ably been at work in the production of the wonderful diversity
of colour and marking we find everywhere, more especially among the
butterflies and moths; and here its chief function may have been to
secure the pairing together of individuals of the same species. In some
of the moths this has been secured by a peculiar odour, which attracts
the males to the females from a distance; but there is no evidence that
this is universal or even general, and among butterflies, especially,
the characteristic colour and marking, aided by size and form, afford
the most probable means of recognition. That this is so is shown by the
fact that "the common white butterfly often flies down to a bit of paper
on the ground, no doubt mistaking it for one of its own species;" while,
according to Mr. Collingwood, in the Malay Archipelago, "a dead
butterfly pinned upon a conspicuous twig will often arrest an insect of
the same species in its headlong flight, and bring it down within easy
reach of the net, especially if it be of the opposite sex."[87] In a
great number of insects, no doubt, form, motions, stridulating sounds,
or peculiar odours, serve to distinguish allied species from each other,
and this must be especially the case with nocturnal insects, or with
those whose colours are nearly uniform and are determined by the need of
protection; but by far the larger number of day-flying and active
insects exhibit varieties of colour and marking, forming the most
obvious distinction between allied species, and which have, therefore,
in all probability been acquired in the process of differentiation for
the purpose of checking the intercrossing of closely allied forms.[88]

Whether this principle extends to any of the less highly organised
animals is doubtful, though it may perhaps have affected the higher
mollusca. But in marine animals it seems probable that the colours,
however beautiful, varied, and brilliant they may often be, are in most
cases protective, assimilating them to the various bright-coloured
seaweeds, or to some other animals which it is advantageous for them to
imitate.[89]

_Summary of the Preceding Exposition._

Before proceeding to discuss some of the more recondite phenomena of
animal coloration, it will be well to consider for a moment the extent
of the ground we have already covered. Protective coloration, in some of
its varied forms, has not improbably modified the appearance of one-half
of the animals living on