106 BOTANICAL GAZETTE [ AUGUST 
The embryo sac passes through the normal 1, 2, 4, and 
8-celled stages (figs. 9, 10, 14, 15). These stages were not all 
seen in sacs which eventually become fertilized, for among so 
many sacs which reach maturity one cannot tell which will be 
the favored one until it is approached by the pollen tube. How- 
ever, if one finds all these stages in embryo sacs arrested in vari- 
ous degrees of development, one may with reason conclude that 
the one later fertilized passed through the same stages, espe- 
cially since some which remain sterile reach a development equal 
to that of the one later fertilized, so far as could be determined. 
As many as twelve, each having a fully developed egg appara- 
tus, were found in a single ovule. 
Within well-developed sacs three cells are organized at the 
micropylar end, forming the egg apparatus (figs. 74, 16, Fe). Te 
jig. 14, C, the cells have not yet collected definitely enough to 
make it certain which will form the egg apparatus. It is sur- 
mised, however, that three of the cells a—d will function as 
such ; ¢ is probably the antipodal polar. From jigs. 10, 74, 15; 
z6, I can see no escape from the conclusion that the egg appa- 
ratus is normal in its origin. These figures are not exceptions 
chosen here and there to prove a point. They can easily be 
duplicated. ig. 76 is indeed the type of a fully developed sac 
in this species. An examination of the egg apparatus in figs. 75; 
16, 77 shows that there is considerable variation in the form of 
the cells composing it in the different sacs. The egg itself can- 
not be distinguished from the synergids in most cases. Some- 
times cases occur showing less than three cells in the egg 
apparatus, but these are few in number, and it seems to me do 
not warrant a conclusion that Casuarina is different in this 
respect from other angiosperms. 
Normally there are three antipodals. Comparing figs. 70 and 
74, the latter is evidently the older group of megaspores. What 
is seen in fig. rg, C, is just what would be expected in the devel- 
opment of a normal sac. Of the eight cells in fig. rg, C, three 
(f) are collected in a depression in the antipodal region, just 
as one might expect, and about them is massed a quantity 
of protoplasm almost, if not completely, separated from the rest 
