such transmission, if possible at all, must occur a certain number of
times in a certain number of cases: it must occur more readily when
both parents are mutilated in the same way, or when the mutilation has
been repeated in many generations, etc. It is extremely improbable that
it would suddenly occur in a case where it was least expected, while it
did not occur in 900 cases of the most favourable kind. Those who
recognise in the doubtful cases of transmission of a single mutilation
present in only one of the parents, proofs of the existence of the
disputed operation of heredity, quite forget that the transmission
presupposes a very marvellous and extremely complex apparatus which if
present at all ought, under certain conditions, to become manifest
regularly, and not only in extremely exceptional cases. Nature does not
create complex mechanisms in order to leave them unused: they exist by
use and for use. We can readily imagine how complex the apparatus for
the transmission of mutilations or acquired characters generally must
be, as I have tried to show in another place. The transmission of a
scar to the offspring e. g. presupposes first of all that each
mechanical alteration of the body (_soma_) produces an alteration in
the germ-cells: this alteration cannot consist in mere differences of
nutrition, only affecting an increased or decreased growth of the
cells: it must be of such a kind that the molecular structure of the
germ-plasm would be changed. But such a change could not in the least
resemble that which occurred at the periphery of the body in the
formation of the scar: for there is neither skin nor the preformed germ
of any of the adult organs in the germ-plasm, but only a uniform
molecular structure which, in the course of many thousand stages of
transformation, must tend to the formation of a soma including a skin.
The change in the germ-plasm which would lead to the transmission of
the scar, must therefore be of such a kind as to influence the course
of ontogeny in one of its later stages, so that an interruption of the
normal formation of skin, and the intercalation of the tissue of the
scar, would occur at a certain part of the body. I do not maintain that
equally minute changes of the germ-plasm could not occur: on the
contrary, individual variation shows us that the germ-plasm contains
potentially all the minutest peculiarities of the individual; but I
have in vain tried to understand how such minute changes of the
germ-plasm in the germ-cells could be caused by the appearance of a
scar or some other mutilation of the body. In this respect I think that
Blumenbach’s condition is nearly fulfilled: he was inclined to declare
himself against the transmission of mutilations, but only if it were
proved that such transmission was _impossible_. Although this cannot be
strictly proved, it can nevertheless be shown that the apparatus
presupposed by such transmission must be so immensely complex, nay! so
altogether inconceivable, that we are quite justified in doubting the
possibility of its existence as long as there are no facts which prove
that it _must_ be present. I therefore do not agree with the recent
assertion[302] that Blumenbach’s condition cannot be fulfilled to-day,
just as it was impossible at the time when it was first brought
forward. But if nevertheless such a mysterious mechanism existed
between the parts of the body and the germ-cells, by means of which
each change in the former could be reproduced in a different manner in
the latter, the effects of this marvellous mechanism would certainly be
perceptible and could be subjected to experiment.
But at present we have no evidence of the existence of any such
effects; and the experiments described above disprove all the cases of
the supposed transmission of single mutilations.
Of course, I do not maintain that such cases are to be always explained
by want of sufficient observation. In order to make my position clear,
I propose to discuss two further classes of observations. First of all,
there are very many cases of the apparent transmission of mutilations
in which it was not the mutilation or its consequences which was
transmitted, but the predisposition of the part in question to become
diseased. Richter[303] has recently pointed out that arrests of
development, so slight as to be externally invisible, frequently occur,
and that such arrests exhibit a tendency to lead to the visible
degeneration of parts in which they occur, as the result of slight
injuries. Since therefore the predisposition towards such arrest is
transmitted by the germ—occasionally even in an increased degree—the
appearance of a transmitted injury may arise. In this way Richter
explains, for instance, the frequently quoted case of the soldier who
lost his left eye by inflammation fifteen years before he was married,
and who had two sons with left eyes malformed (microphthalmic).
