HEREDITY and MEMORY

J. WARD

H431
34

W:^:

(EliF Ell BUI Ctbrarg

5fnrtl| (EaroUna S'tatr
(EoUrg^

QHU31

: W3U

This book was presented by

Mrs. Elsa Winters March

lll'llliilllln ii'; mJ^.m"'!;. ,..?". ""-L LIBRARV

S00141967 "S

THIS BOOK IS DUE ON THE DATE
INDICATED BELOW AND IS SUB-
JECT TO AN OVERDUE FINE AS
POSTED AT THE CIRCULATION
DESK.

FEB 1 0 2006

HEREDITY AND MEMORY

CAMBRIDGE UNIVERSITY PRESS

ILonUon: FETTER LANE, E.G.

C. F. CLAY, Manager

©Hinfturgf): loo, PRINCES STREET

ISttlin: A. ASHER AND CO.

leipjig: F. A. BROCKHAUS

0eto gorfe: G. P. PUTNAM'S SONS

ISomiBS entJ Calcutta: MACMILLAN AND CO., Ltd.

All rights reserved

HEREDITY AND MEMORY

BY

JAMES WARD Sc.D.

PROFESSOR OF MENTAL PHILOSOPHY IN THE
UNIVERSITY OF CAMBRIDGE

BEING

THE HENRT SIDGWICK MEMORIAL LECTURE

DELIVERED AT

NEWNHAM COLLEGE, 9 NOVEMBER 1912

Cambridge :
at the University Press

a a HILL LIBRARY
N. C. STATE UNIVERSITY

CatnbrilJst :

PRINTED BY JOHN CLAY, M.A.
AT THE UNIVERSITY PRESS

HEREDITY AND MEMORY

Writing in the middle of the xviith cen-
tury on " the efficient cause of the chicken,"
William Harvey of Caius College in this
university quaintly remarked : " Although it
be a known thing subscribed by all... that the
egge is produced by the cock and henne, and
the chicken out of the egge, yet neither the
schools of physicians nor Aristotle's discern-
ing brain have disclosed the manner how the
seed doth mint and coin the chicken out of
the egge'' " How much nearer a disclosure,"
asks a writer who quotes this passage, "are
we to-day ?...0n the whole we have [still] to
confess that we do not know the secret of
development, which is part of the larger
secret of life itself^."

1 J. Arthur Thomson, Heredity, 1908, p. 416.

6 The processes of life

Now it is commonly taken for granted
that on this great problem — the problem of
Heredity — psychology can have nothing to
say. But I have come at length to think
that, provided we look at the world from
what I would call a spiritualistic and not from
the usual naturalistic standpoint, psychology
may shew us that the secret of heredity is to
be found in the facts of memory.

But first of all, in accordance with the
observation just quoted, it seems desirable
to enounce a few general propositions true
"of the larger secret of life itself" and
applicable also to this particular part of it.

To begin — we find the processes observ-
able in the world around us can all be ranged
in one or other of two classes, as either
anabolic processes or katabolic — to use in
a somewhat wider sense the terms of a
Cambridge physiologist. The former we take

imply guidance and direction 7

to imply action contrary to, the latter action
along, the line of least resistance. The
processes of the one order build up : those
of the other level down. The one order
implies that direction, in the sense of aim or
end, which we associate with mind as sensitive
and purposive; the other that indifference
which we associate with matter as lifeless
and inert.

Of such guidance or direction — I would
ask you next to note — we have immediate
experience only in the case of our own
activity, as in building a house or organising
a business. It may well seem rash therefore
to attribute such processes as the formation
of chlorophyll in a blade of grass or of albu-
men in a grain of corn to guidance in this
sense. But at all events they are processes
pertaining exclusively to living organisms,
and found nowhere else. If these processes

8 The characteristics of life

should some day be artificially repeated in
a laboratory, as Professor Schafer so con-
fidently expects — even this would imply the
guidance of the living chemist. But still, it
may be asked, what right have we to identify
life and mind; what right, for example, to
credit plants with souls, as Aristotle did?
The right that the principle of continuity
gives us. No sharp line can be drawn be-
tween plants and animals nor between higher
animals and lower.

But here the advocate of Naturalism may
intervene. "Continuity is just as complete
regarded from below as it is regarded from
above," he may urge ; " and if so, surely the
proper method of investigation is to begin
with the simpler and earlier rather than with
the later and more complex." Not necessarily,
we must reply : all depends — as Plato pointed
out long ago — upon where the characters we

clearest in its most advanced forms 9

have to study are clearest and most distinct.
In the case of life there can be no doubt that
this is where they are the most developed.
" It is clear," as G. H. Lewes once said, " that
we should never rightly understand vital
phenomena were we to begin our study of
life by contemplating its simplest manifesta-
tions... we can only understand the Amoeba
and the Polype by a light reflected from the
study of Man^" Moreover if we begin from
the material side we must keep to this side
all through : if matter is to explain life at all,
it must explain all life. But it is evidently
impossible to describe the behaviour of the
higher organisms in physical terms. Indeed
the ablest physicists recognise that the
concepts of physics are inadequate to the
description of life even in its lowest forms.
We may conclude then, that when, as in the

1 Problems of Life and Mind^ 3rd Series, 1879, p. 122.

10 The higher forms of life imply wider

case of life, we are seeking to interpret the
meaning of a continuous series we must start
where that meaning is clearest, where it is
best known and most definite, not where it
is least known and most inchoate.

