the petalody of the stamens (doubling of flowers). In this connection
we must keep in view the fact that every visible character in a plant is
the resultant of the cooperation of specific structure, with its various
potentialities, and the influence of the environment. We know, that in
a pure species all characters vary, that a blue-flowering Campanula or
a red Sempervivum can be converted by experiment into white-flowering
forms, that a transformation of stamens into petals may be caused by
fungi or by the influence of changed conditions of nutrition, or
that plants in dry and poor soil become dwarfed. But so far as the
experiments justify a conclusion, it would appear that such alterations
are not inherited by the offspring. Like all other variations they
appear only so long as special conditions prevail in the surroundings.

It has been shown that the case is quite different as regards the
white-flowering, double or dwarf races, because these retain their
characters when cultivated under practically identical conditions, and
side by side with the blue, single-flowering or tall races. The problem
may therefore be stated thus: how can a character, which appears in the
one case only under the strictly limited conditions of the experiment,
in other cases become apparent under the very much wider conditions of
ordinary cultivation? If a character appears, in these circumstances,
in the case of all individuals, we then speak of constant races. In such
simple cases the essential point is not the creation of a new character
but rather an ALTERATION OF THIS CHARACTER IN ACCORDANCE WITH THE
ENVIRONMENT. In the examples mentioned the modified character in the
simple varieties (or a number of characters in elementary species)
appears more or less suddenly and is constant in the above sense. The
result is what de Vries has termed a Mutation. In this connection we
must bear in mind the fact that no difference, recognisable externally,
need exist between individual variation and mutation. Even the most
minute quantitative difference between two plants may be of specific
value if it is preserved under similar external conditions during many
successive generations. We do not know how this happens. We may state
the problem in other terms; by saying that the specific structure must
be altered. It is possible, to some extent, to explain this sudden
alteration, if we regard it as a chemical alteration of structure either
in the specific