ce.

For purposes of reference, a standard of comparison is required; and
this standard, so long as its nature is not overlooked, may, indeed must
be, to some extent, an arbitrary one. Thus in the phanerogamous plants
there is assumed to exist, in all cases, an axis (stem, branches, roots,
thalamus, &c.), bearing leaves and flowers. These latter consist of four
whorls, calyx, corolla, stamens, and pistils, each whorl consisting of
so many separate pieces in determinate position and numbers, and of
regular proportionate size. A very close approach to such a flower
occurs normally in _Limnanthes_ and _Crassula_, and, indeed, in a large
proportion of all flowers in an early stage of development. To a
standard type, such as just mentioned, all the varied forms that are met
with, either in normal or abnormal morphology, may be referred by
bearing in mind the different modifications and adaptations that the
organs have to undergo in the course of their development. Some parts
after a time may cease to grow, others may grow in an inordinate degree,
and so on; and thus, great as may be the ultimate divergences from the
assumed standard, they may all readily be explained by the operation,
simply or conjointly, of some of the four principal causes of
malformation before alluded to. The fact that so many and such varied
changes can thus readily be explained is not only a matter of
convenience, but may be taken as evidence that the standard of reference
is not wholly arbitrary and artificial, but that it is a close
approximation to the truth.

It has already been said that an arrangement like that here considered
as typical is natural to some flowers in their adult state, and to a
vast number in their immature condition. It would be no extravagant
hypothesis to surmise that this was the primitive structure of the
flower in the higher plants. Variations from it may have arisen in
course of time, owing to the action of an inherent tendency to vary, or
from external circumstances and varied requirements which may have
induced corresponding adaptations, and which may have been transmitted
in accordance with the principle of hereditary transmission. This
hypothesis necessarily implies a prior simplicity of organisation, of
which, indeed, there is sufficient proof; many cases of malformation
can thus be considered as so many reversions to the ancestral form.

Thus, teratology often serves as an aid in the study of morphology in
genera