PROVINCE to a CLASS from a CLASS to an ORDER, from ORDERS to FAMILIES,
and from these to GENERA, until we come at length to the smallest
groups of animals which can be defined one from the other by constant
characters, which are not sexual; and these are what naturalists call
SPECIES in practice, whatever they may do in theory.

If, in a state of nature, you find any two groups of living beings,
which are separated one from the other by some constantly-recurring
characteristic, I don't care how slight and trivial, so long as it is
defined and constant, and does not depend on sexual peculiarities, then
all naturalists agree in calling them two species; that is what is
meant by the use of the word species--that is to say, it is, for the
practical naturalist, a mere question of structural differences.*

[footnote]* I lay stress here on the 'practical'
signification of "Species." Whether a physiological test
between species exist or not, it is hardly ever applicable
by the practical naturalist.

We have seen now--to repeat this point once more, and it is very
essential that we should rightly understand it--we have seen that
breeds, known to have been derived from a common stock by selection,
may be as different in their structure from the original stock as
species may be distinct from each other.

But is the like true of the physiological characteristics of animals?
Do the physiological differences of varieties amount in degree to those
observed between forms which naturalists call distinct species? This
is a most important point for us to consider.