Working in this fashion from ourselves
downward, what then, we may now ask, do we
find as the irreducible m^inimum which all
life implies ; and what are the most general
characteristics that mark every advance from
lower forms to higher ? To the first question
the answer, I trust, may here suffice that life
everywhere implies an individual and an
environment. To changes in this environ-
ment the individual's behaviour is, or tends
to be, so adjusted as to secure its well-being.
For — it is worth remarking by the way —
throughout the realm of life the category of
value, which such terms as well-being and
ill-being imply, is relevant, though irrelevant

and fuller adjustments to environment 11

everywhere else. It is this that gives to
what we have called guidance or direction
its motive and its meaning.

As to the second question — what are the
general characteristics of advancing life?—
this is on the whole admirably dealt with
in Herbert Spencer's Psychology under the
heading General Synthesis. Defining life as
the adjustment of internal relations to
external relations, he shews how, as evolution
progresses, this adjustment extends in range
both spatially and temporally; how at the
same time it increases in speciality and
complexity; how separate adjustments are
more and more co-ordinated and unified;
and so forth. The interval between the
palaeolithic man, whose world was bounded
by his river-valley, who knew little of the
past and planned less for the future, who
lived from hand to mouth content with raw

12 Social progress implies heredity

food and possibly no clothing, who possessed
only the rudest implements and had but a
cave for his dwelling — the interval, I say,
between him and civilised man with his long
traditions, fixed laws, innumerable arts and
organised division of labour, sufficiently
illustrates the character of this progress
when already far advanced. The inheritance
of the permanent achievements of one gen-
eration by the next is obviously the main
factor of such social progress : this we may
call heredity in the literal sense. But we talk
also of heredity as a factor in the biological
progress from the Protista up to Man;
though here the heir and the inheritance can
only be distinguished by calling the individual
the heir and his organism the inheritance,
that is to say by regarding as two what the
biologist conceives as one. It is this bio-
logical heredity that is our problem.

Individual progress implies habit 13

Before attempting to attack this problem
directly there is however still a characteristic
of life or experience that we may for a moment
consider, which seems to throw some light
upon it. Experience I once proposed to
define as the process of becoming expert by
experiment; and recent elaborate observa-
tions^ of the behaviour of the Protozoa shew
that these microscopic creatures frequently
succeed, as we do, only by way of trial and
error. Even plants prove capable to a great
extent of accommodating themselves to
changes in their environment. All this pre-
supposes a certain plasticity, which in turn
implies retentiveness : in other words, just
as later generations inherit from earlier
generations so later phases of the individual
inherit, as it were, from earlier phases. In

^ H. S. Jennings, Contributions to the Study of the Behaviour
oj the Lower Organisms, Washington, 1904.

14 Habit means that function

our own case we get a good deal of insight
into this process in what we can observe of
the growth of habit. What was originally
acquired by a long series of trials and
failures, engrossing all our attention, becomes
at length "secondarily automatic" — to use
Hartley's now classic phrase. Of this such
feats as skating or piano-playing are familiar
examples. This mechanization of habit is
aptly described in the saying that "use is
second nature." It sets attention free for
new advances which would else be impossible.
So natura naturata is the condition of further
natura naturans. The organism gives us a
warrant for the term " mechanization " in the
permanent modification of brain and muscle
which the acquisition of new dexterities en-
tails ; and mutatis mutandis, the same holds
true of the knowledge we know so thoroughly
that, as Samuel Butler said, we have ceased

perfects structure 16

to know that we know it at all. Now this
law of habit we may reasonably regard as
exemplified in the life of every individual
in the long line of genealogical ascent that
connects us with our humblest ancestors, in
so far as every permanent advance in the
scale of life implies a basis of habit embodied
in a structure which has been perfected by
practice.

And now, starting from the analogy just
noted — namely that habit connects successive
phases in the life of one individual as heredity
connects successive stages in the development
of one race — we may pass at length to our
problem. That analogy suggests the possi-
bility of an indefinite advance upwards in
the scale of life without the succession of in-
dividuals which heredity involves — provided,
of course, that a single individual lived
suflBciently long and did not grow old.

