have assumed entirely the dimensions and appearance of
leaves.

Median prolification has occasionally been recorded in flowers that
have, in their ordinary condition, but one carpel, as in _Leguminosae_
and in _Santalaceae_. In _Leguminosae_, as also in _Amygdalus_, it would
seem as if the adventitious bud were strictly a lateral and axillary
production, and moreover that the carpel itself is not strictly terminal
but lateral in position, though apparently terminal from the abortion of
other carpels. In the only recorded instance that I am aware of, of this
malformation affecting the genus _Thesium_, the pistil was altogether
absent, and occupying its place was the new bud or branch.[124]

[Illustration: FIG. 61.--_Daucus Carota_, showing leafly carpels,
prolification, &c.]

As the carpels are not unfrequently absent in cases of median
prolification, it has been thought that the pistil in such cases was
metamorphosed into a stem bearing leaves or flowers. Setting aside the
physiological difficulties in the way of accepting such an opinion, an
examination of any number of cases is sufficient to refute it; for, as
Moquin well remarks, the carpels may frequently be found either in an
unaltered condition or more or less modified.

If the pistil be normally syncarpous, its constituent carpels, if
present at all in the prolified flower, become disjoined one from the
other to allow of the passage between them of the prolonged axis; thus
in some malformed flowers of _Daucus Carota_ gathered in Switzerland
(fig. 61), not only was the calyx partially detached from the pistil,
but the carpels themselves were leaf-like, disjoined, and unprovided
with ovules; between them rose a central prolongation of the axis, which
almost immediately divided into two branches, each terminated by a small
umbel of perfect flowers, surrounded by minute bracts.[125]

Not only are the carpels thus frequently separated one from the other by
the prolonged axis, but they undergo commonly a still further change in
becoming more or less completely foliaceous, as in the _Daucus_ just
mentioned, where the carpels were prolonged into two lance-shaped
leaves, whose margins in some cases were slightly incurved at the apex,
forcibly calling to mind the long "beaks" that some Umbelliferous genera
have terminating their fruits--for instance, _Scandix_. Dr. Norman, in
the fourth series of the 'Annales des Sciences,' vol. ix, has described
a prolification of t