
1887, and 1890 the type form was alone gained from almost
every case of pneumonia — and these were of a benign nature —

1 Hueppe and Wood, Lancet, December 7, 1889. [May it have been that
Hueppe and Wood's harmless saprophytic bacillus obtained from earth was the
B. mycoides of Thiele and Embleton referred to in Chapter III. of Part I. ?]

2 Banti, Lo Sperimentale, xliv., 1890, pp. 349, 461, 573. An observation
of Nikiforoffs may be quoted in this connexion (Zeitschr. f. Hygiene, viii.,
1890, p. 531). From the lung of an influenza patient he obtained a microbe,
identical with the Diplococcus pneumoniae in every respect save that it pre-
served its virulence for longer periods and was fully virulent for mice, while
rabbits were completely refractory, whereas the true diplococcus is virulent for
both mice and rabbits. [It is interesting that renewed study of the pneumo-
coccus has established the existence of four well-defined strains.]

STRAINS OF PATHOGENIC BACTERIA 125

in 1888 and 1889 the type No. 1 was never isolated, but the other
varieties were present, and the cases were severe.

Foa1 has shown that by inoculation variation may be
induced in the diplococcus. If from a rabbit that has died of
inoculation pneumonia two rabbits be inoculated, one from
the fresh fibrinous pneumonic exudate, the other from the
cerebro-spinal fluid, it is found that the disease differs in the
two. The former shows inflammatory oedema of the skin, the
latter none. So, too, if the diplococcus be grown for twenty-
four hours anaerobically the latter type is produced and its
properties persist, although the original virulence and characters
may be restored by simultaneous inoculation of this form along
with the Staphylococcus pyogenes aureus and the proteus vul-
garis. These studies of Foa find their counterpart in Banti's
researches into acute primary meningitis.2 From two cases of
this disease Banti isolated a diplococcus not to be distinguished
from that of Talamon-Frankel, except that it rapidly lost its
virulence both inside and outside the body, so that when the
meningeal exudation was inoculated into a rabbit, and from
this a further series of passages was made, the sixth rabbit had
a mild non-fatal disease.

In the case of typhoid, Babes 3 has made a long series of
observations, and though he handles the subject with what for
him is extreme caution, it is difficult to arrive at any other con-
clusion than that the facts he represents can only be satisfactorily
accounted for on the assumption that, granting there be a specific
micro-organism for typhoid, and granting his results to be reliable,
this micro-organism is capable of modifications so definite that
many races are developed. Babes made numerous cultures
from twelve typhoid corpses, and while in nearly every case he
found the typical form, in every case he discovered also varieties.
In all he describes eighteen of these, all presenting some slight
differences. These were sufficiently permanent to permit him
to declare that he never saw one revert to the original form
(as determined by an original culture from Berlin), although in
many there was a tendency towards the loss of characteristics.

1 Foa, International Medical Congress, Berlin, 1890 (Section of General
Pathology).

2 Banti, Lo Sperimentale, xliii., 1889, p. 138.

3 V. Babes, Zeitschrift f. Hygiene, ix., 1890, p. 323.

126 ON VARIABILITY AND ADAPTATION

While acknowledging that these varieties, with rare exceptions,
produce in small animals lesions similar to those due to the
typical bacillus, he concludes that they are not specific, and that
typical and atypical belong to a group possessing similar condi-
tions of existence. He notes also that the Bacillus coli communis
possesses many natural varieties. Despite this very cautious
conclusion, Gaffky (who with Eberth shares the honour of having
discovered the typhoid bacillus) criticised Babes severely, and
suggested that all his varieties were due to secondary " Ein-
wanderung." x To this Babes replied 2 by showing that if
so the varieties had wandered in during life, for his autopsies
were made upon the fresh corpse very few hours after death.
Surely rather than multiply species to the enormous extent that
these conclusions would demand it is simpler, as it is more
inherently probable, to hold these varieties to be races and of a
common origin : races, it may be, produced by the peculiar
conditions of the contest between organism and microbe. At
the same time I am not prepared to go as far as Arloing,3
who declares that the bacillus of typhoid can be developed from
the Bacillus coli communis in the presence of fermenting faecal
matter.

The subject of the bacillus of typhoid cannot be passed over
without mentioning also Cassedebat's important paper.4 The
relation of epidemics of enteric fever to the water supply has
been so clearly traced that all that was wanting to afford an
absolute proof of the relationship was the demonstration of the
presence of the specific bacillus in the suspected sources. And
this demonstration would seem to have been made by the methods
of Vincent, Rodet, Kitasato, Chantemesse and Widal, Thoinot,
and yet others. But Cassedebat, working at Marseilles where
enteric fever is almost endemic, states that in the water supply —
the Durance — he was unable to discover the typical bacillus,
although in its place he determined three forms all resembling
the bacillus of typhoid in their mode of growth upon potato
(which is very characteristic), and in the form of the colonies
upon agar-agar plates, all equally polymorphous, and staining
with difficulty, all possessing the same movements of transla-

1 Gaffky, Hygienische Rundschau, 1891, No. 12.

2 V. Babes, Centralbl. f. Balct. x., 1891, p. 281.

3 Arloing, International Congress of Hygiene and Demography, London, 1 89 1 .

4 Cassedebat, Annates de VInst. Pasteur, iv., 1890, p. 625.

STRAINS OF PATHOGENIC BACTERIA 127

tion and oscillation, and giving much the same growths upon
puncture of gelatine, while in even so minute a characteristic
as the arrangement of the cilia Loffler's method showed no
difference between No. 1 and the type. (It is not stated how
the others appeared under this treatment.) But there were
differences of rate of growth on potatoes, in gelatine, potato-
juice-gelatine, and broth, differences also in the intensity of
colour of potato cultures, differences in the effect of the bacilli
upon media to which aniline dyes had been added, and the
result on inoculations, which were far from complete, show
that on the whole No. 1 was less virulent for mice than in the
type Eberth-Gaffky bacillus. From these differences Cassedebat
concludes that he is dealing with pseudo-typhoid forms, and
that the results of the bacteriological examination of water must
be received with very great caution, if they are to be accepted
at all. I have already shown that the Eberth-Gaffky bacillus
is capable of very considerable modifications, is most susceptible
to change of medium, etc., and here, again, rather than accept
Cassedebat's conclusions, I would hold that the more satisfactory
explanation of his facts is that he dealt with races — natural and
fairly permanent races — modifications of the type bacillus of
typhoid.

Turning now to cholera and its spirillum, it is already a
well-known fact — to which attention was, I believe, first called
by Zaslein 1 — that cultures of the spirillum derived from
simultaneous epidemics in different localities present recognizable
differences — differences in tint, in rate of growth, in power of
liquefying gelatine, etc. So there are those who declare that from
the appearance of a culture they can state its origin, whether from
Cairo or from Berlin, from Naples or Massowah, from Palermo
or Marseilles. It has, however, been left to Surgeon -Major
Cunningham to give the most remarkable example of these
differences.2 At the beginning of 1890 three of the principal
hospitals in Calcutta contained cases of cholera, not differing
from one another in symptoms or virulence, all giving abundance
of cultivable spirilla, but these not of one but of three " species."
Later, other cases were observed, in all sixteen, and from these

1 Zaslein, Deutsche med. Zeitung, ix., 1888, 759 and 771.

2 D. D. Cunningham, Scientific Memoirs by Medical Officers of the Army
of India, Part VI., 1891, p. 1.

128 ON VAKIABILITY AND ADAPTATION

sixteen, ten " species " were obtained, a to Jc. In only four of
the cases was the form present, even doubtfully, that described
by Koch (Cunningham, bacillus a). In some three different
comma bacilli were present at the same time.

The forms described by Cunningham may be divided into
two groups. The first contains only one which does not liquefy
gelatine, does not show interstitial growth in agar-agar, does
not develop " cholera red " on the addition of acids, and morpho-
logically is distinguished by its large size and great variability
of form and curvature. This I feel inclined to consider as a
truly different species. The other group contains all the other
nine, which liquefy gelatine, are capable of interstitial growth,
give the " cholera red " reaction, but all show fairly permanent
differences in the rate of liquefaction of gelatine. In six the
rate is distinctly rapid, in the rest it is slow. In one growth
upon potato is luxuriant, in others there is rare acclimatization
to this medium, in others, again, acclimatization has never shown
itself. Here, then, one working where cholera is endemic, and
one who is a most capable observer, who had been with Koch
when he studied the disease in India, is so affected by the general
view that species are constant and distinct, that the only con-
clusion that he can draw from these facts is that " whilst denying
that the primary cause of cholera is represented by any species
of intestinal schizomycete which has yet been discovered,"
it may be " that those whose development is favoured by the
existence of the choleraic condition may exert an important
influence on the ultimate outcome of individual cases of the
disease." If we accept the view that the spirillum is specific
I do not see any other conclusion from these observations than
that here is another well-marked instance of the production of
allied races.

Only within the last few days news has come from St. Peters-
burg that the spirillum obtained from patients affected in the
cholera epidemic now raging presents marked differences in
size, etc., distinguishing it from Koch's original form ; while
from the Institute Pasteur I hear that the microbe, isolated
from some of the doubtful cases in Paris, presented no greater
departure from type than is to be seen in many of the growths
obtained from indubitable cases of cholera.

I might enter into the discussion of the relationship between

STRAINS OF PATHOGENIC BACTERIA 129

human, bovine, and avian tuberculosis, or again into the differ-
ences to be made out between the bacilli of pulmonary and
surgical tuberculosis (lupus, tuberculosis of bone and joints,
and scrofula), but though much has been done on these subjects
there is a want of well-confirmed results, and the subject is
still too chaotic for safe treatment. I will conclude this relation
of the occurrence of natural races with a reference to Roux and
Yersin's remarkable and admirable work upon diphtheria.1
Apart from proving that the bacillus isolated by Lofiler 2
is specific, by showing that the sterilized medium of growth
inoculated into animals will induce the characteristic symptoms
of the disease (including the slowly produced paralyses), these
observers made careful studies of the pathogenic qualities of
the microbe. They found that, according to their origin, some
cultures would kill guinea-pigs in twenty-four hours, some in
sixty, some in only three or four days, or even after a longer
interval, and they turned their attention to a form or forms
that had been described by Lofiler as the Bacillus pseudo-diph-
thericus. Lofiler found this in the false membranes along with
the virulent micro-organism, and pointed out that it is practically
indistinguishable from this last, save that it has no toxic effect
upon animals. Hoffmann found it also in the pharyngeal
mucous membrane of cases of scarlatina and measles.3 Its dis-
coverer considered it a separate species, and in this view most
observers joined, though Klein and Flugge admitted that it
was an attenuated form. The colonies on coagulated blood
serum are identical with those of the diphtherial bacillus ; like
that, it grows rapidly at a temperature of 33° to 35° C. ; micro-
scopically the two are the same, the staining reactions are
similar, and growth in alkaline beef-broth is characterized by
the same production of acidity followed by increasing alkalinity.
The differences are, that it is often shorter when grown on blood
serum, that the cultures on broth are more abundant, and so
are those on agar-agar. It still grows at 20° to 22° C, a tem-
perature at which the virulent bacillus ceases to proliferate,
and the alterations in alkalinity and acidity of both cultures
proceed more rapidly.

1 Roux and Yersin, Annales de VInst. Pasteur, iv., 1890, p. 385.

2 Loffler, Congress of Military Surgeons, Berlin, 1887 ; and Cenlralb. f.
Bakt. ii., 1887, p. 105.

3 [This form is now commonly referred to as " Hoffmann's bacillus."]

K

130 ON VARIABILITY AND ADAPTATION

This form is very rare in malignant cases of diphtheria — it
is more abundant in benign, and becomes increasingly common
as severe diphtheria progresses towards cure. But, as I have
already indicated, it is to be discovered in non-diphtherial cases.
At the Hopital des Enfants Malades, Paris, from the pharyn-
geal mucous membrane of 45 children not affected with the
disease it was obtained in 15 cases, and of 59 healthy school
children at Caen, in a school where for long there had been no
diphtheria, no less than 26 gave growths of this form, though it
must be confessed that the colonies were few and far between.
Inoculation into guinea-pigs and rabbits never led to fatal
results, and yet differences were observable. Some guinea-pigs
manifested a very considerable oedema at the point of inocula-
tion, in others a slight oedema was to be seen, others presented
no sign of even local disturbance. It is noteworthy that the
most marked oedema was developed where cultures were employed
whose origin was from cases of measles.

Oedema at the point of inoculation is one of the character-
istic lesions when virulent diphtheria bacilli are injected, and
Roux and Yersin found that if they attenuated virulent cultures
either by the action of air and high temperature, or, again, if
they preserved dried-up false membranes for five months in
a cool place and in the dark, they gained colonies, some of which
caused no oedema, others a little, one alone caused marked
oedema. Such attenuated bacilli, like the Bacillus pseudo-
diphthericus, grew more abundantly at a lower temperature,
and, like this, also rendered broth more rapidly alkaline. In
fact, the only point distinguishing the two was that our observers
succeeded in intensifying the virulence of the attenuated form,
and failed to do this with the Pseudo-diphtherial bacillus. In
every other respect the proof appears conclusive that this Bacillus
pseudo-diphthericus is a natural race, or races, of the Bacillus
diphthericus, that it is no wise a separate species of independent
origin.

What is more, Roux and Yersin suggest an explanation of
the onset of diphtheria — of the natural intensification, that is,
of the virus in man. Just as they found that from a case of
measles presenting no sign of diphtheria proper they gained a
race of the Bacillus pseudo-diphthericus, which exhibited distinct,
if slight, virulence, so have they noticed cases in which, through

EVOLUTION OF VIRULENCE 131

the access of an anginal attack, or of measles, diphtheria, which
was recovering, gained fresh malignancy and became rapidly
fatal.

There is another micro-organism frequently found in the
mouth and saliva of healthy persons, namely, the Talamon-
Frankel diplococcus. May it not be that in both these cases
a natural intensification of the virus is brought about by catarrhal
and other affections of the upper respiratory and pharyngeal
mucous membrane, and that to such intensification, rather than
to the entry afresh of virulent micro-organisms, is to be ascribed
the onset of these diseases ?

I am well aware that in setting in order a series of facts and
observations upon deviations from the normal the tendency is
peculiarly strong to see in such deviations the normal, to see in
the normal the exceptional, and it may be that in the preceding
pages I have not dwelt with sufficient emphasis upon the fact
that employing well-recognized and standard methods the typical
forms of bacterial growth are easily and most usually to be
separated out from cases of disease. Still I shall feel that these
pages have not been written in vain if I succeed in drawing
increased attention to the fact that the bacteria are organisms
acutely susceptible to changes in environment, that as species
they are far from presenting constant characteristics, and that
to a variability which may impress itself upon a greater or less
number of generations is to be ascribed, in part, the differences
between successive epidemics, between the successive stages
of one epidemic, and between individual cases of disease.

REFERENCES

The following papers may also be consulted :

Hansen (E. C). Annales de Micrographie, ii., 1889, p. 214. (" On the Produc-
tion of Varieties among the Saccharomycetes.")

Metchnikoff. Annales de l'lnst. Pasteur, iii., 1889, pp. 61 and 265. (" On a
Remarkable ' Spiro Bacillus ' (Polymorphic) Parasitic in Daphnia.")

Winogradsky. Annales de l'lnst. Pasteur, iii., 1889, p. 244. (" On Poly-
morphism.")

Hueppe. Berliner klin. Wochenschrift, 1884. (" On the Relationship of
Saprophytic to Pathogenic Bacteria.")

Delepine (Sh.). Transactions, Pathological Society, London, 1891. (" Muta-
bility of Pathogenic Aspergillus, as also of the Characters of the B. Leprae.")

Unna. Deutsch. Naturforscherversammlung, Halle, September 1891, and
Centralblatt f. Bakt. xi., 1892, p. 638. ("Races of the Favus Fungus
(Achorion Schoenleinii).")

CHAPTER II

ON THEORIES OF INHERITANCE WITH SPECIAL REFERENCE TO THE
INHERITANCE OF ACQUIRED CONDITIONS IN MAN X

(1901)

Questions of inheritance at the present moment occupy a curious
position in the minds of medical men and in medical literature.
To judge from the medical press, we medical men are very Gallios
— we care for none of these things. And yet, in family, as in
consulting practice, questions concerning heredity must and do
continually present themselves. In attempting to arrive at a
conclusion about a given case, we are bound to ask ourselves how
far the frailties or follies of progenitors are responsible for the
conditions found — how far the accidents or the sins of the indi-
vidual. Each succeeding day you must have this question of
possible inherited defect brought before you ; constantly must
you be forced not merely to inquire whether certain phenomena
are matters of inheritance, but assuredly to recognize that this or
other condition runs through all the members of a family and is
an inherited weakness. And yet, although the lay reviews dis-
cuss the matter familiarly, and although perchance the charming
partner you take in to dinner does the same, we scarce write about
these things, save in connexion with one or two branches of
medicine, and when we do, I have no hesitation in stating, though
it is a bold statement, that much of what is written is misleading.
Even in my own subject of pathology, treating as it does
of the causes, the processes, and the results of disease, in the
discussion of which the laws of inheritance should obviously be

1 The first of a series of annual addresses under the auspices of the Brooklyn
Medical Club, delivered May 17, 1901. Reprinted from the New York Medical
Journal for June 1, 1901, and British Medical Journal.

132

INHERITANCE AND MEDICINE 133

studied with care, if inheritance plays even a debatable part, turn
to any of the text-books in our language and what do we find 1
A single page, or it may be but a single paragraph, is thought
sufficient to introduce and to take leave of the subject. In short,
from a concatenation of circumstances the medical study of in-
heritance is largely " taboo." Why is this ?

It depends upon more than one factor. In the first place,
while, as I shall point out later, the study of man is singularly
well adapted for determining certain points in connexion with
inheritance, the fact that the generations of man follow each
other at such relatively long intervals is against man being for
most purposes a good subject for investigation. In the course of
a long life the investigator can but study the characters of three,
or it may be four, generations in one family, while influences
acting upon the susceptible foetus in utero in man, as in all mam-
malia, introduce complications. The basal facts of heredity have
to be made out in the lower animals, in which generations succeed
each other with fair rapidity, and in which the eggs are fertilized
and from the first developed outside the body.

Unfortunately, too few of us are trained biologists ; the curri-
culum of the past, as of the present, lays too little stress upon the
value of a broad biological training as an aid in preparing us to
discuss those special biological problems which constitute medical
study. As a consequence the medical world in general has to
depend upon the biologists proper — upon the zoologists and
botanists — for its views upon heredity, and the pure and simple
biologists have run riot in their lucubrations upon this subject.
Do not think that I mean to belittle them or to indicate that we
do not owe much to all the investigations and all the writings of
the biologists of the last twenty or thirty years. The facts which
they have elicited have been of the highest value. Without
these facts we would be nowhere, but the contending theories
elaborated by them (perhaps I should be the last to make any
such criticism) have been fearful and wonderful, have started
from morphological rather than physiological conceptions, and
as a result have assumed shapes which would not disgrace the
schoolmen of the Middle Ages. While they have appeared to
collate and harmonize the facts known at a given moment, new
facts have caused them to need modification, and the successive
attempts to utilize the old bottles for the new wine, where they

134 ON VARIABILITY AND ADAPTATION

have not burst the bottles, have led them to assume most grotesque
shapes. In fact there has been developed such a muddle that no
amount of midnight oil and wet cloths bound around the temples
permit the ordinary mortal to disentangle and follow the course
of one theory.

This being the state of affairs, it is little wonder that we
have been unwilling to apply the theories of the biologists
to the problems of medicine, and this all the more because
the trend of these theories has been in apparently strong
opposition to medical experience. Of all the workers of late
years Weismann has had the most influence upon biological
thought, and his theory, if not accepted by all in its
entirety — and if, indeed, now found unacceptable, — has, never-
theless, profoundly affected the general consideration of this
subject of heredity. That theory is very complicated, and with
Weismann's successive publications has not by any means become
easier to follow or epitomize in language devoid of technicalities.
Still, if not every schoolboy, at least every educated man is
supposed to comprehend the general tenor thereof and its main
thesis or conclusion, that acquired characters — characters ac-
quired by the individual — are not and cannot be transmitted to
the offspring. To use Weismann's own expression, " We main-
tain that somatogenic characters (characters originating in the
cells and tissues of the body) are not transmitted, or rather, that
those who assert that they can be transmitted must furnish the
requisite proof. The somatogenic characters not only include
the effects of mutilation, but the changes which follow from
increased or diminished performance of function, and those which
are directly due to nutrition and any of the external influences
which act upon the body. Among the blastogenic characters
(characters originating in connexion with the germ cells), we
include not only all the changes produced by natural selection
operating upon variations in the germ, but all other characters
which result from this latter cause." 1 In natural selection is to
be found the key of the phenomena of inheritance.

Now, up to a certain point, we as medical men are prepared
to accept this teaching. We know from experience that acquired
mutilations are not transmitted ; we acknowledge that no clear

1 Weismann, On Heredity. Authorized translation by Poulton, Schonland,
and Shipley, Oxford, 1889, p. 413.

WEISMANN AS INTERPRETED 135

and satisfactory examples of the transmission of mutilations have
been brought forward. We can realize that the loss of an arm,
for example, has no direct influence upon the germ cells of the
individual, and that so these germ cells if fertilized will develop
into the complete individual. But there is another class of
phenomena specially interesting our profession which appears to
give a direct He to Weismann's thesis. And it is, I think, this
failure of the ruling theory to explain satisfactorily the pheno-
mena in question which has been the main factor in making us
as a profession not enthusiastic of late years to debate the
subject.

For, accepting the theory, we must be prepared to deny
wholesale the transmission of acquired defects of every order
and give ourselves over to a most serious form of fatalism.

If an individual is from the first feeble-minded, that is not
his parents' fault ; it is due either to an unfortunate commingling
of the ids, or ancestral plasms, composing the promiscuous
ovum and spermatozoon from which he is derived, or to characters
which have come down to him from previous generations. If
he, being diseased, begets feeble-minded children, he is in no-
wise to blame — acquired characters are not transmitted. Either
those feeble-minded children are an accident — a spontaneous
variation ; or more probably they represent the summation of
characters inherited from long generations. If a man or a
woman becomes an alcoholic, it is not his or her fault, it is the
result of inherited tendencies ; and if the children of the same
are of weak constitution or idiotic, again the parents are blame-
less : characters acquired by the parents are not transmitted,
the characters of the parents must have descended to them.
If a man is a criminal, again he is not to blame ; criminality
is atavism, is a reversion to an earlier state, is an inheritance
of characters or features peculiar to primaeval man. We are,
so the popular translation of Weismann's theory goes, the de-
scendants of criminals, or at least at a certain stage our ancestors
were of an imperfect and criminal type. And if criminality
appears in the family, with imperfect formation of head and
brain and low mental state, that is due to the fact that by the
fortuitous extrusion of certain ids from ovum or spermatozoon,
the ids of the criminal ancestors have preponderated in the
fertilized cell, and the result has been that the individual has

136 ON VARIABILITY AND ADAPTATION

developed possessing criminal features. However much a man
abuses his soma, or body, is of little moment ; the effect upon
the offspring is minimal.

I put this in strong language and baldly, and it may be
urged that I exaggerate the state of affairs. I do not think
I do. I believe that in making this statement I but give
expression to the general, if confused, ideas of the majority :
nay, more, that I state the received conception of what Weis-
mann's theory means when applied to man and to abnormal
inheritance in man.

Now, if there is one conclusion to which we think experience
surelv leads us as medical men, it is that the sins of the father do
tend to be visited upon the children even mito the third and fourth
generation. We think we see this demonstrated day after day.
But Weismann does not support this view. It is true that if we
study Weismann we find that he does not state this in so many
words ; he admits 1 that the germ plasm is not absolutely
unchangeable, that the nutrition and growth of the individual
must exercise some influence upon its germ cells, " but in the
first place this influence must be extremely slight, and in the
second place it cannot act in the manner in which it is usually
assumed to take place." Certainly he does not make it clear
that he believes in the distinct transmission of any order of
acquired characters.

Weismann's Theory. — Weismann, I need scarce say, explains
inheritance along the following lines : the germ plasm, the
essential matter of the fertilized cell from which the individual
develops, must in the process of fertilization come to contain
portions of the germ plasm of both father and mother, brought
to it by the nuclear material of the ovum and spermatozoon
respectively, and the germ plasm of the father and mother
must contain portions of the germ plasms of paternal and
maternal grandfathers and grandmothers. And so, going back
through a long series of generations, it follows, according to him,
that representatives of the germ plasms of a long series of
ascendants, or progenitors, are contained in the nuclear material
of each ovum and spermatozoon. The constitution of a germ
cell therefore may be represented, for purposes of inheritance,

1 Weismann, On Heredity. Authorized translation by Poulton, Schonland,
and Shipley, Oxford, 1889, p. 170.

WEISMANN'S THEORY 137

as made up of a vast number of ancestral plasms, or "ids,"
derived from the long line of progenitors.

I need not here explain his most ingenious demonstration
of the means whereby at each successive act of fertilization a
certain number of these ancestral ids are discharged, so that the
ovum and spermatozoon each contains half the number originally
present, and so that the number of ids in the fertilized ovum is
kept constant. I need but point out, in passing, that by this
process of reduction the set of ids discharged from one germ
cell is by this theory held to be different from that discharged
from another germ cell of the same individual. And as the
ids in two fertilized ova are not identical, as the same series of
ancestors do not contribute to the germ plasm of the two, so
it is that individual variation originates ; the ids varying, the
individuals controlled or developed by these ids tend to vary.
So it is by this fortuitous commingling that spontaneous varia-
tion is apt to show itself.

Further, it has to be noted that when the ovum undergoes
segmentation, and half of the nuclear material passes into one
cell, half into the other, according to Weismann a certain definite
series of these hypothetical ids passes into each cell, and accord-
ing to the series entering each of them, so (and not by any varia-
tion in environmental influence) are determined the eventual
characters of the tissues to which those daughter cells eventually
give rise. The germ cells of the new individual become differ-
entiated at a very early period, receive the full complement of
the ids, and so carry on the whole series of properties inherited
from the ancestors.

Difficulties in accepting this Theory. — This is a crude recapitula-
tion of the main points of Weismann's theory. I have but
mentioned those portions which especially bear upon my argu-
ment. To explain atavism, or the reproduction of characters
in one generation which had not been recognizable in the previous
generation but had been present in some earlier generations,
you will see that the theory demands that some at least of these
ancestral ids should have remained unchanged through a long
period — it may be for centuries. For Darwin's case1 of the
return of features peculiar to the ancestral rock pigeon, brought
about by crossing barbs, spots, and fantails, is clearly a case of

1 Darwin, Origin of Species, sixth edition, p. 18.

138 ON VARIABILITY AND ADAPTATION

atavism occurring after centuries of domestic breeding and loss
of the features in question. So in such a case of atavism we
must, in the terms of this theory, suppose that in the act of
fertilization there is a summation and preponderance of just
that series of unchanged ids which now in development lead to
the bringing into evidence of the atavistic ancestral features.
In other words, the theory demands that the ids must be singularly
stable in constitution — that they grow and multiply, but retain
the same structure.

But now Weismann has to admit that, under certain condi-
tions, the ids are modified in their structure. This admission
indeed is contained in the idea that the individual hypothetical
ids vary in their properties ; and as, if we trace back these
ancestral ids to their common source, they must all originally
have been identical in structure, we conclude that at the same
time they are both stable and capable of change in constitution.
Here indeed is the crux of the theorv. How are we to define
and realize for ourselves the limits of alteration ? Natural
selection cannot explain the alteration, unless we fall back upon
the far-away hypothesis of multitudinous separate acts of crea-
tion in the beginning of things — affording a large number of
distinct idioplasms, — and even this hypothesis does not work
out satisfactorily.

In the example already given of crossing of the old-estab-
lished breeds of barbs, fantails, and spots, we must imagine
that all the ids of each breed have been, in the germ cells of
successive generations, exposed to almost identical conditions,
and, as a consequence, modified along the same lines. Exposed
to the same influences in the course of many generations, it is
almost inevitable that all must become modified, for if there
were any large number of unaltered ancestral ids contained in
the germ cells it would inevitably occur that spots and atavistic
forms would frequently present themselves. But this does not
happen. Each of these varieties of the pigeon breeds singularly
true. How, in short, are we to picture some of them passing
from germ cell to germ cell through all the long years in an
ancestral condition % Put to this test, the theory breaks
down ; we cannot picture the necessary conditions. It is, in
short, an absurdity to regard the nucleus of the germ cell as
containing a colony of what are, to all intents and purposes,

INCOMPETENCY OF WEISM ANN'S THEORY 139

separate and independent individuals, some of which have for
centuries retained properties of one order, some properties of
another — to conceive the germ cell as a colony of individual
living beings, for this is what the theory demands.1

Driesch' s Demonstration of the Incompetency of the Theory. —
But it may be urged, what is the use of all this argument to
kill a theory already dead ? For dead it is, so far as regards
the ids, and Weismann's theory without the ids is like Shagpat
without the identical. The brilliant observations of Driesch,2
abundantly confirmed as they have been by others, foremost
among whom must be mentioned Professor E. B. Wilson 3
of Columbia, and T. H. Morgan 4 of Bryn Mawr, show that
the conception is untenable. If in a segmenting ovum we find
that normally each of the blastomeres or primitive segmenta-
tion cells gives rise to one special series of organs or tissues, but
if nevertheless the ovum of sundry animals can have its cells
shaken apart at the two-, four-, eight-, and even sixteen-cell
stage, and each separated cell can be found capable of develop-
ing into an entire if dwarfed individual, then obviously, each
time the nucleus segments there is no passage into the daughter
nuclei of particular series of ids destined to lead to the develop-
ment of one particular region of the body. Rather the varia-
tion in structure of the different tissues must be, to employ
Driesch's words, " a function of their relative position " (ihre
prospective Bedeutung ist erne Function des Ortes). The exist-
ence of these hypothetical ids is absolutely disproved. I dwell
upon this theory because here I want more especially to
discuss, on account of its importance from a medical point of
view, this matter of the inheritance or non-inheritance of acquired
characters. I hope that I have proved to you that the ground-
work upon which the negative view is based is of proved un-
soundness. The fact that a theory by which a position is sup-
ported falls through does not, it is true, afford proof that the
position is wrong, but when we find that the dictum of non-

1 For a fuller discussion of the weak points in the Weismann-Roux Theory,
regarded from an embryological aspect, vide Wilson, The Cell in Development
and Inheritance, New York, Macmillan, 1898, and Creswell Shearer, Montreal
Medical Journal, May 1901.

2 Driesch, Zeitsch. f. wissenschaft. Zoologie, liii., 1892, and lv.. 1893, and
Archiv f. Entwichelungsmechanih, 1900, p. 361; ibid. p. 411.

3 E. B. Wilson, Journal of Morphology, viii., 1893, p. 579.

4 T. H. Morgan, Anatom. Anzeiger, x., 1895, p. 19.

140 ON VARIABILITY AND ADAPTATION

transmission of acquired characters does not wholly accord
with medical experience, we may well ask : can we gain a con-
ception of the intimate nature of inheritance which is in accord
with that experience ?

Inheritance True and False. — My only regret is that, in
striving to gain that conception, I shall have to inflict upon
you yet another theory ; my only apology, that that theory does
appear to satisfy the conditions met with in man. First, how-
ever, it is necessary to lay down clearly what is not inheritance,
for in medical writings and in ordinary medical parlance a terrible
confusion prevails upon this point, and much that is certainly
not inherited is commonly spoken of as being hereditary. There
is, for example, no such thing as hereditary syphilis. There
is congenital syphilis, and there are, to employ Founder's term,
inherited " parasyphilitis " lesions, but " hereditary ' and
" congenital " are not and must not be regarded as interchange-
able terms.

The confusion is due to the common error of regarding the
individual as beginning his existence at the moment of birth
and not until then, so that everything occurring before that
moment is grouped in one category, everything after in another.
The chick, so to speak, is not a chick until it breaks open the
shell ; its status from the moment it ceases to be a new-laid
egg — or, more strictly, the egg of commerce — until it emerges
from the shell is not recognized in law, and fresh egg and chick
are commodities of wholly different orders. But the individual
existence of the chicken has already begun even before the egg
is laid, and what is true of the chick is equally true of the human
being. The individual begins the moment that fecundation is
accomplished, the moment that nuclear material of the sperma-
tozoon fuses with the nuclear material of the ovum and these
twain become one. Compared with the event, birth is seen to
be of secondary importance, for the intra-uterine association of
the embryo with the maternal tissues is but one means employed
by a restricted number of species to ensure the satisfactory
nourishment of the individual during the earlier stages of develop-
ment. The recognition of these facts is essential for any serious
study of the problems of human inheritance. Any disturbance
due to influences affecting the individual from without while
in utero is acquired. It certainly must not be spoken of as

INHERITANCE TRUE AND FALSE 141

inherited ; it is an ante-natal acquirement or is of congenital
origin. That alone is inherited which is the property of the
individual at the moment he becomes an individual, which is
a property of the germ plasms from which he originates, or is
produced by the interaction of those germ plasms. The biologist
has no alternative but to define inheritance according to the
principle here laid down, nor have we, dealing with a limited
field of biology, the right tq modify terms in general scientific
use for our own convenience.

Now, when we find that syphilis or tuberculosis acquired
in utero during the later months is peculiarly severe and wide-
spread in its manifestation, it is wholly unjustifiable to premise
that the microbic germs of one or other disease could be present
in either the conjugating ovum or the spermatozoon, could pass
into one or other of the blastomeres, in a latent condition, doing
no harm to the developing ovum. The ovum would surely be
destroyed or at the least be monstrous. Could it conceivably be
present, it is more than debatable whether we could regard
such a fortuitous inclusion as a part and portion of the germ
plasm, and so a strict inheritance. However, I have already,
at the Academy of Medicine in New York, dwelt upon this
subject.1 Suffice it to say that tuberculosis or syphilis of the
new-born must from every valid consideration be an acquired
congenital condition, not an inheritance. And, let me repeat,
only that which is derived from the parental germ plasms is
truly inherited.

It is to the germ plasm, the active matter in the germinal
cells, and to the properties of that germ plasm, that we must
turn in order to gain our basis for any sound theory for inherit-
ance. Weismann has done yeoman's service in emphasizing this
fact. This germ plasm it is which conveys living matter from
generation to generation.

Growth and its Essential Nature. — Now, whatever life is,
the fundamental phenomenon or possession of living matter is
the performance of work coupled with growth — the capacity
manifested by that living matter to assimilate non-living matter
of certain orders, to absorb it, to endow it with like properties,
to convert it into matter like unto itself, into additional living

1 Adami, " Syphilis and the Liver," New York Medical Journal, April 22,
1899.

142 ON VARIABILITY AND ADAPTATION

matter. In other words, difficult as it is to conceive or picture
to oneself the details of the process, growth is essentially a process
of conversion, a chemical process, and any adequate theory bearing
on the phenomena of growth must primarily be along chemical lines.
We are ignorant of what it is in the structure of living matter
that gives it those properties ; we are, if possible, more ignorant
of the physics of the process of growth, of the nature of the
force which, acting upon or inherent in the constitution of living
matter, leads to this continuous process of assimilation and
conversion, and in our ignorance we are unable to separate the
chemistry and the physics of the process ; we must, that is,
regard growth as a property of living matter. We must also for
present purposes speak of the matter which is essential to and
directly concerned in the activity of any one species or individual
as a single substance which, following Nageli, we can refer to
as " idioplasm," and our conception of the individual or of the
separate cell units forming that individual must be that in each
we have to deal with two constituents, the idioplasm, or essential
and directive living matter, and the cytoplasm, which is in the
strictest sense non-living or certainly unable to exhibit the whole
series of vital properties apart from the idioplasm, and which
consists of various formed elements developed and influenced
by the controlling idioplasm, intimately connected therewith, it is
true, but at the same time not an essential part of the same —
the cytoplasm varying in its composition and nature under
varying conditions which affect the idioplasm, the idioplasm
under all conditions retaining certain cardinal properties.

That the constitution of the idioplasm is not absolutely but
only relatively constant has also to be assumed. We are bound
to recognize that it is capable of undergoing modifications
within certain limits without loss of its cardinal properties,
and this from the following reasons : Admitting, as we must,
that the highest forms of animal and vegetable life have been
evolved from the very simplest, that there has been an unbroken
line of development of living forms from the simplest unicellular
to the most complex multicellular ; admitting also that in every
act of reproduction, however simple or however complicated,
the direct conveyance of the living matter of the parent into
the offspring is to be demonstrated, that, in other words, the
idioplasm of the primal living being has been continued on to

A PHYSICO-CHEMICAL THEORY 143

successive and progressive generations ; then we must admit
that the idioplasm of the highest forms, judging from its powers
of controlling and directing the development of the highly
complicated organism, is something very much more complex
than the idioplasm of the unicellular organisms ; that in the
course of evolution this has undergone successive accretions of
properties, and this accretion of properties is the manifestation
and accompaniment of increasing complexity of constitution
of that idioplasm.

This idioplasm must be capable of modification, either by
its environment or under the action of some law of progressive
modification. The fact that there exist to-day forms of life
practically identical in the main details of structure with those
of remote geological ages is against the latter view ; we must
hold that environment determines changes in the constitution
of the idioplasm.

Chemical Theories of Inheritance : the Isomeric Theory. — It
must now be asked : Can we imagine a chemical substance so
constituted as to be capable of modification in its molecular
constitution, and so, in sundry of its properties, without under-
going complete change, without other properties being lost ?

We can. If, as Professor Walker of McGill points out to me,
such relatively simple bodies as lactic and malic acids are capable
of existing in more than one form — as laevogyrous and dextrogy-
rous and optically inactive modifications — and this with no obvious
alterations in their chemical properties (though physically they
show different actions upon polarized light, and physiologically
we find that certain bacteria can act upon one and not the other
form), then, certainly, the much more complex proteid material,
which would seem surely to be the basis of the idioplasm, may
present a very great number of molecular arrangements, and
each of these may be accompanied by slight differences in certain
physical and physiological properties.

Certain recent observations have rendered this additionally
probable. By our present method of conceiving and visualizing
the structure of chemical substances, we regard the carbon
atom in the molecule as being situated at the centre of the
tetrahedron and as capable of linking four other atoms or groups
one to each of the corners or apices of the tetrahedron. A sub-
stance like malic acid having a single central carbon atom may,

144 ON VARIABILITY AND ADAPTATION

therefore, be built up in three ways, with the linked groups
arranged in one relationship to each other, or in the reversed
relationship, or the substance may be a combination of the
two. These are all the forms we can picture as existing
according to this system. But now a fourth form of malic
acid has been found which is more dextrogyrous in the
polariscope than is ordinary dextrogyrous malic acid. And
the salts of this new form of malic acid show slight but distinct
variations, in solubility, etc., from the ordinary malates. If
this is so, there may yet be further molecular modifications of
such a relatively simple substance as malic acid, and a fortiori
idioplasm (with its multitudinous carbon atoms) may be capable
of an enormous number of modifications.

The Side-Chain Theory of Inheritance. — The mode of the
atomic arrangement in the idioplasmic molecule may therefore,
in part, explain the variation in the properties of that idioplasm
seen throughout the animal and vegetable kingdoms. I say
in part, for if we assume that the structure of the individual is
primarily the outcome of the structure and properties of the
idioplasm, then for each different form of living being, nay, for
each individual being, we have to assume a different molecular
structure of the idioplasm. Or, otherwise, we have to assume
that the modifications of this idioplasm are infinite in number.
This, it seems to me, asks too much. The matter cannot be
quite so simple. Each molecular modification may play some
part, it is true, but our conception of the structure and modifica-
tion undergone by the idioplasm must be more elaborate. We
must, I think, formulate a theory of structure somewhat akin
to that laid down by Ehrlich 1 in his now well-known theory
of the nature of immunity. We can picture to ourselves the
primitive idioplasm as composed of a mass of material each
molecule of which is formed of a central ring, to which there
can be attached side-chains, and from which sundry side-chains
can be detached without the central ring being destroyed.

This conception, which upon first encounter appears revolu-
tionary and opposed to our ordinary chemical ideas, is, after a
little deliberation, recognized as being but a statement, in
chemical terminology, of what has been for long years the
accepted physiological conception of the nature of protoplasm.

1 Vide Plimmer, Journal of Pathology, v., 1898, p. 489.

THE SIDE-CHAIN THEORY 145

It is, if I may so express it, the fundamental — the essential —
conception of the constitution of living matter ; it follows
logically and inevitably the postulate of assimilation and growth.
Whether we agree, with the majority, to distinguish between
the idioplasm and the cytoplasm of the cell, or prefer to speak
simply of protoplasm, it must, I think, be recognized by all that
we are bound to assume — for the process of assimilation and,
again, in order to explain the variation in structure and pro-
perties of the various cells of the organism — that there is a
central basal substance to which become linked more or less
permanently, or more or less temporarily, those other secondary
substances, which, not in themselves protoplasm, modify the
constitution of the protoplasm as a whole. And these secondary
substances, we see, are necessary for the full manifestation of
the properties of that protoplasm. This conception is, I acknow-
ledge, difficult to harmonize with the prevalent elementary
conceptions of chemical action and chemical constitution ; but
it is, nevertheless, essential for the physiologist and unavoidable.
What is more (for I do not pretend to be a chemist), I learn that
it is not opposed to the more recent deductions of the chemists
concerning the nature and properties of the more complex
carbon derivatives. In other words, it is by no means heterodox
from an advanced chemical standpoint.

Accepting this view, it is not necessary to regard the mole-
cules of idioplasm as at all times presenting their completed
structure, with every side-chain attached. On the contrary,
we are free to conceive the molecules being laid down and being
transmitted in a relatively simple form, some of the side-chains
only becoming attached when the molecules are brought into
certain particular relationships with their surroundings. It is
not necessary, for example, to hold that already in the ovum
there is idioplasm identical in structure with that eventually
present in the muscle fibres or nerve-cells developed from that
ovum. Rather we must hold that in the ovum there is one
common idioplasm of simple type, to which, when distributed
in the various cells derived from that ovum, different side-chains
become attached, according to the relationships assumed by
those cells, so that the cells of different orders are controlled
and formed around protoplasmic or idioplasmic molecules com-
posed of those central rings plus varying series of side-chains.

L

146 ON VARIABILITY AND ADAPTATION

Indeed, I am prepared to go further and to state that the idio-
plasm possessing its full complement of side-chains must be
regarded as ipso facto incapable of initiating cell multiplication.

I base this statement upon the fact, to which I have more
than once called attention during the last few months,1 that
it is only cells which are undifferentiated, or which have reverted
to a simpler, less differentiated form, that are found to undergo
division. Highly differentiated cells never multiply as such.

Inheritance in Unicellular Forms. — If the primitive uni-
cellular individual formed by this controlling idioplasm divides
into two to form two individuals, each half will contain a portion
of the idioplasm, and if the circumstances affecting the mole-
cules of the new generation are the same as those affecting the
parent forms, then this idioplasm will continue to assimilate
to itself non-living matter and to endow it with its constitution
and properties, and the individuals controlled by this idioplasm
will correspond to the parent form and to each other.

But if the circumstances affecting the filial idioplasm vary
from those affecting the parental, then these more unstable and
loosely connected side-chains will be the first to be influenced.
The very act of assimilation (the surrounding medium varying
in its composition) may lead to the substitution of other side-
chains, to slight variation in the composition of the idioplasm.
And the cell or individual developed or controlled by the idio-
plasm will therefore vary from the parent form, while the pro-
ducts of division of this second generation, containing as they
do this modified idioplasm, will exhibit like structural modifica-
tions. Here we have the simplest example of the inheritance of
acquired characters.

There is, however, yet another property which we have to
assume. It may be laid down as one of the great laws of biology
that characters which are of the most recent acquirement are
those which are most unstable, and are first lost ; those which
are oldest are the last to disappear. In man, for example, the
first signs of generalized systemic weakness and imperfect
development show themselves in connexion with those pro-
perties which differentiate man from the animals most nearly
allied, namely, in weakness and arrest of the higher functions

1 See the chapters "On Growth and Overgrowth" and "On the Causation
of Cancerous and other New Growths " in Part III. of this volume.

ROSS'S POSTULATE 147

of the nervous system, or, again, in susceptibility to certain
specific diseases which are peculiarly apt to affect man, relative
immunity toward which on the part of the healthy individual
must be regarded as a comparatively recent acquirement. As
my old teacher, James Ross, of Manchester, was the first, I
believe, to point out, when there is progressive atrophy of the
cells in the cortex of the brain the first motor cells to show signs
of that atrophy are those governing the muscles which differentiate
man from other animals, namely, the opponentes muscles of
the hand.

Hence it has to be laid down that the attachment of these
side-chains, which are recently acquired, is relatively unstable,
that such recent side-chains are most easily detached, so that
the idioplasm, and the organism developed around that idioplasm,
are prone to return to their former condition or lose characters
gained by recent progenitors. The more side-chains become
attached, the more complicated the structure, the more unstable
the equilibrium ; hence the greater the liability to revert. As
Professor Walker points out to me, parallel conditions are to
be recognized in organic chemistry. The old-established view
of the existence of " radicles " to a large extent admits this
principle ; certain central atomic groups in a compound are
seen to be more fixed and not to undergo change when the
attached groups are removed and replaced. Still more closely
allied is what has been observed in connexion, for example,
with aniline. The composition of this is C6H5NH2. Here the
H atoms of the NH2 group can be replaced by other groups,
by methyl (CH3) and ethyl (C2H5) groups, etc., and it is found
that under certain conditions, in carrying out a series of these
replacements, the last group attached is the first to be split off
and replaced.

We see this principle in action in the lowest forms of life.
Contrary to what Weismann has laid down regarding partheno-
genesis and asexual reproduction, it is comparatively easy to
impress new characters upon the bacteria. By subjecting a
growth of pigment-producing bacteria to the action of a tem-
perature just below that which will cause their death, we can
bring about a loss of pigment production, so that the rapidly
succeeding generations are perfectly colourless, but gradually in
the course of time the cultures made from the original (heated)

148 ON VARIABILITY AND ADAPTATION

tube regain the power of pigment production. This may be
in two or three days, or, again, only after several transplantations
at the end of two or three weeks ; and when we remember that a
bacillus divides and so forms a new generation in, on the average,
something considerably less than an hour, it is seen that the
acquired character may be impressed upon the race for some
hundreds of generations.1 The more intense the alteration
to which the bacillus is subjected, the longer and the more
frequently the race is subjected to the altered temperature
conditions, the longer it is before there is a sign of return to the
normal. Translated into the terms of this theory, heat leads
the idioplasm to have certain side-chains, either modified or lost,
and this modification or loss is inherited ; but return to the
normal environment leads the modified idioplasm in the process
of growth and metabolism eventually to assume side-chains of
the same composition, the conditions of growth being similar to
those under which the species originally acquired the side-chain.
Yet another solution, though it is one which I do not favour,
is that not all of the molecules of the idioplasm undergo modifica-
tion ; a minority retain their original constitution and, under
favourable conditions, gradually come to preponderate.

Sexual Conjugation and Inheritance. — How, next, does this
theory bear upon sexual conjugation and its effects, for this is
also met with in many unicellular organisms ? I shall not take
up the explanation of the development of male and female
characteristics ; this is too large a subject, but I would discuss
how fusion of the idioplasm of two individuals affects inheritance.
In that fusion we have a mixture of the two idioplasms, and,
as already pointed out, these idioplasms, a result of the varying
influences which have acted upon them, tend to vary in certain
particulars in their constitution. This fusion must be either a
mere admixture, so many molecules of one idioplasm becoming
admixed with so many molecules of the other, or a true chemical
combination. If we presuppose a mere mixture, we are led
along the lines of Weismann's theory and have to regard the
idioplasm of the individuals of one generation as being composed
of idioplasmic molecules or ids which have been passed down from
the long line of previous generations. I have already pointed
out the difficulties in which we find ourselves if we accept this

1 See the preceding chapter upon the Variability of the Bacteria.

CONJUGATION AND INHERITANCE 149

view. The other view largely, if not entirely, removes these diffi-
culties, nor does it introduce any fresh crop of serious difficulties.
We may regard, then, the idioplasms from the two parent
forms as undergoing a true chemical combination,1 the resultant
idioplasm of the new generation being in truth a new idioplasm
not possessing the identical properties of that of either parent,
but being intermediate, tending in its characters and constitution
toward the constitution of either one or the other according, it
may be, to the number or chemical activity of the molecules of
one or other parent entering into combination. Weismann
supports his view by pointing out that for a certain period
the maternal and paternal materials in the chromosomes of the
daughter nuclei remain distinct, and reaches the bold and
utterly unsupported conclusion that when the chromosomes fuse
to form the irregular network pervading the nucleus, their
constituents nevertheless remain distinct and sharply separable
from those of the other chromosomes. But we have no evidence,
in the first place, that the chromosomes alone contain the idio-
plasm, that chromatin and idioplasm are identical. Everything
points to the fact that the idioplasm is contained in the
nucleus, but we cannot with certainty advance further. It
may be that the chromatin is but the cystoplasmic framework,
the mechanism, as it were, by which the idioplasm is distributed.

If the germ cells of both parents possess certain loosely
attached or unstable side-chains of more recent acquirement,
which are of like nature, there is no sufficient reason why the
protoplasm of the offspring should not also possess them. We
can thus realize how it is that abnormal features present in both
parents may be equally or more prominent in the offspring.
But in general we must recognize that, the parents varying in
different directions, the tendency of conjugation is to preserve
the mean, to bring about an approach to uniformity in the
constitution of the idioplasm of successive generations of one
species exposed to like influences.

I need scarcely say that in these higher unicellular organisms
which present conjugation the cell already presents a nucleus,
and that everything indicates that this nucleus is the controlling

1 [It will be recalled from what was said in Part I. that I no longer accept
this hypothesis of chemical combination. For some years I have replaced it
by that of interchange of side- chains.]

150 ON VARIABILITY AND ADAPTATION

and directive body in the cell — that it is the nuclei which undergo
conjugation — and that, in short, we have to recognize that the
idioplasm of the individual is contained in the nucleus. It is
by no means necessary to conceive that the whole of the nuclear
material is idioplasm, or that the whole of the idioplasm enters
into chemical combination with the molecules derived from
the conjugating germ cell, molecule for molecule. I mention
this in passing to indicate that it is not necessary to assume
that when polar bodies are extruded, the material forming them
is identical with the idioplasm of the remaining nuclear material.
The difficulties in explaining polar bodies and the reduction of
the nuclear substance are no greater by this than by Weismann's
theory. The exigencies of time demand that I should not enter
in detail into the consideration of these subjects. Rather I
must give the broad outlines of the theory and pass on to consider
inheritance in multicellular organisms.

Inheritance in Multicellular Forms. — Every multicellular
organism arises from a single cell, the fertilized ovum, itself in
all sexually produced forms the result of fusion of a male and
female cell, of the ovum and spermatozoon. Studying the
method whereby this one cell gives rise to the adult multi-
cellular organism, we see that by successive acts of division this
one cell gives rise to all the cells of the body. In the course of
this process its nucleus divides and redivides, and this in such
a way that at each division like portions of the nuclear material
pass into each daughter cell. This regular distribution of nuclear
material affords a sound ground for believing that there is an
equally regular and uniform distribution of idioplasm into each
cell. Now I have already pointed out the value of Driesch's
observations in overthrowing Weismann's contention that there
must be a qualitative difference in the idioplasm distributed to
the daughter cells. We have absolutely no ground for believing
in any such qualitative difference ; on the other hand, it is a
legitimate inference that the idioplasm is modified by its environ-
ment. To take the simplest example — a multicellular organism
composed of a spherical mass of cells. Those cells which come
to form the peripheral layer of the individual are subjected to
a series of reactions quite different from the series telling upon
the centrally situated cells, and it requires no stretch of the
imagination to predict that in the process of assimilation and

INHERITANCE IN MULTICELLULAR FORMS 151

growth on the part of the idioplasm of these outer cells, that
idioplasm tends to be altered in its constitution as compared
with the idioplasm of the centrally situated cells. So that the
results of division of the peripheral cells, if retained in the same
environment, will tend to produce the characters impressed
upon the parent cells, and, if this subjection to the special set
of conditions is continued and impressed upon this order of cells
for a sufficiently long period or with sufficient intensity, even
when pressed or passing into other environment, the cell genera-
tions will retain these modified conditions, and, for example,
cells of mesoblastic origin will tend to maintain characters
different from those derived from epiblast, and this whatever
the ultimate position of the cells in question.

In fact, as a first law, we may lay down with Driesch that
the structure of the cells in a multicellular organism is a function
of their position, and this because the position of the cell deter-
mines the modification undergone by its idioplasm. As a second
law we may lay down that the greater the change impressed
upon the idioplasm of these cells, and the longer that idioplasm
is subjected to the conditions inducing this change, the more
permanently will the daughter cells exhibit the peculiar altera-
tion in the idioplasm, with consequent modified structure wher-
ever they find themselves in the economy. We have, in short,
to recognize that two orders of forces determine the structure
of every cell in the body : (1) the previous influences acting
upon its idioplasm and causing it to be of a particular chemical
constitution ; and (2) the position in which the cell finds itself,
and the forces acting momentarily and immediately upon its
idioplasm. Or, briefly, these two series of forces are inheritance
and environment, and inheritance and environment determine
the constitution of the idioplasm and the structure of the cells.1

1 To the worker in bacteriology the hesitancy on the part of biologists to
accept environment as a most important factor in originating variation is
almost incomprehensible. Nothing is more remarkable in the study of the
lowest unicellular forms of life than this effect of environment in bringing
about changes in character, changes which not only tell upon a limited series
of generations (as I have already pointed out) but are permanent in their
nature. By no means save altered environment could Hansen (Compte rendu
des travaux du laboratoire de Carlsberg, v., 1900, 1 ; Abstr. in Ctbl. f. Bakt.
2te Abt. vii., 1901, 199), taking isolated yeast cells or " spores " (100 per cent
of which, when cultivated under ordinary conditions, gave rise to spore-bearing
forms) and subjecting these to the highest temperature at which growth could
still occur, obtain a race or variety of yeast which now, after twelve years,

152 ON VARIABILITY AND ADAPTATION

Following this line of thought, we can understand how it
comes to pass that the body cells and the germ cells in the
higher and more complex organisms become so sharply divided,
how it is that the body cells are no longer able to reproduce the
whole organism. Their idioplasm has become altered in certain
directions to such an extent that it is able, under favourable
conditions, to divide and reproduce cells of like nature, but the
very extent of the modification it has undergone has taken from
it that constitution or structure which is necessary to allow it
to reproduce each and every order of cells which together form
the individual. The germ cells, on the contrary, as a function
of their position in the organism, undergo no such extensive
changes in constitution, their idioplasm as it grows and is dis-
tributed into the successive germ cells retains its fundamental
constitution with but little alteration, and when these germ
cells are discharged they and their idioplasm, brought into like
relationships to those affecting the parent germ cells, undergo
the like series of developmental changes, and reproduce the
whole series of cells, tissues, and organs characteristic of the
species to which they belong.

In the terms of this theory, therefore, inheritance essentially
depends upon the chemical constitution of the idioplasm or the
life-bearing or biophoric protoplasm of the germ cells, not upon
the number of the separate ids or biophores or ancestral plasms
or pangenes contained in the idioplasm ; and variation, whether
slight and individual, or extensive and leading to the production
of species, is ultimately the expression of modification in the
constitution of that idioplasm brought about by environment.
Whereas Weismann's theory lays stress upon relative fixity in
the constitution of the idioplasm, this theory admits freely the
capacity for change in structure of the same. So long as the

has continued to grow under ordinary conditions without once developing
spores. By no means save altered environment is it possible to explain
Vincent's conversion of the absolutely harmless potato bacillus or, again, the
Bacillus megatherium (by long - continued sojourn within closed collodion
capsules in the peritoneal cavity of animals) into forms profoundly pathogenic,
and fatal to rabbits, mice, and guinea-pigs in the course of a few hours (Ann.
de Vlnst. Pasteur, xii.? 1898, 785).

The argument that phenomena observed in unicellular organisms cannot
be applied to multicellular organisms is, to say the least, a severely strained
argument. The extent to which environment acts as a factor may, it is true,
be diminished in the latter, but surely it cannot be regarded as being eliminated
and rendered negligible.

ATAVISM : MUTILATIONS 153

surrounding conditions are unaltered the idioplasm is unchanged ;
alter these conditions and the idioplasm is liable to variation in
constitution.

Atavism. — Lastly, in regard to atavism and reversionary
degeneration of the cells of the individual, this conception of
the idioplasm with attached side-chains, which are more firmly
or more loosely attached, affords a perfectly adequate explana-
tion. The more advanced the organism, the greater number of
these attached side-chains ; the more recent the attachment of
a side-chain, the more unstable that attachment. As a conse-
quence, any profound disturbance will, according to its intensity,
tend to cause the loosening and removal of the more unstable
side-chains, in general in the order of their stability of attach-
ment. And as the structure of the individual and of the
individual cells is the expression of the constitution of the
idioplasm, of the germinal idioplasm, and of the idioplasm of the
individual cells, so according to the intensity of the disturbance
will there be greater or less reversion to an earlier stage in the
developmental history.

Let us now apply this theory to those special problems
which I have referred to as being of peculiar interest to us as
medical men. Can acquired defects be transmitted ? In
seeking to answer this question, at least three orders of pheno-
mena have to be recognized and distinguished. These are : —

1. The Non-Inheritance of Acquired Mutilations. — To these
I have already referred. We cannot conceive of the direct
transmission of identical lesions of this order from parent to
offspring. At most we can conceive of the possibility of indirect
effect where the mutilation is extensive or affects organs playing
an important part in general nutrition. If there is impoverished
general nutrition we can understand that this can affect the
germ cells along with the other cells of the organism, that the
lack of due assimilation or the excess of sundry internal secretions
which, in consequence of lowered general metabolism, have been
imperfectly neutralized, by telling upon the blood and lymph,
may lead to modification of the idioplasm of those germ cells,
with the possible resultant imperfections in the growth of the
fertilized germ, everything depending here upon the combination
of the modified germ plasm of the mutilated individual with
that of the other parent. If the number of molecules of this

154 ON VARIABILITY AND ADAPTATION

other idioplasm or the constitution of the same is adequate to
counteract the loss of side-chains in the idioplasm of the mutilated
individual, there may be no recognizable effect. Or, on the other
hand, the effect upon this latter idioplasm may be so serious
as to lead to inherited defects in the offspring. But clearly
these defects will be of a different order and a more generalized
type ; they will not be identical with the mutilation. There
will be no direct transmission of acquired defects of this nature.

2. The Indirect Transmission of Acquired Diathesis. — With
reference to diathesis, this also may be laid down, that acquired
disease, and the effects caused by disease, cannot in general
be transmitted in such a way that the offspring presents lesions
identical with those produced in the parent, though it has to be
recognized that there is the possibility of modification in that
offspring due to the parental disease. There is the possibility
of a certain amount of transmission, not of the identical local
lesion caused by the disease in the parent, but of a modified or
impaired condition of the germ plasm. We must recognize
that constitutional disease, by leading to disturbance in the
activity of important organs, tells not only directly upon these
organs, but secondarily upon other organs. It leads, for example,
to an altered condition of the blood, and so to altered nutrition
of all the cells of the body. Among other cells, the germ cells
may be directly affected, their idioplasm modified, and the off-
spring directly influenced. Conditions affecting the parents
are capable of influencing and modifying the descendants. It
is this which is forcibly brought home to us in our medical work.
It is changes of this order which are almost inevitably neglected
by morphologists. for they are not within their ken. The
changes brought about in the tissues by what is essentially
chronic intoxication may be so slight as to be inappreciable.
Microscopical examination may reveal nothing ; only by their
physiological effects can their existence be recognized. It is in
the study of these conditions and their effects that medicine
can afford the most valuable aid in the matter of inheritance.