16 Continuous genesis without heredity

In place then of the innumerable individu-
als a certain genealogical ascent has succes-
sively entailed, let us imagine one individual
accomplishing the whole of it. The final
result as regards structure might conceivably
be substantially the same, nor need the time
required be very different. But there would
certainly be one very important difference.
For the solitary immortal without ancestry
structure would be wholly the result of
fimction. But for the many mortals — who
have a racial history as well as a personal
history — function would be the result of
structure, so far, that is, as the embryonic
stage of their existence is concerned. To
this stage Haeckel accordingly gave the name
of palingenesis ; because in it the structural
acquisitions of our imaginary immortal are
recapitulated. But whereas the same level
of development might have been attained in

Palingenesis with heredity 17

approximately the same time in the two
eases — that of a single persisting individual
and that of a continuous succession of
perishing individuals — the recapitulatory
process, peculiar to the second case, proves
to be vastly more rapid. It took thousands
of years, say, to produce the first chicken,
but the hen's q^^ reaches the same level in
three weeks. To be sure the recapitulation
is not precise and complete in every detail ;
yet most biologists, I believe, allow that it is
very full and substantial. Nevertheless even
in the most complex organisms far the
greater part of it is accomplished within a
year. How may we account for this extra-
ordinary brevity in the repetition of a process
that took so long at first ? The consideration
of this question in the light of our preliminary
investigation may lead us towards a solution
of our problem.

2

18 The recapitulatory process briefer

Let us suppose our imaginary individual,
when he had proceeded but a little way in
the slow and arduous fashion of a pioneer,
to have been set back to the beginning once
more, still however retaining the memory of
Ms former experiences. " We may be sure,"
as I have said elsewhere, " that in that case
he would make good the ground lost in much
less time than he required at first, and also
without following all the windings of the
tentative route into which his previous in-
experience had led him : his route the second
time would be routine^." So, for example, a
man, who had gradually by various improve-
ments adapted a house or a machine to his
purposes, would proceed, if either by some
accident was destroyed. Let us again sup-
pose that after a while, when a still farther
advance had been achieved, our imaginary

1 The Realm of Ends, 2nd edn, 1912, p. 209.

because it is all routine 19

individual had to submit to a like setback
once more; and that after a yet further
advance the same thing recurred again, and
so on indefinitely. The result would clearly
be that the latest acquisitions would be the
least straightened out, the least automatic
and the least fixed. Now we know directly
by observation that in like manner the
memories and dexterities that are acquired
latest are the least engrained and the most
likely to fail. And turning from the indi-
vidual to the race, Darwin has shewn that
here it is specific characters, which are
acquired later, rather than the generic, upon
which they are superposed, that are pecu-
liarly liable to variation.

Now the situation we have supposed
closely resembles that of a new complex
organism. Though not identical with the
old it is nevertheless continuous with it, and

2—2

20 Objections to the inheritance of

it does reproduce the ancestral acquisitions
with the less hesitation or variation, the more
they have become habitual, secondarily auto-
matic or organised. This no doubt implies —
prima facie, at least — that acquired characters
are inherited. But it implies also — a point
too often forgotten — that we should not ex-
pect any clear manifestation of such heredity
till the functions that have led to structures
have passed far beyond the initial stage
where conscious control is essential to their
performance.

This doctrine of the inheritance of ac-
quired characters, enounced, I believe, by
Aristotle, accepted alike by Lamarck and
Darwin, was only seriously called in question
some thirty years ago. But so rapidly has
opinion swung round, that now the great
majority of biologists — and especially of
zoologists — reject this hypothesis, as they

acquired characters so far assumed 21

call it, altogether. If they are right, it is
useless attempting to develope further the
psychological theory I have been trying to
suggest. At this point then we must pause
to examine their objections.

First, it is said, neither observation nor
experiment has so far yielded any really
decisive evidence for the old theory of in-
heritance. But it is equally true that they
furnish no conclusive evidence against it.
There are at present no crucial instances
either for or against it ; but as |I have just
said, we should not expect that there would
be. Even the theory of natural selection
was not established in that way: there too
the argument depends entirely on cumula-
tive evidence and general considerations. It
must be owned that a vast mass of worthless
cases of hereditary transmission has been

22 (i) No decisive evidence,

exploded by Weismann and his followers;
and this has naturally led to a general
distrust of the rest. Yet, as a singularly
fair-minded and acute biologist, Delage, main-
tains, the evidence is still formidable ; for as
another Lamarckian has said, "transforma-
tion... acts as if the direct action of the
environment and the habits of the animal
[the parent animal that is to say] were
the efficient cause of the change, and any
explanation which excludes the direct action
of such agencies is confronted by the diffi-
culty of an immense number of the most
striking coincidences ^"

Yes, the Weismannians reply, we allow
that appearances all point in the Lamarck-
ian direction, but inasmuch as the modus
operandi of the transmission is altogether

^ Professor W. B. Scott, American Journal of Morphology^
1891, p. 395.

(ii) the modus operandi inconceivable 23

inconceivable, we decline to believe that they
are more than appearances. This is their
second argument. But as the " impassioned
controversy," as Delage calls it, has gone on,
this argument, it is hardly unfair to say, has
changed its form. No process of transmis-
sion being conceivable, it is assumed that no
such process is possible. In this way the
Neo-Darwinians relieve themselves of the
difficulty there is on their side of proving a
negative. But assumption is not argument,
and our mere ignorance of the " how " alone
will not justify us in rejecting prima facie
evidence. We do not brush aside the facts
of gravitation because we are utterly ignorant
of the process which they involve.

