efore, it is argued, Darwin and his followers are
profoundly mistaken in representing the principle as one which
_produces_ adaptations. Now, although this objection has been put
forward by some of the most intelligent minds in our generation, it
appears to me to betoken some extraordinary failure to appreciate the
very essence of Darwinian doctrine. No doubt it is perfectly true that
natural selection does not produce variations of any kind, whether
beneficial or otherwise. But if it be granted that variations of many
kinds are occurring in every generation, and that natural selection is
competent to preserve the more favourable among them, then it appears to
me unquestionable that this principle of selection deserves to be
regarded as, in the full sense of the word, a natural cause. The
variations being expressly regarded by the theory as more or less
promiscuous[42], survival of the fittest becomes the winnowing fan,
whose function it is to eliminate all the less fit in each generation,
in order to preserve the good grain, out of which to constitute the next
generation. And as this process is supposed to be continuous through
successive generations, its action is supposed to be cumulative, till
from the eye of a worm there is gradually developed the eye of an eagle.
Therefore it follows from these suppositions (which are not disputed by
the present objection), that if it had not been for the process of
selection, such development would never have been begun; and that in the
exact measure of its efficiency will the development proceed. But any
agency without the operation of which a result cannot take place may
properly be designated the cause of that result: it is the agency which,
in co-operation with all the other agencies in the cosmos, produces that
result.

[42] The degree in which variability is indefinite, or, on the
contrary, determinate, is a question which is not yet ripe for
decision--nor even, in my opinion, for discussion. But I may here
state the following general principles with regard to it.

(1) It is evident that up to some point or another variations _must_
be pre-determined in definite lines. Men do not gather grapes from
thorns, figs from thistles, nor even moss-roses from sweet-briars.
In other words, "the nature of the organism" in all cases
necessitates the limiting of variations within certain bounds.

(2) But when the question is as to what these bou