All infectious diseases are intoxications. If a parent is the
victim of syphilis, it is obvious from febrile and other phenomena
not merely that there are local toxic phenomena at the foci of
growth with multiplication of the germs of the disease, but that
the toxines pass into the general circulation. They may produce

ACQUIRED DIATHESIS 155

no immediate structural changes in the cells and tissues, but we
have evidence that the protoplasm of various tissues is affected,
although the results of the disturbance may only show them-
selves after long years. Indeed, the only explanation we can
give of many remote effects of syphilis is this, that during the
active period of the disease there has been a change wrought
in the constitution of the cells of sundry tissues, slight and
subtle, it may be, but sufficient to lead to premature exhaustion
of the idioplasm of those cells. It would be absurd to argue
that the immature germ cells lie absolutely dormant in the
organism. They need nourishment ; they assimilate ; they
are thus also apt to absorb circulating toxines, and their idio-
plasm must be affected in this act. Hence, while syphilis as
such is not inherited, the toxines of the disease must be regarded
as prone to set up molecular disturbances in the germinal idio-
plasm, and the offspring may show, not syphilitic lesions, but
parasyphilitic lesions — various forms of arrested and imperfect
development of different tissues due to the intoxication, and
therewith modification of the germ plasm while still a portion
of the parental organism.

Parental intoxication, therefore, is seen to be capable of
directly affecting the germ cells, and if there is no direct trans-
mission of the effects of such intoxication, certainly there are
indirect effects. In demonstrating the truth of this statement,
it must be freely admitted that conjugation and intra-uterine
existence introduce grave complications. In fertilization it is
obvious that the idioplasm of the sound parent may largely
neutralise the defective constitution of that of the diseased
parent, while we have constantly to guard against ascribing to
defects in the germ plasm conditions acquired during intra-
uterine life. If the mother is the subject of any toxic state (to
use the broadest possible phrase), not merely may the ovum be
directly affected prior to fertilization, but in the course of foetal
existence the organism of the offspring, deriving its nutrition
as it does from the maternal blood, is liable to be affected and
disturbed by circulating toxic substances, and the development
of the different tissues to be correspondingly influenced. The
only conditions we can safely study are those in which the father
is the subject of disease or intoxication, the mother exempt.
Nor is it easy in cases of infectious disease, or even in frank

156 ON VARIABILITY AND ADAPTATION

intoxications such as the alcoholic, to be perfectly sure regarding
maternal exemption.

Paul's observations, however, upon the effects of lead-poison-
ing afford a most convincing demonstration along the required
lines.1 Plumbism is peculiarly a trade disease. It particu-
larly affects those following certain occupations ; thus often the
male wage-earning member of the family is alone exposed.
Studying the history of thirty-two pregnancies in which the
father was the victim of saturnine poisoning, the mother free
from the condition, Paul obtained the following remarkable
statistics : —

Twelve resulted in death of the foetus before term (eleven
abortions, one child born dead).

Twenty children were born alive, of which eight died during
the first year ; five died during the third year ; one died later ;
two alone were found to be living, one aged twenty years, the
other only twenty-one months.2

In connexion with the effects of paternal syphilis, Fournier 3
has contributed strong evidence along the same lines,
pointing to the great frequency of arrested development of
various orders, from intra-uterine death to infantilism. To his
statistics objection may be made that certain of the infants
recorded by him probably suffered from the actual disease
acquired in utero, secondary to local infection either of the
placenta or of the membranes. Discounting this possibility in
a certain proportion of cases, his figures still remain very re-
markable. But what is needed is a more searching study of
these cases of defective children, the offspring of a healthy
mother and an infected father, to make sure that we are dealing
with parazymotic as distinct from zymotic lesions.4 In con-

1 Paul, Arch. gen. de mid. xv., 1860, p. 513.

2 Legendre, in his Essay in Bouchard's Traite de Pathologie, vol. i. (which in
the first place directed me to Constantin Paul's admirable article), gives a table
of 141 pregnancies which I cannot find in the paper referred to. Possibly
this is from a later article by the same author, but the proportion of abortions
as given by him is so much larger that I doubt if this can be the case.

3 Fournier, quoted by P. Legendre in Bouchard's Traite de pathol. i.
p. 363. Of 103 pregnancies in which the father was syphilitic, the mother
healthy, 41 children were born dead, 19 born definitely syphilitic, 43 children
(not definitely syphilitic) lived but a short period.

4 [It is still a matter of debate as to the relative extent of these inherited,
and acquired, parasyphilitic states. With the discovery of the Treponema
pallidum, which occurred some few years after the delivery of this address,

TRANSMISSION OF ACQUIRED DIATHESIS 157

nexion with the subject, Gheorghiu, studying malformations,
has pointed out the remarkable frequency with which there is,
on inquiry, obtained a history not merely of the mother, but of
either father or mother having been a sufferer from acute or
chronic infectious disease at the time of conception.1

3. The Direct Transmission of Acquired Constitutional
States. — In the above-mentioned series of phenomena we have
dealt, as I say, not with the direct transmission of acquired
conditions, but with the deleterious influence leading to general
defects of development and due to the action of toxic agents
upon germ cells prior to conception. Can we advance further
and see evidences of direct transmission of acquired constitutional
states ? I think we can.

If, for example, an animal acquires immunity to a disease,
we are convinced that the process of acquirement is a chemical
process ; that the action of the toxine of the disease has been
to set up certain molecular changes, certain alterations in the
composition of the cell substance, so that that cell substance now
responds in a different manner when brought into contact with
the toxine, and once this modification in the cell substance is
produced, the descendants of this cell retain the same properties ;
that immunity to one special disease is not merely a momentary,
but is a more or less lasting state of the system. It is true
that it tends not to be permanent ; we see that, where it is
attainable, the more prolonged and the more severe the changes
set up in the process of immunization, the longer it lasts ; we
recognize, further, the action of the law that properties of most
recent acquirement are soonest lost, and that there is a distinct
tendency for the condition, or acquired state, of the cells to pass
away. Nevertheless we admit that inheritance of the acquired
condition has to be granted in the case of the body cells in this
connexion. Here, again, if these processes obtain in connexion
with the body cells, we must logically admit their action in the
case of the germ cells. The idioplasm of body and germ cell is
of like origin, and must be susceptible to like influences.

there has been manifested a tendency to regard all the lesions of so-called con-
genital syphilis as actually of infective nature, due to the transmission of the
Treponema, or Spirochaete from the mother to her offspring. This has not
been proved, and, for myself, I still believe in the existence of certain inherited
parasyphilitic non-infective conditions.]

1 Gheorghiu, UObstUrique, January 1900, p. 63.

158 ON VARIABILITY AND ADAPTATION

The toxines circulating in the blood of the individual under-
going immunization must also affect the germ cells. They
must acquire immunity, and the individuals developed from
them must, subject to the law of loss already noted, have the
same properties. Now, as a matter of fact, this transmission of
acquired immunity has been occasionally noted ; where, for
example, both parents (rabbits) have been rendered immune to
the Bacillus pyocyaneus, the offspring have been found more
refractory to pyocyaneus infection, but in general the observa-
tions have been negative. This, as I have hinted, is only to be
expected on account of the easy detachment of, if I may so
express it, newly-acquired side-chains. It is, however, legitimate
to suppose that successive immunization through several genera-
tions will cause the new side-chains to become more and more
fixed, and that racial immunity is brought about by these means,
a view more probable than the alternative of mere " survival of
the fittest." 1

Conversely, in those cases where immunity is not developed
in the case of chronic conditions like tuberculosis, we can, along
these lines, comprehend how the toxines weaken the germ cells
along the same lines as they weaken the general tissues of the
body ; and as the resistance of the body in general to a special
microbe and its products becomes less and less, so also the
idioplasm of the germ cells becomes less and less resistant, and
so from parental disease the offspring gains a peculiar suscepti-
bility to one special disease. So that, in short, from disease
acquired by the parents, a particular diathesis is developed,
a special susceptibility to the particular form of disease.

Here Weismann would make the somewhat subtle distinction
that we are not dealing with the direct transmission of acquired
parental defects — that the toxines produce these results not by
acting on the body cells, but by direct action on the germ cells
— that the inheritance is blastogenic, not somatogenic.2 This
is a sorry and almost Jesuistic play upon words. Let us grant
that they are of blastogenic origin ; they are nevertheless of
individual acquirement. The individual consists of body plasm

1 It goes without saying that where both father and mother are immunized
through successive generations and the foetus — and its germ cells — acted upon
by the maternal blood and milk, the development of acquired inherited immunity
should become yet more assured.

a Weismann, loc. cit. p. 410.

INHERITANCE OF ACQUIRED IMMUNITY 159

and germ plasm, and whether the defect tell primarily or second-
arily upon the germ plasm of the individual, we have here
examples of conditions acquired by the individual being trans-
mitted to the offspring.

But we can go further. Exogenous and bacterial intoxica-
tions are not the only intoxications. We recognise yearly more
and more the existence of states of truly endogenous intoxication,
auto-intoxications — of disturbed states of the constitution due
to disturbances in glandular activity or to excess of certain
internal secretions, or of the substances ordinarily neutralised
by the same. Such disturbances, acting on the germ cells,
would be truly somatogenic.

If gout and the gouty diathesis are, as many hold them to be,
of the nature of true auto-intoxications, if in a given percentage
of cases (in France 12 per cent, according to Bouchard) the
gouty state shows itself in those giving an absolutely negative
history of gout in their progenitors, then we are at liberty, I
think, to regard the gouty diathesis as an example of truly
somatogenic acquirement of an inherited and inheritable con-
stitutional state.

Defect in body metabolism has led to intoxication of the
germ cells, and the offspring show a peculiar liability to be
the subjects of intoxications of the same order. Here what is
transmitted is a constitutional state, and that constitutional
state may manifest itself in more than one way, but no one will
deny that this is truly inheritance of an acquired condition.

We must therefore, I hold, be prepared to admit the possi-
bility of the transmission in inheritance of certain orders of
acquired constitutional conditions ; we must see that it is not
necessary, with Weismann, to deny strenuously the inheritance
of each and every order of acquired defect, and that along the
lines of some such theory as that here outlined we gain
a fuller harmony between theoretical considerations regarding
the nature of inheritance and the facts as they present themselves
to us day after day.

Conclusion

Within the time at my disposal it has been impossible to
touch upon many aspects of inheritance which interest us
as medical men — upon spontaneous variations and their trans-

160 ON VARIABILITY AND ADAPTATION

mission, upon in-breeding and marriage of consanguines, upon
degeneration as distinct from atavism, upon the particular
problems of inheritance of nervous conditions, to mention
but a few. It seems to me, however, that this conception of
the properties of idioplasm is adequate to bring together and
harmonize the facts we possess concerning all of the above-
mentioned conditions.

Let me conclude with Weismann's apology : " Hypotheses,
even when not absolutely right, may be of value in advancing
our knowledge, if only they are relatively right, i.e. when they
correspond with the state of existing knowledge. They are
like the feelers which the short-sighted snail stretches forth on
its darkened path, testing this way and that, and withdrawing
them and altering the route so soon as they come across any
obstacle." x

I must ask your forgiveness for bringing before you a
subject so far outside the line of general medical thought,
and for having inflicted on you so much that is theoretical.
That was not my intention when I sat down to prepare the
paper. I had intended to indulge in the main in a destructive
criticism, to point out how Weismann's and allied theories fail
to satisfy certain orders of conditions presenting themselves to
medical men ; but as I proceeded with the task it became
obvious that mere destructive criticism was valueless, that it
became imperative to present an alternative theory which for
many months — I may truly say years — had been simmering
within me, unexpressed.

1 Weismann, he. cit., Introductory Note.

CHAPTER III

ADAPTATION AND INFLAMMATION

(1905)

[I republish this chapter as representing an intermediate stage,
one in which, while I recognized fully the fact of the survival of
the fittest, and, at the same time, was convinced that direct
equilibration or adaptation was a very real factor in pathological
processes, I did not see fully how to harmonize the two.

To-day my attitude is expressed in the following theses :

(i.) That individual variation is not primarily due to any
inherent tendency on the part of living matter to vary. On the
contrary, living matter is capable of being varied according to its
environment.

(ii.) That when individuals of a species are exposed to a
particular environment they do not present multitudinous
variants of all orders. On the contrary, alteration of environ-
ment of a particular order gives origin to a particular order or
series of variations, of which that grade will survive and be
perpetuated which represents the most complete equilibration
between the organism and its surroundings.

(iii.) That " chance," it is true, enters in connexion with
the results of gamogenesis, but here the variation from either
parental type due to amphimixis is not progressive : it remains
within the limits of the properties inherent in the particular
species, and what is more, with indiscriminate mating tends to
become less and less marked in successive generations. Gamo-
genesis, that is, tends to preserve the mean, not to encourage

1 Being a reprint of a section of an article upon Inflammation, as revised
in the second edition of Professor Allbutt's System of Medicine, 1905.

161 M

162 ON VARIABILITY AND ADAPTATION

the extreme. It is only those variations (and mutations) induced,
not in one, but in several members of a species by subjection to
a common alteration in their environment, which with gamo-
genesis become progressively more pronounced and are of
evolutionary value.]

It will be seen that the picture of inflammation here given is
very different from the old view in which the dominating idea
was that inflammation is essentially an injurious process leading
to cell — and tissue — destruction. Here we regard the irritant as
capable of causing cell and tissue destruction, and so long as
the irritant is in action so long may this destruction continue.
But inflammation itself we regard as the series and sum of the
reactive processes set up in the tissues, and then bringing about,
not destruction, but the very reverse. Taking as our definition
that inflammation is the response or reaction to injury, we are
inevitably led to see that this response results in counteracting,
or more exactly in tending to counteract, the deleterious effects
of the irritant ; the inflammatory process tends towards repair.
It may not result in repair, for, as we have pointed out in several
instances, too often the reaction is either inadequate or excessive.
The exudation may possess but slight bactericidal powers, or
may be poured out in such quantities that the microbic irritant,
instead of being retained in the region of injury, is conveyed
outside that region ; the wandering cells, instead of destroying,
may undergo destruction ; they may incorporate bacteria, but
not be able to annihilate them ; the fixed cells may either form
an incomplete cicatrix, or continue to proliferate in excess.
Attempt at repair is not repair. Notwithstanding, studying the
various factors involved, it is forced more and more upon us
that each tends in one definite direction, and the sum of the
processes is reparative.

This conception of the process of inflammation has met with
considerable opposition. It is urged that to consider inflamma-
tion as an attempt at repair is teleological, i.e. is to assume that
each reaction in the process is, in itself, purposeful. And,
carrying this objection to its ultimate end, " you conceive,"
say the critics, " that the leucocyte is endowed with intelligence
so that it recognizes in the microbe a foe to the organism ; scents
it from afar ; hunts, seizes, and digests it, and then, its duty

ADAPTATION AND INFLAMMATION 163

done, its mission in life fulfilled, it withdraws its pseudopodia,
and dies contentedly."

It is needless to say that we hold no such views regarding the
intelligence of the leucocyte. At the same time we unhesitat-
ingly regard inflammation as purposeful, every whit as much
as we regard the iris, with its contraction and dilatation under
different intensities of light, as subserving a purpose, or the
acts of feeding and digesting as being with purpose. Inflamma-
tion is a physiological process in so far as it is the calling into
action, in response to accustomed stimuli, of properties normally
possessed by the tissue ; it is a pathological process in so far
that, while the stimuli are in kind not different from normal
stimuli, in intensity they are greater. If we admit physiological
purpose, we must admit pathological. This may, indeed, be
laid down regarding all pathological conditions, namely, that in
them we are dealing not with the effects of new and unaccustomed
factors, but with the ordinary factors telling upon the tissues in
an abnormal way, being either deficient or excessive in their
action. Within physiological limits, the reaction to a given
stimulus is nicely balanced and adequate ; when the stimulus
is excessive, the reaction is liable to be imperfect. The iris
accommodates and adequately protects the retina within certain
limits, but if the eye be exposed to too intense a light, the iris
fails to arrest all the rays and the retina suffers. And so it is
that, in inflammation, when the stimuli are excessive and so
have become irritants, the tendency is for the reaction not to be
perfectly balanced, and the ultimate result to be an incomplete
counteraction of the disturbance. But, to repeat, if we recognize
purpose in the one set of cases, we must recognise it in
the other.

All, it will be seen, depends upon what is our conception of
" purpose " in vital phenomena. That conception is teleo-
logical if and when we regard it as primary — as what may be
termed an intelligent endeavour on the part of the tissue to
accomplish a certain object, a predetermined end. To suppose
that, in inflammation, the vessels dilate and bring about in-
creased exudation in order to flush out the irritant is an utterly
wrong and baseless idea. If, on the other hand, our conception
is along these lines — that in the course of evolution those in-
dividuals survived who, by chance, let us say, happened to

164 ON VARIABILITY AND ADAPTATION

manifest this reaction on the part of their vessels in response
to stimuli of a certain order, whereas those who did not were
more unfavourably placed and so succumbed ; that they con-
veyed the same power to their descendants who also possessed
this advantage over individuals incapable of affording the
reaction ; then we can conceive the development of a race
possessing a mechanism for countering a given stimulus by a
given reaction, a race in which provision is made, or gained, for
dealing with a given order of events ; then it will be seen that
what primarily is accidental becomes secondarily purposeful
through the survival of the fittest and the inheritance of defensive
acquirements. The tissues thus become prepared to respond
to certain alterations in their environment. In other words,
" natural selection " renders what was primarily accidental,
secondarily purposeful.

The whole process of inflammation is an exemplification of
" adaptation," and we would strongly commend the address by
Professor Welch x upon this subject to those desirous of gaming
a right point of view regarding pathological processes in general.
For living matter to survive, it must be adapted to its environ-
ment, and this in the first place happens through inheritance,
through the survival of the fittest, along the lines laid down
above.

Yet in the study of inflammation we are compelled to re-
cognize, not merely inherited, but also individual adaptation.
We cannot otherwise explain why it is that bacteria, which at
first exhibit active local growth within the tissues, become
eventually destroyed on the resolution of the inflammatory
state, unless we acknowledge that the cells, or certain of them,
acquire and, it must be, transmit, increased bacteriolytic and
antitoxic properties. The facts gained regarding the develop-
ment of immunity — the presence in the body-fluids of the
immunized animal of substances inimical to the infective agents,
which are absent, or present in but minimal amounts, in the
fluids of the untreated animal — all demonstrate this individual
adaptation. At the present time, indeed, workers all over
the civilized world are busy in researches bearing upon this very
subject, the production and mode of action of what may be
termed, generically, antibodies.

1 Welch, Amer. Journ. of the Med. Sciences, 1897, cxiii 631.

RESERVE FORCE AND ADAPTATION 165

. Here I can do little more than mention certain general laws
which seem to be at work in bringing about, or favouring,
individual adaptation.

1. The first is that of reserve force. No cell — and no tissue —
normally is in action to the limit of its powers ; on the contrary,
normal activity is far below what the cell is capable of perform-
ing. In other words, normal stimuli do not induce a maximal
reaction, and, therefore, irritants — excessive stimuli — up to a
certain limit, do not overwork, and do not lead to disintegra-
tion of the cell. Perhaps the greatest difficulty encountered by
most students of medicine in accepting the facts of phagocytosis
lies in this, that, regarding the tissues as normally sterile, they
cannot comprehend the assumption of what appear to be totally
new properties by the leucocytes and other cells in inflammation
and infection, namely, the assumption, as they regard it, of
phagocytic powers and the taking up of pathogenic bacteria.
But, as a matter of fact, this is no new property. Throughout
life the cells are engaged in digesting bacteria. We find phago-
cytic leucocytes passing out and free upon mucous surfaces —
any smear or swab from the pharynx will show these leucocytes
with their contained bacteria. As Ruffer,1 Nicholls,2 and
others have shown, bacteria are to be seen in the lymph-glands
and along the lymphatic channels of healthy animals, and these,
most often, within cells. As Ford,3 working in my laboratory,
has demonstrated, using the fullest precautions against con-
tamination, bacteria can be cultivated from the liver and kidneys
of more than 50 per cent of the apparently healthy animals of
the laboratory ; and that the bacteria so obtained are not con-
taminations is proved by the remarkable regularity with which
these organs of each different species present a different bacterial
flora. Having followed Ford's observations and seen the care with
which they were conducted, I cannot accept the observations of
those who refute his work. As Wrosczek 4 has recently shown, if
animals be fed with non-pathogenic germs, or germs setting up
no recognizable intestinal disturbances, colonies of the species

1 Ruffer, Brit. Med. Journ., 1897, i. 1177.

2 Nicholls, Journ. of Med. Research, 1904, xi. 455.

3 Ford, Journ. of Hygiene, 1901, i. 277, with series of tables in Trans.
Assoc. Am. Phys., 1900, xv. 389.

4 Wrosczek, Arch, polonaises d. sc. biol. et mid., 1903, ii. 1796 (Ref. Journ.
de physiol. et path, gen., 1904, vi. 385).

166 ON VARIABILITY AND ADAPTATION

so injected are, later, to be gained from the various organs of
the apparently healthy animals. The tissues are potentially
sterile ; that is, the leucocytes and other phagocytic cells are,
throughout life, engaged in destroying bacteria which have
gained occasional entrance into the tissues.1 When a few
intensely virulent bacteria gain this entrance, or a large number
of a less virulent form manage to grow in some one or other
locality, and thus set up inflammatory changes, the above-
mentioned reserve force in the phagocytic cells comes into play
and permits these cells to take up and digest the greater numbers
or the more toxic forms. It is this circulation of potentially
pathogenic microbes that explains the cryptogenic infections of
internal organs.

2. The second law is that of accustomance. A cell not
actually destroyed by any deleterious agency is apt to become
accustomed to the presence and action of that agency. What is
the basis of this accustomance it is difficult to say, though it is
not difficult to suggest a hypothesis or hypotheses. Here I
simply state that this is an observed law, a law best exemplified
in what has been made out regarding the conversion of a negative
into a positive chemiotaxis.

3. The third and very important law, somewhat closely allied
to the last, is that of habit, or, as Fraser Harris 2 has termed
it, " vital inertia." According to this law, when once, through
a given stimulus, a certain series of molecular changes is set up
in a cell, those changes are liable to continue after the stimulus
has ceased, and, if the stimulus be sufficiently strong or sufficiently
often repeated, the cell acquires the habit, or property, of setting
in action a particular series of molecular changes after a minimal
stimulation. This is, perhaps, best exemplified in connexion
with our subject in the production of antitoxins. It is found
that, once the diphtheritic toxin, for instance, has stimulated
the cells of the organism to produce antitoxins, that production
continues and is wholly out of proportion to the amount of toxin
introduced ; while, similarly, once an animal has gained full im-
munity against any micro-organism, it only needs the introduction
of that micro-organism into the system to induce an immediate
reaction, whereas previously days or weeks had been required

1 Adami, Journ. Amer. Med. Assoc, Dec. 23, 1899.

2 D. F. Harris, Brit. Med. Journ., 1900, ii. 741.

ACCUSTOMANCE AND VITAL INERTIA 167

for accustomance and counteraction to be adequately developed.
The existence of this law was first recognized by Weigert ; 1
it may be said to form the basis of Ehrlich's side-chain theory
of immunity.

1 C. Weigert, Gesammelte Abhandl. Leipzig, 1906, 1.

CHAPTER IV

THE MYELINS AND POTENTIAL FLUID CRYSTALLINE
BODIES OP THE ORGANISM 1

(1906)

The polarizing microscope, simple as it is with its Nicol's prisms
— the two pieces of Iceland spar which can be turned at various
angles one to the other — has not been a popular instrument in
medical science. I take it that my own experience is that of
other medical men. I can remember well a genial and enthusi-
astic colleague inviting me years ago to spend the evening with
him, when he showed me slide after slide of various substances
exhibiting exquisite figures under the Nicol's prisms. I know
I thought the results too pretty to be useful — that the instrument
was peculiarly well adapted for the use of members of microscopic
societies and other amateurs of microscopy, but for the physician
and pathologist it was at most a toy. I would here recant this
early error and would acknowledge humbly that within certain
limits the polarizing microscope shows itself a most valuable
aid in the detection and recognition of the nature of a class of
substances within the tissues which it is difficult, nay almost
impossible, to recognize by other means.

If you take a section of the fresh adrenal of man or of one
of the animals of the laboratory and examine the cortex under
the ordinary microscope, the parenchyma cells have, as is well
known, the appearance of being in the condition of advanced
fatty degeneration — the cell substance, that is, is seen to be
densely packed with small fatty globules. But, as shown by
Kaiserling and Orgler, examine that section between the Nicol's
prisms, and sundry of the globules exhibit an exquisite black

1 Lecture delivered before the Harvey Society, New York, December 1, 1906.

168

PLATE II

Fig. 1. — From juice of adrenal cortex of guinea-pig seen with crossed Nicol's
prisms. The globules with the black crosses are the doubly refracting myelin
globules — the rest are fatty globules (isotropous). Sketch of appearances seen
with the high power (Leitz, f in.).

Fig. '2. — Human liver juice, expressed from piece of liver tissue acted on for 12
hours by absolute alcohol. Large doubly refractive globules are seen under
the crossed Nicol's prisms, along with abundant fatty globules.

THE MYELINS 169

cross between four illuminated sectors. Smear a little of the
juice of the fresh adrenal on a slide, and these can be examined
more narrowly. Under the ordinary lens that juice is found
filled with pure fatty globules varying in size ; with the crossed
Nicol's prisms a few of these now stand out as illuminated
crosses. Turn the prism round, and what had been crosses
appear indistinguishable from the abundant surrounding fatty
globules (Plate II. Fig. 1).

Here clearly we have not to do with ordinary fats. Neutral
fats and fatty acids under no condition afford these charac-
teristic doubly refractive globules. We are dealing with some
other substance, a substance apparently acted on by water,
for the addition of water to the juice causes the crosses to fade
out ; they disappear also if the preparation be desiccated, as
again rapidly if it be treated with absolute alcohol. By that
apparently they are dissolved, for treat an adrenal with alcohol
and now evaporate that alcohol and at a certain stage these
doubly refractive globules make their appearance to disappear
again as the preparation dries up. Where an adrenal has been
hardened in formalin, minute rod-like crystals take the place of
these globules.

The adrenal is far from being the only organ that affords
them, although it is the organ which in the normal state affords
them in the greatest abundance without previous treatment of
any kind. A common morbid state that often yields them in
great abundance is atheroma of the aorta — one has but to scrape
off a little of the broken-down material in an atheromatous
plaque to find again this association of isotropous fatty and
doubly refractive anisotropous globules. Or, again, pound up
the liver or the spleen or the kidney in absolute alcohol, leave
for a few hours, put a drop or two of the fluid on a slide, and
as the alcohol evaporates these remarkable bodies make their
appearance in relatively large numbers (Plate II. Fig. 2).

What are they and what do they signify %

The answer to that question is rather a long story, and a
round-about-one at that ; nor as yet is it in my power — or any
one's — to tell you its conclusion. The most I can hope to do is
to interest you in the story, to show you mto what by-paths of

170 ON VARIABILITY AND ADAPTATION

science it leads, and to set you guessing, and I hope some of
you more than guessing, as to what is the conclusion thereof.
For the matter appears to open up not a few lines of profitable
investigation. I may, it is true, give you immediately an ap-
parent answer. I can, that is, give these bodies a name. I
may call them " myelin globules," and state that they are the
condition assumed by myelin at a certain phase or under certain
conditions. I question, though, whether this will bring much
comfort. For what is myelin ?

The remarkable fact about myelin is that it has been known
for more than fifty years ; that within a few years of its recog-
nition by Virchow, in 1854, it had been determined that bodies
of the nature of myelin could be gained from practically every
organ of the body, and this often in large amounts, and that,
though this is the case, though the pains of pathologists brought
myelin into the world and pathologists mothered, or fathered,
it, though there is quite an extensive literature on the subject,
it is rarely mentioned in polite medical society. Your writers
of text-books on physiology and pathology studiously pass by
on the other side without apparent recognition of its existence.
This possibly because, had they recognized its existence, they
could but confess their ignorance why it was there and what its
function in the economy. And it has to be confessed that myelin
has shown itself a most elusive substance.

If one takes fresh medullated nerve or brain substance and
teases it out in water, the marrowy matter forms into drops
and masses of irregular rounded contour, and as one examines
these they become altered in shape, throwing out blunt rounded
processes with a double contour. We are clearly not dealing
with ordinary fats. Early in the last century Berzelius ob-
served that the cerebrin which he extracted from brain matter
gave similar bizarre forms ; so, too, Drummond, in 1852, noted
a like phenomenon with the alcoholic extract of brain matter.
I find, indeed, that if fresh brain matter be placed in absolute
alcohol for twenty-four hours the development of these bodies
and processes becomes greatly exaggerated. Place a fragment
under a cover-slip and surround with water, and long processes
are shot out from the mass, curving in a most serpentine and
life-like manner, and from them double contoured droplets
become detached. Virchow, in 1854, called attention to the

THE MYELINS : THEIR PROPERTIES 171

fact that by alcoholic extraction similar bodies could be gained
from other tissues : from the blood, from yolk of egg, from the
ovaries of calves, from the normal spleen, goitrous thyroid and
diseased lungs. And, as in their physical properties they
resembled brain marrow, Virchow gave these the class name of
myelin. Whether he was dealing with one or with several
substances, he could not determine ; he was inclined to the
view that, if not a single substance, he dealt with a group which
chemically were as closely allied as the various albumins. And
Virchow summed up their properties as follows :

1. When brought into contact with water they swell up,
and in doing so exhibit a characteristic morphology, being seen
under the microscope to develop processes of irregular and
often bizarre form, globular, rod-like, or curved on themselves
and variously distorted, exhibiting, as already noted, a double
contour, undergoing changes of shape while under examination.

2. They are easily soluble in hot alcohol, becoming, in part,
precipitated on cooling.

3. They dissolve rapidly in ether, chloroform, and turpentine.

4. They are acted on but slowly and to a slight extent by
weak acids and alkalies.

5. Under the action of strong alkalies they shrink, with
eventual loss of their characteristic properties.

6. Under the action of strong acids they first swell greatly
and then undergo dissolution.

Now, whether recognizable immediately in the tissues or cells
or cell debris so soon as water is added, or recognizable only
after an alcoholic extract has been made of the tissues and such
extract treated with water, bodies conforming to these postu-
lates have been found distributed through the organism. And
as a class they possess, with the limitations already laid down,
the power of double refraction.

The history of the recognition of this last property affords
an interesting example of the way in which valuable observa-
tions may, for long years, wholly disappear from remembrance,
and that because they have been originally given to the world
in the pages of an obscure journal. In 1857, if I mistake not,
a society was established in Germany for the advancement of
medical science, and the official organ of that association was
distributed primarily, and it would seem almost entirely, among

172 ON VARIABILITY AND ADAPTATION

the members of the association. That journal, like the associa-
tion, had but a brief existence. In its thirty-first number,
Geheimrath Dr. Mettenheimer published his observations on
myelin, noting its power of double refraction. The observa-
tions were so important that the elder Beneke, in 1862, reprinted
the article in its entirety. But this happened in a monograph
whose title, " The Presence, Distribution, and Formation of the
Constituents of Bile," did not in the least suggest that it was
concerned largely with this subject of myelin. Thus it is that,
reviewing the literature for the next forty-four years, I have
failed to come across a single reference to Mettenheimer's
work, while Beneke's, after a year or two, appears similarly
to have passed into oblivion until quite recently, when, with
filial piety, the younger Beneke drew renewed attention to his
father's work. In the meantime workers in different branches
of biologic science made, as they thought, the independent
discovery of this property of myelin — Apathy, in 1889-90,
working on the histology of the nervous system, G. Quincke,
the physicist, in 1894, Miiller of Marburg in 1898, and Kaiserling
and Orgler of Berlin, later in the same year ; and of these, from
their writings, each seems to have been supremely ignorant
that any one had been before him in making the observation
that myelin is doubly refractive.

That all substances affording myelin forms can be shown
to be doubly refractive I greatly doubt, or, more accurately,
I would say that there are substances, such as the lecithins,
capable of affording myelin forms of the simplest type with
which so far I have been unable to gain doubly refractive
globules.1 I shall have something to say later regarding these
apparent exceptions.

I have here tabulated the distribution of Virchow's myelin
in the organism as it has been recorded by various observers :

1 Since delivering this address I have found that at least one lecithin (from
egg yolk) gives with water exquisite doubly refractive figures, and that after
repeated solution in chloroform and precipitation with acetone, a procedure
which should remove any dissolved cholesterin. This is in opposition to
Beneke's statement that egg lecithin completely purified from cholesterin gives
no myelin figures. It is possible that Beneke is correct, but if so the process of
separating cholesterin from lecithin must be very difficult.

THE MYELINS : DISTRIBUTION 173

TABLE A. — Showing the Distribution of Myelin Substances

in the Organism

1 . Intracellular Myelin Globules :

A. Physiologic, or associated with normal regressive processes :

Cells of suprarenal cortex .... Kaiserling and Orgler.

Granular cells of corpora lutea . . . Kaiserling and Orgler.

Cells of thymus gland .... Kaiserling and Orgler.

Cells of mucous membrane of gall bladder . Aschoff.

B. Pathologic :

Aortic endothelium, fatty patches . . Aschoff.

Atheromatous patches of aorta . . . Mettenheimer.

Lungs : Alveolar epithelium of newborn . Hochheim.

Bronchial epithelium .... Schmidt.

Diseased lung tissue .... Mettenheimer.
Kidney : Epithelium in fatty degeneration . Albrecht, Lohlein.

Epithelium after arterial ligation . . Albrecht.

Crystalline lens, cataract .... Mettenheimer.
Tumours, cells of many cancers and sarcomas Kaiserling and Orgler.

C. Autolytic :

Lung, alveolar epithelium .... Albrecht, Hochheim.

Kidney and liver cells .... Many observers.

Skeletal and heart muscle .... Dietrich and Hegel.

Morning sputum ..... Miiller and Schmidt.

2. Diffused Myelin : (Impure lecithin ?).

Myelin gained from various tissues by digestion
with alcohol ; brain and nerve tissue, spleen, liver,
egg yolk, blood, mesenteric chyle, pus, etc. . Virchow.

3. Myelin in Secretions : (Also gained by alcoholic

extraction.) ....... Virchow.

Bile.

Contents of small intestine after fatty meal . . Beneke.

With what Order of Substance are we dealing ?

My attention was forcibly directed to this class of substances
through certain observations made in my laboratory by Dr.
Oskar Klotz. Studying the experimental production of cal-
careous degeneration, he noted that if permeable celloidin
capsules, containing oleic acid or neutral fats, be placed in the
peritoneal cavity of the rabbit, on removal after a few days
the contents give a relatively considerable proportion of calcium
salts — salts which had not been there previously, which now
are present in definite excess over the normal calcium contents
of the rabbit's blood and lymph. The only conclusion to be
reached was that in the organism under certain conditions
calcium salts may become fixed by fatty substances ; in other

174 ON VARIABILITY AND ADAPTATION

words, that calcium soaps become formed. From these observa-
tions tie was led to study, histologically and chemically, areas of
pathological calcification in the organism in order to determine
if these afforded indications that the fats play a part in the
process of pathologic calcification, and more particularly if there
were indications of the presence of soaps as an intermediate
stage in the process. He found that outside the body the stain
Sudan iii affords a differential staining between neutral fats and
soaps, globules of the latter taking on a paler, more yellowish
tinge, and that in the zone immediately surrounding areas of
active calcareous deposit he could recognize similarly globules
taking on the deeper stain of neutral fats, together with others
taking on the characteristic tint of soaps.

From these studies Klotz concluded that fats play an active
part in the process of calcification : that, first, the affected
area undergoes necrobiosis with fatty degeneration ; that, next,
calcareous soaps become formed, and, finally, that the fatty
acid moiety of the compounds becomes replaced by the chemi-
cally more powerful phosphoric or carbonic acid, calcium phos-
phate and calcium carbonate being the end products. The
indications were that he had not to deal with a simple calcium
soap, but with a soapy compound containing calcium and, as
he held, a proteid constituent. These stages could be well
followed in that commonest seat of calcareous degeneration,
namely, the aortic wall in the course of arterio-sclerosis. But
now, happening to visit Marburg for other purposes, there
Professor Aschoff pointed out to me that Dr. Klotz's description
of these fatty globules, which were not fat but of a soapy nature,
seen in the atheromatous area, corresponded in many respects
with that of the globules seen by him in the arterio-sclerotic
artery, globules which, as Mettenheimer first showed, and as
Torhorst, working in Aschoff's laboratory, had independently
determined, are doubly refractive. They are, in short, myelin
bodies.

Were Klotz's Soaps and these Myelin Bodies

one and the same ?

It was to the solution of this question that Aschoff and I
directed our attention. We found, in the first place, that the

SOAPS AND MYELINS 175

globules in the atheromatous aorta, which under Nicol's prisms
were doubly refractive, take on the differential stain with
Sudan iii ; that Torhorst's myelin and Klotz's soaps are iden-
tical. The methods for isolating fats and soaps are still so im-
perfect, and the amount of the doubly refractive material in the
atheromatous aorta relatively so small, that chemical isolation
and study appeared hopeless. All that was left was to study
the physical properties of various soaps and lipoid bodies so
as to determine which of them approached most nearly in
properties to the myelins. Thus it was that I undertook a
long series of observations, in association with Professor Aschoff,
beginning with the various simple soaps, to observe whether
they possessed the power of forming these characteristic doubly
refractive globules.

That certain soaps under certain conditions produce myelin
bodies has been known for long x and has been the subject of
study, more especially by Quincke. One has but to take a drop
of oleic acid on a slide and surround it with strong ammonia 2
to obtain immediately a brilliant development of myelin figures,
and, what is more, these figures examined between crossed
Nicol's prisms are doubly refractive (Plate III. Fig. 1). We can
produce " myelin forms " from an ammonia soap. This, however,
is not quite the same thing as reproducing the characteristic
doubly refractive spherules such as we have seen in the tissues of
the organism. Briefly, I may state here that with certain simple
soaps it is possible to gain these spherules, and that by very
simple means, namely, by taking the pure soap with a small
quantity of water on a slide, warming it until it dissolves, and
then as it cools it may be that under the polarizing microscope
a perfect rain of spherules shows itself. In some cases these
persist for hours and, indeed, for days ; in others, depending
on the nature of the soap, they are transient, appearing for a
moment and immediately giving place to a brilliant white layer
of formed crystalline plates.

By this means we determined that simple soaps of oleic acid
give these figures — oleate of ammonia, of sodium and potassium,
so also those of calcium — but here appeared to be a difference :

1 Neubauer in the 'sixties seems to have been the first to make the
observation.

2 The experiment is more striking, as noted by Lehmann, if the ammonia
and oil be coloured by contrasting dyes.

176 ON VARIABILITY AND ADAPTATION

the calcium globules would seem to be relatively solid, the others
relatively liquid. But by no means was I able to gain the
phenomenon with simple soaps of palmitic and stearic acids :
on cooling concentrated solutions they passed immediately into
the true crystalline form, and as these palmitic and stearic soaps
are solid and definitely crystalline at the room temperature, it
seemed evident that the spherules seen in the organism at room
temperature could not be of palmitin or stearin compounds, or
at least could not be simple uncomplicated compounds of the
same. It is possible, I would suggest, that compounds con-
taining palmitic or stearic, along with oleic, acid may be fluid
at room temperature.

Here I trust that you will not misunderstand me. I do not
in the least wish to suggest that the doubly refractive bodies
seen in, or gained from, the tissues are simple soaps of oleic acid.
The very instability of the oleates of ammonia, sodium, and
potassium renders this most unlikely. All I wish to show at
this stage is that we have a group of relatively simple bodies
of known composition, of bodies having these curious physical
properties that are likewise possessed by the myelin of the
organism, and to suggest that a study of these simple cases is
calculated to throw light on the more complicated. This
certainly it has accomplished, and here, before referring to our
studies on more elaborate compounds, it is fitting and timely
that I should indicate how the study of the simple soaps afforded
us what I believe to be a most important clue to the nature and
properties of the myelins in general.

I have already mentioned that, studying strong solutions of
the simpler soaps, these remarkable doubly refractive globules
make their appearance as the solution undergoes cooling. Time
and again the appearance is transient. At one moment the
whole field of the microscope may flash out into a rich clustered
constellation of bright crosses to be followed almost immediately
by complete crystallization of the whole area. This fact alone,
not to mention the doubly refractive nature of the globules,
indicates that they are akin to crystals. Obviously they are
not crystals proper ; the form is not that which we associate
with crystals ; they are globular, not angular ; they have all
the appearances of being fluid bodies and not solid ; in water
they do not dissolve as do ordinary crystals, but swell up and

PLATE III

Fig. 1. — The myelin processes formed by the action of ammonia on oleic acid,
seen under the polarization microscope with crossed prisms.

Fig. 2. — Flowing or ductile crystals of azoxybenzoic acid ethylester (after Lehmann).
Cholesterin oleate is apt to exhibit very small figures.

FLUID CRYSTALS 177

gradually lose their doubly refractive qualities. This notwith-
standing, they may appear as an intermediate stage in the
process of crystallization of the pure oleates out of pure watery
solutions.

What is the Meaning of this Phenomenon ?

Not one of those who had worked on the myelins had even
incidentally touched on this question. Nevertheless the solution
had already been given by the physicists. To one of these,
Professor Schenck, working in the very next institute at Marburg,
we went with our inquiries and found that he had been busied
over this very problem for the past few years. Schenck himself,
it is true, had not discovered the solution ; that was due to
Professor 0. Lehmann,1 of the Technical High School in Carls-
ruhe, who two years ago had entombed his findings in a huge
quarto monograph of 250 pages, a superb example of everything
that a monograph ought not to be — verbose, diffuse, wandering,
abundantly polemic, wanting in anything of the nature of a
table of contents, let alone an index ; in short, wholly mediaeval
save for its profuse and admirable illustrations (which neverthe-
less are devoid of legend or key), and for the valuable facts that
can be dug out of its pages. Schenck, who had been working
independently along the same lines, has given to the world a
luminous description of the whole matter, giving clearly in a
few pages all the important data and conclusions gained from
the researches of Lehmann and himself 2 and his pupils.

The popular impression is that a crystal is essentially a solid
unyielding body. It has indeed been known for long that
metals like lead and gold can, under pressure, be forced through
apertures ; the same is true of solid (or crystalline) sodium, of
wax, paraffin, etc. But the general impression has been that
change of shape in these cases is essentially brought about by
translation, by the minute solid crystals of these substances
gliding one on the other, or even — as shown by my colleague,
Prof. F. D. Adams, in his remarkable observations on the
alteration in shape of marble cubes submitted to great pressure

1 0. Lehmann, Flussige Kristalle sowie Plastizitat von Kristallen, etc.,
Leipzig, Engelmann, 1904.

2 R. Schenck, Kristallinische Flussigheiten unci flussige Kristalle, Leipzig,
Engelmann, 1905.

N

178 ON VARIABILITY AND ADAPTATION

— by actual rupture of the crystals, the separate parts gliding the
one on the other.

Undoubtedly this does happen with certain crystalline
substances, but Lehmann, in 1889, first called attention to
another order of phenomena. If solid cholesteryl benzoate,
which is in the form of colourless crystalline plates, be heated
to the temperature of 145*5° C, it melts into a turbid fluid
having the consistence of olive oil. Heated still further, as
Reinitzer first showed, at 178"5° C. it suddenly becomes a
perfectly transparent fluid. Studying the intermediate stage,
Lehmann found that under the polarization microscope the
turbid fluid, despite its fluidity, exhibited double refraction
with the crossed Nicol's prisms — a property hitherto regarded
as associated with the solid crystalline state only ; heated to
178 '5° C, the field became dark and isotropous, like ordinary
fluids. If the opposite process were now undertaken and the
heated fluid subjected to cooling, the whole field became con-
verted into a mass of doubly refractive spherocrystals, showing
here and there little dark crosses ; cooled further, these gave
place to plates of the solid modification, which plates grew in
size until they occupied the whole field.

In fairly rapid succession other substances were determined
having the same peculiarities : — other compounds of cholesterin
with the fatty acid series, such as cholesteryl acetate, cholesteryl
propionate, and cholesteryl oleate, compounds of oleic acid,
sodium, potassium and ammonium oleate, as also methyl-,
dimethyl-, and trimethylamin oleates, and various paraderiva-
tives of anisol and phenetol. In Table B I have transcribed
the list given by Schenck last year, adding thereto certain com-
pounds found by us to possess the same properties, acknowledg-
ing that it is not complete, and that already numerous additions
have been made during the last few months.

All these bodies become fluid on being heated, but the fluid
examined between crossed Nicol's prisms has still the main
optical features of crystalline substances ; it is anisotropous.
Some of these crystalline liquids are thick like olive oil, some
(like £>-azoxyphenetol) are more fluid than water. Poured into
a vessel the surface becomes level ; in tubes it assumes the
characteristic concave meniscus of fluid bodies, and this although
the constitution is crystalline. Heat to a further degree and

FLUID CRYSTALS

179

the fluid becomes wholly isotropous ; it gains all the physical
properties attributed to, and of what we regard as, a true fluid.

TABLE B. — Potential Fluid Crystalline Substances
(Modified from Schenck, with additions)

Melting

Clearing

Name.

point
in degr.

point
in degr.

Observer.

centig.

centig.

Silver iodide

145

450

Lehmann

j9-Azoxyanisol ....

116

134

Gattermann

^-Azoxyphenetol

137-5

168

Gattermann

p-Azoxyanisolphenetol

93-5

149-6

Gattermann

Azin of jp-Oxethylbenzaldehyde

172

199

Gattermann

Anisaldazin

160

180

Franzen

£>-Methoxycinnamic acid .

170

185-7

Van Romburgh

Condensation product from Benz-

aldehyde and Benzidin .

234

260

Gattermann

Condensation product from p-

Toluylaldehyde and Benzidin

231

300

Gattermann

2?-Azoxybenzoic acid ethylester.

113-5

120-5

Vorlaender, Meyer, and
Dahlem

p-Diacetoxyltolluylene chloride .

124

138

Muench

Hydrocarotinbenzoate

• •

Reinitzer and Lehmann

Cholesterylacetate

90-

114-

Reinitzer and Schonbeck

100

114-4

Cholesterylpropionate .

98

114

Obermueller

Cholesterylbenzoate .

145-5

178-5

Reinitzer

Cholesterylbutyrate .

. .

Adami and Aschoff

Cholesterylstearate

• •

Adami and Aschoff

Cholesterylpalmitate .

• s

Adami and Aschoff

Cholesteryloleate

• •

Reinitzer

Potassium oleate

• •

Quincke and Lehmann

Sodium oleate ....

• •

Quincke and Lehmann

Ammonium oleate

. .

Quincke and Lehmann

Dimethylammonium oleate

• .

Quincke and Lehmann

Trimethylammonium oleate

# a

Quincke and Lehmann

Cholin oleate ....

, #

Adami and Aschoff

Neurin oleate ....

Adami and Aschoff

In this intermediate stage, then, we deal with crystalline
fluids, and the individual crystals are " fluid crystals " — though
here a distinction is drawn bv Lehmann and Schenck between
" flowing " or ductile crystals (fliessende Kristalle) and " fluid '
(flussige) crystals proper (Plate III. Fig. 2). The former we
encounter in crystalline fluids of the thicker, less fluid type, the
latter in the more watery fluids. The former, under the micro-
scope, are of definitely crystalline structure, needle-shaped or

180 ON VARIABILITY AND ADAPTATION

prismatic, but tending to have rounded edges and angles. If the
cover-glass be pressed (as I have repeatedly confirmed) they
become distorted, returning to their original shape when the pres-
sure is removed (Plate IV. Fig. 1). The latter under the ordinary
microscope show no signs of crystalline structure ; they appear
as masses of spherical bodies, capable of distortion, lying in a
singly refracting, isotropous matrix. But one and the same
substance may exhibit both forms ; there is no absolute division
into substances exhibiting purely the one or purely the other,
while if any of these bodies be distributed in an inert fluid matrix
the individual aggregations are peculiarly apt to take on the
sphero-crystalline form, appearing as doubly refracting globules
— identical with those that I have described to you as encountered
in the tissues.

It need scarcely be said that these observations, completely
overturning the older ideas of the properties of crystals, have
from many quarters been regarded as heretical, and have en-
countered violent opposition. I am, as I have said, no physi-
cist, and should therefore not presume to weigh the evidence
that has been tendered against these conclusions. I can only
say that I have seen with mine own eyes that these sphero-
crystals and ductile crystals are capable of distortion (vide Plate
IV. Fig. 1) — that they are not solid. A very natural suggestion
has been made that we have here to deal with substances of two
orders, that, on heating, a purer more crystalline matter separates
out from a more inert fluid menstruum of different constitution.
G. Quincke, for example, has urged that the conditions corre-
spond with those found in emulsions, a fluid " skin " surrounding
and leading to the persistence of the separate globules. Tammann
has compared this stage to what is seen in an emulsion of carbolic
acid in water, which from being cloudy becomes transparent
on heating, and holds that here is not a true crystallization but
a depolarization phenomenon. Lehmann has brought forward
proof that we have to deal with true double refraction and not
a depolarization phenomenon, and he and Schenck have shown
that the phenomena present themselves equally well with
chemically pure substances of this order.1

1 Had we to deal with emulsions, centrifugating should separate the two
constituents, or, passing an electric current through the medium, the suspended
particles, if of different constitution, should gather at one or other pole. Schenck
has shown that with pure substances neither of these events occurs.

NATURE OF FLUID CRYSTALS 181

Here, then, for the first time, we gain a satisfactory physical
explanation for the doubly refractive globules seen in, or ob-
tained from, the organism ; they are fluid sphero-crystals.

Can we peoceed any fuether ?

As I have already stated, the investigation of compounds
presenting this intermediate stage is a recent study, the number
found is increasing with relative rapidity, and it is well within
the bounds of the possible that yet other substances, constituents
of the normal organism, will be found to possess it. But what
is not a little suggestive is that the physicists, without any
thought of physiologic problems, have already noted its existence
in two groups of bodies which are normally represented in the
organism, namely, the cholesterin compounds and the oleates.
And studying the list of crystalline fluids and the temperatures
at which the intermediate stage manifests itself, with the excep-
tion of one group these bodies pass into what, for convenience,
I term the intermediate stage at temperatures far above that
of the room or body ; thus save for that one group they cannot
be responsible. The only group containing members which
afford doubly refractive globules at room temperature is that
of the compounds of oleic acid. These compounds are so unstable
— the oleic acid so readily undergoes dissociation — that it is
hopeless or almost hopeless to gain them in a pure state, and as
a consequence it is not possible to state with precision what are
the points of melting and clearing. But certainly cholesteryl
oleate is viscid and buttery at the room temperature, and at this
can be demonstrated to afford the globules, and the simple
oleates of ammonium, potassium, and sodium may likewise con-
tinue to manifest them at the room temperature,1 although in
general the fields show definite crystals. Here it may be noted
that the medium in which the soaps are present is capable of
modifying the melting-point. To this fact I shall have to return
later. Cholesteryl palmitate and stearate, I have found, both
afford these globules and so exhibit the intermediate stage, but
their melting-point is very definitely higher. We are led thus to

1 Evidently these myelin globules of the organism are not doubly refractive
at blood heat. I could not demonstrate those of the adrenal in a warm room
at the Rockefeller Institute (about 75° F.) until the window had been opened
and the room cooled down.

182 ON VARIABILITY AND ADAPTATION

exclude the simpler palrnitates and stearates from the causes of
the phenomenon in the organism, although the possibility must
not be forgotten that bodies like certain of the lecithins which
contain both oleic and palmitic or stearic acid radicals, may
eventually be found to afford the reaction.

These considerations, therefore, so far as it is legitimate to
carry them, distinctly favour the view that the myelin globules
of the organism are probably of the nature of oleic acid com-
pounds— are soaps of oleic acid of greater or less complexity.

It deserves pointing out that from wholly different considera-
tions, namely, from the point of view of chemical analyses of
myelin-containing substances, the earlier workers have arrived
at conclusions which are approximately in harmony with
ours. Liebreich, for example, analysing nerve matter, deter-
mined that the constituent which was the essential cause of
the myelin figure formation — or otherwise the myelin proper —
was the protagon which he was the first to isolate. With Apathy,
I may add, I have found the myelin figures from nerve matter
under certain conditions doubly refractive. Now protagon
dissociates into lecithin, fatty acids and neurin or cholin, and
while crystalline protagon treated with water merely swells,
affording no myelin figures, if a drop of oleic acid under the
microscope be acted on by a solution of cholin or neurin, I gained
an immediate development of exquisite doubly refractive myelin
figures — as exquisite as when strong ammonia acts on oleic acid.
It is not therefore the protagon as such, but dissociated cholin
oleate or neurin oleate 1 that would seem to be the base of the
myelin formation in nerve matter.

There is another body which separates out abundantly in the
alcoholic extraction of nerve matter, namely, cholesterin. This
in itself does not afford the intermediate state, but its com-
pounds with the fatty acids manifest it, and it is quite possible
that such compounds play a part in the myelin formation seen
in fresh nerve tissues. We owe to the elder Beneke in 1862 the
first recognition of the significance of cholesterin in myelin
production. He showed that while olive oil treated with caustic
potash afforded myelin bodies, the more cholesterin he added
to the oil the better was the result ; that alcoholic extract of

1 There is still some doubt regarding the identity of cholin and neurin.
I have, however, gained a compound of oleic acid with both substances.

PROTAGON : LECITHIN : CHOLESTERIN 183

egg yolk freed from cholesterin gave no myelin figures ; that
ordinary soaps on the addition of cholesterin gave them. In
other words, he produced experimentally impure cholesteryl
oleate, palmitate, and stearate. He called attention to the exist-
ence of cholesterin in tissues affording myelin and concluded
that in animal and vegetable tissues " ohne Cholesterin keine
Myelinformen." Liebreich retorted with his observations on
protagon, and Beneke's work became discredited. Neverthe-
less, Aschoff and I believe that Beneke, if too extreme in his
dictum, was largely right, and that in many situations in the
body the myelin globules are of the nature of cholesteryl oleate.
This, it may be noted, as shown by Hiirthle, is a constant con-
stituent of the blood. In the atheromatous patches in the aorta,
in old cataracts of the eye, and in other necrotic areas, it is a
matter of familiar experience that we encounter fatty globules
along with plates of cholesterin, and here, too, we meet with
the myelin globules. An incidental observation on an aorta
showing early atheroma impressed me greatly. The smear
from a small area of slight softening presented numerous fatty
globules and cholesterin crystals, but no doubly refractive
bodies. I heated it gently over the flame, and on cooling there
appeared abundant myelin globules, more particularly in the
neighbourhood of the cholesterin platelets, which now exhibited
a somewhat corroded appearance. Nay more, we would suggest
that cholesteryl oleate is the source of cholesterin calculi in the
gall-bladder.

Cholesteryl oleate excreted from the blood into the bile (and,
as Virchow showed, myelin is present in the normal bile) may
in the alkaline fluid easily undergo dissociation, the oleic acid
combining to form simple diffusible soaps, the cholesterin being
set free. At the last meeting of the German Pathological
Association at Stuttgart, Aschoff demonstrated the presence of
the doubly refractive myelin globules in the cells of the mucous
membrane of the gall-bladder.

We thus have what I regard as strong presumptive evidence
that at least two groups of oleic acid compounds give rise to the
myelin globules of the organism — the cholin or neurin, and the
cholesterin. There is a third group deserving consideration,
namely, the lecithins. Regarding these I would speak with
considerable caution, while at the same time stating my con-

184 ON VARIABILITY AND ADAPTATION

viction that some of them give origin to myelin globules. These
lecithins, I may remind you, dissociate into fatty acid, glycero-
phosphoric acid and cholin. According to Carbone, in dis-
sociation they may give rise to fatty acids, neutral fats, and
cholesterin. Here once more we have this suggestive appear-
ance of the two substances of widely different chemical nature,
cholin and cholesterin, each of which we have found associated
with the development of myelin globules.

From the tissues there have been gained a di-oleo-lecithin, a
di-stearo-lecithin, and a p almi to-st ear o -lecithin.1 They also, it
will be seen, are fatty acid compounds. They afford " myelin
figures " of a simple type and doubly contoured myelin droplets,
but thus far with the pure substances free from cholesterin I
have been unable to gain doubly refractive globules.2 It is
deserving of note that the two tissues affording myelin in greatest
amounts contain normally the most abundant lecithin, namely,
the brain and the adrenals, and that, as Albrecht has pointed
out, the abundant " myelin " obtained from the red corpuscles
is composed in the main of lecithin. More particularly is it in
connexion with autolysis, and the self-digestion of tissues that
the association of lecithin and the appearance of " myelin bodies"
has of late been commented on from various quarters. Time
forbids that I should enter at length into this most interesting
field. I can only very rapidly point out that if a sterile organ
be kept with all aseptic precautions at the body temperature,
its cells in the course of a few hours exhibit abundant small
rounded irregular bodies in their cytoplasm, possessing double
contour. Coincidently, the nuclei become pale and it can be
made out (Albrecht, Dietrich and Hegel) that the chromatin
becomes discharged into the cytoplasm, the appearance of the
myelin bodies bearing a definite relationship to the discharge
of the chromatin.

Some but not all of these afford obscure double refraction.
As indicated by the accompanying table from Waldvogel and
Mette, while in autolysis of the liver the ethereal extract (of
total fatty substances) is not greatly increased, the fatty acids
and neutral fats in the cell undergo great increase at the expense
of the lecithin, as does also the cholesterin.

1 According to the recent very full studies of Thudichum every true lecithin
contains at least one oleic acid radical. 2 Vide note on page 172.

DISSOCIATION OF LECITHIN

185

TABLE C

No. of Days.

Lecithin.

Fatty Acid.

Neutral Fat.

Cholesterin.

Ethereal
Extract.

0
13
44

11-8
6-82
106

0-52
1-80
3-74

006
0-96
3-61

007
1-80
5-41

13-4
130
15-9

With Dietrich and Hegel, it is difficult not to be forced to
the conclusion that the lecithins with their glycerophosphoric
acid are intimately associated with and derived from nuclear
material with its glycerophosphoric acid ; and that the myelin-
like bodies in the cytoplasm are of the nature of lecithins, which
lecithins undergo subsequent dissociation. My present view —
which I admit is subject to revision — is that it is not the leci-
thins proper that afford the doubly refractive globules, but the
products of their dissociation and more particularly those of
the di-oleo-lecithins, and of the interaction between the cholin and
the cholesterin and the oleic acid. For it must be kept in mind
that oleic is a singularly weak acid, that its compounds are
singularly liable to hydrolysis with liberation of free oleic acid.
As a matter of fact, lecithin warmed with cholesterin affords
with water or dilute glycerin exquisite doubly refractive myelin
figures ; as, again, by acting on pure lecithin (Riedel's lecithol)
with alkalies, the doubly refractive globules can be gained.
Schenck, indeed, lays down that the salts (soaps) of oleic acid
never exist in a chemically pure state ; there is always some
water and alkali mixed, free oleic acid, and free alkali. This
natural impurity is very possibly a feature of importance in
the myelin formation by members of this group.

