one generation, is cumulative, and under the
influence of selection is sufficient to keep up the harmony between the
organism and its slowly changing environment.[214]

_The Non-Heredity of Acquired Characters._

Certain observations on the embryology of the lower animals are held to
afford direct proof of this theory of heredity, but they are too
technical to be made clear to ordinary readers. A logical result of the
theory is the impossibility of the transmission of acquired characters,
since the molecular structure of the germ-plasm is already determined
within the embryo; and Weismann holds that there are no facts which
really prove that acquired characters can be inherited, although their
inheritance has, by most writers, been considered so probable as hardly
to stand in need of direct proof.

We have already shown, in the earlier part of this chapter, that many
instances of change, imputed to the inheritance of acquired variations,
are really cases of selection; while the very fact that _use_ implies
_usefulness_ renders it almost impossible to eliminate the action of
selection in a state of nature. As regards mutilations, it is generally
admitted that they are not hereditary, and there is ample evidence on
this point. When it was the fashion to dock horses' tails, it was not
found that horses were born with short tails; nor are Chinese women born
with distorted feet; nor are any of the numerous forms of racial
mutilation in man, which have in some cases been carried on for hundreds
of generations, inherited. Nevertheless, a few cases of apparent
inheritance of mutilations have been recorded,[215] and these, if
trustworthy, are difficulties in the way of the theory. The undoubted
inheritance of disease is hardly a difficulty, because the
predisposition to disease is a congenital, not an acquired character,
and as such would be the subject of inheritance. The often-quoted case
of a disease induced by mutilation being inherited (Brown-Sequard's
epileptic guinea-pigs) has been discussed by Professor Weismann, and
shown to be not conclusive. The mutilation itself--a section of certain
nerves--was never inherited, but the resulting epilepsy, or a general
state of weakness, deformity, or sores, was sometimes inherited. It is,
however, possible that the mere injury introduced and encouraged the
growth of certain microbes, which, spreading through the organism,
sometimes reached the germ-cells, and thus transmitted a