not exhibit
it in its most complete form. In some cases this faculty may not be
exhibited at the top of the main stem, although developed in lateral
branches: in others it begins too late for full development. Much
depends upon nourishment and cultivation, but almost always the
horticulturist has to single out the best individuals and to reject
those which do not come up to the standard.

The internal causes are of a historical nature. The external ones may be
defined as nourishment and environment. In some cases nutrition is
the main factor, as, for instance, in fluctuating variability, but in
natural selection environment usually plays the larger part.

The internal or historical causes are constant during the life-time of
a species, using the term species in its most limited sense, as
designating the so-called elementary species or the units out of which
the ordinary species are built up. These historical causes are simply
the specific characters, since in the origin of a species one or more of
these must have been changed, thus producing the characters of the new
type. These changes must, of course, also be due partly to internal and
partly to external causes.

In contrast to these changes of the internal causes, the ordinary
variability which is exhibited during the life-time of a species
is called fluctuating variability. The name mutations or mutating
variability is then given to the changes in the specific characters.
It is desirable to consider these two main divisions of variability
separately.

In the case of fluctuations the internal causes, as well as the external
ones, are often apparent. The specific characters may be designated as
the mean about which the observed forms vary. Almost every character may
be developed to a greater or a less degree, but the variations of the
single characters producing a small deviation from the mean are usually
the commonest. The limits of these fluctuations may be called wide or
narrow, according to the way we look at them, but in numerous cases the
extreme on the favoured side hardly surpasses double the value of that
on the other side. The degree of this development, for every individual
and for every organ, is dependent mainly on nutrition. Better
nourishment or an increased supply of food produces a higher
development; only it is not always easy to determine which direction
is the fuller and which is the poorer one. The differences among
individuals grown from dif