506 
NATURE 
[SEPTEMBER 21, 1899 
hile an acquired character depends on the nature of the 
stimulus. 
If we imagine a zygote to be a machine capable of working 
out certain results on material supplied to it, then we should 
properly apply the term genetic character to the features of 
the machinery itself, and the words acquired character to the 
results achieved by its working. These clearly will depend 
primarily on the structure of the machinery, and secondarily 
upon the material and energy supplied to it—that is to say, 
upon the way in which it is worked. 
Variations in genetic characters are variations in the 
machinery of different zygotes—that is to say, in the constitu- 
tion—while variations in acquired characters are variations in 
the results of the working of one zygote according to the con- 
ditions under which it is worked. 
For instance, let us take the case of those twins which arise 
by the division of one zygote, and are consequently identical 
in genetic characters, z.e. in constitution. If they are submitted 
to different conditions, they will develop differences which will 
depend entirely upon the conditions and the time of life when 
the differentiation in the conditions occurred. These differences, 
then, will be a function of the external conditions, z.e. of the 
manner in which the machinery is worked, and constitute what 
we call variation in acquired characters. 
Are Acquired Characters Transmissible as such in 
Reproduction ? 
To return to our question, are the so-called acquired char- 
acters ever transmitted in reproduction? Let us consider what 
this question means in the light of the preceding discussion. 
Acquired characters are features which arise in the zygote in 
response to external stimuli. Now the zygote at its first estab- 
lishment has none of the characters which are subsequently 
acquired. All it has is the power of acquiring them. Clearly, 
then, acquired characters are not transmitted. The power of 
producing them is all that can be transmitted ; and this power 
resides in the reproductive organs and in the gametes to which 
the reproductive organs give rise, so that the question must be 
put in another form, Is it possible by submitting an organism 
toa certain set of conditions, and thus causing it to acquire 
certain characters, so to modify its reproductive organs that the 
same characters will appear in its offspring as the result of the 
application of a different and simpler stimulus ? 
For instance, the power of reading conferred by education, 
the hardness of the hands and increased size of the muscles pro- 
duced by manual labour: is it possible that these characters, 
now produced by complex external stimuli applied at a particular 
period of life, should ever in future ages be produced by the 
simpler stimuli found within the uterus, so that a man may be 
born able to read or write, or with hands horny and hard like 
those of a navvy ? 
In trying to find an answer to this question let us first of all 
look into the probabilities of the case, to see if we can relate the 
question to any other class of phenomena about which we have, 
or think we have, definite knowledge. 
When an organism is affected by external agents the action 
may apply to the whole organisation or principally to one organ. 
Let us take a case in which one organ only appears to be 
affected, ¢.¢. the enlargement by exercise of the right arm of a 
man. Now, although in this case it is only the muscles of the 
arm which appear at first sight to be affected, we must not forget 
that the organs of the body are correlated with one another, and 
an alteration of one will produce an alteration in others. By 
exercise of the right arm the muscles of that arm are obviously 
enlarged, but other changes not so obvious must also have taken 
place. The bones to which the muscles are attached will be 
altered ; the blood-vessels supplying the muscles will be en- 
larged, and the nerves which act upon the muscles, and probably 
the part of the central nervous system from which they proceed, 
will also be altered. These are some of the more obvious cor- 
related changes which will have occurred ; no doubt there will 
have been others—indeed it is not perhaps too much to say that 
all the organs of the body will have reacted to the enlargement 
of the arm—but the effect on organs not in functional correla- 
tion with the muscles of the right arm will be imperceptible, 
and may be neglected. Thus the colour of the hair, the length 
and character of the alimentary canal, size of the leg muscles, 
the renal organs, &c., will not show appreciable alteration. 