We certainly are entirely in the dark
as to how structural changes in the body
of the parent can affect at all the structure
of the germ which it nourishes; since the

24 But there are actual operations

two are anatomically entirely distinct. But
it is something to the point to note, as
Cope did years ago, that there is at least
one case of a very precise connexion between
two distinct tissues which is perhaps quite
as wonderful as the connexion between
body-plasm and germ-plasm and hardly less
mysterious — viz. the adjustment of skin-
colouration to ground-surface brought about
through the organs of sight. Of this the
chameleon furnishes the most familiar but
not the most impressive instance. I came
the other day across an account of some
experiments that seem clearly to imply
the intervention in some way of conscious-
ness, in bringing about this adjustment —
an intervention which Cope surmised but
could not prove. Into a tank of flat fish,
whose colour matched its sandy bottom, a
number of pebbles of a different colour

almost as mysterious 25

were introduced. As seen by the fish the
mosaic so produced would appear more or
less foreshortened; but presently for all
that the fish became mottled like the bottom,
not as it appeared to them at rest, but as it
would appear to an observer looking down
from above, like the enemies the fish had
to elude. Also it is perhaps noteworthy that
there is at least a very intimate connexion
one way between body-plasm and germ-
plasm, as the extirpation of the reproduc-
tive glands shews.

But in the third place not only is the
inheritance of acquired characters said to
be without specific evidence, not only is it
declared inconceivable and so incredible
a p7'iori, but in what he calls "germinal
continuity " Weismann claims to have found
direct and positive proof that it is actually
impossible. At the same time he contends

26 (iii) Weismann's counter -doctrine

that the mingling of ancestral characters in
sexual reproduction is suflScient to provide
endless variations on which natural selection
can work, thus rendering it also needless.
Continuity of germ-plasm in some sense is
an obvious fact in any case ; and Harvey's
position, omne vivum e vivo, is one that no
biologist is concerned practically to deny.
What Weismann means by germinal con-
tinuity, however, is such an absolute conti-
nuity of the germ-plasm as entails its absolute
discontinuity from the body-plasm. Let the
changes acquired by the parents be what
they may, they can, he maintains, make no
difference to the offspring. The circular
character of the argument is here again
apparent. Till the impossibility of the
Lamarckian position is proved, germinal
continuity or continuity of the germ-plasm
may exist, but the absolute isolation of this

of Germinal Continuity 27

germ-plasm from the body-plasm is still open
to question; and apart from this isolation
the Lamarckian position remains tenable.
In fact Romanes, who has criticised Weis-
mann at length with much acumen, represents
him not as attempting to prove but as simply
''postulating'' the absolute non-inheritance
of acquired characters and deducing the
absolute continuity of the germ-plasm in
his sense from this.

A brief examination of Weismann's main
position and a word or two on an important
supplementation, to which he was eventually
driven, may I trust suffice to bring us round
again to the theory of " organic memories "
or "engrams^" to which our preliminary

1 This theory was first definitely broached by Professor Ewald
Hering in a lecture, Concerning Memory as a general function
of Organized Matter, delivered at Vienna in 1870. It has
recently been revived and developed by Professor R. Semon of
Munich (to whom the word ' engram ' is due), in Die Mneme ah
erhaltendes Prinzip im Wechsel des organischen Geschehens,
2te Auf. 1908.

28 Natural Selection and Amplimiixis

propositions led up. In order to see more
clearly the issue raised, we may begin by
comparing our imaginary immortal with some
mortal double, who has reached the same level
of biological development within a few short
years. The bodies of both are related to
the simplest form of life — the one directly
the other indirectly — the one having de-
veloped from such a form by continuous
interaction with its environment, the other,
according to Weismann, having developed
from a continuous stock of germ-plasm,
entirely cut off from the environment since
the unicellular stage. Now in both cases
there is an "immortal" concerned — our im-
aginary individual in the one case, and
Weismann's hypothetical germ-plasm in the
other. Also the result attained is in both
cases the same ; and certainly it is the most
wonderful instance we know of what we call
an end, a Ti\Q%. The fundamental factors in

here replace experience 29

the one ease are (1) the environment, and
(2) experience: the latter an essentially
positive or teleological factor, since it is
always directed towards self-conservation or
betterment. But the factors fundamental in
the other case are (l) natural selection, and
(2) amphimixis or the periodical blending of
two ancestral strands of the immortal germ-
substance ; and both alike are non-teleo-
logical.

Can it be that one and the same end can
be reached by such disparate means ? Its at-
tainment through experience without natural
selection is conceivable : its attainment with-
out experience, by natural selection and for-
tuitous variation alone, is surely inconceivable.
That both experience and natural selection
have co-operated seems indeed to be the fact,
as Darwin himself at all events assumed. But
this co-operation Weismann with Wallace

30 Discontinuity from body-plasm the

and the rest of the Neo-Darwinians denies.
And this denial it is that entitles us to say
that d'iscontinuity, or isolation from the body-
plasm, is the essential feature of Weismann's
germ-plasm. This germ-plasm persists indeed
continuously ; but it persists, so to say, under-
ground, screened like the sacred queen-bee^
from all the functioning with the external
world which the offsets it periodically throws
out discharge. Nevertheless it alone — not
intercourse with the environment — is sup-
posed to determine their structure. Can the
manifold adaptations such structures display
and the endlessly diversified series of them —
continuously advancing in complexity of
adaptation — which the evolution of life as a
whole displays, can all these, we ask again,
be explained from such a standpoint?