Let me, in conclusion, state that I have not brought up this
matter of the doubly refractive myelin globules as a method
by which the existence of myelin in the tissues of the body
can surely be recognized. That is by no means the case. The
formation of these globules is no constant reaction. What I
would emphasize is that under certain conditions, not always
easily obtainable, Virchow's myelin can be shown to possess
this very characteristic reaction, and that so, the myelin figures
and myelin globules in the organism must be regarded as of
fatty nature, and more, that these belong to that remarkable

186 ON VARIABILITY AND ADAPTATION

class of substances possessing an intermediate state in which
they are present in the form of fluid or of ductile crystals.

This in itself, while a matter of interest, would not perhaps
be of great importance were it not that associated with this
property, as pointed out by Schenck, is a further one which we
are led to regard as of great significance. Schenck has called
attention to the fact that whereas ordinary crystalline sub-
stances permit of mixture with other substances to a very limited
degree, all the members of this class, even when of widely different
chemical composition, unite in all proportions, the melting-
point of the mixture being then determined by the relative
amounts of the two substances present. If, then, we admit
that more than one substance present in the organism belongs
to this class, it is by no means assured that a given doubly re-
fractive sphero-crystal (in the adrenal, for example) is com-
posed of a single substance. It may be an admixture of two
or more. Nor is this everything. Their power of mixing
with and absorbing 1 other substances is very great, not to say
extraordinary. With water, for example, the oleates, the
lecithins, protagon, and allied bodies do not in the first place
dissolve. They absorb it and swell up. Only after they have
swollen up greatly does solution, or what appears to corre-
spond to solution, show itself, for there is still debate among
the physicists regarding the solubility of these bodies. What
is true of water is true of a large number of other substances ;
oleic acid, for example, and neutral fats are absorbed, and
within certain limits, despite the presence of these foreign sub-
stances, the globules continue to exhibit double refraction.
Such admixture, for example, occurs in the adrenal. Studying
the adrenal juice under the polarizing microscope, one of the
earliest facts that strike the observer is that the globules are
of varying lustre, some bright and clear, others pale, others
faint, just discernible shadows. And, in addition, one finds
aberrant globules. Fig. 2, Plate IV. shows some of the forms I
have seen.

Remembering that these lipoid myelins are widely distributed
through the organism, this power of admixture and absorption
appears to be most significant. To this Albrecht has already
called attention in connexion with the abundant myelin of

1 [Or in modern parlance " adsorbing."]

PLATE IV

Fig. 1. — From the same material as Fig. 2, Plate II., after five days. The
myelin globules were in coherent masses (from partial evaporation of the
alcohol). They exhibit well-marked distortion.

Fig. 2. — Aberrant forms of the myelin globules of human adrenal seen under the
crossed prisms of the polarization microscope.

ADSORPTION AND THE MYELINS 187

the red corpuscles. It is these properties which favour the
action of the erythrocytes as common carriers of the organism.
It is not necessary that diffusible bodies become chemically
combined with the substance of the red corpuscles ; they may
be merely absorbed and easily yielded up when the surroundings
become altered.

Most suggestive of all seems to me the observations of
Albrecht and Dietrich and Hegel on the one hand, that the
myelin of the cells in autolysis makes its appearance in the
cytoplasm coincidently with the loss of the nuclear chromatin —
and our own observation, that outside the body it is possible to
gain union between oleic acid and nitrogenous bases, such as
cholin and neurin. It is true that so far no one has been able
to demonstrate the existence of protein-fatty-acid compounds.
Brucke, Quincke, and Klotz have all concluded that they must
exist. This demonstration of ours of the existence of cholin
and neurin oleate is, I would suggest, a step in this direction.

If fats can be taken into the protein molecule — if the lecithin-
like bodies of the nucleus and the cytoplasm exist there normally
in intimate association with the protein constituents — then
we gain a valuable insight into the most perplexing matter of
fatty degeneration. Fats, that is, appearing in fatty degenera-
tion and necrobiosis, are not necessarily or entirely due to
absorption from the blood and lymph, as Rosenfeld would hold,
but some at least are products of the disintegration of the com-
plex molecules of living matter. With many workers on auto-
lysis we have to recognize a succession of steps from the most
highly organized nuclear materials through the myelins to
neutral fats, fatty acids, and cholesterin. There is a myelinic
preceding the fatty degeneration ; or, more accurately, in true
cell degeneration, as distinct from infiltrative processes, the
disintegration of the cell substance may well be in two stages,
bodies of the myelin type being formed first, and these being
followed, with further dissociation, by the appearance of fatty
bodies of simpler type.

CHAPTER V

THE DOMINANCE OF THE NUCLEUS X
(1906)

There are, it seems to rne, two alternative reasons which should
govern the choice of a topic for discussion at the meetings of
Sections of the British Medical Association : either to afford to
the general medical public an expression of opinion by specialists
upon topics of the time, or, on the other hand, to direct the
attention of the public to matters in which it is well that they
should be interested. These discussions are not merely for the
benefit of the participants ; they are published in extenso in
what has become the organ of the British practitioner throughout
the world ; and this public aspect must be kept in sight, nor
should the debate be allowed to narrow itself into the discussion
of minutiae.

It must be frankly admitted that nuclear function is not
exactly a burning question of the day. Your ordinary medical
man is little concerned about it ; your routine physiologist is
concerned in the main with mass effects ; your pathologist sees,
it is true, certain changes in the nucleus in various conditions of
cell disturbance, but what these changes indicate are scarcely
discussed in his text-books or journals. It is for the second of
the above-mentioned reasons that this topic has been chosen for
to-dav's discussion. Though we have not what has become a
topic of the time, we have a matter which it is timely to bring
forward.

For years individual observers in zoology and botany, cyto-

1 Being the opening address of a discussion on the Physiology and Patho-
logy of the Nucleus at the 74th Annual Meeting of the British Medical Associa-
tion, Toronto, August 1906.

188

THE DOMINANCE OF THE NUCLEUS 189

logists and students of " Entwickelungsmechanik," physiological
chemists and morbid histologists have been recording facts
regarding the nucleus, and these facts brought together point to
the one conclusion that the nucleus is the dominating structure
in the cell — dependent, it is true, upon the cytoplasm, or cell-
body, but nevertheless dominant. The time has come to realize
that general advance lies in a recognition of these foremost
properties of nuclear matter, to recognize the fact that within
the unit, the cell, is the more intimate unit, the nuclear matter,
so that the physiology and the pathology of the future are destined
to be nuclear rather than cellular. Or, to be more exact, while
the cell remains our natural unit, within that cell the modifications
that have taken place must receive their explanation primarily
in terms of nuclear change. Possibly this may seem to be a
matter of little moment to the practitioner. So, I doubt not,
appeared fifty years ago Virchow's insistence upon the all-
importance of the cell. We can but say here that to the thought-
ful man, ever seeking the why and wherefore of things, even if
the ultimate answer is never to be reached, each successive step
onwards towards that ultimate answer is a notable achievement,
and this because each such step affords wider generalizations and
the recognition of a fuller harmony of phenomena.

And there are other and weighty reasons, first amongst which
is the opportunity this choice affords as a means of rapprochement
between the physiologists and pathologists, and, if the remark be
not impertinent, as a means of encouragement to the former.
It is good and natural that these two branches of medicine
should come together. For many years they have tended to
drift apart ; the problems which have interested the one have
had little compelling interest for the other ; and I fear it must
be admitted that there has been a feeling on the part of patho-
logists, and of medical men in general, that the teaching in the
one subject has too often not been in the direct line of preparation
for the study of other branches of medical science. In short,
physiologists were already embarked in the study of mass effects
before the cellular structure of tissues was discovered, and they
had so large a field before them that for long years organs
and their properties occupied their whole attention. Modern
pathology, developing later under the guidance of Virchow, has
been essentially based on the cell theory ; it is the cell and not

190 ON VARIABILITY AND ADAPTATION

the tissue that has formed its unit. Only now are there in-
dications, with the development of finer methods and the relative
completion of the work upon mass effects, that physiologists in
general are by a natural process gravitating from the study of
the tissue, its functions and its chemistry, to that of its com-
ponent cells. Physiology is becoming and must inevitably
become more cellular. And it is peculiarly fitting that we should
inaugurate this discussion in Toronto, in recognition of the
pioneer part played by Professor A. B. Macallum in emphasizing
the importance of cellular physiology. It is no exaggeration
to characterize Professor Macallum's long-continued work upon
the nucleus, its histology and its chemistry, as the most im-
portant series of contributions to medical science that has
proceeded, not merely from Toronto, but from Canada at large ;
no exaggeration to refer to him as the first English-speaking
physiologist to consecrate his activities to work along these lines.

It is a sincere pleasure to me, coming from another Canadian
city, and occupying in this respect the vantage ground of not
being a Torontonian, that I can with propriety direct attention
to a matter in which Toronto is among us facile princeps. I
take it that in opening this discussion I shall perform the greater
service if I devote myself to a rapid review of the various findings
which together compel the conclusion that the nucleus is the
centre of cell activity, leaving it to those who follow me to enter
more particularly into the evidence of one or other order.

Such a general survey is more especially demanded because,
to my knowledge, it has not yet been attempted ; or more cor-
rectly, when attempted what I regard as the inevitable con-
clusions have not been drawn. While individual workers have
demonstrated the controlling powers of the nucleus in one or
other respect, there has been a curious disinclination to bring
the various orders of data together and deduce their full signifi-
cance. But here certain limitations must be introduced. The
activities of living matter are to be divided into two categories,
intrinsic and extrinsic, or vegetative and functional. The
observations which have been made upon the nucleus in con-
nexion with the vegetative activities, with cell multiplication and
reproduction, are very abundant. To discuss these along with
the data bearing upon the role of the nucleus in the functional
activities of the cell would make our debate altogether too

VEGETATIVE ACTIVITIES OF THE CELL 191

diffuse. It has been thought wiser, therefore, to confine ourselves,
save in one respect, to the latter — the functional activities.
Nevertheless, if I have correctly interpreted my duties as in-
troducing the subject in order to place in a clear light the con-
trolling influence of the nucleus in the life of the cell, I cannot
leave these vegetative activities out of account. As opener I
must as briefly as is possible, consistent with lucidity, bring
forward the evidence of nuclear predominance as afforded by
studies upon cell and individual reproduction. It was the
studies upon mitosis that first revealed the high importance of
this constituent of the cell.

We can, perhaps, best treat this section of the subject by
means of a series of theses :

1. The properties which distinguish the individuals of any
race or family from the individual of any other race or family
are to be traced back to the constitution of a single cell, the
fertilized ovum, from which that individual has been developed.

2. There must, therefore, be something in the constitution
of the germ matter of the parent stock which differentiates it
from the germ matter of other stocks. Nay, more, no two
individuals appear to possess germ matter of absolutely identical
constitution.

3. In individuals of gamogenetic origin, resulting from
sexual union, the material contributed to the ovum by the
paternal spermatozoon and the maternal ovum is, physiologically
speaking, of equal value. As demonstrated by Mendel in his
observations upon hybrids, like orders of offspring result whether
the male cell of stock A be employed to fertilize the ova of
stock B, or the female cells of stock A be fertilized bv the male
cells of stock B.1

It is evident, therefore, that matter of like order is contributed
to the fertilized ovum by the two parents.

4. In studying more narrowly the process of fertilization we
find that the only matter contributed correspondingly by both
parents is nuclear matter. Ovum and spermatozoon are cells of
widely different appearance, and the result of fertilization is
that the female cell affords the cytoplasm, or cell substance, of
the fertilized ovum ; the male cell provides the centrosome.

1 In mammals intrauterine existence would seem to introduce a factor of
differentiation.

192

ON VARIABILITY AND ADAPTATION

The nucleus of the fertilized ovum or new individual is formed
of corresponding amomits of nuclear matter (chromatin) from
both parents.

5. Not only is this the case, but most significantly — I shall
take up a probable exception immediately — each supplies a
like number of chromatin loops or chromosomes, and as the
fertilized ovum undergoes development and proceeds to divide
and redivide, the like process of distribution is continued, so
that each separate body cell of the fully developed organism
contains equivalent parts of chromatin of paternal and maternal

origin .

6. We can proceed yet further and recognize that in certain
species at least the chromosomes supplied by or derived from
either parent while pairing with like chromosomes from the

other parent are not
all identical in appear-
ance and size, but varv
among themselves, the
variation being con-
stant ; that is to say,
the same types of
chromosomes are found
in successive genera-
tions of cells. This
peculiar variation, as
has been pointed out
more particularly by
American observers
(Montgomery and

Fig. 13. — Spermatocytes from human testis (after
Moore and Arnold) to demonstrate the various
forms of chromosomes contributed to the future
spermatozoa, and the presence of the various
types in the different cells. These distinct types Sutton), is frequent in
a, b c,d are always found in the human sperm ; t - ^ R

mother cell when undergoing maiotic division. «-*»w»w x±x tuv/ _

which ultimately give
rise to the germ cells. As Moore and Arnold of Liverpool have
just shown, a like constancy is to be made out in the types of
chromosomes seen in the spermatocytes of mammals, even of
man himself. The constancy of the particular varieties present
in individual species suggests that the chromosomes of different
orders possess different properties and determine different
characters, or sets of characters, in the cells to which they are
distributed, and in the individual formed from the aggregation

Fig. 1. — Schema of the stages of fertilization of the ovum, to indicate the
«,„, respective parts played by contributions from the ovum and spermatozoa to
the fertilized ovum or new individual (after Boveri).

(1) Entrance of spermatozoon into ovum. (2) Head of spermatozoon enlarging to form
the male pronucleus, the neck-piece forming the centrosome of the cell. (3) Division
of the centrosome ; male pronucleus undergoing further enlargement. (4) Further
enlargement of male pronucleus, until in (5) the male and female pronuclei are
approximately equal. (6) Formation of equivalent chromosomes in male and female
pronuclei. (7) Formation of nuclear spindle, equivalent chromosomes of paternal
and maternal origin becoming distributed, so as eventually to pass into the two
primary blastomeres of the segmenting ovum. (8) The two primary blastomeres with
their nuclei, each receiving thus equivalent portions of chromatin from both parents.

THE CHROMOSOMES 193

of these cells. In support of this hypothesis are the remarkable
observations, first, of M'Klung of Kansas, and, later, of E. B.
Wilson, of New York, that the spermatozoa of sundry insects
are of two orders, though there is but one type of egg. The one
order of spermatozoa gives rise to males, the other to females,
the difference between the two being in their chromosomes. In
the maturing spermatocytes which give origin to the sperma-
tozoa, either the one set of cells possess an accessory chromo-
some, or, in other cases, a particular chromosome in one half
the maturing spermatozoa is large, in the other half is minute.
To quote M'Klung :

" A careful consideration will suggest that nothing but sexual
characters thus divide the members of one species into two
well-defined groups, and we are logically forced to the conclu-
sion that the peculiar chromosome has some bearing on the
arrangement."

Here we are not discussing sex, and I do but note these
observations in passing. There are other cases, not as yet
fully worked out, in which, as in the aphides, there would appear
to be one type of spermatozoon and two types of ova.

The natural conclusion to be reached from all these data is
that the nuclear matter conveys and determines, or controls, the
inherited 'peculiarities of the individual ; further, the conveyance
is through matter contained in the chromatin loops or chromosomes,
while it may be that these individual loops, varying among them-
selves, determine particular conditions.

What we know concerning the spermatozoon points very
definitely to the conclusion that the groups of chromosomes
distributed to the spermatozoa derived from a single spermato-
cyte are not identical, each spermatozoon receiving only one
half the number of chromosomes proper to the primordial
germ cell, and to the cells in general of any particular species.
The ovum on its part exhibits a like reduction. To inquire
further into this remarkable reduction process would lead us
into the discussion of variation and the Mendelian doctrine.
I do but mention these matters here to call attention to the fact
that not merelv inheritance but variation is seen to be most
intimately associated with the nuclear material, and that, if
we can trust our eyes, the one morphological constituent involved
in and responsible for all cases of inherited peculiarities and

o

194 ON VARIABILITY AND ADAPTATION

gamogenetic variation is included in the nuclear chromatin.
That the other constituents of the cell have an influence or can
have an influence we do not deny. If in the fertilized ovum the
nucleus influences the cytoplasm, so, conversely, the constitution
of the cytoplasm must tell upon the nucleoplasm. The facts in
our possession indicate that the latter is the subordinate process ;
the influence of the nucleus is dominant. This is best indicated
by Boveri's remarkable observation that if the nucleus be
removed from the sea-urchin's egg and the enucleated mass of
cytoplasm be fertilized by the spermatozoon of another species
of echinoderm the resultant larva is of the type of the species
that afforded the spermatozoon, that is, the nuclear material ;
this has conveyed and determined the specific properties of the
individual.

Now, if this be so, it must follow that the nuclear matter
controls all the essential cell activities, and this because, studied
narrowly, it is seen that the morphological properties of a cell are
the expression of the constitution of the cell ; it is the constitution
that determines the properties and functions of that cell. All
are bound together every whit as much as are the properties of
any given salt and the constitution of the same. What is true
of the cell holds also of the multicellular individual ; the specific
properties of the individual are the summation of the properties
of its component cells. If, therefore, nuclear composition dominates
the morphology of the individual cell it dominates, likewise, the
properties of the individual.

It must now be asked : What evidence do we possess estab-
lishing that this is really the case ? That evidence may be
dealt with under many heads. We have to deal with the
evidence afforded by : (1) the natural and experimental enuclea-
tion of cells ; (2) gross changes observed in the nucleus as the
result of cell activities ; (3) the finer changes in the same which
may be seen to follow functional activity ; (4) the histological
changes in the nucleus associated with morbid conditions ; (5)
the chemistry of nuclear and cytoplasmic matter respectively ;
and (6) the ferment actions of the cell and their relationship to
nuclear activity.

Let me now lay before you the main data that have been
gained under each of these headings, and the conclusions that
may reasonably be deduced.

j

FUNCTIONAL ACTIVITIES OF THE NUCLEUS 195

1. The Effects of Removal of the Nucleus

The cell which, like the erythrocyte, undergoes natural loss
of its nucleus may continue to exist for a considerable period,
and during that time actively perform function. The mammalian
red corpuscle, for example, according to W. Hunter, Quincke,
and others, exists from fifteen to thirty days. While it exists
we see no evidence of growth, and certainly it never propagates
itself. The same holds good for cells artificially deprived of
their nuclei ; they do not necessarily undergo immediate dis-
organization ; they can be the seat of certain metabolic activities.
According to Klebs, the enucleated cells of the alga, Spirogyra,
can in the sunlight produce new starch granules ; can, that is,
synthesize starch from the carbon, oxygen, and water absorbed,
the starch thus formed in the sunlight being used up in the dark ;
and this may continue for as long as six weeks. They may
further continue to exhibit motion in response to external
stimuli (Lacrymaria olor, Verworn) ; they may actively ingest
food particles. But, on the other hand, the testimony is unani-
mous that higher metabolic activities are incomplete. Unlike
nucleated portions of a vegetable cell, the enucleated is unable
to develop a cell wall of cellulose. Among protozoa, also,
Verworn has noted that enucleated pieces of foraminifera show
not the slightest capacity to form the internal calcareous skeleton.
If the enucleated cytoplasm of Thalassicola pelagica ingest
foreign particles, it is unable to digest them wholly, and while
the enucleated cytoplasm can develop a new centrosome (E. B.
Wilson), it cannot give rise to new nuclear material. It may
be laid down that if it can form new paraplasmic substances
like starch, it cannot form new cytoplasm and cell substance
proper ; that is to say, it cannot increase in bulk and undergo
cell division and multiplication, or, otherwise, these observations
conclusively prove that the nucleus is essential, not merely for the
vegetative activities, but also for the higher metabolic activities of
the cell and their due co-ordination.

That the nucleus alone, deprived of surrounding cell substance,
cannot regenerate the cell is another matter. It has freely to be
admitted, with Verworn, Boveri, and Lillie, that there must be
a certain minimal quantity of cytoplasm associated with the
nucleus before regeneration can take place. But what this

196 ON VARIABILITY AND ADAPTATION

proves is not that the nucleus is not the dominating portion of
the cell complex, but only that the association of nucleus and
cytoplasm is essential for full cell activity. By the lack of per-
ception of this distinction it may be noted that Verworn's treat-
ment of the whole subject of cell processes is greatly weakened
if not vitiated. His facts prove that nucleus and cytoplasm
are equally essential for the full function of the cell, not that
they are of equal value. We may as well argue that in the
community of bees the individual drone or worker is of importance
equal to the queen, because we find that the queen-bee, if separ-
ated from the rest of the community, is incapable of obtaining
food for herself, and so starves to death. I shall refer later to
what I regard as the right conception of the relationship between
cytoplasm and nucleus.

2. Gross Changes in the Nucleus during Activity

Among these may be noted, (1) alteration in the position
of the nucleus in cases in which there are indications of localized
as distinct from diffuse cell activities, and (2) alteration in size
and shape of the nucleus accompanying active function.

In the animal organism possessing cells with a body which
is small in proportion to the size of the nucleus, examples of
the first order would appear to be rare, though they are not
entirely wanting. Thus Korschelt has shown that in the egg
rays of the water scorpion (Nepa), with their cells having remark-
able nuclei, long branches from two adjoining nuclei send out
processes which come into close proximity. In the space between
these a chitinous deposit gradually shows itself, and when the
mass of chitin is fully formed the processes are withdrawn. In
the plant, movement of the nucleus towards the area of new
formation in the cell is relatively common ; thus when there is
the active formation of a thick cell membrane along one aspect
of the cell, it has been noted that the nucleus becomes eccentric
and approximated to the region of new development. There
is a similar eccentric localization of the nucleus during the
development of root hairs (Haberlandt). I need but mention
instances of the second, namely, of alteration in size — they are
now so well known. The earliest observations were those of
Heidenhain years ago upon the different appearance of the nuclei

FUNCTIONAL ACTIVITIES OF THE NUCLEUS 197

of salivary glands when at rest and after stimulation. In more
recent years we have had the striking observation of Hodge,
confirmed by Gustav Mann, Lugaro, and others, upon the nuclear
alterations in the motor ganglion cells of bees, birds, cats, and
other vertebrates brought about by natural and experimentally
produced fatigue.

These observations also clearly demonstrate that the nucleus
is not merely the vegetative centre of the cell, but is involved in its
functional activities.

3. Finer Changes occurring in the Nucleus during
the Course op Cell Activities

If I am not mistaken it was a native of what we regard as
the youngest of the civilized great countries of the world —
Professor Ogata — who first, in 1883, clearly recognized the
finer nuclear changes associated with secretory activity. He
called attention to the granules, or plasmosomes, appearing in
the nucleus at the beginning of secretory activity — granules
which take on the characters of nucleoli, and pass from the
nucleus into the cell body. In these he held that the zymogen
granules are developed which eventually become (part of) the
protoplasm of the cell. In 1887 Lukjanow made confirmatory
observations. He noticed in the secreting cell outside the
nucleus the agglomeration of little spherules which in form, size,
and reaction to dyes were closely related to certain nuclear
bodies (Kernkorperchen). He drew the cautious conclusion
that " it appears in any case that the hypothesis of a connexion
between the nucleus and the cell body has in itself nothing
improbable — a connexion shown outwardly by certain structural
elements of the nucleus passing over into the cell body, and
there undergoing further change." In the following year F.
Hermann noted the apparent discharge of similar minute globules
in mucous goblet cells during secretion, and also called attention
to the fact that these in staining powers resemble the nucleolus.
These he found were absent from the resting cell. In 1890
Professor Macallum made his first report upon similar phenomena.
He pointed out that in the nuclei of developing ova of Necturus
(the lake lizard, found here in Lake Ontario), as also in that of
the frog, at one stage the chromatin is principally collected in

198 ON VARIABILITY AND ADAPTATION

the form of nucleoli at the periphery immediately beneath the
nuclear membrane. These nucleoli are usually spherical, and
vary somewhat in size. At this stage yolk granules are absent
from the cell. With an indigo carmine dye he found that the
nucleus and cell body stained red, whereas the nuclear bodies
took on a deep blue. At what appeared to be clearly a later
stage, yolk spherules made their appearance, and when this
happened the whole ovum stained blue, the nuclei being dimin-
ished in size. What appeared to be an intermediate stage was
seen in ova in which the nucleoli and the cell substance in their
immediate neighbourhood exhibited a blue stain, while the rest
of the nucleus and the main mass of the cytoplasm still stained
red. It was difficult from these observations to arrive at any
other conclusion than that the nuclear matter becomes differ-
entiated into nucleolar, and that this diffuses gradually through
the nucleus and then into the cell substance, the diffusion coin-
ciding in point of time with the formation of the yolk granules.
Macallum thus regarded the yolk granules as formed by the
union of a derivative of the nuclear chromatin with a constituent
of the cell protoplasm. And we here note that these yolk granules
chemically are composed in the main of lipoid material, of
lecithin, a compound to which I shall refer later. In the pan-
creatic cells Macallum found — and Steinhaus has made similar
observations — that the nuclei possess safranophilous nucleoli,
while the rest of the nucleus with double staining takes on a
deeper red colour of haematoxylin. As the nucleus loses its
safranophilous substance, the cell protoplasm acquires safrano-
philous granules. He concluded that the chromatin of the
nucleus gives rise to a substance prozymogen ; sometimes it is
dissolved in the nuclear substance, sometimes collected in masses
(plasmosomes) ; finally it diffuses out into the cell protoplasm,
there meeting with a constituent of the latter to form the zymogen
proper.

I might proceed to detail a long series of confirmatory obser-
vations by Carlier, by Bensley — made here in Toronto — by
Maximow, Solger, Nicholas, E. Miiller, Krause, Galeotti, Vigier,
Gamier, Greenough, and others, all agreeing — save in minor
details — and all bearing upon the processes seen in gland cells.
All describe the smallest and first seen granules as situated in the
immediate neighbourhood of the nucleus ; described these as

SIGNIFICANCE OF NUCLEOLI

199

identical in character with the plasmosomes or nucleoli seen
within the nuclear membrane, and have observed that as they
pass to a farther distance from the nucleus they enlarge into
definite secretory granules. It is with the exact stages of this
process that there has been and still is some debate : whether
they project as buds from the nuclear membrane or make their
way out from pores opening into the same ; whether they finally
dissolve within the cell or undergo solution when discharged into
the external medium. But Professors Macallum, Carlier, and
Bensley are all here, and I must not further steal their fire. I
would only add that what has been determined in the animal

Fig. 14. — Cells of the salivary gland (after Maximow) to show (a) development of
the plasmosomes within the nucleus, (b) passage of these out of the nucleus into
the surrounding cell substance, with subsequent conversion of the same into
granules or globules. Cell 3 is abnormal, showing exhaustion of nuclear material.
In the cell are globules of various sizes, some still taking on the nuclear stain,
others taking on a contrast stain, while others (c) show a centre taking on the
stain of nuclear chromatin, with outer zone staining the same as the cell secretion.

cells holds for the plant cell also. Thus Torrey has described
a succession of changes in connexion with the nucleus and cell
body in the germinating maize seed associated with the production
of diastase. The processes are of an identical nature : deep-
staining granules are first seen in the nuclei, whence these exude
in small streams into the cytoplasm ; scattered at first through
the cell, these later become collected at that end next to the
endosperm, where they become ultimately dissolved. It is follow-
ing upon their dissociation that the first action of a ferment upon
the cell wall and matrix of the endosperm becomes evident.

Nor is it only in connexion with secretions possessing ferment
action that we have evidence of nuclear function. In plants
Schniewind Thies has observed nuclear changes in the nectar

200 ON VARIABILITY AND ADAPTATION

cells of flowers in the process of the elaboration of nectar. In
animals the curious vacuoles in the nuclei of fat cells which
have been known for several years have more recently been shown
by Shattock to contain and to give the reaction for fat.

These data almost justify us in accepting Claude Bernard's
remarkable prevision of more than a quarter of a century ago,
that the cell substance is the seat of vital expenditure, while in
the nucleus resides the power of organic synthesis. This does
not, however, in our opinion, exactly represent the relationship,
for the nucleus is also the seat of expenditure, nay, appears
often to determine that expenditure. But clearly, the indications
are that the higher syntheses, those associated with growth and those
governing the specific enzyme actions of the different forms of cell,
are determined and initiated by the nuclear matter.

4. The Nucleus in Pathological Conditions

of the Organism

Purposely when passing in review vegetative and proliferative
phenomena I did not call attention to the evidence afforded
by the study of the nucleus in cases of aberrant cell growth. It
appeared advisable to consider the pathology of the nucleus by
itself and from all aspects, and that, more particularly, because
while the normal vegetative activities are not subjects for our
discussion, the abnormal come within our purview. At this
point we have to call attention to the evidence of nuclear domin-
ance afforded (1) by cases of abnormal cell growth, (2) by cases
of disturbed function.

Regarding the first of these I shall be brief.

It may be stated unhesitatingly that the majority of patholo-
gists at the present moment regard neoplasia or blastomatosis
as essentially a condition of aberrant cell growth, brought about
not by the constant stimulus of intracellular parasitism, but by
some primary alteration of cell environment. As a consequence
of such alteration, if I may quote myself, the energies which,
had the cells remained in their normal relation, would have been
devoted to functional, become diverted to vegetative and prolifer-
ative activities. Your active malignant tumour cell has charac-
teristically all the attributes of a vegetative cell, or, as it is usual,
perhaps unfortunately, to express it, is of the embryonic type.

THE NUCLEUS IN PATHOLOGICAL STATES 201

Associated with this we find that the growing tumour exhibits
abundant mitoses, and, what is more, the growth being aberrant,
we find a well-pronounced tendency for the mitoses also to be
irregular. We thus encounter a great variety of changes : (1) dis-
persion of chromosomes in the cell body as the result apparently
of rupture of the threads of the achromatic spindle ; (2) asym-
metrical mitoses; (3) multipolar mitoses; (4) hypochromatosis with
diminution either in the number or in the size of the chromo-
somes ; (5) hyperchromatosis, with increase whether in number
or size of the chromosomes ; (6) associated with degenerative

5 6 8 B

Fig. 15. — Various forms of aberrant mitosis in cells from malignant growths (after
von Hansemann and Pianese). 1 and 2, Dispersed chromosomes ; 3 and 4,
multipolar mitoses ; 5 and 6, asymmetrical mitoses ; 7 and 8, hypochro-
matic mitoses ; 9, hyperchromatic mitosis.

changes and rapidly growing tumours we may encounter the
development of paranuclear bodies (Nebenkerne), sometimes
of large size and modified staining properties, lying in the cyto-
plasm, and clearly derived from the nuclear matter.

The existence of these abnormal nuclear conditions in con-
nexion with tumour growth is most significant. Beyond this
statement, that it is difficult to arrive at any other conclusion
than that there is an intimate relationship between these nuclear
vagaries and the abnormal cell growth seen in malignant tumours,
I feel it is unsafe to venture ; for, as Dr. Bashford has frankly
acknowledged, more advanced hypotheses based upon these
abnormalities have not stood the test of extended investigation.

Turning now to observations upon the nucleus in pathological

202 ON VARIABILITY AND ADAPTATION

conditions other than those associated with aberrant growth,
it may, in the first place, be noted that cases may be recalled
bearing upon the cell when it passes into a latent or dormant
condition. While we cannot go as far as Grawitz and accept
the existence of " slumber cells," in which the nucleus and its
chromatin have become so shrunken as to be invisible, we can,
I think, note that with the arrest of cell function and passage
into an inert state, the nuclei undergo shrinkage, becoming
extremely small and attenuated, as in the fully formed con-
nective tissue, fully formed fat cells, etc.

It is in association with cell irritation and the commoner
acute degenerations that the nuclear changes become most
evident. It is a matter of familiar knowledge that pronounced
changes take place in connexion with cloudy swelling and, to
employ the old term, fatty degeneration, as distinct from fatty
infiltration of the cell. In cloudy swelling which so commonly
accompanies the acute fevers and conditions of intoxication,
we note, more particularly in the cells of secretory glands, that
the nuclei, which in the first stage of irritation may become
more intensely stained, rapidly lose their staining property and
become indistinct, and the cell body becomes filled with granules
of albuminous nature. Stolnikow was apparently the first to
make accurate studies upon the changes that occur in these
degenerative processes ; many others have since noted the same
collection of the chromatin in the region of the nuclear membrane ;
the discharge into the cytoplasm (well seen in the liver cells in
phosphorus poisoning) ; have described these little masses as
first staining like nuclear substances, and later losing the nuclear
stain completely, the cell body becoming filled with shell-like
clear-staining globules. The more recent work of Schmaus and
Albrecht, Lubarsch, and others, has confirmed and extended
these observations, the former observers calling particular
attention to the formation of nuclear buds, as also to the hyper-
chromatosis and karyorrhexis in gradual death of the cells of
various organs. There are, needless to say, other changes seen
in the degenerating cell — pyknosis, or contraction and clumping
of the nucleus and nuclear material ; karyolysis, or complete
disappearance of the chromatin. These are evidently post
mortem conditions (that is, in the cell), and need not here be
considered. From those first mentioned it would seem that the

THE CHEMISTRY OF THE NUCLEUS 203

cell may recover. They represent exaggerated conditions of
normal processes, but where the latter stages show themselves,
regeneration of the cells becomes hopeless.

As to the significance of this discharge of nuclear material,
I shall have a little to say after we have discussed the chemistry
of the nucleus. Professor Ewing and others will, I trust,
discuss the relationship of these modified nuclear discharges to
the intracellular appearances which by many have been regarded
as cancer and vaccine or variolous organisms.

5. The Chemistry of Nuclear and Cytoplasmic Matter

respectively

Here, in studying the chemical composition of the two com-
ponents of the cell, we meet with certain remarkable facts, for
not a few of which we are indebted to Professor Macallum.
There are certain substances of great chemical activity bound
up in the nuclei which are present to but slight extent, if, indeed,
at times they can be recognized in the cell body. Notably is
this the case with phosphorus (Lilienfeld and Monti, Macallum),
as also with " masked " iron — iron, that is, in fairly firm com-
bination, so that it is only loosened and made to respond to the
tests for free iron after having been subjected to preliminary
dissociative treatment. On the other hand, certain substances
found to be present in the cell body are absent from nuclear
matter. Among these, as Macallum has pointed out, are potas-
sium and chlorides. When now we come to study the proteid
contents of the nuclei, we find that these, unlike ordinary proteins
of the cell body, are undigested by gastric juice, and that the
undigested material consists of the nuclear network and its
chromatin and the nucleoli. We owe especially to Kossel's
investigations the explanation of these peculiar features. Cell
nuclei, that is, contain as a main constituent a special group of
proteins — the nucleoproteins. These nucleoproteins split up into
albumen (histon) and nucleins, and it is these nucleins in par-
ticular that resist the action of gastric juice, and further, are
characterized by high phosphorus content. These, like the
nucleoproteins, are of a proteid nature ; upon further decom-
position they yield albumen and nucleic or nucleinic acid, and
can be further broken down into the xanthin bases or purin

204 ON VARIABILITY AND ADAPTATION

bodies. It is more particularly the existence of phosphorus
and these xanthin bases that differentiate the nucleus from the
cell body. How the iron is combined is as yet undetermined.
We know at most from Spitzer's observations that it is the
iron-containing products of dissociation of the nucleoproteins
that retain the oxidative properties. But clearly in the nucleus
we have as essential constituents compound proteins of great
complexity of organization. As Spitzer, Herter, and others have
indicated, the iron is of the utmost importance in bringing about
oxidative processes, while the phosphorus likewise would appear
to favour oxidative changes. These and other chemical considera-
tions tend to the conclusion that nuclear material possesses in itself
potentialities superior to those of any ordinary constituent of the
cell body, and again support the view that the nucleus is the
centre or source of the higher cell activities.

6. The Ferment Actions of the Cell and their
Relationship to Nuclear Activity

Jacques Loeb, indeed, has been led to the conclusion that the
nucleus is the centre of the oxidative processes of the cell, and
the correctness of this view has of late been demonstrated by
his pupil Lillie. It would open up too large a field to detail •
and weigh the data indicating that nuclear matter is the essential
source of those bodies which afford the enzyme actions of the
cell. We would merely note in passing that it is now universally
accepted that much of the cell function — I do not say all — is
the outcome of enzyme action, and I would recall the data already
brought forward to show that in the absence of the nucleus the
higher specific cell activities are at a standstill ; the evidence
also of the relationship of the nucleus to the formation of
zymogens.

Referring to the discharge of plasmosomes or spherules of
nuclear matter into the cell body it may now be asked, What
chemical processes do these indicate ? It is suggestive that
under normal conditions this discharge has been noted in cells
affording specific secretion, and in abnormal conditions accom-
panied by the accumulation in the cell body of modified para-
plasmic granules or globules. It is at least suggestive that in

THE NUCLEUS AND FERMENT ACTION 205

autolysis (the self-digestion of tissues removed from the body
under aseptic conditions) we note a diffusion out of nuclear
chromatin, and following upon this the formation in the cell
body of myelin granules and masses. Everything indicates that
these myelin masses so formed are complex lipoid bodies ; they
contain fatty acids, more particularly oleates, and studying the
composition of what is regarded as the simplest group, the
lecithins, we find that they are compounds of a nitrogenous base
(cholin), with glycero-phosphoric acid and a fatty acid. Where
these make their appearance in the cell undergoing autolysis
(and probably in other conditions), we must conclude that the
glycero-phosphoric acid is of nuclear origin, and, leaving aside
for the moment the question of the seat of origin of the nitro-
genous base, remembering that the nucleus of the ordinary cell
is devoid of fat, we are led to regard these lecithins as com-
binations between matter of nuclear origin and fatty matter
from the cell body. These lecithins are bodies having very
remarkable properties, both chemical and physical ; they have
great powers of holding other substances in solution, and this is
true of all the myelin bodies. It may well be that the suggestive
series of nuclear changes and cell accumulations which we find
in the cloudy and fatty groups of degenerations, represents
successive stages in which the development and dissociation of
bodies of this type play the essential part. In our studies in
Montreal during the last three years on calcareous and fatty
degeneration this matter of the formation of compounds of
albumen and fat has constantly been brought before us. Dr.
Klotz (in this following upon the conclusions of Briicke long
years ago) has brought forward data favouring the view that
direct union may occur between the two ; but he will be the first
to admit that an absolute chemical proof of the existence of such
compounds is singularly difficult to adduce. It is true that,
working with Professor Aschoff at Marburg, we have recently
demonstrated the combination between nitrogenous bases, such
as cholin, and oleic acid, but this is another matter — nitrogenous
bases while built up into proteins are not proteins. But if we
are not as yet wholly certain of the existence of oleates of albumen,
it is a well-ascertained chemical fact that lecithin can combine
directly with albumen to form albuminates. Thus lipoids of
the nature of the lecithins afford us the necessary linkage bodies

206 ON VARIABILITY AND ADAPTATION

between various albumens and between albuminous and fatty
acids.1 As regards their importance in this connexion we
would only call attention to Preston Kyes's remarkable observa-
tions upon the part played by lecithin as complement, or linkage
body, between certain serum proteids and cell proteins and
snake venom. It is interesting to note how almost simultaneously
during the last few months independent workers in Germany,
France, the United States and England, approaching the subject
from widely different points of view, have converged to the same
conclusion — that the lipoids are of singular importance to the
cell and in relationship to metabolic processes. We seem at
the threshold and in its shadow, and see already the light within.
But here at the threshold I must stop.

The Nucleus and Growth

Before closing, however, there is a question which I doubt
not has arisen in your minds, and one which must be answered.
" You arrogate," it will be said, " all these powers to the nucleus.
What part is played by the cytoplasm ? " To this I would
answer that, passing further and further backwards in our
endeavour to comprehend what is life, if we believe in living
matter and that vital phenomena are the expression of the
effects of physical and chemical forces acting upon that matter,
then our ultimate conception of life must be that it is the function,
or the sum of functions, of a special order of molecules. For
convenience, we would term these ultimate molecules of living
matter biophores. However much we strain our imagination it
would seem impossible to conceive the existence within the cell
of two orders of molecules of widely different type, but of equal
value, which, by their interaction, initiate vital processes. We
must premise that there is in each form of life one primal order
of living matter. If so, the biophores must be contained either
in the nuclear matter or in the cytoplasm ; and as we have shown

1 The mucins would seem to represent a parallel group of carbohydrate-
proteid compounds, and the histological observations of Steinhaus, Maximow,
and others demonstrate most clearly that nuclear matter is concerned in their
development : indeed, in goblet cells, according to Steinhaus, there is a total
conversion of the old nucleus of the cell into mucinogen. The figures given by
this observer are of the same order as those afforded by Ewing of nuclear changes
in epithelial cells in connexion with vaccinia and variola.

THE NUCLEUS AND GROWTH

207

that the higher vegetative powers of the cell are intimately
associated with nuclear matter, it is in the nucleus that we must
locate these biophores, and we must therefore regard the cyto-
plasm as composed of subordinate matter, and as having what
must be termed subvital functions.

Now, the simplest conception that we can form of these
biophores — and even in the very lowest forms of life they must
be singularly complex — is that they are rings or rings of rings,
carbon- and nitrogen-containing and of the benzol type. The
only satisfactory conception of growth, by multiplication of
these molecules, is that the
pre-existing rings possess un-
satisfied affinities, and attract
side -chains of various ions,
simple and compound, from
the surrounding media, and
that these become grouped in
a manner identical with the
grouping present in the pre-
existing biophore. In other
words, we must regard the
building of the new biophoric
molecules as obeying laws of
the same order as those which
determine the building of ions
out of a solution to form crystals of a particular form of salt, but
with this difference, that so far we have no evidence of biophores
becoming formed anew save under the influence of pre-existing
biophores — we know no case of spontaneous generation. Thus,
growth demands affinities and side-chain formation on the part of
the biophores. As with evolution the biophoric molecules have
become more complex, we would suppose that ions and radicals
have become attracted and attached not in ring arrangement
but in loose series and loose connexion with the biophores. As
in growth new biophoric molecules are formed in association
with the pre-existing, the result is an inevitable tendency towards
the grouping of the biophores in a central mass surrounded by a
zone of other attracted matter. With the development of such
a complex system the biophoric molecules proper are no longer
in direct and immediate relationship with the outer medium ;

208

ON VARIABILITY AND ADAPTATION

Nucl.

there is interposed between the two an intermediate mass. The
direct attraction of new matter is, in the main, accomplished by
the intermediation of this outer cytoplasmic zone. So that
eventually we reach the stage in which with increasing com-
plexity of organization
Cytopi. the biophoric molecules
proper, deprived of the
outer cytoplasmic zone,
are unable to attract
ions to themselves in
the proper order — these
must first have been
built up into particular
orders of radicals within
the cytoplasm. In other
FlG* 1'# words, the presence of

preformed cytoplasm becomes essential for the continued exist-
ence and growth of the nucleus — of the nuclear-biophoric matter.
Each becomes essential for the continued existence of the cell
as a whole.

This, frankly, is all hypothetical, but it is the hypo-
thesis which seems best to throw light upon and to harmonize
the data we possess regarding the function and the relative
importance and nucleus and cytoplasm respectively. Nay
more, it is in harmony with what we know concerning
the very lowest forms of life, and their imperfect nuclear de-
velopment.

I feel I shall have done some service if I have demon-
strated the dominance of the nucleus and impressed you with
the conviction that the future will see not merely a cellular but
a nuclear pathology and physiology. From the omne vivum ex
vivo to the omne ovum ex ovo and the omnis cellula e cellula of
our predecessors we now reach the omne chromosoma e chromo-
somate of the modern student of development and see before us
surely the conclusion omne biophorum ex biophoro ejusdem generis.
If this be the ultimate conclusion of the investigator, it is
at the same time the point from which chemist and physicist,
anatomist and physiologist, pathologist and physician1 must
start to develop harmoniously, each along his respective line,

1 [And, I would add, zoologist and botanist.]

NUCLEAR AND CYTOPLASMIC RELATIONSHIP 209

their various conceptions of vital processes, and, as the indications
are that these biophores exist in the nucleus, so it is to the
nucleus and its alterations that each of us, whatever his particular
branch of biological science, must apply himself for the fullest
intimate grasp of the succession of changes that take place in
health as also in disease.

I

CHAPTER VI

THE REDUCTIO AD ABSURDUM OF WEISMANN'S THEORY

(1907)

There have of late years been abundant theories of inheritance
brought forward, but none has attracted greater attention than
that of Weismann, none has been so fully worked out ; and as
the distinguished author of the theory in the evening of his life
has given us this theory in a rounded and complete form, which
has, further, been admirably translated,2 it will be well to take
this as a basis.

Weismann accepts, as indeed he was one of the first to em-
phasize, that inheritance is conveyed by germ cells which have
from the first been set apart for reproductive purposes. These
germ cells as such take no part in the building up of the parental
individual of whose organism they form a part ; only when
liberated and fertilized do they proceed to multiply, giving rise
to one set of cells which form the body as a whole, and to another
smaller set — the germ cells. In other words, he holds that the
germ cells contain within them two constituents, the somatic
or body plasm and the germ plasm. The tissues built up bj
the somatic plasm are mortal, subject to death ; the gerrn
plasm is potentially eternal — it is carried onward from generatior
to generation. Here is the first weak point in Weismann's
argument — a false conception which vitiates the whole subsequent
train of thought. The germ plasm is not potentially eternal.
If, as already indicated, the individual human being derived

1 Abstracted from article upon " Inheritance and Disease " in the System
of Medicine by Sir W. Osier, Bart., F.R.S., F.R.C.P., and T. McCrae, M.D.,
F.R.C.P., Philadelphia, Lea Bros. & Co., vol. i., 1907, p. 36.

2 August Weismann, The Evolution Theory, translated by J. Arthur
Thomson and Margaret R. Thomson. London, Ed. Arnold, 1904.

210

WEISM ANN'S MAIN TENETS 211

from a solitary spermatozoon and a solitary ovum can produce
on the average, as already indicated, 85,000,000,000 other
spermatozoa each resembling the primordial spermatozoon in
size and properties, it is clear that — leaving the body cells out
of account — the germ plasm contained in that primordial sper-
matozoon has multiplied itself several thousands of millions of
times ; or otherwise it has in the process of growth assimilated
vast quantities of other material to itself, and rearranged it so
that that new material has come to possess the same constitution,
and, with this, the same properties as itself. With growth there
is constant new formation of molecules forming the individual
cells. In other words, the germ plasm is not eternal ; it is
constantly being renewed. What are the nearest to being potenti-
ally eternal are the chemical and physical properties of the germ
plasm. And this very fact, that growth implies constant re-
arrangement and assimilation of constituent molecules, makes
the chemical composition " eternal " only so long as the assimil-
able material remains the same. To this I shall revert. Weis-
mann, it is true, admits that there is growth ; he does not,
however, realize all that this necessitates.

He next demonstrates very clearly that this heritable germ
plasm is contained in the nuclei of the conjugating ovum
and spermatozoon, and regards as a proved proposition
that the nuclear chromatin is the hereditary substance. This
is present in the germ cells of every species in the form of a
definite number of chromosomes which, in cells destined for
fertilization, is reduced to one-half the number, so that the
single nucleus — the segmentation nucleus — of the fertilized ovum
contains the number of chromosomes characteristic of the cells
of the individual of any particular species. The hereditary
substance of the child is therefore formed half from the maternal,
half from the paternal substance, and as, at each succeeding
cell-division, each of the paternal and maternal chromosomes
doubles by dividing, each cell of the individual contains hereditary
material derived from both father and mother.

This very fact of reduction, he points out, indicates that half
the chromosomes contain all the essential primary hereditary
constituents, or, otherwise, that it is not the chromatin as a whole
that conveys hereditary properties. The rather, this must be
made up of a series of constituents or ids, each of which is repre-

212 ON VARIABILITY AND ADAPTATION

sentative and capable of conveying all the paternal or maternal
properties.

Let us pass back a generation. The germ cells of the parent
are similarly formed by a combination of chromosomes from the
two grandparents. These must be represented, as indeed must
also be the chromosomes of the great-grandparents, etc. What
is more, in the process of reduction he holds that certain chromo-
somes derived from certain ancestors are cast out, others retained,
and that it is the variation in the series of retained chromosomes
that in the main determines the variations between the members
of the same generation. The hereditary substance in the fertilized
ovum thus consists of several complexes of primary constituents
contained in the chromosomes, or " idants," each of which
complexes (an " id ") comprises within itself all the primary
constituents of a complete individual.

Each " id " Weismann conceives as being composed of a
mass of different kinds of parts, each of which bears a relation
to a particular part of the perfect animal, and so to some extent
represents its primary constituents, although there may be no
resemblance between these Anlagen and the finished parts.
These representatives of individual parts contained in each
" id " he terms " determinants ' (BestimmungsstucJce). In
each "id," therefore, there must be as many determinants as
there are regions in the fully formed organism capable of in-
dependent and transmissible variation at all stages of develop-
ment— for the caterpillar stage, for example, as well as for the
butterfly ; determinants even for the egg, because eggs, cater-
pillars, and butterflies are seen to vary independently. If, as
has been noted, the individual hairs on the antennae of insects
are capable of transmissible variation, and if, in man, the con-
formation of particular teeth is inheritable, then determinants
there must be for these individual hairs and teeth. If a little
patch of scales on the butterfly's wing is peculiar to one variety,
for those few scales there must be a determinant.

And lastly, each individual determinant must be made up
of molecules of living matter, or biophores. Such biophores, he
holds, must be larger than any chemical molecule ; they must
consist of groups of molecules, some large and complete, others
simpler and more minute ; so that ultimately it is the interaction
of these biophores — the casting out of some from one parental

IDS, DETERMINANTS, AND BIOPHORES 213

stock, the retention of others in their containing determinants
and ids — that determines variation. No two individuals contain
identical groups of determinants and identical ids, and, as these
control development, therefore no two individuals are identical.
If in reversion the characters of an ancestral form, it may be
hundreds of generations back, are reproduced, this is because
in the process of reduction, followed by fusion, of the two parental
germ plasms, the ancestral ids come to predominate to the
exclusion or casting out of the ids of more recent generations.
Why this should tend to happen in a special order of cases the
theory does not venture to explain.

The reader from these data should be able to apply the theory
to particular cases. It has also to be added that Weismann
holds that the development of the individual from the ovum
proceeds in such a way that by nuclear division it is brought
about that the germ cells are assured of possessing a complete
set of ids, whereas the body cells do not gain this complete set.
There is a qualitative differentiation of the chromatin passing
to what are to be the eventual tissues of one or other order, so
that ultimately the particular determinants find themselves in
control of particular groups of cells, destined to produce specific
tissues or parts of tissues.

There is, we confess, something that savours of mediaeval
scholasticism in this conception, something remote from our
general conception of the order of natural events, and, as a matter
of fact, the whole edifice of ids, determinants, and biophores
collapses when confronted with the findings of physical science.
Weismann's biophores, in brief, composed as he imagines them
of numerous molecules, some of them complex and of large size,
must be smaller than are individual atoms ! We can in certain
nuclei recognize rows of granules forming the individual chromo-
somes— little bodies 0*5 fju or less in diameter — bodies considerably
smaller than the micrococci of suppuration. These Weismann
regards as the individual ids. Each id is, he postulates, made
up of determinants, of which, as each region capable of variation
is supposed to be represented by a separate determinant, there
must in the human id be thousands rather than hundreds. Each
determinant is made up of biophores or ultimate units of living
matter ; each biophore, according to him, consists of a group
of molecules. Each molecule is, we may add, composed of

214

ON VARIABILITY AND ADAPTATION

numerous atoms. Now Lord Kelvin, in his convincing investi-
gation into the size of the molecule of water (by a study of the
thinnest possible films of a bubble), has proved that in a line
0-5 fA in length there could be only about 150 molecules of water.
Let us be generous and compute the ids as being not on the average
0-5 but 1»0 ja in diameter, and, again being generous, let us com-
pute the ultimate molecule of living matter as being only thirty

Fig. 18.

A. Nucleus undergoing mitosis, the "skein" of chromatin broken up into V-shaped loops

or chromosomes.

B. Nucleus undergoing mitosis, in earlier skein stage, in which (as occasionally seen)

the skein is seen to resolve itself into a double row of chromomeres (Weismann's
" ids ") prior to breaking up into loops.

times the size of the molecule of water — from every considera-
tion an absurd underestimate. It will be seen upon calculation
that the id (supposing it to be spherical) can contain only about
as many molecules as presumably Weismann requires for one or
two biophores, or at the most economical rate for a single repre-
sentative determinant ; and not one, or two, or three, but several
hundreds of determinants ought to be compressed into a
respectable id.

We have, in short, the reductio ad absurdum of Weismann's
theory.

Addendum

[In the second edition of my Principles of Pathology, published
in 1910, with the publication of further studies by the physicists
upon the size of the molecule of water, I modified the treatment
of the subject as follows : *

" We employed previously Lord Kelvin's estimate of the size
of a molecule of water, pointing out that, according to his figures,
in the chromomeres, or bead-like granules seen in certain chromo-
somes, which have been taken to represent Weismann's ' ids.'
there could be stretched across the diameter only about 150

1 P. 136.

THE PHYSICAL REDUCTIO AD ABSURDUM 215

molecules of water, and that when the highly complex molecules
of the nucleo-proteins have a molecular weight of not less than
15,000, the number of protein molecules capable of occupying
this diameter must be very much smaller. But, in the opinion
of many modern physicists, this estimate of Lord Kelvin's repre-
sents if anything the maximum and not the minimum possible
size of a molecule of water. Nernst, indeed (Theoretical Chemistry,
translated by Palmer, pp. 348 et seq.), accepts Van der Waals'
calculation, based upon the molecular kinetic theory, that
the magnitude of the molecule (of water) is ^ millionth of a
micro-millimeter (0*000002 //,), or otherwise that along a line
0-5 fju in length there could exist not 150 but 2500 molecules of
water. Even taking this estimate, Weismann's conception is
still outside the limits of the possible, when the huge size of the
nucleo-protein molecule is taken into account as compared
with that of a simple molecule such as water.

" If the biophores are, according to Weismann's conception,
not simple molecules of proteid type, but aggregations of the
same, the determinants, composed of aggregations of biophores,
should be recognizable under the highest powers of the micro-
scope, and the ' id ' formed in the higher animals of thousands
and ten thousands (of determinants) must inevitably be a body
of from 30 to 300 times the diameter of the determinant — so
large, that is, that if it existed it must have been recognized
from the moment the nucleus of the cell was first observed.
And if, as Weismann supposes, the nucleus of the ovum contains
hundreds of ids derived from numerous ancestors, that nucleus
would fill the whole field of the microscope ! Needless to say
this is not the case ; nor, we may add, does the coarseness of the
nuclear structure vary materially according to the complexity
of the animal. Physically, therefore, Weismann's conception
is an impossibility, and, as Weismann has carried the conception
of preformation to its logical outcome, it follows that in demon-
strating the impossibility of his theory we simultaneously destroy
all less fully developed theories of preformation.

" Determinants, in Weismann's sense, cannot exist." *]

1 [And this despite Professor Arthur Thomson's inability to entertain the
view that it is possible to dispense with any postulate of "Representative
particles " {Heredity, 1912, p. 454).]

CHAPTER VII

A LECTURE ON LIFE x

(1905, 1909)

To all of us mortals presents itself the question, Why are we :
what is the meaning of this state of living : what is life ? — to
none more often than to the physician. His duty it is to tend
the lamp of life. Time and again, as day follows after day, he
sees it burning low ; now in danger of being blown out rudely
by some all-unexpected gust ; now guttering down with weak
and weaker flame, flaring up momentarily, but it may be only
momentarily, until the cold blue flame, scarce giving light,
ushers in the darkness of death. His duty it is to keep the lamp
alight and burning brightly as long as possible — to prevent that
darkness. Is it surprising that he asks himself, again and again,
What is life ? — what is this flame to which we minister ?

Think of it ! Think of the countless ages — for we cannot
count them — since man first became capable of abstract thought,
and so became truly man ! Think of the generations that have
come to life, have grown up, and had their day, and passed !
To each in turn this question has presented itself — to each.
And direct answer has been found by none of all of them. And
the generations have consoled themselves by the thought that
this infinite mystery of existence, abysmal, dark, is purposeful ;
that it is the God who has created fife, that He knows, and
that suffices ; He knows and we, His creatures, cannot attain
unto His knowledge. The generations have had those of robust
mind, those whom this philosophy did not wholly satisfy, those
who have asked why this should be. Solomon has succeeded

1 Delivered before the McGill Medical Students' Society in 1905, and
redelivered by request and with additions before the Ottawa Valley McGill
Graduates' Association in March 1909.

216

WHAT IS " LIFE " 1 217

Job, Goethe has followed Kit Marlowe, the author of the City
of Dreadful Night has followed the author of the Rubdiydt, and
Herbert Spencer, Lucretius ; but no one has lifted the veil.

To-night I wish to range myself anew among the inquirers.
I would not say we can lift the veil, for in such inquiries, as in
all other investigations into nature, we can reach but to a certain
point. It seems to me, however, that it is possible to penetrate
farther into the mystery now than ever in the past ages. I hold
it right also that, being endowed with minds, we should strive
to understand all that we can regarding that which is of so deep
importance. All the same, it is with no little hesitation that I
have determined to make this the subject of my talk to you,
and that because, however much we may desire to treat the
purely scientific aspect of the subject, religion and the views
we have imbibed from earliest childhood inevitably obtrude.
It is not given to the majority, as it was to Darwin and to
Pasteur, to separate in all humility their scientific from their
religious lives and thoughts.

I can recall an open-air dejeuner a la fourchette in a little
Parisian courtyard over on the " rive gauche," the sun glinting
through the trees upon the napery and glassware of the table :
a luncheon with Emile Roux, the great pupil of the great master.
The conversation had turned upon Pasteur and his modes of
work and habits of thought. That happened close upon a score
of years ago, but I remember it as though it were yesterday.
And Eoux then traced what long years of intimate fellowship
had taught him were the mainsprings of the great master's
activities. He spoke of his sincerity, his earnestness, the deep-
seated religiousness of his character, his attachment to the
Church, and the beautiful faith which dominated the family life.
It was a revelation to me, and I said as much. " No," said
Roux, " M. Pasteur never alludes to these matters in his writings.
He holds firmly that a man's faith and his knowledge of science
are two wholly different parts of his existence which it is pre-
sumption on his part to try to harmonize. Humbly and not
with pride should we regard our scientific knowledge and acumen .
The facts we have garnered are few compared with the vast bulk
of hidden knowledge ; the deductions we draw from those facts
are at the most to be treated as working hypotheses, liable to be
modified by further accumulation of facts."

