298 The Botanical Gazette. [July. 
The mature embryo-sac is perfectly normal; the antipodal 
cells are large, and increase in volume with the subsequent 
growth of the embrye-sac. 
Anemonella thalictroides.—In an early stage of the ovule 
of this species there was observed in several instances a num- 
ber of hypodermal cells well marked off from the cells below, 
suggesting a primitive archesporium (figs. 49, 50). It is 
very probable that there is here but one initial cell, as only 
one was found to proceed in its development, and that the 
sharp definition of these cells is due to regularity in growth; 
yet the possibility is not excluded that there is present 
a tendency of more than one ceil to become reproductive. 
As to the origin of the inner integument at z—1, fig. 49, 
there can be no question. In all ovules examined the num- 
ber of cells produced by the initial cell was three (fig. 52). 
The nucellus is rather long and narrow impressing this char- 
acter upon its cells. 
The antipodal end of the gradually enlarging embryo-sac 
is more or less pointed, and the large antipodal cells adapt 
themselves to this space. Extending backward through the 
nucellus from the antipodal end of the newly formed embryo- 
sac is a strand of long narrow cells resembling the rudiment 
of a vascular bundle. This strand is gradually absorbed by 
the encroaching embryo-sac which makes its way wedge-like 
to the base of the nucellus. The nuclei of the antipodal cells 
are large and undergo division. 
Thalictrum dioicum.—In this genus the development of 
the embryo-sac agrees closely with that of Anemonella. It 
is evident (fig. 54) that the great bulk of the nucellus owes 
its origin to the hypodermal layer of cells. This statement 
will apply generally to all the preceding genera. 
As far as observation extended, the initial cell gives rise to 
three cells only, but in all probability four occasionally result 
from two successive divisions, as in the other genera. 
The nucellus is less elongated than in Anemonella; the em- 
bryo-sac more oval in form and it reaches maturity sooner 
with reference to the corresponding growth of the integu- 
ments. The antipodal cells are large, but not so pronounced 
as in Anemonella. 
FHlepatica acutiloba.—The earlier stages in the development 
of the embryo-sac of Hepatica were not observed, as my ma- 
terial, collected about the middle of November, was too far 
