on the parentage and present fertility of the several
breeds of the dog, of Indian and European cattle, sheep, and pigs. Hence it
would be unreasonable to expect that races formed under domestication {411}
should acquire sterility when crossed, whilst at the same time we admit
that domestication eliminates the normal sterility of crossed species. Why
with closely allied species their reproductive systems should almost
invariably have been modified in so peculiar a manner as to be mutually
incapable of acting on each other--though in unequal degrees in the two
sexes, as shown by the difference in fertility between reciprocal crosses
in the same species--we do not know, but may with much probability infer
the cause to be as follows. Most natural species have been habituated to
nearly uniform conditions of life for an incomparably longer period of time
than have domestic races; and we positively know that changed conditions
exert an especial and powerful influence on the reproductive system. Hence
this difference in habituation may well account for the different action of
the reproductive organs when domestic races and when species are crossed.
It is a nearly analogous fact, that most domestic races may be suddenly
transported from one climate to another, or be placed under widely
different conditions, and yet retain their fertility unimpaired; whilst a
multitude of species subjected to lesser changes are rendered incapable of
breeding.

With the exception of fertility, domestic varieties resemble species when
crossed in transmitting their characters in the same unequal manner to
their offspring, in being subject to the prepotency of one form over the
other, and in their liability to reversion. By repeated crosses a variety
or a species may be made completely to absorb another. Varieties, as we
shall see when we treat of their antiquity, sometimes inherit their new
characters almost, or even quite, as firmly as species. With both, the
conditions leading to variability and the laws governing its nature appear
to be the same. Domestic varieties can be classed in groups under groups,
like species under genera, and these under families and orders; and the
classification may be either artificial,--that is, founded on any arbitrary
character,--or natural. With varieties a natural classification is
certainly founded, and with species is apparently founded, on community of
descent, together with the amount of modification which the