i aaa 
1903] EMBRYO SAC OF CASUARINA 105 
rise, respectively, to wall tissue and sporogenous tissue. The 
occurrence of massive sporogenous tissue, as reported by Treub, 
is an indubitable fact. Its limits are well defined laterally by 
the larger cells and larger nuclei of the sporogenous tissue. At 
the ends it grades more or less into the surrounding parenchyma, 
and the cells near the microple appear younger than those 
toward the chalaza. I am inclined to believe that the nucellus 
as it elongates carries with it the primary sporogenous cells, each 
of which leaves behind by its own division a train of sporoge- 
nous cells. The formation of sporogenous tissue near the chalaza 
from other than sporogenous cells, as suggested by Treub, was 
not observed, and is believed not to occur in C. stricta. Accord- 
ing to Treub, some of the sporogenous cells absorb others, but 
nothing of the kind was found in C. stricta. Either this species 
differs from those studied by him, or the absorption of sporoge- 
nous cells has been confounded with the absorption of mega- 
spores. C. stricta forms no tracheids in the sporogenous tissue, 
agreeing in this with Treub’s report for C. suberosa. The cells of 
the sporogenous tissue are several times as long as wide (fig. 6), 
while later, when the embryo sacs begin to form, only approxi- 
mately isodiametric cells are apparent. This makes me believe 
that Treub was right in his statement that the formation of four 
megaspores occurs here. A further reason for my conclusion, 
and a stronger one, is that later groups of four in a row are 
recognizable ( figs. 8, 9, 70). 
The differentiation of megaspores begins quite soon after 
their formation. Usually one of each group of four starts, but 
sometimes more (figs. ro, 74). Many get no further than an 
enlarged nucleus, while others reach various stages of develop- 
ment, up to an apparently fertilizable sac. A row of four, as in 
Jig. To, certainly presents a strong argument, if any further argu- 
ment is required, for the megasporic nature of the cells compos- 
ing such a row in angiosperms. The sterile megaspores are not 
all resorbed; but some certainly are, and it seems to me that 
most of them are. The number of sacs reaching maturity varies 
greatly in different ovules, but ranges from two to twelve. These 
are mostly those of central location, forming an axial core in the 
megasporic tissue. 
