1897] PLACE OF ISOLATION IN ORGANIC EVOLUTION 248 
uncertain and irregular. There is still another reason for coming 
to the same conclusion. If it be good for a plant to have its flowers 
fertilised by pollen from other plants, then the grouping of flowers 
into a head or spike must be injurious, because it almost insures 
that the flowers shall be fertilised by pollen from other flowers of 
the same inflorescence, which Darwin says does little or no good; 
and yet plants with capitate flowers are numerous and prosperous. 
That cross-fertilisation is useful is proved by the fact that all 
the higher animals are bisexual; but that it is not essential is also 
proved by the existence of large numbers of asexual organisms, and 
by the fact that the highest plants are hermaphrodite. The first 
phanerogams were, no doubt, diclinous and anemophilous. As the 
number of species increased the individuals of each species would 
diminish, and, consequently, fertilisation by the wind would become 
more uncertain ; and, probably, it is for this reason that the higher 
angiosperms became hermaphrodite. But this hermaphroditism was 
tempered by dichogamy, the origin of which is not connected with 
insect fertilisation. The visits of insects came later. They com- 
menced to cultivate, as it were, the flowers for their own use, each 
species trying to preserve the honey for its exclusive benefit ; for it 
is evident that all these floral arrangements—including colour, capi- 
tate flowers, scent, etc.—are very useful to the anthophilous insects, 
for whom the honey is preserved. But these flowers, elaborated by 
insects for their own benefit, have secured complete isolation for the 
plants to which they belong, and the variations have therefore been 
preserved, whether they were useful or indifferent, or even when they 
were injurious, as in the reduction of stigmatic surface in the 
orchids, the abortion of one half of each anther in Salvia, and the 
asexual condition of the ray-florets in some of the Compositae. All 
the changes, however, are useful to those insects which alone can 
fertilise the flowers, and Dr Hermann Miiller thinks that different 
kinds of insects have evolved different kinds of flowers suited to 
their tastes. In fact, these flowers have been cultivated by moths 
and bees, just as ants have domesticated some beetles and aphides. 
The plants that have escaped from their cultivators have run wild 
again, like rabbits in Australia and New Zealand. 
Isolation by Elimination—or natural selection in Darwin’s sense 
—anust always have a utilitarian cause, because the elimination is for 
the benefit of the remainder—that is, for the selected. It may be 
a struggle for food, or it may be a struggle for protection against 
eneinies, or it may be a struggle to secure the persistence of the 
species ; but in all cases it must be a struggle with death as the 
penalty for being vanquished, because, without elimination by death, 
there can be no selection and no isolation. It is only the struggle 
