42 BOTANICAL GAZETTE [JULY 
3. The embryo sac at the eight-celled stage is small, somewhat 
pointed at the micropylar end, and rests upon a stalklike portion 
of the nucellus. A tapetal layer of cells from the inner integument 
completely surrounds it. The egg apparatus in this stage is not 
conspicuous; the polars are large and striking in appearance; the 
antipodals are found with so much difficulty that it. is probable that 
one or more of the cells is often lacking. 
4. The studies of pollination and fertilization are not complete. 
Little evidence of pollination has been found and none of fertili- 
zation. The production of seedless fruit is probably involved in 
the problem of pollination. 
5. After the flowers fall, the whole ovule increases rapidly in 
size. The egg enlarges and becomes filled with densely staining 
globules. The primary endosperm nucleus divides early and the 
endosperm fills the sac, and then crowds the inner integument 
quite up to the dense outer one. 
6. The embryo is late in appearing. The earliest stage identi- 
fied was a three-celled one in the extreme micropylar region. The 
tendency to variation seen in many of the stages is shown here in 
the two types found, the freak embryos and the case of poly- 
embryony. 
7. Pollen mother cells were found on a tree a week before the 
older flowers opened. The mother cells are large and the whole 
mass is surrounded by a single tapetal layer. The spindle in the 
tetrad formation is small, the chromosomes being 30 or more. 
The pollen grains show some difference in size, and frequently 
only one nucleus could be distinguished. 
I am very much indebted to Professors Joun M. Courter 
and CHARLES J. CHAMBERLAIN for assistance in the preparation 
of this paper, and also to the many friends who have so generously 
supplied me with material. 
AUBURN, Inp. 
