revert, but rather that their
capacities in this respect, as well as for change, are different in
different kinds, so that often reversion may only show itself at the end of
very long periods indeed.

Yet some of the instances given as probable or possible causes of reversion
by Mr. Darwin, can hardly be such. He cites, for example, the occasional
presence of supernumerary digits in man.[117] For this notion, however, he
is not responsible, as he rests his remark on the authority of a {123}
passage published by Professor Owen. Again, he refers[118] to "the greater
frequency of a monster proboscis in the pig than in any other animal." But
with the exception of the peculiar muzzle of the Saiga (or European
antelope), the only known proboscidian Ungulates are the elephants and
tapirs, and to neither of these has the pig any close affinity. It is
rather in the horse than in the pig that we might look for the appearance
of a reversionary proboscis, as both the elephants and the tapirs have the
toes of the hind foot of an odd number. It is true that the elephants are
generally considered to form a group apart from both the odd and the
even-toed Ungulata. But of the two, their affinities with the odd-toed
division are more marked.[119]

Another argument in favour of the, at least intermitting, constancy of
specific forms and of sudden modification, may be drawn from the absence of
minute transitional forms, but this will be considered in the next chapter.

It remains now to notice in favour of specific stability, that the
objection drawn from physiological difference between "species" and "races"
still exists unrefuted.

Mr. Darwin freely admits difficulties regarding the sterility of different
species when crossed, and shows satisfactorily that it could never have
arisen from the action of "Natural Selection." He remarks[120] also: "With
some few exceptions, in the case of plants, domesticated varieties, such as
those of the dog, fowl, pigeon, several fruit trees, and culinary
vegetables, which differ from each other in external characters more than
many species, are perfectly fertile when crossed, or even fertile in {124}
excess, whilst closely allied species are almost invariably in some degree
sterile."

Again, after speaking of "the general law of good being, derived from the
intercrossing of distinct individuals of the same species," and the
evidence that the pollen of a distinct _variety_ or race is prepotent