increasing the beauty of either species by means of selection. The
variability is here almost confined to the male sex; but Mr. Wallace
and Mr. Bates have shewn (24. Wallace on the Papilionidae of the
Malayan Region, in Transact. Linn. Soc. vol. xxv. 1865, pp. 8, 36. A
striking case of a rare variety, strictly intermediate between two
other well-marked female varieties, is given by Mr. Wallace. See also
Mr. Bates, in Proc. Entomolog. Soc. Nov. 19, 1866, p. xl.) that the
females of some species are extremely variable, the males being nearly
constant. In a future chapter I shall have occasion to shew that the
beautiful eye-like spots, or ocelli, found on the wings of many
Lepidoptera, are eminently variable. I may here add that these ocelli
offer a difficulty on the theory of sexual selection; for though
appearing to us so ornamental, they are never present in one sex and
absent in the other, nor do they ever differ much in the two sexes.
(25. Mr. Bates was so kind as to lay this subject before the
Entomological Society, and I have received answers to this effect from
several entomologists.) This fact is at present inexplicable; but if it
should hereafter be found that the formation of an ocellus is due to
some change in the tissues of the wings, for instance, occurring at a
very early period of development, we might expect, from what we know of
the laws of inheritance, that it would be transmitted to both sexes,
though arising and perfected in one sex alone.
On the whole, although many serious objections may be urged, it seems
probable that most of the brilliantly-coloured species of Lepidoptera
owe their colours to sexual selection, excepting in certain cases,
presently to be mentioned, in which conspicuous colours have been
gained through mimicry as a protection. From the ardour of the male
throughout the animal kingdom, he is generally willing to accept any
female; and it is the female which usually exerts a choice. Hence, if
sexual selection has been efficient with the Lepidoptera, the male,
when the sexes differ, ought to be the more brilliantly coloured, and
this undoubtedly is the case. When both sexes are brilliantly coloured
and resemble each other, the characters acquired by the males appear to
have been transmitted to both. We are led to this conclusion by cases,
even within the same genus, of gradation from an extraordinary amount
of difference to identity in colour between the two sexes.
But it may be asked whether the difference in colour between the sexes
may not be accounted for by other means besides sexual selection. Thus
350