1 An illustration used by Dr Francis Darwin in his Presidential
Address to the Brit. Assoc. 1908.

essential feature of Weismann' s germ-plasm 31

Yes, they can, was Weismann's answer:
which, however, as I said, he afterwards
quahfied. Any detailed examination of his
theory would require a volume, has in fact
occupied many volumes. But the disconti-
nuity just referred to — duly considered —
suffices, I think, summarily and yet success-
fully to dispose of it. Let me try to make
this clearer. First of all I must ask you to
note that, so long as we confine our attention
to the unicellular organisms, there is no
special problem as to the inheritance of
acquired characters, nor in fact as to in-
heritance at all ; for, as Weismann himself
has shewn, there is here no natural death, and
therefore — strictly speaking — no succession
of generations. In his own words, " Natural
death occurs only among multicellular or-
ganisms, the single-celled forms escape it."
Eeferring to the well-known cell-division by

32 Experience the determining factor

which the one individual becomes two, he
remarks : " In the division the two portions
are equal, neither is the older nor the younger.
Thus there arises an unending series of in-
dividuals, each as old as the species itself, each
with the power of living on indefinitely, ever
dividing hut never dying^T

And now, in the next place, "it should
also be borne in mind that," as one of
Weismann's ablest supporters has observed,
"many of these unicellular organisms... are
highly differentiated, i.e. [are provided] with
great complexity of structure... and that
many have very definite and interesting
modes of behaviour, such as swimming in
a spiral, seeking light or avoiding it... trying
one kind of behaviour after another— func-
tional peculiarities, some of which cannot
be described without using psychological

^ Italics mine.

in the development of Unicellulars 33

terms ^." Lowly as these forms of life appear
when compared with our own, still the
enormous diversity to be foimd among them,
their wide range in space and their prodigious
antiquity, together suggest that between the
highest of them and the very beginnings of
life lies a long, long history of " one kind of
behaviour after another." In the course of
that history functions were mechanized and
structures fixed.

All this while the situation is that of our
imaginary immortal developing by intercourse
with its environment, save that here the im-
mortal is real and not imaginary; and also,
instead of remaining solitary, is continually
manifolding itself into new individuals who
start equipped with all its acquisitions. All
this while too the whole organism of every
individual is in touch with the environment

^ J. Arthur Thompson, Heredity, p. 33.

34 The devdopmeiit of Multicellulars

and every new acquisition is passed on to
all the individuals into which any one is
manifolded. If we like to call this inheri-
tance, then it is inheritance of acquired
characters, and there is no other. And all
this while too there is no ground for dis-
tinguishing between body-plasm and germ-
plasm; unless the cell nucleus — without
which the cell never divides and which
always divides with it — were to be called
germ-plasm on this accoimt.

But now so soon as we advance to the
Metazoa and the MetapJiyta — that is to the
multicellular animals and plants — we are
supposed, according to Weismann, to be
suddenly confronted with an absolute dis-
continuity of mortal body and immortal
germ. The one can no longer bequeath,
the other can no longer inherit. Surely in
such a case what we should naturally expect

not depeiide^it on experience 35

would be not the enormous advance from
jelly-fish to mammal, which is what we find,
but rather a practically stationary state.
For consider the parallel case of social pro-
gress referred to earlier. The main factor
in such progress we found was heredity.
Imagine then, if you can, what would happen
if now from this time forth every new gene-
ration had to begin where the old began and
not where it left off"; if no single human
product from now onwards outlasted the
individual who produced it ; if in short all
tradition and inheritance were from hence-
forth no more. Such a breach of social
continuity between the future and the past
would surely be startling; and yet it is
an even greater discontinuity that Weismann
seemed to imagine as marking off* the Pro-
tozoa from the Metazoa. The Lamarckian
factor good up to that point has, in his

3—2

36 Amphimixis as a mingling

view, been absolutely inoperative beyond it.
Natural selection and the blending of an-
cestral traits, more or less diverse, have
henceforth sufficed.

The essentials of this process of amphi-
gony or mixed generation, as Weismann
conceived it, are simple enough. Two packs
of shuffled cards he has himself suggested
as representing the paternal and maternal
germ-plasm. Now imagine half of each
rejected and a new pack made of the re-
mainders. A portion of this new pack will
remain unaltered, though it will grow in-
definitely: this portion is the latent germ-
plasm that continues undeveloped until
the advent of a new generation, when the
processes of mixing, shuffling, and reducing
division are repeated. From the remaining
portion the new individual arises. Any one
of the cards would suffice for its complete

of ancestral germ-plasms Zl

development, for each corresponds to a single
ancestor. But there are many of them,
and so a struggle ensues. The card or plasm
of one ancestor — a paternal grandmother
say — succeeds in shaping the nose : another
— perhaps a maternal great-grandfather —
provides the mouth, and so on. Though
any one of them might conceivably have
furnished the whole, it is far more likely
that all or most of the pack have a share
in it. In all this — the mingling of ancestral
plasms, the reducing division, the reversion
and atavism — we are, so far, in the region
of fact. But when Weismann asks us to
regard these ancestral plasms as themselves
quite stable, we enter the region of fable.
And even in this region, if we try to picture
out — not everyday instances of heredity —
but the primitive heredity, the link, that is
to say, which long ago connected the

