14 ON THE SEXUAL ORGANS OF THE CYCADACEEX. 
cular layer is not generally completely exposed ; there still adheres an 
excessively thin layer of cellular tissue which may belong (as has been 
already stated) to the epithelial layer of the nucleus, or perhaps is 
formed of the first cells of the endosperm. 
It is generally known that the nucleus of Cycads, which is more or 
less ovoid, is united for about two-thirds of its height with the coat, 
but that it terminates above in a free summit, which is more or less 
conical (Hooker calls it the cone in Welwitschia). This free portion 
is sometimes higher, sometimes shorter, but generally it is only slightly 
projecting at first, and rises as the nucleus grows, so as to have its 
summit sometimes immediately below the tubular exostome. Ordi- 
narily conical in form, it is sometimes abbreviated, sometimes pro- 
longed into a tube. When the endosperm increases in breadth, it is re- 
duced to a lower level, but in the latter period of the formation of the 
seed, the endosperm pushes it eompletely upwards, and in this case it 
presents itself under an entirely different form. Near its base the cone 
is intimately united with the coat, the internal surface of which, where 
it is not united with the nucleus, is lined with a kind of epidermis, 
which ultimately forms a brown and smooth layer. Externally the 
cone is covered with a layer of dense cells (Schleiden’s epithelium of 
the nucleus, Grundzüge, ii. p. 349; Gotische’s covering.of the nuclear 
protuberance, l.c. p. 380). This becomes less distinct towards the 
base, on the part of the nucleus which is adherent to the coat, but may 
be recognised on the surface of the nucleus after boiling. Although 
this layer, at the point where the nucleus becomes free, is intimately 
united with the free internal surface of the coat, one cannot, especially 
when taking the independent existence of the nucleus into considera- 
tion, regard it as a continuation of the external epidermis; since there 
would thus be a structure in addition of which I know no other ana- 
logue.* 
At the summit of the cone this layer rises a little above the internal 
tissue, and surrounds it like a ring (Plate XCI. fig. 7 f, and fig. 8 and 9). 
The apex of the summit which is not covered by this epithelium, and 
on which the pollen-grains fall to complete their development, and 
* I must on this point differ advisedly from Vogt gr pte eins ii. 
h 395). The point is more evident on comparing the ovules of Conifers 
or umso: pesi daran o. Bot. Zeit. 1855, pl. ii. ind. e numerous Spuren 
ur in the writings ‘of Hofmeister. 