Microphthalmia is an arrest of development. The soldier did not lose
his eye simply because it was injured, but because it was predisposed
to become diseased from the beginning and readily became inflamed after
a slight injury. He did not transmit to his sons the injury or its
results, but only microphthalmia, the predisposition towards which was
already innate in him, but which led in his sons from the beginning,
and without any obvious external injury, to the malformation of the
eye. I am inclined to explain the case which Darwin in a similar manner
adduced, during the later years of his life, in favour of the
transmission of acquired characters, and which seemed to prove that a
malformation of the thumb produced by chilblains can be transmitted.
The skin of a boy’s thumbs had been badly broken by chilblains
associated with some skin disease. The thumbs became greatly swollen
and remained in this state for a long time; when healed they were
malformed, and the nails always remained unusually narrow, short, and
thick. When this man married and had a family, two of his children had
similarly malformed thumbs, and even in the next generation two
daughters had malformed thumbs on both hands. The case is too
imperfectly known to admit of adequate criticism; but one may perhaps
suggest that the skin of different individuals varies immensely in its
susceptibility to the effects of cold, and that many children have
chilblains readily and badly, while others are not affected in this
way. Sometimes members of the same family vary in this respect, and the
greater or less predisposition towards the formation of chilblains
corresponds with a different constitution of the skin, in which some
children follow the father and others follow the mother. In Darwin’s
instance a high degree of susceptibility of the skin of the thumb was
obviously innate in the father, and this susceptibility was certainly
transmitted, and led to the similar malformation of the thumbs of the
children, perhaps very early and after the effect of a comparatively
slight degree of cold[304].
The last class of cases which I should wish to consider, refer to
observations in which the mutilation of the parent was certain, and in
which a malformation similar to the mutilation had appeared in the
child, but in which exact investigation shows that the malformations in
parent and child do not in reality correspond to each other.
In this class I include an instance which has only become known during
the present year (1888), and which has been observed as exactly as
possible by an anthropologist and physician, who has worked up the
history of the case. Dr. Emil Schmidt communicated to this year’s
meeting of the German Anthropologists’ Association at Bonn a case which
indeed seems at first sight to prove that mutilations of the human ear
can be transmitted. As Dr. Schmidt has been kind enough to place at my
disposal all the material which he collected upon the subject, I have
been able to examine it more minutely than is generally possible in
such cases; and I will discuss it in detail, as it seems to me to be of
fundamental importance in the history of human errors upon this subject.
In a most respectable and thoroughly trustworthy family, the mother
possesses a cleft ear-lobe upon one side. She quite distinctly
remembers that when playing, between the ages of six and ten years,
another child tore out her ear-ring, and that the wound healed so that
the cleft remained. Later on a new hole was made in the posterior part
of the lobe. She had seven children, and the second of these, who is
now a full-grown man, has a cleft ear-lobe on the same side as the
mother. It is not known whether the mother possessed an innate
malformation of the ear before it was mutilated, but, judging from the
present appearance of the ear, this is extremely improbable.
Furthermore, the existence of an innate cleft in the ear-lobe has never
been previously observed. The parents of the mother did not possess any
malformation of the ear. The conclusion seemed to be therefore
inevitable that the transmission of an artificial cleft in the ear-lobe
had really taken place.
But we must not be too hasty in forming an opinion. When we compare the
figures I. and II., representing the two ears, we are first of all
struck by the fact that the malformation of the ear of the son has an
entirely different appearance from that of the mother. The ear-lobe of
the latter is quite normally formed; it is broad and well-developed,
and only exhibits a median vertical furrow which is the result of the
mutilation. The ear-lobe of the son, on the other hand, is extremely
minute, one might even maintain that it is completely wanting. In my
opinion a cleft is not present at all, but the far higher posterior
corner of the ear forms the end of its posterior margin—the so-called
helix. But even if another opinion were pronounced with regard to the
interpretation of this part, there is one other circumstance to be
taken into account, which appears to me to be absolutely conclusive,
and which completely excludes the interpretation of this malformation
as the transmission of a mutilation.
