n full-grown. No doubt most variation,
(not depending on habits of life of individual) depends on early
change{161} and we must suspect that at whatever time of life the
alteration of foetus is effected, it tends to appear at same period.
When we <see> a tendency to particular disease in old age transmitted by
the male, we know some effect is produced during conception, on the
simple cell of ovule, which will not produce its effect till half a
century afterwards and that effect is not visible{162}. So we see in
grey-hound, bull-dog, in race-horse and cart-horse, which have been
selected for their form in full-life, there is much less (?) difference
in the few first days after birth{163}, than when full-grown: so in
cattle, we see it clearly in cases of cattle, which differ obviously in
shape and length of horns. If man were during 10,000 years to be able to
select, far more diverse animals from horse or cow, I should expect
there would be far less differences in the very young and foetal state:
and this, I think, throws light on above marvellous fact. In larvae,
which have long life selection, perhaps, does much,--in the pupa not so
much{164} There is no object gained in varying form &c. of foetus
(beyond certain adaptations to mother's womb) and therefore selection
will not further act on it, than in giving to its changing tissues a
tendency to certain parts afterwards to assume certain forms.

{160} The following: "Deaths of brothers <when> old by same peculiar
disease" which is written between the lines seems to have been a
memorandum which is expanded a few lines lower. I believe the case
of the brothers came from Dr R. W. Darwin.

{161} See the discussion to this effect in the _Origin_, Ed. i. pp.
443-4, vi. p. 610. The author there makes the distinction between a
cause affecting the germ-cell and the reaction occurring at a late
period of life.

{162} Possibly the sentence was meant to end "is not visible till
then."

{163} See _Origin_, Ed. i. pp. 444-5, vi. p. 611. The query
appended to _much less_ is justified, since measurement was
necessary to prove that the greyhound and bulldog puppies had not
nearly acquired "their full amount of proportional difference."

{164} <The following discussion, from the back of the page, is in
large measure the same as the text.> I think light can be thrown on
these facts. From the followin