erms, or oftener the half-germs, it
contains would have reproduced in their development. If this
modification does not involve the production of substances capable of
changing the germ-plasm, or does not so affect nutrition as to deprive
the germ-plasm of certain of its elements, it will have no effect on the
offspring of the individual. This is probably the case as a rule. If, on
the contrary, it has some effect, this is likely to be due to a chemical
change which it has induced in the germ-plasm. This chemical change
might, by exception, bring about the original modification again in the
organism which the germ is about to develop, but there are as many and
more chances that it will do something else. In this latter case, the
generated organism will perhaps deviate from the normal type _as much
as_ the generating organism, but it will do so _differently_. It will
have inherited deviation and not character. In general, therefore, the
habits formed by an individual have probably no echo in its offspring;
and when they have, the modification in the descendants may have no
visible likeness to the original one. Such, at least, is the hypothesis
which seems to us most likely. In any case, in default of proof to the
contrary, and so long as the decisive experiments called for by an
eminent biologist[48] have not been made, we must keep to the actual
results of observation. Now, even if we take the most favorable view of
the theory of the transmissibility of acquired characters, and assume
that the ostensible acquired character is not, in most cases, the more
or less tardy development of an innate character, facts show us that
hereditary transmission is the exception and not the rule. How, then,
shall we expect it to develop an organ such as the eye? When we think of
the enormous number of variations, all in the same direction, that we
must suppose to be accumulated before the passage from the pigment-spot
of the Infusorian to the eye of the mollusc and of the vertebrate is
possible, we do not see how heredity, as we observe it, could ever have
determined this piling-up of differences, even supposing that individual
efforts could have produced each of them singly. That is to say that
neo-Lamarckism is no more able than any other form of evolutionism to
solve the problem.

* * * * *

In thus submitting the various present forms of evolutionism to a common
test, in showing that they all strike ag