ures in the morphology of plants and animals. The attempt to
substitute these laws for those of variation and natural selection has
failed in cases where we can apply a definite test, as in that of the
origin of spines on trees and shrubs; while the extreme diversity of
vegetable structure and form among the plants of the same country and of
the same natural order, of itself affords a proof of the preponderating
influence of variation and natural selection in keeping the many diverse
forms in harmony with the highly complex and ever-changing environment.

Lastly, we have seen that Professor Weismann's theory of the continuity
of the germ-plasm and the consequent non-heredity of acquired
characters, while in perfect harmony with all the well-ascertained facts
of heredity and development, adds greatly to the importance of natural
selection as the one invariable and ever-present factor in all organic
change, and that which can alone have produced the temporary fixity
combined with the secular modification of species. While admitting, as
Darwin always admitted, the co-operation of the fundamental laws of
growth and variation, of correlation and heredity, in determining the
direction of lines of variation or in the initiation of peculiar organs,
we find that variation and natural selection are ever-present agencies,
which take possession, as it were, of every minute change originated by
these fundamental causes, check or favour their further development, or
modify them in countless varied ways according to the varying needs of
the organism. Whatever other causes have been at work, Natural Selection
is supreme, to an extent which even Darwin himself hesitated to claim
for it. The more we study it the more we are convinced of its
overpowering importance, and the more confidently we claim, in Darwin's
own words, that it "has been the most important, but not the exclusive,
means of modification."

FOOTNOTES:

[Footnote 198: See the Duke of Argyll's letter in _Nature_, vol. xxxiv.
p. 336.]

[Footnote 199: _Journal of the Anthropological Institute,_ vol. xv. pp.
246-260.]

[Footnote 200: The idea of the non-heredity of acquired variations was
suggested by the summary of Professor Weismann's views, in _Nature_,
referred to later on. But since this chapter was written I have, through
the kindness of Mr. E.B. Poulton, seen some of the proofs of the
forthcoming translation of Weismann's Essays on Heredity, in which he
sets forth