corn reared under different conditions.]

We now recognize that this statement contains an important truth, but we
have found that it contains only a part of the truth. Any one who repeats
for himself this kind of selection experiment will find that while his
average class will often change in the direction of his selection, the
process slows down as a rule rather suddenly (fig. 80). He finds, moreover,
that the limits of variability are not necessarily transcended as the
process continues even although the average may for a while be increased.
More tall men may be produced by selection of this kind, but the tallest
men are not necessarily any taller than the tallest in the original
population.

[Illustration: FIG. 79. Curves showing how (hypothetically) selection might
be supposed to bring about progress in direction of selection. (After
Goldschmidt.)]

Selection, then, has not produced anything new, but only more of certain
kinds of individuals. Evolution, however, means producing more new things,
not more of what already exists.

Darwin seems to have thought that the range of variation shown by the
offspring of a given individual about that type of individual would be as
wide as the range shown by the original population (fig. 79), but Galton's
work has made it clear that this is not the case in a general or mixed
population. If the offspring of individuals continued to show, as Darwin
seems to have thought, as wide a range on each side of their parents' size,
so to speak, as did the original population, then it would follow that
selection could slide successive generations along in the direction of
selection.

[Illustration: FIG. 80. Diagram illustrating the results of selection for
extra bristles in D. ampelophila. Selection at first produces decided
effects which soon slow down and then cease. (MacDowell.)]

Darwin himself was extraordinarily careful, however, in the statements he
made in this connection and it is rather by implication than by actual
reference that one can ascribe this meaning to his views. His
contemporaries and many of his followers, however, appear to have accepted
this _sliding scale_ interpretation as the cardinal doctrine of evolution.
If this is doubted or my statement is challenged then one must explain why
de Vries' mutation theory met with so little enthusiasm amongst the older
group of zoologists and botanists; and one must explain why Johannsen's
splendid work met with such bitt