brs Pa ot Tee) Pee Lene 
RO OSL Se Date ee Saree a eae © eter a 
Be Sea Ma dca set ys A Aes 
DEST Ey eee 
DR. J. D. HOOKER ON WELWITSCHIA. 37 
Ripe Seed.—This (Plate VII. fig. 13) occupies the centre of the pericarp (figs. 3, 4), is 
obovoid, compressed, terminated by the calyptriform membranous integument of the 
ovule, with its rigid styliform apex. Removing the calyptriform integument (fig. 14), 
the cone comes into view, the outer tissues of which are continuous with the walls of the 
nucleus, now the only integument of the albumen. The granular albumen, with its 
fleshy neck, is quite free ; and the suspensor forms a coil within or over the mouth of the 
neck, on the summit of the contained embryo; the connexion between the suspensor 
and cone (by means of the tubular prolongation of the secondary embryo-sac) is usually 
ruptured, but sometimes persists (fig. 15). 
The embryo (figs. 15, 16) is linear, terete, or slightly flattened, about $ths the length of 
the albumen, and occupies the cavity in the axis of the latter. The lower end is contracted 
into two small flat cotyledons, which are closely applied to one another, and include no 
plumule. The radicular end swells out into an uneven fleshy mass, which is the cellular 
base of the suspensor, and is lodged in the equally fleshy annular neck of the albumen. 
I have throughout described the thick vascular investing coat of the seed, which is 
developed as the endosperm enlarges and descends, as appertaining to the nucleus, because 
it is continuous downwards uninterruptedly from the body of the nucleus as that nucleus 
existed previous to impregnation, at the time when the insertion of the ovular integument 
coincided with the base of the ovule; but Prof. Oliver has forcibly directed my attention 
to the claims which this seminal integument may have to be regarded as an urceolate 
prolongation of the axis of the flower, and not a development of the nucleus, which the 
parts above the insertion of the calyptriform integument indisputably are. In favour of 
regarding it as belonging to the axis are, 1, the two or more vascular bundles which are 
carried up in its walls, and whose terminations coincide at all periods with the base of the 
calyptriform integument, thus marking in the seed, as they did in the ovule, the boundary 
of what is certainly ovular, and what axial ; 2, some important considerations suggested 
by the embryogeny of Loranthacee, upon which Prof. Oliver is at present engaged. 
Against this axial view may be urged, 1, the absence of any ring or thickened 
margin on the integument of the seed, at the point of union of the axial and ovular 
portions, and the unbroken continuity of the tissues of the integument, externally with 
the calyptriform integument, and internally with the substance of the cone; 2, the cir- 
cumstance that, of the two integuments investing the albumen of Gnetwm, the inner 
one corresponds with that of Welwitschia, and must thus also be considered axial; which 
requires us to believe, either that the axis of the flower of Gnetum is produced upwards 
within the base of the outer integument of the ovule (whose insertion, unlike that of 
the inner coat, is found at the base of the seed), or that this outer coat of Gnetum is 
not ovular at all; 3, that in respect of the carrying up of the integument immediately 
investing the ovule of Welwitschia, that ovule differs only in degree from those of 
many angiospermous plants, and that under this view the two vascular cords do not 
inaptly represent a double raphe of very anomalous development and character. 
I am unable further to discuss this curious point, which, for further elucidation, must 
await Prof. Oliver’s researches on the embryogeny of Loranthacee and Gnetacee, both 
