against his grandson's, I have always intended to support. With
Charles Darwin, on the other hand, there is indeed cunning, effort,
and consequent use and disuse; nor does he deny that these have
produced some, and sometimes even an important, effect in modifying
species, but he assigns by far the most important role in the whole
scheme to natural selection, which, as I have already shown, must,
with him, be regarded as a synonym for luck pure and simple. This,
for reasons well shown by Mr. Spencer in the articles under
consideration, is so untenable that it seems only possible to
account for its having been advanced at all by supposing Mr.
Darwin's judgment to have been perverted by some one or more of the
many causes that might tend to warp them. What the chief of those
causes may have been I shall presently point out.

Buffon erred rather on the side of ignoring functionally produced
modifications than of insisting on them. The main agency with him
is the direct action of the environment upon the organism. This, no
doubt, is a flaw in Buffon's immortal work, but it is one which
Erasmus Darwin and Lamarck easily corrected; nor can we doubt that
Buffon would have readily accepted their amendment if it had been
suggested to him. Buffon did infinitely more in the way of
discovering and establishing the theory of descent with modification
than any one has ever done either before or since. He was too much
occupied with proving the fact of evolution at all, to dwell as
fully as might have been wished upon the details of the process
whereby the amoeba had become man, but we have already seen that he
regarded inherited mutilation as the cause of establishing a new
breed of dogs, and this is at any rate not laying much stress on
functionally produced modifications. Again, when writing of the
dog, he speaks of variations arising "BY SOME CHANCE common enough