218 ON VAKIABILITY AND ADAPTATION

We are, to paraphrase Roux's words and employ Carlyle's
simile, but sticklebacks in a puddle. What can the stickleback
in his insignificant pool know of the workings of the great
universe ? Faith is essential to and inherent in our human
nature : it depends upon and grows upon that which is not
demonstrable : the realm of the spirit is apart from the realm
of science. To presume, with this imperfect knowledge, to test
and criticize revealed religion is futile, to presume upon this
insecure foundation to build up our faith is absurd. Keep
therefore the two apart ; strive ever to gain a deeper insight
into the truths of the natural world, and at the same time nourish
what is spiritual within us ; but do not waste time and energy
in attempting to harmonize things which can only be harmonized
when all is open and all is known,

And Faith, triumphant, ceases to exist,
Transmuted into Knowledge absolute.

Now I am convinced that in this attitude of Pasteur there
is a profound truth. The blatant infidelity of the present day
is, it seems to me, founded upon this vain attempt, this in-
evitable failure, to harmonize knowledge and faith — things
which to-day cannot be discussed the one in terms of the other.
Whether the attempt be made by those deeply religious, or the
reverse, the result is almost equally disastrous. At the most I
would say that the studies of the individual worker upon nature
and natural phenomena must inevitably influence the life and,
through the life, the faith also of that individual. This, however,
is one thing. To go into the market-place — or magazine — and
discourse dogmatically concerning these matters is quite another.

Few, however, have attained unto this philosophy, sound
though it be, and thus it is with some temerity that to-night I
take up this discussion of life. Too few realize that religion is
assuredly not based on matter, or, to put it in another form,
that all things have their spiritual as well as their material aspect.
Let me impress upon you that I have to deal with the material
aspect of life only, and that doing so, while acknowledging its
existence, I do not venture to discuss the spiritual aspect ;
that thus I do not come before you as a materialist : and if to
some who have not reached thus far, if to those who cannot
dissociate the spiritual from the material in living matter, it

SCIENCE AND RELIGION 219

may seem almost impious to probe into the constitution of
living matter, let me reassure them. In the old days it was
accounted to Galileo as an offence against religion that he should
demonstrate that the earth was not the centre of the universe ;
that the sun did not travel round it, but it round the sun. We
all now accept Galileo's teaching, and our religion is in no whit
weakened thereby. Less than a century ago our forefathers
regarded as heretics those geologists who taught that fossils
were the remains of living beings, and that therefore the earth's
age, instead of being an odd six thousand years, as Archbishop
Ussher and others had computed, must be some hundreds of
thousands of years, if not millions. Every one nowadays
accepts the geologist's evidence without thereby being accounted
an enemy of revealed religion. Fear not, therefore. True
religion is unaffected by results of research upon natural
phenomena.

So now to come to my subject — What is life ? In the first
place, if we analyse living matter, or, more accurately, matter
that has been endowed with life, whether we take the most minute
vegetable or the largest animal, from the simplest to the most
complex, we gain one particular order of substances as the result
of our analysis, an order only found in nature in connexion
with matter that has been living, and these substances we
speak of as proteids, or proteins. Save for water and
certain very simple salts of sodium and potassium, these
proteins are the only bodies common to all forms of matter
that have been endowed with life. There are plenty of other
substances which we may gain from certain orders of living
matter — chlorophyll, starches, fats, and so on — but these are not
universally distributed. The proteins are the one order of
substances derivable from all animate bodies. We may express
this in another way by saying that life is immediately associated
with the presence of proteins. This, however, is not absolutely
correct ; we are not convinced that the proteins as such are
actually present in living matter — in fact, when we isolate these
proteins they do not exhibit the properties which we associate
with life : they cannot move, they cannot grow, they are in-
sensitive to stimuli. We only know that dead organic matter
yields proteins. It is more correct to say that life is associated
with the presence of proteidogenous matter — of matter which

220 ON VARIABILITY AND ADAPTATION

in dying, as again in certain of its activities while living, yields
proteins.

So universal, so essential is this association, that clearly
the first step to a comprehension of living matter must be
gained through a study of proteins and their properties,
for the phenomena of living are clearly bound up with the
processes of association and dissociation of bodies of this par-
ticular order. Over more than fifty years the physiological
chemists have been working at the problem of the constitution
of these proteins. At first the problem seemed hopeless. It
was found that they were formed of carbon, hydrogen, oxygen,
nitrogen, sulphur, but the formula of constitution was some-
thing appalling. Common salt, NaCl, for example, consists
of one atom of chlorine (CI) joined to one atom of sodium (Na).
But in these proteins the amount of sulphur to be obtained is so
minute compared with the amount of the other constituents that
obviously the molecules are of enormous size. Take, for example,
one of the proteins which, since it can be obtained in a crystalline
form, must be regarded as among the less complex, namely
haemoglobin, the protein which gives the red colour to the cor-
puscles of the blood. Its molecular composition is somewhere
in the neighbourhood of

H

N

0

Fe

S

1130

214

245

1

2

712

The molecular weight of water, formed of two parts of hydrogen
to one of oxygen, is 16 ; the average molecular weight of the
proteins has been estimated at 15,000, or, otherwise, this molecule
is about 1000 times as weighty as is the molecule of water. I
say in the neighbourhood of these figures, for no two samples
yield identically the same results. It seemed impossible to
think of building up experimentally such hugely complex bodies.
Next it was found that these molecules are, as I may express
it, conglomerates. Haemoglobin, for example, can be split up
into an iron-containing protein, haematin, and an iron-free
globulin ; and in the 'seventies and 'eighties a material advance
was made in the study of the products of splitting up the mole-
cules by the action of acids and of the digestive ferments. The
peptones and albumoses so obtained were found to be less
complex. In the 'nineties Kossel discovered a group of proteins,

THE PROTEINS, THEIR CONSTITUTION 221

the simplest so far obtained — namely, the protones — bodies
allied to the peptones, whose molecules are much simpler. Thus
Sturin, obtained from the sturgeon, is

C H N 0

36 69 19 7

Only yesterday my colleague Professor Ruttan showed me a fat
formed from the hydrocarbon C H which he had built up in
the laboratory ; so here it will be seen that we are within the
limits of possible synthesis. Kossel showed that these can be
broken down into yet simpler bodies of the amino-acid group,
Histidin, Arginin, and Lysin, C6H9N302, etc. In fact, the re-
searches of Curtius, Hofmeister, and Emil Fischer have demon-
strated that the proteins in general are composed of these
amino-acids — that they are compound molecules composed of
these amino-acids joined in series. I do not wish to enter too
deeply into so complicated a subject. I will only say that
the amino-acids are compounds of carbon, hydrogen, nitrogen,
and oxygen that have the structure of fatty acids — bodies
of the butyric and acetic acid group — to which nitrogen-
containing amine molecules (of NH2) have become linked, giving
them the remarkable property of being at the same time basic
as well as acid, so that they can enter into combination at one
and the same time with acids and bases. It is this particular
property that would seem to be at the bottom of their striking
characteristic of forming huge compound molecules. Thus,
suppose we regard them as bricks, having at one end an acid
affinity which will attract and attach a basic body, and at the
other a basic affinity which will attract and combine with an
acid body (or perhaps better as a succession of magnets, attracted
the positive pole of one to the negative pole of the other), it will
be seen how thereby it is possible to build compound molecules
formed of long chains of these amino-acids. And proteins are
bodies of this nature.

This is no longer a matter of theory. The long-continued
studies of the great German chemist Emil Fischer have in the
last few years been crowned with brilliant success. Briefly, he
has not only isolated a long series of these amino-acids, but,
following Curtius, has been able to synthesize them, that is, to
make them in the laboratory from simpler substances ; but also,

222 ON VARIABILITY AND ADAPTATION

having done this, he has succeeded in Unking them together in
series. In one case he linked together as many as eighteen
amino-acid radicles. These polypeptides, as he has termed them,
in appearance, reactions (such as the characteristic Biuret
reaction, which we use commonly to detect the presence of
proteins), and in behaviour towards acids and alkalies, so closely
resemble the true peptones that, to quote Fischer, they must
be regarded as their nearest relatives. And Fischer has gone
further than this. In the Faraday lecture delivered by him in
London in 1907, he announced that one of the bodies artificially
built up by him (I. leucyl- triglycyl- I. tyrosin) has all the pro-
perties of the albumoses — of certain simpler proteins which we
gain by the peptic and tryptic digestion of muscle.

\

/ \> ■ \ d> ^

o

/ C<HTH-CH-COf^ \ \ ^*

c% ' co0^ -

C.H.OH AClp

<*°>*» „ ASPA*TI0

TYROSIN. *

Fig. 19.

Thus at last there has been accomplished the building up of
these specific organic substances which in nature are found solely
as the outcome of life — nay more, form the material basis for
the manifestation of life. It is the most notable achievement
of the new century.

Here let me again emphasize the fact that these proteins
which we are now in a position to build up, or, if you like
the term, manufacture in the chemical laboratory, are inert
bodies. They are not living as we understand the term : the
living matter is not proteid, but proteidogenous. Can we form
a chemical or physical conception of the difference between the
two — between living and dead organic matter ?

"Here is a fact, the meaning of which is of far-reaching
significance. I show you two tubes. Each contains a small

PROTEINS, PEPTONES, AMINO-ACIDS 223

quantity of a white powder — about half a teaspoonful. Each
powder consists of the same elements, oxygen, hydrogen, nitrogen,
and carbon. One is practically harmless ; the other contains
within it the power of death to a thousand men. The one is
quinine, the other aconitia — the alkaloid which makes so deadly
the plant whose flower our ancestors called monkshood, in the
far-off days when the original was often before their eyes. It is
an almost startling fact that in this minute quantity of powder,
hardly visible to those at a distance, there is such a potentiality of
death. Picture to yourselves a thousand men. That which is
in this tube would end the life of every one of them. Here is a
latent power beside which the lightning flash is feeble, and to
which the earthquake might give place, as far as the comparison
depends on lethal certainty.

" But the resemblance in the aspect of these two substances
is not all. As I said, each consists of the same elements — each
is made up of carbon, nitrogen, oxygen, and hydrogen. Each
consists of the elements which compose air and water, with
carbon added. Why is one almost harmless and the other a
most deadly poison ? I might ask the question regarding many
other substances composed of the same elements, but between
these two the resemblance is strikingly close. The answer to
my question may be given, ' It depends upon the chemical
constitution.' True, but this takes us a very little way. When
we discern that the difference depends upon the way in which
the elements are arranged in molecules, and the molecules are
grouped together, we are not much nearer an explanation. We
see a little more, however, when we realize that chemical con-
stitution means that energy is held ' latent ' (as it is said), ready
to be released when the elements form simpler, closer com-
pounds. All vital function of the body depends on a like simpler,
closer union of the elements which make up complex organic com-
pounds. As far as we can see, all the energy which is released in
the animal body, is released in consequence of chemical action
under the mysterious influence of life. Where such closer union
of the elements and such release of latent energy are going on,
the process may be changed entirely by the contact of molecules
of allied constitution, with latent energy on the point of release,
so held as to blend with that which is being set free in the living
tissue. Blending with this, it may augment or oppose. Re-

224 ON VARIABILITY AND ADAPTATION

member that difference in chemical constitution means difference
in the readiness with which the elements separate and reunite
and release their energy. Remember also that minute differences
in constitution enable these chemical compounds then to blend
with the vital action in one structure, or to be absolutely inert.
It must depend on differences in the vital chemistry which
underlies function, although these differences which determine
affinity or indifference we can discern only by the result.

" Nerve force, as far as we can see, is the result of chemical
change occurring under the influence of life in the molecules
which compose nerve tissue. Chemical processes, the breaking
up of complex compounds, and the formation of simpler com-
pounds, with consequent release of the energy held latent in the
former, is the constant element in the production and conduction
of nerve impulses. Some chemical compounds may come into
relation with the tissue in which the change is occurring without
exerting the slightest influence upon it. But another substance
may come even in amount inconceivably minute, whose molecules
are so arranged as to fit, as it were, with the changing molecules
of the living tissue. The energy the new molecules bear seems
to blend with that which is in process of ordered release in the
living tissue, and to blend so effectively as to derange it entirely.
Such an influence as I have spoken of seems to be exerted widely
in the case of aconitia. Its contact with some living nerve
structures seems to be so instant and precise as to induce the
production of an excess of energy, sweeping all before it ; on
others, to oppose the process, to induce a sudden stillness among
the changing molecules, and to arrest all action. Among the
nerves thus influenced may be those on which depend the action
of the heart, and with a sudden spasm or a sudden stillness, the
heart stops and life is ended."

I have quoted these last three paragraphs from a clinical
lecture delivered several years ago by Sir William Gowers. I
have not altered his words ; they approach near enough to what
I wish to impress upon you to serve my purpose. They lay down
in a striking manner that the sole difference that we can determine,
from a. chemical point of view, between the living, palpitating
matter and protein, between " imperial Caesar " and his own
dead " clay," is brought about by chemical combination, by
the entrance of certain molecules into combination with the

PROTEINS AND PROTEIDOGENOUS MATTER 225

living or biophoric molecules of certain controlling cells of the
organism, and forthwith, from being active and reactive, these
become inert — dead — protein. That some similar change takes
place in connexion with the death of the tissues in general is
indicated by the change in reaction when any cell passes from
the living to the dead state. Living matter has a feebly alkaline
reaction ; with the onset of death, the reaction becomes acid.
Or, otherwise, in passing from the relatively unstable proteido-
genous to the dead, relatively stable proteid state, the biophoric
molecules either take up alkaline molecules (or ioDs) from the
surrounding cell sap, or give up acid ions to the surrounding
fluid ; they surely undergo chemical change.

Now the very constitution of the protein molecules, as re-
vealed to us by the researches of Hofmeister and Fischer, explains
how this must be. These amino-acid radicles which compose
the protein molecule are all built up along the same lines. They
have multiple affinities. Possibly I here delve too deeply for
some to follow me, but an elementary knowledge of chemistry
and of chemical nomenclature is a part of modern culture, and
therefore I presume to venture, and the accompanying diagrams
may help to explain my meaning. From our knowledge of the
constitution of the protein molecule we may regard the biophoric
or living molecule as made up of a series of amino-acid radicles
joined together in ring form. Fischer's studies have taught us
the mode of junction of these radicles ; it is by the acid carboxyl
(CO) group of the one radicle to the alkaline amine (NH) group
of the other. To this extent the radicles or nuclei are relatively
firmly united. The other components of the different amino-
acids must then form free swinging or side -chains, and it is
according to how these side-chains are built up that we obtain
the different nuclei, or amino-acid components. These are
capable of replacement and modification according to the ions
or compounds attracted from the surrounding medium. They
may be regarded as less stable, able to be detached and
replaced.

In discussing what life is, we may therefore lay down in the
first place that all vital manifestations are manifestations of chemical
change in proteidogenous matter, are, in short, the outcome of
arrangement of that matter with the necessary liberation or
storing up of energy. To this extent all vital phenomena

Q

226 ON VARIABILITY AND ADAPTATION

resemble phenomena of surrounding inanimate nature ; they
differ from those only in degree, not in kind. There is not one
vital activity which can be mentioned that demands for its
explanation something over and above chemical change ; 1 and,
to this extent, inanimate and animate nature are one. There is,
however, an apparent, most important difference between the
results of vital and non-vital phenomena. This has been well
put by Earl. " Every living organism may be regarded as a
centre at which energy is being constantly transformed. It is
by the nature of this transformation that we recognize it as a
living organism. But," he continues, " the continuous operation
of these transformations in the region of the organism is dis-
tinctive. In all exchanges of energy between inanimate bodies
there is a speedy attainment of equilibrium, whereas the organism,
so long as it lives, is incessantly disturbing the equilibrium
which would otherwise arise between itself and its environment.
In other words, living organisms are not ordinary conservative
systems, and the extent to which they diverge from the principle
of the conservation of energy is another indication to us that in
the organism we come in touch with phenomena which are not
yet, at all events, reduced to physical laws."

I wish to discuss how far this statement is true, for, if true,
it immediately defines the difference between phenomena of
animate and inanimate creation. It will be best, I think, to
study the subject by analysing that property of living matter
which is the most evident outcome of the incessant disturbance
of equilibrium above mentioned — I mean growth.

If we seek to test the relative importance of the various forms
in which vital activity manifests itself — motion, sensation,
assimilation, excretion, reproduction — we are bound to see that
one and all of these subserve growth. If the individual moves,
an ultimate analysis shows that the primary object in motion
is either to obtain more food, or, more accurately, the primary
result of that movement is to approach and assimilate food-stuffs,
and that food obtained is of benefit as it can be used for further
growth ; or is to place itself at a greater distance from disintegra-
tive forces. The same is true also with regard to sensation.

1 Even memory has been explained by Hering and others as a reproduction
under particular stimuli of particular relationships between particular mole-
cules, so that now they set in order an identical series of reactions in the cerebra
cells.

THE CONSERVATION OF ENERGY AND GROWTH 227

That is of benefit primarily in order to acquaint the individual
living unit with, on the one hand, its closeness to food-stuffs, or,
on the other hand, with the presence of physical or other agents
deleterious to the organism. Assimilation and excretion are
but the auxiliaries in the due utilization of materials which aid
growth, and in removing from the organism all materials whose
continued presence would disturb the process. Growth, then,
is the central or essential phenomenon of life, and to understand
life it is necessary that one gains a clear idea of what is the
essential nature of this process of growth.

Let us then consider what growth means. It means quantita-
tive increase in the individual matter endowed with life, increase
in the living substance. That individual may consist of a single
cell, may be an almost infinitesimal micrococcus, for example, or,
at the other pole, say in the elephant or in the whale, may consist
of a huge aggregate of countless billions of cells, all associated
and depending the one upon the other. In this latter case it is
the separate cells which, some or all of them, increase in size,
and with this increase also in number. Each cell of such a
multicellular organism is, we know, derived from a primitive
fertilized ovum by repeated division of the original single cell ;
and in this process of division there is a partition of the bioplasm,
of the vital matter. During the period of most rapid growth,
in the stage of development, this increase in size of the individual
cells and their multiplication is most rapid, and when we compare
the adult with the ovum from which it sprang, when we know
that in suitable environment, a single bacillus, for instance,
dividing, can, in twenty-four hours, give rise to hundreds and
thousands, not to say millions of bacilli equal in size and identical
in properties to the original bacillus — one observer has estimated
that in the breeding of the unicellular Rotifer, if all the possible
progeny could be preserved, there would in the course of a year
be developed a mass of organic living matter as large as this
world of ours — we can have no doubt, not merely that growth
means a heaping up, through the agency of the essential vital
portion of the protoplasm, of secondary substances, not in
themselves living and vital, but that that living matter itself is
marvellously increased in amount. There is no other possibility
open to us. But now, can we picture to ourselves the nature
of such increase in the vital substance, the actual bioplasm ?

228 ON VARIABILITY AND ADAPTATION

I think we can, and that by the use of current chemical con-
ventions and symbols. The actual process must, I admit, be
vastly more complicated than the diagram we employ can re-
present. But that doesn't matter. If, by the use of a simple
diagram, we can depict and can grasp the nature of the process,
then we have gained a great step in advance ; only, while using
such a simple diagram, we must remember that it is but a symbol,
it is like using the symbol tt to represent that interminable
fraction 3*14159265 . . . etc.

V
•A

I vi.

V,

<»•-•••

».•

Y\

ll]^>-^llllip^-

« / »

• t • '

H N

Fig. 20. — Diagram of growth of biophoric molecules.

Thus far I have pointed out that all vital processes are
manifestations of energy, and that such manifestations of energy
surely indicate chemical change. Thus the substance or bioplasm
endowed with life may be regarded as a single chemical substance,
varying, it is true, in its properties in the different species and
forms of living beings. If a chemical substance, then it is formed
of molecules. I have already given you the conception of the
structure of the biophore or molecule of living matter. Let us
reduce this to its simplest form — as a ring of carbon-containing
nuclei built up after the type of a benzole ring, with which the
chemists are familiar in the large group of benzene compounds.
Each nucleus of such a ring may, for our present purposes, be
represented by one of the more complex amino compounds
already referred to. Such nuclei are polyvalent — they have,
that is to say, multiple affinities which can be satisfied by the
attachment of other atoms as radicles. It will be seen that to
make a ring two at least of the affinities of each amino-acid
nucleus must be satisfied by junction with other nuclei, leaving,
however, other unsatisfied attachments. And it is in accordance
with the wav in which these unsatisfied arms become satisfied —

PHYSICAL SELECTION AND GROWTH 229

according to these side-chain combinations — that the different
benzene derivatives are formed.

It is interesting to note that even among these relatively
simple carbon compounds — those familiar with chemistry will
appreciate the virtue of that term " relatively " — we find already
that there is indication of selective activitv. I mean this, that
once one of these bodies or derivatives has been formed and
becomes partially broken down, it is fomid more easy to obtain
new associations after the original type rather than new associa-
tions of a different order. Even among the simpler carbon
compounds we see the faint origin of what, biologically, we speak
of as habit. The same is true of radicles in general. Once
certain compounds are formed, the adhesions or attachments are
more close, more fixed. So now to continue. Let us suppose
this simple annule of living matter with certain side-chains
already attached, floating about in a fluid medium in which,
through a concatenation of circumstances, various odd ions are
similarly floating free. The unsatisfied arms of the ring are
liable to be satisfied, that is, to attract and fix on certain other
ions. All that we have to do is to suppose this particular annule
attaches to one unsatisfied arm a floating ion of carbon, that this
then so attaches itself to other carbon ions until a second ring is
formed in association with the first. Then this ring similarly
attaches to itself side-chains in the identical order seen in the
first instance.

Now behold, there is built up a second annule of identical
nature, which, when once formed, may break loose. In place of
one annule we have two, and as the properties of any chemical
compound depend upon its composition, the composition deter-
mining how, and how much, energy is liberated under certain
conditions, so, if the composition of the first annule, as we
agreed, was such as to confer upon it the properties we term
vital, the second annule will possess the identical properties.
In place of one living molecule we have two ; in short, we have
growth.

But stop, you will be saying. By what earthly right do you
assume that chance ions floating in a fluid medium come to
attach themselves one to the other in due and regular order, so
as to form a ring with side-chains identical with the original
ring with its side-chains, with which the first of these wandering

230 ON VARIABILITY AND ADAPTATION

ions became connected ? You admit at the start that this
original ring and side-chains are immensely complicated, and
that our diagram represents an absurdly simple case ; how can
you have the face to make any such proposition ?

All I can say, in reply, is that I have precedent upon
my side. What happens in crystallization ? You have a
watery solution, say, of common salt. You now know that
though you place solid sodium chloride, a definite chemical
compound, into that water, in the very act of solution, the
ions of chlorine dissociate themselves from the ions of sodium.
Molecules of NaCl break up, they are no longer there as such.
You now cause this solution to evaporate and become concen-
trated ; what happens ? As the water volatilizes do the ions of
the gas chlorine escape into the air, leaving the heavier sodium
ions behind ? Not a bit of it. Inevitably, when the concentration
reaches a certain point, some ions of the sodium, aided by some
sharp point or inequality in the surface of the vessel, once more
join themselves each to an ion of chlorine, and the process of
crystallization begins. The very existence of one crystal clearly
seems to cause other junctions to occur in its neighbourhood,
and just that one particular series of junctions necessary to form
sodium chloride, so that, although there may be present in the
solution various other substances, various other ions, yet they
are not attracted, they remain still in solution to a very great
extent ; we gain pure, or almost pure, crystals of the one substance.
Here we have a process of the very same order as that which I
contemplated above, a process of definite, inevitable, or selective
attachment of ions in a certain order, given certain definite
conditions. Certain particular ions unite and build up the
molecules and crystals of one particular inevitable form, of one
particular composition.

Here again the example is very simple ; I might, it is true,
have mentioned solutions of salts of a more complicated nature,
but again I hold that the simplest case is good enough, nay, is
best, upon which to base my contention, which is that we have
to recognize the existence of this affinity or attraction, or what-
ever you please to call it, whereby certain freely floating ions,
and combinations of ions under certain definite conditions, tend
to attach themselves one to the other in a definite order, to
form bodies of a definite composition. And in this connexion,

CRYSTALLIZATION AND GROWTH 231

is it not, to say the least, suggestive, that a fluid men-
struum is essential for vital processes ? Such fluid menstruum,
we now know, is essential for the breaking up of salts into
their constituent ions, so as to bring about the formation of
new combinations, and that in the absence of great heat. Is it
not suggestive that we, for example, are over 70 per cent
water ; that the actively functioning, as apart from the inert
tissues, contain still greater proportions — and the same is true
throughout living nature ; that we have in the living cell just
those conditions favouring alternate ionization and crystallization
of the contained molecules, and that the dominant and char-
acteristic element present in the living matter is one which we
find relatively wanting in this earth of ours, save in connexion
with living matter, or matter like coal, which has been alive ;
and that element, carbon, is, unlike the majority, tetravalent,
and therefore so peculiarly liable to form extensive and com-
plicated attachments ? It would almost seem as though all the
carbon present on the surface of the earth inevitably becomes
utilized to form and to give rise to living matter. The same is
largely true, too, as regards the nitrogen.

In short, were I a Frenchman, I would say : La vie c'est
la crystallisation. Life is the by-play of carbon-nitrogen com-
pounds now breaking up and liberating energy, now attaching
to themselves other ions, and so storing up energy, deduplicating
themselves, and undergoing growth, or, as I have expressed it
elsewhere, life is to be regarded as a state of persistent and
incomplete recurrent satisfaction and dissatisfaction of certain
proteidogenous molecules.

If we accept this view, and the more I ponder over it the more
convinced am I that it is along the right lines, where is the wide
difference I quoted to you as being observed between animate
and inanimate nature ? Where is that violent, not to say
indecent, assault upon the otherwise universal and ever to be
respected principle of the conservation of energy ? Life and
growth come under the category of chemical processes such as
we are familiar with. All that we have to recognize as separating
what we term animate from what we term inanimate nature is
that, from the intimate constitution of carbon, it is able to form
combinations of a marvellous complexity — unsatisfied compounds
which, therefore, are in a state of unstable equilibrium, which

232 ON VAKIABILITY AND ADAPTATION

easily break off certain side-chains already attached, thereby
giving off energy, which as easily, under certain conditions of
environment, attach other side-chains to them in definite order,
thereby storing up energy and undergoing growth.

I would dwell for a moment upon this point made by Earl,
and already referred to, namely, the remarkable condition of
persistent, unstable equilibrium of living as distinguished from
non-living matter. The distinction, I would say, not so absolute
as he lays down. Once more, it is a matter of degree. Non-
living matter is only in a condition of relatively stable equilibrium.
Our earth, for instance, taken as a unit, is only relatively stable.
If astronomy teaches anything, it is that each planet, each sun,
each solar system, has its periods ; that each begins in a gaseous
state, a state of enormous instability, enormous liberation of
energy ; that from the nebulous stage, as a result of the mutual
attraction and combination of its elements, it undergoes a stage
which is strictly comparable to that of growth and then settles
down into partially — and only partially — independent systems,
with a certain storage and a certain dispersal of energy. But
our earth, for instance, is subject to progressive change, and is
far from being eternal. Sooner or later in the aeons that are to
come, our sun will become exhausted as an energy-distributing
centre ; the system will be dead. Sooner or later, in aeons far
far distant, the attraction ever present between planet and sun,
and sun and other suns of yet greater systems, will draw planet
unto sun, and sun unto other suns, and with each act, so colossal
will be the force exerted, the force of impact, that what is now
solid and fixed will once more be dissipated ; the equilibrium is
but temporary, the return to the nebulous stage inevitable.

But true it is, all the same, that the equilibrium of living
matter is most markedly unstable, and this instability is essential
for active dvnamic life. We are never satisfied. Render the
living molecule a satisfied body, with all its side-chains complete,
and unable to make attachments with other ions, so that no
further interactions are possible between the molecular system
and its environment, and it must come to rest, and function be
at a standstill, all indications of life ceasing. It is quite possible
that, as Gowers indicates, certain poisons act, and thereby induce
death, by combining with the molecules of certain all-important
centres in such a way as to satisfy those molecules, and so arrest

INSTABILITY OF LIVING MATTER 233

function and further activity. Other poisons and physical agents
may likewise cause death by the contrary process of decom-
posing and breaking down the molecules, whereby they become
so gravely disorganized as to be unable to continue the pro-
gressive series of attachments and of liberation of side-chains
which are the underlying features of the vital process.

I could expand this conception and dilate for long upon its
many bearings. Once we accept this chemical theory of growth,
if I may so term it, it is wonderful how it illuminates and
harmonizes a whole host of phenomena regarding which there are
hosts of theories, having this in common, that they do not
adequately explain. I could spend some hours of your time
applying this conception to the subjects of evolution, of descent,
and of inheritance, the inheritance or non-inheritance of acquired
properties, the inheritance of disease, the immunity from disease,
and so on. But I will desist ; it may seem to you that I have
already gone too far. But in defence of myself, I am prompted
to quote some words of the good old Stephen Hales, Vicar of
Teddington, near London, in the reign of George L, who was the
founder of modern exact physiology, based upon accurate
measurements, and quantitative rather than qualitative studies.
" In natural philosophy," he says, " we cannot depend on any
mere speculations of the mind ; we can only, with the mathe-
maticians, reason with any tolerable certainty from proper data
such as arise from the united testimony of many good and
credible experiments.

" But it seems not unreasonable, on the other hand, though
not far to indulge, yet to carry our reasoning a little further than
the plain evidence of experiments will warrant ; since at the
utmost boundaries of those things which we clearly know, there
is a kind of twilight cast from what we know on the adjoining
borders of terra incognita. It seems therefore reasonable to
indulge conjecture there ; otherwise we should make very slow
advance in future discoveries, either by experiments or reasoning."

I have, it may be, conducted you to this twilight land on the
borders of terra incognita. I shall be satisfied if I have set you
thinking ; if I have indicated to you, though vaguely, that life
is not a thing peculiar and apart : that it is a portion of a larger
whole ; that animate and inanimate nature conform to the same
laws, the same principles of chemical association and dissociation.

234 ON VARIABILITY AND ADAPTATION

Vaguely, but no less surely, we find the same laws at work
throughout nature in all its aspects. The processes which
determine the structure of the atom, determine the properties
of all aggregations of atoms ; what rules the mighty sun, rules
also the less than visible microbe.

CHAPTER VIII

ON HABIT, SYMPTOMS, AND DISEASE1

(1911)

It has struck me that it might be serviceable to take up the
subject of habit in its relationship to disease : to call attention
to the probability that much commonly regarded as symptomatic
of functional, not to say anatomical, disturbance may, after
all, have no serious anatomical basis, but be one or another
manifestation of habit, and, therefore, wholly curable provided
the habit can be interrupted ; and to indicate how, nevertheless,
habit sufficiently long continued may well induce anatomical
change, from which stage onward cure can be only relative.

Placed thus on paper, there is nothing new in my text. These
are truisms which we all must have had impressed upon us in
the course of our work, now vaguely, now more clearly. The
whole group of tics and hysterical manifestations is immediately
called to mind, these affording examples of both stages. But
what I want to do is to point out that the nervous system is not
alone involved, and that the tissues generally are creatures of
habit ; further, to investigate the principles underlying habit
phenomena and the anatomical or histological outcomes of the
same.

This matter was brought to my attention years ago in con-
nexion with a case of incipient pulmonary tuberculosis. The
patient had, shortly after child-birth, gone rather rapidly down-
hill, giving the old story of profuse lactation, loss of strength,
intractable cold with profuse expectoration, and afternoon rise
of temperature. Everything pointed to tuberculosis, and, what

1 An address delivered before the Medical Research Club at the University
of Pittsburgh, April 6, 1911. Reprinted from International Clinics.

235

236 ON VARIABILITY AND ADAPTATION

is more, our ablest diagnosticians recognized a small area of
dulness at the right apex ; they were doubtful as to whether
there was a still smaller area to the left. But no bacilli could
be discovered. The patient was sent to Lakewood, and there
at last, after two negative reports, a search through something
like seven slides revealed a single definite tubercle bacillus,
with a suspicious second. This was years before the development
of the finer modern methods of diagnosis, but it has always
seemed to me that here was a case in which the disease was
suspected and then recognized in its earliest stage. The patient
was immediately packed off to Dr. Trudeau at Saranac Lake,
and from the first week put on weight, gaining fifty pounds in
five months, and rising from one hundred to one hundred and
fifty pounds in weight. The effect of treatment, that is to say,
was immediate. Only once in the first fortnight of her stay
there was a doubtful bacillus or two recognized in the sputum.
From this time onward the sputum was uniformly negative.
That was twelve years ago, and the patient has never shown
a sign of set-back.

The point that I want to make is this : that here there was
undoubtedly tuberculosis, but that of the very slightest we
could then diagnose, and, with proper treatment, the bacilli,
never present in any number in the sputum, disappeared within
a week or two. The fever went down, the weight went up.
Notwithstanding this, for more than six months — I do not
believe I exaggerate when I say for at least a year — that patient
not merely had a cough, but brought up, particularly in the
morning, an excessive amount of thin mucoid expectoration.
The cough I could understand — the healing of the tuberculous
process in the lungs and the formation of fibrous tissue might
well set up a certain amount of local irritation, and reflexly in
this way induce cough. Nevertheless, the cough very soon took
on what we term the nervous character. The patient was
highly strung, and this might well be a nervous habit. But I
could not well explain the abundant expectoration continuing
all these months after the evident healing of the lesions, save on
the theory of habit, the theory, namely, that a definite irritant
had in the first place stimulated the production and discharge
of mucus, with associated congestion of the peribronchial vessels,
and that the irritant had continued active for a sufficiently long

PATHOLOGICAL HABITS 237

time to set up a habit of heightened activity on the part of the
secreting cells, so that when the irritant disappeared and was no
longer in action, the congestion remained, and with it the cells
continued to secrete abundantly.

A few months ago my colleagues, Drs. Birkett and Meakins,
contributed a paper, " Vaccine Treatment in Chronic Inflamma-
tory Disease of the Accessory Sinuses of the Nose," to a most
interesting discussion on " Vaccine Therapy " at the Triennial
Congress of American Physicians and Surgeons at Washington.
They reported a short series of cases in which from the sinuses
a growth of one or other pyogenic organism was obtained prior
to treatment with vaccines, and in which, as the result of the
treatment, the discharge became wholly sterile. But in the
majority of these cases, while the character of the fluid changed
under treatment from a mucopurulent to a purely mucoid fluid,
the discharge, it is noted, had continued for months, with all the
distressing symptoms of hypersecretion and retention. To
explain this, Dr. Meakins calls attention, first, to the chronic
thickening of the submucous layer of the lining of the sinuses,
as the result of continued inflammation ; and, secondly, to a
partial closure of the ostia by the swollen mucous membrane
inducing defective drainage ; thirdly, he comes to the same
conclusion that I had reached in the case just described, namely,
that through repeated stimuli (at first bacterial) hypersecretion
has been set up, which after a time has become a true habit,
independent of bacteria, as evidenced by its perpetuation after
the cavity has been practically sterilized.

As I have already hinted, we are all so familiar with the
establishment of nervous habits, with the development of habit
tics and those graver habits of lack of co-ordination, and appar-
ently of arrest of active communication between various centres,
which would seem to be the basis of hysteria, that I must not
here tire you with an enumeration thereof. It is well recognized
that it is easier for nervous stimuli of different orders to travel
along well-worn paths, and that thus there may be established
a condition in which a minimal afferent impulse, travelling
without interruption along a particular path, is sufficient to set
in action first one and then another of a group of co-ordinated
centres whereby a minimal stimulus may eventually produce
a maximal result ; per contra, the act of inhibition or arrest of

238 ON VARIABILITY AND ADAPTATION

the passage of a given impulse, if repeated, would seem to oppose
so strong an obstruction to the passage of impulses along a
particular path, that soon a maximal stimulus may produce no
result. Here there is no anatomical change : it is a matter of
habit. Put the patient under an anaesthetic and remove the
inhibiting mechanism, and now stimulus is followed by the normal
reflex muscular act. The worst of it is that, too often, the patient
is in absolute ignorance of the fact that he or she is exerting
this inhibition. I recall vividly a very delightful old lady, a
woman of great personal charm, and, if I may so express it,
eminently virtuous, who had been in a terrible railway accident
on the north coast of Wales in which many people were killed,
who received so strong a nervous shock as to be incapable of
using her lower limbs, and to be bedridden for twenty years. I
have said that she was virtuous ; that is to say, that within a
year after the accident she was awarded by the courts one
hundred thousand dollars (£20,000), and she did not, as is pain-
fully often the case — and that in a way that, perhaps at times
mistakenly, makes us contemn poor humanity — recover her
powers so soon as the court had awarded her this compensation.
No, she remained bedridden twenty years, and then for the
first time in her reign Queen Victoria was about to visit a near-by
city. The old lady had never seen the Queen, who was just of
the same age, so the occasion appealed to her very peculiarly.
She longed to see her, and that with a great longing ; and, sure
enough, the afternoon before the great event she got out of
bed and walked, went to the celebration next day, and after
that walked until she died.

I do not say that she walked vigorously, but if a long spell
of arrested function had some effects, certainly nerve tracts had
not been destroyed, or otherwise her paralysis was not due to
anatomical changes in the nerve cord. It was an inhibitory
habit. But, as I say, I do not want to go into these nervous
conditions ; these we all accept. The instances of automatic
cellular activity are of more immediate interest, and once I
mention one or two of these to you, I doubt not that you who
are in practice will call to mind individual cases of your own
which are best explained along this line. Thus my colleague,
Dr. Meakins, has quoted to me more than one case in which a
small dose of mercury has induced abundant salivation. We

PATHOLOGICAL HABITS 239

know that salivation is associated in the first place with discharge
of the mercury from the system, but all the mercury in small
doses must disappear in the course of a few days, nevertheless
the ptyalism in some cases persists, and is noticeable for six
months or even a year. Some cases, at least, of persistent gastric
hyperacidity and excessive secretion would seem to be of the
same order, as also, I would suggest, the opposed habits of
constipation and looseness of the bowels, the one due to the
habit of excessive absorption of the fluid matter of the faeces,
the other to defective absorption. Professor Klotz has just
quoted to me four cases of obstinate dysentery in which un-
doubtedly there had been an original bacillary disturbance, but
repeated bacteriological examinations failed to disclose anything
of the nature of a specific agent. There is also a singularly
intractable, obstinate, and elusive intestinal condition which
I would suggest is most simply grasped if we regard it as a habit
disorder — a condition devoid of indications of bacterial nature, of
insidious onset, becoming progressively more severe, unassociated
most often with febrile disturbance, most often showing itself
in the female, and then in high-strung girls verging upon the
hysterical — I refer to mucous colitis. There is something
striking in the way this persists once it manifests itself, and
that in the absence of sign of any abiding source of irritation.
I would suggest that here, following upon a catarrh of moderate
intensity in an individual of unstable equilibrium, the cells of
the mucous membrane of the colon have acquired the habit of
excessive formation and discharge of mucin.1

1 After this address had been delivered Dr. Lichty, of Pittsburgh, called my
attention to an important paper of his (American Medicine, iv., 1902, p. 223)
upon the " Etiology of Mucous Colitis," in which he has drawn attention to
the intimate association between abdominal ptosis and mucous colitis, and in
which he draws the conclusion that with this ptosis and hypostasis and further
interference with the blood supply of part, at least, of the large bowel, a condi-
tion of chronic congestion of the mucosa is set up, with a disturbance of function
and excretion of mucus. Dr. Lichty recognizes fully that not all cases of
splanchnoptosis exhibit accompanying mucous colitis, pointing out that 2558
consecutive examinations of men, women, and children afforded 313 cases of
splanchnoptosis — 41 males and 272 females ; but only 21 of these had mucous
colitis, and he concludes that ordinarily a condition of compensation is estab-
lished, such as is so often seen in the circulatory system when there is disease
of the heart and the kidneys. When, however, this compensation is lost or
disturbed, the symptom-complex of mucous colitis appears, and he gives
instances in which, in those affected with splanchnoptosis, mucous colitis
showed itself in one case after an attack of acute tonsilitis, in others after typhoid

240 ON VARIABILITY AND ADAPTATION

Perhaps here wrongfully I speak for myself, and do not
represent the general views of others, but it seems to me that
commonly we are apt to regard every functional act of the various
tissues as an immediate response to some stimulus, and, doing
this, to overlook and neglect the abundant cell activities that
are truly automatic on the part of individual cells, and cell-
collections or tissues ; are apt not to realize that according to
causation there are three orders of cell activities : (1) those
determined by nervous stimulation, which we may term neuro-
genic ; (2) those determined by direct stimulus of the individual
cells by anything modifying their immediate environment —
the environmental ; and (3) the automatic, namely, activities
which proceed in spite of or despite absence of sustained stimuli,
whether neurogenic or environmental.

Now, it is these last, the automatic activities, to which I
would particularly direct your attention — to their nature, their
mode of origin, and their relationship to the other two. I
believe that when we come to look narrowly we find that they
are quite common. It seems to me that the development of
cell habit is essentially the development of cell automaticity.

Primarily, as determined by a study of the lowest unicellular
form of life, and of isolated cells of multicellular animals, such
as leucocytes, it is external agencies that initiate cell activities ;
the nervous system is a higher co-ordinating mechanism de-
veloped in multicellular organisms, and so neurogenic cell activi-
ties are of secondary and later development. Further, we must
regard all automatic activities as not of independent development,
but as initiated by the individual cell becoming subjected, in the
first place, to either environmental or neurogenic stimuli. To
afford a basal instance : the ovum while fully matured lies latent
and incapable of growth until such time as a spermatozoon, or, as
Loeb has shown, alteration in the tonicity of the surround-
ing medium, initiates active nuclear changes ; once initiated,

fever. While thus Dr. Lichty affords an anatomical basis for the supervention
of mucous colitis, it will be seen that his views are not in opposition to the
suggestion here put forth, but, on the contrary, tend to confirm it. Dr. Lichty,
that is, recognizes that some active disorder leading to lowered vitality and
increased congestion is necessary to originate the condition ; that, once origin-
ated, the excessive discharge of mucus lasts for weeks and months after the
exciting cause has passed away. Accepting Dr. Lichty's views, we may say
that the state of chronic congestion of the colon favours the development of
the habit of excessive secretion of mucus.

THE ORDERS OF CELL ACTIVITIES 241

these continue. All cell activities are at bottom chemical or
physical changes in the relationships of the cell molecules and
their constituents, and so living matter is similar to matter of
all other orders in that molecular changes and activities require
to be set in motion by the influence of forces acting from without.

To this broad statement it may be objected that the remark-
able investigations of late years upon radium and allied bodies,
notably those of Rutherford and other pupils of Sir J. J. Thomson,
have demonstrated that every individual atom of matter exhibits
an inherent activity of its constituent electrons, and, therefore,
energy from without is not an essential pre-requisite ; that in its
very essence matter is a storehouse of energy, and that thus
even inert non-living matter possesses the potentiality of auto-
matic activity. Here, truly, we plunge into very deep waters.
To such objections I would reply that the studies upon radium
so far only exhibit the possibility of these inherent atomic activi-
ties leading to the dissociation of electrons and coincident libera-
tion of energy, whereas the striking feature of living matter
is characteristically its capacity to associate, to build up from
its surroundings, other matter like unto itself. Nevertheless
my colleague, Professor Barnes, who, as a co-worker with Ruther-
ford at the time when he was making his remarkable discoveries,
speaks with high authority, assures me that the activity of
the electrons, so far as we can determine with our present know-
ledge, is wholly uninfluenced by, and presumably, therefore,
not initiated by, influences acting upon the atoms from without.
Up to the present, the greatest heat and the greatest cold to
which radium has been subjected have not been found to influence
in the slightest the rate of emanation of the radium rays. In
other words, the rate of motion of the electrons is unaltered by
extreme temperature changes. It may, therefore, be that in
this automatic activity of the atom we have the primal basis
of the automatic activities which here I am discussing ; that,
just as there are inherent attractions and repulsions within
the atom, so there are inherent attractions and repulsions of
the molecules that constitute the cell.

If this be the case, how are we to picture the development
of automatic activities ? In this way, namely, that the cell is
not a single substance, but a complex of many, with central
nuclear matter, and a cytoplasm of proteid matter which presents

R

242 ON VARIABILITY AND ADAPTATION

within it material of various orders, i.e. (1) bodies in the process
of being broken down to provide food-stuffs, (2) the molecules not
yet wholly organized or built up into the cell structure, (3) other
molecules which in the dissociation of the food-stuffs are separ-
ated off as waste products, (4) other molecules which are the dis-
sociation products of nucleus or cytoplasm, and are the resultants
of functional activity, and along with these, (5) framework sub-
stances, such as fibrils and the striated material of the muscle-cell,
developed as the cell becomes differentiated, the outcome of that
differentiation. The activities within the nuclear membrane
are, from the nature of that membrane, very different from those
outside. Surface tension has also led to the development of a
potential, if not an actual, membrane between the cytoplasm
and the external medium.

The cell, in short, is a microcosm — a little world in itself.

We in pathology always go back to the cell to determine its
elementary properties, and from them to base solidly our views
regarding the disturbances that may affect collections of cells,
namely, the tissues and organs of the body. Has it ever struck
you that we can go much further than this ? There is room for
a good full essay upon the study of cytology as the essential
basis of sociology and political economy. It stands to reason ;
it is, if I may so express it, a " dead easy " exercise in elementary
logic to prove that it must be so. The nation is a collection
of allied and similar individuals, and national sentiment, which
expresses itself in national action, can only be the expression
of the sum of individual sentiments, at most gaining impetus
and direction by the interaction of the sentiment of one individual
upon the sentiment of the other. It is obvious, therefore, that
the political economist, in order to understand social tendencies,
must base himself upon, and must first study, the 'individual
and his tendencies. But now, what is the individual but a
collection of cells ? By a like process of reasoning, it is in-
evitable that to understand the individual we have to get right
back to the unit cell, its properties, its reactions, and its inter-
actions with the other cells of the corporation. Wherefore it
follows that the only sound method of mastering political economy
is through an expert study of cytology.

I might dilate upon this theme ; might, for example, point
out how a study of the cell throws light upon the existence of

CYTOLOGY AND SOCIOLOGY 243

two great political parties, and show how the " mugwump " is
the outcome of a peculiarly exact balancing in the one individual
of those two opposing properties of all living matter ; namely,
of heredity in the strict sense, whereby, despite environment, the
living matter of one generation tends to reproduce with exactitude
the characters impressed upon preceding generations, and varia-
bility, whereby that living matter is keenly responsive to environ-
ment and liable to modification. In most of us one or other
of these two properties is in the ascendant. In the words of
Gilbert (and Sullivan), we are born " either a little liberal or
else a little conservative " ; in the mugwump, thanks to the
parental amphimixis, the two properties are accurately balanced.

However, this is all parenthetic. What I want to impress
upon you is that just as nowadays we have learned to regard
the atom not as a solid, if almost infinitesimal, mass of matter,
but as exhibiting the active play of numerous electrons, so,
taking the cell as a unit, we must realize that it is not merely
a mass of homogeneous protoplasm, but it is an extraordinary
complex of molecules, a little world in itself.

We have to imagine these molecules as in a state of con-
tinuous interaction. Given a stimulus from outside, they set
in motion a particular chain of interactions, and we are forced
to recognize that, once started, the interaction may continue
after the stimulus which has set them in motion has ceased to
act. We have, in the first place, then, to recognize the existence
of what Dr. Fraser Harris has spoken of as the inertia of living
matter, which might perhaps more accurately be spoken of as
the momentum of living matter. It is the principle whereby
the wheel set rolling continues to rotate until friction brings
it to rest. It is the principle that explains the latent period of
muscle. If the muscle be resting, as every first-year student
knows, a definite period elapses after the passage of the stimulus
down the nerve before the muscle passes from the resting to the
contracting state. Here it seems to me is the first step ; but
something more than this is necessary to induce automatic cell
action, and to gain an explanation we pass from the cell to the
larger world in a search for parallel phenomena, for, as I have
laid down, we may safely draw such parallels.

Centuries ago when a comet appeared in the sky it had a
profound influence on those who beheld it. It was unusual,

244 ON VARIABILITY AND ADAPTATION

therefore it betokened something unusual. It was a portent,
but its effect was temporary. People spoke of it after it was
gone ; as it were, a condition of inertia continued with steadily
diminishing force after the stimulus had passed away. Then by
chance in the seventeenth century a comet came into the ken of
an astronomer who, through his special knowledge, was, if we
may so express it, peculiarly responsive ; its presence stimulated
him to a more than ordinary degree. He looked up the earlier
records and found that a similar comet had been reported as
naming into prominence at regular intervals. This led him
further to study the path of the comet, and not merely to deter-
mine the course of this one comet, but that of other comets and
of meteorites in general. And now everybody knows all about
comets, can predict their return ; the series of observations of
this one astronomer, Halley, has permanently and continuously
affected civilized man.

So it is with the cell molecules. One particular stimulus
may have but a temporary effect ; another, acting upon the cell
in a particular state, may initiate a particular recurrent cycle
of intracellular changes, may set up a habit.

Not all influences upon cells will set in action these recurrent
and cyclic changes, but undoubtedly the dominant constituents
of cells, the proteins, by their very nature and complexity of
structure, favour the development of cell habits. Let me
illustrate. A diffusible foreign body, itself of proteid affinities,
gains entrance into the blood — a toxin ; and, provided that
it be not exhibited in excessive amount, it is in the first place
neutralized, and next leads to the presence in the blood of anti-
toxins. We know that the toxins gain entry into the cells ;
know that the cells, or certain of them, actively discharge anti-
toxins. There may have been not a sign of specific antitoxins
in the body fluids primarily, but now they are produced in
abundance — in amounts far in excess of the amount of toxin
introduced. If at first they were modifications of the toxins
introduced, brought about by cell activity, certainly the later
crop cannot be of this nature. And here is the point : they
continue to be produced days and weeks after the original toxins
have been neutralized or destroyed.

It may be that our presumption is wrong, that the toxin
molecules, once within the cell, persist there, acting as enzymes

DEVELOPMENT OF CELL HABITS 245

or ferments, converting one after the other molecular groups of
a particular order into antitoxin molecules. These are matters
not yet determined. It fits in better with our knowledge to
suppose that the toxin molecules determine a modification or
rearrangement of certain side-chain molecules, which in their
turn lead to the building up of other side chains of like con-
stitution. But, whatever the exact process, here is an exquisite
example of a cell habit of the automatic type. Once started
in their new work of producing antibodies, the cells continue
to produce them, and this without external stimuli, developing
them from the usual food-stuffs assimilated by the cell.

It follows, thus, that we may recognize two orders of cell
habit of the automatic type : first, as afforded in my original
instance, the exaltation of a property already possessed by the
cells ; and, second, as exemplified by this last instance, the
acquirement by the cell under abnormal environment of new
properties — an acquired cell variation becoming, if I may so express
it, converted into a cell heredity.

Underlying both we must see a manifestation of that most
potent cause of metabolism, namely, enzyme action. In other
words, it is the existence of intracellular enzymes that permits
both continuous and cyclic automatic cell activities. If the
view be accepted which I have put forward elsewhere, namely,
that organic enzymes are of the nature of detached side chains
of the proteidogenous molecules — able, on the one hand, to
attach themselves to the main ring ; on the other, to molecules
of food-stuffs, bringing about their dissociation, with eventual
building up of like side-chain material in series, — then we can
realize how, so long as side-chain molecules of a particular order
are present in the cell, and that cell possesses its usual nutrition,
the enzyme action will continue within that cell until arrested
by the concentration of the products of that action ; to become
active again so soon as those products become used up, diffused
out, or otherwise removed. Once enzymes appear or become
developed within a cell, we have the conditions most favourable
for continuous cell activity, irrespective of alteration in environ-
ment or external stimulus, save that afforded by the exhibition
,nd regular absorption of food-stuffs. And when abnormal
bodies of proteid type — toxins — gain entrance into the cell, I
would lay down that they have affinity to side chains of the

246 ON VARIABILITY AND ADAPTATION

cell substance, and either themselves act as enzymes, or more
probably form compounds of a new type within the cell which
act as side chains of a new order, and are capable of building up
new side-chain molecules in series, the excess of which become
discharged as antitoxin molecules.

However, from the pursuit of such ideas I again must desist,
for I draw you into the depths. I only wish to impress upon you
that the known existence and extent of intracellular enzyme
activity and the development of the same afford us a full physio-
logical basis for the assumption by the cell of habits of
activity.

Thus far I have discussed only those habits associated appar-
ently with no anatomical or histological change in the parts
that are involved in the habit. This is, however, by no means
necessary, and the last few minutes of this discourse may be
spent in calling attention to the habit of growth — conditions
associated with distinct structural changes in the affected tissues.
The whole somewhat limited group of what we term metaplasias
are really growth habits. They are conditions in which, always,
it would seem, under the influence of altered environment,
one or other tissue undergoes a metamorphosis, and continues
to develop in the metamorphosed state. I had a beautiful
example of this only a few months ago, in which, apparently
under the influence of chronic inflammation, it could be seen
that the perichondrium of the tracheal cartilages had given
origin to or had undergone conversion into cartilage cells within
its substance, which cells, undergoing proliferation, had now
projected outward into the submucosa to form small nodular
growths, and these in their turn had undergone bony transforma-
tion. Here, then, one and the same tissue, under different
conditions of environment, gave rise to connective tissue, cartil-
age, and true bone. Nor was this all. These multiple bony
nodules projected into the lumen of the trachea, and, doing
this, interfered with both the nutrition and the function of the
overlying epithelium. As a result, over each nodule the mucosa,
instead of being formed of a single palisade layer of columnar
cells, had become converted into a well-formed squamous epi-
thelium, while between the nodules was the normal columnar-
celled epithelium.

The cells here under different environment had assumed a

HABIT AND METAPLASIA 247

totally different habit. It may be asked whether here I am
using the term in the same sense as previously, whether what I
here term habit is not the result of continued alteration of environ-
ment. It may be asked whether in these cases return to the
normal environment would not bring about return to the normal
type of tissue, or whether this altered habit of growth persists.

To this I would answer that undoubtedly it tends to persist.
As evidence I may cite the example of a region in which epithelial
metaplasia is not uncommon, namely, the gall-bladder. There,
as a result of the chronic irritation set up by the presence of gall-
stones, it is not uncommon to find larger or smaller areas of
squamous epithelium taking the place of the normal columnar-
celled mucosa, and this more particularly in the region of the
fundus. Now, as is well known, 90 per cent or more of cases of
cancer of the gall-bladder are associated with cholelithiasis, and
the majority of those cancers originate in the fundal region ;
and the interesting point is that I have encountered (and there
are several similar cases recorded in the literature) cases in which
that cancer, instead of being of the adenocarcinomatous type
which one expects to originate from a columnar-celled epithelium,
approximated to the squamous epitheliomatous type. They
were not absolutely typical epitheliomas ; epithelial pearls, for
example, have been wanting. My own case, for example — and
others have made the same observation, — exhibited a mixture
of alveolar masses of the epitheliomatous type, along with other
alveoli of adenocarcinomatous appearance. But certainly these
were not ordinary adenocarcinomas. There is no epithelium in
the neighbourhood of the developing gall-bladder of squamous
epithelial nature ; no chance, therefore, that these tumours
develop from a squamous epithelial cell rest. The only adequate
explanation is that the metaplastic squamous epithelium has
its characters so firmly impressed upon it that, when it gives
rise to a cancer, its cells, as they infiltrate into the surrounding
tissues, do not revert to the columnar type, but give rise to a
squamous epithelial new-growth. Again an acquired variation
becomes hereditarily transmissible among the cells. They have
" got the habit."

Some of you may know that many years ago I applied this
habit conception to the elucidation of tumour growth in general,
and though in the meantime I have amplified my views, I still

248 ON VAEIABILITY AND ADAPTATION

believe that my original hypothesis is sound. I would conclude
by recalling this to you. There are many examples that I might
take as illustrative, but perhaps the best is that worked out by
my late colleague, Dr. Wolbach,1 now assistant professor at
Harvard — his study of the mode of origin of that modern disease,
X-ray carcinoma. He had at his disposal material from a long
series of X-ray burns, and so was able to follow well all the
stages.

Paradoxically, the X-rays have singularly little effect upon
the epidermis and epidermal cells, but they influence profoundly
the underlying corium. There in the earliest stage there are
set up marked lesions, telling especially upon the vessels and
lymph spaces — proliferation and desquamation of the endothelial
cells, thromboses, and other evidence of severe vascular dis-
turbance ; and it is secondary to this that the epiderm is affected
and tends to ulcerate. Those changes, to be brief, lead to a
form of inflammatory fibrosis of the subepidermal tissue, and
as a result the nutrition of the epiderm is gravely reduced. In
some places the reduction is so great that the cells necrose, and
ulceration is the result. In other places the effects are not
so intense, and one gets what I would term the " poor curate '
phenomenon. You will have observed that the well-fed million-
aire and his wife are apt to be childless, whereas the curate and
his wife have a family in inverse size to their capacity to support
them. That appears to be a law of nature : there is a certain
nutritive minimum at which, it would seem, the preservation of
the species becomes of greater consequence than that of the
individual, and multiplication ensues, to the detriment of the
individual, but on the off-chance that a few out of the abundant
progeny may survive. And so it is that in these cases where
the epiderm is not wholly necrosed, its cells multiply ; and now
is observed a phenomenon apparently of the same order as that
first observed by Bernhard Fischer after inoculating a mixture
of olive oil and Sudan iii. into the tissues of the rabbit's ear
(an observation abundantly confirmed), namely, the growth
downward of the epithelial cells toward the oil. In these X-ray
burns the actively proliferating epithelial cells grow downward
through the dense fibrosed cutis toward the nearest source of
nutrition — the underlying vessels.

1 Wolbach, Journ. of Med. Research, xxi., 1909, 415.

HABIT AND NEOPLASIA 249

But doing this, obviously, they sacrifice their function as
epidermal cells. Cells growing down into underlying tissues
cannot function as an epithelium. And thus, coincidently, in
these cells we have the vegetative powers exalted and the func-
tional activity depressed. Now, while there is a certain mean
or level at which moderate function, by stimulating increased
assimilation, may stimulate increased cell growth, it must be
recognized that growth and function are diametrically opposed
states of cell activity. Growth demands that matter taken as food
is built up into the cell substance and becomes associated, with
coincident storage of energy ; function, on the contrary, demands
dissociation, with liberation of energy. And, as a matter of
fact, your actively functioning cell shows no signs of proliferating.
Your actively vegetative cell is of embryonic type, i.e. shows
little signs of differentiation, and therefore of functional capacity.
We observe, thus, that these down-growing epithelial cells —
and the same is true of all early malignant growths — present
coincidently depression of function and exaltation of vegetative
power. And I would lay down that it is this modification of the
cell activities that is the basis of neoplasia ; that what is charac-
teristic of the true tumour is that the component cells, from
one or other cause — displacement, irritation and so on, — have
lost the habit of function, and coincidently, as the nutritive
material taken in needs to be used up, the very accumulation
of the newly formed cell substance, disturbing, as it does, the
surface-to-mass relationships of the cell, becomes in itself the
stimulus to cell division, in order that the due relationships may
be restored. Thus is developed the habit of growth, replacing
the habit of function, and it is, I hold, this habit of growth that
characterizes all true tumours. It is at least suggestive that
the one non-surgical method of treating malignant growths
that is giving definite results at the present moment, when
appropriately applied, essentially depends upon a specific reduc-
tion in the growing power of the tumour cells. It is now well
established that radium and X-rays act especially upon the
vegetative cells of the organism, and, characteristically, it is
the actively vegetative malignant tumours, rather than the
benign with their more highly differentiated cell elements, that
are arrested and undergo absorption after treatment with either
radium or X-rays.

