d exceedingly irregular in its incidence and
severity. It is most severe and fatal to the young; but when an animal
has once reached maturity, and especially when it has gained experience
by several years of an eventful existence, it may be able to maintain
itself under conditions which would be fatal to a young and
inexperienced creature of the same species. The examples adduced by Mr.
Mivart do not, therefore, in any way impugn the hardness of nature as a
taskmaster, or the extreme severity of the recurring struggle for
existence. (See _Nature_, vol. xxxix. p. 127.)]

[Footnote 46: _Origin of Species,_ p. 72.]

[Footnote 47: Darwin's latest expression of opinion on this question is
interesting, since it shows that he was inclined to return to his
earlier view of the general, or universal, utility of specific
characters. In a letter to Semper (30th Nov. 1878) he writes: "As our
knowledge advances, very slight differences, considered by systematists
as of no importance in structure, are continually found to be
functionally important; and I have been especially struck with this fact
in the case of plants, to which my observations have, of late years,
been confined. Therefore it seems to me rather rash to consider slight
differences between representative species, for instance, those
inhabiting the different islands of the same archipelago, as of no
functional importance, and as not in any way due to natural selection"
_(Life of Darwin_, vol. iii. p. 161).]

[Footnote 48: See _Variation of Animals and Plants_, vol. i. p. 86.]

[Footnote 49: _Journal of the Linnean Society, Zoology,_ vol. xx. p.
215.]

[Footnote 50: In Mr. Gulick's last paper (_Journal of Linn. Soc. Zool._,
vol. xx. pp. 189-274) he discusses the various forms of isolation above
referred to, under no less than thirty-eight different divisions and
subdivisions, with an elaborate terminology, and he argues that these
will frequently bring about divergent evolution without any change in
the environment or any action of natural selection. The discussion of
the problem here given will, I believe, sufficiently expose the fallacy
of his contention; but his illustration of the varied and often
recondite modes by which practical isolation may be brought about, may
help to remove one of the popular difficulties in the way of the action
of natural selection in the origination of species.]

CHAPTER VII

ON THE INFERTILITY OF CROSSES BETWEEN DISTINCT SPECI