EMBRYOLOGY OF THE AMENTIFERJE, 411 
development, though such a case is shown in fig. 9, e.s., e.s. I have not been able to trace 
the organie base of these axial strands in material gathered later than Мау 14, ten days 
after the date of the ovule drawn in fig. 39. In the case referred to, the tendency to 
develop one strand more prominently than the rest, which we sometimes observe even 
іп Fagus, is shown by the semi-diagrammatic figure (fig. 5), which should, however, be 
contrasted with that given in fig. 6. 
The embryo-sac most frequently develops from a cell that lies about fourth or fifth 
down from the epidermal cap. "There is no other mark by which to distinguish it from 
its sister cells above and below it in an axile row than the formation of the two nuclei 
which take up their position at either end of the cell (fig. 7, e.s.). But we may, perhaps, 
from analogy with other Angiosperms, tentatively regard the cells intervening between 
it and the epidermis as tapetal. 
Pollination occurred on Мау 9, just three weeks before the pollen-tubes reached the 
embryo-sac. On May 24 the ovarian cavity was crowded with branching pollen-tubes, 
three or four of which have been seen to enter a micropyle simultaneously (fig. 11). 
They enter the epidermal cap and make their way to the apex of the embryo-sac. Here 
one was observed to penetrate for a short distance between the two synergidz which 
abutted on the apex of the sac, but I have not succeeded in observing the transfer or 
even presence of the male nucleus. The converging tiers of cells of the epidermal cap 
are doubtless of service in directing the course of the pollen-tube, and their marked 
turgidity is probably also advantageous, for the cells collapse and flatten after fertilization 
has taken place. Fig. 12 shows a pollen-tube plunging between these receptive cells. 
The embryo-sae presents nothing abnormal up to the period of fertilization. Аб this 
time the antipodals acquire a thick cellulose wall, and are thus prevented from entering 
the cecum, which is now extruded from the base of the вас. "The definitive nucleus 
enters this cecum, which sometimes reaches to the base of the nucellus, taking its course 
down one of the tiers of cells already described (fig. 13). 
То sum up :—Fagus is normal in the structure of its embryo-sac and in the course taken 
by the pollen-tube. - Its sporogenous tissue agrees far more with Strasburger's well-known 
description of Rosa livida than with Treub’s account of Casuarina, and the cecum is 
only formed оп, or shortly before, fertilization. 
CasrANEA. (Pl. LXVIII. figs. 14-21; РІ. LXIX. figs. 22-27.) 
. Тһе species examined has been Castanea vulgaris, Lam. (C. sativa, Mill.), and to this I 
refer when employing the generic name. The appearance of the young nucellus is best 
understood by a reference to fig. 14. Essentially the same structure obtains as in Fagus, 
but the lateral strands no longer, in their early stages, suggest a sporogenous tissue. The 
central strand of five to eight cells is obviously the archesporium. The two to four upper 
square cells, the uppermost of which abuts immediately on the epidermis, appear to be 
very evidently tapetal. The embryo-sac lies immediately below them, and we find two 
sister cells continuing the row, which I have several times observed to be bi-nucleate 
(fig. 18,5.с.). Occasionally these cells undergo further division, as is shown in fig. 20, ат, 
They enlarge and compress the neighbouring cells, some of which may be observed to 
3M 2 
