goniums from the Cape are equally
accommodating, and vary just when and where and how we require them.

Turning to animals we find equally striking examples. If we want any
special quality in any animal we have only to breed it in sufficient
quantities and watch carefully, and the required variety is _always_
found, and can be increased to almost any desired extent. In Sheep, we
get flesh, fat, and wool; in Cows, milk; in Horses, colour, strength,
size, and speed; in Poultry, we have got almost any variety of colour,
curious modifications of plumage, and the capacity of perpetual
egg-laying. In Pigeons we have a still more remarkable proof of the
universality of variation, for it has been at one time or another the
fancy of breeders to change the form of every part of these birds, and
they have never found the required variations absent. The form, size,
and shape of bill and feet, have been changed to such a degree as is
found only in distinct genera of wild birds; the number of tail feathers
has been increased, a character which is generally one of the most
permanent nature, and is of high importance in the classification of
birds; and the size, the colour, and the habits, have been also changed
to a marvellous extent. In Dogs, the degree of modification and the
facility with which it is effected, is almost equally apparent. Look at
the constant amount of variation in opposite directions that must have
been going on, to develop the poodle and the greyhound from the same
original stock! Instincts, habits, intelligence, size, speed, form, and
colour, have always varied, so as to produce the very races which the
wants or fancies or passions of men may have led them to desire. Whether
they wanted a bull-dog to torture another animal, a greyhound to catch
a hare, or a bloodhound to hunt down their oppressed fellow-creatures,
the required variations have always appeared.

Now this great mass of facts, of which a mere sketch has been here
given, are fully accounted for by the "Law of Variation" as laid down at
the commencement of this paper. Universal variability--small in amount
but in every direction, ever fluctuating about a mean condition until
made to advance in a given direction by "election," natural or
artificial,--is the simple basis for the indefinite modification of the
forms of life;--partial, unbalanced, and consequently unstable
modifications being produced by man, while those developed under the
unrestrained a