CHAPTER IX

PAKENTERAL DIGESTION AND IMMUNITY1

(1914)

Asked to open this discussion upon immunity, I will, I think,
be of most service if I place before you in outline the more recent
developments in our views regarding the essential nature of the
processes which lead up to the production of the immune state.
All medical men of our generation will have, doubtless, a vivid
memory of the earlier phases through which we have passed. The
first phase was the empirical and lasted up to the recognition of
the existence of pathogenic microbes : during this it was admitted
that one attack of many infections was followed by resistance
to subsequent attacks, and there stood out the one solitary ap-
plication, namely, Jennerian vaccination. The second phase,
in the 'eighties and early 'nineties, was one of confusion, when,
recognizing that the infections were of microbic causation, and
being successful in isolating the causative agents, we found
that each different pathogenic agent had a distinct modus operandi.
On the one hand it was found that organisms like the diphtheria
bacillus and tetanus bacillus in their growth outside the body
produced diffusible toxins, and Behring, Roux, and others
discovered that the immunity which developed was antitoxic,
and on the other hand the majority of pathogenic microbes
were found to produce no noticeable ectotoxins, and such im-
munity as was obtainable was not antitoxic but bactericidal,
the fluids of the immunized animal bringing about the solution
and death of the causative bacteria.

1 Contributed to the discussion upon Recent Advances in the Theory and
Practice of Immunity, held at a meeting of the Montreal Medico -Chirurgical
Society, December 4, 1914. Reprinted from The Canadian Medical Associa-
tion Journal, July 1915.

250

PARENTERAL DIGESTION AND IMMUNITY 251

Rapidly following upon this came the discovery in France
and Germany by Bordet and PfeifTer, respectively, that the
latter form of immunity demanded the presence of two orders
of substances in the body fluids of the immunized animal. How
great was the confusion is indicated by the fact that for one
of these, Pfeiffer's " immune body " and Bordet's " substance
sensibila trice," the other names of " amboceptor," " fixateur,"
" intermediate body," " sensitizer," " preparator," " desmon,"
" copula," were proposed by various workers. Nor was the
confusion lessened by the continuing fight between the German
humoralists, who were satisfied to study the phenomena mani-
fested by the body fluids alone, and the Franco-Russian cellular-
ists, led by MetchnikofI, who, with broader histological insight,
laid stress upon the cellular reactions to infection.

The third phase was that of the ascendency of Ehrlich's
" side-chain theory " of immunity. It was found that antibodies
present in the blood serum neutralized the toxins ; all the in-
dications were that chemical interaction went on between the
two, but toxins were active in such extraordinary minute quanti-
ties, and also were produced in such minute quantities, that
chemical methods of analysis revealed nothing beyond the
probability that they were to be regarded as protein degradation
products. The biological test was far more delicate than the
chemical. We were, in fact, beyond the realms of chemical
technique. If confusion were not to be worse confounded, it
was necessary to have some working plan whereby to visualize
the processes that went on in the reaction between the micro-
organism and its products and the organism. This Ehrlich
afforded in his celebrated side-chain theory. Some may to-day
criticize that theory, and may gird at its toxins, toxoids, and
toxones, its haptophores of the first, second, and third orders,
its end-piece and mid-piece of the complement. But unquestion-
ably it afforded a working scheme, whereby a common basis
for expression of phenomena was afforded, and, equally un-
questionably, it promoted an enormous burst of activity in the
study of immunity problems. What is more, although Ehrlich
cautiously refrained from suggesting or premising any particular
element in the cell as being provided with side-chains, his theory
was based upon the only rational conception of cell metabolism —
that which regards the molecule of living matter as of extra-

252 ON VARIABILITY AND ADAPTATION

ordinary complexity, and in a state of constant unsatisfaction,
built up by linking on other simple molecules, and as constantly,
in the performance of function, giving up or discharging into
the surrounding medium these and other molecular complexes
which it has elaborated.

In my Principles of Pathology, published in 1908, I called
attention to certain defects in Ehrlich's theory, and pointed
out how it might be brought into fuller conformity with actual
knowledge of general cell activities and of enzyme action. I
laid down that throughout the whole study of immunity we deal
with " the methods by which living matter, whether animal
or vegetable, reacts towards other living matter, whether animal
or vegetable, and towards the products of the same which come in
contact with it, and although," I continued, " the statement may
at first encounter appear to be both novel and extreme, further
thought will confirm it ; the problems of immunity narrow
themselves down to special problems bearing upon the assimila-
tion and digestion of unusual proteid matter." I might have
added, for my treatment of the subject was consistently along
these lines, that they resolve themselves into studies of enzyme
action.

Long years ago, in the early 'eighties, I remember Sir Michael
Foster teaching us that the assimilation of food-stuffs evidently
demanded that more complex molecules of food-stuffs were not
built into the cell direct, but were first broken down into their
simpler constituent molecules, and of these the molecules common
to the constitution of the foreign food-stuffs and to the cells are
taken up by the latter, and used for building-up purposes. This
view to-day is universally accepted, more particularly as the
result of Emil Fischer's studies upon the chemical constitution
of proteins, and his demonstration that these are complexes
formed by the union of varying proportions of mono- and diamino-
acids. The Germans speak of this as the " Baustein," or building-
stone theory.

It is a familiar fact that proteids taken as food, while
they disappear from the intestinal contents, are not to be
detected in the blood and lymph coming from the intestinal
tract. And this is true not only of the fully formed proteins,
but also of the peptones and proteoses of the alimentary canal
resulting from the disintegration of the larger protein molecules

IMMUNITY AND DIGESTION 253

into simpler soluble proteins of the second order. Indeed we
have this remarkable fact that if, instead of allowing them to
be absorbed and acted upon by the intestinal mucosa, we intro-
duce these peptones and proteoses directly into the circulation,
they are found to be highly toxic. There is evidently, therefore,
a pronounced difference between enteral or intestinal, and 'paren-
teral digestion, the breaking down, that is, of foreign proteins
introduced into the organisms outside of, and without the
intermediation of, the digestive tube.

We owe to Victor Vaughan, of Ann Arbor, and his long con-
tinued studies from 1901 onwards, the demonstration of the fact
that outside the body, by the simple procedure of digestion with
several times its bulk of absolute alcohol in the presence of 2 per
cent of sodium hydroxide, any protein — even so apparently
harmless a body as egg albumen — is split up into a poisonous
and a non-poisonous moiety, the former soluble in alcohol, the
latter insoluble, and this poisonous moiety, when introduced
into the system other than through the intestine, is rapidly fatal
in doses so minute as from 8 to 100 mg. per guinea-pig. Evidently
in the process of passage through the intestinal mucosa, either
this poisonous moiety of the protein molecule is detoxicated,
or protein disintegration there is so carried on that it is never
given off.

The high significance of these observations lies in this, that
the bacterial bodies are largely compounded of proteins, that
bacterial bodies similarly treated give these same poisonous
and non-poisonous moieties : that the process of destruction
of bacteria that have gained entrance into the tissues is one of
parenteral digestion, and that the phenomena associated with
parenteral digestion must therefore be closely studied in order
to understand the nature of infection and the phenomena of
immunity. It is this study that has ushered in the latest phase
of our subject ; and in this subject two names stand out par-
ticularly, those of Victor Vaughan and Abderhalden. Perhaps
it would be equally correct to state that this latest phase corre-
sponds with the study of anaphylactic phenomena, and of their
relationship to immunity, for, as I shall proceed to show, the two
are most intimately connected.

254 ON VARIABILITY AND ADAPTATION

Anaphylaxis

If I may judge from the answers received recently from
the candidates for Dominion Medical Registration, anaphylaxis is
evidently regarded by the majority as one of those subjects that
no fellow can be expected to understand ; and addressing a general
medical audience it is essential to describe its broad phenomena.
Briefly, then, whereas, as everybody knows, the introduction of
foreign proteins into the economy is in general followed by the
development of immunity, the system gaining the power of
rendering harmless many times the fatal dose of that protein,
it is found that, employing minute doses of the protein for the
first injection, if now, at the end of a week or ten days a second
dose be injected, larger than the first, but still many times less
than that producing serious symptoms on first injection, then
instead of showing evidences of immunity, on the contrary the
animal has become hypersensitive. It may, indeed, die within
three to five minutes. In the guinea-pig, which shows most
strikingly these phenomena, there is an intense contraction of
the muscular walls of the bronchi, whereby dyspnoea is brought
about, with overfilling of the lungs so extreme that even when
the chest wall is freely cut open immediately after death, the
lungs remain expanded. With this there are muscular cramps
and obstruction to the onflow of blood with capillary congestion
leading to scattered haemorrhages. This state can be induced
by inoculation of any foreign protein, even one ordinarily so
harmless as egg albumen, or, on the other hand, by the employ-
ment of pathogenic bacteria and their proteid products.

Now, here is the significant point. The symptoms and lesions
produced by the poisonous moiety of split protein are, as Vaughan
was the first to point out, identical with those seen in anaphylaxis.
In other words, anaphylaxis is due to the sudden setting free
in the economy of poisonous bodies of the same nature as the
poisonous disintegration moiety of proteins. On the doctrine,
therefore, that like results are produced by like substances, we
arrive at the conclusion that this hypersensitiveness or ana-
phylaxis is essentially due to a disintegration of the introduced
proteins with rapid liberation of the poisonous moiety.

But the first dose introduced sets up none of these phenomena,
the second does. It is obvious that, following upon the intro-

VAUGHAN AND ABDERHALDEN 255

duction of the foreign protein, the organism has gained a new
power, that of splitting up the foreign protein.

It is at this point that Abderhalden's observations acquire a
high significance. In 1910 Abderhalden with various pupils
published a remarkable series of " serological studies by means
of the optical method." Starting from the fact that the different
proteins and peptones present each in the polariscope a specific
index of rotation, he demonstrated first that, as determined by
the polariscope, the ingestion through the alimentary tract of
any particular protein has absolutely no effect upon the blood
serum ; its rotatory power is unaltered from the normal. But
inject any of the more common proteins subcutaneously or
directly into the blood, and in twenty-four or forty-eight hours
remove some of the blood, separate the plasma, add to this a
little of the protein, and now a remarkable change is seen :
the index of rotation of the plasma undergoes progressive altera-
tion, a clear indication that the protein is being broken down,
and this is confirmed by the fact that by dialysis peptones can
be separated off. In other words, within forty-eight hours
after the introduction of a foreign protein, the system has re-
sponded either by elaborating a proteolytic or proteoclastic
enzyme, or, it may be, by the rapid increase in production and
discharge of already existing proteoclastic enzyme present in
certain cells. What I would especially emphasize is the fact
that the enzyme thus first liberated is not specific. It is an
indifferent proteoclastic ferment : it will act not only on the
particular protein inoculated, but on other proteins : it will
break down indifferently globulin, casein, edestin, gelatin, and
their peptones, with the evolution of toxic peptone-like bodies.

It might be laid down that this indifferent ferment first pro-
duced is the cause of the anaphylactic phenomena. Against this
view is the fact that Abderhalden's ferment is recognizable in
twenty-four to forty-eight hours, whereas anaphylactic pheno-
mena are not obtainable until the sixth to the tenth day. This
in itself is not absolute disproof, for one of the factors governing
the rate of enzyme action is the amount of enzyme present, and
this might well depend upon the accumulation of a sufficient
amount of the enzyme. The amounts should be large to cause
dissociation of the foreign proteid at such a rate and such an
extent as will lead to the production of symptoms.

256 ON VARIABILITY AND ADAPTATION

There are, however, serious objections to this view. As
pointed out by Abderhalden, the anaphylactic phenomena are
specific ; it is only the protein first introduced that sets up
disturbance ; on subsequent injection other proteins are not
affected, whereas Abderhalden's proteolytic and peptolytic
enzymes are active in respect to any protein. The only safe
conclusion is, therefore, that in these data we have evidence of
the development of two stages :

1. Elaboration and discharge into the blood plasma, within
forty-eight hours, of indifferent proteoclastic and peptolytic
enzymes, affording Abderhalden's phenomenon.

2. The gradual production of more highly developed specific
proteoclastic enzymes, having the power of rapid dissociation
of the particular protein introduced, with liberation in the course
of a few minutes of sufficient toxic disintegration product or
products to set up symptoms.

I should note in passing that it cannot be regarded as ab-
solutely settled as yet whether it is the enzymes liberated into
the blood plasma that are the essential cause of the anaphylactic
symptoms, or whether the phenomena are due to intracellular
dissociation in specific groups of cells, such as certain cell groups
and centres in the nervous system. If these cells under the
action of the first dose have been stimulated to produce and
consequently store the non-specific enzyme, the moment the
foreign protein comes into contact with their cytoplasm, or is
absorbed by them, it will undergo degradation. An extra-
ordinarily minute dose of foreign protein in this way, by acting
on or being acted on by specific cells, might set up fatal results.
The rapidity with which anaphylactic phenomena may show
themselves, and that upon the introduction of extraordinarily
small amounts of any particular protein, rather favours the latter
view : it suggests that there is a selective taking up of the protein
by particular cells, cells whose due and ordered action is essential
for continued existence.

Immunity

What I want to bring specially before you is that the two
stages above postulated lead up to a third ; in other words, that
anaphylactic phenomena afford us explanation of the nature
of immunity. By cautious reinjection of the foreign disharmoni-

ANAPHYLAXIS AND IMMUNITY 257

ous protein we can pass through the anaphylactic to the succeed-
ing stage of immunity. Now the characteristic of a true bacterial
(as distinct from an antitoxic) immunity is that the body fluids
gain remarkable digestive or fermentative powers. Any one who
has observed Pfeiffer's phenomenon — has seen living typhoid or
cholera organisms swell up and dissolve like pieces of sugar in
the serum of a highly immunized rabbit or guinea-pig — cannot
fail to be impressed by the extraordinary digestive powers that
have been acquired by the body fluids through this process of
immunization.

Everything here indicates that we deal with proteoclastic
enzymes. Only now, in the immune animal, the process no longer
leads to acute poisoning. I admit that more than one explanation
may be sought and given for this fact, but, choosing the simplest,
we see here a further evolution of the same process, namely,
the development by the organism of further enzymes which
disintegrate the poisonous moiety, and so carry on the degrada-
tion of the foreign protein to a further and harmless stage. It
is pleasant to think that we owe to one of our own workers,
Dr. Fraser B. Gurd, the first experimental evidence of the co-
existence of these two orders of enzymes (lysins) in the immune
animal.1 As he points out, it is not that the earlier enzymes are
completely replaced : they are still active and indeed abundant,
and evidence of the presence of these sensitizing bodies can be
obtained in immune animals. But after they have broken
down the specific protein with the production of the poisonous
substance, this is immediately acted on by the new enzyme and
rendered innocuous.

To me it is not a little curious that Abderhalden, Thiele, and
Embleton, and the other workers in this territory, while familiar
with the existence of the great numbers of individual enzymes
that are to be isolated from the organism, have not taken up and
considered this probability of, or recognized the indications
pointing to, the production of a succession of progressively more
specific and more active enzymes in the development of the
immune state.

1 Amer. Jour. Trop. Dis. and Preven. Med. i., 1914, 776 ; Jour. Med. Fes.
xxxi., 1914, 205.

258 ON VARIABILITY AND ADAPTATION

Enzymes and Ectotoxins

There is one point which, before closing, I must take up,
namely, the nature of ectotoxins.

In the first full study of these ectotoxins, that, namely, by
Roux and Yersin a quarter of a century ago upon the diphtheria
ectotoxins, it was clearly pointed out how many properties these
bodies possess in common with the group of organic enzymes :
their activity in extraordinarily minute doses ; the impossibility
of isolating them in a pure state ; their destruction by a tempera-
ture of 60° centigrade ; their precipitation by phosphates and
other salts. It is not a little interesting how, notwithstanding,
bacteriologists in general have shown unwillingness to accept
this view. Long years ago Sidney Martin in London brought
forward further evidence to the same effect, that the diphtheritic
membrane was not in itself highly toxic, that it contained an
enzyme whose activity gave origin to toxic albumoses, and that
the internal organs, while showing no evidence of the toxin proper,
might come to contain these toxic bodies. The significance of
these observations lies in the fact that, as everybody knows,
diphtheria bacilli are present in quantity in the membrane,
and are almost absent from the internal organs. I have not
come across a single leading German work on bacteriology that
has taken any serious notice of these observations of one of
the most brilliant and capable English workers. The stumbling-
block has evidently been in the prominence obtained by the
phenomena of antitoxin production. This reactive production
through the agency of ectotoxins of bodies which combine with
the ectotoxins to render them inert, is whollv outside our usual
conception of enzyme activities, and yet it deserves pointing
out that, if we take solutions containing what we recognize as
enzymes foreign to the animal body, enzymes, for instance,
obtained from plants, and if we inoculate these into the animal
body, anti-enzymes are developed, identical in all their properties
to the antitoxins. Dr. Zinsser of Columbia University, in his
excellent work upon Injection and Resistance, lays down that there
is an important difference between ectotoxins and enzymes in their
mode of action. He states : " While the toxins are apparently
bound or neutralized by the tissues they attack, the action of
an enzvme seems rather to be a process in which the enzvme

THE ENZYME NATURE OF ECTOTOXINS 259

unites with the substance it acts upon, is released as this result
is obtained, and freed for further action without noticeable
loss of quantity."

Here there is this very confusion that I have already noted,
that, namely, of dwelling only upon the combination of toxin
and antitoxin in the one case, and of the action of the enzyme
upon the substrate in the other.1 Only two paragraphs lower
down Zinsser calls attention to the incubation period in the action
of toxins, pointing out that this is longer when small doses are
given, shorter when large doses are administered, but always
evident. Thus in the horse he points out that it may be four
or five days before the action of tetanotoxin shows itself : in
mice, as De Waele points out, animals extremely susceptible
to tetanotoxin, a minimal lethal dose gives an incubation period
of thirty-six hours ; if 3600 lethal doses be administered this is
shortened to twelve hours, and whatever the dose, the interval
cannot be shortened below eight or nine hours. Such facts as
these cannot be explained as due to the activity of an ordinary
mineral or alkaloid poison : ordinary poisons have no incubation
period. It is wholly consistent with the action of a proteoclastic
enzyme which, present in the blood or becoming absorbed by
certain cells, by acting upon some specific proteid substrate,
splits off a toxic moiety which, accumulating under the continued
action of the enzyme, at length is present in sufficient amount
to set up symptoms. The full proof of the enzyme nature
of ecto toxins has recently been afforded by Abderhalden, who
has demonstrated in vitro the proteoclastic properties of these
bacterial toxins. In fact we must conclude that just as the animal
body gives off free enzymes along the course of the alimentary
canal, whereby foreign proteins of the food-stuffs are split up,
and so made capable of absorption, so certain bacteria have a
like power of discharging enzymes which can act upon the proteins
of the animal economy and prepare them to be assimilated by
the bacterial body, and it is in this process of disintegration of
the proteins of the organism that the toxic moiety is liberated,
thus inducing the symptoms of disease. It is these, and not
the so-called ectotoxins, which are the immediate causative
factor of these morbid phenomena.

1 [He passes over, without notice, the phase of disintegrative activity of
the toxin.]

260 ON VARIABILITY AND ADAPTATION

Thus, to conclude, I would lay down the following as summing
up the views here put forward as most satisfactorily harmonizing
the data at present in our possession.

1. Whatever the nature of enzymes, the organism has the
power of elaborating new orders of these bodies in and from its
cells in response to the entrance into the tissues of foreign pro-
teins, and potential food-stuffs in general.

2. The development of anti-bacterial immunity in an animal
is essentially the development of the power of parenteral digestion
by the tissues of the constituents (mainly protein) of the patho-
genic micro-organisms.

3. This is rendered possible by the elaboration of a succession
of (mainly) proteoclastic enzymes.

4. Three stages are to be recognized in this process :

(a) The development of indifferent proteoclastic enzymes.

(b) The development of specific proteoclastic enzymes,
splitting the bacterial proteins into highly poisonous and non-
poisonous moieties.

(c) The development of toxoclastic enzymes which render
the disintegration products innocuous.

5. Ectotoxins are to the micro-organism what discharged
enzymes are to the animal. They have all the properties of
enzymes.

6. They are not themselves toxic, but proteoclastic, and it
is the products of their activity upon the proteins of the organism
that are the essential toxic substances.

7. Antitoxic immunity is thus wholly distinct from anti-
bacterial immunity. It is a process not primarily of digestion
and disintegration, but, rather, of combination and fixation of
enzyme, so that this is no longer capable of attacking and dis-
integrating successive complex proteid molecules ; or, it may be,
one of development of toxoclastic enzymes by the tissues, which
disintegrate the toxic proteins as they are produced, rendering
them innocuous : or both.

PART III
ON GROWTH AND OVERGROWTH

261

CHAPTEK I

ON GROWTH AND OVERGROWTH AND ON THE RELATIONSHIP
BETWEEN CELL DIFFERENTIATION AND PROLIFERATIVE
CAPACITY ; ITS BEARING UPON THE REGENERATION OF
TISSUES AND THE DEVELOPMENT OF TUMOURS *

(1900)

There are in medicine and other sciences not a few beliefs and
ideas which we have taken up we know not how, and which in
general we no more think of discussing than we do, for example,
the subject of good manners. Thus, just as it is difficult, if not
impossible, for us to state how or where we gained any single
article of our code of personal ethics, so it is with certain of
these general ideas in medicine ; and while we have never
discussed these beliefs, we feel individually assured that they are
the current ideas of other workers along the same lines.

One of these tacit beliefs or comprehensions relates to the
subject of the growth of tissues. Asked off-hand we should
assuredly, each one of us, state as his familiar belief that muscle
arises from muscle, epithelium from epithelium, nerve cell from
nerve cell, and so on, and the mental picture which we form of
the process of growth is, I fancy (though here, of course, I speak
under correction), that the fully formed epithelial cell undergoes
mitosis and divides into two, and so with the cells of other tissues.
Yet if we think a little longer and recall what we have actually
seen under the microscope, this mental picture is seen to be
incorrect or, at least, imperfect. Thinking recently over this
matter it has been impressed upon me that if we obtain a correct
idea of what occurs during the process of growth of tissues, we
not only realize that there is a very broad biological law under-

1 Published in the Festschrift in honour of Dr. Abraham Jacobi, New York,
1900, 422, and Medical Chronicle, Manchester, June 1900.

263

264 ON GROWTH AND OVERGROWTH

lying this process of growth from the earliest stages of the embryo
upwards, but further, applying this biological law to the subject
of pathology, we gain a deeper and a fuller comprehension of
certain matters bearing upon regeneration and degeneration of
tissues, and, what is more, bearing very directly upon certain of
the phenomena of tumour growth.

While the individual, formed as he is of a marvellous complex
of various tissues, is a reproduction of the tissues and organs
present in his paternal and maternal ancestors, he is the outcome
and development, not of the combined fully developed tissues,
nor again of any one highly differentiated cell or cell compound,
but of a single undifferentiated cell — the fertilized ovum — a
cell neither of the component parts of which, ovum or sperma-
tozoon, has ever through the whole course of the ages been
derived from other than similar undifferentiated cells ; or, in
other words, these apparently simple germ cells are capable of
giving rise to the whole series of cells forming the whole mass
of tissues from the simplest connective up to the most highly
differentiated nerve tissue.1 If we study the process of the
development of the embryo, whether in plant or animal life, we
see again a somewhat similar phenomenon.

a

Mother Cells " and Tissue Growth

From a very early period in such growth we recognize that
certain cells alone appear to be actively dividing and to be
actively proliferative, whereas other cells, the products of these,
while they take on characteristic appearances, do not thus divide.
Very early indeed in the developing embryo we recognize this
existence of what may be termed " mother cells," — cells which
themselves remaining embryonic in type give rise by division
to other cells which assume more highly differentiated characters.
And this is true not only of animal organisms, but still more

1 Throughout this article I refer to cytoplasmic and not nuclear differentia-
tion. It is possible that while the cell body of these germ cells is relatively
undifferentiated, the nucleus is peculiarly highly elaborated. We have, indeed,
not a few indications that this is the case. Thus it may be that nuclear and
cytoplasmic differentiation, in germ and " mother " cells, and in specific
tissue and " daughter " cells respectively, are in inverse ratio. We have not
as yet reached a point at which any definite statement can be made in this
connexion.

((

MOTHER CELLS " 265

obviously of plants : we have but, with Sachs, to study the
growing points of plants to realize the existence of these " mother
cells."

Now when we pass to the fully formed individual, we must
still recognize this fact, which we are apt to pass over, namely,
that the specific cells, if I may so term them, of the different
tissues — that is, the highly differentiated and characteristic cells
of those tissues — do not themselves give rise directly, and by
division, to other specific cells, but that in each tissue there are
(1) these more or less undifferentiated mother cells, which more
especially have the power of proliferating, and it is the daughter
cells which assume the full specific properties in the different
tissues, or (2) if, as is seen to be the case in some instances,
the active cells of certain not very highly differentiated tissues
themselves proliferate, they only do this after a preliminary
reversion to a " mother cell," or more embryonic type.

Let me here give a few examples. Immediately I mention
them, every one will realize that these are facts thoroughly
familiar. One of the simplest cases is that of the skin : in this
it is the lowest layer of the epidermis, the Malpighian layer,
whose cells undergo active proliferation, and this constantly,
throughout the whole period of life, and the cells which are most
active in the process are those of the very lowest layer — cells
which are of a permanently embryonic type.

After injury, it is true, we may come across occasional mitotic
increase, more especially in the lower animals, in the region of
the prickle cells, but, when we do, we see that it is accompanied
by reversion of the cell to a simpler type : the prickles or
bridges disappear, the cell loses its connexion with the
neighbouring cells, and passes back to a simpler stage prior to
multiplication.

In bone the case is the same. Production of bone tissue is
not brought about by multiplication of the pre-existing and
characteristic bone cells, but by osteoblastic tissue ; or, in other
words, by the mother cells of bone lying in the immediate
neighbourhood of the vessels of the inner layer of the periosteum,
or which course through the Haversian canals and between the
laminae. When through injury or otherwise a stimulation is
afforded, it is these mother cells which, applying themselves to
the pre-existing bone, multiply, and there governing the spaces

266 ON GROWTH AND OVERGROWTH

around them lead to the deposit of a calcareous or bony matrix
and gradually assume all the characters of typical bone corpuscles.

From the variability of the structure of the glandular organs
it is difficult to lay down any general rule with regard to re-
generation of glandular epithelium, so that it must be confessed
that, partly on account of this variability, partly on account of
the fact that the different observers upon glandular regeneration
have not had this conception of mother cells before them in their
studies, such glandular epithelium in the present state of our
knowledge affords the least satisfactory demonstration of the
principle. There are, however, so many instances concerning
which we have positive information as to render it more than
probable that throughout the same principle is in action.

Of the mother cells of the type seen in the Malpighian layer
of the epidermis, several instances may be called to mind : the
mother cells of the testicular epithelium and (though differing
somewhat in plan) of the ovarian follicles, of the sebaceous and,
apparently, from Tornier's studies, the mammary glands. In
both of the latter organs there is, during activity, a giving off of
cells, and mitosis occurs especially in the deeper, more external
layers.

In this connexion should, I think, be mentioned the columnar
and ciliated epithelia of mucous membranes. It is remarkable
how one histological text-book after another figures such columnar
epithelium as formed of a palisade of fully developed columnar
cells in regular series situated upon the basement membrane,
without indicating or even referring to the presence of embryonal
mother cells lying between the bases of the fully differentiated
columnar cells. Certainly under pathological conditions one is
impressed by the fact that these basal cells are present, and that
there is a development of new cells from beneath to take the
place of the degenerating fully formed cells. While in the simpler
acinous glands it is difficult to recognize such permanent mother
cells (indeed they are probably absent in the adult), here at
least they are present.

In the lymph glands we may have another arrangement. In
them the mother cells are collected together in the form of small
nodes in the centres of the individual follicles ; a somewhat
similar arrangement of foci of embryonal cells is met with in the
thyroid, while in the pancreas, Langerhans's bodies or " cell

MOTHER CELLS AND DEDIFFERENTIATED CELLS 267

accumulations " would seem to be of a similar nature, although
this is not as yet absolutely determined.1

Closely allied to this condition is what is seen in the lung,
and again in the serous coat of the peritoneum (though neither
of these, it is true, comes strictly under the category of true
gland structure). In the former, certain more embryonic cells
persist, especially at the angles of junction of the alveoli, and
these are generally regarded as centres for the normal proliferation
and regeneration of the alveolar epithelium. On the peritoneal
surface also there are found clumps of smaller cells capable of
active proliferation. With regard to both these cell layers,
the flattened — differentiated — cells are liable, under irritation,
to revert to the fuller, more rounded embryonic condition and
then undergo mitosis and proliferation.

Judging from the fact that mitoses are not observed in the
fundal portions of several acinous glands, while they are frequent
in the neck regions of the same, it would seem at least probable
that mother cells, or cells capable of reverting to the mother cell
type, are situated in the latter area. But generally in acinous
glands of simple type there is, as Bizzozero has pointed out, little
evidence of continued proliferation under normal conditions,
once they are fully formed. In connexion with these it is
that more exact studies are requisite ; certainly in some, e.g. in
the kidney tubules, slight continued irritation leads to the
differentiated cells undergoing proliferation, but at the same
time these cells are seen to revert to the simpler, more embryonic
type. Thus, while confessing that it is not as yet possible to
demonstrate in every case that either mother cells are present
or that for regeneration the differentiated cells revert to a more
embryonic type, the number of examples that can be brought
forward, where either one or other of these processes is seen to
occur, is so large that we may assume that a general principle
obtains in connexion with all.

Passing now to another tissue, namely muscle, the process of
regeneration, as pointed out by Barfurth, varies largely according
to the age of the individual. Originally the muscle fibres, it is
needless to say, are recognized as arising from a series of cells

1 [Despite the abundant studies upon Langerhans's bodies which have been
made in the last seventeen years, the " mother cell " nature of these bodies is
still a subject of debate.]

268 ON GROWTH AND OVERGROWTH

known as sarcoblasts, relatively large cells with abundant proto-
plasm, the nuclei of which tend to proliferate ; and along one
aspect of, or generally around, the periphery there is gradually
developed first a longitudinal fibrillation, and then a transverse
striation. In young larval forms of amphibia, for example,
after injury such sarcoblasts may be seen to be present, to undergo
multiplication, and to give rise to typical muscle fibres ; here,
therefore, very clearly there are mother cells. The existence
of such in the adult is still a matter of dispute, for in the older
individuals two processes have more especially been recognized :
one, that in which the fully formed muscle fibres become the
seat of an active nuclear proliferation, and then in certain areas
of the fibres the striation, more especially toward the end, or
laterally, becomes indistinct and little buds containing nuclei
are given off, which become the precursors of the new growth.
Or secondly, as pointed out first by Weismann, there may be
recognized, even in perfectly normal muscle of adult animals,
certain occasional spindle-like bodies, consisting of short striated
muscle fibres, characterized by very abundant nuclei, and,
according to Kolliker and others, these bodies represent centres
from which by longitudinal division new fibres are given off.
Even in this case it will be seen that these bodies represent a
lower imperfect stage of differentiation, and the richly nucleated
mass may strictly be regarded as a " mother cell." 1

Thus from a study of the process of regeneration as seen in
the different tissues in man and vertebrates generally, it is
possible to lay down the following laws : —

1. The fully differentiated cells of a tissue proper never arise

from cells that are themselves fully differentiated.

2. Under the normal conditions of growth and during physio-

logical regeneration, the fully differentiated cells would
seem in nearly every case to be developed from " mother
cells " — undifferentiated or but partially differentiated
cells which themselves throughout the term of life of the
individual never attain to the full differentiation peculiar
to the tissue to which they belong and which indeed
they produce. For these mother cells by division give

1 [Since 1900, when this paper was written, the general consensus of opinion
has veered round to a belief that these muscle spindles are not associated with
regeneration.]

THE LAWS OF CELL MULTIPLICATION 269

origin to the daughter cells, and it is the daughter cells
which attain full differentiation and form the specific
cells of the tissue. More rarely the functional cells
themselves by reversion to a more embryonal type can
take on the properties of mother cells.

3. Under abnormal conditions, the fully differentiated function-

ing cells of certain tissues are capable of proliferation and
giving rise to cells of like nature, but this is only after
a preliminary reversion to a simpler, more embryonic
type. The fully differentiated cell as such is incapable
of proliferation.

4. Or, otherwise, the energy stored up by the cell may be

expended in one of two directions, but not in both —
either in functional activity or in preparation for pro-
liferation ; and, the structure of a cell being the expression
of the activity of that cell, the expenditure of energy
in either direction is attended by corresponding mor-
phological or structural differences in the cell.

5. The more highly differentiated a cell, the more highly

elaborated its structure, the less the ease with which it

reverts, and the less the liability to reversion to a simpler

reproductive form ; the simpler the cell, the greater

the ease and the greater the liability to such reversion.

It is thus possible, at the one extreme, to conceive cells

so simple in function and in structure that functional

or reproductive activity may be called into play without

recognizable preliminary structural alteration, and, at

the other, cells so highly differentiated that the capacity

for proliferation has become entirely lost.

The law with its corollaries here given does not so far appear

to have obtained general recognition. Certainly I fail to find it

explicitly referred to by writers upon morbid anatomy, and had

it been currently accepted by biologists it would ere this have

become part and parcel of pathological teaching. It is only

after some little search and almost by chance that I find its

main paragraph laid down by Kolliker in 1885, and that only,

as it were, in passing, in an article not dealing directly with

regeneration, but dealing with the part played by the nucleus in

inheritance. In that article Kolliker states : ' In all cases in

which an organ or a tissue is capable of regenerating itself, it

270 ON GROWTH AND OVERGROWTH

must contain elements of an embryonal character, or at least
such elements as can take on that character." And again, in
the same article : " These cells then are dependent upon the
same law as, in the embryo, is the formation of organs." 1

But in the general portion of his great Handbuch der Gewebe-
lehre des Menschen, published in 1889, I cannot find that he
again refers to this principle, hence it would appear that he did
not realize its wide application.

Thus, although the law when once grasped is simple, almost
to the verge of being a truism, it merits being stated for once in
as clear terms as possible that its high importance may be fully
recognized.

Regeneration

Applying now these considerations to certain pathological
conditions, it is in the first place necessary to say but little
concerning pathological regeneration and repair, for I have
already applied some of the facts gained from a study of such
regeneration in the development of my argument. This much,
however, may be said, that the more we study the regeneration
of various tissues, the more we see that the phenomena of the
process fall into place in consonance with the law above laid
down. The cells of a regenerated tissue are derived from the
pre-existing cells (and mother cells) of that particular tissue,
and the more highly differentiated the tissue the less is its capacity
for regeneration.

Of all cells, those which are the most highly differentiated,
namely the nerve cells, show the least capacity for proliferative
regeneration ; there is a remarkable lack of evidence of fully
developed nerve cells having been observed in the process of
mitosis, and, corresponding with this lack, when we consider the
histological basis of motor, sensory, and mental acts, it is a
priori hard, if not impossible, to conceive a given cell in any of
the higher nerve centres undergoing proliferation without there

1 Kolliker : " In alien Fallen, in denen ein Organ oder ein Gewebe fahig
ist sich wieder zu erzeugen, muss dasselbe Elemente von embryonalem Charakter
enthalten oder wenigstens solche, die diesen Charakter anzunehmen im Stande
sind." " Diese Zellen bedingen dann nach denselben Gesetzen, wie beim
Embryo, die Organgestaltung " (" Die Bedeutung der Zellenkerne fur die
Vorgange der Vererbung," Zeitschrift f. wiss. Zool. vol. xlii., 1885, quoted
by Barfurth " Zur Regeneration der Gewebe," Arch. f. mikrosk. Anat. xxxvii.,
1891, p. 424).

REGENERATION 271

being simultaneous arrest, not to say loss, of function. Every-
thing indicates that of all the cells of the body the neurones are
those endowed with the longest life period.

During foetal or larval life, prior to the attaining of full
function, such proliferation has been recognized (Barfurth), or
otherwise the incompletely differentiated nerve-cells may take
part in the regenerative process ; though it may be considered
an open question whether the new development of neurones in
experiments upon the spinal cord of the larval newt and other
amphibia is the result of proliferation of developed neurones or
of mother cells.

That such mother cells must exist even in the human nervous
system has been forcibly brought home to me by a study of
certain sections of the spinal cord to which Dr. Shirres,1 working
in my laboratory at the Royal Victoria Hospital, recently called
my attention.

These sections are from a case of porencephalus in which
apparently, from all indications, the atrophy and disappearance
of practically the whole motor area in the left hemisphere dated
from the last month or two of foetal life. Despite this practically
complete loss of the entire motor area on the left side, the result-
ing evidence of paralysis and disturbance of function were little
more than a relatively slight want of development of the bones
and muscles of the right side of the body together with paresis
of the musculature of the right hand. In the cord the difference
between the two motor tracts is most marked, there is entire
absence of the crossed pyramidal tract on the left side, of the
direct pyramidal tract on the right, and, corresponding with this,
there is a most marked atrophy of those groups of cells towards
the periphery of the anterior horn of the grey matter on the
right side, which everything indicates as being connected with
the crossed pyramidal tract and being what I may term the
spinal motor centres for the upper extremity. But in sections
in which this atrophy is most marked there is a distinct increase
in the number of ganglion cells in the deeper and more central
areas of the grey matter ; here the number of cells is much above
the normal, the cells are large and more numerous than those
of the left side of the same section.

For our present purposes it is clear that these abundant cells

1 [Studies from the Royal Victoria Hospital, Montreal], No. 1, 1901.

272 ON GROWTH AND OVERGROWTH

cannot be regarded as due to proliferation of the atrophied cells
— their relative position forbids such an assumption. Their
existence, however, is strongly in favour of the view that they
have been derived from mother cells in the grey substance.

The regeneration of nerve fibres is another matter. So far
as regards the axones, or axis-cylinder processes themselves, it
is strictly comparable with the new growth of the cell substance
of the amoeba when a part of that cell substance has been removed
without direct injury to the nucleus. As regards the investing
sheath of the medullated nerve fibre, regeneration here is in
strict accordance with the third law ; there is a preliminary
period of reversion ; the formed substance (myelin), governed
or elaborated by the cell, breaks down, the nuclei become more
prominent and surrounded by an increased amount of protoplasm,
and then in place of a single nucleus of a node of Ranvier one
gets numerous nuclei, and the new cells evidently surround the
newly developed axis cylinder as it grows downwards from the
seat of injury.

I have already referred to the regeneration of muscle, but if
we now pass to the other extreme and consider what happens in
the simplest of all tissues, namely, white fibrous connective
tissue, we find that even here for pathological regeneration to
occur there is a preliminary stage of swelling of the connective
tissue cells, the nuclei become larger and more prominent, the
surrounding protoplasm becomes increased in amount, the
surrounding fibrils tend to disappear, and now, only after this
reversion to a simpler, more embryonic condition, do we have
proliferation and the subsequent building up of new connective
tissue.

To enter here into a discussion as to whether connective
tissue can be formed from wandering cells, and whether any
special form of wandering cell is capable of taking part in such
new formation, would be imperative if I were dealing more
especially with the law that tissues arise from pre-existing specific
tissue cells ; here rather, dealing with the process and the law
of growth, with the law of mother cells, if I may so term it, the
very limits to which this article is confined must be an excuse
for not dealing with this subject.

METAPLASIA AND NEOPLASIA 273

Metaplasia

ifew words, however, must be said with regard to metaplasia
m §e\eral, for the law of reversion to a more embryonic type is
distinctly in evidence throughout, and indeed helps us to under-
stand iho^ for example, a columnar-celled epithelium can be
converted into a squamous-celled epithelium, and one form of
connective tissue into another ; or, otherwise, metaplasia is
never direct, but is only brought about by preliminary reversion
to a more embryonic type, or, where mother cells are present,
by the modified development of cells derived from the mother
cells, the influence of environment altering the character of those
daughter cells during the period of growth.

Neoplasia

If the principle be correct that the more highly differentiated
a cell, the less its capacity for proliferation ; or, in other words,
the more the daughter cells depart from the type of the mother
cells, the less their proliferative capacity, then the converse
would seem to hold — that the more the daughter cells in a given
tissue retain the characteristics of the mother cells, the greater their
proliferative capacity. Or — to continue the argument — if through
any cause the daughter cells in a tissue do not attain full specific
differentiation, then they are peculiarly liable to proliferate and
function, not as specific cells, but as mother cells. And further,
as Bizzozero has pointed out, tissues in which the cellular elements
exhibit frequent mitosis are those which more especially are liable
to be the seat of excessive growth and tumour formation.

I am far from saying that this is the one principle concerned
in the development of the blastomata, or tumours proper, but
I would urge that it is one important principle. It is outside
the limits of this article to consider what is the essential stimulus,
or what are the stimuli, leading to neoplastic growth ; accepting,
however, for the moment, that, whether temporarily or per-
manently, some stimulus or stimuli be in action, at least we
gain a comprehension of why in the first place histoid tumours
(in which one or other tissue is more or less faithfully reproduced)
are essentially benign ; and in the second place why malignancy
and excessive cell proliferation, which is the essence of malig-

T

274 ON GROWTH AND OVERGROWTH /

nancy, go hand-in-hand with tumour formation of a prono1 onced
embryonic type. j

The failure to recognize the normal existence of mothe: f cells
in those tissues which are prone to tumour formation k> very
obvious in reading over Cohnheim's chapter upon turnouts and
in following the course of the argument which led him to ccvnclude
that tumours arise from latent superfluous embryonal 'tissue
lying either within a tissue of the same nature or heterotopic.

While we must admit the existence of such embryonal rests
in order to explain the development of such tumours as, for
example, myomata of the kidney and primary cancerous growths
developing in bone, the permanent existence of these mother
cells as the normal constituent of the tissues throughout life is
wholly adequate to explain most benign and malignant tumours.
To this larger conception of what is " embryonal tissue " we
must, I think, inevitably come, and with Senn and others we must
regard tumours as products of tissue proliferation of embryonic
cells of either congenital or post-natal origin, even though we at
present continue unable to state definitely what it is that im-
mediately induces these cells to undertake excessive growth.
Granting this, that in tumours we have aberrant tissue growth
and that that tissue growth is due to certain cells assuming
excessive proliferative properties wholly outside the needs of
the economy, then such proliferating cells are to be considered
as being derived :

1. From embryonic " cell rests " which have for a shorter or
longer period remained latent in one or other tissue and then
have taken upon themselves a rapid proliferation leading to
tumour formation.

2. From the mother cells of a tissue which, remaining un-
differentiated, but capable of active proliferation throughout
life, now assume excessive proliferative powers, their daughter
cells retaining to a greater or less extent the characters and
the properties of the mother cells.

3. From differentiated cells which reverting to a simpler,
more embryonic type, with this reversion gain the capacity for
active and excessive proliferation.

Possibly I may add that the tendency to the development
of glandular cancer in later life bears some relationship to the
reversion and degeneration of gland cells at this period. As the

PROLIFEROUS CELLS AND NEOPLASIA 275

tissues become exhausted, the more highly differentiated cells
tend to become structurally simpler, revert, that is to say, to a
simpler type, and, with this simplification of structure accom-
panying atrophy, there may be, I would suggest, along the lines
already laid down, a greater liability for those cells to assume
proliferative powers.

CHAPTER II

ON THE CAUSATION OP CANCEROUS AND OTHER NEW GROWTHS *

(1901)

Just as to the bright spirits of the sixteenth century, in the
expansion and enthusiasm of the time, everything seemed
possible and men dreamt of the perfect commonwealth, so to us
in this age of medical renaissance the discoveries that have been
made in connexion with the causation and modes of prevention
of so many important diseases appear to point to the time when
the most dread scourges of humanity shall one after the other
be subjugated, and our race rise to a level of physical well-being
hitherto beyond the bounds of credence. So it is that, regarding
the most awe-inspiring and painful of these scourges, cancer and
malignant growths in general, while admitting our complete
impotence to deal with these conditions save when we have the
good fortune to recognize them at their very onset, we nevertheless,
to-day, are not wholly cast down ; rather are we sanguine. If we
have discovered the cause of so fatal and widespread a condition
as tuberculosis, sooner or later we hope to find a specific cause
for cancer. It is, we encourage ourselves, but a question of
study and of time. Analogy and the prevailing hopefulness
cause us to expect to find some microbic cause, and, being in
this frame of mind, we accept gladly each possible indication
pointing to the infectious nature of malignant growths. Already,
indeed, we seem to be on the very threshold of discoveries rich
in relief to suffering humanity. Discover the cause, say we,
and then inevitably we must bring the disease under the yoke.
Are we justified in taking this position ? Are our studies

1 Delivered before the Yale University Medical Alumni Association, New
Haven, Conn. Reprinted from the British Medical Journal, 1901.

276

THE PARASITIC THEORY OF NEW GROWTHS 277

being directed aright ? Are the arguments employed in support
of this theory of the infectious origin of malignant neoplasms
thoroughly sound ? Do the almost innumerable observations
which have been made during the last few years bear out the
assumption, or have we been neglecting other important factors
in the etiology of new growths in the keenness of our desire to
determine some specific causative agent ?

These are the matters I wish to discuss. The time, indeed,
is opportune for such discussion when of late, upon this side of
the ocean, two well-known surgeons like Roswell Park of Buffalo *
and Collins Warren of Boston 2 have placed themselves in evidence
as favouring the theory of infectious origin, and when, on the
other side, a Cancer Number of that well-known journal, the
Practitioner? with its series of articles by responsible writers deal-
ing with different phases of the subject, most assuredly leads up
to the conclusion that cancer is an infectious disease.

The Parasitic Theory

It will be well, in the first place, to adduce the main argu-
ments in favour of the parasitic origin and infectious nature of
malignant growths. These, stated briefly and impartially, are
the following :

1. That the increase in frequency in malignant tumours in
civilized communities during the last four decades is wholly out
of proportion to the possible action of any factor, save the
gradual spread of some infective agent. It is in excess of any
lessening of the death-rate, and not to be accounted for merely
by the increased longevity of civilized races — is not, that is, to
be explained by the fact that a larger number of individuals
now reach the cancer age. The increase is to be seen in country
as well as in urban districts, in European as well as in American
communities, and cannot therefore be attributed to any one
alteration in the mode of fife, to altered habits, altered environ-
ment. There is no other common factor capable of explaining the
increase, hence, per exclusionem, we must fall back upon infection.

2. That the incidence of the disease is frequently found to
be peculiarly localized — certain districts, mainly low-lying, are

1 Roswell Park, Trans. Amer. Surg. Assoc, xvi., 1808, p. 182.

8 Collins Warren, Boston Med. and Surg. Journ. vol. cxliii., 1900, p. 25.

* Practitioner, Cancer Number, April 1899.

278 ON GROWTH AND OVERGROWTH

seen to be especially affected, certain villages, and even certain
houses in those villages — and this apart from blood relationship
between the affected individuals. The incidence is, if I may
so express it, of a miasmatic type, resembling, for example,
that of malaria.

3. That the lesions produced by the growth of malignant
tumours within the organism are comparable with those induced
by certain known infective agencies, and that a close analogy
can be drawn between the tubercle and the dissemination of
tuberculosis, and the primary cancer nodule and the metastatic
cancerous or sarcomatous growths. Just as in chronic affections
there are, as a rule, single primary foci of origin, so with the
malignant neoplasms ; and just as there is extension of the
tuberculous process by continuity (with the destruction of the
surrounding tissues and their replacement by new tissue), by
the lymphatics, or by the blood stream, so it is in connexion
with neoplasms and their progressive development in the
organism.

4. That the more carefully material from malignant tumours
is examined, the larger is the proportion of cases in which certain
intracellular and extracellular bodies are to be recognized. The
series of forms seen, while large, is, according to each individual
observer, remarkably constant, and not to be regarded as in-
dicative of cell-degeneration, though observers differ between
themselves as to the specific series of forms. It is generally
agreed that these bodies are most numerous in the young growing
edge of the tumours rather than in the older central cells in
which degeneration might be expected to show itself. While
but a few years ago the observations of Sjobring, Ruffer, Metch-
nikoff, and others led to these bodies being regarded as parasitic
sporozoa, and so as being of animal nature, among upholders
of the parasitic theory the trend of opinion at the present time,
following Russell, Roncali, and Sanfelice, is to regard them as
of vegetable origin, as blastomycetes or yeast-like growths.
But within the last few weeks there have been indications that
the sporozoon theory is of late being actively supported.1

1 [In the years that have elapsed since this address was delivered, as one
after another other species of pathogenic microbes have been recognized, with
a striking inevitability, one after another these have been announced as the
causative agents of cancer-spirochaetes, amoebae, gregarines, chlamydozoa
and filterable or ultra-microscopic viruses.]

INFECTION AND MALIGNANCY 2ri

5. While the cases on record of apparent direct infect 3e
from one individual to another are so few as to be explai rm
by the law of chance, and as being but the coincidental oc- 0f
rence of the same disease in two associated individuals, d\v .
while the experimental inoculation of fresh cancer mat are
from one individual to the other (whether these be of the £
or of different species) has almost uniformly failed, neverthehur
in rare cases, this would have appeared to have succeeded ;
and, according to Sanfelice, the inoculation of a pure culture of
a yeast isolated by him has, in a small proportion of cases, led
to the production of definite neoplasms. Two factors are neces-
sary for successful experimental inoculation — the virulence of
the parasite and the susceptibility of the tissues ; and it may
well be that if malignant growths are brought about by parasites
of a very low state of virulence, a very special susceptibility
or special condition of the tissues is necessary for the inoculation
to be successful. To this lack of special susceptibility of the
tissues is to be attributed the frequent failure of experimental
inoculation, either with cancerous material direct, or with pure
cultures grown outside the body.

These, I take it, are the main arguments in favour of the
parasitic origin and infectious nature of malignant growths.
Stated thus, they present a very strong prima facie case in favour
of the parasitic theory. There is, however, I need scarcely
say, another side to the question, and each of the above arguments
in turn may be, and has been, strongly assailed.

(1) Some capable observers still hold that increased longevity
is capable of explaining the increase. (2) A fuller study of low-
lying and estuarial regions shows that while cancer is very common
in some, it is equally uncommon in others. (3) The analogy
between the mode of dissemination of tuberculosis, for example,
and the development of metastatic cancerous nodules is recognized
by all histologists as being thoroughly unsound. (4) Further,
it is difficult to refute the careful observations of Fabre-Domergue
and of Pianese,1 that the so-called cancer bodies of varying types
described bv different observers are so manv different forms
of degeneration products within the cancer cells. When, by
employing special staining methods which afford a differential
stain for mucoid, pseudo-mucoid, hyaline, and amyloid material,

1 G. Pianese, Zeiglefs Beitrdge, xx., 1896.

230 ON GROWTH AND OVERGROWTH

se£an be shown that a succession of forms can be made out,
hosing imperceptibly from conditions within the cells which
bet'iot possibly represent the stages in the life-history of some
s0 ^cellular parasite to others, which the parasitologists regard
tha^dicating stages in the life-history of one or other microbic

;- — the histological evidence that cancer is due to parasites
turjmes, to say the least, singularly frail.

I shall not, however, now take up these various arguments
and endeavour to weigh their value ; rather I would attempt
to approach the matter from a different point of view, would
seek to determine the relationship of malignant to other forms
of growth in the organism, and from a study of this relationship
would endeavour to determine whether we are justified in holding
this advanced view in regard to the parasitic origin of malignant
growths.

Tumoues Benign and Malignant

I would in the first place ask : Can we draw a fine between
tumours which are essentially benign and tumours essentially
malignant ? The answer to this question is an unhesitating
" No." If we understand by malignancy the property of local
invasion of surrounding tissues, together with that of the forma-
tion of new growths of like nature in distant organs — the other
associated properties being subsidiary to these two — then all
growths which we classify as true tumours may take on malignant
properties. Even so highly differentiated and (ordinarily)
benign a neoplasm as the chondroma 1 may exhibit metastases.
It is true that very often the assumption of malignant properties
indicates and follows a change in the morphological characters
of the tumour or parts of the same. Thus the lipoma,2 as such,
is not malignant, but portions of a lipoma may take on a more
embryonic character, and the rapidly growing simpler cell forms
may infiltrate surrounding tissues and may lead to metastases.
Nevertheless, we have no doubt that this sarcomatous tissue is
the direct outcome of the cells forming the primary lipoma.
There is no question that here the lipoma has taken on malignant
characters. It can scarcely be questioned, also, that malignancy
is not a primary but a superadded property of the tumour, that

1 [A tumour composed of cartilaginous tissue.]
8 [A tumour composed of fat-cells.

BENIGN V. MALIGNANT TUMOURS 281

growth up to a certain point is of the benign type. Hence,
if any theory of the parasitic causation of tumours is to conform
to the facts already acquired by us as to the life-history of
tumours, that theory must assume one of three forms, namely :

1. That all true tumours, benign as well as malignant, are
of parasitic causation.

2. That infection is but one of a series of causes of tumour
growth, both benign and malignant.

3. That while tumours, as such, need not primarily be of
micro-parasitic origin, the assumption of malignant properties
is due to infection by parasites, these leading to that active
and excessive purposeless cell growth which is the basis of
malignancy.

Let us consider now these possibilities. Are all tumours,
benign as well as malignant, of parasitic origin ?

Definition of the Term " Tumour "

Before seeking to answer this question, it is necessary to
lay down a definition of this term " true tumour." This in
itself is no easy matter, for into that definition for our present
purposes — and, I may add, equally for general purposes — no
definite statement as to causation should be admitted. So
long as we are uncertain as to the causation of these growths,
our definition must be strictly confined to facts. Ziegler's
statement that " a tumour is a new formation of tissue possess-
ing atypical structure, not exercising any function of service
to the body, and presenting no typical limit of growth," is, on
the whole, adequate, though in the existence of such forms as
certain adenomata, osteomata, and chondromata the use of the
term " atypical structure " requires a little explanation.

I prefer the definition of C. P. White : " A tumour proper
is a mass of cells, tissues or organs resembling those normally
present, but arranged atypically. It grows at the expense of
the organism, without at the same time subserving any useful
function." 1

1 C. P. White, Journ. of Path, and Bacter. vi., 1899.

282 ON GROWTH AND OVERGROWTH

Limitations of the Parasitic Theory

Accepting this as our definition — and I may say that other
attempted definitions agree in including or tending to include
the same provisions — then immediately we are compelled to
recognize that all true tumours are not of parasitic origin. There
is the group of teratomata, due to the inclusion within the
tissues of one individual of tissues derived from another. These
tissues become organically connected with those of the host,
and grow at the expense of the host, throughout subserving no
function of use to that host. They form a well-recognized group,
which I need not here describe more fully.

Thus, then, on the one hand there are tumours which assuredly
are not of parasitic origin, but, on the other hand, we cannot go
to the opposite extreme and say that no tumours are due to the
action of parasites in the tissues. It may be laid down that
physical and toxic agencies, when acting below the point neces-
sary to induce cell exhaustion and cell destruction, may act as
stimuli rather than as irritants, and doing this may bring about
increased cell proliferation. It may be — it has been — argued
that these bacteria and their products cannot be the direct
cause of cell proliferation. Weigert and his pupils, for example,
strenuously deny such direct action, but the fact remains that
the bacterial toxins can be the initiating, indeed in certain cases
an essential, cause ; and the recent studies by Mallory,1 more
especially upon the changes occurring in the endothelium of
vessels and lymphatics during typhoid and other infectious
disease, seem to prove this with absolute clearness.

Bacterial toxins, which when concentrated lead to excessive
cell necrosis, in a less concentrated condition may induce active
cell growth. In the central portion of the abscess, where there
are abundant bacteria and their products, there is abundant
evidence of cell necrosis. At the periphery, where by sundry
mechanisms the bacteria are hindered from existing, and where
the toxins diffusing out are less concentrated, there is evidence
of active mitosis and cell growth.

All the tissue components, it is true, are not alike in their
reaction. The simplest forms of cells may be stimulated to

1 Mallory, Journ. of Ezper. Med. iii., 1898, p. 611, and Trans. Assoc.
Amer. Phys. xv., 1900, p. 224.

PARASITISM AND TISSUE OVERGROWTH 283

proliferate by. substances which destroy or arrest the activity
of the higher cells in the same neighbourhood. Thus it is that
we can trace various stages of bacterial action from those in which
destruction predominates to those in which, as in the infective
granulomata, the higher tissue elements undergo destruction, but
endothelial and connective tissue elements show, it may be in
the very earliest stages of the process, well-marked proliferation.

Further than this it would seem that we cannot advance
with the bacteria. In other words, so far as our present know-
ledge permits us to conclude, bacteria — schizomycetes — below
a certain degree of toxicity of their products cannot maintain
an existence in the tissues of animals. Thev succumb to the
antibacterial mechanisms of the organism, and, succumbing,
any proliferative activity they may have induced comes to an
end. In plants, it is true, we meet with definite bacteria living
a symbiotic existence in the rootlets, and causing well-marked
overgrowths.1 It may eventually be found that bacteria are
capable of initiating progressive cell growth in the higher animals,
but of this power proof is so far wanting.

There are, however, other forms of life which, while of pecul-
iarly low toxicity, are capable of existing and actively proliferat-
ing within the tissues of animals. Coccidia, for example, are
peculiarly common in the rabbit, but the irritation they set up
in the tissues is of so mild a type that in many regions the majority
of rabbits, young and old, show the results of their presence in
the tissues, and that without any sign of disturbance of general
health. I need scarcely say that the evidence of their action
is especially found in the liver, where, growing within the cells
lining the bile ducts, they lead to a proliferation of the same
with surrounding overgrowth of the connective tissue, and so
give rise to what are truly " cystadenomata." 2 Similar growths
of like causation have in rare instances been found in the human
liver. It is to be noted that the growths so produced are benign
and not malignant. For some reason — what reason remains
confessedly problematical — the irritation and the reaction to

1 [Since this paper was written Dr. Erwin F. Smith of the Bureau of Agricul-
ture at Washington has demonstrated conclusively (1907-1916) that a specific
bacterial form, the Bact. tumefaciens, is the causative agent of large localized
overgrowths in plants, crown galls, of beets, Paris daisy, roses, etc.]

2 [An adenoma is a tumour reproducing the structure and cell arrangement
of one or other type of glandular organ.]

284 ON GROWTH AND OVERGROWTH

the growth of these microbes is not sufficient to bring about
such an active proliferation and continued vitality on the part
of the cells that, carried to other regions, these cells reproduce
the primary growth.

The nematode worm, Bilharzia, an organism vastly higher
in the scale of animal life, and one infesting a very large section
of the human race, may exist within the organism for long
years without setting up any very serious disturbance ; but in
those parts of the body which by preference it infests it is clearly
capable of producing cell proliferation of an adenomatous,
papillomatous, or even of a cancerous type. Living in the radicles
of the portal and pelvic vessels, its abundant ova make their
way mechanically (on account of their shape) through the vessel
walls into the submucosa and mucosa of the lower bowel and
the bladder, and so into the lumen of the intestines and the
bladder respectively. Numerous cases are on record of extensive
neoplasms of the rectum and bladder which are clearly secondary
to the continued cell irritation induced by these ova. All, I
think, are nowadays prepared, from such cases as these, to
agree that there are parasitic agents capable of inducing tumour
growth of a benign type ; and, what is more, that this growth
does not merely occur in cells which are congenitally displaced,
but in those which, prior to irritation, have been portions of
strictly normal tissues.

