it draws the inference that new characters arose in the form in which they
are found to be inherited, as complete units, and not by gradual,
continuous increase, that specific characters are due to mutations, and
that all evolution has been the result of similar hereditary factors,
arising by some internal process in the divisions of reproductive cells,
and not determined by external conditions. Some Mendelians maintain that
if the mutations are not compatible with the existing conditions of life,
the organism must either die or find new conditions in which it can live.

Bateson remarks (_Mendel's Principles of Heredity_, 1909, p. 288):
'Mendelism provides no fresh clue to the problem of adaptation except in
so far as it is easier to believe that a definite integral change in
attributes can make a perceptible difference to the prospect of success,
than that an indefinite and impalpable change should entail such
consequences.' Here the distinction between adaptive and non-adaptive
characters is recognised, but both are emphatically attributed to the same
origin.

The American evolutionist, T. H. Morgan, also a specialist in Mendelism,
goes further, and maintains, not merely that mutations which happened to
make a 'difference to the prospect of success' survived, or were selected,
but that if a mutation arising from a change in the gametes was not
compatible with the conditions of the animal's life at the time, it either
died, or found other conditions, or adopted new habits which were adapted
to the new character or structure. He takes Flat-fishes as an example, and
suggests that having by mutation become asymmetrical, and having both eyes
on one side, etc., the fish adopted the habit of lying on the ground on
one side of its body. This is, of course, the exact opposite of the older
conception: the structure of the animal has not been changed by new habits
or conditions, but new habits and conditions have been sought and found in