408 BOTANICAL GAZETTE [NOVEMBER 
means an unusual feature in angiosperms. With the beginning 
of enlargement of the functioning megaspore the degeneration 
process becomes apparent (figs. 31, 32). It is always characterized 
by increasing density of both cytoplasm and nucleus, plasmolysis of 
the cytoplasm, and final collapse into a distorted strand in the 
center of the cell (figs. 36, 37). The disturbance occurs with 
increasing frequency through the various stages of embryo sac 
formation, 2- (figs. 39, 40, 41), 4- (fig. 43), and 8-nucleate (fig. 45) 
stages being found in more or less advanced degeneration, until by 
the time the sac should be mature, scarcely one remains untouched. 
In all the sections made and examined, I have seen just one normal 
appearing embryo sac. It is not a case of having overlooked the 
stage, for repeatedly late buds and open flowers showed the remains 
of sacs in all stages of disorganization (fig. 47). 
At first, and in early embryo sac stages, only the sac itself is 
involved, but later the entire ovule becomes shriveled, with the 
individual cells deeply staining. If the process has begun early 
enough, it involves the entire ovary also. An open flower, therefore, 
may show all parts of the ovary normal except the embryo sac 
(fig. 13), or both embryo sac and ovule may be degenerating (fig. 14), 
or the entire ovary may be shriveled up, brown, and dead, a mere 
husk projecting up from the base of the flower (fig. 15). 
Degenerations do not halt here, but may attack those ovaries 
in which embryo formation has begun. At any stage in the 
development of the fruit the ovule may collapse about the embryo 
and endosperm, which then take on the characteristic dense, 
deeply staining appearance. The ovary itself appears sometimes 
to continue normal development, at least for a time, or it may 
follow the other parts in degeneration. Certainly not more than 
10 per cent of the flowers examined in section contained embryos, 
and of these not more than 10 per cent had the appearance of 
being able to reach maturity in a normal manner. 
It is very striking that in every case a plate of dense impervious 
cells, several layers thick, was formed across the chalaza, con- 
necting at the edges with the impervious layer over the outer 
integument and the ovary walls. Other and irregular patches of 
similar tissue appear in the funiculus, often seeming to involve 
