Embryo-Sac and Embryo of Geissoloma marginata.347 
of embryo-sac found in that order can be considered as more 
primitive than the normal Angiospermous type, or as derived from 
it. In the Penaeaceae, both the structure of the embryo, apparently 
highly adapted to xerophytic conditions, 1 and the comparatively 
feeble development of the endosperm, probably point to a higher 
degree of specialisation than is the case in Geissoloma. The embryo- 
sac of the Penaeaceae then might be expected to show a corres¬ 
pondingly higher degree of development. As the embryo-sac of Geis¬ 
soloma is, if anything, rather specialised from the normal type (as is 
shown by the marked differentiation in the egg-apparatus, and the 
evanescence of the antipodals), it may be inferred that the 
Penaeaceae probably show a still further specialisation. Thus the 
evidence derived from the study of this form supports the con¬ 
clusion that in the embryo-sac of the Penaeaceae we have, not a 
relatively primitive type, but rather one which is derived from 
the normal by the inclusion of four megaspores in its development. 
Summary. 
Geissoloma marginata is the only representative of the order 
Geissolomaceae, which is closely allied to Penaeaceae. The embryo- 
sac is probably derived from one of a row of megaspores. Its 
development is that of a typical Angiosperm, but the antipodal 
cells are very evanescent. It contains much starch. The pro¬ 
embryo is at first pear-shaped, later spherical, with no suspensor. 
The embryo lies in a copious endosperm. 
The fact that this close ally of the Penaeaceae has a normal 
embryo-sac tends to uphold the view that Geissoloma is to be 
regarded as a representative of a separate order rather than as an 
anomalous genus of the Penaeaceae. 
This investigation has been carried out in the Cambridge 
Botany School, by kind permission of Professor Seward. 
1 Stephens, E. L. loc. cit., p. 369. 
EXPLANATION OF PLATE 
Illustrating Miss Stephens’ Paper on the Embryo-Sac and Embryo 
of GEISSOLOMA MARGINATA. 
Fig. 1. Binucleate embryo-sac ; the disintegrating mass above may represent 
the remains of the other megaspores, x 430. 
Fig. 2. Three nucleate embryo-sac. x 430. 
Fig. 3. Four nucleate embryo-sac. x 430. 
