38 DR. J. D. HOOKER ON WELWITSCHIA. 
of which Orders present many correspondences in the structure and functions of their 
organs of fertilization with Welwitschia. It must, however, be borne in mind, that as 
yet no “ Ovular Theory” has been advanced that meets general acceptance, and that, 
until one shall be forthcoming, all such questions can only be answered by guess: at 
present, the whole subject is in a state of confusion, to which the ovule described in the 
present monograph adds not a little. Under many points of view, every ovule must be 
considered a terminal organ, and hence axial in relation to the vascular system of the 
plant, whether developed in the apex of a floral axis, as in Welwitschia, or on the sur- 
face of a carpellary leaf, as in Nymphea, &c., or on the edges of the same, as in most 
plants. Its integuments may correspond to floral discs, or may represent hairs or 
other epidermal appendages, or may be foliar in origin. However this may be, the 
desideratum is, some general view of the relations of the ovule to the other floral 
organs, under which all the apparent deviations from any hitherto received theory of its 
origin shall be harmonized. 
I am unable to indicate an exact counterpart to the process of impregnation and 
embryo-formation above described, amongst either angiospermous or gymnospermous 
plants. There is a general agreement in many most essential particulars with Cycadee 
and Conifer, especially in the structure of the ovule, its simple integument, the 
application of the pollen-grain to the nucleus, the free embryo-sac being filled with 
endosperm-cells previous to fertilization, the numerous secondary embryo-sacs, the 
position of the germinal vesicle at the base of these sacs, and in the high development of 
the long tortuous suspensor. The prominent differences are,—the absorption of the 
membrane of the upper part of the embryo-sac; the growth upwards of secondary 
embryo-sacs from out of the embryo-sac into the cone of the nucleus, apparently without 
the intervention of “rosette cells ;” the impregnation of those secondary sacs in the 
cone of the nucleus when far removed from the embryo-sac (impregnation thus being, in 
one sense, extra-uterine); and the germinal vesicle at the base of the secondary sacs not 
giving origin to a plurality of suspensors and embryos. In this character of the secondary 
embryo-sacs, which answer in function to the embryo-sac itself in angiospermous plants, 
being for the most part developed and always fertilized in the nucleus outside the primary 
sac, Welwitschia presents an embryogenic process intermediate between that of Gymno- 
sperms and Angiosperms. 
There seems to be much variety in the early development of the secondary embryo-sacs 
in Conjfere. In Juniperus especially many irregularities are described, the sacs sometimes 
originating in a remarkable increase of size in the deep-seated cells of the endosperm ; 
and sometimes a secondary sac opens in the middle of one of the lateral surfaces of the 
endosperm (Hoffmeister, ‘On the Higher Cryptogams,’ &e., Currey’s transl. p. 410). This 
is a step towards what occurs in Welwitschia when they actually leave the endosperm. 
In Conifere it would appear that there are always two separate periods of the elonga- 
tion of the pollen-tube; that is to say, that after the pollen has been placed on the apex 
of the nucleus, its tube descends and rests within the nucleus at a certain distance from 
its apex and from the embryo-sac, and that, after a considerable interval, the tube elon- 
se ee 
