r in the two
upper petals; and that when this occurs, the adherent nectary is quite
aborted; when the colour is absent from only one of the two upper petals,
the nectary is only much shortened.

With respect to the difference in the corolla of the central and exterior
flowers of a head or umbel, I do not feel at all sure that C. C. Sprengel's
idea that the ray-florets serve to attract insects, whose agency is highly
advantageous in the fertilisation of plants of {146} these two orders, is
so far-fetched, as it may at first appear: and if it be advantageous,
natural selection may have come into play. But in regard to the differences
both in the internal and external structure of the seeds, which are not
always correlated with any differences in the flowers, it seems impossible
that they can be in any way advantageous to the plant: yet in the
Umbelliferae these differences are of such apparent importance--the seeds
being in some cases, according to Tausch, orthospermous in the exterior
flowers and coelospermous in the central flowers,--that the elder De
Candolle founded his main divisions of the order on analogous differences.
Hence we see that modifications of structure, viewed by systematists as of
high value, may be wholly due to unknown laws of correlated growth, and
without being, as far as we can see, of the slightest service to the
species.

We may often falsely attribute to correlation of growth, structures which
are common to whole groups of species, and which in truth are simply due to
inheritance; for an ancient progenitor may have acquired through natural
selection some one modification in structure, and, after thousands of
generations, some other and independent modification; and these two
modifications, having been transmitted to a whole group of descendants with
diverse habits, would naturally be thought to be correlated in some
necessary manner. So, again, I do not doubt that some apparent
correlations, occurring throughout whole orders, are entirely due to the
manner alone in which natural selection can act. For instance, Alph. De
Candolle has remarked that winged seeds are never found in fruits which do
not open: I should explain the rule by the fact that seeds could not
gradually become winged through natural selection, except in fruits which
opened; so that the individual plants producing {147} seeds which were a
little better fitted to be wafted further, might get an advantage over
those producing se