[Illustration: FIG. I. _H._ Helix. _Cr. Ah._ Crura anthelicis._ Ah._
Anthelix. _Cch._ Concha. _Hl^1_. and _Hl^2_. Holes 1 and 2 for
ear-rings. _Lob._ Ear-lobe. _Sp. H._ Spina helicis. _Inc._ Incisura
intertragica. _Tr._ Tragus. _Atr._ Antitragus.]
If we compare the ears with each other, that of the mother with that of
the son, not only the anatomist but every trained observer will at once
be struck by the fact that they are totally different in their outlines
as well as in every detail. The upper margin of the ear is very broad
in the mother, in the son it is quite pointed: the so-called _crura
anthelicis_ are normally developed in the mother, in the son they can
hardly be distinguished and open in an anterior direction, while in the
mother they are directed upwards. The concha itself, the _incisura
intertragica_, in short everything in the two ears, is as different as
it can possibly be in the ears of two individuals.
But this fact obviously proves that the son does not possess the ear of
his mother, but probably that of his father or grandfather.
Unfortunately the father and grandfather have been now dead for a long
time, so that we cannot obtain certain evidence upon this point. At all
events, the son does not possess the ear of his mother, and it would be
very rash to suppose that he has inherited the ear from the father, but
the malformation of the ear-lobe from the mother—a malformation which,
as it seems to me, is certainly quite different from that of his
mother’s ear. I said that this case was of fundamental importance
chiefly because it shows very distinctly, on the one hand, how
difficult it is to bring together the material which is absolutely
necessary for the correct understanding of a single case, and on the
other hand, how carefully the abnormalities must be compared and
examined if we wish to escape wrong conclusions. Such precautions have
hitherto been rarely taken with the necessary accuracy; people are in
most cases satisfied with the knowledge that an abnormality is present
in the child on the same part which had been malformed by mutilation in
the parent.
But if we are to speak of the transmission of a mutilation, it must be
shown, before everything else, that the malformation of the child
corresponds precisely to the mutilation of the parent.
For this reason the older observations upon this subject are, in most
cases, entirely valueless.
The readiness with which we may be deceived is shown by the fact that I
myself nearly became a victim during the past year (1888). A friend of
mine, in order to convince me of the transmission of mutilations,
called my attention to a linear scar on his left ear, which extended
from the upper margin of the helix for some distance upon the posterior
part of the anthelix, giving it the appearance of a small, rather sharp
ridge. The scar had been caused by a cut from a duelling sword, which
the gentleman had received during his residence at the University.
Strangely enough, the left ear of his daughter, who is five years old,
exhibits a similar peculiarity. The posterior part of the anthelix
forms a rather sharp and narrow ridge like that of the father, although
the scar itself is wanting.
I must admit that I was at first rather puzzled by this fact, but the
mystery was soon solved in a very simple manner. I asked the father to
show me his right ear, and I then saw that this ear possessed a similar
ridge on the posterior part of the anthelix. Only the scar was absent,
which in the left ear brought the crest of the ridge into still greater
prominence. The ridge was therefore only an individual peculiarity in
the formation of the ear of the father,—a peculiarity which had been
transmitted to one ear of the child. No transmission of the mutilation
had taken place.
In the same manner, many of the so-called proofs of the transmission of
mutilations would be shown, by a careful examination, to be deceptive.
We must not expect to succeed in all of them, for in most cases the
investigation cannot be completed, chiefly because the condition of the
part in question in the ancestors is not known or is only known in an
insufficient manner. This is the reason why fresh examples of such
so-called proofs continue to appear from time to time,—proofs which do
not admit of a searching criticism because something, and in most cases
very much, is invariably wanting. But it will be admitted that even a
very large number of incomplete proofs do not make a single complete
one. On the other hand, it may be asserted that a single instance of
coincidence between a mutilation in the parent and a malformation in
the offspring, even if well established, would not constitute a proof
of the transmission of mutilations. Not every _post hoc_ is also a
_propter hoc_. Nothing illustrates this better than a comparison
between the ‘proofs’ which are even now brought forward in favour of
the transmission of mutilations and the ‘proofs’ which supported the
belief in the efficacy of so-called ‘maternal impressions’ during
pregnancy, a belief which was universally maintained up to the middle
of the present century. Many of those ‘proofs’ were simply old wives’
fables, and were based upon all kinds of subsequent inventions and
alterations. But it cannot be denied that there are a few undoubtedly
genuine observations upon cases in which some character in the child
reminds us in a striking manner of a deep psychical impression by which
the mother was strongly affected during pregnancy.