38 The only ancestral plasms so far

evolution of mortal multicellular organisms
with the wholly distinct evolution of immortal
unicellular organisms, we are utterly at a
loss to find any resemblance between the two.
No doubt amphimixis is a potent source
of congenital variations, and such variations
are the indispensable 2?om< cTappui for natural
selection. But what avails that if all possible
variations are confined to such unit-cha-
racters as the unicellular organisms display ?
As Henry VIII long ago remarked: "You
can't make a silk purse out of a sow's ear,"
turn it how you will. And how, in view of
the stability of germ-plasm and its discon-
tinuity from body-plasm — which, according
to Weismann, are maintained throughout
multicellular evolution — the higher levels of
life were reached is a mystery. For, as De-
lage has said : — " Without the inheritance of
acquired characters there can be no new

those of the unicellular organisms 39

ancestral plasms, and without ancestral
plasms more complicated than those of the
Protozoa there can be none of the superior
animals ^" But Weismann rejects the in-
heritance of acquired characters and so cuts
oflF the first possibility : also he maintains —
or rather began by maintaining — the absolute
continuity and isolation of the germ-plasm,
and so cut off the second.

Simultaneously with Delage's criticism
however — in 1895 that was — Weismann made
the complete change of front to which I
have just alluded, abandoning — or at least
essentially modifying — both his main posi-
tions, viz. the stability of germ-plasm and

1 Yves Delage, L'Heredite et les grands Problem es de la
Biologic genSrale, 1903, p. 560. But to Professor Hartog of
Cork belongs the credit of first conceiving "cette objection
capitale sous la forme du dilemme," as Delage expressly allows.
Op, cit. p. 559 n. See Professor Hartog's letter in Nature,
vol. xliv, 1891, entitled, "A Difficulty in Weismannisra," pp. 61 3 f.,
and a second letter, vol. xlv, pp. 102 f.

40 Weismann evades a dilemma

its discontinuity from body-plasm. In the
same year, H. P. Osborn^ had said: "If
acquired variations are transmitted there
must be some unknown principle in heredity,
if they are not transmitted there must be
some unknown factor in evolution." "A
perfectly correct conclusion" Weismann at
once replied, and set to work to find the new
factor by the simple method of extending
the range of natural selection to the ultimate
constituents of the germ-plasm itself He
supposed that the drama of the world without
is here repeated on a minute scale. The
determinants or ultimate constituents of the
germ-plasm struggle with each other for
nutriment. Some at length succumb and
the successful survivors are thus "selected."
So variations arise within the germ itself,

1 In a lecture entitled The Hereditary Mechanism and the
Search for the Unknown Factors of Evolution.

by assuming " intra-germinal selection " 41

followed of course by corresponding varia-
tions in the body that they eventually build
up. This, his latest theory, the theory of
intra-germinal selection, has practically con-
vinced nobody; though it has been hailed
by opponents of Darwinism as sounding the
knell of the theory of natural selection ^
Why, even if we allow the propriety of
regarding Weismann's hypothetical deter-
minants as organisms struggling for nutri-
ment— and facts seem to be against it^ —

* "With a 'rehabilitation' of natural selection in the real
Darwinian meaning and only fair application of the phrase the
new theory has nothing whatever to do. It is, much more,
a distinct admission of the inadequacy of natural selection to do
what has long been claimed for it." Kellogg, Darwinism to-day ^
1907, p. 199.

2 "Actual experimentation on the influence of food-supply in
development does not bear out the assumption on which the
theory of germinal selection rests. Weisraann himself gave the
larvae of flies... an abnormally small food supply..., with the only
result that the mature individuals were dwarfed ; that all their
parts were reduced in size, but the actual proportions... were

42 His whole position is thus surrendered

why, even then, it mu^st follow that useful and
consentient variations will appear just when
and where they are wanted, is more than
Weismann has seriously attempted to prove ;
though he confidently asserts that it is so.
What is important about his new theory
however is the surrender both of the an-
cestral continuity and also of the somatic
discontinuity of the germ-plasm, a surrender
that, as Delage and many others have urged,
undermines his whole position. In short,
while the ground on which was based his
direct and positive proof of the impossibility
of the inheritance of acquired characters is
abandoned, his full and definite admission
of the need for some equivalent of that
Lamarckian factor remains. We may then
now resume our consideration of what I have

unchanged." Kellogg, Op. cit. p. 201. But of. especially Plate,
Die Bedeutung des Darwinischen Selectionsprincipes, 2te Auf.
1903, pp. 164-70.

arui we resume our inquiry 43

called the psychological or mnemic theory
of heredity. Obviously we muct begin from
the biological side, for we have no direct
experience in the matter. Moreover it is
not our memory, but a so-called organic
memory with which we are concerned.

