llustration: FIG. 142.--Leafy ovules of _Trifolium repens_, showing
formation of nucleus, &c. After Caspary.]

On the other hand, there are cases in which the leafy coat of the ovule,
in place of being a distinct organ, seems to originate from the margin
of the carpellary leaf itself--to be, as it were, a lobule or small
process of the carpel, and not an absolutely new growth. Thus,
Planchon[274], from an examination of some monstrous flowers of _Drosera
intermedia_, was led to the inference that the ovules are analogous to
hairs on the margins of the leaves. This acute botanist was enabled to
trace all the gradations between the simple cup formed by the
confluence of four glanduliferous hairs and the concave leaf and the
perfect ovule.

Brongniart[275] records ovules of _Delphinium elatum_ existing in the
form of marginal lobes of the carpellary leaf itself; so that each ovule
corresponds to a lobe or large tooth of this leaf, the funiculus, as
well as the raphe, being formed by the median nerve of the lateral lobe.
M. Clos[276] mentions a similar instance in _Aquilegia Skinneri_; and
another is figured in Lindley's 'Elements of Botany,' p. 88, f. 180.

[Illustration: FIG. 143.--Portion of an open foliaceous carpel of
_Delphinium_, with ovules on the lobules.]

Cramer[277], from an examination of several ovular malformations, as
well as from the investigation of the mode of evolution of the ovules,
is led to a similar conclusion with reference to the production of
ovules from the modified lobes of the carpellary leaf. Figs. 143-145,
copied from Cramer, show how the nucleus of the ovule is formed as a new
growth from the surface of the lobes of the leaf in _Delphinium elatum_.

[Illustration: FIG. 144.--Section through marginal lobe of carpel
(_Delphinium_), showing the nucleus (_n_).]

[Illustration: FIG. 145.--Section through marginal lobe of carpel,
showing nucleus and tegument (_Delphinium_).]

[Illustration: FIG. 146.--1. Placenta of _Dianthus_, bearing ovules and
carpels. 2. One of the ovaries separated.]

[Illustration: FIG. 147.--Ovules of _Dianthus_ passing into carpels.]

One of the most singular instances of ovular malformation in record is
that cited by the Rev. M. J. Berkeley, in the 'Gardener's Chronicle,'
September 28th, 1850, p. 612. The plant was a carnation, and its
placenta bore, not only ovules, but also carpels (fig. 146), the latter
originating in a perverted development of the former, so