body, and the relative shortening or the
throwing forward of the central portion containing the ribs (frequently
one less in number) and the sternum. The whole body of the pigeon is so
much under the control of artificial selection, that every precaution
must be taken to guard against such possible sources of error.[35]

Under domestication there would be a suspension of the previous
elimination of reduced breast-bones by natural selection (Weismann's
panmixia), and a diminution of the parts concerned in flying might even
be favoured, as lessened powers of _continuous_ flight would prevent
pigeons from straying too far, and would fit them for domestication or
confinement. Such causes might reduce some of the less observed parts
affected by flying, while still leaving the wing of full size for
occasional flight, or to suit the requirements of the pigeon-fanciers. A
change might thus be commenced like that seen in the rudimentary keel of
the sternum in the owl-parrot of New Zealand, which has lost the power
of flight although still retaining fairly-developed wings.

SHORTENED FEET IN PIGEONS.

Darwin thinks it highly probable that the short feet of most breeds of
pigeons are due to lessened use, though he owns that the effects of
correlation with the shortened beak are more plainly shown than the
effects of disuse.[36] But why need the inherited effects of disuse be
called in to explain an average reduction of some 5 per cent., when
Darwin's measurements show that in the breeds where long beaks are
favoured the principle of correlation between these parts has lengthened
the foot by 13 per cent. in spite of disuse?

SHORTENED LEGS OF RABBITS.

In the case of the domestic rabbit Darwin notices that the bones of the
legs have (relatively) become shorter by an inch and a half. But as the
leg-bones have _not_ diminished in relative weight,[37] they must
clearly have grown _thicker_ or denser. If disuse has shortened them, as
Darwin supposes, why has it also thickened them? The ears and the tail
have been lengthened in spite of disuse. Why then may not the ungainly
hind-legs have been shortened by human preference independently of the
inherited effects of disuse? By relying on apparently favourable
instances and neglecting the others it would be easy to arrive at all
manner of unsound conclusions. We might thus become convinced that
vessels tend to sail northwards, or that a pendulum oscillates more
often in one direc