In these flowers the placenta seemed to be composed of several funiculi soldered together, and bearing imperfect ovules. In other cases no traces of ovules are visible, but the funiculi are in a foliaceous condition. Moquin also alludes to a case of the same nature in _Cortusa Mathioli_, in which the funiculi bore little rounded leaves. Brongniart has described some malformations of _Primula sinensis_ in which the ovules were transformed wholly or partially into small leaves with three to five lobes.[269] Dr. Marchand[270] mentions similar changes in _Anagallis arvensis_ and _Lonicera Periclymenum_. Cramer[271] figures ovules of _Primula sinensis_ in the form of stalked leaves, often becoming infolded at the margins, and giving origin to a small nucleus on their inner surface. M. Tassi[272] records an instance in _Symphytum officinale_ wherein the ovules were replaced by two small linear leaves arising entirely from the axis, and not from the carpels. In most of the foregoing illustrations the foliar portion of the ovule must have been independent of the carpel; this independence is less manifest, though probably as real in the cases now to be mentioned. In _Sinapis_ and in _Brassica oleracea_ foliaceous ovules may occasionally be seen, attached to the placenta by long stalks. No trace of the nucleus is visible in these specimens. [Illustration: FIG. 140.--_Sinapis_, replum and ovules; the dotted line shows the position of the carpels.] Griffith, in alluding to a similar case in _Sinapis_,[273] describes the ovules as foliaceous, and having their backs turned away from the axis, the raphe being next to the axis and representing the midrib the funicle corresponding to the petiole. The outer tegument of the ovule, according to Griffith, is a leaf united along its margins, but always more or less open at its apex. No inversion can, therefore, really take place in anatropous ovules, but the blade of the leaf is bent back on the funicle, with which its margins also cohere. Caspary, in an elaborate paper on phyllomorphy occurring in _Trifolium repens_, figures foliaceous ovules springing from the edge of an open, leafy carpel. The nucleus of the ovule, in these cases, appears to originate as a little bud from the surface of the leafy ovule (figs. 141, 142). [Illustration: FIG. 141.--Leafy ovules, &c., _Trifolium repens_.] In a species of _Triumfetta_ (see p. 260), of which I examined dried specimens, the ovary was open and partly foliaceous; it bore on its infolded margins ten erect leaflets, representing so many ovules; each leaflet was conduplicate, the back being turned towards the placenta. [Illustration: FIG. 142.--Leafy ovules of _Trifolium repens_, showing formation of nucleus, &c. After Caspary.]