Above all, the other arm will not be affected, or if it is affected 
the alteration will be so slight as not to be noticeable. Now, 
NO. 1560, VOL. 60] 
we know that homologous parts, whether symmetrically homo- 
logous or serially so, are in some kind of close connection. For 
instance, when one member of an homologous series varies, it is 
commonly found that other members of the same series will also 
vary. Yet in spite of this connection which exists between the 
right and left arms and between the right arm and right leg, 
there is no similar alteration either in the left arm or in the right 
leg. Now, if parts which from these facts we may suppose to 
be in some connection are not affected, how can we expect the 
reproductive organs, not only to be modified, but also to be so 
modified that the germs which are about to be budded off from 
them will be so affected as to produce exactly the same char- 
acter—in this case enlarged muscle, &c.—without the applic- 
ation of the same stimulus, viz. exercise? Thus, while I freely 
admit that every alteration of an organ in response to externa) 
agents wlll react through the whole organisation, affecting each 
organ in functional correlation with the affected organ in a way 
which will depend upon the function of the correlated organ, 
and possibly other organs not in functional correlation in an in- 
definite way and to a slight extent, yet I maintain that it is very 
hard to believe that it will have such a sharp and precise effect 
upon every spermatozoon and ovum subsequently produced that 
not merely will these products be altered generally in all their 
properties, but that one particular part of them—and that part 
of them always the same—will be so altered that the organisms 
which develop from them will be able to present the same 
modification on the application of a different stimulus. It is in- 
conceivable ; unless, indeed, we suppose that the very mole- 
cules of the incipient organs in the germ are more closely cor- 
related with corresponding parts of the parent body than are the 
homologous parts of the parent body with one another. 
Now, to prove the existence of such a remarkable and inti- 
mate correlation would surely require the very strongest and 
most conclusive evidence. Is there any such strong evidence ? 
I think I may fairly answer this question in the negative. The 
evidence which has been brought forward in favour of the so- 
called inheritance of acquired characters is far from conclusive. 
That such evidence! exists I do not deny, but it is all, or almost 
all, capable of receiving other interpretations. 
Effect of Changed Conditions upon the Reproductive Organs. 
On the other hand, all the certain evidence we have concern- 
ing what happens when the reproductive organs are affected, 
either directly or by correlation, by a change of conditions— 
and, as we have seen above, they must be affected if there is to 
be any change in the offspring—tends to show that there is not 
any relation between the effect produced on the parent and that 
appearing in the offspring. 
The only means of judging whether the reproductive organs 
are affected by external conditions is by observing any change 
which may occur in their function. Now, only two such 
physiological effects of a change of conditions are certainly 
known; these are (1) the production of sterility or of partial 
sterility ; (2) the production of an increased but indefinite vari- 
ability in the offspring. With regard to the first of these effects 
One of the most common, or at any rate one of the most notice- 
able, alterations in an organism, effected by change in the 
external conditions, is an alteration of the reproductive system, 
an alterationof such a kind that organisms which had previously 
freely interbred with one another are no longer able to do so. 
One of the most common results of removing organisms from 
their natural surroundings is to induce sterility or partial sterility. 
There is no reason to doubt that this sterility or tendency to 
sterility is, broadly speaking, due to an affection of the repro- 
ductive system. In the case of the higher animals, it may in 
some cases be due to an action upon the instincts, but in the 
lower animals and in plants we can hardly doubt that it is due 
to a direct action upon the reproductive organs. Indeed in 
plants these organs are often visibly affected. Among animals, 
however, there does not appear to be any satisfactory evidence 
on the point, and it is not known what organs are affected, 
whether it is the actual gametes, or the reproductive glands, or 
some of the other organs concerned.” 
The other result of changed conditions which is certainly 
known is to induce an increased amount of variability of the 
genetic kind, though not immediately, often indeed not untib 
1 For a good statement and discussion of the evidence in favour of this. 
view, see Romanes’ * Darwin and after Darwin,” vol. ii. chaps. 3 and 4. 
2 The exact cause of this sterility in the higher animals is a point which 
specially needs investigation. 