Thus a trustworthy person told me of the following case. A well-known
medical authority cut his leg above the ankle with a knife: his wife
was present at the time and was much frightened. She was then in the
third month of pregnancy: the child when born was found to have an
unusual mark upon the same place above the ankle. People almost forget
nowadays the tenacity with which the idea of maternal impressions was
kept up until the middle of this century; but it is only necessary to
read the received German text-book on physiology of fifty years ago,
viz. that of Burdach, in order to be convinced of the accuracy of this
statement. Not only does Burdach give a number of ‘conclusive’ cases in
man and even in animals (cows and deer), but he also attempts to
construct a theoretical explanation of the supposed process. This is
undertaken in the following manner,—‘Imagination influences the
function of organs;’ but the function of the embryo is the ‘tendency
towards development, and hence the influence [of maternal imagination]
can make itself felt only as variations in the mode of development.’
Thus by exchanging the conception of function for that of the
development of organs, Burdach comes to the conclusion that ‘homologous
organs of the mother and the embryo are in such connexion’ that when
the former are disturbed a corresponding ‘change in the formation of
the latter may arise.’
It seems to be not without value for the appreciation of the questions
with which we are dealing to remember that the idea of ‘maternal
impressions’ was only comparatively recently believed to be a
scientific theory, and that the proofs in support of it were brought
forward in form and language as scientific proofs. In Burdach’s book we
even meet with detailed ‘proofs’ that violent mental shocks produced by
maternal impressions may not only exercise their influence upon one but
even upon several children born successively, although with diminishing
strength. ‘A young wife received a shock during her first pregnancy
upon seeing a child with a hare-lip, and she was constantly haunted
with the idea that her child might have the same malformation. She was
delivered of a child with a typical hare-lip: her next child had an
upper lip with a less-marked cleft; while the third possessed a red
mark instead of a cleft.’
Now what can be said about such ‘proofs’? We may probably rightly
conjecture that Burdach, who was in other respects a clever
physiologist, was in this subject somewhat credulous: but there are
also instances about which there is not the slightest doubt. I may
remind the reader of a case which has been told by no other than the
celebrated embryologist, Carl Ernst von Baer[305].
‘A lady was very much upset by a fire, which was visible at a distance,
because she believed that it was in her native place. As the latter was
seven German miles distant, the impression had lasted a long time
before it was possible to receive any certain intelligence, and this
long delay affected the mind of the lady so greatly, that for some time
afterwards she said that she constantly saw the flames before her eyes.
Two or three months afterwards she was delivered of a daughter who had
a red patch on the forehead in the form of a flame. This patch did not
disappear until the child was seven years old.’ Von Baer added, ‘I
mention this case because I am well acquainted with it, for the lady
was my own sister, and because she complained of seeing flames before
her eyes before the birth of the child, and did not invent it
afterwards as the “cause” of the strange appearance.’
Here then we have a case which is absolutely certain. Von Baer’s name
is a guarantee for absolute accuracy. Why then has science, in spite of
this, rejected the whole idea of the efficacy of ‘maternal impressions’
ever since the appearance of the treatises by Bergmann and
Leuckart[306]?
Science has rejected this idea for many and conclusive reasons, all of
which I am not going to repeat here. In the first place, because our
maturer knowledge of the physiology of the body shows that such a
causal connexion between the peculiar characters of the child and, if I
may say so, the corresponding psychical impressions of the mother, is a
supposition which cannot be admitted; but also and chiefly because a
single coincidence of an idea of the mother with an abnormality in the
child does not form the proof of a causal connexion between the two
phenomena.
