It may be concluded from the whole account which I have given of the structure of the ovulum, that the more important changes consequent to real, or even to spurious fecundation, must take place within the nucleus:  and that the albumen, properly so called, may be formed either by a deposition or secretion of granular matter in the utriculi of the amnios, or in those of the nucleus itself, or lastly, that two substances having these distinct origins, and very different textures, may co-exist in the ripe seed, as is probably the case in Scitamineae.

On the subject of the ovulum, as contained in an ovarium, I shall at present make but one other remark, which forms a necessary introduction to the observations that follow.

ON THE STRUCTURE OF THE FEMALE FLOWER IN CYCADEAE AND CONIFERAE.

That the apex of the nucleus is the point of the ovulum where impregnation takes place, is at least highly probable, both from the constancy in the appearance of the embryo at that point, and from the very general inversion of the nucleus; for by this inversion its apex is brought nearly, or absolutely, into contact with that part of the parietes of the ovarium, by which the influence of the pollen may be supposed to be communicated.  In several of those families of plants, however, in which the nucleus is not inverted, and the placentae are polyspermous, as Cistineae,* it is difficult to comprehend in what manner this influence can reach its apex externally, except on the supposition, not hastily to be admitted, of an impregnating aura filling the cavity of the ovarium; or by the complete separation of the fecundating tubes from the placentae, which, however, in such cases I have never been able to detect.

(Footnote.  This structure of ovulum, indicated by that of the seed, as characterizing and defining the limits of Cistineae (namely, Cistus, Helianthemum, Hudsonia and Lechea) I communicated to Dr. Hooker, by whom it is noticed in his Flora Scotica (page 284) published in 1821; where, however, an observation is added respecting Gaertner’s description of Cistus and Helianthemum, for which I am not accountable.)

It would entirely remove the doubts that may exist respecting the point of impregnation, if cases could be produced where the ovarium was either altogether wanting, or so imperfectly formed, that the ovulum itself became directly exposed to the action of the pollen, or its fovilla; its apex, as well as the orifice of its immediate covering, being modified and developed to adapt them to this economy.

But such, I believe, is the real explanation of the structure of Cycadeae, of Coniferae, of Ephedra, and even of Gnetum, of which Thoa of Aublet is a species.

To this view the most formidable objection would be removed, were it admitted, in conformity with the preceding observations, that the apex of the nucleus, or supposed point of impregnation, has no organic connexion with the parietes of the ovarium.  In support of it, also, as far as regards the direct action of the pollen on the ovulum, numerous instances of analogous economy in the animal kingdom may be adduced.