Now if it is once admitted that the body
influences the germ-plasm in one way — i.e,
by way of nutrition — the possibility of its
influence in other ways can hardly be denied.
Even the influence in the one way may mean
a great deal more than Weismann surmised,
may involve not merely quantitative differ-
ences but qualitative differences as well.
Truly, as Mephistopheles said : " Blut ist ein
gam hesondrer Saftr For example, through
the blood there circulate certain secretions,
called hormones, destined — as our Regius
Professor of Medicine has said — "for the

44 The role of the nervous

fulfilment of physiological equilibrium."
"Thus," as he goes on to say, "the reci-
procity of the various organs, maintained
throughout the divisions of physiological
labour, is not merely a mechanical stability,
it is also a mutual equilibration in functions
incessantly at work on chemical levels, and
on those levels of still higher complexity
which seem to rise as far beyond chemistry
as chemistry beyond physics^.

But now the most striking instance of the
equilibration of its functions that an organ-
ism displays is that which brings about the
adjustment of internal relations to external
relations. In this adjustment indeed we may
say, as I have already urged, that life essen-
tially consists. But this adjustment in the
higher organisms, where the characteristics

^ Sir T. Clifford Allbutt, Ency. Brit.^ vol. xviii, art. Medicine,
pp. 57 f.

system and its retentiveness 45

of life are most distinct, is directed and sus-
tained by a nervous system. Now to this
system belongs in a pre-eminent degree that
retentiveness and modifiability which life and
experience everywhere imply — characters
which we figuratively express by reference
to plasticity, as in such terms as protoplasm,
idioplasm, germ-plasm and the like. Plas-
ticity, in a word, though most pronounced in
the higher organisms, where the physiological
division of labour has developed a distinct
nervous system, is present in all organisms :
all alike, though in divers degrees, live and
learn. But now a multicellular organism, it
is generally allowed, originated in, and still
consists of, a colony of unicellular organisms
— a colony that has become more of a
commonwealth the further its functional
division of labour has advanced. It is im-
portant then to insist on two points. First,

46 The role of the cell-nucleus

every living cell, whether living in isolation or
as a member of a complex organism, must be
credited with that " organic memory " which
all life implies. Next, we can set no limit to
the consentient interaction between such
cells when they have become "members one
of another." " Even in normal circumstances "
— to quote Sir Clifford AUbutt once more —
"their play and counterplay, attractive and
repellent, must be manifold almost beyond
conception."

Now the reactions of the body-plasm in
the higher organisms are, as said, guided
mainly by the nervous system; and there,
we know, facility and familiarity become
automatic, more or less "unconscious"
through repetition. But in the germ-plasm
the role of regulating ontogeny, it is allowed,
belongs mainly, if not solely, to the cell-
nuclei or chromatin, Weismann's "germinal

analogous to that of the brain 47

substance." " May we not therefore consider
it probable that the nucleus plays in the cell
the part of a central nervous system ? " was
the question raised by Dr Francis Darwin
in his masterly Presidential Address to the
British Association in 1908. I am aware of
no smallest detail in which the analogy
between the two fails, and — in my opinion
at all events — Dr Darwin was amply justified
in contending, as Hering in his classic paper
had done before him, "that ontogeny — the
building up of the embryo — is actually and
literally a habit."

Further, as Dr Darwin goes on to remark,
this so-called mnemic theory is "strongest pre-
cisely where Weismann's views are weakest —
namely in "giving a coherent theory of the
rhythm of development." The prodigious
complication of Weismann's germ-plasm with
its idants ids, determinants and biophores —

48 The mnemie theory:

a vast army without a general — only spells
confusion, when we try to picture it in
motion, a plight like that of the builders
of Babel after the confusion of tongues.
According to the mnemie theory, on the
other hand, the germ-cell is a definite unity,
the counterpart of the structural alterations
wrought by habit in the parental organisms
with which it has been in sympathetic rap-
port all along. It is potentially what Leibniz
would have called an automate spirituel ou
formel, the latent entelechy of a future
organism. We may compare it to a com-
pany of actors awaiting behind the scenes
the call to begin their play. Each one knows
his part by heart and also knows his cue.
The routine or orderly rhythm of the per-
formance is thus ensured and the play,
though continually condensed at one end
and extended at the other, has been so often

ontogeny a habit 49

repeated as to be acted without hitch or
hesitation save perhaps in the cases of its
latest amendments.

And now two or three last remarks by
way of summary and conclusion.

