168 HERBERT SPENCER

He first points out that when a structure is altered
by a change of function the modification is often
unobtrusive, and its transmission consequently difficult
to detect. "Moreover, such specialities of structure
as are due to specialities of function, are usually
entangled with specialities which are, or may be, due
to selection, natural or artificial. In most cases it is
impossible to say that a structural peculiarity which
seems to have arisen in offspring from a functional
peculiarity in a parent, is wholly independent of some
congenital peculiarity of structure in the parent,
whence this functional peculiarity arose. We are
restricted to cases with which natural or artificial
selection can have had nothing to do, and such cases
are difficult to find. Some, however, may be noted."

When a plant is transferred from one soil to
another it undergoes " a change of habit"; its leaves
may become hairy, its stem woody, its branches
drooping. " These are modifications of structure
consequent on modifications of function that have
been produced by modifications in the actions of
external forces. And as these modifications re-
appear in succeeding generations, we have, in them,
examples of functionally-established variations that
are hereditarily transmitted." But this is a non
sequitur, since the modifications may re-appear merely
because they are re-impressed directly on each successive
generation.

Spencer notes that in the domestic duck the bones
of the wing weigh less and the bones of the leg
more in proportion to the whole skeleton than do the
same bones in the wild duck; that in cows and goats
which are habitually milked the udders are large;l units covers them all ? Theg on it modifica-intrinsic incoherence; worthless as