I do not doubt that among the many thousands of present and past
students in German Universities, whose faces are covered with scars,
there may be one with a son who exhibits a birth-mark on the spot where
the father possesses a scar. All sorts of birth-marks occur, and why
should they not sometimes have the appearance of a scar? Such a case,
if it occurred, would be acceptable to the adherents of the theory of
the transmission of acquired characters; it would in their opinion
completely upset the views of their opponents.
But how could such a case, if it were really established, be capable of
proving the supposed form of hereditary transmission, any more than von
Baer’s case could prove the theory of the efficacy of ‘maternal
impressions’?
I am of opinion that the extraordinary rarity of such cases strongly
enforces the fact that we have to do with an accidental and not a
causal coincidence. If scars could be really transmitted, we should
expect very frequently to find birth-marks which correspond to scars
upon the face of the father,—viz. in almost all cases in which the son
had inherited the type of face possessed by the father. If this were so
we should have to be seriously concerned about the beauty of the next
generation in Germany, as so many of our undergraduates follow the
fashion of decorating their faces with as many of these ‘honourable
scars’ as possible.
I have spoken of ‘maternal impressions’ because I wished to show that,
until quite recently, distinguished and acute scientific men have
adhered to an idea, and believed that they possessed the proof of an
idea, which has now been completely and for ever abandoned by science.
But in addition to this, there is a very close connexion between the
theory of the efficacy of maternal impressions and that of the
transmission of acquired characters, and sometimes they are even
confounded together.
Last year a popular scientific journal quoted the following case as a
proof of the transmission of mutilations. I do not, however, wish to
imply that the editor must be held responsible for the errors of a
correspondent. ‘In November, 1864, a pregnant merino sheep broke its
right fore-leg, about two inches above the knee-joint; the limb was put
in splints and healed a long time before the following March, when the
animal produced young. The lamb possessed a ring of black wool from two
to three inches in breadth round the place at which the mother’s leg
had been broken, and upon the same leg.’ Now if we even admitted that a
ring of black wool could be looked upon as a character which
corresponds to the fracture of the mother’s leg, the case could not
possibly be interpreted as the transmission of a mutilation, but as an
instance of the efficacy of maternal impressions; for the ewe was
already pregnant when she fractured her leg. The present state of
biological science teaches us that, with the fusion of egg and
sperm-cell, potential heredity is determined[307]. Such fusion
determines the future fate of the egg-cell and the individual with all
its various tendencies.
Such tales, when quoted as ‘remarkable facts which prove the
transmission of mutilations,’ thoroughly deserve the contempt with
which they have been received by Kant and His. When the above-mentioned
instance was told me, I replied, ‘It is a pity that the black wool was
not arranged in the form of the inscription “To the memory of the
fractured leg of my dear mother.”’
The tales of the efficacy of ‘maternal impressions’ and of the
transmission of mutilations are closely connected, and break down
before the present state of biological science. No one can be prevented
from believing such things, but they have no right to be looked upon as
scientific facts or even as scientific questions. The first was
abandoned in the middle of the present century, and the second may be
given up now; when once discarded we need not fear that it will ever
again be resuscitated.
It is hardly necessary to say that the question as to the transmission
of acquired characters is not completely decided by the unconditional
rejection of the transmission of mutilations. Although I am of opinion
that such transmission does not take place, and that we can explain the
phenomena presented by the transformation of species without this
supposition, I am far from believing that the question is settled,
simply because the transmission of mutilations may be dismissed into
the domain of fable. But at all events we have gained this much,—that
the only facts which appear to directly prove a transmission of
acquired characters have been refuted, and that the only firm
foundation on which this hypothesis has been hitherto based has been
destroyed. We shall not be obliged, in future, to trouble about every
single so-called proof of the transmission of mutilations, and
investigation may be concentrated upon the domain in which lies the
true decision as to the Lamarckian principle, it may be concerned with
the explanation of the observed phenomena of transformation.