It would appear, therefore, that the evidence at present in
our possession confirms the second possible theory that parasites
are but one of the series of causes of tumour growth.

Now, between these two extremes — forms assuredly due to
independent cell growth without microbic intervention, and
forms assuredly initiated by microbic intervention — we have a
very broad debatable territory, wherein we encounter the main
mass of neoplasms. How are we to regard these ? Do they
belong to the one or to the other class ?

Classification of Tumours

It will be well in the first place to attempt a rapid classi-
fication of the different forms, according to the tissues implicated ;
according, that is, to the tissues and the cells which give origin
to the different forms of tumours. Such a classification is

THE OKDERS OF TRUE TUMOURS 285

possible. To attempt to indicate the class to which each in-
dividual tumour belongs would immediately involve us in a
long and, for present purposes, absolutely fruitless discussion.
And this, it would seem, very largely because a tumour of one
histological type may be produced under different conditions.
Carcinoma, for example, may develop in connexion with the
f unctionless or relatively f unctionless cells of the ovarian dermoid ;
or, again, may originate from the active and functional cells of,
say, the gall-bladder or the splenic flexure of the colon.

The following classification appears to me to include every
form even if this be the case, and even if at the present time
we may differ as to the class in which certain varieties of tumours
should be placed.

A. Teratomata

Tumours composed of the products of growth of one individual
within the tissues of another individual of the same species.

I. Teratomata of the first order — Twin Teratomata. Tumours
due to the continued growth, within the tissues of one individual,
of the tissues and organs of another individual which has become
included in the former during foetal life.

II. Teratomata of the second order — Filial Teratomata.
Tumours due to the continued growth, within the tissues of one
individual, of tissues and organs which are the product of growth
of one of the germ cells of that individual.

(i.) Parthenogenetic. — Tumours formed of the products of
the unfertilized germ cell. The more recent studies
of Repin x and Arnsperger 2 most strongly support the
view that ovarian and testicular dermoids can only
be regarded as examples of abortive parthenogenesis.

(ii.) Gamogenetic. — Tumours formed of the products of the
fertilized germ cell. The recent work of H. Peters 3
has settled the angry discussion as to the nature of

1 Repin, " Origine parthenogenetique des kystes dermoldes de l'ovaire,"
These de Paris, 1892.

2 Arnsperger, Virch. Arch. Bd. clvi., 1899, S. 1 ; see also Wilms, Dent.
Arch. f. Bin. Med. lv., 1895, S. 219 ; Zeigler's Beitr. xix., 1896, S. 233 and
367.

3 H. Peters, Die Einbettung des menschlichen Eies, Leipzig und Vienna
(Deuticke), 1899.

286 ON GROWTH AND OVERGROWTH

the syncytial cells of the placenta ; they are, as many
have long held, of foetal origin, derived from the
outer layer of the foetal epiblast. They have, physio-
logically, well-marked powers of eroding or breaking
down the uterine tissue, and through their agency it
is that the villi penetrate into the maternal blood
sinuses. Physiologically, that is, they possess what
we regard as malignant properties. The highly
malignant tumour, formed as a result of their over-
growth, the so-called deciduoma or syncytioma
malignum, is thus clearly an example of cells which
are the product of one individual invading the tissues
of another individual. Placental moles belong also

to this class.

B. Blastomata

Tumours composed of the products of aberrant growth of cells
and tissues of the individual in whom they develop.

I. In which abnormal cell relationship precedes the neo-
plasia by a definite interval, and apparently predisposes to the
aberrant purposeless growth. These abnormal relationships may
be (a) inherited, (b) acquired, and then either of ante-natal or
post-natal acquirement.

(i.) Heterochronic. — Tumours in which the primary abnormal
cell relationship is due to the persistence of tissues
from a previous stage of development, which tissues
normally undergo atrophy and disappear. Examples :
Tumours developing from the thyroglossal duct,
Gartner's duct, branchial clefts, etc.
(ii.) Heterotopic. — In which the abnormal cell relationship
is due to the displacement of cells and tissues during
the process of growth, so that the tumours develop
from " cell rests." Examples : Cancer of accessory
thyroids and other tumours developing in accessory
glands, rhabdomyomata,1 and suprarenal tumours of
the kidney, etc.
(iii.) Heterotrophic. — Tumours in which the abnormal cell
relationship is due to unequal growth of different

1 [Tumours composed of striated muscle elements.]

THE BLASTOMAS 287

elements composing a tissue. Examples : Congenital
angiomata * and lymphangiomata, in which from
obstruction or other cause the vascular elements of
the tissue become unduly prominent, while the normal
parenchyma of the part is imperfectly developed.
II. Tumours in ivliich one and the same agency leads to abnormal
cell relationship and to tumour growth.

(i.) Tumours in which no primary disturbance of cell re-
lationship is to be recognized, the growth originating
apparently from functional cells and tissues, and in
which so far the causative agent has not been deter-
mined. This group would seem to include a very
large number of malignant growths — epitheliomata
of the skin, tongue, etc., carcinomata 2 of the stomach,
colon, and so on. The more we study the early
stages of cancer the more it is impressed upon us that
these would seem frequently (though by no means
necessarily always) to develop from the cells of tissues
that are or that have been normal and functional,
(ii.) Tumours originating from normal and functional tissues
through known microbic agency, for example, cocci-
diosis of the liver of the rabbit, bilharzia tumours of
the bladder and rectum.
Eeviewing the various classes of tumours here rapidly set
in order, we recognize a series with at the one extreme tumours
originating in misplaced tissue, and growing and continuing to
grow without microbic irritation, and at the other extreme
tumours originating in normal tissues, and growing as a conse-
quence of the low form of irritation induced by parasites.
Between these two extremes some approximate and apparently
belong to the former group, others show evidences of relationship
to the latter.

Accepting this as the case, it must next be asked, Can we
determine anything in common — a common denominator, as it
were — associating these various groups ? Structurally, tumours
belonging to the two extremes may be of like nature and histo-
logical character, and common denominator there must be, and

1 [An angioma is a tumour composed of vessels and vascular elements.]
a [A carcinoma is a malignant tumour containing the elements of a glandular
epithelium.]

288 ON GROWTH AND OVERGROWTH

if we can only determine it, then we can proceed to establish
the true theory of tumour formation.

I need not say that there have been many attempts to form
such an adequate theory, from that of Cohnheim, who, ignorant
of the existence of the one group, sought to ascribe everything
to " cell rests," to those of present-day writers, who, heedless
of the existence of the other group, seek to ascribe everything
to microbes. Neither extreme, as I have shown, is possible.
I trust that I shall be forgiven if neglecting, on account of the
demands of time, to put before you what I may term intermediate
theories in full detail (of which those most discussed pro and
con at the present time are the theories of Ribbert 1 and Hanse-
mann.) 2 I here give rein to my own line of thought.

Let me, in the first place, put before you in concise language
the problem that has to be solved : (1) Certain tumours arise
from misplaced cells ; (2) certain tumours arise from cells origin-
ally in normal position. What is the cause of either category
of cells taking on excessive growth independent of the needs of
the organism ?

The Relation of Growth to Functional Activity op Cells

To solve this problem it is clear that in the first place we
must have right ideas as to the nature of cell growth and pro-
liferation in ordinary. Now, as I pointed out a few months ago
(in the Jacobi Festschrift 3), we are apt to have a vague idea, to
say the least, of the nature of the normal growth of tissues.
The more one studies what occurs in the various tissues, the
more obvious it is that multiplication and the active performance
of other function by the cell are incompatible, or otherwise, that
the actively functioning and fully developed cell, as such, does
not undergo mitosis and show evidences of multiplication.

In not a few tissues (the epidermis, periosteum, etc.) there
are present what may be termed proliferous or " mother cells " —
cells which themselves throughout life do not attain full differ-
entiation, but which give off daughter cells, and the daughter
cells it is which develop into the fully differentiated functional

1 Ribbert, Das patJiologische Wachstum der Oewebe, etc., Bonn (Cohen), 1896.
1 Hansemann, Die mikroscop. Diagnose der bosartigen Oeschwulste, Berlin
(Hirschwald), 1897. 3 See p. 263.

GROWTH V. FUNCTIONAL ACTIVITY 289

cells. In other tissues the fully differentiated cells are capable
of division and proliferation, but this only after reversion to a
simpler type, or, as we term it, a more embryonic state, in which
evidently their functional secretory activity is reduced to a
minimum. Thus, after the injection of indigo-carmine, Mar-
tinotti x has pointed out that in a kidney, portions of which have
been removed in order to study the compensatory overgrowth
of the remaining part, all the cells of the convoluted tubules in
the remaining part take up the pigment, save and except those
in a condition of active mitosis : these remain perfectly colourless.
Even in so simple a tissue as connective tissue, the cells exhibit
this return to a more " embryonic " state prior to multiplication.2

Professor Carlier of Birmingham has, I may add, noted cases
apparently contrary to what is here stated ; cells of the gastric
glands may show mitosis and the appearance of secretory granules
in their protoplasm at one and the same time.3 But these may
well be stored-up granules. In a letter to me — which in conse-
quence of the long time it would take to gain his necessary
permission, I trust I may be permitted to quote — he states that
in his opinion a dividing cell may be functionally active. He
adds, however, that when this is the case, he is inclined to believe
that the nuclear changes characteristic of mitosis become sus-
pended for the time being.

Thanks to the researches of Hodge,4 Mann,5 and other neuro-
logists, of Macallum of Toronto,6 and of his pupils, Scott and
Bensley, and of Carlier, and of other physiologists and cytologists,
we are recognizing more and more that the nucleus plays a con-

1 Martinotti, Centralbl. f. allg. Path, i., 1890.

2 Here let me urge that the use of this term " embryonic " as applied to
cells in an adult organism is of questionable value, and indeed is responsible
for not a little of the want of comprehension of the phenomena of normal and
abnormal cell growth. I would suggest in preference the employment of the
term " vegetative " or " proliferous " to include the whole series of these cells.
For, strictly speaking, in these cases of reversion to a simpler state, it is a mis-
nomer to speak of adult cells becoming embryonic, and a continuance of or
reversion to this simpler histological condition is not an indication of an em-
bryonic condition, but is essentially correlated to the mode of activity of the
cell ; or, otherwise, the so-called embryonic cell is a cell which specially is in a
position to proliferate, and is not in a position to perform specialised function.

3 Carlier, British Medical Journal, September 15, 1900, p. 740.

4 Hodge, Amer. Journ. of Psychol, iii., 1890, p. 530; Journ. of MorphoL vii..
1892, pp. 99 and 449 ; Journ. of Physiol, xvii. p. 129, 1894.

6 Mann, Journ. Anat. and Physiol, xxix. p. 100, 1895.

6 A. B. Macallum, British Medical Journal, 1898, ii. p. 778.

U

290 ON GROWTH AND OVERGROWTH

trolling part, not merely in cell division, but also in the function
of the cell. With activity the nuclear chromatin becomes used
up and discharged into the body of the cell, there to combine
with other substances to build up prezymogens and other bodies,
which are eventually discharged as the specific secretion of the
cell. This being the case, we can readily understand that the
higher specific functions of the cell cannot be carried on by a
nucleus whose nuclear material is being utilized to its fullest in
mitosis. The process of cell division and the performance of
the higher functions of the cell are incompatible, and the cell
engaged in the active performance of its special functions cannot
undergo division.

Yet, paradoxically, under normal conditions after birth, it
is the actively functioning tissue that undergoes hyperplasia
and takes on increased growth. This is to be explained in most
cases by the fact that the new units of that tissue are derived,
not from the actively functioning cells, but from mother cells
present therein ; in other cases, probably by the fact that such
hyperplasia occurs where the activity is not continuous, but
interrupted, growth occurring in the intervening periods of rest.
One may quote, for example, the late Sir James Paget's well-
known illustration of the growth of a corn on the foot being due
to an interrupted irritation of the skin by the pressure of a tight
boot, that irritation which otherwise would lead, if continued, to
atrophy, being removed at night, leads to hyperplasia of the
epidermis.

For a cell to divide actively there must be at least temporary
arrest of the specific, as distinct from the vegetative functions.
Or, conversely, arrest or disturbance of the specific functions
of the cell, if of such a nature as not to arrest vegetative activity,
favours cell multiplication. The nuclear activities, unemployed
in one direction, become diverted into another ; the nuclear
material, not being discharged and converted into the specific
secretions of the cell, tends to become heaped up, and, accumu-
lating up to a certain point, is then used in mitotic processes.

Influence of Environment on Cell Activity

While heredity surely plays an important part in determining
the structure of the cell, we are forced to see that, underlying

THE INFLUENCE OF ENVIRONMENT 291

and determining heredity, the eventual structure, as again the
specific functions of a cell, are determined by the sum of the
forces acting upon that cell ; structure and function are adapta-
tions to environment. It is not only the nutritive material
absorbed by the cell that determines the characters thereof,
but its position relative to other cells in the economy. From
the first the cell continues to proliferate until the sum of the
above-mentioned forces brings it sooner or later to a condition
of equilibrium, until the amount of nutrition absorbed (and the
resultant energy) is fully utilized in maintaining the cell in the
status quo. And when this point is reached, the products of
nuclear and cell metabolism, as they are formed, become used
up in the performance of function. There is no sufficient surplus
or redundancy of material to permit the cells to undergo simul-
taneously those material transformations which result in cell
division. Disturb the environment of the cell, remove certain
of the forces acting upon it so that the cell energy previously
devoted to counteracting the effects of those forces may now be
diverted into and employed in other directions, and then un-
employed energy may be utilized for growth, or perhaps, more
exactly, the material heaped up by constant assimilation, by
its very presence in the nucleus, stimulates this to undergo
mitotic changes. As a matter of fact, numerous examples can
be adduced of overgrowth of tissues due to removal of forces
normally acting upon these tissues, overgrowth which continues
until such time as the sum of the forces acting upon the newly
formed cells becomes equal to the utilization of the sum-total
of the cell energies ; then continued growth comes to an end.

This, on the one hand, namely, that removal of forces acting
upon the cell from without leads to its passing from the functional
to the proliferative state ; or, if it be in the proliferative state,
leads to its continuing in the same ; on the other, increased
absorption and utilization of nutritive material affords the
opportunity for increased exhibition of energy, increased
katabolism.

Inertia of Cells

Now increased absorption and assimilation is in general a
response to previous or concurrent increased katabolism and
increased activity. So long as this increased assimilation

292 ON GROWTH AND OVERGROWTH

corresponds to and makes up for the loss of substance brought
about by this increased activity, for so long is there no tendency
towards cell multiplication. It, however, may happen that
after stimulation the amount of absorption and assimilation
may be in excess of the needs of the cell. We have, in short,
to recognize the existence of the principle of inertia.

Inertia, physically speaking, is one of the properties of
matter in general. Now, just as according to physical termino-
logy there are two kinds of inertia — inertia of matter at rest
(or inertia of mass) and the inertia of matter in motion (or
momentum) — so in relation to the protoplasm of the cell we notice
the same kinds of inertia in evidence. As Dr. Harris points out,1
protoplasm which is at rest cannot be instantly caused to change
that state to respond to stimuli. We have the familiar latent
period, and numerous examples can be given of that functional
inertia which corresponds with the inertia of movement (momen-
tum), " the inertia that makes the wheel rotate long after you
have ceased to spin it." The isolated heart of the frog will
continue to beat long after its removal from the body. Hairs
continue to grow after the death of the body ; indeed, all cases
of organs and tissues continuing to perform functions after their
blood supply has been cut off are cases of functional inertia.

And coming now to the immediate problem before us, it may
be said that active katabolism, with the active performance of
function, leads to corresponding increased anabolism, and that
if by nervous or other mechanism the performance of f miction
be arrested, the absorption of nutrition and assimilative processes
may still continue for a time. If there be continued access of
nutriment, this, according to the above principle, is liable to
continue over and above the needs of the cell. In other words,
there is a liability for reserve material to be heaped up in the
nucleus and body of the cell. It is, I would suggest, under these
conditions that proliferation may replace functional activity.
If the cell be not called upon to perform its normal functions, be
not called upon to work after previous stimulation has led to
increased absorption and anabolism, that cell is in a favourable
condition to divert its energies from function to proliferation.

In this connexion I would point out that we are forced to
recognize that the stimulus for the performance of special function

1 D. F. Harris, British Medical Journal, September 15, 1900, p. 741.

FUNCTIONAL V. VEGETATIVE ACTIVITIES 293

is not intrinsic : it does not arise within the cell but comes from
without. Vegetative functions, on the other hand, including
cell division, would seem to be of intrinsic origin and automatic,
determined only indirectly by external conditions, directly by
conditions within the cell. The normal cell, in short, stands in
very much the same relationship to the organism as does the
normal citizen to the body politic : the work that individual
performs is determined by his position in the body politic, and
is for the benefit of society at large, but at the same time the
individual citizen works for himself and for the continuance of
his species. The one, it is true, depends upon and is intimately
connected at every point with the other function of the individual,
but these two functions of citizenship and self-preservation are
clearly distinct, the one determined by external conditions, the
other being individual and inherent.

So long as the cell is not called upon to work, and for so long
as the surrounding conditions are unaltered, we have the inertia
of rest, and for so long there is no call either for increased
activity and breaking down of its tissue, or for subsequent build-
ing up in excess of vegetative needs. A cell lying functionless
and latent does not therefore show any tendency to proliferate.
Alter these surrounding conditions and a condition of increased
metabolism may be induced.

Position of the " Cell Rest " Theory

Thus a " cell rest " 1 may remain latent for years or for life,
•showing little or no signs of growth. The very fact that it is
not in a position to perform adequate function is against growth,
so long as the conditions are unchanged. Change the conditions,
and this same fact that it is not in a position to perform adequate
function is evidently the reason why the cells forming that
"rest" show a peculiar tendency towards proliferation. They
are stimulated from without by physical and other means ;
their relationships to other cells, whether merely of tension or of
tension plus position, become altered, and the reactive increased
anabolism and storing-up of cell material which cannot be used
up in active work to the same extent as occurs in normally
situated cells, is the essential cause why now the stored-up

1 [The persistent remains of a foetal organ, or group of rolls of an oi
displaced during the course of development.]

294 ON GROWTH AND OVERGROWTH

energy becomes utilized in another direction — namely, that of
multiplication.1

Here we arrive at the strongest objection to the " cell rest "
theory pure and simple of tumour growth. It is one thing to
have " cell rests ' and foetal remains present in the various
parts of the body, quite another thing to have tumour growth
originating in these. It is probable that the vast majority of

cell rests ' never take on an excessive purposeless growth.

Cell rests ' may show signs of active proliferation from the
first ; they may, on the other hand, lie latent for long years,
only then showing tendencies to excessive overgrowth ; lastly,
they may never be found in other than what I may term " a
resting stage." So that, granting the existence of " cell rests,"
we cannot grant that these misplaced cells are the primary cause
of tumour formation — at most they are a predisposing cause ;
some other cause — a something acting upon the " rests " — is
necessary to explain their aberrant and excessive growth. We
have to demand some change in the surrounding conditions.

Conditions necessary for Cellular Hyperplasia

Keeping in mind these two conceptions, namely, that the
stimulus to increased functional activity comes from without,
and that the cell in its actions exhibits marked inertia, we can
thus realize the conditions under which increased cell proliferation
may be excited in a tissue, whether normal or containing mis-
placed elements. These conditions are either : (1) that the cells
be subjected to periodic irritation leading to increased functional
activity, increased secretion, etc., followed by periods of rest in
which there is storage of cell material to such an extent that the

1 Along these same lines we best explain the proliferation of mother cells
in functioning tissues ; though these give rise to fully active cells, their rela-
tionships lead them to live a vegetative existence. Alterations in their sur-
roundings brought about by increased activity of the surrounding daughter
cells must tell upon their relationships, and altered tension and environment
must here again lead to the same reactive increased anabolism, and this must
initiate cell division. Where again we have cells once functionally active
becoming through senile changes latent and somewhat atrophied, here also any
stimulus, whether mechanical, chemical, or physical, acting upon those cells
and calling for increased activity, may in like manner render those cells pecul-
iarly liable to proliferate. The change of relationships in the cells brought
about by their atrophy and the somewhat altered relationships of the surround-
ing parts may render them incapable of performing their normal functions, and
increased anabolism may be followed instead by multiplication.

CONDITIONS INDUCING OVERGROWTH 295

vegetative activities of the cell are called into play to utilize
that material in proliferation ; or (2) that the irritation be of
such an extent and continued for so long a period, that in con-
sequence of the increased functional activity of the cell, of the
increased secretion, and the increased blood and lymph brought
to the part, the relationships of the cell to those in its immediate
neighbourhood are altered to such an extent that, while there is
adequate or even increased assimilable material which it can
absorb, the very alteration of environment and the increased
tension to which it is subjected hinder the proper performance
of function, and the stored-up energy becomes diverted from
the performance of specific function to proliferative activity.

Now, considering the first condition, the very property of
inertia must be a preventive of sudden change from the functional
to the proliferative stage of the cell, while again, the environment
of the cells continuing relatively unaltered, they must easily
respond to the recurrent stimulation to perform their normal
functions, and responding, the tendency to multiply must be
liable to arrest. Thus I am inclined to consider that while this
condition may be possibly in action to induce that moderate
and orderly amount of overgrowth which follows upon increased
tissue activity, it is incapable of inducing or explaining neoplasia.

A Starting-point common to all Forms of Tumour

It is the second state, or condition, that affords the starting-
point for excessive local overgrowth in previously normal tissues.
And, accepting this view, we at last reach the position of obtain-
ing one common starting-point for all forms of tumour proper.
This starting-point is that condition in which the cells of the
part, whether functional or lying latent, upon being stimulated
and as a consequence undertaking active assimilation and
anabolism, cannot from their relations carry on their peculiar
functions to a corresponding extent, so that the accumulated
energy is instead utilized in mitotic changes and cell division,
rather than in the normal katabolism.

Herein, however, we do but explain the first stage in the
development of a tumour proper. How is it that once started
the cells continue to proliferate ? What are the conditions
under which we can recognize and explain the development of

296 ON GROWTH AND OVERGROWTH

the condition of uncontrolled proliferation ? It is clear that
at first at least that same stimulus which had led to increased
assimilation in the parent cells must continue in evidence, for,
if this were not so, then the new cells would either lead but a
latent existence, or the tissue in which they are present reverting
to a normal state, these cells would gain the same relationships
as were possessed previously by their parent and " sister " cells ;
they would develop into normal functional cells with specialized
structure. We are bound, therefore, to assume that this is not
the case — that the primary stimulus continues to act for a certain
period, and that as a result of this continued action there is
developed a colony of cells all tending actively to proliferate.
But now the greater the proliferation the more inevitable must
it be that a certain number of new cells assume relationships
more and more removed from those proper to the parent cells.
Some of these new cells, at least, must pass to a greater distance
from the blood capillaries and blood supply, from the lymph
channels and the terminal nerve supplying the part, and becoming
thus still further removed from the ordinary relationships of
the cell forming the particular tissue which has given them birth,
still less are they capable of performing normal functions ; still
more, granted adequate nutrition, are they liable to proliferate.
If becoming thus heterotopic their nourishment is cut off, they
of necessity die or at least their continued proliferation is arrested.
If also the verv abundance of the new cells formed leads to
marked increase of the pressures acting upon those cells, growth
is equally arrested.

According, therefore, to (a) the time of continuance of the
primary stimulus (under which term, be it remembered, for the
time being I here include everything capable of inducing modified
cell relationships to an extent sufficient to permit continued
anabolism with perverted katabolism) ; according also to (b)
the relationship assumed by these new cells one to the other ;
and (c) the tensions to which they are subjected, so do we
have or not have the conditions favourable for continued and
increasing growth and development of new cells. It is this
continuance of the primary stimulus coupled with continued pro-
liferation which gives us the second stage in the evolution of the
tumour. These newly proliferated cells may exhibit various
departures from the structure and appearance of the fully formed

CONDITIONS FAVOURING OVERGROWTH 297

functional cells, according to the extent in the alteration of their
relationships and the extent of the stimulus. Under one order
of conditions, inertia and heredity may lead the new cells to
assume specific characters very closely approximating those of
the parent cells, and the more the cells are of specific functional
type the slower will be the growth ; under another, the departure
from type may be so extensive that it is difficult to determine
from the individual cells the tissue in which they originated.

Theory of Continuous Tumour Growth

But even at this point we have not explained tumour forma-
tion. We have but reached the stage met with in infective
granulomata in which destroy or inhibit the toxic cause of the
cell overgrowth and that overgrowth ceases. We have still to
account for the continued automatic growth of tumours proper.
For such, it seems to me, there are two possible explanations.
Of these the first and most obvious is that the modified relation-
ship of the parts, or the stimulus which originated the growths
in the first place, continues persistently in action. In those
cases in which, apparently, modified relationships of the cells
to each other are sufficient to act as the primary cause of growth,
it is not, I think, possible to conceive the continuance of the
same initial disturbance throughout the life of the tumour ; the
very growth and massing together of the cells must remove the
old and introduce a new series of relationships ; we cannot, for
example, apply this idea of altered environment alone to explain
the production of metastatic growths. It is not merely the
outward relationships of the cells, but something in their con-
stitution that can alone explain this active proliferation of cells
in regions wholly away from their primary seat of growths, in
regions where they are exposed to totally new conditions.

It is, however, quite possible to conceive the continued
existence of microbic parasites within the new growth or within
the cells of that growth, the products of which by irritating the
cells and modifying their functions would continually stimulate
the cells to multiply along the fines already laid down. It is
possible to conceive these cells and parasites being conveyed to
distant parts by the blood or the lymph stream, and when the
cells come to rest the associated or symbiotic parasites still by

298 ON GROWTH AND OVERGROWTH

their products affecting the cells in their new relationship, and
thereby leading to continued growth. This, I take it, is the
usual conception of those who propound the parasitic theories of
cancer formation.

But herein comes the difficulty. Continuous purposeless
growth is characteristic of all forms of tumours proper. Never-
theless, as I have already stated, in a very large group of tumours
all the evidence we possess points surely to the fact that microbes
have not initiated the growth, and are not concerned in the
continuance of the growth.

It would be absurd to urge that microbic agencies play a
constant part in the production of the group of teratomatous
tumours. No one would suggest that such are the cause of
post-operative implantation cysts, or of simple congenital
dermoids, such as the little hairy dermoids upon the sclerotic,
or of the different forms of angioma. And we can recognize
no sharp dividing line, histologically, between the tumours just
mentioned and another group including the rest of benign
and malignant neoplasms. Histologically all are of the same
type.

Further, we can recognize no sharp dividing line between
the benign and the malignant forms of such tumours ; there is
no one stage in which it is possible to say, thus far one set of
conditions has been at work leading to purposeless cell growth,
beyond this microbic irritation enters in. In both benign and
malignant tumours the growth is purposeless, and the only feature
separating the two forms is the rate of proliferation and stage
in which that occurs. And certainlv we encounter everv transi-
tion between tumours of frankly benign and those of frankly
malignant type.

The " Habit op Geowth " in Cells

Here let me be clearly understood. I am not arguing that
parasites, intracellular or extracellular, may not be regarded
as originating a certain number of tumours. I am pointing out
that certain features are common to all tumours ; that microbic
parasites and the irritation they may induce are not common to
all tumours, and that this last factor is therefore inadequate to
explain those characteristics which are common to tumours in

THE HABIT OF GROWTH 299

general. Thus it is that I am led to favour a theory which I
believe is applicable to all tumours, whatever their origin.
Briefly, this theory is based on the fact that cells and their
descendants which for long periods have been subjected to certain
influences, whereby their properties and structures have become
modified, eventually retain those properties after the influences
referred to have ceased to act upon them. I base it again upon
that principle of inertia already indicated — a principle which
some years ago (1896) in this connexion I referred to as the
" habit of growth." 1

Here let me cite certain examples illustrating the existence
of this principle in connexion with tumours :

It is a familiar fact that a columnar-celled epithelium, sub-
jected to altered conditions of certain orders, is liable to become
converted into a squamous, many-layered epithelium. Such
transformation is often noted in the uterus that has been prolapsed
and everted ; it has been noted in the gall-bladder after chronic
inflammation, and again in the larynx. What is more — and
this is a point to which I would especially call your attention —
from the uterine mucosa under these conditions, it has been
observed that if cancer develops, it tends not to be of the columnar
or adeno-carcinomatous type usual in the uterus, but to be
definitely epitheliomatous and of the squamous - celled type.
The development of primary epithelioma has been more than
once recorded in the gall-bladder, and, as there is no squamous
epithelium anywhere in the neighbourhood of the normal gall-
bladder, the generally accepted conclusion is that there has been
a pre-existing chronic inflammatory condition, which has caused
metaplasia of the columnar - celled mucosa into a stratified
epithelium. So also where an epithelioma arises in the larynx,
away from regions where — as along the edge of the vocal cords —
such stratified epithelium is normally present, the most satis-
factory conclusion is that a similar metaplasia has preceded the
cancerous development.

Now certainly in the first of these cases, and most probably
in the others, long-continued irritation and modified conditions
of environment have led to a change in structure, and this
change has impressed itself upon the affected cells to such an

1 Adami, " The Habit of Growth," Montreal Medical Journal, xxiv., 189G,
581.

300 ON GROWTH AND OVERGROWTH

extent that even when these cells take on cancerous proliferation,
and their descendants penetrate into the deeper tissues — away
from the conditions which affected their progenitors — they still
retain well-marked evidences of their acquired characters, and
they produce, not a columnar-celled, but a squamous epithelial
cancer. I fail to see how such cases can possibly be explained
unless we accept this principle of what we may term the inheritance
of acquired characters by the individual cells of the organism.
Pathologists in general freely accept this as the explanation of
the occurrence of primary squamous-celled epithelioma ta in the
uterus and the gall-bladder. Apparently, so far as I have seen,
the significance of these tumours, as indicating an important
property of the cancer cell, has not been hitherto realized.

It is nowadays almost needless to point out that to explain
immunity we are forced to admit a similar well-marked inheritance
of acquired characters by various cells of the organism. And so
I urge that in connexion with cancers and with tumours in
general these facts may be applied to the elucidation of tumour
growth. When cells have for long been subjected to those
modified conditions under which thev multiplv activelv and
rapidly, then, provided the stimulus or cause of this state acts
over a sufficiently long time, the cells take on the " habit of
growth." They continue to proliferate in a purposeless manner
long after the factors which started the process have ceased to
act. Their growth and proliferation is no longer inhibited by
the performance of specific functional activities ; it becomes
wholly independent of the needs of the organism, and the more
extreme this independence of growth and the more the cells
lose the signs of their primordial function, the more malignant
the growth. Provided that there is adequate nutrition it matters
little where these cells find themselves ; they grow for them-
selves. Assimilation and mitosis are the two cell f mictions
retained and impressed upon them. They have cut adrift from
the old relationships and tensions which determined the primor-
dial, activity of the parent cells. They have reverted from the
specialized structure impressed upon them by that primordial
activity, as by heredity, until but a trace of that structure is left.
They assume the form common to cells undergoing mitosis, a
form we term embrvonic, and that because this is the form
characterising actively prolif era ting cells in the embryo. This

INHERITANCE OF ACQUIREMENTS BY CELLS 301

form, let me repeat, is not so much embryonic as proliferous.
It is peculiar to actively dividing cells at all times of life.

Whatever the origin, therefore, of the tumour proper, however
it is started, what makes the tumour is the assumption by the prima ry
cells of that tumour of the habit of growth in place of the habit of
work, and, according to the extent of this replacement, so do we get
the various grades of tumour formation from the most benigyi to
the most malignant.

Summary of Argument

Let me now sum up the various points brought forward one
after another in arriving at this conclusion. They are :

1. The katabolic activities of the cell are of two orders :
those determining the relationship of the cell with the exterior,
and those that are vegetative determining the continued existence
and multiplication of the cell ; the former excited by stimuli
of various orders from without, the latter only indirectly so
excited, being more directly called into play by conditions
obtaining within the cell.

2. The controlling agency in at least the higher katabolic
activities of the cell, both " functional " and " vegetative," is
the nucleus, and nuclear activity is accompanied by break-
ing down and discharge, or by rearrangement of the nuclear
molecules.

3. The changes which occur in the nucleus during the active
performance of the specific functions of the cell are of a character
so different from those observed during the process of cell division
that proliferation and active performance of specific function,
the one precluding the other, are obviously to a large extent
incompatible. The cell engaged in the active performance of
function in response to external stimulation cannot simultane-
ously proliferate.

4. It follows, therefore, that active cell division and cell
proliferation occur only in conditions in which the cell cannot
fully utilize the assimilated material (and the energy stored up
in the assimilation of that material) in the performance of its
specific functions.

5. Such conditions are to be met with where the tensions
acting on the cell are reduced and certain energies which before

302 ON GROWTH AND OVERGROWTH

were necessary to counteract opposing forces are freed and
become thus capable of diversion from their purpose, or, again,
where stimulation from without results in increased assimilation
and storage of nuclear and cell material which now from any
condition cannot be utilized in the performance of specific
function.

6. In either case the cells will continue to proliferate so long
as the primary modification of physical relationships or the
primary stimulus continues to act, so long as there is adequate
nutriment, and so long as the tensions exerted upon the cells
do not become excessive.

7. Provided that these conditions are observed, the greater
the amount of cell proliferation, the greater is the tendency for
certain at least of the newly formed cells to be projected from
the relations proper to cells of the tissue giving them origin, the
less will be the opportunity for such cells to carry on their
primordial function, and the greater the liability to proliferation.

8. The longer the cells are diverted from their proper extrinsic
functions to proliferative activity, the greater the momentum
acquired by them to continue performing the proliferative act
until the functional activities become largely suspended and the
" habit of growth " is set up.

9. When this habit of growth is inaugurated the cells can
continue to grow and multiply in the complete absence of those
conditions which initiated their proliferation in the first place,
and we obtain that purposeless functionless cell growth char-
acteristic of the true tumour.

10. According to the stage of cell development in which this
habit becomes impressed upon the cell, so do we have the various
grades of benign and malignant tumour formation.

Application of Theory

How now does this theory apply itself to the possible microbic
origin of at least certain forms of malignant and other tumours ?
It applies itself thus : According to this theory, microbes and
their products may be one of the causes originating localized
cell proliferation in the first place, provided that (1) they bring
about stimulation rather than irritation, or irritation of so mild
a type that the cells are stimulated to an increased metabolism

THE " HABIT OF GROWTH " THEORY 303

which does not go on to exhaustion and excessive breaking-down
of their protoplasm ; provided, also, that the microbes and their
products continue in action for a sufficiently long time to set up
the habit of growth. It is quite conceivable that such microbes
might continue to exist in the tumours they originated, exerting
a cumulative effect. The more the cells depart from type the
greater the effect of these microbes and their products in pro-
ducing a tissue of rapidly proliferative and malignant type.

This continuance and persistence of microbic action, however,
must not, I think, be regarded as essential. The very fact that
after all these years, and after the hosts of careful observations,
we are still in very grave doubt as to whether any of the bodies
seen in tumours are really parasites, the fact that no growths of
these bodies have surely been obtained outside the organism,
and then upon injection have induced tumour formation, although
by no means proof absolute, may be quoted in favour of the view
that if microbes originate malignant tumours, they do not
continue in the living state.

I was not a little interested to note that while Sanfelice * was
apparently able in his observations to induce the formation of
neoplasms in two out of rather a long series of animals, following
upon inoculation of these animals with his blastomycete — the
Saccharomyces neoformans — when he came to make cultures
from the tumours and from the other tissues of the animals
possessing these developed tumours he was unable to gain any
growths. An observation somewhat to the same effect I have
come across recorded by Dr. T. Harris,2 of Manchester, in a case
of cancerous growth of the bladder associated with and evidently
secondary to bilharzia disease. Harris calls particular attention
to the fact that although there were abundant ova of the bilharzia
in the surrounding vesical tissue, the tumour itself stood out
completely free from any sign of these parasites.

It is quite possible, according to this theory, that certain
specific forms of microbic life originate certain forms of tumour
growth — of cancer and sarcoma, for example ; that, like other
pathogenic microbes, these may show a predilection to attack
special tissues under special conditions ; and if it be true (for
some doubt continues to be thrown upon the matter) that

1 Sanfelice, Centralbl. f. Bali, xxiv., 1S0S, 8. 166.
2 T. Harris, Trans. Path. Soc, London, vol. xxxix. p. 183, 1888.

304 ON GROWTH AND OVERGROWTH

malignant growths are specially common in certain localities,
then such microbic origin becomes eminently probable.

But if this theory be true, it does not follow that, discovering
the causative microbe, we shall be able to arrest the development
of cancer bv antimicrobic or antitoxic means. At least, in the
present state of our knowledge, I fail to see how we can. The
most we can look forward to is, on the one hand, the discovery
and employment of means whereby to destroy or arrest the
continuous growth of cells which have taken on this functionless
growth ; and, on the other hand, studying their habits outside
the body, to exterminate the causative agents, just as nowadays
it is being attempted with marked success to exterminate the
mosquitoes in some malarial regions.

I would urge, then, that at the present time, when after all
these years of labour no causative agent of malignant growth in
general has with certainty been determined, that line of research
which promises surer results and greater profit on the part of
clinicians, as well as of laboratory investigators, lies in the direc-
tion of testing various methods of arresting the growth of the
tumour cells without injury to the organism in general. Herein
it seems to me that Coley 1 has chosen the better part. But over
and above all I cannot but feel that the greatest benefit to the
patient, the greatest triumph and satisfaction to the practitioner,
will for yet some years to come be the recognition and successful
removal of malignant tumours at the earliest possible date.
Aye, and what is more, the removal of benign tumours in general
before they have taken on possible malignant characters.

I wish I could be more optimistic, for optimism is good and
healthy, but the above deductions are the necessary outcome
of the considerations I have put before you. As to the hypo-
thesis here put forward, I am quite prepared to find that many
will at first have difficulty in grasping it, but when grasped, I
cannot but believe most firmlv, it will be found to include and
to explain the largest number of individual cases — that it will
be found to satisfy the conditions of sound theory.

1 By his treatment with mixed streptococcus and pyocyaneus vaccines.

CHAPTER III

ON THE CLASSIFICATION OF TUMOURS 1

(1902)

Once it was established by histological studies that the different
forms of neoplasms arise from different tissues, it became possible
to group tumours according to their origin. Once again, when,
through the observations of the embryologists, it was recognized
that the tissues could be grouped according to their origin from
the primitive cell layers, it became possible to group tumours
in the same way. This was done more especially by Waldeyer,
and we obtained thus a classification of tumours into those of
mesoblastic origin and those of epiblasiic and hypoblastic derivation,
the tumours derived from the two latter cell layers being grouped
together, for it was soon found that they were of the same general
type.

Here was a broad and very important generalization, and,
what is more, in the then stage of histological and embryological
knowledge, it appeared not only to be founded on a sure and
scientific basis, but to fulfil to the fullest the needs of the worker.
For it appeared to separate two sharply differentiated orders of
tumours — those of connective tissue origin and connective tissue
type from those of epithelial and glandular origin and epithelial
and glandular type. So that for long years this distinction
remained dominant ; even to-day, in at least one text-bo< >k
published during the last twelve months, that by Dr. Nicholas
Senn, this is given as the acceptable classification.

It is, however, scarcely necessary to say that, with increasing

1 The main body of this paper was contributed as an address to the Toronto
Pathological Society, January 4, 1902. Reprinted from the Journal of Patho-
logy and Bacteriology, Edinburgh, June 1902.

305 X

306 ON GROWTH AND OVERGROWTH

knowledge, manv cases were discovered which did not fit into this
scheme. A classification which brings together unlike bodies,
making them members of one group, is, on the face of it, in-
adequate and faulty. Now, more especially during the last ten
years, pari passu with a recognition of the bearing of the fuller
and more recent findings of the embryologists, this classification
has been found to have the above failing. It is, for example,
generally accepted that the specific and characteristic cells of
several tissues of the glandular type — of the kidneys, suprarenal
bodies, ovaries, testes, and uterine mucosa — are of mesoblastic
origin, but these, nevertheless, give rise to tumours which may
and often do resemble most closely those of hypoblastic and
epiblastic origin.

There has been grave doubt as to the embryogeny of the
organs in question. The idea that tissues of glandular type can
only be derived from the two primary cell layers is very firmly
fixed, and in one direction the attempt has been made to show
that the organs in question are of hypoblastic or epiblastic origin ;
on the other, to make out the distinction between these organs
and what have been termed " true glands." But I am only
expressing the general opinion of modern embryologists and
histologists when I say that all these organs now are accepted
by the majority as being definitely derived from mesoblast.
And thus the cancer-like tumours which originate in these organs
must be accepted as being mesoblastic.

On the other hand, the gliomata have a structure which
brings them into close alliance with the atypical or malignant
connective tissue tumours, and yet the neuroglia, from which
they are derived, is of epiblastic origin. The notochord, again,'
is an organ of hypoblastic origin. According to Ribbert, and the
view is becoming accepted, the remains of this foetal organ may
give rise to tumours somewhat resembling myxomata,1 that is
to say, to tumours which, though of hypoblastic origin, are of
connective tissue type. Histologically, and for practical purposes,
the first series above mentioned ought to be grouped along with
the adenomata and carcinomata, and the second with the sarco-
mata and connective tissue tumours, but the old embryological
classification forces us to make the very opposite arrangement.

1 Tumours of mucoid type, resembling the connective tissue of the foetus
as seen in " Wharton's jelly " of the umbilical cord.

HISTOLOGICAL CLASSIFICATION 307

These difficulties have induced so strong a reaction that we
have only to read the recent text-books and articles published
during the last ten years to recognize that pathologists in general,
nowadays, refuse to consider embryogeny in their schemes of
classification, and from Thoma, or even earlier, from Hamilton
in 1889 onwards, through Ribbert and Lubarsch — the list is
so long that I need not give it — the tendency has been to divide
the autonomous neoplasms into those of typical and atypical
connective tissue appearance, and those of typical and atypical
glandular appearance. Some, like Hansemann,1 would go so
far as to declare that tumours must be described purely according
to their histological appearance, and while certain terms in general
use must continue to be employed — terms such as adenoma,
sarcoma, etc. — nevertheless, the only right classification at the
present time must be by the organ, the tumours originating
from one or other tissue being grouped together. So that, for
example, we must group together the adenomata and carcinomata
of the stomach. In short, they urge that the topographical
classification is the only one possible at the present time.

There is undeniably a virtue in this position, provided that
it is assumed in a proper spirit and regarded in the right light ;
not as a final stage, beyond which it is impossible to advance, but
as a temporary stage of careful collection and collation of all the
facts bearing upon the tumours proper to each individual organ,
to the end that we may, from the knowledge so gained, proceed
to further and sounder generalizations, that we may utilize the
facts so amassed, to formulate broad statements concerning
neoplasms, their relationship one to the other, and their mode
of growth.

But against such a position this has to be said : All these
years, whatever the scheme of classification popular at one time
or another, pathologists have not been idle, so that we are already
in possession of an enormous amount of material, and, what is
more, of accurate drawings of the same. Hence, if preconceived
notions as to embryological relationships modified the earlier
conclusions reached concerning one or other form, the details
have been honestly described, and the descriptions and the
accompanying illustrations help us to determine where the

1 D. von Hansemann, Die mikroskopische Diagnose der bdsartigen Qe-

schwulste, Berlin, 1897, p. 22.

308 ON GROWTH AND OVERGROWTH

earlier conclusions need correction. Nay, more, we already
possess exhaustive studies upon the various forms of tumours
affecting one or other organ, the mammary gland, uterus, kidney,
skin, etc. Taking everything into consideration, the time ought
to be ripe for attempting more than this individualizing method.
So, to repeat, I believe that we have by us embryological and
anatomical observations which permit us to proceed further.
What is more, I believe that, for a sound classification, we must
inevitably pass backwards to the developmental relationship
of the different tissues, that we must accept an embryogenetic
basis, but one more in consonance with our present knowledge.

It is not necessary here to dwell upon the relationship of
morbid to normal processes, and to show that the former in each
case are but exaggerations in one or other direction of the latter ;
nor need I point out the importance of a knowledge of the develop-
ment of each tissue in arriving at a determination of the modi-
fications which may be undergone by that tissue. In his Middle-
ton-Goldsmith lecture in New York, Minot * has, during the last
year, treated this latter subject in so masterly a manner that
anything I could say would be but a feeble reflection of his
admirable presentation of the bearing of embryology upon
pathology. What I wish now to point out more especially is
that we pathologists — and we are not by any means the only
ones to blame — have during these years continued to hold fixed
and stereotyped views with regard to the exact nature and
development of the different germinal layers, and it has been this
misconception of these layers and the changes undergone by
them, and of the mode in which the various tissues have been
derived from them, that has brought us to this stage of discarding
classifications constructed along embryological lines.

We have, that is, held as a body, that from a given layer,
as, for example, the epiblast, only tissues and tumours of one
general type are developed, and in practice we have found that
this is not wholly the case. We have concluded that embryo-
genesis is a broken reed, and this despite our willingness to dis-
cover in it the basis for sound classification.

In order to indicate how a classification which is primarily
embryological can be developed, it will be necessary for me to

1 " On the Embryological Basis of Pathology," Science, U.S., xiii., 1901,
481.

EMBRYOGENETIC CLASSIFICATION 309

indicate rapidly what are the leading facts with regard to the
earlier stages in the development of the different types of tissue,
and I must recapitulate matters of a most elementary nature ;
nevertheless, if by doing this I can make my argument clear, I
trust that I shall be forgiven.

The earliest stage to be recognized in the development of
the fertilized ovum, once it has proceeded to segment, is the
production of a morula, in which the blastomeres form a cluster
or group of cells of the same order, with almost entire lack of
differentiation. Rapidly this gives place to a second stage, in
which the component cells arrange themselves into two layers,
the epiblast and the hypoblast, so that, at a singularly early
stage, the future epiderm and endoderm are recognizable. The
next stage to be noted is that the hypoblast, or internal of the
two primitive layers, gives rise by proliferation of its cells to a
group or mass of cells which now lie intermediate between the
primitive epiblast and the hypoblast, and form the " Anlage "
of the mesoblast and of the organs derived from that layer, the
hypoblast itself still remaining as a distinct lining membrane.
What I would here note is that we are perfectly willing to advance
thus far and recognize these two primary and the third secondary
cell layer. But there, in our appreciation of embryology as bear-
ing upon pathology, we have been strangely apt to stop.

But now, just as the hypoblast gives origin to the mesoblast,
so it is perfectly legitimate for us to recognize a similar process
on the part of the epiblast ; for the epiblastic cells in the im-
mediate neighbourhood of the primitive groove proliferate rapidly,
and, in so doing, project in part below the original line of the
epiblast, and, being forced inwards, a regular mass of cells is
developed, in the central portion of which, around the spinal
canal, there still remains evidence of its origin, in the form of
a definite epithelial lining. This second portion becomes cut
off completely from the superficial epiblast to form the mother
tissue of the nervous system. Similarly, the hypoblast gives
off a second localized mass of cells to form the notochord.

Professor Minot has called my attention to the fact that,
while possessing special features, the cells forming the notochord
throughout retain epithelial characters, that they are grouped
together without an intercellular matrix, that no vessels penetrate
between them, and that, although in degeneration these cells

310 ON GROWTH AND OVERGROWTH

assume a somewhat dubious appearance, they can most assuredly
not be regarded (as I find, on reading von Hansemann and other
recent writers, I had not been alone in regarding them) as assum-
ing a connective tissue type. The nature of the cells forming
the chordoma of Ribbert makes him more than doubtful as to
the origin of this form of tumour from notochordal remains. I
have thus, in deference to his authority, modified to some extent
my original statements regarding this.

The point I wish here to indicate is that mesoblast, " neuro-
blast," and notochord are derived from the two primitive cell
layers, and the first two at least lose the lining-membrane charac-
teristics of these two earliest layers, and take on a less differentiated
condition prior to further evolution.1 At a somewhat later period
the mesoblast repeats the process of differentiation, and, from
being a simple undifferentiated cell mass which we may compare
with the morula, certain of its cells growing outwards between
the epiblast and hypoblast become arranged into a definite layer,
to form or enclose the primitive body cavity.2 From this point
onwards we can distinguish two structures of mesoblastic origin —
the mesothelium, or lining membrane portion of the mesoblast ;
and the mesenchyme, or, as I may term it, the mesoblastic pulp.

It will be seen that I here make no note of the separation of
mesoblastic elements into archiblast and parablast, as laid down
by His, and made the basis for a classification by Klebs and
Williams. His's conception of the parablast, as arising from
the elements of the white yolk and from the " granulosa cells,"
is now known to be wrong, and indeed he has himself withdrawn
his earlier hypothesis as to its origin. Add to this, that so hope-
less a confusion has arisen among writers as to what is archiblastic
and what parabiastic, that we have no option but to discard these
terms. As pointed out by Minot, the recognition of the separa-
tion of the mesoblast into mesothelial and mesenchymatous
elements respectively, suffices for all practical purposes to indicate

1 [To-day, with all deference to the late Professor Minot's deep knowledge,
I would support my original contention that the human notochord, and the
chordomas originating therefrom, assume the connective tissue type of cell
relationship.]

2 [Even if the body cavity or coelum had its primitive origin as a pouching
of the hypoblast, or endoderm lining the enteron, there is, to my knowledge,
no clear evidence that in the body cavity of higher animals this is formed other
than by a splitting of the mesoblast with subsequent differentiation of the
surface cell lining into a mesothelium.]

THE CELL LAYERS 311

that which is of real importance in His's observations, namely,
the recognition of the ultimate evolution of the primitive meso-
blast into two distinct series of cellular constituents.

For our purpose it is unnecessary to choose definitely between
the two main contending views as to the origin of mesenchyme.
Whether all the primitive mesoblast first passes into a mesothelial
stage, from which by further proliferation the mesenchyme is
derived ; or whether, on the other hand, a portion of the meso-
blast does not undergo conversion into mesothelium, but con-
tinues directly to develop into mesenchyme, is for us a relatively
secondary matter. The important point is that we have to
recognize that the primitive mesoblast is eventually separated
into these two sets of cells, and of these the mesothelium is
differentiated into a layer of the lining membrane type.

At a still later date masses of mesothelial cells again accumu-
late, and, as was the case with the epiblast and the hypoblast
in the earlier stage, they give off on either side a mass of more
undifferentiated cells, and these masses form the mother tissue
or Anlage of the eventual striated muscle. Later, though still
in this embryonic period, with the development of the first
vessels, the mesenchyme gives off a series of cells of the lining
membrane type, which form the eventual lining cells, or endo-
thelium, of the vascular and lymphatic systems.

There is still some little uncertainty as to the exact relation-
ship of the vascular endothelium, whether it be directly derived
from mesothelium or from the mesenchyme. As Professor
Minot has pointed out to me, His has of late indicated that it
is of relatively very early development in certain forms. On
the other hand, I learn from Professor MacBride that relatively
high up among the forms of animal life it may be wanting, as,
again, it may only show itself at a period definitely later than
the development of the vascular channels. What I have stated
above, thus, may be taken as representing, as accurately as is
possible at the present time, the generally accepted relationships
in time of endothelial to other embryonic developments.

Thus during early embryonic life we obtain a series of difl<
entiations of the primitive cell layers leading to the production
of two sets of tissues ; one which we may term the lining nu in-
brane tissues, the other the pulp tissues. I do not wholly like
the latter expression, but can think of none other which more

312 ON GROWTH AND OVERGROWTH

nearly expresses the conception which I wish here to impress
on the reader, namely, that in this very earliest stage the mother
cells from which certain tissues are derived already have a definite
order and position, as constituting membranes, whereas the
mother cells of the other group of tissues lack this definite order
and exhibit no marked differentiation.

If, now, we follow up these two orders of tissues to their full
development, we find that from the lining membranes are de-
veloped tissues of one character ; from the pulp, tissues of a
different character. These lining membranes, from whatever
layer they originate, may either remain as functional membranes
covering an extensive surface, or they may become modified
so as to form the various highly specialized and specific con-
stituents of various organs. To reach this latter stage the
epithelium either undergoes extensive infolding, or sends down-
ward tubular processes, which branch to a greater or less extent,
or, again, which sends downward solid processes, as occur, for
example, in the development of the suprarenal. In quite a large
number of instances it would seem that the first process in the
production of the tubular glands is one of budding, or projection
downwards into the underlying mesenchyme, of solid processes,
which subsequently became hollowed out and tubular (e.g. sudori-
ferous glands and renal tubules). Thus it comes to pass that
the various glandular organs are formed of a parenchyma originat-
ing from one of the primitive lining membranes and a stroma
of connective tissue which is of mesenchymatous origin. And
even in cases where there is the widest divergence from the original
type of lining membrane, we find that this distinction still holds,
that the parenchymatous cells form layers or groups into which
the vessels do not penetrate, and in which there is an absence of
stroma betiveen the members of the cell groups, the cells being at
most united by bridges and by a fine cement material. While,
contrariwise, regarding tissues originating from the embryonic
pulp, we notice that in them the prominent characteristic is
that there is an intercellular ground substance, either homogeneous
or fibrillated, separating the specific cells of the tissue.

Accepting this conception of the different characters of the
different tissues, it will be seen that these may be divided into
two great groups, which we may term, provisionally, the lining
membrane group and the body pulp group. This division

PLATE VI
Schema of Tissue Relationships

Epiblast
lepidic

Cutaneous

Mesenchyme
hylic

Spinal canal
Epiblast
/*' iridic

Mesothelial
skeletal muscle
hylic

Vascular

endothelium

lepidic

Liver
lepidic

Mesothelium
lepidic

Epibla&1
NeurencephalOD

/"/lie

Notochord

hypoblast
hylic

Mesotlu'liiiiii
glandular
organs
lepidic

Pleuro-
peritoneal

cavity

Lumen of alimentary
canal

Mucous
membrane

Hypoblast

lepidic

LEPIDIC AND HYLIC TISSUES 313

appears to me most important. While some pathologists, like
Buxton x and 0. Israel,2 have already noticed this distinction,
I do not know that histologists and embryologists have called
adequate attention to it. Though I am strongly against the
coinage of new scientific terms, there are occasions when this is
absolutely necessary, and this appears to be pre-eminently one
of these occasions. Thus, I would term the lining membrane
tissues lepidic, from \e7rt?, XeirlSo^, a rind, skin, or membrane ;
and what I have termed the pulp tissue hylic, from v\r], crude
or undifferentiated material or matter.3 We can go further
and subdivide each of these main groups according as to whether
the tissues are of epiblastic, hypoblastic, mesothelial, mesenchy-
matous, or endothelial origin. On this basis we obtain the
following classification of normal tissues :

I. Lepidic or Lining Membrane Tissues

in which the blood-vessels do not penetrate the groups of specific
cells, and in which there is an absence of definite stroma between
the individual cells, although such stroma, of mesenchymatous
origin, may be present between the groups of cells.

1. Epiblastic.

Epidermis. Epidermal appendages — hairs, nails, enamel of
teeth, etc. Epidermal glands. Epithelium of the mouth,
salivary glands. Epithelium and glands of nasal tract and
associated spaces. Epidermal portion of hypophysis cerebri.
Lens of eye. Epithelium of membranous labyrinth of ear,
anus, male urethra (except prostatic portion).

2. Hypoblastic.

Epithelium of digestive tract and glands connected with it.
Specific cells of liver, pancreas, tonsils, thymus, thyroid.
Epithelium of trachea, lungs, bladder, female urethra, male
urethra (prostatic portion).

3. Mesothelial.

Lining cells of pleurae, pericardium, peritoneum. Specific cells
of suprarenals, kidneys, testes, ovaries (Graafian follicles).
Epithelium and glands of Fallopian tubes, uterus, vagina,
vasa deferentia, vesiculae seminales, etc.

1 Buxton, Journ. Cutan. and Gen.-Urin. Dis. xviii., 1900, 74.

2 Israel, Berliner klin. Woch. xxxvii., 1900, 609, 644, 667.

3 I would here acknowledge my indebtedness to Principal [now Sir William]
Peterson of McGill University for aid in the selections of these terms.

314 ON GEOWTH AND OVERGROWTH

4. Endothelial.

Lining endothelium of blood-vessels, lymphatics.

II. Hylic or Primitive Pulp Tissues

Organs and tissues in which the special characteristic is that
the specific cells lie in, and are separated by, a definite stroma,
homogeneous, or fibrillar, in which there may or may not be blood
and lymph vessels.

1. Epiblastic.

Nerve cells, neuroglia.

2. Hypoblastic.

(Notochord. See note, p. 310.)

3. Mesenchymatous.

Fibrous connective tissues, cartilage, bone, reticulum of lymph
glands, bone marrow, fat cells, involuntary muscle tissue,
spleen, blood-vessels, blood corpuscles.

4. Mesothelial.

Striated muscle, including cardiac muscle.

With this conception of the two great groups of tissues, we
can now proceed to classify the tumours, by which term I refer
here to what Thoma has called " autonomous neoplasms." Of
these there are two great orders, the Teratomata and the Blasto-
mata. The former I have elsewhere defined as " tumours
composed of the products of growth of one individual within
the tissues of another individual of the same species," the latter
as " tumours composed of the products of aberrant growth of
cells and tissues of the individual in whom they are developed." x
It is not necessary here to discuss the correctness of these defini-
tions, for, however defined, I wish here to leave the Teratomata
very largely out of consideration ; they form a class by them-
selves, and whether we accept or do not accept the definition
above given, we find that their mode of development and their
characteristics follow — with complications — the lines about to
be laid down with regard to the Blastomata. These latter form
the more important class, and it is with them that I wish specially
to deal.

Following this scheme of the classification of the normal
tissues, we may now divide these into two main genera — the

1 British Med. Journ., London, 1901, i. 621. See pp. 285, 286 of this
volume.

LEPIDIC AND HYLIC TUMOURS 315

Lepidomata, originating from the above lining membrane
tissues, and the Hylomata, originating and derived from tissues
developed from the embryonic pulp.

I. Lepidomata or " Rind " Tumours

A. Primary Lepidomata

1. Epilepidomata.

Tumours whose characteristic constituents are overgrowths of
tissues, derived directly from the epiblastic lining membranes,
or true epiblast.

(a) Typical. — Papilloma, epidermal adenomata (of sweat,

salivary, sebaceous, and mammary glands, etc.).

(b) Atypical. — Epithelioma proper (of skin), carcinoma of

glands of epiblastic origin.

2. Hypolepidomata.

(a) Typical. — Adenoma and papilloma of digestive and

respiratory tracts, thyroid, pancreas, liver, bladder,
etc.

(b) Atypical. — Carcinoma developing in the same organs

and regions.

B. Secondary Lepidomata

3. Mesolepidomata.

Tumours whose characteristic constituents are cells derived in
direct descent from the persistent mesothelium of the embryo.