No doubt when we try to ascertain the
details of this process the difficulties in our
way are, as Dr Francis Darwin candidly
allows, " of a terrifying magnitude." But in
all exploration the first thing is to secure, if
possible, a general survey, a bird's eye view
of the whole. If we are to see the wood,
we cannot be among the trees. Now it is
entirely with this preliminary problem that
we have been concerned ; and as befits such
an inquiry we have tried — not to scrutinize
the details of life — but to look at it as a
real, concrete whole. Such a whole implies
continuity: absolute breaks are impossible.
Leibniz's maxim, Natwa nan facit saltus,

4

50 The way of experience the one

is ours too. We find then no ground for
separating organic life from psychical life:
for us all life is experience. We cannot
therefore assume that experience has no
part in the building up of the organism,
and only begins when a viable organism is
already there. For us, ontogeny and here-
dity are aspects of a single process — a
\^ process that only experience will explain.
Again, the principle of continuity forbids
us to assume that this process, by which
an organism is built up, abruptly changes
when we pass from unicellular organisms to
multicellular. The way of trial and error
and eventual success — function determining
structure — followed in the earlier stages of
the progress of life — in phylogeny, as it is
technically teniied — has equally been the way
of its progress ever since. Every organism
has proceeded from an organism ; yet among

way tlirougliout all life 51

unicellular organisms though there has been
progress there has been no genealogy. Among
multicellular organisms we find both: while
the offspring is still unicellular at the start,
the parents from which it sprang are uni-
cellular no more. The greater this difference
in complexity between the offspring and its
parents, the longer the way its common an-
cestors will have traversed phylogenetically
or historically, and the longer too the way
that it will have to traverse ontogenetically
or automatically, before attaining to the
parental level and beginning a pergonal
history of its own.

This pre-natal, so to say, prehistoric life,
is called a heritage. And why? Because,
looking broadly at the whole record of mul-
ticellular genealogies, there appears to be
everywhere more or less correspondence and
nowhere a positive deviation between the two

52 What habit is for the iyidividual

itineraries — let us say — the historical and the
automatic, the route of the original struggle
and the routine of its recapitulation after
many repetitions. But the more repetitions
the more fixity. ^In short, what habit is for
individual life that is heredity for racial life.
But the one lags enormously behind the
other: the repetitions that will suffice to
make a habit automatic for a lifetime are
very far from sufficing to ensure heredity
for future generations. And the more ad-
vanced the race, the farther heredity lags
behind acquisition. At the unicellular stage,
they are on a par, in other words there is as
yet neither distinction nor interval between
body and germ. At the multicellular stage
there is both, and so, as the scale of life
rises farther, the greater becomes the dis-
parity between the still unicellular germ
and the mature organism of ever increasing

heredity is /or the race 53

complexity; the greater therefore is the
number of intermediaries through whose
hands — so to say — the acquisitions of the one
have to pass before they can be imparted
to the other. And not only will these engrams
— as they are called — be fainter on this ac-
count, and so require more repetitions to give
them any permanence ; but also for the same
reason they will lose in definiteness and detail.
Language, for example, will be inherited
not as speech but as mere tongueiness
and babble; art not as technical skill
but as mere handiness or pure mischief;
and so on. But if there is evidence of such
inheritance of forms of behaviour impossible
of acquisition at the unicellular stage, can
we say that the continuity between body-
plasm' and germ-plasm ceased then? Can
we at all understand such facts without
recognising this continuity still?

54 The mnemic theory presupposes

But what exactly is this continuity be-
tween body and germ, and how are new
acquisitions passed on? The continuity is
what it always was, the continuity of mem-
bership in a commonwealth, where the whole
is for the parts and the parts for the whole ^ ;
where all are more or less en rapport. The
key to all this is to be found, I believe, in
social intercourse, not in physical trans-
mission. Unhappily however — as it seems
to me — most of those who uphold the
mnemic theory of heredity seem to hanker
unduly after a physical explanation of the
modus operandi, Hering and Haeckel
talked of peculiar vibrations : if they were
writing to-day they would probably refer to
wireless telegraphy, as Professor Dendy has

1 A Protozoan, we must remember, though unicellular, is far
from simple, and an absolutely simple form of life is beyond
our ken.

not only records hut minds 55

recently done^. But it is meaningless to talk
of memory unless we are prepared to refer
it to a subject that remembers. Records or
memoranda alone are not memory, for they
presuppose it. They may consist of physical
traces ; but memory, even when called " un-
conscious," suggests mind; for, as we have
seen, the automatic character implied by
this term " unconscious " presupposes fore-
gone experience. But it is possible that a
subject may impart his knowledge or dex-
terity to another without dragging his pupil
through all the maze of blundering which
his own acquisition cost. The mnemic theory
then, if it is to be worth anything, seems to me
clearly to require not merely physical records

^ Outlines of Evolutional Psychology^ 1912. Signer B.
Rignano, founder of the international review Scientia, in particular
has developed a very elaborate theory of " organic memory " to
explain the inheritance of acquired charactei-s, Sur la transmis-
sibilite des caracteres acquis^ Paris, 1906.

56 Pampsychism

or "engrams" but living experience or tra-
dition. The mnemie theory will work for
those who can accept a monadistic or pam-
psychist interpretation of the beings that
make up the world, who believe with Spinoza
and Leibniz that "all individual things are
animated albeit in divers degrees." But
quite apart from difBculties of detail, I
do not see how in principle it will work
otherwise.

CAMBRIDGE : PRINTED BY JOHN CLAY, M.A. AT THE UNIVERSITY PRESS.

^'fe

'Y^i

•"^itVi

i^

4>n^