functionless at first, when it began to appear;" and the conclusion is,
that, _qua_ functionless, such a modification cannot have been caused by
natural selection. I will consider these two points separately.

First as to the premiss, it is not true that every modification of
structure must necessarily be functionless when it first begins to
appear. There are two very good reasons why such should not be the case
in all instances, even if it should be the case in some. For, as a
matter of observable fact, a very large proportional number of incipient
organs are useful from the very moment of their inception. Take, for
example, what is perhaps the most wonderful instance of refined
mechanism in nature--the eye of a vertebrated animal. Comparative
anatomy and embryology combine to testify that this organ had its origin
in modifications of the endings of the ordinary nerves of the skin. Now
it is evident that from the very first any modification of a cutaneous
nerve whereby it was rendered able, in however small a degree, to be
differently affected by light and by darkness would be of benefit to the
creature presenting it; for the creature would thus be able to seek the
one and shun the other according to the requirements of its life. And
being thus useful from the very moment of its inception, it would
afterwards be gradually improved as variations of more and more utility
presented themselves, until not only would finer and finer degrees of
difference between light and shade become perceptible, but even the
outlines of solid bodies would begin to be appreciated. And so on, stage
by stage, till from an ordinary nerve-ending in the skin is evolved the
eye of an eagle.

Moreover, in this particular instance there is very good reason to
suppose that the modification of the cutaneous nerves in question began
by a progressive increase in their sensitiveness to temperature.
Wherever dark pigment happened to be deposited in the skin--and we know
that in all animals it is apt to be deposited in points and patches, as
it were by accident, or without any "prophecy" as to future uses,--the
cutaneous nerves in its vicinity would be better able to appreciate the
difference between sun and shade in respect of temperature, even though
as yet there were no change at all in these cutaneous nerves tending to
make them responsive to light. Now it is easy to see how, from such a
purely accidental beginning, natural selection would have had f