(a) Typical. — Adenoma of kidney, testicle, ovary, uro-

genital ducts ; adenoma of uterus and prostate ;
adenomas originating from the serous membranes,
" mesothelioma " of pleurae, peritoneum, etc.

(b) Atypical. — Cancer of the above-mentioned organs ;

squamous endothelioma, so called, of serous sur-
faces, epithelioma of vagina.

4. Endothelial Lepidomata.

Tumours originating from the endothelium of the blood and
lymph vessels ; endothelioma, perithelioma.

II. Hylomata or " Pulp " Tumours

1. Epihylomata.

Tumours whose characteristic constituents are overgrowths of
tissues derived from the embryonic pulp of epiblastic origin.

(a) Typical. — True neuroma, glioma.

(b) Atypical. — '* Gliosarcoma."

316 ON GROWTH AND OVERGROWTH

2. Hypohyloniata.

Tumours derived similarly from embryonic pulp of hypoblastic
origin.

(?) Chordoma.

3. Mesohyloniata.

A. Mesenchymal Hylomata. — Derived from tissues originating
from the persistent mesoblastic pulp or mesenchyme.

(a) Typical. — Fibroma, lipoma, chondroma, osteoma,

myxoma, leiomyoma.

(b) Atypical. — Sarcoma (derived from mesenchymatous

tissues), with its various subdivisions, fibrosarcoma,
spindle - cell sarcoma, oat - shape cell sarcoma,
chondro- sarcoma, osteo- sarcoma, myxo - sarcoma,
melanotic sarcoma, etc.

B. Mesothelial Hylomata. — Tumours which are overgrowths
similarly of tissues derived from embryonic pulp of definitely
mesothelial origin.

Rhabdomyoma.

It will be seen that in this classification I do not include
the deciduoma malignum. As I have pointed out elsewhere,1
accepting the view that the syncytium is of foetal origin and
not maternal, these tumours have to be classed with the Tera-
tomata, i.e. with the tumours originating in the growth of the
cells of a second individual within the tissues of an individual
of the same species.

If this classification be studied, it will be seen that we have
done away with that deficiency in the earlier embryological
classifications whereby tumours of unlike orders and histological
appearances were grouped together and those of like characters
separated. Gliomata, for example, come to be placed close to
the mesenchymatous tissues ; the gland-like tumours of meso-
blastic origin become grouped along with those of epiblastic
and hypoblastic origin.

Have we, in accomplishing this, introduced any new diffi-
culties ? One objection will undoubtedly present itself, namely,
that among the Mesolepidomata we have grouped together
tumours some of which are of a strongly epithelial or glandular
type, for example, the cancers of the uterus, with others like the
endotheliomata, which tend to be distinctly of a sarcomatous
type. But further consideration will show that this, instead of

1 Loc. cit. See pp. 286 and 326 of this volume.

Fig. 1. — Diagrammatic representation of a lepidic tissue, «, with its underlying
hylic connective tissue ; b, Capillary lined by lepidic endothelium.

Fig. 2. — Diagrammatic representation of a hylic tissue, with intercellular ground
substance of fibrillae in a ho nogeneous matrix separating tbe specific cells.
a, Nuclei of specific tissue cells, b, A capillary lined by Lepidic endothelium.

CARCINOMA AND SARCOMA 317

being a weakness, is a strong point in this classification. We
have, that is, to recognize that among these Mesolepidomata,
as above defined, we meet with several forms of tumours of
transitional type, tumours which in their least aberrant portions
show characters which approximate them to the Carcinomata,
and in their more aberrant portions are indistinguishable from
Sarcomata. And, indeed, it is only by a study of the embryogeny
of the tissues from which these tumours are derived that we gain
any satisfactory comprehension of the why and wherefore of
these peculiar characters.

Here let me point out that, employing the terms here intro-
duced, based as they are upon the embryogeny of the different
tissues and the tumours derived from them, we may allow the
terms " carcinoma " and " sarcoma " to revert to their earlier
and purely histological significance. And I would emphasize
that it must not be understood that these terms, carcinoma
and sarcoma, are to be regarded, and are by me regarded, as being
synonymous with " atypical lepidoma " and " atypical hyloma "
respectively. Rather, I would lay down that, accepting this
nomenclature, we may safely speak of any tumour of the aberrant
glandular type as carcinoma, whether it be of epiblastic or
mesothelial origin, and any tumour of aberrant and so-called
embryonic connective tissue type as sarcoma, whether derived
from the mesenchyme, the epiblast (e.g. gliosarcoma), or even
from the endothelium or mesothelium.

Of late years there has been an ineffectual attempt to restrict
these two terms. Thus, many authorities have refused to speak
of malignant adenomata of the kidney and suprarenal as being
true carcinomata, and others have strenuously opposed the
employment of the term gliosarcoma. Nevertheless, the same
authorities, while refusing to speak of a cancer or carcinoma
of the kidney, freely refer to carcinoma of the uterus, although,
like the renal tubules and the suprarenal, the uterine mucosa is
of mesothelial origin. In short, it has been proved impossible
to employ these terms with embryogenetic limitations, and this
introduction of a nomenclature which is based upon embryogeny
ought, if accepted, to permit us to use them, as they ought to
be used, in the purely histological sense.

It will be seen that I in no sense urge that (were it possible !)
the use of these terms (sarcoma and carcinoma) be done av

318 ON GROWTH AND OVERGROWTH

with, for, accepting the purely histological use of the terms, we
can unreservedly agree with Israel that it is a matter of taste
whether we speak of a given tumour as an " endothelioma
carcinomatosum or sarcoma endotheliale." For routine clinical
purposes they are most valuable. When a tumour has assumed
a carcinomatous or a sarcomatous appearance it is coincidently
locally malignant, if not in all cases generally malignant also.
The terms, therefore, have a clinical significance and value. Only,
let me repeat, they are valueless for purposes of relationship and
classification, and must bear no embryogenetic signification.

To explain these peculiarities of lepidomatous tumours let
me point out that —

1. After the embryonic period, it would seem that hylic
tissues never take on lepidic characters. We have no instances,
that is, in which, after embryonic life, we recognize that lining
membranes or glands become differentiated from connective
tissues.

2. It is generally held that the converse is also true.1 With
regard to tumours we find the same principle evidently in
operation.

3. We may confidently lay down that all tumours and portions
of tumours containing cell layers or cell groups of the lepidic
type have been derived from pre-existing lepidic tissue.
Possibly this so-called principle is more of the nature of a
postulate than of a proved law ; we take it for granted, and
may not be able to prove our position in every case. We have,
that is, fairly numerous examples of neoplasms of lepidic type
developing in situations in which normal lepidic tissue is not
present — adenomatoid tumours of the bone, gland-like follicles
in the midst of uterine fibroids, cysts, or tubular spaces fined by
cubical or columnar epithelium in gliomata — and the fist might
be lengthened.

1 True, that is, until Beard's rather startling observations upon the origin
of leucocytes is confirmed. According to this observer the first, and indeed the
main, development of leucocytes is to be found as a process of proliferation and
metamorphosis of the hypoblastic epithelium of the follicles of the foetal thymus
gland. Beard's observations were published eighteen months ago, but to the
best of my knowledge they still lack confirmation. Leo Loeb also (Arch. f.
Entwicklungsmech. d. Organ., Leipzig, vi., 1898, and Medicine, April 1899) ha3
thought to see connective tissue cells undergoing origin from the Malpighian
layer of the skin : his observations have not gained acceptance. In lower
forms of life, however, some definite cases of regeneration of hylic from lepidic
tissues have of late been brought forward.

THE LEPIDOMAS 319

In by far the larger number of cases of this order, either the
structure of the tumour so conforms to known neoplasms, the
origin of which has been traced positively to some glandular
organ, that we are convinced that the growth has originated
from the inclusion of a portion of such glandular tissue, or we
recognize that the growth occurs in some region, in which, during
foetal life, there has existed some duct or portion of lepidic tissue,
in a region, likewise, in which in the adult we occasionally en-
counter the persistent remains of the same. But pathologists,
I believe without exception, agree to the sense of this postulate ;
we may not know, in all instances, what is the original tissue
from which a heterotopic glandular tumour has had its origin,
but we are absolutely sure that that original tissue giving rise to
a glandular tumour has not been of the connective tissue type.
We may, therefore, I think, lay this down with confidence, that
tumours in which the cell arrangement is of the lepidic type,
presenting columns or groups of cells, devoid of any stroma
between the members of these columns or groups, have been
developed from one or other lepidic tissue, and not from one
of the hylic or pulp tissues.

4. We cannot, with equal confidence, make the converse
statement, that tumours of the hylic type have always originated
from hylic tissues and not from lepidic.

4a. I believe that I am right in stating that lepidic tumours
of epiblastic and hypoblastic origin, however rapidly they grow,
however extensive and distant be their metastases, always, even
in their most aberrant portions, retain lepidic properties. Wher-
ever two or three of the specific cells of such a tumour are gathered
together, they form alveoli with no stroma and no interstitial
capillaries. To the best of my knowledge, no case is on record
in which it has been satisfactorily proved that peripheral or
metastatic growths of a typical carcinoma of epiblastic or
hypoblastic origin have assumed definitely sarcomatous characters.

Possibly the careful studies now being made in connexion
with the melanotic tumours originating in connexion with the
skin may prove that this statement will need eventual revision.
Authorities are still so much divided as to the exact origin of
sundry tumours of sarcomatous type belonging to this group,
that it is not possible to make an absolute statement concerning
their origin. If, however, it can be proved that under certain

320 ON GROWTH AND OVERGROWTH

conditions tumours originating from the Malpighian layer of
the skin can in their growth lose their lepidic characters, i.e.
can become possessed of a definite stroma passing between the
individual cells, then we shall have to acknowledge that epiblastic
lepidomata can, in the course of their growth, revert to a more
undifferentiated sarcomatous type.

In one very interesting tumour of the prostate, examined by
me some years ago, I was for a time of opinion that I could
recognize such a transition. Sections taken from the prostate
itself showed beautifully the existence of a very typical carcinoma
almost scirrhous in appearance ; passing towards the bladder
the type became that of carcinoma simplex. The abundant,
apparently rapid, growth, forming a projection into the bladder
itself, was so wholly cellular as to resemble a round-cell sarcoma,
but more careful examination showed that even here the cells
were arranged in alveoli, although the stroma between the
individual groups was so delicate as to be traced with considerable
difficulty.

46. With mesothelial and endothelial lepidomata the same
is not always so ; the older or earlier portions of the tumour
may show distinct adenomatoid or carcinomatoid characters —
occasionally the whole tumour is typically adenomatous ; but
more rapidly growing portions are peculiarly liable to depart
so far from type, so peculiarly liable to take on the appearance
of embryonic connective tissue, that it becomes impossible,
basing our terminology upon histological appearances, to say
whether we are dealing with a carcinoma, or a sarcoma, or a
mixed growth — a carcinoma sarcomatodes, or a sarcoma car-
cinomatodes. More particularly the increased recognition of
the frequency of endothehomata and peritheliomata has forced
us to see the difficulties in our present mode of classification.
The perithelioma when developing characteristically, apparently
as an endothelioma of the perivascular lymphatics, may strongly
resemble an adenoma in the regular columnar arrangement of
its cells, and yet other parts of the same tumour may be absolutely
sarcomatous in type. And while the ordinary endothelioma,
such as one so commonly meets with forming tumours in con-
nexion with the membranes of the brain, is in general char-
acteristically sarcomatous in structure, areas are to be detected
here and there indicating its origin as a squamous proliferation

DEDIFFERENTIATION OF TUMOURS 321

of the lymphatic or blood-vascular endothelium. Without, I
trust, taking anything from the interest and value of his forth-
coming article upon this subject, I would here note that Dr.
P. G-. Woolley, Fellow in Pathology at McGill University, has
just completed a most elaborate and minute study of a tumour,
originating in the zona fascicularis of the suprarenal, in which
a similar transition from adenomatoid to purely sarcomatous
structure is to be followed without the possibility of doubt.1

Here, then, are tumours which, showing in the least aberrant
regions indications of origin from a lining membrane or lepidic
tissue, are apt to take on the appearance and structures more
characteristic of " pulp " tumours.

Now this difference in behaviour between the Epilepidomata
and Hypolepidomata on the one hand, and sundry of the Meso-
theliomata and Endotheliomata (Mesolepidomata) on the other,
is but consonant with embrvological observations and the broadest
biological principles. One great principle which we see con-
stantly in evidence is that those structures and properties which
are of oldest acquirement are those which are last to be lost ; it is
the most recent acquirement which tends to be the earliest to dis-
appear. We see this exemplified continually in connexion with
blastomas in general. The more rapid the growth — the more the
cells of a tumour depart from their normal mature environment
— the more do we observe that those features of the tumour cells
which are specific for one or other tissue tend to disappear.
In the most rapidly growing and most aberrant tumours the
individual cells afford us little or no clue to the tissue of origin.
It is the general arrangement of the cells that aids us in making
our diagnosis, and even the general arrangement is not so much
that peculiar to the fully grown tissue as that common to con-
nective tissues in general, or to glandular or lepidic tissues in
general. We recognize a reversion to an earlier, simpler, or,
as we express it, embryonic type. As I have pointed out else-
where, a distinction must be recognized between functional
and proliferative activity with loss of those features which
are directly associated with the performance of function.

I would now suggest that we carry the working of this principle
a little farther back. The first lining membranes to be differ-

1 [Trans. Assoc. Amer. Phys. xvii., 1902, 627, and Virch. Arch. 172,
1903, 301.]

322 ON GROWTH AND OVERGROWTH

entiated are the epiblast and hypoblast : their differentiation,
indeed, is one of the earliest events in developmental history ;
and this being the case we should expect — and we find — that
tissues, whether normal or neoplastic, derived in direct line
from these two layers, are singularly tenacious of their properties
as lining membranes, and so it is that Epilepidomata and Hypo-
lepidomata always show evidences of their lepidic nature. We
should not expect — and we do not find — that where this direct
line has been departed from, where, for example, hypoblast has
given off masses of cells to form mesoblast, and the epiblast
similar masses to form the neuroblast, that, in reversion, tissues
derived from mesoblast and neuroblast respectively should again
enter upon the lining membrane stage. Where in the process
of development an organ or part is formed by the cells of tissue
of a higher order assuming a less differentiated condition, and
from this lower state proceeding to develop along special lines,
we do not find that in reversion and degeneration that tissue
passes beyond the less differentiated stage, and then proceeds
to show characters of a primary more differentiated condition.
Thus it is that the mesenchymatous tumours and sarcomata in
general show no tendency to assume fining membrane or lepidic
characters, even though, without exception, all these tissues have
primarily arisen as derivatives from either epiblastic or hypo-
blastic lining membrane.

If occasionally in gliomata we find cysts lined with columnar
epithelium, this is not an example of such reversion to the more
primitive epiblastic characters of the glial tissue. Those who
have studied cases of this order have, without exception, ascribed
such conditions to inclusions of rests of embryonic tissue con-
taining portions of the Anlage of the central nervous canal.1

But now, coming to the mesoblast, we know that it is at a
later date in the history of the embryo that this becomes
differentiated into mesothelium and mesenchyme, and the
development of the mesothelium is in the direction of increased
specialization. Thus we should expect — and we find — that in
processes of a reversionary type these more lately acquired
characters are more capable of being lost, so that growths formed

1 [I have since pointed out (Principles of Pathology, vol. i. 1st edit. p. 564)
that in gliomas areas of necrosis with autolysis of the dead tissue come to form
cysts lined by a false epithelium. The lining cells become modified as a function
of their position, but present no basement membrane.]

MESOLEPIDOMAS : THEIR PROPERTIES 323

from organs of mesothelial origin are more liable to pass back,
and to assume characters approaching those of the primitive
mesenchyme and mesoblast, than are growths from hypoblastic
and epiblastic organs to revert to what we might term the morula
stage. And if it be correct to regard the endothelium as a still
later development from the mesenchyme, then we can understand
how it is that Endothelioma ta are peculiarly liable to take on
the characters of the primitive pulp tissue from which endothelium
became differentiated.

Israel * recognizes fully this same dependence of the characters
of the endotheliomata upon the embryogeny of the mother
tissues, for he remarks (the italics are mine) : " Diese (endotheliale)
Deckzellen haben sich nicht unter alien Umstanden in ihren
Eigenschaften von diejenigen ihrer Intercellularmasse bildenden
Stammesgenossen so weit entfernt, dass nicht auch sie gelegentlich
wieder befaliigt wurden, Intercellularmasse hervorzubringen,
deren Qualitat von den ererbten Eigenthumlichkeiten abhangig
ist. . . . Gelegenliche Vorkommnisse in den Endothelgeschwiil-
sten und auch bei gewissen entziindlichen Neubildungen zeigen,
dass die Fahigkeit Intercellularsubstanzen zu produciren, manchen
in Tumoren gewachsenen Endotheldescendenten, nicht unwieder-
bringlich verloren gegangen ist, wie den Epithelien. Das ist aber
auch der fur die Diagnose praktische bedeutsamste Unterschied
endothelialer Krebse gegenuber den epithelialen, und er macht
von alien die Ubergangsformen vom Eudotheliom zum Sarkom
verstandlich."

I am far from saying that this is the one and sufficing cause
why certain orders of the Mesolepidomata have this marked
tendency to assume more sarcomatous characters. I do not
think that this is everything ; nevertheless it does, I think,
materially aid us to understand why this peculiarity in the
progressive development of these tumours manifests itself ; it
is one factor. Indeed, it is not all the Mesolepidomata which
present these same tendencies. To give one example : we
never, to my knowledge, find that carcinoma ta of the uterus,
either in their more rapidly growing parts or in their metastases,
show other than well-developed cancerous structure. In fact,
there is singularly little distinction to be drawn between the
characters of cancers originating from the uterine mucosa and

1 Israel, 0., loc. cit. p. 668.

324 ON GROWTH AND OVERGROWTH

those of epiblastic and hypoblastic origin. The same is true
with regard to the tumours of the Fallopian tube. And, so far
as I see, we have to lay down also that the high grade of specializa-
tion and differentiation of the lining cells of a membrane is
accompanied, pari passu, by a lack of capacity to undergo
reversion to the simplest type. While, for example, the glial
cells are apt to undergo rapid proliferation and form tumours
of a simple cellular type, true Neuromata are extraordinarily
rare, and in general we may lay down that the high degree of
differentiation which has been reached by the neurone has taken
from it, to a very large extent, the power of proliferation. This
is, it is true, an extreme example. The cells of the uterine
mucosa are not nearly so highly differentiated ; they are, how-
ever, distinctly more specialized than are endothelia and the
cells lining the serous cavities. And, I would add, more
differentiated than, for example, the cells of the cortex of the
suprarenal, which but attain an arrangement in solid columns
resembling that seen during the process of development of certain
other glandular organs.

Thus it may be that simplicity of type is one factor determin-
ing why certain tumours revert from a more carcinomatous to
a more sarcomatous type. I mention this as a possible factor,
and would not dwell too strongly upon it ; rather I would admit,
and that willingly, that here we are in the region of analogy and
hypothesis. Just as we are unable to explain why in one case,
in a given organ, we have produced, for example, a relatively
benign adenoma, in another, in that same organ, an actively
growing and malignant carcinoma ; so it may be that full con-
sideration will show that here also we have to recognize, without
possibility of explanation, that, in connexion with tumours of
one organ, the lepidic properties are retained with remarkable
persistence, while in tumours of another organ, derived from the
same cell layer, these same properties are lost with comparative
ease. But it seems to me not without its use to point out
tentatively a possible factor in the development of the different
properties of different tumours. As a group it has to be admitted
that mesothelial and endothelial tumours exhibit this tendency
to assume a more sarcomatous appearance.

We are thus justified in separating Lepidomata into two
main divisions — the primary lepidomata, wherein we include

TRANSITIONAL LEPIDOMAS 325

the tumours derived from tissues of direct descent from the
epiblast and hypoblast, and the secondary or transitional lepido-
mata, which include tissues of indirect descent from the same,
and which may as a consequence show what I may term transi-
tional characters. Making this division, I leave it open as to
whether we speak of the tumours of the first order only as the
true carcinomata, and refer to those of the second order as at
most carcinomatoid, or speak of all tumours of what I may now
without confusion refer to as having lepidic characters, as being
carcinomata. My preference, as I have already stated, is for
employing this term purely in a histological sense.

Recognizing that many years have passed since I was engaged
in the active study of embryology, I have not ventured to publish
this paper without consulting those most capable of pronouncing
authoritatively upon the embryological problems here involved ;
and I would here express my sincere thanks and sense of deep
obligation to Professor Charles Sedgwick Minot of Harvard,
and Professor Carl Huber of Ann Arbor, for their most kind
criticism and suggestions in connexion with the views here
elaborated. I very gladly acknowledge also that I owe more
particularly to a correspondence with Professor Huber, now
some months ago, my recognition of the importance, from a
pathological point of view, of distinguishing between the meso-
thelial and mesenchymatous derivatives of the mesoblast.

CHAPTER IV

SYNCYTIOMA MALIGNUM : ITS BEAKING UPON THE ESSENTIAL

NATURE OF MALIGNANCY1

(1902)

Asked most courteously to contribute from a distance some-
thing to the debate upon the nature of cancer, and recognizing
the difficulties under which both my audience and myself must
labour when I accept, it has seemed to me better not to attempt
an academic essay covering the whole territory of the possible or
presumed relationship of parasites to malignant growths, but to
consider more in detail one small portion of the vast field before
us, in the hope that this will not already have been traversed by
other participators in the debate : in the hope, further, that the
deductions which may logically be drawn from the study of a
single form of tumour may throw light upon the essential nature
of tumours in general. And in advance let me say that I am
well acquainted with the ancient Greek parable of a certain
man who, having a house to sell, circumambulated the city
bearing in his hand a brick as sample. I trust that when my few
words have been delivered the brick will not be thrown in my face.
Until recently we have been in the dark as to the relationship
between the embryo and the maternal tissues in the womb ;
now certain important points are well established. The observa-
tions of H. Peters, of Vienna, upon the earliest human ovum
which has thus far been discovered in utero — an ovum of about
the end of the first week — have proved in a manner admitting
of no dispute (when compared with what we know concerning
mammalian ova in general) that even at this early date the cells

1 A contribution to a discussion, held by the Chelsea Clinical Society, upon
Cancer. Reprinted from the Clinical Journal, June 18, 1902.

326

THE TROPHOBLAST 327

of the ovum have a marked influence upon the uterine mucosa
at the spot where the ovum becomes arrested. The epithelial
cells of that mucosa are so acted upon that they become dis-
solved and disappear, and the ovum from being on the surface
sinks down into a depression, becomes imbedded in the uterine
wall, so that now it lies in close apposition to the capillaries
underlying the epithelium. These observations have been largely
confirmed by my late colleague, Dr. J. C. Webster, now of the
University of Chicago.

This is but the first step in the formation of the placenta.
The outer layer or layers of the epiblast of the rapidly growing
embryo, termed by Hubrecht the Trophoblast, continues this
process of erosion and infiltration in a most remarkable way.
This trophoblast literally eats into the maternal tissue opposed
to it, forming at first solid cellular processes, purely epiblastic ;
later, as the chorionic mesoblast and the allantois with its vessels
become developed, these processes gain a mesoblastic and vascular
core ; so are evolved the characteristic chorionic villi. Both the
more primitive solid processes and the villi, through this phago-
cytic, or more truly extra-cellular-digestive, action of the tropho-
blast, make their way into the maternal uterine veins and venous
sinuses, whereby the cells of the embryo come to lie within the
maternal blood-spaces and gain their nourishment, and nourish-
ment for the embryo in general, directly from the maternal
blood.

There was for some years violent debate, more especially in
Germany, as to the nature of the trophoblast, whether it was of
foetal or maternal origin. This culminated in an important
debate between anatomists, zoologists, gynaecologists, and
obstetricians at the 1897 meeting of the German Naturforscher-
versammlung. The most that the most violent opponents of
the foetal view can now bring forward is that not all the cells
considered by some as trophoblastic, or syncytial, are of foetal
origin. Some, they say, are maternal. Nay, more, in the course
of a most painstaking and minute study of the development of
the rabbit's placenta, my colleague, Dr. Chipman,1 has shown
in a conclusive manner that these trophoblastic or syncytial

1 Through the liberality of the Governors of the hospital, Dr. Chipman 's
Edinburgh thesis, with its abundant photographic illustrations of this and
kindred points, [was] published in the series of Studies from the Royal Victoria
Hospital. Montreal [No. 2, 1902].

328 ON GROWTH AND OVERGROWTH

cells, when they gain entrance into the maternal blood sinuses,
passing along the walls, digest and replace the endothelium of
those sinuses, so that eventually there is developed a most
complicated system of passages lined by cells of foetal origin and
containing maternal blood ; or almost, in the words of Matthias
Duval, the placenta is to be regarded as "a maternal haemorrhage
encysted by foetal tissue." By this remarkable " adaptation "
the foetus gains nourishment from the maternal blood.

I have described this at some little detail because I wish to
impress upon you that here, as a normal or physiological con-
dition, we have passing those identical processes of invasion and
infiltration which we have been accustomed to regard as peculiar
to, nay, as the essential feature of malignant growth. The only
difference between what is here seen and malignancv is one of
degree, or perhaps, more correctly, of orderliness. This in-
filtration is due to the inherent properties of the syncytial cells.
Nor would this appear to be by any means the only case of
antagonistic " phagocytic ' cellular activity. Metchnikofr has
of late been engaged in collecting evidence to prove that such
activity is fairly general within the organism, and is the basis
of senile changes.

Under ordinary conditions this syncytial invasion is strictly
limited ; it only affects a definite area of the uterine wall, and
with parturition and the separation of the placenta the whole
area of invasion is separated off and discharged. In other words,
to explain these normal phenomena (if the phrase be permissible),
we have to recognize the existence and interaction of two forces
— on the part of the foetal cells of invasive properties, on the part
of the maternal tissues of protective properties.

Having laid this down, I may now approach my main point.
Occasionally — and apparently not so very occasionally, for
Marchand collected from the literature some sixty-five cases
recorded between the beginning of 1895 and the middle of 1898 —
these syncytial cells, after normal labour, are found to take on
aberrant growth, to trespass beyond the placental site, to infiltrate
extensively the uterine walls, the vaginal tissues, the pelvic
tissues in general, and, what is more, to form secondary growths
in distant organs, the lungs, the spleen, the brain, and so on.
These svncvtial cells are large and characteristic ; there is
nothing quite like them. We have, that is to say, as nearly

SYNCYTIOMA MALIGNUM 329

perfect histological evidence as it is possible to possess that at
least one group of primary malignant tumours of the uterus — a
group, I may add, presenting " exquisite " or " eminent "
malignant properties — consists of the aberrant growth of foetal
cells within the maternal organism.

This view of the foetal nature of the cells of the deciduoma
malignum was, I need scarce say, violently opposed for many
years by an important section of pathologists ; but now Mar-
chand, whose opinion cannot be passed over, has been forced
to accept the view : in short, no other view is possible at the
present time. The typical deciduoma is a syncytioma or chorio-
epithelioma malignum. I greatly doubt whether any committee
of the London Obstetrical Society could be found in these latter
days to report, as one reported only five years ago, that a group
of specimens of this form of tumour was of connective tissue
origin, and so sarcomatous.

Can we explain this most remarkable form of neoplasm by
any parasitic theory ? To do this it is obvious that we must start
from the assumption that micro-parasites, in some way or other,
exalt the already existing invasive properties of the syncytial
cells, so that now they invade and infiltrate to an excessive
extent. That must be the starting-point.

For myself I am perfectly prepared to accept this much,
namely, that parasites and their products are capable of stimulat-
ing cell growth. I see that in his recent Milroy Lectures Mr. C.
Powell White discovers a poorly defended spot in my argument
published last year,1 in that there I make this same admission,
while simultaneously I lay down that the vegetative and pro-
liferative activities of the cell are inherent, the functional activi-
ties set up by stimulation from without. Upon first consideration
the objection appears valid. Further consideration, I think,
shows that it does not hold. We can imagine stimuli leading
to building-up — anabolic — processes in the cell, processes which,
under ordinary conditions, are preparatory to functional and
katabolic processes, which, in the absence of circumstances
favouring katabolism, favour cell proliferation. Now, in con-
nexion with this very form of tumour we are considering, it is
not a little interesting to note that the growth does not occur in
the placenta during the period of pregnancy, during the period,

1 [Vide p. 288.]

330 ON GROWTH AND OVERGROWTH

that is, when the syncytial cells are actively performing their
functions of absorption and excretion, but afterwards, when
these same cells, not having been shed, are left in the condition
of gaining nourishment — of being stimulated from without also —
and of having no opportunity to perform their normal functions.
Under such conditions microbic products might favour overgrowth.
It will be seen that the two statements are not contradictory.

But is it necessary to call in micro-parasites to explain the
aberrant growth ? There, I confess, looking at the matter in
the abstract, I feel more than doubtful. For, in the first place,
it will be seen that this assumption — and remember that it is
still only an assumption, and that, consequently, one is permitted
to discuss it in the abstract — this assumption demands that
microbes of one or other order have the peculiar power of so
acting upon the cells of one organism — the embryo or foetus —
that they invade the tissues of another organism — the mother.
Our knowledge of the phenomena of transplantation and im-
plantation shows that it is not necessary to call in such external
agencies. That micro-parasites should have this power is, to
say the least, unexpected. At most we must admit that it is
remotely possible.

In the second place, one of the favourite arguments of those
who hold parasitic theories is the similarity or analogy between
the local growth and metastases of malignant neoplasms, and, in
the case of bacterial infection, the primary focus of infection, its
local extension and metastatic, granulomatous, or abscess develop-
ments. Undoubtedly the parallelism is close, so close that
we may admit parasitism as the active agent in both conditions.
Indeed, these syncytiomas demonstrate most convincingly why
this is so. Bacteria are, we know, the causative parasites in
the one : what these syncytiomas show, and show without
possibility of denial, is that the trophoblastic cells — the tumour
cells — are the parasites in the other. They are cells of foreign
origin ; it is they that invade, they that gain entrance into the
blood-stream, that get carried to remoter parts of the organism,
and there set up degenerative and reactive processes, and so
bring about a condition of affairs resembling what is seen when
bacteria are the invading agents. The tumour cells are the
parasites ; and if we admit this — and we must admit it — then the
analogy becomes absolutely valueless as an argument in favour

THE PARASITIC HYPOTHESIS 331

of the micro-parasitic origin of malignant neoplasms. To urge
that in order to induce the aberrant growth there must be para-
sites within the parasites, is to pile Ossa on Pelion — it is super-
erogatory.1

Nevertheless, these considerations, while demolishing the
argument based upon analogy, do not wholly do away with the
possibility that micro-parasites play some part in the production
of these and other malignant growths. There is yet another
method of approaching the subject.

As I have already laid down in discussing the normal growth
of the syncytium, we have to see that there are two factors
constantly opposing each other — the invasive powers of the
syncytial cells and the self-protective powers of the maternal
tissues. This is true, also, of the abnormal growth of the same.
It is possible, therefore, that the aberrant growth of the syncy-
tioma may depend not so much upon an exaltation of the invasive
powers of the component cells as upon a lowering of the self-
protective powers of the surrounding maternal tissues. Indeed,
the observations of Beatson and others, even if therapeutically
they have not been crowned with complete success, can only
be regarded as demonstrating that this second factor has to be
taken into account. Alteration of general metabolism following
upon removal of the ovaries does at times arrest malignant tumour
growth elsewhere. At the present time I feel even more strongly
than I expressed myself a year ago that it is along the lines of
research into the exaltation of the protective mechanism of the
organism, and into the diminution of the proliferative powers of
the tumour cells, that our energies should now be directed, and
coincidently I note that the last twelve months have not yielded
any more promise than showed itself then of distinct help to be
gained from this weary and unsatisfied hunt after specific cancer
parasites.

I do not take the stand of denying the existence of such. In
this special connexion we can observe a parallelism between the
virulence of the microbe in bacterial disease and the protective
powers of the organism on the one hand, the invasive powers of
the cancer cells and the protective powers of the organism on

1 It is worthy of note, further, that in the known cases where we have
intracellular parasitism — e.g. coccidiosis — we have no sign of the cells them-
selves taking on these parasitic functions.

332 ON GROWTH AND OVERGROWTH

the other. But have we any sound grounds for assuming that the
presumed cancer parasite acts by stimulating and augmenting
the invasive powers of the cells, or by doing the reverse, namely,
by lowering the vitality of the tissue about to be infiltrated ?

Looked at thus closely, it will be seen that this conception
of parasite acting within parasite leads to marvellous complica-
tions. I may be wrong, but I think that I have noted time and
again that when we determine fundamental laws and funda-
mental phenomena they are wonderful in their simplicity, so
that now when I find that a conception followed up leads us
deeper and deeper into the mire, I shrewdly suspect that the
conception is wrong. So it is with the parasitic theory of malig-
nancy as now held by so many. It may be right, some one may
discover the germ or germs, but neither on theoretical grounds
nor on the results of practical observations can I base myself
with any satisfaction. Certainly Gaylord's results upon this
side of the Atlantic do not carry conviction, and neither he nor
Plimmer, so it seems to me, have proved that their bodies are
any other than degenerative products, the abundant presence
of numerous forms of which in tumours was first demonstrated
by Fabre-Domergue, and later and most fully by Pianese.1

My inclinations at present, it will thus be seen, in the absence
of more definite discoveries, lead me to doubt very gravely the
permanent presence of micro-parasites in malignant growths ;
nor do I think that a study of the deciduoma malignum can lead
to any other than a firm belief that alterations in cell properties
and assumption of the parasitic habit by cells, I will not say, is
at the bottom of malignancy — for we have still to determine
what initiates the alteration in property — but must be recognized
and must be taken as the basis for future advance.

Thus, to sum up what, I think, may be legitimately deduced
from a study of the syncytioma malignum :

1. There exists one form of tumour of highly malignant type
in which the infiltrating cells are not those of the organism itself,
but are derived from another organism.

2. The infiltrative and invasive properties of these cells are
not a new acquirement, but are an exaltation of properties
normally possessed by them ; or, more exactly, under normal

1 Such bodies, described as Blastomycetes, hare, I should add, been figured
by Rossi-Doria as found by him in two cases of deciduoma malignum.

THE CELLS AS PARASITES 333

conditions we observe that there is an interaction of two forces :
the invasive properties of these cells, the protective properties of
the surrounding maternal tissues, by which interaction the extent
of invasion of the cells is strictly limited to the placental site.

3. The development of the syncytioma malignum must there-
fore be attributed to either an increase in the invasive properties
of the syncytial cells, or a lessened resistance on the part of the
maternal tissues, or a combination of the two.

4. If micro-parasites play any part in the production of the
tumour, that must be either by exalting the infiltrative powers
of the one or by lowering the resistant powers of the other.

5. Proof positive of the existence of such specific micro-para-
sites is still wanting, and in its absence it is difficult to conceive
how specific micro-parasites should bring about these results.

6. Admitting that micro-parasites and their products are cap-
able of inducing cell proliferation, we must also admit that other
external agencies have the same capacity, so that micro-para-
sites, at most, may be one of the causes of aberrant cell growth.

7. The prevalent conception of cancer parasites, as existing
within the cancer cells, involves the idea of malignant growths
being the product of parasites acting within parasites ; for, as
shown by this study of the syncytioma, the tumour cells them-
selves are essentially parasitic in the organism.

In short, tracing the present popular idea to its legitimate
conclusion, the most that can be said for it is, as the association
of parasitic ideas may already have reminded some of my
audience, that it is in harmony with the great generalization of
the plagiaristic poet, that —

Great fleas have little fleas upon their backs to bite 'em,
And little fleas have lesser fleas; and so ad infinitum.

Is this generalization also to be applied to explain the action
of pathogenic bacteria ? Have they their stimulant parasites ?

I am willing to see, in the long-continued action of micro-
parasites, a process which may, like other modes of stimulation,
initiate aberrant and neoplastic cell growth, but beyond this
point I cannot advance. We seem to be asked to contemplate
a most extraordinary state of affairs, so that I cannot accept
the view that micro-parasites are in constant action in malignant
growths, and are essential for the extension of the same.

CHAPTER V

UNIPOTENTIALITY, PLURIPOTENTIALITY, AND TOTIPOTENTIALITY
OF CELLS : A NOTE UPON THE CLASSIFICATION OF TUMOUES 1

(1907)

A few years ago I published a paper upon the classification of
tumours based upon the embryogeny of the different tissues.2
That classification, I confess, was somewhat elaborate in appear-
ance, although simple in principle, based upon the fact that each
primitive cell layer gives rise to tissues of two orders, to denote
which I was compelled to introduce a new set of terms. I did not
like to do this, nor was it my idea that these terms should in any
way attempt to replace in general practice those honoured by
long use. Nevertheless, mere length of use has brought about
such vagueness in the employment of the latter that for scientific
and exact purpose they have very largely lost their primitive
value. These terms notwithstanding, it is some satisfaction to
see that that classification is coming to be accepted by writers
of systemic works on pathology as the most rational so far pro-
posed. To-day I wish to call attention to what I regard as a
further advance in our conception of the relationship of the
different forms of tumour which permits us to make in some
respects a still more exact classification.

It may be remembered that in my older classification I divided
neoplasms into two great groups of the teratomas (the tumours
due to the development within the body of one individual of
tissues and parts belonging to another individual) and the

1 A paper read before the Pathological Section of the Canadian Medical
Association, Montreal, September 1907.

2 Journal of Pathology and Bacteriology, iv., 1902, 243 [pp. 305 et seq. of
this volume].

334

" MIXED TUMOUES " : EMBRYOMAS 335

blastomas (or tumours due to the aberrant growth in the organism
of one individual of cells and tissues of that individual himself).
So far as it went this division was fairly satisfactory : it left,
however, a certain group of tumours in the debatable ground.
I refer more particularly to that group which has been studied
so ably by Wilms, the group to which he has given the name of
" Mischgeschwulste," or mixed tumours, and to that section
of the Mischgeschwulste to which the name embryoma has been
applied. The type example of this growth is to be found in
connexion with the kidneys of children and young individuals.
Here there develops a cell-mass formed of a mixture of several
distinct elements. There may be striated and plain muscle,
gland tubules of renal type, cells of connective tissue type, of a
sarcomatous type, and, very rarely, bone cells or cartilage cells
have been encountered. Wilms has pointed out very clearly
the mode of origin of these tumours. The Anlage of the glandular
cortical portion of the permanent kidney is of mesoblastic origin,
and at a very early stage in the embryo this " nephrotome "
is cut off or differentiated from the myotomes, or cell-masses
lying on either side of the median line which will eventually
give origin to the main musculature of the trunk axis and to the
bone and cartilage of the vertebral column. Now, there have
been cells at a yet earlier stage which have been the ancestors
both of the future muscle ceils and of the glandular portion of
the kidney cells, and these ancestral cells have, by progressive
multiplication, given origin to both. We can best comprehend
the tumours in question by supposing that in the course of
development one or more of these ancestral cells has become
snared off, or has come to occupy an unusual position with regard
to its fellows, so that its development cannot follow the normal
line. We may suppose that such a cell becomes carried along
into the developing future kidney and that in time that cell
begins to grow, and in growing retains the properties and the
potentialities given to it at its origin. Such a " cell-rest," to
apply Cohnheim's term, in its proliferation will tend, therefore,
to produce both muscle cells and renal epithelium, and the tumour
derived from it will of necessity be of a mixed type.

We encounter mixed tumours of a similar nature in several
parts of the body — in the parotid, in connexion with the genital
tract, and so on — and it is along these lines that they gain their

336 ON GROWTH AND OVERGROWTH

simplest explanation. Where are we to place these tumours ?
The cells that give origin to them have not the power, even under
the most favourable circumstances, to reproduce all the different
forms of tissue ; they have not the potentiality to form, even in
the most favourable circumstances, the complete individual.
And, on the other hand, there are no signs in these tumours of
a combination of tissues derived from all three germinal layers.
There may, indeed, be only the representative of one layer, the
mesoblast. But we have in this order of tumour two or more
distinct forms of tissue, in addition to the connective tissue
stroma. They are not teratomas, they are not typical blastomas.

It is here that the terminology introduced by the embryologist
Barfurth, in his Handbuch der Embryologie, appears to me to
be of singularly high value and to afford the basis of an adequate
classification. Barfurth has pointed out that, following the
development of the typical ovum, one recognizes a first stage in
which each individual blastomere or product of the cell division
of that ovum may be shaken apart from its fellows, as has been
done by Driesch, Roux of Breslau, E. B. Wilson, Morgan, Jacques
Loeb, and many others, and each cell so separated has the power
of giving origin to a complete, even if somewhat dwarfed, in-
dividual. Such cells he terms totipotential.

Following upon this stage, he notices that as the cells of the
dividing ovum develop into the different germinal layers these
cells no longer possess the individual power of giving origin to
the complete individual. But, obviously, in this next stage the
individual cells of the primitive streak region in differentiating
into epiblast, mesoblast, or hypoblast, according to their position
become the ancestors not of a single tissue, but of various groups
of tissues. If epiblastic in position, they give origin to the
cutaneous, glandular, and nervous elements ; even if at an early
stage they give off cells which become mesoblastic, and so origin-
ate members of the group of what we term familiarly the con-
nective elements. Nor is this capacity to give origin to more
than one form of tissue confined only to the familiar three-layer
period. The cells may be regarded as retaining this capacity
throughout the whole of what Ballantyne has termed the germinal
period of the embryo, until, that is, they become so far differen-
tiated by division and relative site that they give origin to the
matrices and anlagen of the different organs of the body. We

BARFURTH'S ORDERS OF CELLS 337

have already given an example of this type of cell in what we
said regarding the primitive cells of the myotomes. Such cells
Barfurth terms pluripotential.

With progressive development, once the anlagen of the
different tissues have been laid down the cells composing these
become more and more differentiated, and now those cells have
the capacity to produce one type of tissue only : they become
unipotential.

Consideration will show that tumours group themselves into
three great groups exactly corresponding to those three groups
of cells laid down by Barfurth. As regards tumours derived
from the parasitic growth within the body of one individual, of
the products of a separate ovum — twin teratomas — and " filial "
teratomas derived from aberrant growth of cells of the germinal
type, the products of the reproductive organs of the individual,
the cells from which these tumours originate are most simply
regarded as being totipotential. It is a far less simple conception
to suppose that such tumours are derived from the segregation
of a group of cells derived from all three primitive layers. That
conception has now been wholly given up.

Similarly we may regard the whole group of mixed tumours of
the Wilms type, the embryomas, as derived from pluripotential
cells. If we speak of the first class as teratomas, we may con-
veniently refer to these as terato-blastomas, to differentiate them
from ordinary tumours formed of one specific type of constituent,
with a framework or stroma of indifferent connective tissue and
blood-vessels. The latter we must regard as derived from
unipotential cells. These we may continue to speak of as the
blastomas.

This method of regarding tumours appears to me both simple
and straightforward and helpful in dispersing the doubt and
muddle that has existed regarding these transitional types and
their relationship.

We may thus classify new growths as follows : —

I. Teratomas. — The products of the growth within one in-
dividual of what is primarily or potentially another individual.

(1) Twin or Germinal Teratomas. — Due to the growth within
the tissues of the one individual (the host) of the products of
another fertilized ovum, coeval with the ovum giving origin
to the host.

338 ON GROWTH AND OVERGROWTH

Example : Foetal inclusions.

(2) Filial Teratomas. — Due to the growth within the tissues
of the one individual of the products of a totipotential cell derived
from that individual.

(a) Blastomeric Teratoma, due to the segregation of a (toti-

potential) blastomere in the early embryonic or
" germinal " period (Ballantyne) and subsequent
growth.
Examples : Three-layered epignathus and congenital sacral
teratoma.

(b) Germ-cell Teratoma, due to independent, aberrant growth

of a totipotential germ cell at any period after the germ
cells, as such, become segregated from the somatic cells
of the embryo, and before the nuclear reduction pro-
cesses take place which lead to the production of
oocytes and spermatozoa.
Examples : Ovarian and testicular dermoids .

II. Terato-blastomas. — The products of growth of a segregated
pluripotential cell of the individual.

Examples : Two-layered and simpler forms of epignathus
and " congenital sacral teratoma " ; also " Mischgeschwulste "
(embryomas of Wilms) of kidney, parotid, etc.

III. Blastomas. — The products of growth of a segregated
unipotential cell.

(1) Teratogenous, due to continued aberrant growth of uni-
potential cells of one individual (the embryo, or, rarely, of a
teratoma) within the tissues of the other (the parent).

Examples : Placental moles and chorio-epithelioma malignum.
Carcinomas, etc., originating in ovarian dermoids.

(2) Common, derived from unipotential cells of the host.
Examples : The ordinary neoplasms formed of one type of

tissue — fibroma, adenoma, etc., sarcoma, endothelioma, car-
cinoma, etc.

It is these blastomas that can be subdivided according to
my previous classification into the lepidic or lining membrane
tumours (lepidomas) and the hylic or matricial tumours (hylomas),
whether of epiblastic, hypoblastic, or mesoblastic (mesothelial
and mesenchymatous) development.

IV. Blastomatoid Growths. — There is yet a fourth class of new
growths so called, not coming into any of the above categories,

THE CLASSIFICATION OF TUMOURS 339

which remains to be noted. I refer to those conditions in which
not individual unipotential cells of a tissue but the whole tissue
undergoes a local overgrowth, resulting in the formation of
tumours that are (1) ill-defined, (2) tend to be diffuse or multiple,
and (3) to retain in their structure the ordinary tissue relation-
ships. These are the conditions of idiopathic " Riesenwuchs "
of some authorities. They are commoner and more varied than,
I think, is usually imagined. Examples of such conditions are :
fibromatosis (multiple fibroids), lipomatosis, chondromatosis
(multiple ecchondroses) and osteomatosis (multiple exostoses),
myelomatosis (myeloma multiplex), gliomatosis, lymphoma-
tosis (including lymphatic and myelogenous leukaemia), etc.
Whether the termination " omatosis " should be applied to these
conditions is debatable. It is that which, apart from this
recognition of the distinction here drawn between these con-
ditions and the blastomas proper, has hitherto been employed
to designate the members of this class, and this so generally
that to introduce another terminology seems a work of superero-
gation. Such growths form a link between the tumours proper
and the hypertrophies.

CHAPTER VI

ON THE RELATIONSHIP BETWEEN TUMOURS PROPER (BLASTOMAS)

AND HYPERBLASTOSIS x

(1913)

To-day more fully than ever before it is realized that to com-
prehend the abnormal we must first understand the normal.
Thus it is that modern research into the etiology of cancer is
more and more occupying itself with inquiries into the pheno-
mena of normal growth. The results already gained from these
inquiries render it timely to reconsider a particular form of
overgrowth, to which German authorities have given the name
" Riesenwuchs." This is so commonly regarded as blastomatous,
as belonging to the tumours proper, that neither in English nor,
to my knowledge, in French has an adequate term been so far
afforded. Merely to translate the German term and speak of
" giant growth " conveys no well-defined meaning to the English
mind. C. P. White, it is true, has labelled it " progressive
hypertrophy," but this is a description rather than a name :
I, too, have spoken of it as " blastomatoid," but the expression
is adjectival and not substantive. And yet, as I hope to show,
this type of tissue overgrowth has properties so characteristic
as to separate it sharply from the true blastomas, to necessitate
its recognition as an order apart, and to demand a precise name
whereby to ensure that recognition. Nay, I would say further
that it is a notable aid to our understanding of the etiology of
malignant growths to make this recognition.

Here, to emphasize the distinction, it must be pointed out
that judged by the many definitions that have been afforded of

1 Communicated to the Pathological Section of the Seventeenth Inter-
national Medical Congress, London, August 1913.

340

" GIANT GROWTH " OF TISSUES 341

a " true " or autonomous tumour our stereotyped conception
of such is that it originates as a circumscribed overgrowth of cell
elements, not exercising any function of service to the body, or
at least becoming separate from the normal tissues in its physio-
logical and functional relationships. Even if we cannot with
Cohnheim regard it as derived always from a matrix of super-
abundant or erratic deposit of embryonic elements, we all, I
think, are accustomed to accept Ribbert's view that it is self-
confined, dependent upon the organism for its nourishment, but
otherwise largely if not quite independent. That is our stereo-
typed mental picture of what constitutes the tumour proper.

Let me approach my subject from what may appear to be a
novel, but what I believe to be the correct, standpoint.

The researches upon the functions of ductless glands pursued
with increasing activity during the last twenty years have
demonstrated in a wholly unexpected manner that sundry of
these glands have a remarkable influence over the growth of
particular tissues. This is now so well recognized that it is
needless for me here to give a detailed statement. Briefly, the
outstanding results are these :

1. That specific ductless glands and their internal secretions
influence the growth, not of all tissues equally, but of particular
tissues. Thus the experimental removal of the whole of the
anterior portion of the pituitary is associated with defective
growth of the bones, whereas, per contra, excessive development
of the anterior portion of the same gland is associated with
excessive growth of bone. Overgrowth of the adrenal cortex
is more particularly associated with premature development
and overgrowth of the organs of generation. Hypoplasia of the
thyroid is associated more particularly with overgrowth of the
subcutaneous connective tissue ; hypopituitarism with localized
or generalized adiposity along with genital deficiency.

2. Such overgrowth or arrest of growth of particular tissues
may have associated with it a coincident overgrowth of asso-
ciated tissues and parts. In osseous giantism associated with
excessive development of the anterior portion of the pituitary
we find coincident increased development of surrounding tissues ;
along with the premature development of the essential organs
of generation seen to accompany benign adenomatous tumours
of the adrenal cortex we note an over-development of the second-

342 ON GROWTH AND OVERGROWTH

ary organs of generation, as also of the muscular system. But
this associated growth must be regarded as secondary and
co-ordinate. In each of these cases we are struck by the fact
that one particular tissue shows excessive growth, whereas other
associated tissues, while presenting increased growth, have that
growth proportioned rather than over-proportioned to the
excessive development of the particular tissue.

3. While recognizing thus that excess or defect of particular
internal secretions exercises this specific action on particular
tissues, it is noteworthy that the particular tissues do not present
a universal hyperplasia or hypoplasia throughout all areas of
their distribution. In acromegaly, for example, it is the bones of
the face, hands, and feet that are more particularly involved ;
in pituitary giantism the bones of the limbs show excessive growth
rather than those of the trunk. Let it be admitted that mechani-
cal and other possibly deeper reasons exist for this regional
overgrowth of one or other tissue. Let us even admit that, as
has been suggested, in acromegaly it is a plethoric blood supply,
or an activity of the blood-forming organs, in short a primary
haemic change that induces bony overgrowth ; it still remains
that under the influence of altered internal secretions the growth
of one or other tissue is seen to lack proportion as regards the
relative regional development of that tissue, just as it lacks pro-
portion as between that tissue and the other tissues of the body.

From this group of cases we pass to another so similar in basal
properties that we must, I think, conclude that it is of the same
type, the group, namely, to which it has been customary to limit
the name " Riesenwuchs." Of this the most typical example
is adiposis and the allied conditions. Even where adiposis is
generalized every pathologist is forced to recognize that there
is an individual variation in the laying down of the fatty tissue.
Some cases, for instance, presenting a large panniculus adiposus
exhibit but a meagre deposit of fat in the mesenteries and omen-
tum ; and vice versa. We encounter, however, striking examples
of local overgrowth of the fatty tissue. I need but recall the
symmetrical lipomatosis seen more often in males, involving it
may be especially the neck, the submaxillary, or parotid region,
but in other cases seen in the mammary region, the perineum
and scrotum, or the inguinal region. With these must be noted
the characteristic distribution of the fatty overgrowths in

HORMONES AND OVERGROWTH 343

Dercum's disease (adiposis dolorosa). While these may be
nodular in type, more frequently the deposits while localized
are diffuse over the supraclavicular, inframammary, and lower
scapular regions. How regional is this distribution is frequently
noticeable in the extremities when, to quote Sir Dyce Duckworth,
" the hands appear to come out as from a cuff, and the foot from
a pantaloon." To the same order belongs the remarkable group
of perirenal, retroperitoneal, and mesenteric lipomas, so called.
These are one and all overgrowths of the fatty tissues normally
present in these regions. They are not, let me emphasize,
blastomas proper. They are diffuse hyperplasias which only
from their extreme extent give the impression of being distinct
tumours, but if carefully examined they are seen to respect the
boundaries of the normal tissue, and in fact to pass imperceptibly
into the normal fatty tissue around, without any sign of limiting
capsule. They cannot be regarded as autonomous, independent
developments ; they do not come within the accepted definitions
of tumours proper, or blastomas. They are conditions of hyper-
blastosis, that is to say, of the state of hyperplasia of an individual
tissue.

As I say, these localized regional overgrowths of fatty tissue
constitute the type example of " Riesenwuchs," but what is of
particular interest for the development of my thesis is that as a
class they appear to be due, not to any local irritation, but to
internal secretory disturbances, to some lack of equilibrium
between the internal secretions, resulting, as in our previous group
of cases, not necessarily in a generalized but in a regional over-
growth of this particular tissue. Notably in the case of Dercum's
disease, almost every case which so far has come to autopsy has
been characterized by thyroid or pituitary changes or both,
while conversely, if I may so express it, we have the authority
of one whom I may term the leading authority on obese states
and their treatment-^Chune Fletcher — that administration of
thyroid extract is the one method of treating Dercum's disease
that yields favourable results.

Passing to the other members of this group it is true that so
far we possess no distinct evidence that they too are associated
with internal secretory or metabolic disturbances. Anatomically,
however, they present the same general characteristics, and this
in so striking a manner that we are forced to consider them as

344 ON GROWTH AND OVERGROWTH

belonging to the same class. Let me enumerate rapidly the
more important members, only treating with somewhat more
detail conditions which, while not generally included as coming
under this category, not only, I hold, are rightly so included, but
find their proper place and proper relationships when so included.

One prominent and characteristic sub-group is in association
with the nervous system. In this belongs a most striking form
of " Riesenwuchs," which until recently has been regarded as a
fibromatosis. I refer to that condition of multiple subcutaneous
and perineural growths to which so many names have been
given : molluscum fibrosum, multiple neuro-fibromas, neuro-
fibromatosis, etc. The observations of Durante, Kohn, Bard,
and Verocay taken together demonstrate convincingly, I think,
that these growths are hyperplastic developments of the cells
of the sheaths of Schwann, of cells, that is, of neuroblastic origin.
These overgrowths beginning often in early fife, and developing
slowly over long years, respect the normal boundaries of the nerve
sheaths, merge imperceptibly at either pole, without defining
capsule, into the tissues of the nerve along which they have
developed, and in every respect conform in their properties with
the multiple fatty growths already enumerated. Closely allied
embryogenetically is the condition of gliosis or gliomatosis.
Here we have the same slow progressive growth, the same diffuse
nature and lack of delimitation ; indeed, my own experience
leads to the conviction that the majority of the so-called gliomas
belong to this category. What is perhaps the most well-marked
example of this condition is seen associated with, and apparently
underlying, the condition of syringomyelia.

Passing over certain less important examples such as leio-
myosis or leiomyomatosis (which, I would point out, is the more
correct nomenclature for what gynaecologists and modern text-
books wrongly term adenomyoma of the uterus), and endotheliosis
(such as is seen notably in the spleen in Gaucher's type of spleno-
megaly, and some cases of Banti's disease),1 I would at greater

1 There can now be no question as to the functional, or, indeed, superfunc-
tional nature of this striking overgrowth of the sinus cells in the spleen, since
Banti has demonstrated that operative removal of the organ terminates the
anaemia — prevents, that is, the excessive destruction of the red corpuscles.
(These cells lining the splenic sinuses possess normally the property of taking
up and destroying the red corpuscles. When they are developed in excess
the destruction becomes also excessive.)

EXAMPLES OF HYPERBLASTOSIS 345

length call your attention to a most important group of hyper-
blastoses, those, namely, affecting the lymphatic tissues and the
bone marrow.

That I may not be accused of forcing my point, I would here
quote the description of one of these allied conditions given by
a recent writer whose position is sufficiently indicated by the
fact that my quotation is taken from the " Referat " which he
was invited to give before last year's meeting of the Deutsche
Pathologische Gesellschaft. " The lymph nodes," states Pro-
fessor Eugen Frankel, " are liable to be most intensely affected,
swelling up into huge packets, and when it is the more superficial
lymph nodes that are involved, an immediate diagnosis can be
made. Often enough at autopsy it is determined that besides
the peripheral lymph nodes the internal collections are implicated :
those at the hilus of the lungs, in the mesentery, in the retro-
peritoneal tissue, exhibit similar change. It is in no wise necessary
that the nodes throughout the bodv are involved to the same
degree, although total exemption of one or other group is scarcely
ever observable. . . . The spleen also shows in general a notable
increase . . . indeed at times the enlargement of this organ
predominates to such a degree that the lymphoid enlargement
passes relatively into the background. . . . Cases, however, are
not infrequent in which the spleen remains small." And Frankel
points out that microscopically we deal in these cases with what
is purely a hyperplasia of the lymphoid tissue. This is a descrip-
tion of the condition which from a primary misfortune in nomen-
clature, from fastening the attention upon the outcome rather
than upon the underlying state, has for years been a source of
confusion : the condition, namely, described by Cohnheim as
pseudoleukaemia, by others as aleukaemic leukaemia, or as the
pre-leukaemic stage of leukaemia. For now many years my
teaching has been that lymphatic leukaemia is a blastomatoid
condition,1 and I willingly accept as the preferable nomenclature
that supported by Aschoff, Schridde, Hirschfeld and Nageli,
namely, lymphadenosis or lymphadenia. The time, indeed, has
come when, for clear thinking, the term leukaemia should be relegated,
by clinicians and 'pathologists alike, to its proper symptomatic rank :
we should speak of lymphadenosis with or without leukaemia,

1 Vide Adanii, Principles of Pathology, first edition, vol. i., 1908, p. 678
et sea.

346 ON GROWTH AND OVERGROWTH

distinguishing, I should add, between lymphadenosis (lymphatic
leukaemia) and myelosis (myelogenous leukaemia). For what is
true of the lymphoid tissue proper obtains also in respect to the
elements of the bone marrow ; these also may undergo diffuse
regional hyperplasia, with the complication that in the marrow
there exist elements of more than one type, and one or other of
these may exhibit hyperplasia. Thus in general lymphadenosis
the lymphocytic or lymphoblastic elements of the bone marrow
may be coincidently involved. In what I think we may term
myelosis proper (myelogenous leukaemia) it is the myelocytes
that in the main are hyperplastic, and this to such an extent that
the immature myelocytes are discharged into the blood stream,
and, at times, such is the stimulus to the formation of this type of
cell that organs like the spleen and liver resume the power they
possessed in embryonic life, and once again become the site of
myelocyte production.

The limitations set to communications before this Congress
forbid that I should do more than sketch the broad outlines of
my subject. I can merely state that lymphadenosis (leukaemic
and aleukaemic lymphatic leukaemia) and myelosis (myelogenous
leukaemia) are system diseases, hyperblastoses of the lymphoid
and myeloid tissue respectively of the whole organism, which,
under the influence of localization and regional intensity of the
process in the different areas of the body, are apt to induce
numerous variations in the facies of the disease. Saving this
it is necessary, parenthetically, to exclude Hodgkin's disease from
this group of hyperblastoses : with the majority of recent workers,
I regard this as of irritative and chronic inflammatory origin, as
a lymphogranulomatosis.

We must now pass on to the consideration of another feature
of these hyperblastoses, which for the sake of clearness I have
so far studiously kept in the background. I refer to their liability
to present malignant change. This may exhibit itself either as
a primary or a secondary phenomenon, and may be of localized
origin or generalized. It is characteristic not only of the lymph-
adenoses and myeloses, but of the hyperblastoses as a group,
that, instead of the constituent cells being of fully-formed adult
type, they may either locally or diffusely exhibit actively vegeta-
tive or " embryonic " characters. This, after all, is only what
might be expected from what we know of the general phenomena

LYMPHADENOSIS AND LEUKAEMIA 347

of growth : it is the cell that has normal relationships that is
most apt to attain and to retain complete differentiation.
Hyperplasia connotes cell proliferation, and where two cells
take and retain the place of one, ODe, if not both, of those cells
must fail to preserve the normal relationship to nutrient vessel,
stroma, etc. Hyperplasia thus favours anaplasia of at least a
portion of the cell elements of the affected part. Thus we find
that a portion of or all the cells of a gliosis may take on a more
sarcomatous type (I use the term here strictly in a histological
sense) ; areas of a liposis (e.g. of a retroperitoneal " lipoma ")
or of a leiomyosis may become sarcomatous ; and as regards the
myeloses and lymphadenoses, we observe a very interesting set
of conditions.

The so-called multiple myeloma, for example, has all the
features of a hyperblastosis — excessive development occurring
in certain bones only, and absence of limitation save by the
natural boundaries of the involved areas — with this in addition,
that the constituent cells are of " embrvonic " character, so
embryonic that here and there they are liable to exhibit active
malignancy, and may not merely absorb the bony trabeculae
but may infiltrate the periosteum and surrounding tissues, and
even, if rarely, may give rise to metastases in other organs at
a distance, or to quote Berlinger, " myeloma is becoming to an
increasing extent regarded as a systemic disease of a malignant
type." So vegetative is the type of cell in many of these cases
that, as well known, there is active debate regarding their origin,
whether they belong to the lymphoblastic or to the myeloblastic
type. The indications appear to be more and more convincing
that there may be a specific hyperblastosis involving each distinct
element of the bone marrow — the lymphoblasts, the myeloblasts,
and the erythroblasts (as in Bibbert's well-known case of megalo-
blastic overgrowth which he held to be an erythroblastoma
(more accurately an erythroblastomatosis).

Of peculiar interest in this connexion are the more recent
observations upon Chloroma. Here there are the same regional
ill-defined overgrowths most often occurring in early fife, and
involving particularly the skull, ribs, or sternum. Constantly
where the blood has been examined, the picture has been that of
leukaemia, with predominance of the large or relatively large
non-granular mononuclear cell, although in a small proportion

348 ON GROWTH AND OVERGROWTH

of cases there have been granular mononuclears. The picture
has been that of either acute lymphatic or acute myelogenous
leukaemia. By the employment of the oxidase (indophenol)
test,1 which differentiates between lymphocytes and myelocytes,
my colleague, Professor Burgess, has demonstrated clearly that
the marrow growths and the characteristic blood cells in this
condition are myelogenous, not lymphogenous. The condition
is a myelomatosis, or, to use ordinary terminology, an " acute
myelogenous leukaemia," and confirming Schultze, Longcope and
Cooke, and others, Burgess points out that cases of so-called
acute lymphatic leukaemia with " large lymphocytes " are truly
cases of acute myelogenous leukaemia (or strictly of acute myelo-
matosis with leukaemia).2

Parallel to these conditions affecting the bone marrow of
particular localities with their liability to malignancy, we may
cite the not infrequent mediastinal tumour, which originating
in the thymus (and perhaps sometimes in the mediastinal lymph
nodes) shows a striking liability to become locally malignant,
infiltrating all the surrounding tissues.

From these intermediate forms we pass on by imperceptible
gradations to cases of primary malignant hyperblastosis, to cases
of diffuse lymphosarcomatosis and myelosarcomatosis.

If the views here laid down and the relationships of this
group of conditions be correct, we arrive at certain interesting
conclusions. Namely, and first, that whatever be the essential
cause of the autonomous blastomas, we have in these hyper-
blastoses a group of diffuse overgrowths which by analogy must
be regarded as due to disturbances of metabolic equilibrium ;
which, further, in their simple non-malignant stages at least,
may possibly be combated eventually (certainly not to-day)
along the lines of organotherapy. Secondly, that every transition
is observable in this series between the development of over-
growths of fully-differentiated tissue, non-malignant grades of
anaplasia, and diffuse malignant infiltrative growths. This

1 Burgess, Journ. of Med. Research, xxvii., 1912, 133.

2 I must, however, disagree with my colleague in regarding leukaemia as
a blood metastasis. The symptom leukaemia is not an index of malignancy ;
the discharge of leucocytes into the blood is not of the nature of a malignant
infiltration ; on the contrary, it is best regarded as the outcome of chemiotactic
phenomena, as an attraction of marrow cells by something circulating in the
blood. There may be extensive lymphadenosis or myelosis without leukaemia,
or leukaemia may be present for a time, and then disappear — and reappear.

HYPERBLASTOSIS AND MALIGNANCY 349

suggests strongly that the causation of malignancy is not to be
sought for in the entry and action of external agencies, but in
stimulation of the growth properties of specific tissues by changes
in what I may term tissue equilibrium. And lastly, that the
more we study, the more we become impressed by the fact that
the number of conditions of hyperblastosis, whether simple or
malignant, is relatively considerable, so considerable as to
deserve separate consideration.

APPENDIX I

EXPERIMENTS WITH B. TYPHOSUS re ADAPTATION

BY

F. B. BOWMAN,1 M.B. (Toronto),
Major, C.A.M.C., O.C. No. 1 Canadian General Laboratory, Folkestone.

At the request of Lieutenant-Colonel Adami, the following experi-
ments were conducted to prove or disprove certain conclusions
reached by him regarding the effect of environment on certain cultural
reactions of micro-organisms.

Owing to the short time which was allowed for making these
investigations before the date set for the delivery of the Croonian
lectures, it was very questionable whether results of any interest
would be obtained. However, the following notes and charts seem
to bear out the contentions made by him, and to prove that instead
of chance variations in reaction we were dealing with direct adapta-
tion of the organism to its environment.

No investigation of the biochemistry or the raison d'etre of results
will be made, but simply a statement of facts from observation of
the reactions from day to day.

It is well known of course that certain bacteria in process of
growth cause fermentative reactions with different sugars, that is
that certain sugars would seem to be assimilable by certain organisms,
and that other sugars are useless as nutriment for their regular
growth and multiplication. The typhoid- colon group perhaps
illustrates this faculty better than any other, and for this reason
B. typhosus was used in this work.

1 Major Bowman delivered his charts and results personally to me on the
morning of the delivery of the lecture, in which reference was made to them.
Upon forwarding the proof of these charts to Major Bowman, with the suggestion
that without explanatory notes they appeared somewhat bare, he writes :
19/10/17 : " Perhaps the enclosed could replace the proof you have sent me.
However, I have changed the proof slightly if you should still use that, and have
initialled corrections." It is these expanded notes that are here given.

The observation given in the final paragraph is new to me, and deserves
further investigation. The probability is that the sugar in the neighbourhood
of the colonies had been used up, and that so acid production had ceased. —
J. G. A.

351

352

THE STUDY OF EVOLUTION

A pure culture of B. typhosus was obtained from Colonel Harvey
of the Royal Army Medical College. This organism was found to
agglutinate specifically in 1-5000 dilution. Fermentation tests were
then done and observations made for several days, and it was found
to ferment only mannite and glucose (see Chart I.). Azolitmin broth
(sugar free) was used as a base, and the different sugars added to this
medium.

CHART I
(Acid Production)

Sugar.

24 Hours.

48 Hours.

72 Hours.

4 Days.

Remarks.

Glue 086

+ +

+ +

+ +

+ +

Culture used re-

Lactose

-

—

—

—

acts specifically

Saccharose

-

-

—

—

with glucose and

Mannite

+ +

+ +

+ +

+ +

mannite only.

Dulcite

—

—

—

—

Isodulcite

—

—

—

—

Now, the question at issue seemed to be whether by reducing the
nutriment in the culture medium, i.e. the peptone, and at the same
time adding a non-fermentescible sugar, the organism would grow
by adapting itself to the food present in the medium. This would
be shown by fermentative reactions.

A culture medium of very low nutritive value was prepared as
a base, i.e. 0.5 cc. sugar-free peptone broth, added to 100 cc. of
distilled water, sterilized, and coloured with azolitmin. Cultures
of B. typhosus sown in this medium grew very slowly, but by
adding a fermentescible sugar in even a very small amount, growth
became much more rapid with prompt change in colour from
blue to red. By adding a non-fermenting sugar the growth was
very slow.

It was decided at first to gradually withdraw a fermentescible
sugar from the medium, and to substitute one non-fermentescible,
and to plate at daily intervals. Because of the short time at our
disposal this was only continued for a few days, but the results were
encouraging.

The following procedure was finally adopted :

100 cc. flasks of dilute peptone broth were prepared as above,
and 1 per cent isodulcite added, this being non-fermentescible, and
the whole sterilized on three successive days with steam, and azolitmin
added to colour. At the same time azolitmin-isodulcite agar was
prepared, only sugar-free broth being used.

Several flasks of dilute azolitmin isodulcite broth were inoculated
with B. typhosus. In 24 hours any change in reaction was noted,

APPENDIX I

353

and plate cultures were made, using azolitmin isodulcite agar. These
individual plates were themselves observed for several days.

In 48 hours a second plate was made from each flask and so on
for 6 days. The following Chart II. illustrates what occurred
graphically.

CHAET II

Isodulcite-azolitmin Agar Plate Cultures from 100 c.c. Flasks.

Time.

24 Hours'

48 Hours'

72 Hours'

Growth.

Growth.

Growth.

24 hours

No change in

No change in

No change in

colour

colour

colour

48 hours

do.

do.

do.

72 hours

do.

do.

do.

4 days

Slight pink tinge

Slight pink tinge

Pink tinge

5 days

Marked growth.

Marked growth.

Marked growth.

Distinct pink halo

Distinct pink halo

Distinct pink halo

around colonies

around colonies

and coloration

and some colora-

of medium

tion of medium

6 days

Marked growth.
Distinct pink
coloration

do.

do.

In 4 days there was no change in colour in the flask culture, but
in 5 days a very slight pink could be noticed, which had not increased
in 6 days. It will be seen that no change occurred even after
3 days in the 24-hour, 48-hour, or 72-hour plates, and only a slight
change in the 4-day plates. However, marked growth and distinct
fermentation occurred in the 5-day plates, and in the 6-day plates.

It would seem that direct ' adaptation to environment ' had
occurred. On examining these plates by chance three weeks later
it was seen that they were blue again, and apparently the organism
had lost its affinity for isodulcite, and was growing again on the agar
without producing acid.

2a

APPENDIX II

SIR E. RAY LANKESTER REBUKES RUDENESS: A
CORRESPONDENCE IN THE PAGES OF THE BRITISH
MEDICAL JOURNAL

I. From the British Medical Journal, July 14, 1917

To the Editor of the British Medical Journal

Sir — It would take an undue amount of your space and of my
time were I to state fully the grounds which I have for regretting
the tone and the matter of Dr. Adami's two Croonian lectures pub-
lished by you in your issue of June 23. Nevertheless I ask your
permission to lay them briefly before your readers. Those grounds
may be classed as matters of taste and matters of fact. With regard
to the first, Dr. Adami offends (a) by citing without my permission
(and in a garbled form) a private communication made by me to
him ; (b) by professing that " the time is ripe " for him to instruct
the biological world in elementary facts as to the experimental
modification of the activities and forms of pathogenic bacteria which
(as I had pointed out to him) are really familiar to biologists, and
are not and never have been — as he persists in asserting, in spite of
plain information to the contrary — treated with " superb indiffer-
ence " and neglect in this country ; (c) further by comparing Professor
Bateson, for the purpose of vulgar ridicule, with a bumble-bee in a
greenhouse ; and also (d) by making use of the unworthy method
of suggestio falsi in recklessly stating that the evolution of a new
property by direct acquirement is " contrary to the hypotheses and
dogmas of Professor Bateson and Sir Ray Lankester." My objection
to this is that Dr. Adami must know, if he knows anything about the
matter, that Professor Bateson's views and mine on this subject have
little in common. It is unfair to both of us to suggest that they are
identical, whilst it is merely rhetorical abuse to speak of those views
as " dogmas." Further, it is the fact that Dr. Adami was cate-
gorically informed by me eighteen months ago that the frequent
evolution of a new property in a race of bacteria by " direct acquire-

354

APPENDIX II 355

ment " was maintained by me forty years ago, and supported by
facts published by me and at that time novel, concerning the peach-
coloured bacterium (B. rubescens or roseo persicinum) and the frequent
pleomorphism of the bacteria, and that I, in common with many
other biologists — including my regretted friend, filie Metchnikoff —
have continued to hold and advocate that view in spite of the ad-
hesion of " medical bacteriologists " to Koch's doctrine of the fixity
of bacterial " species."

Dr. Adami now endeavours to give his Croonian lectures a flavour
of novelty by claiming for some medical bacteriologists who have
recently accepted the long-known views of myself, Warming, ZopfT,
and Metchnikoff, the merit of a new discovery. He actually, at this
late hour of the controversy, sets out, whilst announcing himself
as a " militant " reformer and missionary, to teach us what we have
done our best to teach him from his student days onwards.

So much for matters of taste. Now as to some of the matters of
fact misrepresented by Dr. Adami.

(a) Under the heading, " Nature of variation " (p. 838) Dr.
Adami confuses the term " variation " and " variability," and
thereby renders all his argument " suspect." No one disputes (as
Dr. Adami asserts that some do) that variability is a primary quality
of living matter inherent and not acquired. On the other hand, no
one denies that actual variations are brought about by the influence
of forces acting from without upon this labile variable living matter.

(b) Farther on (at the top of p. 840) Dr. Adami states that I
deny that there can be external influences of such a nature that
specific variation may be induced by them. I have never at any
time denied, but on the contrary always maintained, the truth of
this elementary proposition. Biologists do not require Dr. Adami
to pose, as he says, " under the shade of Harvey," and to cite well-
known facts in order to be persuaded of this. It is really unfortunate
that Dr. Adami should pretend that these facts are unknown to
those whom, with some cryptic implication, he terms c academic
biologists."

(c) Moreover, Dr. Adami is hopelessly wrong in his use of Herbert
Spencer's term " direct adaptation." He says that the worker in
pathogenic bacteriology is impressed with the fact that ' direct
adaptation " is one of the basal phenomena of living matter. This
is the great discovery which he thinks it his duty to teach to academic
biologists. He is singularly unfortunate, for he proceeds at once to
show that he has never understood what Herbert Spencer meant by
" direct adaptation." Dr. Adami says that direct adaptation is
" specific modification in response to a specific alteration in environ-
ment." That is not so. As Mr. Herbert Spencer is at some pains
to point out, a specific modification in response to a specific altera-
tion in environment is not necessarily an " adaptation." It may in

2a2

356 THE STUDY OF EVOLUTION

certain cases happen to be so, but on the other hand it may have no
such quality. Such a modification may be destructive of the life of
the organism, or it may be disadvantageous rather than adaptive,
or it may be, so to speak, neutral and without significance. Such
specific modifications in the case of the simplest living matter have,
it seems probable, never been directly adaptive, but some of the
many modifications so set up would, under some of the many further
changes of environment, prove of value in the struggle for existence
to the primitively simple living thing so modified, and would lead
to its survival by natural selection. This is called by Mr. Herbert
Spencer " adaptation by indirect equilibration " as contrasted with
" adaptation by direct equilibration." I need not pursue the subject
further. It is sufficiently clear that Dr. Adami is under the illusion
that mere modification of an organism in response to change of
environment is described by the term " direct adaptation," which
it is not. Accordingly in this case also, as in regard to the words
" variation " and " variability," he misunderstands what other
people have said on the subject on which he lectures, and, in con-
sequence, solemnly gives utterance to propositions which have no
serious meaning.

(d) Dr. Adami proceeds to maintain that the retention of its
changed character by a strain of experimentally modified bacteria
multiplying by fission is an example of the " transmission " of an
acquired character. He is probably aware that it is very widely
admitted that no case of the transmission of what are called
" acquired ' ' characters from parent to offspring has been demon-
strated in so far as those higher animals and plants which multiply
by means of specialized egg cells and sperm cells are concerned.
The well-known retention of induced change of character in very
simple fissiparous organisms does not go far towards rendering it
probable that transmission occurs in the case of elaborate multi-
cellular organisms with specialized reproductive cells. Dr. Adami
offers us a dissertation on the word " individuality," which is of no
assistance in the matter, and later proceeds to state that Weismann
" violently opposed " the doctrine of the transmission of acquired
characters. This is a baseless charge. Weismann was a patient
investigator, and anything but a " violent " controversialist. He
came to the conclusion that not only had no case of such trans-
mission on the part of higher organisms been demonstrated, but that
the mode of development and structure of the reproductive cells was
such as to make it improbable that such transmission could be
brought about. Nevertheless, Weismann would have examined
fairly and judicially any attempted demonstration by a competent
investigator of an instance of the transmission of such characters,
and so at any time would the " academic biologists " of this country.
Weismann would not have been interested in Dr. Adami's recent

APPENDIX II 357

orations, since he was well acquainted with the life-history of both
Protophyta and Protozoa.

It would not be right, since Dr. Adami has made a point of
alluding to me, that I should leave his statements unnoticed, as I
should have done had he not referred to me by name. — I am, etc.,

E. RAY LANKESTER.

London, W., July 4.

II. From the British Medical Journal, July 21, 1917

To the Editor of the British Medical Journal

Sir — Sir E. Ray Lankester, it appears, falls foul of the term
" academic biologist." By labelling a man as " academic " it is in
general meant to imply that such a one is more concerned with up-
holding the teaching and tradition of the Schools than with the
advance of his subject. Had I wished to demonstrate to your readers
what I meant by this term, I could not possibly have afforded a
more perfect instance than, by his letter in your last issue, Sir Ray
has himself presented.

Take, for example, his treatment of my use of the term " direct
adaptation," ascribed to Spencer, and of my demonstration that
this surely exists among the bacteria. Herbert Spencer wrote
(Principles of Biology, 1898 edition, vol. i. page 528) : " There go on
in all organisms certain changes of function and structure that are
directly consequent on changes in the incident forces — inner changes
by which the interchanges are balanced and equilibrium restored.
But ... we see that the modified conditions to which organisms may
be adapted by direct equilibration are conditions of certain classes
only. Besides direct there must be indirect equilibration." Where
is the difference between my " specific modification in response to a
specific alteration in environment within limits to be presently 1 laid
down " (Sir Ray manages to leave out the phrase in italics) and
Spencer's " certain changes . . . that are directly consequent," etc. ?
Later, in the abstract of my second lecture, which was in Sir Ray's
hands, it is pointed out that experiments can be so made as to remove
all possible question of chance variation and survival of those forms
alone which had exhibited variation in a favourable direction
(Spencer's indirect equilibration), so " that there was within certain
limits direct adaptation in the Spencerian sense, direct equilibration

1 In my fourth lecture. By the closing paragraph of my first, as again by
the invitation sent him to be present at the series, Sir Ray was advised that
other lectures were to follow in which my theme would be developed, but of
this he takes no notice.

358 THE STUDY OF EVOLUTION

•
between the organism and its environment." Sir Ray, it will be

observed, passes this over in complete silence, and makes no refer-
ence to my further demonstration of direct adaptation in the matter
of acquirement of pathogenic powers. But by a casuistic mingling
of Spencer's statements and his own opinion he implies that I confuse
direct and indirect equilibration, and states that I hold the view
that mere modification of an organism in response to change of
environment is in every case " direct adaptation." This is the
method of the academic biologist, not of the seeker after scientific
truth.

I freely admit that Sir Ray registers two definite hits. First
as to the confusion between "variation" and "variability." The
two are distinct, and the use of the latter in the circumstances was
inaccurate. It will be observed that the section is headed " The
Nature of Variation." I have before now employed the simile that
explosiveness, or explosibility, is one of the properties of nitro-
glycerine, but that it requires a force from without to produce an
explosion of this body. It would have been more accurate had I
written, " whether we deal with an inherent tendency to vary, or a
capacity on the part of living matter to he varied." But every fair-
minded reader will realize that this was the drift of my argument,
and will see that the misuse of the word variability weakened instead
of strengthening it. So, also, to those of us who have met Weismann
I confess that " vigorously " rather than " violently " would have
been the more appropriate adverb. But this again has no bearing
on the main argument. His hits are " outers."

As to the other more personal matters in Sir Ray Lankester's
letter, these again are all evasions of the main issue, such as a contro-
versialist employs to darken counsel. They turn upon a triangular
correspondence between X (an old friend of Sir Ray's), Sir Ray, and
myself, in which, at the request of X, I wrote my views upon adapta-
tion to him ; X showed my letter to Sir Ray ; and Sir Ray's reply,
while addressed to me, was forwarded to X to read and send on to
me. To quote loyally the sense of a statement is not to " garble "
that statement, with what that implies, and I challenge Sir Ray to
publish the correspondence in full elsewhere, to show that my refer-
ence was not substantially correct, and that I was not justified in
my action. My two letters are, and have always been, at his dis-
posal for this purpose.

For myself I desire no German naval victory in which, before the
action is fully developed, the fleet returns in triumph to its home
port sheltered by a cloud of inky smoke. Nor do hard hits irritate
me so long as they are clean. But I do feel most strongly that this
matter of adaptation and its mechanism is a matter that is not the
possession of zoologists and biologists alone, for them to dream
dreams about and hold unseemly parochial wrangles over at meetings

APPENDIX II 359

of the British Association. It is one in which medical men and the
community at large are to-day very immediately concerned, one in
which our voice has a right to be heard and our experience to be
seriously weighed and considered, instead of being contemptuously
spurned as coming from outside the pale ; one in which we look for
dependable leadership from men of the status of Sir Ray Lankester,
and not receiving it must take the lead ourselves. To quote from
Herbert Spencer's final words on the subject,1 " And now I must
once more point out that a grave responsibility rests on biologists
in respect of the general question (inheritance), since wrong answers
lead, among other effects, to wrong beliefs about social affairs and
to disastrous social actions. In me this conviction is increasingly
strengthened. Though the Origin of Species proved to me that the
transmission of acquired characters cannot be the sole factor in
organic evolution, yet I have never wavered in the belief that it is
a factor, and an all-important factor. And I have felt more and
more that, since all the higher sciences are dependent on the science
of life and must have their conclusions vitiated if a fundamental
datum given to them by the teachers of this science is erroneous, it
behoves these teachers not to let an erroneous datum pass current."
To understand the full significance of what is here written I would
ask my readers to turn to Sir Ray's letter of last week's issue and
read it once again. — I am, etc.,

J. G. ADAMI,

London, W. Lieut. -Colonel, C.A.M.C.

III. From the British Mjedigal Journal, August 4, 1917

To the Editor of the British Medical Journal.

Sir — Dr. Adami is suffering from an illusion. I have never
had, and at present have not, the intention of discussing a scientific
theory with him. His manners and methods render that impossible.
My purpose in writing to you was to expose (as I explained in my
letter of July 14) the offences against the laws of social intercourse
of which he has been guilty in making public use of a confidential
communication from me without my permission and in using vulgar
ridicule and rhetorical abuse when entrusted by a learned and digni-
fied College with the privilege of addressing it. I also took occasion
to show that Dr. Adami erroneously claimed novelty for the view
that the activities of the bacteria are susceptible of change under
changed environment — a view which forty years ago I was one of the
first to advance, although Dr. Adami goes out of his way to declare

1 In the Appendix to the last edition of the Principles of Biology.

360 THE STUDY OF EVOLUTION

" without rhyme or reason " that I am opposed to it. I showed that
his confusion of thought on this and related matters of fact is due to
his misapprehension of the significance of the words " variation,"
" variability," and " adaptation."

Dr. Adami admits the offences and the blunders to which I have
drawn attention ; indeed he cannot deny the plain evidence of his
own words. There is not (as Dr. Adami erroneously asserts) a
" main issue " or " main argument ' beyond these facts, in which
I am concerned.

When a man stands convicted on his own confession, as Dr. Adami
does, he either has the grace to express his regret and offer an apology
for his offences and blunders, or he has not. Dr. Adami has not
shown that grace ; until he has, I decline to pursue the matter any
further. He writes airily about " registering hits," and " outers,"
and " a German naval victory " as though the College of Physicians
and the columns of a great medical journal were a sort of children's
playground in which he is at liberty to " run amok " with a pretence
of importance accompanied by violence to bystanders, and then run
away without apology or penalty.

Even after his exposure Dr. Adami continues to write with
regrettable flippancy. He commences his letter to you of July 21
by stating that by " academic " he means a man who " is more con-
cerned with upholding the teaching and tradition of the schools
than with the advance of his subject." And then he is so far reckless
of his own reputation for discrimination and knowledge of character
as to apply this description to me. He forgets that others can
judge of the aptness of his definition and illustration, and also that
in his first lecture he ventured to invoke " the shade of Harvey,"
one of the greatest of " academic biologists." — I am, etc.,

E. RAY LANKESTER.
London, W., July 26.

IV. From the British Medical Journal, August 11, 1917

To the Editor of the British Medical Journal

Sir — It is a piteous exhibition that Sir Ray Lankester has made
of himself, and were it not that he has again touched upon points of
personal honour I would have preferred to be silent and not further
discover his shame. But by his deliberate misrepresentations and
reiterated and unfair attempts to place me in the wrong, he has left
me no option. He forces me to point out :

1. That he repeats and does not withdraw his accusation " that
Dr. Adami erroneously claimed novelty for the view that the activities

APPENDIX II 361

of the bacteria are susceptible of change under changed environ-
ment." I made no such claim. On the contrary, I called attention
to the fact that Pasteur, Roux, and Chamberland had demonstrated
changes of this order. It was not necessary for me to inform my
audience that the demonstration was afforded in the early 'eighties,
or that Pasteur died before this century opened. What I did claim
was that " the latter-day investigations in medical science " showed
that certain of the above-mentioned changes were of the nature of
direct adaptation, or direct equilibration — a very different matter.

2. That he repeats the charge that I misapprehend the signi-
ficance of the word " adaptation," and states that I admit the
blunder. Far from making any such admission, I, on the contrary,
proved to the hilt, by a reference to Herbert Spencer's own words,
that the misapprehension was his, that my use of the term was con-
sistent and correct, and that it was he and not I who misused it.
He must himself admit, now that the heat of the moment has passed,
that this is not clean fighting. It is inexcusable.

3. That as for his repetition of the charge of an offence against
the laws of social intercourse by making use of a confidential com-
munication, I must point out, in fuller detail than in my last, that
it was Sir Ray Lankester himself who, while knowing my address,
by transmitting his letter, addressed to me, to a third person for
prior perusal, converted what would otherwise have been a private
into an open correspondence. I have his letter of transmission
before me as I write. It may interest him to know that previous
to the delivery of my first lecture I put the case before two friends
of his, both men of distinction, one of them a Fellow of the Royal
College of Physicians. Both held that I would be justified in referring
to the correspondence, and this, in the case of one of the two, after
he had read the manuscript and seen the exact terms of the reference.

It will be observed that Sir Ray Lankester takes no notice of my
challenge that he should publish this earlier correspondence and
prove that I had, as he asserted, "garbled" my reference to his
views. He cannot afford to refuse to discuss so vital a matter as
this of the evidence recently obtained regarding direct adaptation
with its bearing upon heredity and social matters of the first order.
He cannot do this without tacitly resigning his position as a leader.
To preserve his own honour I again ask him to publish the triangular
correspondence already referred to, that the points at issue may be
clearly defined.

Surely I need say no more. — I am, etc.,

J. G. ADAMI.

London, W., August 6.

INDEX

Abderhalden, and anaphylaxis, 38 ;
serological studies by the optical
method, 91, 265 ; on enzyme nature
of ectotoxins, 259

Abrin, 47; effects on offspring, 62

Accustomance, law of, 166

Achorion schoenleinii, 23

Acquired characters, Weismann on
non - transmissibility of, 134 ; as
distinct from inherited, 140 ; ther-
mogenic acquirements, 151 (note)

Acquired constitutional states, direct
transmission, 157

Acquirement of conditions of defect,
66, 153

Acromegaly, 342; antiquity of, 16
(note)

Adami and Aschoff, 82

Adams, F. D., 177

Adaptation, direct, Sir E. Ray Lan-
kester on, 10, 355, 357, 361;
Lamarck's views, 10 ; the crucial
experiment, 351 et seq.

Adaptation, in the bacteria, 15, 28
et seq. ; to increased temperature,
40; of harmless bacteria to growth
in the tissues, 41 ; in protozoa, 45 ;
to disease-producing agencies in
higher animals, 45 et seq. ; leuco-
cytic, 53 ; and inflammation, 116
et seq.

Adenoma, malignant, of kidney, 317 ;
of suprarenal, 317

Adenomyoma of uterus, 344

Adiposis, 342

Adsorption by fluid crystals, 186

Albrecht, 184, 187

Alcohol, effects on offspring, 63

Algae, marine, antiquity of, 17

Algonkian limestones, mode of for-
mation, 17

Amboceptor, 53, 251

Amino -acids, 74, 221

Amphimixis, 13, 94

Anaphylaxis, 254 et seq. ; its nature,
38 ; relation to immunity, 42 ;
and inheritance, 67 ; mechanism of,
92 ; a stage in development of
immunity, 256

Anaplasia and depression of function,
249

Anaspides, 18

Andrewes, E., on streptococci from
throat, 24, 36

Antibodies, 50

Antitoxins, production of, 48, 50, 244 ;
and toxins, relation of, 50

Apathy, 172, 182

Arnsperger, 285

Arthritis, rheumatoid, antiquity of , 16 ;
in tertiary horse, 16

Aschoff, J., on myelins, 174 ; their
relationship to gall-stones, 183 ;
on compounds of nitrogenous bases
and fatty acid, 205; on lympha-
denosis, 345

Atavism, 137 ; explained by side-
chain theory, 153

Atom, size of, 214

Auto -intoxications, inheritance of, 159

Autolysis and myelin production,
187

Automaticity of the cell, 240

Axolotl, conversion to Amblystoma,
67 (note)

Babes, V., 125

Bacillus anthracis, 35, 115, 119, 124

„ botulinus, 23 (note)

,, coli, 126; mutations of, 29

„ cyanogenus (syncyanus), 117

,, diphtheriae, 129

„ hoffmanni, 129

„ leprae, 39

„ megatherium, 152 (note) ; in-
duced virulence of, 37

363

364

THE STUDY OF EVOLUTION

Bacillus mycoides, 39
„ perturbans, 31
„ prodigiosan, 56, 111, 116
„ pseudo-diphthericus. See B.

hqffmanni

„ pyocyaneus, 109, 111, 113,

117, 158

ruber (of Kiel), 110, 113, 114,

117, 119 (of Plymouth), 113

„ rubescens. See B. roseo-

persicinum
„ tuberculosis, retention of pro-
perties, 57 ; strains of, 25
„ typhosus, 28, 109, 112, 125;
and adaptation, 351 et seq. ;
mutations of, 32
„ zopfii, 108
Bacillus of swine erysipelas, 116; of

haemorrhagic septicaemia, 123
Bacteria, what constitutes the in-
dividual, 12; antiquity of, 16;
pathogenic, allied non - virulent
strains, 24; experimental produc-
tion of specific variations, 33 ;
changes in fermentative powers,
36 ; variability, 103 et seq. ; fixity
of species, 104; polymorphism,
108; chromogenic, 110, 112, 147;
production of races, 1 14 ; natural
races, 121 ; of haemorrhagic septic-
aemia, 123 ; in healthy tissues,
165
Bacterial species, doctrine of fixity of,

24
Bacterium calcis, 17

„ roseo-persicinum, 108, 355

Ballantyne, 336

Banti, on races of pneumococci, 124
Banti's disease, 344
Barber, 29

Barfurth, on proliferation of nerve

cells, 271 ; on regeneration of

muscle, 267 ; on totipotential, pluri-

potential, and unipotential cells, 336

Barnes on activities of the electrons,

241
Bashford, 201

Bateson, W., his doctrine of evolution
by loss, 6, 34; his statical concep-
tion, 97 ; on backward evolution,
97
" Baustein " theory, 252
Bayliss, W. M., 86; on unicellular

organisms, 12
Beard, J., 318 (note)
Beatson, 331

Beneke, 172, 182

Berlinger, 347

Bernard, Claude, on unicellular organ-
isms, 12 ; on nucleus and cell
substance, 200

Bernard of Chartres, 5 (note)

Bilharziosis, 284, 303 ; antiquity of, 16

Billings, F., 37

Biophores, 206 ; their constitution,
77 ; growth of, 80 ; interaction
between, 95

Bioplastic cell activities, 84

Birkett and Meakins, 237

Bizzozero, 267 ; on neoplasia, 273

Blastomas, 335, 337 ; classification,
286 ; and hyperblastosis, relation-
ship between, 340 et seq.

Blastomatoid growths, 338, 340

Blastomycosis, 21

Bone, proliferation of, 265

Bordet, on nature of bacteriolysis, 251

Boveri, 193, 195

Bowman, F. B., 33; experiments with
B. typhosus re adaptation, 351

Bretonneau, 42

Briicke, 187, 205

Brunton, Sir L., and Macfadyean, 30,
111

Buchner, 109

Burgess, 348

Buxton, on the two orders of tissues,
313

" Cancer bodies," 278, 332

Cancer, causation of, 276 et seq. ;
parasitic theory, 278, 302, 329, 333

Car bone, 184

Carcinoma, definition of, 317 ; sarco-
matodes, 320 ; of uterus, 32

Caries, antiquity of, 16

Carlier, 198 ; on cell proliferation and
secretion, 289

Carriere, on effects of tubercle bacillus
toxins, 62

Cassedebat, 126

Cell activities, neurogenic, environ-
mental, and automatic, 240

Cell and tissue differentiation, 93

Cell heredity, 59, 245, 247

" CeU rests," 274, 288, 293, 335

Cell, functional versus vegetative
activities, 84, 191, 269, 293; de-
generations and the nucleus, 202 ;
automaticity of, 240 ; paraplasmic
substances, 242; and enzyme
action, 245; metaplasia, 246;

INDEX

365

depression of function (anaplasia),
249 ; influence of environment upon,
290 ; inertia, 292 ; hyperplasia,
conditions originating, 294 ; in-
heritance of acquired characters
by, 300 ; microbic stimulation of
growth of, 329 ; as the parasite in
malignant neoplasms, 330 ; uni-
potent, pluripotent, and totipotent,
334 et seq.

Centrosome, regeneration of, 195

Ceratodus, 18

Chance and variation, x ; in nature, 30

Chalk, mode of formation, 17

Chantemesse and Widal, 112

Charrin, 109

Chipman, W., 327

Chloroma, 347

Cholera, antiquity of, 20, 21

Cholesteryl oleate, 181

Cholin oleate, 182

Chordoma, 306

Chorio -epithelioma malignum. See
Syncytioma

Chromatin, conveyor of heritable
matter, 78

Chromosomes, different forms of, in
cell, 192; accessory, 193

Coccidiosis, 283, 331 (note)

Cohnheim's theory, 274, 288, 293,
341

Coley, 304

Colitis, mucous, 239

Collie, Sir J., 58 (note)

Complement, 53

Conditions of defect, acquirement of,
66

Congenital versus inherited conditions,
140

Conjugation and inheritance, 148

Conservation of energy, and life, 226,
231

Continuity of the germ plasm, 83, 210

Correns, 95 (note)

Creighton, 19

Crookshank, 121

Crystallization, process of, 230

Crystals, fluid and flowing, 179 ;
fluid, adsorption by, 186

Cunningham, D. D., 127

Curtius, 221

Cyanophyceae, antiquity of, 17

Cytology and sociology, 242

Dallinger, 40

Dallingeria drysdalei, 40 (note)

Darwin, Charles, the pangenesis hypo-
thesis, 69, 72 ; atavism by crossing
pigeons, 137, 153

Darwin, Erasmus, 6

De Bary, 85

De Waele, 259

Deciduoma. See Syncytioma

Dedifferentiation and proliferation,
265; in tumours, 321

Defects, acquired, non-inheritance of,
153

Degenerations, cellular, 202

Dercum's disease, 342

Diathesis, inheritance of, 60, 154

Dietrich and Hegel, 184, 187

Differentiation of tissues, 93

Digestion, enteral and parenteral, 253

Dinwiddie, 25

Diphtheria, origin of, 42 ; antitoxic
immunity, 49 : variations in virul-
ence, 104

Diphtheroid bacilli, 24

Diplococcus pneumoniae, 112 ; races
of, 124, 131

Direct adaptation, proof of, in patho-
genic bacteria, 44

Drew, 17

Dreyer, 55

Driesch, 139

Ductile crystals, 179

Ductless glands, influence of, upon
growth, 341

Durante, on neurofibromatosis, 344

Duval, Matthias, 328 ; and Couret, 39

Earl, on life and the conservation of

energy, 226, 232
Ectotoxins, diphtherial, 258 ; of

tetanus, 259 ; shown to be enzymes,

259
Ehrlich, on strains of typhoid bacilli,

28 (note) ; on immunity against

ricin, 48 ; on orders of cell poisons,

50 ; side -chain theory of immunity,

144, 251
Electrons, 241
Elephant, evolution of, 19
Elliot, Hugh, 10 (note)
Embryoma, 335, 337
Endothelioma, 318, 320, 321, 323
Entamoebae, adaptation in, 46
Enteral and parenteral digestion, 253
Enteric fever, immunity against, 51
Environment and acquirement of

properties, 151 ; influence of, on cell

activities, 290

366

THE STUDY OF EVOLUTION

Enzyme action, 86 et seq. ; and the

nucleus, 204; as the basis of cell

acquirements, 245
Enzymes, proteoclastic, indifferent,

and specific, 256 ; and antitoxins,

258
Eohippus, 19

Epidermis, proliferation of, 265
Epithelioma of gall-bladder, 59, 299 ;

squamous, of larynx, 299
Equilibration, direct and indirect, 10,

357
Erythroblastomatosis, 347
Erythrocyte, life period of, 195
Eurypteridae, 18
Evolution of new diseases, 41
Exaltation of virulence, 35, 116

Faber, Kniest, 40

Fabre-Domergue, 279, 336

Fatty degeneration and the myelins,

187, 205
Favus, 23
Fear neuroses, 58

Ferment actions. See Enzyme action
Fertilization of enucleated egg, 193
Firmin-Didot, 5 (note)
Fischer, B., 248
Fischer, Emil, 74, 221
Fletcher, Chune, 343
" Flowing crystals " and " fluid

crystals," 179
Fluctuations, 28, 29
Fluid crystals, 168 et seq.
Foa, 125
Ford, W. W., 51 ; on bacteria in

healthy tissues, 165
Foster, Sir Michael, on digestion,

252
Fournier, 156

Frankel, C, 116 (note), 123
Frankel, E., 121 ; on leukaemia,

345
Frankel and Simmonds, 109
Frothingham, 25
Fuligo, 46
Functional activities, 84

Gaffky, 126

Galton, Francis, on non-inheritance
of mutilations and use acquire-
ments, 72

Gall-bladder, epithelioma of, 247

Gamaleia, 123

Garwood, 17

Gaskell, W. H., studies upon verte-

brate descent, 4 ; on anabolism and
katabolism, 85

Gaylord, 332

Geikie, Sir A., 18

Geddes, Sir Auckland, 16 (note)

Germ cells, extrinsic and intrinsic
influences on, 68 ; influence of
position on, 152

Germ layers, 309

Germ plasm, continuity of, 83, 210 ;
and somato - plasm, 93 ; and the
nucleus, 191 ; multiplication of, in
spermatocytes, 211

Gessard, 111, 117, 119

Gheorghiu, 157

Giant growth, 340

Gland cells, proliferation of, 266

Glioma, 306, 324 ; false epithelium in,
321 (note)

Gliosis, 344

Goodsir, 23

Gout, inheritance of, 159

Gowers, Sir W., on structure and
properties of alkaloids, 222-4

Grawitz, 202

Growing point in plants, 265

Growth, 79, 227 ; and the biophores,
80 ; its essential nature, 141 ; of
bioplasm, 227 ; and crystallization,
230 ; habit of, 247, 298 ; and over-
growth, 263 et seq. ; and functional
activity, 288 ; giant, 340 ; relation-
ship of ductless glands to, 341

Gurd, Fraser, 93; on lysins, 257

Haberlandt, 196

Habit, the law of, 55, 166 ; and
symptoms, 57 et seq. ; movements,
58 ; of growth, 59, 247, 298 ; symp-
toms and disease, 235 et seq.

Haemorrhagic septicaemia, 123

Hales, Stephen, on the value of hypo-
thesis, 233

Halley's comet, 244

Hamilton, D., 307

Hansen, 151 (note)

Harris, Fraser, 243, 292; on law of
inertia, 55, 166

Harris, T., 303

Haureau, 5 (note)

Heidenhain, R., 196

Hering, on memory, 226 (note)

Hermann, F., 197

Herter, 204

His, W., on archiblast and para blast,
310

INDEX

367

Higgins, C, 37

Hodge, 197, 289

Hodgkin's disease, 346

Hoffmann, 129

Hog cholera, 22

Holman, 36

Hofmeister, 75

Horse, evolution of, 19

Horsley, Sir Victor, 63 (note)

Hort, 23

Hrdlicka, 20

Huber, C, 325

Hubrecht, 327

Hueppe, 112, 123; and Wood, 124

Hunter, W., 195

Hurthle, 183

Hylic tissues, 313, 314

Hylic tumours (hylomas), 315

Hyperblastosis, 340 et seq.

Hyperplasia, 294

Hysteria and habit, 58, 235, 237

Id, size of, 214

Idioplasm, 142

Immune body, 53, 251

Immunity, an adaptation, 47 et seq. ;
antitoxic, 47 et seq. ; bacteriolytic,
51 ; the mechanism of, 90 et seq. ;
inheritance of, 158; antitoxic and
bacteriolytic, 250 ; as a special
case of the digestive process, 252 ;
and anaphylaxis, 256 ; bacteriolytic,
stages of, 256 ; stages of, 260

Individuality in the bacteria, 12

Inertia of the cell, 55, 166, 243

Infectious diseases, antiquity of, 15
et seq.

Inflammation and adaptation, 161
et seq.

Inheritance, of acquired properties,
doctrine of, history, 6; of acquired
conditions in higher animals, 60
et seq. ; of acquired conditions,
influence of hormones, 68; theories
of, 132 et seq. ; true and false, 140
et seq. ; physico-chemical theory of,
143 et seq. ; effects of conjugation,
148; part played by cell nucleus,
149 ; in multicellular forms, 150 ;
of diathesis, 154 ; of infections,
154; indirect, 155; of acquired
constitutional states, 157 ; of im-
munity, 158 ; of endogenous intoxi-
cations, 159

Intoxications, endogenous, inheritance
of, 159

Israel, O., on the two orders of tissues,
313; on properties of the endothe-
liomas, 323

Jequirity, 47

Joannes Sarisbcriensis, 5 (note)

Kaiserling and Orgler, 168

Kamen, 109

Kammerer's experiments, 66

Karyolysis, 202

Katabiotic cell activities, 84

Kelvin, Lord, on size of the atom, 214

" Kernkorperchen," 197

Kidney tubules, proliferation in, 267

Klebs, 195, 310

Klein, E., 105

Klotz, O., on B. perturbans, 31 ; on
production of calcium soaps in
organism, 173 ; on protein fatty-
acid compounds, 187 ; on assump-
tion of habit in dysentery, 239

Koch, R., on fixity of bacterial species,
25

Kolliker on muscle proliferation, 268 ;
on regeneration, 269

Korschelt, 196

Kossel, 74; on chemistry of the
nucleus, 203

Kroon, 50

Krumweide, 26

Kurth, 108

Kyes, Preston, 206

Lacrymaria olor, 195

Lamarck, on nerve paths, 58 ; on
inheritance of identical induced
change, 69

Lamarckian doctrine, ix, 10

Langerhans's bodies, 266

Lankester, Sir E. Ray, and La-
marckism, 10; on Lamarck and
Darwin, 69; on B. roseo-persicina,
108; rebukes rudeness and is re-
buked, 354 et seq.

Laurent, 110, 117

Lead poisoning, effects on germ cells,
61, 156

Leech, D. J., 89 (note)

Lecithins, 184, 205

Lehmann, O., on fluid crystals, 81,
177

Leishman's experiment, 54

Leiomyomatosis, 344

Lepidic tissues, 313

Lepidic tumours (lepidomas), 315 ;

368

THE STUDY OF EVOLUTION

primary and secondary or transi-
tional, 325

Levy, 122

Leucocytes, and development of
immunity, 42 ; and adaptation, 53 ;
phagocytic activities of, 166

Leukaemia, 345

Lichty, 239 (note)

Liebreich, 182

Life, a lecture on, 216 et seq. ; and
crystallization, 231 ; definition, 231

Lillie, 195, 204

Limestones, mode of formation, 17

Limulus, 18

Lingula, 18

Lipoids, 186, 205

Lize, on nitrate of mercury poisoning,
62

Loeb, Jacques, 204 ; effects of altera-
tion of tonicity of medium, 240

Loeb, Leo, 318 (note)

Loftier, 116, 122, 129

Lorenz, 122

Lubarsch, 202, 307

Lugaro, 197

Lukjanow, 197

Lustig, 62

Lymphadenosis, 345

Lymph nodes, proliferation in, 266

Macallum, A. B., 289 ; on cell-chem-
istry, 190; extrusion of nucleolar
matter, 197 ; on chemistry of the
nucleus, 203

MacBride, E. W., on influence of
hormones on the germ cells, 69 ; on
origin of the endothelium, 311

Mackenzie, 33 (note)

McKlung, 193

Malignancy in the hyperblastoses, 346

Malformations, influence of parental
infection in causation, 157

Mallory, 282

Malm, 116

Mann, Gustav, 197, 289

Marchand, 328

Marmorek, 35

Marsh, 19

Martin, Sidney, on toxic albumoses,
258

Martinotti, 289

Massini, 31

Maximow, 198

Meakins on assumption of habits, 238

Measles, in Pacific Islands, 21

Melanotic tumours, 319

Memory, 226 (note)

Mendel, and the Mendelian doctrine, 9

Mendelism and the biophores, 95

Meningococcus, allied strains, 24

Metabolism, 85

Metaplasia, 246 ; and cell habits, 58 ;

the law of reversion, 273
Metchnikoff, on tetanotoxin, 51 ; on
phagocytosis, 53 ; and Bordet, 53 ;
on cancer parasites, 278, 332; on
phagocytosis by tissue cells, 328
Mettenheimer, 172
Mesenchyme, origin of, 310
Mesothelioma, 320, 321
Microbacillus prodigiosus, 111, 113
Micrococcus indicus, 110, 111

„ pneumoniae. See Diplo-

coccus
„ pyogenus, 114

Minot, Sedgwick, on the embryologi-
cal basis of pathology, 308 ; on
origin of endothelium, 311 ; on the
notochord, 309
Mischgeschwulste, 335, 337
Modifications, 28 ; transient, of bac-
teria, 108
Molluscum fibrosum, 344
Momentum of living matter, 243
Monboddo, Lord, 6
Montgomery, 192
Moodie, 16
Moore, Norman, 20
Moore and Arnold, 192
Morgan, T. H., 139
Morris, Sir EL, 20 (note)
" Mother cells," 264, 288 ; and neo-
plasia, 273; their proliferation in
functioning tissues, 294 (note)
Mouse septicaemia, 122
Mucous colitis, 239
Mucous membranes, proliferation of,

266
Miiller (of Marburg), 172
Muller, B,., 31 (note)
Muscle, proliferation in, 267 ; re-
generation, 267
Musgrave and Clegg, 46
Mutations, 28 et seq.
Mutilations, non -inheritance of, 153
Myelins, the, 168 et seq.
Myeloma, multiple, 347
Myelosis, 346
Myxomycetes, adaptation in, 46

Nageli, 108, 142
Natural selection, x

1

INDEX

369

Nautilus, 18

Neoplasia, and cell habits, 59; and
the nucleus, 200 ; and the habit of
growth, 247; and anaplasia, 249;
and dedifferentiation, 273 ; benign
and malignant tumours, 280; the
primary cause, 295 ; the " habit of
growth " theory, 301 ; microbes as
one causative agent, 302

Nepa, 196

Nephritis, Epidemic War, 24

Nernst, on size of atom, 215

Nervous habits, 237

Neubauer, 175 (note)

Neurin oleate, 182

Neurofibromatosis, 344

Neurogenic cell activities, 240

Neuroma, 324

Neurones, existence of " mother cells "
for, 271

Nicholls, A., 165

Nikiforoff, 124 (note)

Nitrate of mercury, effects on off-
spring, 62

Nucleolus and cell activities, 198

Nucleoproteins, 203

Nucleus, and cell activities, 79, 289;
and inheritance, 149 ; dominance
of, 188 et seq. ; and germ plasm,
191 ; in cell growth and multiplica-
tion, 191 ; effects of removal, 195 ;
relationship to cytoplasm, 195, 207 ;
changes during activity, 196 ; and
functional activities, 196 ; and
secretion, 197; in pathological con-
ditions, 200; chemistry of, 203;
and enzyme actions, 204

Ogata, 197

Ontogeny and phylogeny, 93
Osier, Sir W., 28 (note)
Osteoblasts, 265
Osteomyelitis, 16

Paget, Sir James, 290
Paleopathology, 16
Parallel induction, 66
Paranuclear bodies, 201
Paraplasmic substances, 242
Parasitic theory of cancer, 278, 302,

329, 333
Parenteral digestion, and immunity,

250 et seq. ; and enteral digestion,

253
Park, Roswell, 277
Pasteur, on adaptation in bacteria, 31 ;

asporogenous races of anthrax

bacillus, 35 ; method of protection

against rabies, 41 ; and Thullier,

116 ; on science and religion, 217
Pasteur, Chamberland, and Roux,

115, 119
Paul, Constantin, on lead poisoning,

61, 156
Pearson, Karl, 63
Peckham, Miss A. W., 31
Penfold, 31
Perithelioma, 320
Peters, H., 285, 326
Pfeiffer on Vibrio metchnikovi, 123 ;

on nature of bacteriolysis, 251
Pfeiffer's reaction, 52, 257
Phagocytosis, 53

Phytotoxins and bacteriotoxins, 49
Pianese, 279, 336
Picardy sweat, 19
Piltdown skull, 16 (note)
Placental syncytium, 286
Plasmosomes, 197, 204
Plimmer, 332

Pneumococcus. See Diplococcus
Polypeptids, synthesis of, 74, 222
" Poor curate " phenomenon, 248
Porencephalus, 271
Protagon, 182

Proteidogenous matter, 74, 219
Protein fatty-acid compounds, 187
Proteins, their constitution, 73 et seq.,

220, 225 ; the essential constituent

of living matter, 219
Prozymogen, 198
Pseudoleukaemia, 345
Pulmonary epithelium, proliferation

in, 267
Purpose in vital phenomena, 163
Pyknosis, 202
Pyorrhoea alveolaris, antiquity of, 16

Quincke, G., 172, 175, 180, 187, 195

Rabbit, in Australia, 21

Races, production of, in bacteria,

114
Radziewski, 52
Regeneration, 270 et seq.
Reid, Archdall, on alcohol and the

race, 63 (note)
Renault, R., 17
Repin, 285
Reserve force, 165
Reversion, 269 ; senile, and neoplasia,

274

370

THE STUDY OF EVOLUTION

Ribbert, 288, 307, 341 ; on erythro-
blastoma, 347

Ricin, 47

" Riesenwuchs," 339, 340, 342

Rorner, 50

Roncali, 278

Root hairs, 196

Rosenfeld, 187

Rosenheim, 33 (note)

Rosenow, on induced modifications in
streptococci, 36 ; on elective localiz-
ation of streptococci, 43

Rossi-Doria, 332 (note)

Ross's law, 57, 147

Roux, E., on loss of spore formation,
115, 120; on Pasteur, 217; and
Yersin, 49, 129, 258

Ruffer, Sir M. A., on paleopathology,
16; on bacteria in healthy tissues,
165 ; on cancer parasites, 278

Russell, 278

Rutherford, Sir E., 241

Salamandra maculosa, 66

Sanderson, Sir J. Burdon, on phago-
cytosis, 54 (note)

Sanfelice, 278, 279, 303

Sarcina erythromyxa, 113
,, ventriculi, 23

Sarcoblasts, 268

Sarcoma, definition of, 317; carcino-
matodes, 320

Schenck, 67; on fluid crystals, 177;
on adsorption by fluid crystals,
186

Schiller, 109

Schloss and Foster, 39

Schmaus and Albrecht, 202

Schoenlein, 23

Sedgwick, A., 84

Senn, N., 274, 305

Sex and the accessory chromosome,
193

Shaw, W. V., 39

Shell shock, 58

Shirres, D. A., 271

Side-chain theory of inheritance, 144

Sjobring, N., 278

S. Joaquin Valley, 21

" Slumber cells," 202

Smith, Elliot, 20

Smith, Erwin F., on crown galls in
plants, 283 (note)

Smith, Theobald, 25

Soaps and myelin, 175

Sociology and cytology, 242

Sparrow, in North America, 21
Spencer, Herbert, on inheritance of

acquirements, 6 ; on the deter-
minants of the peacock's .tail, 73 ;

and direct adaptation, 355, 357,

359
Spermatozoa, multiplication of, 83,

211
Sphero -crystals, 181
Spirillum cholerae, 52, 111, 112, 113,

123, 127
Spirillum F inkier -Prior, 113

„ metchnikovi, 53, 113, 123
Spirogyra, 195
Spitzer, 204
Splenomegaly, 344
Stahl, 46

Staphylococcus pyogenes, 114
Steinhaus, 198
Stockard's experiments, 63
Stolnikow, 202
Streptococcus, exaltation of virulence,

35, 40 ; elective localization of, 43 ;

erysipelatosus, 121
Strong, 21
Sudhoff, 20

Survival of the fittest, 11
Sutton, 192

Sweating sickness, 19, 42
Swine erysipelas, 116
Syncytioma malignum, 286 ; its

bearing in nature of malignancy,

326 et seq. ; demonstration of its

nature, 328
Syncytium (placental), 286
Syphilis, antiquity of, 20 ; mode of

origin, 42 ; not inherited, 140 ;

parasyphilitic inheritance, 155 ;

Fournier's observations, 156

Teleological conceptions, what con-
stitute, 163

Terato-blastomas, 337

Teratomas, 282, 285, 334

Tetanotoxin, 51, 259

Thalassicola pelagica, 195

Thiele and Embleton, 39

Thies, S., 199

Thoma, 307

Thomson, Arthur, 215 (note) : on
determinants, 71 (note)

Thomson, Sir J. J., 241

Thompson, D'Arcy W., on fortuitous
variations, 11 ; on growth, 79 ; on
fluid crystals, 82

Thudichum, 184 (note)

INDEX

371

Tics, 58, 235

Tissues, the, lining membrane and
pulp tissues (lepidic and hylic),
311

Torhorst, 174

Tornier, 266

Torrey, 199

Toxins as causes of cell prolifera-
tion, 282; and antitoxins, relation
of, 50

Trench fever, 22, 42

Trench shin, 22, 42

Trophoblast, 327

Tuberculosis, antiquity of, 16 ; bovine,
in man, 26 ; not inherited, 141 ;
inheritance of diathesis, 158 ;
pulmonary, assumption of habit
in, 235

Tuberculous diathesis, inheritance of,
61

Tumours, causation of, 273 et seq. ;
benign and malignant, 280 ; defini-
tion, 281 ; due to parasites, 283 ;
in plants, 283 ; starting - point
common to all, 295; the " habit of
growth " theory, 301 ; microbes as
a causative agent, 302; melanotic,
319; the cells as parasites, 330;
mixed, 335

Tumours, classification of, 305 et seq.;
embryogenetic, 284 ; Waldeyer's
classification, 305 ; by cell potency,
334 et seq.

Twort, 31

Van Tieghem, 17

Variability, of bacteria, 103 et seq. ; in
protozoa, Weismann on, 106; tran-
sient, 108

Variation, chance and, x; inherent
or acquired, 6 ; influence of sexual
conjugation, 13 ; by loss of factors,
34 ; in pigment production by
bacteria, 56 ; and variability, 355,
358

Variations, favourable, 10 ; fortuitous,
11 ; impressed, 28 ; impressed, as
against chance, 33 ; specific, ex-
perimental production of, in
bacteria, 33

Vaughan, Victor C, on protein split
products, 38, 91, 253 ; anaphylaxis,
254

Vegetative activities, 84

Verogay, 344

Verworn on the enucleated cell, 195 ;

on nucleo-cytoplasmic relationship,

195
Vibrio metchnikovi. See Spirillum
Vincent, 37 ; on B. typhosus, 109 ;

on B. megatherium, 152 (note)
Von Chauvin, Marie, 67 (note)
Von Hansemann, 59 ; on tumours,

288; on classification of tumours,

307
Virchow on cells as vital units, 85 ;

on myelin, 170; and the develop-
ment of pathology, 189
Virulence, exaltation of, 35, 116 ;

development of, in non-pathogenic

bacteria, 39
Vital inertia, 55, 166

Walcott, C. D., 17

Waldeyer, 305

Waldvogel and Mette, 184

Walker, 143, 147

Walker, Ainley, on mutations, 29
(note) ; on changes in fermentative
powers of bacteria, 36; on adapta-
tion by bacteria, 43 (note)

Walter, 66

Warren, Collins, 277

Wasserzug, 116

Watson, 62

Webster, J. C, 327

Weigert, C, on law of inertia, 55, 167 ;
on cell activities, 85; on cell
proliferation, 282

Weismann, A., against the inheritance
of acquirements, 6 ; his influence,
72 ; on biophores, 80 (note) ; on
continuity of the germ plasm, 83,
210 ; on variability in protozoa,
106 ; his theory of inheritance, 134,
136 et seq. ; on somatogenic in-
fluences on germ cells, 136; on
blastogenic inheritance, 158; on
the value of hypothesis, 160; the
" Reductio ad absurdum," 210 et
seq. ; on muscle spindles, 268 ; Sir
E. Ray Lankester on, 356

Welch, W. H., on adaptation by
bacteria, 43, 164

Whitman, 84

White, C. P., definition of true
tumours, 281 ; on microbic stimula-
tion of cell growth, 329 ; on pro-
gressive hypertrophy, 340

Willey, A., 18 (note) '

Williams, 310

Wilms, 335

372

THE STUDY OF EVOLUTION

Wilson, E. B., 139 ; on accessory chro-
mosome, 193; on regeneration of
centrosome, 195

Wolbach, on X-ray carcinoma, 248

Woolley, P. G., on tumours of adrenal
cortex, 321

Wright, Sir Almroth, on opsonins, 54

Wrosczek, 165

X-ray cancer, 248

Yeasts, thermogenic production of
races, 151 (note)

Zaslein, 127
Ziegler, 281
Zinsser, 258

THE END

Printed by R. & R. Clark, Limited, Edinburgh.